thread_key,title,primary_bucket,attention_score,maintainer_next_action,active_item_count,context_item_count,active_issue_numbers,active_pr_numbers,context_issue_numbers,context_pr_numbers,relationship_edge_count,thread_hash
thread:issue:188812:pr:188801,pr #188801: [MPS] Leak MetalShaderLibrary bundled singleton to avoid exit-time destructor crashes,waiting on contributor,151,Wait for contributor update on PR #188801; keep related issue/PR context attached.,2,0,188812,188801,,,1,a3bf660df1c69f086e3d8fd8e680a6b0154f5476f9305f8811f08b4be130e680
thread:issue:158861:issue:188900:pr:189122,pr #189122: Fix sparse-dense mul dropping data when broadcasting a size-1 sparse dim,ready for maintainer decision,139,Review active PRs and linked issues as one maintainer work thread.,3,0,"158861, 188900",189122,,,4,460e127c1dec1e9fc3b7d239622770a5d842cc9ecf37aa76e546f59436e621c7
thread:issue:186535,"issue #186535: Windows, gloo: Access violation (0xC0000005) in ProcessGroupGloo::enqueue when calling allreduce on CUDA tensors — GlooAllreduceRegistry has no kCUDA creator",needs triage,139,needs triage,1,0,186535,,,,0,4589fb50821f36395b185b8e48fa752cb9494aa33b7a85dc29b991676c6002a8
thread:issue:188323,issue #188323: [Inductor][CPU] dynamic=True convolution lowering crashes with ValueError: Exponent must be non-negative,needs triage,139,needs triage,1,0,188323,,,,0,45573294530ec58bf705a63feef08896e95de263ca1cabfc9ec0d8b06edbdaa1
thread:issue:157668,issue #157668: NCCL error caused due to use of NVLS in torch 2.7.1-cu128 on aarch64 gb200 cluster,needs triage,127,needs triage,1,0,157668,,,,0,be1cd8fce244206e05b69b9f2f9ac6d7f0c46fedba7503b53781ae8408f549f7
thread:issue:187912,issue #187912: [CPU] Concurrent `cpublas::brgemm` calls can crash in the AMX path when the underlying oneDNN ukernel is shared,needs triage,127,needs triage,1,0,187912,,,,0,9198ae20d77b32c855b641422922e5a0d25b5c7d91cfbf7e4669dc8fc3156b6e
thread:issue:150022:issue:150371:pr:185730,pr #185730: Fix dynamic shapes for variadic kwargs,PR blocked,121,Review active PRs and linked issues as one maintainer work thread.,3,0,"150022, 150371",185730,,,5,e825da66735fbec4314c576939feec9a1fbb57a37f70184e1906e59681e6b236
thread:issue:116254:issue:162476,issue #116254: C++ API `at::quantized_max_pool2d`: Heap-buffer-overflow,needs triage,121,needs triage,2,0,"116254, 162476",,,,1,356b1d6a7d4827805028373a8130032160586cfb9cc1d8154321454028bdd218
thread:issue:162422,issue #162422: Runtime failure when running torch.compile() & using GCC 11.5.0 on Neoverse V1,needs triage,121,needs triage,1,0,162422,,,,0,f791895686cc5f00bc83b3adc6e2e2d507edbffb2953092f8832d7d3f9b259f3
thread:issue:154297:issue:162178:issue:162745:issue:162748,issue #154297: Hangs and timeouts on dist.reduce_scatter on B200 GPU,needs maintainer decision,115,needs maintainer decision,14,0,"154297, 162178, 162745, 162748, 162820, 162871, 162897, 162917, 162940, 163429, 165170, 165685, 187158, 189065",,,,21,4191e919c63590f4dfe2fa95e8da1c65296683f2cd005e2608fa451d2f36845a
thread:issue:55655:issue:57118:issue:68768:issue:72948,issue #55655: Performance debugging / warning mode,stale/low urgency,115,stale/low urgency,5,0,"55655, 57118, 68768, 72948, 75725",,,,4,cdee6597b7655ca266eab42545209c64b9f252abda88de970d6b0c490c7b995c
thread:issue:66504:issue:68407:issue:73332,issue #66504: BatchNorm runtimeError: one of the variables needed for gradient computation has been modified by an inplace operation,needs triage,115,needs triage,3,0,"66504, 68407, 73332",,,,3,78d5f28595b7306d30f9d8271f5d8891a18e95ed28276d9f21e7a6a7d6d3d3bf
thread:issue:144965:issue:69078,"issue #144965: RuntimeError ""global alloc not supported yet"" when using TorchScript optimization.",needs triage,115,needs triage,2,0,"69078, 144965",,,,1,1214e1c9672e8873e03347dcde07f2872332c99becd9778190b8d51fff08921b
thread:issue:106164,issue #106164: distributed.batch_isend_irecv() crash when send/recv refers to itself,needs triage,115,needs triage,1,0,106164,,,,0,454a57b4ea49987c295830a93f52e92f1b5c1d439e4edb5bcf435c644c58a436
thread:issue:116423,issue #116423: PyTorch Distributed Elastic Launch Segmentation Fault with Python 3.12,needs triage,115,needs triage,1,0,116423,,,,0,11bb25750ec6837ba3c59a137b97e66a4d0096091a7b36ad2570a0f740c8a57b
thread:issue:119845,issue #119845: Segmentation fault in dataloader worker sub-process,needs triage,115,needs triage,1,0,119845,,,,0,0f1591593006e0a4832df42c548b01f84fce527d8197a1b48d20ea4aae89f816
thread:issue:145610,issue #145610: mmap fails on 64k page aarch64 systems for AOTI model loading,needs triage,115,needs triage,1,0,145610,,,,0,a51c9753724e6505823958586d1ec2bde4607da876135773187747373c3c365e
thread:issue:153517,issue #153517: [CI][CUDA][Distributed] test_non_blocking_with_eager_init timeout,needs triage,115,needs triage,1,0,153517,,,,0,0ec276a5c2eca0df29156e642f603b86e3d519ba613ce01fdc531a2fdaa3dea6
thread:issue:162731,issue #162731: DTensor cached op propagation results can be mutated when propagating other ops,needs triage,115,needs triage,1,0,162731,,,,0,a5bc902d427cc4e517bb97ade57bb7691ba47b40b7279377d62fd51b72a247dd
thread:issue:167693,issue #167693: DISABLED test_side_stream_backward_overlap_cuda (__main__.TestAutogradStreamSynchronizationCUDA),needs triage,115,needs triage,1,0,167693,,,,0,25457a3be87483761b160e0011a6852e489b861723f883b9ad93022f5a8e71f5
thread:issue:170003,issue #170003: batch_isend_irecv with nccl causes illegal memory access depending on P2P ordering,needs triage,115,needs triage,1,0,170003,,,,0,4e9426a63bf1286912bf22207d7e399b3b5dc95f194cbb1f0ffb9a86d5f54014
thread:issue:173921,issue #173921: [Inductor] Significant numerical divergence (4.7% relative error) in Conv2d on CPU with torch.compile,needs triage,115,needs triage,1,0,173921,,,,0,d5a0991c6cf536e90ed64b9179b3a5097fd9211ab95a0556af731982db5aa694
thread:issue:174288,issue #174288: [distributed] Batched isend/irecv with NCCL backend hangs on high load,needs triage,115,needs triage,1,0,174288,,,,0,39f025dbba7d0f303fe75085a9c1750bb5d142069abfb5fdac17ef9442311b65
thread:issue:179502,issue #179502: [DTensor] sharded view incorrectly passes when redistribution is needed,needs triage,115,needs triage,1,0,179502,,,,0,7c425c70e395ec446318a49292862b127a641808902acc624fc3b812f7baa82c
thread:issue:180088,"issue #180088: [DTensor] RNG tracker does not advance state for CPU tensors on CUDA mesh, causing trunc_normal_ infinite loop",needs triage,115,needs triage,1,0,180088,,,,0,5f9e35ad3a10a9b55ac7f681b355d684ed6195d237636c6d3763475e00571154
thread:issue:183459,issue #183459: Regression in torch.distributed._functional_collectives.all_to_all_single in 2.9.1 -> 2.11,needs triage,115,needs triage,1,0,183459,,,,0,84094c1a34f3b447115fe5d7148bbe6b1fb5066179f9961b957de2fef668a2c9
thread:issue:189126,"issue #189126: Report a bug, as requested by the error message",needs triage,115,needs triage,1,0,189126,,,,0,0ab2094909fb97acb60ac33f7154a47b6c8c4e53a32fa393f0c27651483fcc74
thread:issue:189144,issue #189144: [XPU][B580] `flex-attn-causal` performance drop with `last_level_cache_size` cache clear between runs,needs triage,115,needs triage,1,0,189144,,,,0,641fbcf35fbb907651315a0013ec77c874c84be47c1ffe896db152d8cec00464
thread:issue:189150,issue #189150: SIGSEGV / cudaErrorIllegalAddress when replaying a CUDA graph that captured multiple training iterations of two trainers over deeply unrolled recurrent modules,needs triage,115,needs triage,1,0,189150,,,,0,eb6b34928e54ac40f4bd54b1d68011d065dff069e4f29c5bf509d912db34cecd
thread:issue:189194,"issue #189194: [ROCm] torch 2.13 wheel: ""Can't detect vectorized ISA for CPU"" in torch.compile smoke test on non-ROCm image (regression vs 2.12.1)",needs triage,115,needs triage,1,0,189194,,,,0,9f5e91dd9991ed40f0c9adaf5acc75ef3f1a51c967b9c5ca8411ccbdb74e163e
thread:issue:189239,issue #189239: [Inductor]Unable to apply layout optimization on convolution backward causes training performance regression.,needs triage,115,needs triage,1,0,189239,,,,0,171265045b6716feac30e0c6765edf1feb3fc77f48583d9edf378e6819c1dd93
thread:issue:189281,issue #189281: [CPU] operator_benchmark: embedding / embeddingbag ~70-130x slower since ~Nov 2025 (x86 + aarch64; possible measurement issue),needs triage,115,needs triage,1,0,189281,,,,0,f5489b942b916b1b5a8ccdb57fb65a1e83b69171caf419214cc247b8e9b7efae
thread:issue:189282,"issue #189282: [CPU] operator_benchmark: backward-pass (add, + batchnorm on aarch64) ~200-500x slower since ~Nov 2025 (x86 + aarch64; possible measurement artifact)",needs triage,115,needs triage,1,0,189282,,,,0,05d93d1b072ffb60f4091869ca25b590d9fea78486940462bfe51b247a3056f5
thread:issue:71303,issue #71303: [RFC] Cross-Process Performance Analysis: Straggler Detection,needs triage,115,needs triage,1,0,71303,,,,0,dabcb6662052f241b91a3a0f3ad49f88ed49985711e1d04471a115101204ab7d
thread:issue:76449,issue #76449: Enhance _verify_param_shape_across_processes,needs triage,115,needs triage,1,0,76449,,,,0,068787c6ed37843da777870ac7f0dfc809c9d9d4707636af6c179ae6afd5f354
thread:issue:81684,issue #81684: Message exchange failure when perform alltoallv (cpus),needs triage,115,needs triage,1,0,81684,,,,0,74f17e00575a1f79c1d4827abc0afc13f22e9be68ab5ed648a028e8259343caf
thread:issue:85088,issue #85088: reentrant torch.utils.checkpoint does not work with NamedTuple outputs,needs triage,115,needs triage,1,0,85088,,,,0,86e111a118e90b2aa5d89b49f173d3aa1f8c7ffb1064e732bb62060adf59334e
thread:issue:124110:pr:186461,pr #186461: Fix SymBool equality handling in symbolic shapes,ready for maintainer decision,112,Review active PRs and linked issues as one maintainer work thread.,2,0,124110,186461,,,1,55b2924959b6fd36b6b348e0f1488c4b1b6f873caa8cd86d9064440accddd0a4
thread:issue:127153:pr:186429,pr #186429: Fix ONNX export of pad_sequence with symbolic split lengths,ready for maintainer decision,112,Review active PRs and linked issues as one maintainer work thread.,2,0,127153,186429,,,1,ffbdb3b87545a7acb7dfa7036bfd95016c668c20ce0778519e5adb5032a2baa6
thread:issue:151510:pr:185691,pr #185691: Avoid UB in float to signed integer casts,ready for maintainer decision,112,Review active PRs and linked issues as one maintainer work thread.,2,0,151510,185691,,,1,791d817891002c9033ef34e0bd28fa356f6028baaad6b6c443f823a7569cb5e4
thread:issue:178127:pr:184716,pr #184716: Fix conv_transpose2d meta output padding validation,ready for maintainer decision,112,Review active PRs and linked issues as one maintainer work thread.,2,0,178127,184716,,,1,8ea684b27731386885348d7bbd6dd69295e5db47efe3cdfb217e657a6566f1bf
thread:issue:188544:pr:188632,pr #188632: Fix Dynamo opaque object staticmethod guards,ready for maintainer decision,112,Review active PRs and linked issues as one maintainer work thread.,2,0,188544,188632,,,3,5ebbe10413f12e5181dcb1a622abf6428adde5faf07db56a4d1a34c68fb056e5
thread:pr:184939,pr #184939: Support SymInt steps for linspace/logspace export,ready for maintainer decision,112,ready for maintainer decision,1,0,,184939,,,0,89137d6e2a9aac6b008ca548c9b4459361c28189dd3cb8550579dd7533c913c3
thread:pr:184648,pr #184648: [Test] Add has_sufficient_memory() hook to DeviceTypeTestBase,PR blocked,108,PR blocked,1,0,,184648,,,1,2b9c9479cbe58d0f7ba8574d9effbd6e4b58f02a4b37746fba1414fa427eafa6
thread:pr:185882,pr #185882: [PrivateUse1] Return `False` instead of `None` when PU1 backend is not available,PR blocked,108,PR blocked,1,0,,185882,,,0,7b956bb502a3bc469c432fe8e5e6875217e19666da2c245bcd4d2ad30420472b
thread:pr:186927,pr #186927: [MPS] Gemv kernels,PR blocked,108,PR blocked,1,0,,186927,,,0,1a1ce47c21ef57066db9ff44b73194675f03398dbd6d7e109b320123e19709a0
thread:pr:187501,pr #187501: [Stable C Shim] Use error message retrieval shim if available at runtime,PR blocked,108,PR blocked,1,0,,187501,,,0,2033b677facc67f60c343135e65a8f198be35091ae2e163cf155eb26479d0d6e
thread:pr:188237,pr #188237: Optimize CSR SpMM CPU grain size and inner accumulation,PR blocked,108,PR blocked,1,0,,188237,,,1,0502b4836142236e4eed5b5b3ff4c981ef2c7c850ff65ee2d40fa490e0bd3b90
thread:pr:188301,pr #188301: [inductor] Fix inductor dropping ordering dep between effectful ops with different kernel types,PR blocked,108,PR blocked,1,0,,188301,,,2,3a59e7c77533978fa4b7dc21343a3d36d4ab763445111a51f4acb52f5cf1cc1d
thread:pr:189307,pr #189307: [CI/CD] Copy newer CUPTI headers into manywheel binary-build images,PR blocked,108,PR blocked,1,0,,189307,,,0,c2b21bf6f396bf82a9d28defa7a397d03d487c79d2c6f16f3f76e4ea9fab0386
thread:pr:189312,pr #189312: [CUDA][cuBLAS] Change cuBLAS default workspace size for SM 11.0 to 32 MiB,PR blocked,108,PR blocked,1,0,,189312,,,0,9f165e85636ffa0bac0797f8038516c820aac0ac72064b52b0674bddfdfc8891
thread:issue:153410:pr:188948,pr #188948: Fix triangular_solve for sparse CPU tensors on non-MKL platforms,waiting on contributor,107,Wait for contributor update on PR #188948; keep related issue/PR context attached.,2,0,153410,188948,,,2,7ecef50b4ca5e3051a52d8691d4dc243f7265c707cd91b8b1fdee1f8fafd807f
thread:issue:180397:issue:181374:issue:188970:pr:181720,pr #181720: [MPS] Make pin_memory return CPU-aliased storage backed by a unified MTLBuffer,has linked issue,106,Review active PRs and linked issues as one maintainer work thread.,5,0,"180397, 181374, 188970","181720, 189256",,,9,acf73cf161637e8bd3f935ad90cd656c8da098d94f1a439aef082dbaacd06d49
thread:issue:188892:pr:189043,pr #189043: Preload full bundled cuDNN set with RTLD_GLOBAL to prevent sublibrary version mismatch,waiting on contributor,106,Wait for contributor update on PR #189043; keep related issue/PR context attached.,2,0,188892,189043,,,2,e42e4273af33a67a1f5c01f17916dc41410a5f91ef26576611048884d4aa5dc5
thread:pr:135631,pr #135631: [scan] Autograd,PR blocked,100,PR blocked,1,0,,135631,,,0,eec761d639dbffba5b1fd908cc4daa8a724732d4f25248cf598f8470ac3ca091
thread:pr:188115,pr #188115: [Inductor][TEST] Align `test_main_loop_scaling` with H100 support surface,PR blocked,98,PR blocked,1,0,,188115,,,0,4b7d6ad7fa955eba7da9f5d0fd4c32ce63b8a762ff2b441fa983cafdbc156ade
thread:issue:103625:issue:104011:issue:132845:issue:15963,issue #103625: DISABLED test_backward_ddp_outside (__main__.TensorPipeDdpUnderDistAutogradTest),needs maintainer decision,97,needs maintainer decision,11,0,"15963, 16295, 51062, 58856, 64093, 69178, 69179, 94003, 103625, 104011, 132845",,,,10,be77a37a11df6ad8662929dfb6d2fd0bb524df553cce253d944efefae2c090ec
thread:issue:104506:issue:124950:issue:88103,issue #104506: TImeout in NCCL doesn't work,stale/low urgency,97,stale/low urgency,3,0,"88103, 104506, 124950",,,,2,fd6264ec76be0809123748a1165dd05415d097e616e0c13fe6fd9e6d7ee2a0d6
thread:issue:42258:issue:46944:issue:46946,issue #42258: Cannot load certain function from dumped Torchscript file,needs triage,97,needs triage,3,0,"42258, 46944, 46946",,,,2,ddaa12c269b49136a77d0f74c88f966c9257f8fce7c4a10c285e5cd8e98b7074
thread:issue:58111:issue:62998:issue:70865,issue #58111: Static graph training fails if forward is called multiple times before backward,needs triage,97,needs triage,3,0,"58111, 62998, 70865",,,,2,918b814bcb92cb1300bc4a4c4895df5cc512ec0ac6b2bbd6f0f0d47ae6df2054
thread:issue:64636:issue:64905:issue:65357,issue #64636: [docs or feature request] torchelastic: OOM recovery / skipping batches (e.g. if inf loss or nan gradients),stale/low urgency,97,stale/low urgency,3,0,"64636, 64905, 65357",,,,2,e7673238e5cb0ac6fa1f79b7d64c2ba710d63c86d9cb17fd99ae082104bd9b57
thread:issue:121219:issue:64345,issue #121219: Torch profiler corrupted names with Python 3.11,needs triage,97,needs triage,2,0,"64345, 121219",,,,1,6a5bd3bbc7da3f37f9e63ea463fcb1111ddb38970aef7d1c49363c2c68c4afad
thread:issue:128693:issue:128694,issue #128693: Segmentation fault (core dumped) in `torch.fused_moving_avg_obs_fake_quant`,needs triage,97,needs triage,2,0,"128693, 128694",,,,1,e719448a516feea56e8b7a5fcec416aa1569bf469e66b82f8d780e376103cbce
thread:issue:25888:issue:34329,issue #25888: Forward/backward hooks for C++ torch::nn modules,stale/low urgency,97,stale/low urgency,2,0,"25888, 34329",,,,1,850665525f5282766bea62eacac436dc3eeb947e26ba43873d4e13faf65c5b41
thread:issue:39351:issue:46564,issue #39351: Clean up RPC request callback implementation,needs triage,97,needs triage,2,0,"39351, 46564",,,,1,d03f8bef8e19d8ce6027ccc54dd5273f39c4dd5225fd2f3150633a081bd9e0ee
thread:issue:47260:issue:80832,issue #47260: DDP doesn't work with retain_graph = True,needs triage,97,needs triage,2,0,"47260, 80832",,,,1,13a8255173b5b05cb9a25491e3720e543fcf4a2a602785a666767eec9ed45a6e
thread:issue:55905:issue:56090,issue #55905: Attempt to use jited `torch.isnan` hit internal assert,needs triage,97,needs triage,2,0,"55905, 56090",,,,1,90490f9e3eacd344c6e01158fa7b750186b6a8a2d0195839f93b65e996898c74
thread:issue:100253,issue #100253: profiler.export_stacks doesn't return stack trace unless experimental_config is provided,needs triage,97,needs triage,1,0,100253,,,,0,24e426d878b3f4593312758e97394b44cf4b81198b93654082a810051825fc0d
thread:issue:103947,"issue #103947: [torch.fx] Deserialization Error - TypeError: ones() received an invalid combination of arguments - got (tuple, device=Attribute)",needs triage,97,needs triage,1,0,103947,,,,0,04e33174ec16cc891a2c22cd617dc1afa5fd55134edaf7a4a763d21e8260bdf1
thread:issue:113572,issue #113572: Deepcopying an exported model changes numerics for MobileNetV2,needs triage,97,needs triage,1,0,113572,,,,0,570b33428363341d922e7525f8402c20ed8274798612ad5c7e73c7a10ec242a3
thread:issue:115640,"issue #115640: Quantisation results in a ""Illegal instruction (core dumped)""",needs triage,97,needs triage,1,0,115640,,,,0,bd3f8d2a8e73d90cb734194d66bf7f166b3626a42d69cb2e63fe3e9fd53424b3
thread:issue:115818,issue #115818: Fx graph recompilation along with partitioner produces different node order even though graph operations are same,needs triage,97,needs triage,1,0,115818,,,,0,0361cca9edd5f327c5ddafa7f953efbecdbb8b0f13833405ae18d111542508b5
thread:issue:116351,issue #116351: `torch.quantize_per_channel`: FPE,needs triage,97,needs triage,1,0,116351,,,,0,d178fd882ac6d784e64a4845457110d2067a54191bd8af569a945a8dccf58f3e
thread:issue:122019,issue #122019: `F.max_pool3d`: Segfault due to incorrect `dilation` check in kernel,needs triage,97,needs triage,1,0,122019,,,,0,72fdf8b915566fdea870225a869fc8867288007198eb5196d627df4ac797759f
thread:issue:135835,issue #135835: A segmentation fault will be raised when using the `torch._C._jit_to_backend`,needs triage,97,needs triage,1,0,135835,,,,0,7c22a5e165ba3e4a61e61abecf81a5ebcb4b675cdd105ab8e234e347ddb71145
thread:issue:136712,issue #136712: Segmentation fault (core dumped) in `torch.ao.nn.quantized.dynamic.LSTMCell/GRUCell`,needs triage,97,needs triage,1,0,136712,,,,0,703164339c436c9a2725f6d3b84ca4e67cf086331736547b2e221dddc471a796
thread:issue:136728,issue #136728: Aborted (core dumped) in `torch.package.package_exporter.PackageExporter`/`torch.package.PackageExporter`,needs triage,97,needs triage,1,0,136728,,,,0,79db3c160b8ea1b3692eb1b7b9dbbfd697593020f74ecdf8fcc458902323bfd3
thread:issue:145088,issue #145088: [torchbench] torch._dynamo.exc.Unsupported: Graph break due to unsupported builtin None.morphologyEx,needs triage,97,needs triage,1,0,145088,,,,0,5b2a04af469f4c53ebfc0fdf6cc1e7a7c9f0c249993cd0e6677a3f2a252f213e
thread:issue:146788,issue #146788: Segmentation Fault in `torch.choose_qparams_optimized` with Invalid Parameters,needs triage,97,needs triage,1,0,146788,,,,0,bf27e34485f3ab641322bf236d34bb31f8ae0d202a5d355ad97efeb82c364377
thread:issue:149447,issue #149447: DTensor slicing on sharded dimension leads to replication,needs triage,97,needs triage,1,0,149447,,,,0,3e4719bf3f5447f09291973c7cd1e57a91578eb8c8e1858383e447fb81f4eac6
thread:issue:152108,issue #152108: [inductor][cpu] AMP static shape default wrapper AOTInductor performance regression in 2025_04_20 nightly release,needs triage,97,needs triage,1,0,152108,,,,0,c36cabe403c0625b3d90871dd0e34413c6c42038edbf8744ebddd0bf2d73ead7
thread:issue:153145,issue #153145: Segmentation faults with CPU FlexAttention,needs triage,97,needs triage,1,0,153145,,,,0,7012e2c2f9225f753cb9fc710c70197c2921d501c9667a1f73848e56b3424ff1
thread:issue:153260,issue #153260: Export shouldn't run TS under the hood.,needs triage,97,needs triage,1,0,153260,,,,0,6314de5cdf45f5581e32adb591e9e728b21a6c8ed05c6c5867684e04ffeca2cd
thread:issue:153941,issue #153941: TorchScript Fails Type Inference When Using .item()-Derived Scalar in Tensor Construction,needs triage,97,needs triage,1,0,153941,,,,0,8e8c7e2a729f19d97bb3d15cb07d4eecca19a6ba57d63cbae719909ce23e2dcf
thread:issue:155146,issue #155146: Segmentation fault when using torch.profiler,needs triage,97,needs triage,1,0,155146,,,,0,e6b0bf668eefa36bc3c21a255a9fb8dec55b7b79a17c136a4fc91e67ad61669a
thread:issue:155147,issue #155147: torch.profiler raises Aborted (core dumped) failurer related with GIL (gilstate_tss_set),needs triage,97,needs triage,1,0,155147,,,,0,29370817ee70f09dd87fa447a91308ae70bd47b1b23eca2da49655fa507dc9de
thread:issue:156563,issue #156563: Segmentation fault (core dumped) in `torch.profiler.profile`,needs triage,97,needs triage,1,0,156563,,,,0,f4fa10c7ea6dd52b3bd14f301972708263b70522a86773625e7edf64d467968e
thread:issue:159238,issue #159238: Performance regression: torch.jit.trace() significantly slower on RTX 5090 than RTX 4060 (cu128 nightly),needs triage,97,needs triage,1,0,159238,,,,0,11fe0b6529f9f3697de5782b974b7672504d2f3f049e64171b16c670ff9e59f5
thread:issue:164929,issue #164929: get_optimizer_state_dict modifies optimizer state,needs triage,97,needs triage,1,0,164929,,,,0,04caa4ec6c3d891cc319e3b08b6eecd4d70e6af9bb5bce41ed0cca7a2a22f690
thread:issue:28786,issue #28786: RPC couldn't match torch.ones with requires_grad=True,needs triage,97,needs triage,1,0,28786,,,,0,c0e2530849556dcd7b4c57316e1c73025078b2f3a41a16d6d228b2f6e14d0bf8
thread:issue:34116,issue #34116: autograd with TorchScript does not match finite differences,needs triage,97,needs triage,1,0,34116,,,,0,92a13e3bf7dc7a159bbb30d71077dedb27bbf5776b21d662e2f6ddd78d3ab374
thread:issue:35998,issue #35998: JITed GRU too slow,needs triage,97,needs triage,1,0,35998,,,,0,833fecfd7ecf59d099be85102a377a93e47c4870e2e14fdcfa9a947fca38b8c0
thread:issue:37015,issue #37015: Converting to Torch Script: cpp_module does not match nn_module,needs triage,97,needs triage,1,0,37015,,,,0,bed122c55d2f93b05b50d17f16e42c962282c4bb61a32ba6cbe674c460e2fa25
thread:issue:40867,issue #40867: [JIT] Infer type of argument foo = None to be Optional[Tensor],needs triage,97,needs triage,1,0,40867,,,,0,059e350595a7a960d833bcaaa84443bd7e7d7e0abb879825dff0c40c5f031c93
thread:issue:41289,issue #41289: List[Any] support in TorchScript,needs triage,97,needs triage,1,0,41289,,,,0,c03614e87334dcf94721f911d39140f10621ec9e604376124a989e3f07120fef
thread:issue:41695,issue #41695: Improve overload resolution order rules for `invokeOperatorFromPython`,needs triage,97,needs triage,1,0,41695,,,,0,2a19d7fc03058c245f83aeed614ff6de234d95079a3209af964b76c3ba8610db
thread:issue:42317,issue #42317: Instance norm annotation is incorrect,needs triage,97,needs triage,1,0,42317,,,,0,313dd0ef867fcdf8c2d1b4cef78ea0ff881d388882bb49b73e2e9d2df8c206bd
thread:issue:43548,issue #43548: JIT tracing incorrectly records some slice bounds,needs triage,97,needs triage,1,0,43548,,,,0,bb6712d7fa0c79fd9901019215ed571027f0bb831b1510ecb1b64755b60d8f80
thread:issue:45549,issue #45549: Can't initialize NCCL/GLOO process group if default process group is MPI,needs triage,97,needs triage,1,0,45549,,,,0,42a801f6b0c8cc3657e4161751db1887b9b5981dcdf197fa5c45b55a1afb89d7
thread:issue:46166,issue #46166: Error with DistributedDataParallel with specific model,needs triage,97,needs triage,1,0,46166,,,,0,602e9a3df84d683834f94797976809f4cf2f6d5f349def999faa2121e068002c
thread:issue:47218,issue #47218: [JIT] Alias Annotations for Native Schema List Ops Is Subtly Wrong,needs triage,97,needs triage,1,0,47218,,,,0,6692cf6ab457501f06288fe790409348efd851997157769fbfb8b7d3337ea261
thread:issue:47999,issue #47999: Torch DDP seems not prefer IPv6 over IPv4?,needs triage,97,needs triage,1,0,47999,,,,0,b4306555af7c136220638a042d34f2bd2c2e4ecd5ef974b4fcc3bc670835cbde
thread:issue:54524,issue #54524: Store created by dist.new_group doesn't appear to respect timeout,needs triage,97,needs triage,1,0,54524,,,,0,b7f161ed5d9c9e1a47098abff518a4da4b7ad9335c915ddfddc7b9fa25772451
thread:issue:62780,issue #62780: Seeing unit test failures for ProcessGroupShareTensorTest,needs triage,97,needs triage,1,0,62780,,,,0,2a7a5731a48bd0ddee97a4579e4138214e8fd0f344a9eb3a59f0bda50ac82592
thread:issue:63917,issue #63917: Add tests for DDP broadcast_buffers=True,needs triage,97,needs triage,1,0,63917,,,,0,d449d4822f995161a5c6a1bf5d509bcaccac8e73a3484a89404e226053ee3440
thread:issue:64518,issue #64518: TensorPipeDistAutogradTest is frequently failing,needs triage,97,needs triage,1,0,64518,,,,0,94a8ca3539e9f5c855cb6031b7b75d67fd599232a959fe3ba781667584f74fe4
thread:issue:66894,issue #66894: test_nccl_barrier_timeout_new_group_non_member fails intermittently,needs triage,97,needs triage,1,0,66894,,,,0,f3b7731b536939b458c691cf159faebd5385b2f317c225cdca89f7517761569b
thread:issue:67389,issue #67389: Missing doc for torch.distributed functions,needs triage,97,needs triage,1,0,67389,,,,0,72f8b1842f4f5121b0bc0f6b53268ce7d77a49c5b2842489ba7cab381dce88df
thread:issue:67541,issue #67541: TestBenchNetwork::test_forward[resnet50_jit-profiling-te] is failing intermittently,needs triage,97,needs triage,1,0,67541,,,,0,56f495254eed35d0bc53d9b05b577cd0551f729c54781b76e418b90c08eb8909
thread:issue:67646,issue #67646: test_jit_fuser_legacy and test_jit_fuser_te are failing with SIGIOT,needs triage,97,needs triage,1,0,67646,,,,0,96cbbdaeb65649a90d1c083340dba3ad11cccd53c37c3217699d51ceee055363
thread:issue:7036,issue #7036: [Caffe2] compilation linker error `libtbb.so.2: undefined reference to std::__exception_ptr`,needs triage,97,needs triage,1,0,7036,,,,0,96d1c3c0f401cc2637062c0e6e47fc397b171b70fa1f859039da2567f8a197de
thread:issue:71855,issue #71855: Regression in multi-node training speed with Transformers + PyTorch,needs triage,97,needs triage,1,0,71855,,,,0,ef49a857306ed4db3dac67beb199fc52dad176f340d16d623a2a04a97ce94225
thread:issue:72859,issue #72859: Profiler crashes with ProfilerActivity.CUDA on AWS p4d.24xlarge with A100-SXM4-40GB,needs triage,97,needs triage,1,0,72859,,,,0,f69f4854102ea1fb7f3cbc617ef0821150dcb24d9604088ba7e3aeaebe5a9efe
thread:issue:73352,issue #73352: Segfault on unloading a model,needs triage,97,needs triage,1,0,73352,,,,0,a0b693376c486ce118957389fbbb9261896a3f89864ee5683ec18e19691d3250
thread:issue:73602,issue #73602: Report DDP efficiency metric to users and guide them for setting up DDP correctly,needs triage,97,needs triage,1,0,73602,,,,0,084a42a0279d480eda7e66e838d4111263b6732093ed81de382804ae73a6c109
thread:issue:73885,issue #73885: operation not supported crash when initializing RPC tensorpipe,needs triage,97,needs triage,1,0,73885,,,,0,23ea14f2651154778d9e458155619e60249c66a830478fc8476ebd002b40c4e7
thread:issue:74115,issue #74115: DistributedDataParallel high peak memory usage with find_unused_parameters=True,needs triage,97,needs triage,1,0,74115,,,,0,a22359b2f48f2dfdc28e1863ed36cdc3c64d71eb0da88e0e23a4b0479dab68ec
thread:issue:80151,issue #80151: activation checkpointing with non_reentrant implementation memory leaks,needs triage,97,needs triage,1,0,80151,,,,0,a83cf845d0d02ed7fcee558ade382cc7ea53bb43e89d9451f81f4f897fc416d7
thread:issue:80349,issue #80349: Distributed Store `get` doesn't work well with `add`,needs triage,97,needs triage,1,0,80349,,,,0,4735ec777f3a7fba18be7a59b8dc7190565ec34f4d154ddc315aa7097cc73f19
thread:issue:81085,issue #81085: RuntimeError: required keyword attribute 'value' is undefined,needs triage,97,needs triage,1,0,81085,,,,0,09d8a3f76fc23904e5dcaa11900901f68698b19e9d7ea7767db2abff12a7de45
thread:issue:81115,issue #81115: torch.package can not be used to serialize `resnet18` from TorchVision-0.12,needs triage,97,needs triage,1,0,81115,,,,0,d3ed3f49c6563322fcd6a7aea9f129a9313fec59d341c331280075a3f0a486d5
thread:issue:83289,issue #83289: Improvements to ProcessGroupGloo monitored_barrier,needs triage,97,needs triage,1,0,83289,,,,0,a7754254bdc681b825d44b387852193a49f99669871d41bf6d5685f9d5dd71bc
thread:issue:83606,"issue #83606: test_profiler_experimental_tree_cuda_detailed is too unstable, and as its CUDA only difficult to regen",needs triage,97,needs triage,1,0,83606,,,,0,96c93d48e95c270883165c622ec2838735e6e79d30082c06bb2f6a2ba8d0bd51
thread:issue:86532,issue #86532: Conv2d will crash by using `jit.trace`,needs triage,97,needs triage,1,0,86532,,,,0,c52a339b8c4746358e4bf2e728fc685ae704ec4c03b335e0433fc7ef169a52d5
thread:issue:86594,issue #86594: JIT model with mean will crash when computing the gradients on cuda,needs triage,97,needs triage,1,0,86594,,,,0,b40bc56740fbd3cf9bcb81f8b1a56a22daaedd6efae2925275c6edea949d9bab
thread:issue:86595,issue #86595: JIT model with `relu+div+sgn` will crash when computing the gradient,needs triage,97,needs triage,1,0,86595,,,,0,d07724482307b178e880a367afa5ac8cab3097947c27ebf6ef85e6b380c70ef8
thread:issue:91165,issue #91165: [FSDP] FSDP with CPU offload consumes `1.65X` more GPU memory when training models with most of the params frozen,needs triage,97,needs triage,1,0,91165,,,,0,50b035434a669bc644bcdf799dcaea44a1eecf0fa132d98db3f10fc7e04b1248
thread:issue:178040:issue:183988:issue:186348:issue:187988,pr #181726: [xpu][1/4]Implement scaled_mm_v2 for MXFP8/MXFP4/NVFP4 on XPU,PR blocked,94,Review active PRs and linked issues as one maintainer work thread.,35,0,"178040, 183988, 186348, 187988, 188477, 188675, 188704, 188706, 188707, 188708, 188709, 188710, 188711, 188712, 188713, 188714, 188715, 188716, 188717, 188721, 188891","181726, 181727, 181728, 183511, 184276, 187315, 187318, 187989, 188268, 188931, 188996, 189149, 189186, 189297",,,69,8529652079345f40402b054597bf0cc5df8d03dc9ae10ecd7427bb8a384da412
thread:issue:178680:issue:188133:issue:93617:pr:184481,pr #184481: Guard synthetic-base input alias layouts,waiting on contributor,94,"Wait for contributor update on PR #184481, #184694; keep related issue/PR context attached.",6,0,"93617, 178680, 188133","184481, 184694, 185891",,,16,b710e9618ea2e351e6a90f2e55acf9b2d956f163c680d6328af74140fc82df76
thread:issue:16706:issue:185240:issue:186374:pr:185946,pr #185946: [CUDA] Warn instead of assert in `reportProcessMemoryInfo` to prevent crash on Tegra OOM,PR blocked,94,Review active PRs and linked issues as one maintainer work thread.,5,0,"16706, 185240, 186374","185946, 186375",,,7,555d0c020520f9120c35c29f289d91cddb99708e6d7e2fb2f36e9889b778271f
thread:issue:185470:issue:185484:pr:185846:pr:186358,pr #185846: [decomp] Match eager's device-dependent threshold comparison dtype,PR blocked,94,Review active PRs and linked issues as one maintainer work thread.,5,0,"185470, 185484","185846, 186358, 187609",,,5,43dab3f8ce14a67a6ab0c3dec62133c0b7f4e83b55b1603031dade481e6c1188
thread:issue:171356:issue:187429:pr:185756:pr:187908,pr #185756: [clamp] Fix float16 scalar overflow check inconsistency between CPU and GPU,waiting on contributor,94,Wait for contributor update on PR #187908; keep related issue/PR context attached.,4,0,"171356, 187429","185756, 187908",,,6,9fe83834785a4f8ad7c1d0cc7ba2f3c75f924bd9c8a37285bcb0235b529fffbd
thread:issue:173059:issue:189034:pr:189294:pr:189295,pr #189294: [CI] Realign scipy after nvmath install to fix H100 smoke NumPy ABI crash,PR blocked,94,Review active PRs and linked issues as one maintainer work thread.,4,0,"173059, 189034","189294, 189295",,,4,60c1a50acdee639d26d5d116f242eca37aee10fdd0ee037cec0990c330ea0937
thread:issue:182649:issue:182651:issue:182652:pr:184621,pr #184621: Fix Dynamo torch.Size from tensor shape inputs,waiting on contributor,94,Wait for contributor update on PR #184621; keep related issue/PR context attached.,4,0,"182649, 182651, 182652",184621,,,19,9c3ca9806cb7b6fd5d96ccc1a7ed2c37a47c2eb7ad16aa09470d9be0a5c42db9
thread:issue:187332:issue:188545:pr:187354:pr:188556,pr #187354: Match Inductor i0/i1 infinities to eager CUDA,ready for maintainer decision,94,Review active PRs and linked issues as one maintainer work thread.,4,0,"187332, 188545","187354, 188556",,,4,31fbe6dbe3eccfccb06f399be410aa598fd2019face4caf3506a87ab25a18204
thread:issue:188492:issue:188841:issue:189287:pr:188998,pr #188998: [inductor] Fix bf16 matmul accumulation precision in Triton templates (#188492),PR blocked,94,Review active PRs and linked issues as one maintainer work thread.,4,0,"188492, 188841, 189287",188998,,,4,852e79e5f04119c685f076bf1e5cb87e382b019ad5c11413d2164171a481bea4
thread:issue:40932:issue:41508:issue:55056:pr:189045,pr #189045: Fix NaN gradients in nn.MultiheadAttention for fully-masked rows (#41…,has linked issue,94,Review active PRs and linked issues as one maintainer work thread.,4,0,"40932, 41508, 55056",189045,,,6,2a3313d9ec22135e9602bad91e3a1e742c186e94db54d23af2de7b04d5ae5521
thread:issue:119883:issue:187027:pr:184392,pr #184392: Fix ModularIndexing printer semantics,PR blocked,94,Review active PRs and linked issues as one maintainer work thread.,3,0,"119883, 187027",184392,,,2,7ad6cf7d174f5c969a69097d92a866905f22e59a71a04f7a5f04d7f34dd96c49
thread:issue:139500:issue:139518:issue:176592,"issue #139500: Custom operators registered via decorator slower than ops registered via `torch.Library.{define, impl}`",has linked PR,94,has linked PR,3,0,"139500, 139518, 176592",,,,3,de8c0cee2e369b4660ccd8615e586b65d5a71e1ddea86bd6133e8080825f94ca
thread:issue:146719:issue:155044:pr:185496,pr #185496: Fix export autograd.grad saved tensor tracing,waiting on contributor,94,Wait for contributor update on PR #185496; keep related issue/PR context attached.,3,0,"146719, 155044",185496,,,15,69b05b9f5401ae7bb24e5a42f8e6db7f81f49b5fc1b639e0a3fa2055b5556af4
thread:issue:181696:issue:183120:pr:183654,pr #183654: Fix Inductor CPU group norm affine precision,PR blocked,94,Review active PRs and linked issues as one maintainer work thread.,3,0,"181696, 183120",183654,,,5,786aeceb1d3070954576a0b82878b0665841e595ff21034f746c16ddb4238922
thread:issue:182200:issue:184101:pr:184632,pr #184632: Fix CUDA fake strides for mixed dtype pointwise ops,waiting on contributor,94,Wait for contributor update on PR #184632; keep related issue/PR context attached.,3,0,"182200, 184101",184632,,,5,66c516e055e9172408b836c5bc6ab28496a61828b519a55401da30fab4d6abf9
thread:issue:182217:issue:183369:pr:184628,pr #184628: Fix SingletonInt static guard evaluation,PR blocked,94,Review active PRs and linked issues as one maintainer work thread.,3,0,"182217, 183369",184628,,,5,2c4ab41bf59687e624c898d54cdd83acf7ee30aa4f13f39a49c0c29c1e80f8ce
thread:issue:117826:issue:177116,issue #117826: [MPS] Probable incorrect gradient for einsum,stale/low urgency,94,stale/low urgency,2,0,"117826, 177116",,,,1,8b3d61f07db9e9283702d91c2cd0ebca7d8109c1cee48d1974517985815e1b97
thread:issue:120597:issue:121343,issue #120597: floating point exception on `torch.fmod` when x is large negative integer and y is `-1`,high-priority/blocker,94,high-priority/blocker,2,0,"120597, 121343",,,,1,218553e11f2f17d9b2a46f4f198700d1a2f370ff41a4eec7d95f925777a7716e
thread:issue:124863:pr:186454,pr #186454: Warn on unsupported custom op schema types,ready for maintainer decision,94,Review active PRs and linked issues as one maintainer work thread.,2,0,124863,186454,,,1,371dc5c831a958bfd5a4b6453691ba8cadca02d7c43d28b4b28da92a1e983a3a
thread:issue:146569:pr:184153,pr #184153: [Inductor] Clone cudagraph-owned grads before replay,waiting on contributor,94,Wait for contributor update on PR #184153; keep related issue/PR context attached.,2,0,146569,184153,,,1,56132d3745f1db98b01e1ae8fd0a04fa824357d16842192c67d5af7e5bc6548a
thread:issue:160018:pr:185349,pr #185349: Fix flex decoding BLOCK_M validation for short GQA,PR blocked,94,Review active PRs and linked issues as one maintainer work thread.,2,0,160018,185349,,,1,ccc66fe268cd5c64d6963dfc5c59ddae8dfd19efa2c5e591084668e35b33e829
thread:issue:162358:pr:181516,pr #181516: Fix heap-buffer-overflow in torch.linalg.eig on subnormal real inputs,PR blocked,94,Review active PRs and linked issues as one maintainer work thread.,2,0,162358,181516,,,1,1c66aced34e63137caf980f9c5485feb9d943f6f998ec49837447afc3b148c05
thread:issue:168126:pr:186622,pr #186622: Fix bf16 autocast precision barriers in Inductor,PR blocked,94,Review active PRs and linked issues as one maintainer work thread.,2,0,168126,186622,,,1,ce5fc75f9133313cbaa9feeb087ae995219616cdccfafaefc77b419018f88334
thread:issue:171894:pr:184970,pr #184970: Preserve pin_memory in generated view copy kernels,waiting on contributor,94,Wait for contributor update on PR #184970; keep related issue/PR context attached.,2,0,171894,184970,,,3,c8565332fd449e8d1fc632d45e26185b9cbceb99a251e12b2eb2a97f051ee9a6
thread:issue:175154:pr:184848,pr #184848: Fix nearest upsample scale shortcuts in compile,PR blocked,94,Review active PRs and linked issues as one maintainer work thread.,2,0,175154,184848,,,1,d420b03c8d5ccf3703bb34ef3500de9951d6bc56b3a92750c08d5a118ff00bb6
thread:issue:175190:pr:189233,pr #189233: [MPS] Fix avg_pool2d/adaptive_avg_pool2d backward on channels_last inputs,PR blocked,94,Review active PRs and linked issues as one maintainer work thread.,2,0,175190,189233,,,2,8c12f02ecdb8a58a9713d5dfcd689b9bbbe5e2b324e0fc85d4ee9a7d10527e3c
thread:issue:175370:pr:187104,pr #187104: Fix embedding_bag CPU segfault when offsets is empty and indices is not,has linked issue,94,Review active PRs and linked issues as one maintainer work thread.,2,0,175370,187104,,,2,90f7638e09e20c9fec1ce2701feaae9dd05a53bc14f06a49ee57ddb6f6014316
thread:issue:175978:pr:176070,pr #176070: [rnn] Prevent LSTMCell segfault on input/hidden shape mismatch,PR blocked,94,Review active PRs and linked issues as one maintainer work thread.,2,0,175978,176070,,,1,cee494394dbc89a0e8c69c8b2efda33438c4d5f4f7d67e02ba4d7baaa8939e41
thread:issue:177118:pr:183881,pr #183881: Fix online softmax epilogue fusion for T5,waiting on contributor,94,Wait for contributor update on PR #183881; keep related issue/PR context attached.,2,0,177118,183881,,,3,cd98af8f86cde31d8861f5ece0f15aeba3fe6b3333a8924b7a5af58d16310a0d
thread:issue:178089:pr:181036,pr #181036: Clamp spdiags diagonal length to zero for out-of-range offsets,ready for maintainer decision,94,Review active PRs and linked issues as one maintainer work thread.,2,0,178089,181036,,,4,4e5c06fe0f02b5f9969dea9469d761aba3480ac0940a789c8943f2b6d14e6009
thread:issue:181690:pr:183701,pr #183701: Fix CPU vectorized reduction tail stores,ready for maintainer decision,94,Review active PRs and linked issues as one maintainer work thread.,2,0,181690,183701,,,1,7d73d6f730062def7048979b8ec517f3be8ba06cdaad3a061d168aab9411c0cf
thread:issue:184408:pr:186252,pr #186252: [inductor] Fix truncdiv off-by-one on CUDA with _div_rn,PR blocked,94,Review active PRs and linked issues as one maintainer work thread.,2,0,184408,186252,,,2,1d16ea6087d7bb23b47b3a074c0b5b8fb74cc84c100805daa99fae84559f4c78
thread:issue:186084:pr:186395,pr #186395: Preserve overlap deps without blocking fusion,PR blocked,94,Review active PRs and linked issues as one maintainer work thread.,2,0,186084,186395,,,3,1276070ab442c599ebf04f532afe89a617f6fb7de03da7fd2c7fbab3bf1cf187
thread:issue:186241:issue:187175,issue #186241: TMA descriptor loads use stale `xoffset` in mix-order reduction loop,has linked PR,94,has linked PR,2,0,"186241, 187175",,,,1,5b63821c379488a17e9af6a2bbe7519eac89eb0c83e74d91b49e0e8433a111c8
thread:issue:186256:pr:176234,pr #176234: [autograd] Explicitly error when layer_norm computes 3rd order derivatives,PR blocked,94,Review active PRs and linked issues as one maintainer work thread.,2,0,186256,176234,,,2,54b0c960771a46f704db30a656e25035d96d412b49a9e7c84fbfd1789f8aa0ae
thread:issue:186537:pr:186583,pr #186583: Propagate contextvars to Python autograd functions,ready for maintainer decision,94,Review active PRs and linked issues as one maintainer work thread.,2,0,186537,186583,,,4,31c9d84a858768e3a6a23752fffd4dc74d63c5c4423ea6bf089545bc3e846301
thread:issue:187284:pr:187311,pr #187311: Preserve forward AD tangents for Inductor dual inputs,PR blocked,94,Review active PRs and linked issues as one maintainer work thread.,2,0,187284,187311,,,1,03f7ac23426dca121533659527eb5158c003619a49e51178c15734f4b2684671
thread:issue:187340:pr:187493,pr #187493: Fix Inductor nextafter for low precision CUDA tensors,PR blocked,94,Review active PRs and linked issues as one maintainer work thread.,2,0,187340,187493,,,1,a43dfb6757d01fa4fe64559118cb3714770144aa1fb5abd8a67f1d0f02efbdf2
thread:issue:187415:pr:187486,pr #187486: Fix unbacked memo reuse during fake replay,PR blocked,94,Review active PRs and linked issues as one maintainer work thread.,2,0,187415,187486,,,1,4ee30e18baf72b0ab2b407cfda112e932651daae3db04eed31528a671d7c996a
thread:issue:188008:pr:187955,pr #187955: Make expandable-segment Handle a move-only RAII owner and exception-safe,waiting on contributor,94,Wait for contributor update on PR #187955; keep related issue/PR context attached.,2,0,188008,187955,,,2,775b3504eed2e78491b93491b8404ca097660d110d0f69b975d2a6610daaa81f
thread:issue:188023:pr:188047,pr #188047: fix(dlpack): raise Python RuntimeError instead of aborting for negative DLPack strides,PR blocked,94,Review active PRs and linked issues as one maintainer work thread.,2,0,188023,188047,,,3,8172be8917e10294e23de76f6d690c363ec0690f56d1e46f673ca15ce7a0fc69
thread:issue:188034:pr:188039,pr #188039: autograd: fix PyObject* refcount leak in setup_context path,has linked issue,94,Review active PRs and linked issues as one maintainer work thread.,2,0,188034,188039,,,2,dd98fd699320992d8ecd6839d2c1bdf1b01c765ee2848808699e0c1210dac07f
thread:issue:188113:pr:188114,pr #188114: Fix CK SDPA flash attention on gfx1200/gfx1201 (RDNA4),ready for maintainer decision,94,Review active PRs and linked issues as one maintainer work thread.,2,0,188113,188114,,,5,123c5ab756e710a426e1c8be6868a782e6648af4d705b92932aae7cc64502bc1
thread:issue:188541:pr:188546,pr #188546: [Inductor] Fix log_sigmoid_forward flushing float32 subnormals to +0.0 on CUDA,PR blocked,94,Review active PRs and linked issues as one maintainer work thread.,2,0,188541,188546,,,2,ffdb88809c0422c36e8ed1719cdcd331903be6631f11564241ec2ec6dbe0f3b5
thread:issue:188547:pr:188605,pr #188605: Fix infinite recursion with pybind11 enum conversions,waiting on contributor,94,Wait for contributor update on PR #188605; keep related issue/PR context attached.,2,0,188547,188605,,,3,b3c0b7eed7e09720a4f1f8a94b2609db1e3738d42edd17afadb7fde836a54912
thread:issue:188792:pr:188791,pr #188791: Fix MKL VML first-call dispatch race producing silent low-precision results,has linked issue,94,Review active PRs and linked issues as one maintainer work thread.,2,0,188792,188791,,,1,62a0aa9e97f079820b6c41acd0bf11fc75b782cb284bcdbe18f13ebde5b12839
thread:issue:188890:pr:189129,pr #189129: inductor: Fix non-deterministic gradients in slice_scatter backward fusion,PR blocked,94,Review active PRs and linked issues as one maintainer work thread.,2,0,188890,189129,,,2,670d2bfd67fddbcfacc5a9d4195dae17a1cc8b97a277388cdb0054eab525eaa8
thread:issue:189267:pr:189273,pr #189273: Fix nn.functional.threshold OpInfo reference for strict NumPy uint8 casting,draft/noise,94,Review active PRs and linked issues as one maintainer work thread.,2,0,189267,189273,,,3,73e032606153c24d3563049f0b986b00d5d01efb68833d87552ae10fa21ac27a
thread:issue:142454,issue #142454: Segmentation fault (core dumped) in `torch.max_pool1d`,has linked PR,94,has linked PR,1,0,142454,,,,0,87869d538bd79fcc6323525628bec8abf8671d4ddf06a761e02a02ecde637374
thread:issue:145064,issue #145064: `_pdist_forward` causes segmentation fault for 3D tensor with last dimension of size 0,has linked PR,94,has linked PR,1,0,145064,,,,0,1a28183645fe3e0c0adea6c8d03cafa1afb1ce1f52cff9a9c00382eed1c0a8d9
thread:issue:148317,issue #148317: compile SageAttention faing error C2872: “std” for latest torch nightly,has linked PR,94,has linked PR,1,0,148317,,,,0,29268a10ead72f718522f24b7adad87e0cdd572791f19439267fb0f6c59aa97e
thread:issue:153044,issue #153044: Unexpected float32 overflow for amp training with torch.compile,has linked PR,94,has linked PR,1,0,153044,,,,0,da657fe5de35b4a12710972929f9494442ca58580c4af837830f746f8acff016
thread:issue:161783,issue #161783: torch.compile has runtime overhead on small graphs,has linked PR,94,has linked PR,1,0,161783,,,,0,ef387738157185d9ddc4a77241df845ae656895fb5baf8ce7a450ac8d037cdd3
thread:issue:171838,issue #171838: `torch.nn.functional.pixel_shuffle` throws SIGFPE (Floating Point Exception) with large upscale_factor,has linked PR,94,has linked PR,1,0,171838,,,,0,eb1ba9f6e40d5848537ccb98af0d173e7eeb0ea2fe2d6f5e46fe9f40c838296e
thread:issue:173318,issue #173318: [docs] `torch.cuda.Stream` link broken on 2.10+,has linked PR,94,has linked PR,1,0,173318,,,,0,724fffdea23de3567d804c6f629d472f243b22a4d3579a5e0fcd705abc5cd549
thread:issue:173946,issue #173946: Segmentation fault in aten.lstm.data results if hx shapes are invalid (when running on CPU),has linked PR,94,has linked PR,1,0,173946,,,,0,d481979106859c08421a5d3a4bf7e55b90c5b1601cd8829cecaf14ab2913b29d
thread:issue:180740,issue #180740: [Inductor] functorch_dp_cifar10 accuracy regression in dynamic_inductor_torchbench,has linked PR,94,has linked PR,1,0,180740,,,,0,a8ced3a752329d4cbeb3ea4abe89b68b8980ffa4dab56348416cee6602afe5da
thread:issue:181510,"issue #181510: torch.nn.utils.weight_norm triggers a UBSan null pointer error for nn.Linear(0, 1) on CPU",has linked PR,94,has linked PR,1,0,181510,,,,0,66fe0e6cb0d18a8c729d07ee2585fb52f91bca52b469066e65eddb92414643db
thread:issue:182283,issue #182283: INTERNAL ASSERT FAILED at /pytorch/c10/core/Scalar.cpp:42,has linked PR,94,has linked PR,1,0,182283,,,,0,bc2628dbfd9623864e526335440330c8b0729038126edb156e274d696c3e4d74
thread:issue:183806,issue #183806: `torch.linalg.eig` and `torch.linalg.eigh` 30-130x slower on 2.11.0+cu130 vs 2.5.1+cu124,has linked PR,94,has linked PR,1,0,183806,,,,0,29cb3c17801f2e2ce8c95b364226a65b50145d45fb9e7d739055d252dd0e171a
thread:issue:184032,issue #184032: torch.UntypedStorage.<from_file/torch.load(weights_only=True)/_new_shared_cuda>.new() segfaults via null Allocator dereference at c10/core/StorageImpl.h:84,has linked PR,94,has linked PR,1,0,184032,,,,0,761485d94230b93c12a649ac9948dba9f064e8358a6230ca1615c21f4f7a2c02
thread:issue:184876,issue #184876: SIGSEGV when assigning to nn.Parameter.data inside a torch.func transform,has linked PR,94,has linked PR,1,0,184876,,,,0,8bc6ce487e6d1a92d9d62be648632d803dcb53a7e4afe257ae733c3f7d290435
thread:issue:185247,issue #185247: Segmentation fault in `torch.linalg.eigvals` triggered by invalid `avg_pool3d` divisor,has linked PR,94,has linked PR,1,0,185247,,,,0,df34b27dab96187c3353edd780b3d2ef50b2e8143073ef7bc808d17d2e9b6312
thread:issue:185482,issue #185482: [inductor]`torch.cond` Branch with sliced view and scatter_add crashes with non-tensor input,has linked PR,94,has linked PR,1,0,185482,,,,0,9bf262fbdd3c7d724b643b91b6b94d0048eb6dbc0196ff279d10590ba96eb028
thread:issue:185715,issue #185715: INTERNAL ASSERT FAILED: NYI SymInt equality in c10/core/Scalar.h when using torch.compile(dynamic=True) and torch.autograd.grad with torch.pow,has linked PR,94,has linked PR,1,0,185715,,,,0,0c45bafbfaaffee24732acb69221b74ecf0d7cceb9c698d8b5302352f2963cc1
thread:issue:185886,issue #185886: [torch.compile] Clarification/regression: Dynamo cache lookup guard eval is much slower in 2.12 vs 2.7 after `inline_inbuilt_nn_modules` became always True,has linked PR,94,has linked PR,1,0,185886,,,,0,052cf1a0f75c147421e8aba6da3307375166ad708c3a76addbd8eadcb70e1966
thread:issue:186233,issue #186233: C++ API: at::range crashes due to division-by-zero in RangeFactories.cpp,has linked PR,94,has linked PR,1,0,186233,,,,0,6f425b0322041f6ccdae8eed020605d32b951b39b459edce82cc16d8a1865094
thread:issue:186453,"issue #186453: INTERNAL ASSERT FAILED at ""/pytorch/c10/core/AllocatorConfig.h"":124 on python torch import",has linked PR,94,has linked PR,1,0,186453,,,,0,7ef070aa4c007a4c2fc32636384f63206754f4b701a4d917d5b6d98e190ce322
thread:issue:187399,issue #187399: [MPS] torch.baddbmm — INTERNAL ASSERT for size-0 batch dimensions,has linked PR,94,has linked PR,1,0,187399,,,,0,80dab505c956f62d6dd0e7f2174e776d66969d9c23a955de6fef50e39ab609bc
thread:issue:187567,issue #187567: [MPS] torch.linalg.qr is 25-250× slower than CPU on tall-skinny matrices,has linked PR,94,has linked PR,1,0,187567,,,,0,61f6199b0bef0e6937fdda0104fbdb0fecd960c1dffa48be2475afe4ad66f798
thread:issue:188033,issue #188033: [BUG] THPStream_richcompare casts arbitrary objects to THPStream* without type check,has linked PR,94,has linked PR,1,0,188033,,,,0,1d67a797c7ff3320e4eb2b1c1f798244a614800d470aa1dc54072f8c8f0c9aaf
thread:issue:188069,"issue #188069: torch.compile(mode=""max-autotune"") crashes with too many positional arguments during bmm autotuning when template inputs are deduplicated",has linked PR,94,has linked PR,1,0,188069,,,,0,887753eb09995d232c2905a97e9694948e6cd80b7d46e34e90cb1255210a47a1
thread:issue:188198,issue #188198: [MPS] cat op dropped performance in torch==2.10 release,has linked PR,94,has linked PR,1,0,188198,,,,0,4d89099eee6036e02f6ad84b3a74fab5c94f77b58803d324bc15fa7ed3d96317
thread:issue:188231,issue #188231: [inductor] `TypeError: A Boolean argument can only be used in Eq and Ne` during CPP codegen,has linked PR,94,has linked PR,1,0,188231,,,,0,34b7bbfd0042483cd8bc5dafae7ea29355c3723d019c9a980a81cfedd836cd7d
thread:issue:188261,issue #188261: [inductor][XPU] PR #184366 causes AMP fp16 training accuracy regression on XPU,has linked PR,94,has linked PR,1,0,188261,,,,0,7ef1b524e317d2e39c7f998998772bd77345a0d588627d7f1cc10a8364f69034
thread:issue:188358,issue #188358: [Dynamo] Eager and torch.compile behavior mismatch in F.batch_norm with multidimensional bias,has linked PR,94,has linked PR,1,0,188358,,,,0,d654cd5513886f113e59b8e867272473c22b812efedd2b59881a2c7cdc2a8f5c
thread:issue:188578,issue #188578: [Inductor] Optimize split reduction heuristics on GB200,has linked PR,94,has linked PR,1,0,188578,,,,0,5a5d16e280d6597cd800847168bd129f904208286120321f283ee430997a4993
thread:issue:65761,issue #65761: torch.utils._pytree -> stable,has linked PR,94,has linked PR,1,0,65761,,,,0,7e1e9bcc216331af7440b225ccdd06271c86a010a68b7dadde65858efe7f2fe6
thread:issue:124502:issue:170127:issue:174608:pr:184983,pr #184983: Make SDPA backend checks safe for unbacked head dims,ready for maintainer decision,89,Review active PRs and linked issues as one maintainer work thread.,4,0,"124502, 170127, 174608",184983,,,3,d2ebe4994cb16343ef816fd253b1ee84d017b884b2a4de5c5a22ed56d8365024
thread:issue:175467:issue:184229:pr:184508,pr #184508: Fix export with unused DTensor param attrs,PR blocked,89,Review active PRs and linked issues as one maintainer work thread.,3,0,"175467, 184229",184508,,,3,c78cdeee7f049c5385f62c0443ffc9a7eebecc7caa71e0fc2709a746950fee2e
thread:issue:143157:pr:186043,pr #186043: Lower symbolic bool not to torch.sym_not,PR blocked,89,Review active PRs and linked issues as one maintainer work thread.,2,0,143157,186043,,,1,e8f0b773a67d1240542ecff9c415a69c4e0bb64410a8be701a0b2c357717a230
thread:issue:147170:pr:188965,pr #188965: Fix non-deterministic node ordering in get_source_partitions(),PR blocked,89,Review active PRs and linked issues as one maintainer work thread.,2,0,147170,188965,,,3,381df906b1c919d96321e8ab4da0bf149ea2ce1ebf5838c4e5bb22dce2b78d27
thread:issue:147326:pr:185804,pr #185804: Fix non-strict export tensor len handling,PR blocked,89,Review active PRs and linked issues as one maintainer work thread.,2,0,147326,185804,,,1,8333b48a734fe77c8e8f71cbbf74617b4bde0a844333d9afce3776cd23dbcc98
thread:issue:160399:pr:185324,pr #185324: Reject mixed-device fake convolution inputs,PR blocked,89,Review active PRs and linked issues as one maintainer work thread.,2,0,160399,185324,,,3,9c23c859f97739578af35606df2a93e88cea6daaf779d3c1ab58e091c2d602dc
thread:issue:162700:issue:184996,"issue #162700: RuntimeError: The tensor has a non-zero number of elements, but its data is not allocated yet during torch.backward with InternVL",stale/low urgency,89,stale/low urgency,2,0,"162700, 184996",,,,1,55bfbfb0bfc0bb8d874b40425508759112e34a3910ed5c41074f7f82de0acba8
thread:issue:163189:pr:185108,pr #185108: Prevent non-strict export from leaking fake buffers,PR blocked,89,Review active PRs and linked issues as one maintainer work thread.,2,0,163189,185108,,,1,a757129552f2befc26ae1ecdc06d4ac3e2b31eb3617203cef8eb00d7f46ff087
thread:issue:166093:pr:185069,pr #185069: Fix AOTI Windows cross-compile package loading,waiting on contributor,89,Wait for contributor update on PR #185069; keep related issue/PR context attached.,2,0,166093,185069,,,5,faaf3de1f2c83708c68fd9732116e326b1580de1252ff63d9f464b3c698cc3eb
thread:issue:171897:pr:184967,pr #184967: Fix export CIA decomps with backend kernels,PR blocked,89,Review active PRs and linked issues as one maintainer work thread.,2,0,171897,184967,,,1,1c8628bfd80823493cb171b187f3c56ee119340a7fab85f457912b9204a236b2
thread:issue:171964:pr:184966,pr #184966: Preserve unflattened module forward signatures,ready for maintainer decision,89,Review active PRs and linked issues as one maintainer work thread.,2,0,171964,184966,,,1,cf300fc534bda162ad4d35c312a3005b8a761dfd75d6ca85f04d94649f868233
thread:issue:171974:pr:184964,pr #184964: Handle tensor scalar slice bounds in export setitem,PR blocked,89,Review active PRs and linked issues as one maintainer work thread.,2,0,171974,184964,,,1,92bca56b429a2dca2590e4588c2d8f92aff8b4e8361d8917ef9996ec4f303759
thread:issue:171977:pr:184963,pr #184963: Fix FakeTensor shallow copy state,waiting on contributor,89,Wait for contributor update on PR #184963; keep related issue/PR context attached.,2,0,171977,184963,,,5,096b0cf64fe4d06f93b117f3d687850c74e80d98128b2031ff297cf538462133
thread:issue:171978:pr:184961,pr #184961: Fix packed RNN export fake metadata,waiting on contributor,89,Wait for contributor update on PR #184961; keep related issue/PR context attached.,2,0,171978,184961,,,1,894883e7b9c1f53b107afd78d452a8dbb86c49ab41d937469412f66a998975d6
thread:issue:172711:pr:184950,pr #184950: Fix mkldnn transposed conv pointwise shape inference,PR blocked,89,Review active PRs and linked issues as one maintainer work thread.,2,0,172711,184950,,,2,1011cc4861c10b0a3d39c1793d42055876bc60eabbdeac61541dade698c513df
thread:issue:172839:pr:184948,pr #184948: Fix dynamo export for tensor closures in output pytrees,ready for maintainer decision,89,Review active PRs and linked issues as one maintainer work thread.,2,0,172839,184948,,,1,f41f77bb33234d4f02d94cd06e35bcec107ec7799e9f5131fa55500205708681
thread:issue:176428:pr:184787,pr #184787: [export] Deserialize PT2 schema in C++ during load,ready for maintainer decision,89,Review active PRs and linked issues as one maintainer work thread.,2,0,176428,184787,,,1,3b38775bdb4720ea7acf9f25906743dd4940f8d2fb185a208a6f295fa8455745
thread:issue:178522:pr:182224,pr #182224: Enable skipped static_runtime tests on AArch64,PR blocked,89,Review active PRs and linked issues as one maintainer work thread.,2,0,178522,182224,,,1,5fd821c8dc2c6fd8268104f33acc5da3184c5bafad81b6ccf25b1f925d6eb94b
thread:issue:178798:pr:178844,pr #178844: [c10d] Support stacked output tensors in Gloo all_gather_into_tensor …,has linked issue,89,Review active PRs and linked issues as one maintainer work thread.,2,0,178798,178844,,,2,dda4591dac838a167781a8a603f13115a63f54f7c3756b03d1c862261bdfbcfb
thread:issue:180354:pr:184667,pr #184667: Fix cond branch tensor constants in run_decompositions,ready for maintainer decision,89,Review active PRs and linked issues as one maintainer work thread.,2,0,180354,184667,,,5,8e1a69205d41b8130293e60bcdcd376b9d049be22e49766964219e19d0e72e39
thread:issue:184841:pr:185892,pr #185892: Improve export device mismatch errors,waiting on contributor,89,Wait for contributor update on PR #185892; keep related issue/PR context attached.,2,0,184841,185892,,,10,2aedef6fdac3ec5d9a2d72605395d46147e11a7f6ae60e91afbd4e88e525ddca
thread:issue:187480:pr:187481,pr #187481: Add overflow check to FBGEMM version of quantize_val_arm,ready for maintainer decision,89,Review active PRs and linked issues as one maintainer work thread.,2,0,187480,187481,,,2,4b348ee634bb1845baded4942e19f1e41866b224e687643e01a7daba206f439c
thread:issue:188772:pr:188698,pr #188698: Make Fsspec filesystem checkpointing public and exposing cache_staged_state_dict to enable async save,ready for maintainer decision,89,Review active PRs and linked issues as one maintainer work thread.,2,0,188772,188698,,,5,fb35ec9b0343dff805e1d088ee08527591bf0d0e8684e1e7a352647b9d65a9c7
thread:issue:189121:pr:189124,pr #189124: [inductor] Cache DSR-expanded single-config reductions,has linked issue,89,Review active PRs and linked issues as one maintainer work thread.,2,0,189121,189124,,,4,dcf7805c2920911ea42c05d34fbc2c66f3beafd2889233d7ed019323675b02bb
thread:issue:130861,issue #130861: torch.fx.Tracer.record_stack_traces is broken in torch 2.4.0,needs triage,89,needs triage,1,0,130861,,,,0,7d7d76cd5266aeb23d68d8e7027cafc5c2d1a656c786bd50829d3c1ecf9f6f9a
thread:issue:143580,"issue #143580: RuntimeError: expect_autograd_hooks_ INTERNAL ASSERT FAILED at ""../torch/csrc/distributed/c10d/reducer.cpp""",needs triage,89,needs triage,1,0,143580,,,,0,c6afcb3763ce463d53ff2a638875aabf3320033b9444bb69a46786a988a59f06
thread:issue:153614,issue #153614: Add opentelemetry traces,needs triage,89,needs triage,1,0,153614,,,,0,18429cf241d490c07dac6ae27fff7c40634d93724441271ffb55c8e692958a0b
thread:issue:171870,issue #171870: [s390x/ZVECTOR] C++ compilation failure in argmax/argmin operations with torch.compile inductor backend,needs triage,89,needs triage,1,0,171870,,,,0,be6eff83836ddd34eed3b274510b80b8bc382dd5d4d5c74c8a60f0eb55bf10bd
thread:issue:172930,issue #172930: torch.export failed while exporting alpamayo model (Qwen3_vl) with symbolic error,needs triage,89,needs triage,1,0,172930,,,,0,dd44767b2a0e4c26310cb630ef4abdb36ae1668c6141e67a5f57d5ebda20356e
thread:issue:173495,issue #173495: `torch.fbgemm_linear_quantize_weight` crashes with segmentation fault (SIGSEGV) with a cuda tensor,needs triage,89,needs triage,1,0,173495,,,,0,93b1e9276928d60965c1387d2f8337c11c2b09a8cc06c49cdee8546780d9dd5a
thread:issue:173643,"issue #173643: Typo ""emitDefautlArgsWithOutArgs"" in torch/csrc/jit/serialization/export_module.cpp",needs triage,89,needs triage,1,0,173643,,,,0,b2cea86e6abcfdbfcd69b6ee50abadf136e44b55368d4308a081bc68bbc9b1ad
thread:issue:177510,issue #177510: [export] re-raising augmented exceptions resets traceback to except block,needs triage,89,needs triage,1,0,177510,,,,0,797ee69a3aa8a0687c0e09e24739091876a4c55c6484f127546d284e517d1c2c
thread:issue:178473,issue #178473: ProcessGroupNCCL scalable communicator init may reuse stale UniqueNCCLID store keys,needs triage,89,needs triage,1,0,178473,,,,0,5d0dae310cc7c6b68438c9d502e33e44d81e2e29eb5bc3545747a1728cc2995b
thread:issue:178618,issue #178618: torch.export in PyTorch 2.10 emits INT64_MAX instead of None for aten.slice.Tensor end bounds,needs triage,89,needs triage,1,0,178618,,,,0,e719105f8039ec249b2dae9264c7d9e48e475ee84956c972c363be74a9dfe992
thread:issue:181061,issue #181061: [RFC] M-to-N cast primitive (for RL weight rollout & pipeline parallel),needs triage,89,needs triage,1,0,181061,,,,0,eb10c6ecf4495ab0b4be3b24ffb2f4032b0291b2e8eee151a91a5f01f33ee9ab
thread:issue:181238,"issue #181238: torch.cuda.synchronize() syncs on device 0 when current device is not explicitly set, causing incomplete profiling results",needs triage,89,needs triage,1,0,181238,,,,0,8c839269b7643939d053cb02b87b911dbc53376766eae35756052f51896523bb
thread:issue:182031,issue #182031: DTensor.local_map inconsistently handles pytree on input versus placements,needs triage,89,needs triage,1,0,182031,,,,0,91863b457e078710023c1b9c646be70341c2137b5b726126199d187d12dd29e8
thread:issue:182102,issue #182102: DeviceMesh pickling warning spam in DCP process,needs triage,89,needs triage,1,0,182102,,,,0,94b20f11e3c6514a9eb769ba7bd404117a61c349a9cc7c75196216f0213a4f57
thread:issue:182915,issue #182915: Add NVFP4-aware fused scaled_mm_v2 + reduce_scatter support in symmetric memory,needs triage,89,needs triage,1,0,182915,,,,0,5959e07162f1131bf95d49b3901b13c10e79e2cae51c2618dd28b0acb37ee82d
thread:issue:183207,issue #183207: FSDP2 with ignore_params does not work with foreach optimizers,needs triage,89,needs triage,1,0,183207,,,,0,5d06dc2d01f598499a96619d7440e687d9de91bef100c422806147cbc70625da
thread:issue:183520,issue #183520: Cannot JIT a deprecated class,needs triage,89,needs triage,1,0,183520,,,,0,56982c4290b45d25534116d288cfe3a3d29c83c5246ef5329f2a8347ab72dc29
thread:issue:185167,"issue #185167: DTensor: `F.nll_loss(weight=..., reduction='mean')` over a sharded batch returns wrong result",needs triage,89,needs triage,1,0,185167,,,,0,9b34380e5292069d79a2bc5a691a55b9b3571389aeb03309f1cfb0b6ee3238d3
thread:issue:185441,issue #185441: [JIT] Null pointer write in SourceRangeDeserializer::deserialize_source() via unchecked text_table indices,needs triage,89,needs triage,1,0,185441,,,,0,e293333b9e558669e69bb5f9290453e98a10874ca9f55e8f5bbada5f93f85c63
thread:issue:186938,"issue #186938: ShardedTensor.device hardcodes torch.cuda.current_device(), breaks FSDP checkpoint resume on XPU/non-CUDA builds",needs triage,89,needs triage,1,0,186938,,,,0,149e4e9f7dcd02f2be2d9c42a57049a28fd8a9dc7b949453a828f02e3fa22671
thread:issue:187048,issue #187048: [Bug][Quantization] Inconsistent numerical behavior in _choose_qparams_per_tensor and redundant code in test_quantized_tensor.py,needs triage,89,needs triage,1,0,187048,,,,0,28526dfaac2be2be35debb8edce2c095412baa9ec9e879603e72c9b2d8744784
thread:issue:187138,"issue #187138: [DCP] load silently casts checkpoint tensors to the template dtype, truncating precision without any warning",needs triage,89,needs triage,1,0,187138,,,,0,a9037f3c3857838597493d603c25e7e6040ce7db20a6394f8e0d51e3909b3791
thread:issue:188193,issue #188193: torch::cuda::nccl::gather/scatter currently implemented as O(world_size) p2p connections on root,needs triage,89,needs triage,1,0,188193,,,,0,5fc5710a0dc11455b27643b696f30d4bf9ec9d67193a5ecf219267f1985787ea
thread:issue:170039,issue #170039: PyTorch crashes with std::__1::system_error: mutex lock failed: Invalid argument,high-priority/blocker,88,high-priority/blocker,1,0,170039,,,,0,757a906fc6d80c745ee453459bc73d8a084bcc18186dcccb74031fe25a078fb5
thread:issue:161053:pr:185300,pr #185300: Fix AttrProxy module reuse in Dynamo,PR blocked,83,Review active PRs and linked issues as one maintainer work thread.,2,0,161053,185300,,,1,e026393d88f25aa28f23599a0657f4fbb0eeccc7d8959efa1de634471d2e9669
thread:issue:167872:pr:185025,pr #185025: [export][serde] Initialize symbol ranges before constants,ready for maintainer decision,83,Review active PRs and linked issues as one maintainer work thread.,2,0,167872,185025,,,1,856012468f0cfb3545f98f9aa548f501695c77ecd284ee04b3001b79580a6f11
thread:issue:131941,issue #131941: [export] Schematize nn_module_stack serialization,needs triage,83,needs triage,1,0,131941,,,,0,9cb0ab44ed9d398d7d08b23bf1b56184999f0a091dac590e133a0a4e29d2ff43
thread:issue:177730,issue #177730: `nanosleep()` call in `socket.cpp` delay method cannot handle signal interrupts (e.g. from sampling-based profilers),needs triage,83,needs triage,1,0,177730,,,,0,b6eb375f0cdf21f88a449c25cfc2dd05384bacc556c823f21324e08f4c4ed406
thread:issue:51856,issue #51856: Assignment target is transposed when using jit.script and avanced indexing,needs triage,83,needs triage,1,0,51856,,,,0,cbfa2f26b6263edc3fd2ca5f69d8daa182eba566b4d32699c7888078d8feb9a2
thread:issue:113180,issue #113180: Higher train loss and worse evaluation metrics when using `torch.compile()`,high-priority/blocker,82,high-priority/blocker,1,0,113180,,,,0,e95a60ad5c97f253fa783ff6caa13b6e40a9cfe4ed5dab4684413401a8ef662f
thread:issue:121101,issue #121101: Segmentation error for torch==2.2.1 on MacOs,high-priority/blocker,82,high-priority/blocker,1,0,121101,,,,0,93c3455866fb54671be8256a0f9b7bcafbaebbdb4b992f14cb0d6dbba813e9c9
thread:issue:169823,issue #169823: Abnormal CPU usage with ComfyUI,high-priority/blocker,82,high-priority/blocker,1,0,169823,,,,0,04c3945a11a6dc9b805c33e80e8d53b81390808c79e9a3b7ab94de747673dd45
thread:issue:171606,issue #171606: cos CPU ops deterministic not ensured,high-priority/blocker,82,high-priority/blocker,1,0,171606,,,,0,4833354b2daeab2f6a65721820c96967327fce3596ffd3cdad001b289ee0228c
thread:issue:174468,issue #174468: Memory leak by torch.compile,high-priority/blocker,82,high-priority/blocker,1,0,174468,,,,0,bdfbd20ae71ec93951d7ef39a6cf4a91627753ce7cee457e6f5023aecba1cd78
thread:issue:175614,"issue #175614: Segfault during device_map=""auto"" weight materialization on RTX 5090 D v2 (GB202-240, sm_120)",high-priority/blocker,82,high-priority/blocker,1,0,175614,,,,0,184a05e05d4d0b15dd7f4bac65912d1cb79f837e4ca30e93e56786d66e0301ad
thread:issue:176426,issue #176426: Triton kernels with 2 or more tl.load() calls segfault at runtime on sm_120 (NVIDIA RTX PRO 6000 Blackwell). The kernels compile without errors but produce invalid code that crashes on execution.,high-priority/blocker,82,high-priority/blocker,1,0,176426,,,,0,663a7661132ab6a401d8881a478eaefe47d3ac83a95c4c540f93e888e70234a8
thread:issue:177227,issue #177227: Floating point exception (core dumped) with AMD processor,high-priority/blocker,82,high-priority/blocker,1,0,177227,,,,0,35a0c815ffa29506a093222cecb27bd6fda4cd42f4947a9bf19baed9ecbb91de
thread:issue:177561,issue #177561: CUDA driver error: invalid argument in fft backward when output used in subtraction,high-priority/blocker,82,high-priority/blocker,1,0,177561,,,,0,0d4bab8630999819bcc5c3bd0c75814413d8d38709a15a238d4462acf2304fa6
thread:issue:177906,issue #177906: CPU: 2x CPU utilization regression in 2.9+ with no proportional wall-clock improvement,high-priority/blocker,82,high-priority/blocker,1,0,177906,,,,0,733cacb7c1d9fd99162a1e23b7d321117fdedc91cac6308ee6441ecf5e17e334
thread:issue:178038,"issue #178038: torch.dot crashes with SIGFPE (exit code 136) on RTX 5060 Ti (Blackwell, sm_120)",high-priority/blocker,82,high-priority/blocker,1,0,178038,,,,0,fc1f930d93a63f4af2125a8655077c2b80d1a4016e62987e7fd39e444cb96876
thread:issue:178538,issue #178538: [BUG] [CRASH] Torch cause notebook crash on any exception (import torch),high-priority/blocker,82,high-priority/blocker,1,0,178538,,,,0,0fbad427aac153c038ad231ca1fd3673436c58909cc355028a16be2d29ce8ff8
thread:issue:178841,issue #178841: Importing torch before QApplication on macOS prevents Qt from discovering cocoa platform plugin (inverse of #11326),high-priority/blocker,82,high-priority/blocker,1,0,178841,,,,0,3bc19ecec018a872ef275b3eb4c118b209c56757c978c4d031fdc89f96e71829
thread:issue:178892,issue #178892: [Windows] torch.save triggers 0xC0000005 Access Violation on RTX 4090 Laptop (WDDM Driver Conflict),high-priority/blocker,82,high-priority/blocker,1,0,178892,,,,0,5afc6c0bae4faf22a60987b261203e1ee60dd2b3e4cc339cb463ac844880d0a7
thread:issue:179214,"issue #179214: cudaErrorLaunchOutOfResources in CTC loss backward on RTX 5090 (sm_120, Blackwell) with CUDA 13.0",high-priority/blocker,82,high-priority/blocker,1,0,179214,,,,0,01285eec8f238b4fccf40647f61808e45e9b307c3afcac1484a1b9d3dd3c1f74
thread:issue:180458,issue #180458: [Bug] ValueError: I/O operation on closed file during torch.save() on Windows (PyTorch Nightly + CUDA 13.2),high-priority/blocker,82,high-priority/blocker,1,0,180458,,,,0,b506dbf92038689b0d73052e789e39787b7165a8786c0c613db776dffc50ca3c
thread:issue:181534,issue #181534: CTCLoss backward is inconsistent with its forward derivative with respect to the documented log_probs input.,high-priority/blocker,82,high-priority/blocker,1,0,181534,,,,0,579184b8531d32f03f15c3e83cb4de9ffea129f954d2143df22b058e95cffd13
thread:issue:181853,issue #181853: [Bug] DataLoader worker segmentation fault in multi-task training with long sequences and numerical instability,high-priority/blocker,82,high-priority/blocker,1,0,181853,,,,0,32750931cb198865fc93da4767c7b9fcd5d9ec492603ac634427204c73ed3b2e
thread:issue:182279,issue #182279: Segmentation fault on RTX 5090 during repeated PyTorch CUDA forward passes,high-priority/blocker,82,high-priority/blocker,1,0,182279,,,,0,a2e1ecb88778a1cbdbcf9938f5b7b7becef55b8091827e78a6183747f29ecf6f
thread:issue:183251,issue #183251: PyTorch Segmentation Fault on M2 Mac - Bug Report,high-priority/blocker,82,high-priority/blocker,1,0,183251,,,,0,68f1dec9cbfca7048e32c3020e6a0d7b034259b3e941c1686416c527191a9ce3
thread:issue:183812,issue #183812: `torch.scatter_reduce` CPU performance regression (~15–58×) introduced in 2.13.0.,high-priority/blocker,82,high-priority/blocker,1,0,183812,,,,0,700861ce02536833915a637ff941157b7d5b887a12fd3bc729443192a3ca157c
thread:issue:183815,issue #183815: `F.conv2d` CPU performance regression (~4–16×) for bfloat16 inputs introduced in 2.13.0.,high-priority/blocker,82,high-priority/blocker,1,0,183815,,,,0,b0fa6d1faf64ff752deebe5ec02df8aa6dc86b01d77b1ab828ac88cdf7d72e62
thread:issue:188255,"issue #188255: Crash while import torch, in some special virtualization situation.",high-priority/blocker,82,high-priority/blocker,1,0,188255,,,,0,146d7f3fe85710fd5ee7c74fd051c87ba9e2d87b77a2435591380fa13ad5fe8f
thread:issue:10043:issue:101699:issue:103352:issue:103581,pr #186051: Fix disabled no-op module pre-hooks in compile,PR blocked,80,Review active PRs and linked issues as one maintainer work thread.,185,0,"1359, 1369, 2129, 3877, 8853, 9222, 9674, 10043, 12308, 12498, 13246, 14945, 17902, 18904, 19149, 20433, 22402, 23159, 24041, 24185, 27312, 28090, 28307, 28308, 28318, 28319, 28320, 28329, 28341, 29137, 29758, 29973, 31837, 31945, 32097, 32288, 32293, 32570, 32998, 33041, 34279, 34651, 35529, 35641, 40471, 40568, 40570, 41128, 41793, 43116, 43949, 44634, 45833, 45897, 45996, 46164, 46225, 46544, 46829, 46999, 48088, 48628, 50010, 50342, 50343, 50344, 50345, 50382, 51450, 51509, 51720, 51962, 53441, 53704, 53900, 54766, 55279, 55299, 55366, 56356, 56485, 56586, 56697, 56764, 57610, 57947, 58489, 58734, 58736, 58741, 58742, 58743, 58745, 58770, 58828, 59058, 59099, 59101, 59104, 59545, 59786, 59787, 59868, 60858, 61417, 61453, 61474, 61486, 61490, 61523, 61582, 61653, 62352, 62784, 62812, 63293, 63870, 64023, 64208, 64254, 64359, 64897, 68105, 68291, 69491, 70906, 70910, 70914, 70915, 70916, 70919, 70920, 70921, 70924, 70925, 71209, 71386, 71403, 72824, 76644, 80553, 83818, 86055, 87085, 87358, 87366, 87448, 89255, 92141, 94779, 101699, 103352, 103581, 105465, 106584, 106614, 106667, 106991, 108565, 108977, 109017, 109108, 114951, 115819, 123082, 124505, 127622, 132559, 133676, 135704, 135724, 135826, 136264, 139792, 142358, 143588, 145624, 148077, 150612, 151912, 156308, 165612, 187389","186051, 188012",,,205,011518df9bb8a80846dc85d7889569740bf5150ce879d272337973e4a4c33d86
thread:issue:189133:pr:188100:pr:188299:pr:188573,pr #188100: Support callable msg lambda in TestCase.assert* methods beyond assertEqual(),PR blocked,80,Review active PRs and linked issues as one maintainer work thread.,5,0,189133,"188100, 188299, 188573, 189286",,,18,a620a44baf1e2a4829fdd782585ff58260cdf9e0034323246f0fdeb80c0a03be
thread:issue:185955:pr:185989:pr:186969,pr #185989: Bump onnx to 1.22.0rc1 to integrate ONNX 1.22.0 release branch,draft/noise,80,Review active PRs and linked issues as one maintainer work thread.,3,0,185955,"185989, 186969",,,2,8ae248c9170676176adf2a1ac9adf4953306f2d9321617fd8aa6cad1b89e2948
thread:issue:118747:pr:186502,pr #186502: Make context decorator frames identify the context manager,ready for maintainer decision,80,Review active PRs and linked issues as one maintainer work thread.,2,0,118747,186502,,,1,a2a0b8efc7f0734f7c128e3b5c0c43bd489ea7b14c00cd0751727a4322537f6c
thread:issue:137285:pr:188967,pr #188967: Update logging docs to match current API,ready for maintainer decision,80,Review active PRs and linked issues as one maintainer work thread.,2,0,137285,188967,,,4,d5f9838ca78c85983b3b0a51039811622a1c9520a836c2bc759b2c0ccde10cdc
thread:issue:187170:pr:187171,pr #187171: [inductor] Fix test_comprehensive_addbmm_cpu_float16 precision mismatch (#187170),ready for maintainer decision,80,Review active PRs and linked issues as one maintainer work thread.,2,0,187170,187171,,,3,cde69ef46ec2d14b81bf78baa71ae737dfe744b39280470260a0cabbb1c27347
thread:issue:188989:pr:188990,pr #188990: [FSDP] Fix to_cpu when flat_param is already on CPU,ready for maintainer decision,80,Review active PRs and linked issues as one maintainer work thread.,2,0,188989,188990,,,2,e63da1c7dc0b95d5372346af006313e5f27dc5a28086bfd465cf938cc04eeb93
thread:issue:189106:pr:189142,pr #189142: fx: Fix incorrect return_annotation for tuple types in operator schemas,ready for maintainer decision,80,Review active PRs and linked issues as one maintainer work thread.,2,0,189106,189142,,,2,86cdcd2f3ea71a7e8832f8c841854660331d84d19fb40a1336cbd815d8fe423f
thread:pr:154279,pr #154279: Add Support for transposed convolution with Padding Mode 'same' and 'valid',ready for maintainer decision,80,ready for maintainer decision,1,0,,154279,,,0,d7b4c5c8f69da5e99af2bdb6c2fd231660cc959573a2626f1581dddab0304cf2
thread:pr:169249,pr #169249: [ARM] Disable SVE for GCC 11.5.0 to workaround compiler bug,ready for maintainer decision,80,ready for maintainer decision,1,0,,169249,,,0,17b3927881d8e9be20042f57e796fa618aa185e719d4083928feba54801eb367
thread:pr:174198,pr #174198: Adjust test_linalg_lstsq_input_check to be compatible with other devices,ready for maintainer decision,80,ready for maintainer decision,1,0,,174198,,,0,69360712294c1ba7d6a9252d14b87e45a80e7a9ad81e120ad99c1e51eec12c8b
thread:pr:177069,pr #177069: [xpu] Increase tolerance for tests that fail due to non-deterministic operators behavior.,ready for maintainer decision,80,ready for maintainer decision,1,0,,177069,,,0,6564a15e2a912e796c64bc0600931f7c562005a0322ad1897007b3ec0f2f16a0
thread:pr:178100,pr #178100: docs: mention torch.cuda.memory.mem_get_info in CUDA memory management notes,ready for maintainer decision,80,ready for maintainer decision,1,0,,178100,,,0,07cd82336ad896518d028857fa7fcf54849fbae563d932460865f24277dd46e1
thread:pr:178768,pr #178768: Pass tensor data pointer to dlDeviceToTorchDevice from maybeCopyTensor in DLConvertor.cpp,ready for maintainer decision,80,ready for maintainer decision,1,0,,178768,,,0,6b647a741e77e4fdb195743927c6757a9b6ff6b9a766258506e29d26380bdc08
thread:pr:178781,pr #178781: [MPS] Remove stale @skipIfMPS from test_nn.py,ready for maintainer decision,80,ready for maintainer decision,1,0,,178781,,,0,543fad945cf2add3a24164229e452cfee5bca71884bd95d671a432e79c990d7f
thread:pr:179528,pr #179528: [Codemod][PSS3] Upgrade fbcode/mobile-vision to Python Scientific Stack 3 (#179528),ready for maintainer decision,80,ready for maintainer decision,1,0,,179528,,,0,e421fd3be53b213e3b04f424f3ba58e7706d808ef47e83303cdeaf4605f1e869
thread:pr:179732,pr #179732: Fix torch.compile istft failing when length exceeds reconstructed sig…,ready for maintainer decision,80,ready for maintainer decision,1,0,,179732,,,0,70ff91f651c5374226ea6b0af87ddb6218c8e982f41950e3f08e140bbfe9738b
thread:pr:180742,pr #180742: [autograd] lazily start worker threads in the autograd engine,ready for maintainer decision,80,ready for maintainer decision,1,0,,180742,,,0,3745d7d14494c7a97b90175e9ec0e5854d286fca96eb7564baea4f6daf478ea3
thread:pr:182034,pr #182034: Add no-tree-loop-optimize pragma for apply_grad_input on GCC/AArch64,ready for maintainer decision,80,ready for maintainer decision,1,0,,182034,,,0,224e344a81b5d1cd232cb6cc68d5dba06e812f16592f6f5293de5d13c04c684c
thread:pr:182323,pr #182323: [Mergerules][Temp] Restrict docker changes to just maintainers,ready for maintainer decision,80,ready for maintainer decision,1,0,,182323,,,0,1081b33102722d577f34f905feae36a5f30f1369866d44db989d67c3800ad14d
thread:pr:182344,pr #182344: Inductor: Fix Searchsorted/Bucketize Sliced View Boundaries,ready for maintainer decision,80,ready for maintainer decision,1,0,,182344,,,0,835221ebcec9fd8c7a76ff90a7eef22acec47220801ac8e62b899afa8850c1f6
thread:pr:182546,pr #182546: Fix sparse compressed set_data device metadata,ready for maintainer decision,80,ready for maintainer decision,1,0,,182546,,,0,5a3a6589fafac3c1c2fac1f14fce6bd84694824aa1f35c13e0b9724aa42de651
thread:pr:182937,pr #182937: Align XPU unit tests,ready for maintainer decision,80,ready for maintainer decision,1,0,,182937,,,0,e96e3cb28787a461f4dfe487bf6e6dba3b488af99539924c3e665211d40db669
thread:pr:183697,pr #183697: Fix SVE blendv mask handling after bool fill,ready for maintainer decision,80,ready for maintainer decision,1,0,,183697,,,0,1bd6c26f1021cfeea2416d6b948e961b24334feb3e53c7143dee86ffab9a48a8
thread:pr:184745,pr #184745: distributions: fix Kumaraswamy mode formula,ready for maintainer decision,80,ready for maintainer decision,1,0,,184745,,,0,473cf5488eb1ab712102ce8a76e43a1d5a7546dffa37d7c93ee6a21ac310da8e
thread:pr:184881,pr #184881: Fix SIGSEGV from functorch key contamination via Tensor.set_data,ready for maintainer decision,80,ready for maintainer decision,1,0,,184881,,,0,fd234a18ffad906bf4e1cda0499840f7284b2b73bbc7d25aba7dee9dfe6cf713
thread:pr:185190,pr #185190: Make constant SymInts hashable,ready for maintainer decision,80,ready for maintainer decision,1,0,,185190,,,0,62e0ad2e1699c886d6ffd9a6573d147b6e2902067ae4cc6b20993fdf4f03b3dd
thread:pr:185812,pr #185812: Fix arange int64 fractional truncation,ready for maintainer decision,80,ready for maintainer decision,1,0,,185812,,,0,38b7d36b417f40c1261943ae83a4db178f65d63210efa25b00519babc0576b3f
thread:pr:185913,pr #185913: [BugFix] Fix torch.kron crash on non-contiguous input (Issue #185650),ready for maintainer decision,80,ready for maintainer decision,1,0,,185913,,,0,3178321bb82375f900b5ccdc54b6e9b8303d85b4437fb35e537bf222d16906f1
thread:pr:186133,pr #186133: [ROCm] remove redundant @skipIfRocm from test_cudnn_attention_broken_166211,ready for maintainer decision,80,ready for maintainer decision,1,0,,186133,,,0,39fa264c73bb155819c7f501af3072adcde81624b2273869ad3499998fc37077
thread:pr:186346,pr #186346: Handle invalid CUDA JITerator cache entries,ready for maintainer decision,80,ready for maintainer decision,1,0,,186346,,,0,84e1898f2854ed0349652e0b9de2977ab8180b8faf46a904e9760dbedb91cca3
thread:pr:186489,pr #186489: Fix division by zero in range when step becomes zero after conversion,ready for maintainer decision,80,ready for maintainer decision,1,0,,186489,,,0,ea1e4791557eac312637f14a47932fcfbde0ddc9aebbb2826df38429b8007ebc
thread:pr:186492,pr #186492: [inductor] Fix stale xoffset for TMA in mix-order reduction loop,ready for maintainer decision,80,ready for maintainer decision,1,0,,186492,,,0,e44270baa073cc545fd81f06ec8e38cb33872ae8223aa9c1895b9ee7de51c411
thread:pr:186566,pr #186566: Fix ldexp gradient for negative integer exponents,ready for maintainer decision,80,ready for maintainer decision,1,0,,186566,,,0,35febc01f4bf3ad54ee2df963cbdc5f20b5ecc23982a082532275a557eef9b0e
thread:pr:186570,pr #186570: [distributed] Fix Windows gloo allreduce crash by moving ProcessGroupGlooCuda.cpp to base_sources,ready for maintainer decision,80,ready for maintainer decision,1,0,,186570,,,0,6be342118614ae36f936b4f7313430960948ea7b8a3f8a289ced4d997d7558e8
thread:pr:186836,pr #186836: Fix Cauchy/Gumbel icdf boundary values at q=0/q=1 (#186824),ready for maintainer decision,80,ready for maintainer decision,1,0,,186836,,,0,1bc2c74ee152018c8b5708fda685827be7c73fd5d864cf1d4f14624c4634603e
thread:pr:187326,pr #187326: Fix torch.special.zeta returning NaN for q=+inf,ready for maintainer decision,80,ready for maintainer decision,1,0,,187326,,,0,b3df05a98d5620cac38366372daf819f9972ec4a942db6d0c871eecbc4c3dbd6
thread:pr:187396,pr #187396: Fix integer linspace decomposition to match eager,ready for maintainer decision,80,ready for maintainer decision,1,0,,187396,,,0,215bcb945ade483d089596deb51d1ec9b6fcef9eec1e9d47f7759e94ad63bb0c
thread:pr:187928,pr #187928: test: generate native function type hint checks,ready for maintainer decision,80,ready for maintainer decision,1,0,,187928,,,0,a03a1b9bb06d584dc944d3287bd5c4646afb324105a763cefdae5532b1fccf33
thread:pr:187969,pr #187969: Enable traceable DTensor RNG ops on XPU,ready for maintainer decision,80,ready for maintainer decision,1,0,,187969,,,0,4cd4a7a2f8be281e594621bdb8c0566eb1c25ca264ec2e2302eb0993e9346f63
thread:pr:188036,pr #188036: Fix THPStream_richcompare reading off the end of non-Stream objects,ready for maintainer decision,80,ready for maintainer decision,1,0,,188036,,,0,36c5b2465ba9ac8b5785ee067061cf9e3be8b61ddf8c100d3f9adf580f69b93f
thread:pr:188287,pr #188287: Improve FX CodeGen process_inputs typing,ready for maintainer decision,80,ready for maintainer decision,1,0,,188287,,,0,576318faaf582f934f3868775d0f036b109c500fbec3c478f09ddfefbd86a029
thread:pr:188321,pr #188321: [c10d] Integrate ncclScatter into PT,ready for maintainer decision,80,ready for maintainer decision,1,0,,188321,,,0,b6291521cbfd5446efdf663f6e36961820380f89bd733cd97dccee18ee0f0318
thread:pr:188378,"pr #188378: Fix linspace/logspace docstrings: start/end are Number, not float",ready for maintainer decision,80,ready for maintainer decision,1,0,,188378,,,0,52194991a6833442c4deb9be92c4812cba99ad51051f5f21162666d8072145a5
thread:pr:188799,pr #188799: Fix callable value-opaque objects causing graph break on method access,ready for maintainer decision,80,ready for maintainer decision,1,0,,188799,,,0,331b62e953656802235e4f739d188037b7b83429d8e50faf40db516f10c63126
thread:pr:189265,pr #189265: test/inductor: relax FlexAttention invalid-block-size regex,ready for maintainer decision,80,ready for maintainer decision,1,0,,189265,,,0,40bb5f42c9ea8a271a33c83b9a35bb85056b371297365b81f2411edb9f70e90a
thread:issue:171804:issue:172500,issue #171804: [On Demand Profiling] Early CUDA initialization caused by `import torch` when `KINETO_USE_DAEMON=1`,needs maintainer decision,77,needs maintainer decision,2,0,"171804, 172500",,,,1,30fbf189078d08d04012f91c23b4a4e0ae1ae58cbc45062f1048a5a94fd028cd
thread:issue:124007,issue #124007: gpu memory normal but one of gpus GPU-Util is 0 so cant train,needs triage,77,needs triage,1,0,124007,,,,0,416a00d31a7d1a9e876d5c7d5b9e4c11b5c662efde265bf50d5f7374c3e089a1
thread:issue:150011,issue #150011: [Windows][CPU] 51 UT of inductor/test_torchinductor_opinfo.py failed with CPP wrapper,needs triage,77,needs triage,1,0,150011,,,,0,d1060d0269a8c4e30a4075beb8a847c35d15654accb9fa5b82676d797ee40aee
thread:issue:162466,issue #162466: Launch one kernel take long time.,needs triage,77,needs triage,1,0,162466,,,,0,4ce7413d23d061d5b60f2313523a26b80944680b3791e3432929429652dfd91a
thread:issue:172254,issue #172254: Python 3.14 test error: `AttributeError: 'MobileNetV2' object has no attribute '__annotations__'.`,needs triage,77,needs triage,1,0,172254,,,,0,5dfa0b0618db4dc480b5d1258319608b6fce6d3a00f1d36c1e47a427f5ef3ac0
thread:issue:173907,issue #173907: [ARM][QNNPACK] quantized::conv2d.new not supported for TorchScript VisionTransformer models,needs triage,77,needs triage,1,0,173907,,,,0,fecaa03770124f7561bff622ce8828b8d08eb7be2040f22886a3e08bd7b05496
thread:issue:178085,issue #178085: DDP mode: CUDA error: an illegal memory access was encountered,needs triage,77,needs triage,1,0,178085,,,,0,54b6232e9b54649f545f0570bcf93e57563d22ad773a1fee832f8a474258bb78
thread:issue:178766,issue #178766: torch.distributed._functional_collectives.reduce_scatter_tensor gets a wrong result,needs triage,77,needs triage,1,0,178766,,,,0,cdc52ca68956c28d51d3a79e8fc3ae438d460273453bbdd686456f92be458cbf
thread:issue:168288:issue:172445:issue:172632:pr:166876,pr #166876: [ci][cpu] Update AArch64 manylinux compiler to GCC14,PR blocked,76,Review active PRs and linked issues as one maintainer work thread.,4,0,"168288, 172445, 172632",166876,,,3,9c694bd3eb7a3fb3b47d49a955c55068c93f3d109977eccc1e4debcb397a39ae
thread:issue:123800:issue:145555,issue #123800: add `FlopCounterMode` documentation,has linked PR,76,has linked PR,2,0,"123800, 145555",,,,1,9945962c35d0cb65b69d400d8c157189fc0080db88095d2c789129af0f06798b
thread:issue:143649:pr:186021,pr #186021: [Inductor][CPU] Guard compiled integer div by zero,PR blocked,76,Review active PRs and linked issues as one maintainer work thread.,2,0,143649,186021,,,1,a36df910eb443b4321f1a2bd8b8e5430151f1f3f207b8d8c3a36159f41d0d827
thread:issue:145220:pr:185984,pr #185984: Fix Dynamo out= resize for internal tensors,waiting on contributor,76,Wait for contributor update on PR #185984; keep related issue/PR context attached.,2,0,145220,185984,,,10,06a6d6b0eb166e81f6f2e887965bf5570101516c54e9f3227eb294ce68ef18e3
thread:issue:146790:pr:181062,pr #181062: Validate grad_output shape in multi_margin_loss_backward,has linked issue,76,Review active PRs and linked issues as one maintainer work thread.,2,0,146790,181062,,,3,a73f8112192d6ddeadae3aeeb3110cc6968bd5f370046b08cc86a7281ddfbf6c
thread:issue:147077:pr:185809,pr #185809: Fix as_strided_copy storage bounds in Inductor,PR blocked,76,Review active PRs and linked issues as one maintainer work thread.,2,0,147077,185809,,,1,3541655af546b07b9b58b89024b0d3a7c39f8f54807a9ceaca90ae61bf77507e
thread:issue:139789,"issue #139789: compiling ""input mutation"" followed by ""input mutation + aliasing"" does not properly recompile",has linked PR,76,has linked PR,1,0,139789,,,,0,0312cff741fdc4446e20f755751398878c970d1a374bc9cd25718074ad3fc490
thread:issue:150779,issue #150779: Decorator `skipIfXpu` disables tests when used on class,has linked PR,76,has linked PR,1,0,150779,,,,0,655ed0c42f25d22092cc7371bff60366ea14ac18c15df71a898ecabe1e15d08f
thread:issue:167086,issue #167086: Unchecked uint32 overflow in CUDA scan (cumsum/cumprod/cummin/etc),has linked PR,76,has linked PR,1,0,167086,,,,0,bc2a1992b0950271a9788dc41962e83c35990593fe60356b4e452ce2109014b5
thread:issue:170877,issue #170877: AArch64 unit test failure test_transpose_mxn_16_16_bf16_fp16 on Neoverse-V1,has linked PR,76,has linked PR,1,0,170877,,,,0,650410b598685c290336ce729fc6d1a210db03a1c584ff046811cd42b31d85b5
thread:issue:27614,issue #27614: scatter_add allows index tensor that doesn't match input size in forward pass but fails on backward pass,has linked PR,76,has linked PR,1,0,27614,,,,0,8b9d8664f0612c60f041040ce5a450fecd95bf04465ceae95e6b49981d909295
thread:issue:101082:issue:102839:issue:103444:issue:103588,pr #186539: [Dynamo] Limit FX node growth from loop unrolling,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,82,0,"2628, 23756, 26288, 29697, 30291, 31557, 32021, 35666, 37410, 41081, 41243, 41353, 42959, 46168, 46184, 50122, 50688, 54389, 55135, 58997, 61686, 62448, 62530, 64730, 66073, 66707, 67586, 67760, 67761, 67958, 68332, 68648, 68978, 68979, 69972, 70173, 71683, 72146, 78917, 79197, 82464, 83159, 86590, 89116, 92884, 94652, 95408, 96136, 101082, 102839, 103444, 103588, 104849, 105255, 106624, 110761, 111003, 111441, 112303, 112901, 113287, 113914, 114206, 115626, 118758, 118966, 125531, 128046, 129889, 148819, 158371, 158968, 159765, 161302, 162591, 166208, 166209, 168044, 171302, 173559, 175891",186539,,,102,b590c3d3e3e3772f427f3e54319807aadc8b331d7006eb3843e2b104aec243a4
thread:issue:122512:issue:156075:issue:162151:issue:167729,pr #160585: Add dim=None support to torch.logsumexp,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,35,0,"122512, 156075, 162151, 167729, 167732, 168393, 170487, 170496, 170834, 171158, 174929, 182842, 187806","160585, 163503, 171826, 181680, 184974, 185033, 185034, 185149, 185524, 186100, 186476, 187468, 187469, 187531, 187532, 187707, 188173, 189091, 189092, 189234, 189291, 189324",,,107,fc84a380edf95485a4386abe975d8acfd87c74718702f60a5f546a745dc989b8
thread:issue:130182:issue:136216:issue:137520:issue:140842,pr #183884: Fix compile worker quiesce draining,waiting on contributor,73,"Wait for contributor update on PR #184161, #185433, #185976; keep related issue/PR context attached.",17,0,"130182, 136216, 137520, 140842, 144360, 145093, 145231, 158088, 186031, 186136","183884, 184161, 185433, 185976, 186011, 186094, 186361",,,48,0653ab0e187c323c2a8f6533135e8bcc17bd2e6a4a6dd321ce98e7ae764e4fe9
thread:issue:145529:issue:153056:issue:170137:issue:170770,pr #184975: Support FakeTensor module materialization via to_empty,ready for maintainer decision,73,Review active PRs and linked issues as one maintainer work thread.,12,0,"145529, 153056, 170137, 170770, 170781, 186796, 187008","184975, 184976, 185953, 186801, 187090",,,15,16ac55002947114814d4ecc70ee7cd0014163019778f3fdb2a5f2b321aa3328f
thread:issue:13716:issue:18631:issue:187473:issue:70954,pr #184193: Preserve FX graph cache guard provenance,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,8,0,"13716, 18631, 70954, 73764, 75747, 95604, 187473",184193,,,7,71133202bf966529ad090ec9592defff755d8c1d05afbd3c973fe723dc49a997
thread:issue:162859:issue:167026:pr:185057:pr:188755,pr #185057: Support torch.cuda.use_mem_pool in compile,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,8,0,"162859, 167026","185057, 188755, 189123, 189173, 189175, 189230",,,14,3b085762c23c4261f080bb97c967e180c6ee263be39302af25528ef42084799f
thread:issue:161111:issue:161113:issue:161119:pr:184051,pr #184051: Honor custom op layout constraints in fallbacks,waiting on contributor,73,Wait for contributor update on PR #184051; keep related issue/PR context attached.,6,0,"161111, 161113, 161119","184051, 185280, 185288",,,12,4d34b7b24831cbd95eb9f5ccbc965382849938c64fdbc2b5192be5ba84a030b8
thread:issue:139908:issue:141265:issue:141916:issue:142250,pr #184185: Use memory stats for persistent reductions,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,5,0,"139908, 141265, 141916, 142250",184185,,,13,64fcc8c36f2f7a3de1b8d37514c89f5a6b8c5c8c121214f626c04acf406ea7c0
thread:issue:140845:issue:149909:issue:91439:pr:185902,pr #185902: Enable torch.compile for RNN modules by default,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,5,0,"91439, 140845, 149909","185902, 186091",,,6,7423be8f30e48c7255c95131246ef9ab5ebb239527e7f12d4ff38efaece3284f
thread:issue:123470:issue:162687:pr:185132:pr:186466,pr #185132: Fix custom op autograd defaults with tensorlist outputs,ready for maintainer decision,73,Review active PRs and linked issues as one maintainer work thread.,4,0,"123470, 162687","185132, 186466",,,5,bcb7ce4455450809efc2e6759104de43c4a3173cfae649919e4f8401539851ee
thread:issue:126024:issue:168042:pr:168999:pr:187589,pr #168999: [Pt. 1] Make some aspects of Dynamo/Inductor global state thread safe,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,4,0,"126024, 168042","168999, 187589",,,3,fa1b3784bd6d776691879f67e17a38fa3cd09f50cc88331cafa9233fca69304b
thread:issue:139124:issue:144247:issue:144362:pr:186016,pr #186016: Reject invalid bitwise dtypes in Inductor lowering,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,4,0,"139124, 144247, 144362",186016,,,3,5fa856815e6fe0e1ce5b7c4c644e3ca10496ead53eae3037c5b0a9517f5140ea
thread:issue:141184:issue:158007:issue:185135:pr:186614,pr #186614: Delay RNN graph break until module call,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,4,0,"141184, 158007, 185135",186614,,,3,c9adf2ffde922a976b117388d2fbf5f2a3f44255e1e3d8db9c0ea17b3bbdff0d
thread:issue:141258:issue:146951:pr:185815:pr:186069,pr #185815: Fix Dynamo NumPy scalar identity handling,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,4,0,"141258, 146951","185815, 186069",,,3,50dd8fd620c9497edd0f6429b7fc6adc5ea0deb2a21c1b2e0c39b5437873e43a
thread:issue:145687:issue:145709:issue:145711:pr:185939,pr #185939: Fix int8 WoQ GEMM residual epilogue fusion,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,4,0,"145687, 145709, 145711",185939,,,5,bee808e6ce43627c7cd031751af50a1943bc23ef9be01289d884996a7ae835c0
thread:issue:157217:issue:169538:issue:183887:pr:184990,pr #184990: [dynamo] Route fake RuntimeError through user exception handlers,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,4,0,"157217, 169538, 183887",184990,,,4,9503f3c792be8ab43eb84c1397e6a510ce092b3362afc559fc93b0d1252ae44a
thread:issue:103417:issue:103891:pr:186538,pr #186538: Fix CUDA driver linking for Inductor CUTLASS builds,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,3,0,"103417, 103891",186538,,,2,2c19b03303a6be8cd7a6f9a6a0ca563fd5136ca610a088221297ac8667ee4707
thread:issue:120115:issue:154259:pr:185528,pr #185528: Fix Dynamo tracing of compile support checks,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,3,0,"120115, 154259",185528,,,2,3654a6454ee154fc96cdc504819dacd740ffd6701cba4954d08ff90409b7e1d5
thread:issue:125171:issue:164922:pr:185085,pr #185085: [dynamo] Support datetime.now attrs as runtime inputs,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,3,0,"125171, 164922",185085,,,2,d6bc0044a201111bc9b4a8f1ab2c82e494b14b1a3d0eaee6a49ec008139ac607
thread:issue:128830:issue:134885:pr:186406,pr #186406: Fix invalid source in AOT repro generation,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,3,0,"128830, 134885",186406,,,2,4a3380dbc324d2a5c22ecc30ce08a1547f0dcadb186d2b1306cba9660c41b473
thread:issue:129131:issue:97894:pr:186403,pr #186403: Avoid CUDA init during CPU torch.compile,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,3,0,"97894, 129131",186403,,,2,2e6bb1c740c4fcd35423665d2cea272c809e9ead0e48b7473074952513006432
thread:issue:132133:issue:145014:pr:132135,"pr #132135: [torch.special] Adding betainc, betaincc, betaincinv, betainccinv, betaln and beta with backward operation",PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,3,0,"132133, 145014",132135,,,2,a88f5dd7fd8ec6a0762ccd676f56a11e3f961f7a59ba79a9c1db3129ce609f33
thread:issue:137869:issue:139232:pr:186180,pr #186180: Fix Dynamo cProfile under active Python profilers,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,3,0,"137869, 139232",186180,,,2,2a9e6803025a07822cbb81b2c2406e06d8bd2bfc51a82a127ba8f5beb324c4b6
thread:issue:140884:issue:156162:pr:185462,pr #185462: Fix NumPy ndarray operators under default device contexts,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,3,0,"140884, 156162",185462,,,2,bbbc4c6b80b5b15976868b23ba1db8722ea40de0bc1cef9184b16b06e018a53a
thread:issue:159843:issue:160697:pr:185356,pr #185356: Fix DTensor embedding mask state under compile,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,3,0,"159843, 160697",185356,,,2,6873af38eea774fd318bcdc95868b0edd8a49dbf6c2aaf054457ddd2c10fd2c1
thread:issue:163143:issue:165981:pr:185116,pr #185116: Thread effect tokens through cond branches,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,3,0,"163143, 165981",185116,,,2,fd8b3fd243d57d01393b615026855cb62fa538e014596f31ee2b4efc5b7aa28c
thread:issue:179368:pr:179372:pr:183862,pr #179372: [Inductor] Fix view_as_complex stride requirement in Inductor backward by forcing contiguous layout,ready for maintainer decision,73,Review active PRs and linked issues as one maintainer work thread.,3,0,179368,"179372, 183862",,,6,7962de921bfff606c610b042dd1c898c33b0f1ab169b61def2f00413655e1ad5
thread:issue:180651:pr:181096:pr:186857,pr #181096: inductor/flex_attention: fix fake block-mask generator for autotuner,waiting on contributor,73,Wait for contributor update on PR #181096; keep related issue/PR context attached.,3,0,180651,"181096, 186857",,,2,fd8ec9d9e26f7cc3402f8e3884d25aa1474cbb5ced37d20d61c06a39ee2920e2
thread:issue:184195:pr:184308:pr:186920,pr #184308: Route privateuse1-overrideable ops through AOTI dispatcher (,draft/noise,73,Review active PRs and linked issues as one maintainer work thread.,3,0,184195,"184308, 186920",,,2,1c875c8f8b852270c74e44da499849e43f012656342f5104388a075f40e88083
thread:issue:185888:pr:186205:pr:187890,pr #186205: [dynamo] Add as_strided_ handler to fix metadata propagation (#185888),waiting on contributor,73,Wait for contributor update on PR #186205; keep related issue/PR context attached.,3,0,185888,"186205, 187890",,,3,1ca87bd33bc7c729ba0c7f1e81a59a04b3ba5c10f6ff9b46760c8dc597486d05
thread:issue:187336:issue:187721:pr:187353,pr #187353: Preserve signed zero in Triton neg lowering,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,3,0,"187336, 187721",187353,,,2,8077620ee70c1eb87893357ce1a518a0a2943535a9720ad86df0506de12393dd
thread:issue:187810:issue:187811:pr:189224,pr #189224: [xpu][skip] Skip test_combo_kernel_dynamic_scale_rblock on XPU,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,3,0,"187810, 187811",189224,,,5,cd1d83af3a268de72756c1798d01cb9233444ecc72b3c45c2000c5c4777c005e
thread:issue:188422:issue:188423:pr:188636,pr #188636: Handle reciprocal SymPy powers in shape range eval,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,3,0,"188422, 188423",188636,,,2,d714ea55f3f8c9f3d9f4e5046a52bae3a6035104750a1a9c6161d0dc4d7ccabc
thread:issue:102870:pr:184460,pr #184460: Remove obsolete Python wrapper symbol extraction,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,102870,184460,,,1,29f6c80449c56478a22c101f87fb7ba5a142d0beb61aa5388a4162e30defcea7
thread:issue:105768:pr:186534,pr #186534: Account for fanout in Inductor reuse realization,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,105768,186534,,,1,aa271806c14b0a9f1f2136bc166874a4df308b873a9ee2d1163dbee3ddc89955
thread:issue:111385:pr:186530,pr #186530: [dynamo] Restore autocast dtype after trace-time updates,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,111385,186530,,,1,f2dfe66b6418d49dd903fc5b4f5782b3f14e0b32de234da40e555eba56edb2e4
thread:issue:111414:pr:186529,pr #186529: Preserve grad state in Dynamo minifier repros,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,111414,186529,,,1,b4878c175a345206395bc07e38542dc958c7582e36d195de3074b53332ebc1fb
thread:issue:114296:pr:186520,pr #186520: Preserve SymInt bindings in dynamic-shape minifier repros,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,114296,186520,,,1,99bf53060b92c19b4d40f37d33ee9d3b82d5dea2e31f80d2372d775e01dd8b5f
thread:issue:115484:pr:186512,pr #186512: [dynamo] Graph break on shape-changing Parameter.data assignment,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,115484,186512,,,1,7a6298d657bc6430390d3a1766dc79d2d183583627a18004988eb3f767839d67
thread:issue:117265:pr:186507,pr #186507: Invalidate Dynamo caches when module hooks change,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,117265,186507,,,7,58be23f1779b19ba28b28de5104f4e375c943a50b0090fcc5447214f2557aed1
thread:issue:118332:pr:186505,pr #186505: Simplify mod-one symbolic shape guards,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,118332,186505,,,1,86194d9b0bd8901c010eb4b63f92389144bb42b0a7945ac7cbb51f996e208f8e
thread:issue:118334:pr:186504,pr #186504: Allow HOP speculation through torch autograd.Function skipfiles,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,118334,186504,,,1,8232b3c041335fe38053838b55d2c4f82fd964fcb8b1df70299ada669dd41e4b
thread:issue:119191:pr:186501,pr #186501: Collapse foreach input mutation copies in AOTAutograd,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,119191,186501,,,9,e34782f4b38f3efe26ba3e8b965085e5c616c2ffe65ee248e656f748509171b7
thread:issue:120375:pr:186500,pr #186500: Fix Dynamo resume after f-string logging graph breaks,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,120375,186500,,,1,cfa57692efa013ea539bb9d3f41e2c8aea3251063109fccab93ac499a8d67f06
thread:issue:120911:pr:186498,pr #186498: Add dynamic shape test assertions for input dims,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,120911,186498,,,1,6b9266318f1e160a15158958a6c9233cefa58f316dd4d50fcc628a77cf80289b
thread:issue:121526:pr:186494,pr #186494: [Dynamo] Constant-fold CUDA Tensor.is_pinned(),PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,121526,186494,,,1,44c5c0d41b96341ecc69cd1b14cf07d323faf0df1d4d80f2e93f8f977471c7a9
thread:issue:122029:pr:186493,pr #186493: Optimize AOTAutograd compiled forward wrapper,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,122029,186493,,,1,989db9686e5c02acaf0272e4ec1ceeeb36f8a85c4a59c833aaf3f772dbf8a3ca
thread:issue:122200:pr:186487,pr #186487: Run Dynamo compilation with empty functorch stack,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,122200,186487,,,1,85cc334c2e968ded072c86b5f34844104876564cf03ceb25a4c9654818042766
thread:issue:122294:pr:186485,pr #186485: Fix export dynamic dims with size-one examples,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,122294,186485,,,3,c6bda78b63b219c1e73c79487260c5af9bc125fd093f6ac0224cb7d88972f34e
thread:issue:122578:pr:186472,pr #186472: Graph break on dynamic unspecialized ModuleList indexing,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,122578,186472,,,1,3aeb2a52124f6818f7f23f1d7c1b9a06682174ddb7b992aa50f99003d33d5fa9
thread:issue:123374:pr:186469,pr #186469: Wrap Dynamo-generated bytecode runtime errors,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,123374,186469,,,1,6cbd6f9275feacfca3c8c856934868d03768320b3b62b97afeb81994a010f0d7
thread:issue:123411:pr:186467,pr #186467: Preserve non-persistent buffers in compiled modules,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,123411,186467,,,1,62296b075bffc67b0be0511e3018e28777ef98e0f60d7708e90d17fa82929172
thread:issue:123972:pr:186462,pr #186462: Allow cond branch mutation during export,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,123972,186462,,,1,38a54cf83c231aad6bb8d05a13e1a098ecd42235da5ebea44b9d7115e974df43
thread:issue:124163:pr:186460,pr #186460: Support dynamic mode in Dynamo explain,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,124163,186460,,,1,0f9eda19292b7748e9fbd3d96c0fbbf3c5eb616452ee7847e3d9a9fbd8221278
thread:issue:124344:pr:186457,pr #186457: Fix torch.ops float parsing for scalar tensor args,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,124344,186457,,,1,cda3b1fe1c53735619c4f45324ed058b3387efe055e49a9faaac55011233a842
thread:issue:124747:pr:186455,pr #186455: Fix AOTAutograd Dynamo memory leak repro,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,124747,186455,,,1,9e424ad7ce16b64c9df8c377ff7657bbadf0f1dcde37a12114d7bba1caad1802
thread:issue:125277:pr:184343,pr #184343: Prune slow Hopper WGMMA autotune configs,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,125277,184343,,,1,d8de67a4fe70319b6fe72fa08820acdcf297887f0b57820c8c511f8101892208
thread:issue:126474:pr:186446,pr #186446: Fix fake overlap handling for Dynamo try/except,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,126474,186446,,,1,3c17ea5b10ee945396d34ab816b945411feae940ee326756598c5da23ff5eb58
thread:issue:127863:pr:184323,pr #184323: Share foreach combo kernel bodies,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,127863,184323,,,1,cff03091e8282c95ddf21d5b5fa8613825e9a2746787f84db6c245aaa55eefc1
thread:issue:128160:pr:186425,pr #186425: Skip Dynamo inside non-traceable subclass dispatch hooks,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,128160,186425,,,1,1fe84373c5d101cbf4033c1ba31453aa015d46131a990908ad53364431a906c3
thread:issue:128556:pr:186415,pr #186415: Fix Dynamo Parameter construction in new modules,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,128556,186415,,,3,89c5bb749a21e0c241e49dd86d64b2cd4cb2bf4819a0ad2149e0b89bf6bd0fb3
thread:issue:129418:pr:186380,pr #186380: Run python decompositions under functionalization,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,129418,186380,,,2,14d7d7613a4e4c7b97009746875bb3a848368f2513e95d63505539993ae1b81b
thread:issue:129648:pr:186372,pr #186372: Fix torch.Tensor construction from traced NumPy arrays,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,129648,186372,,,1,8111ec02b17c09da4d3fec2c40ffca4b67b2da1c37c0dbcd4648ead8f83120aa
thread:issue:129936:pr:186364,pr #186364: Prevent guard validation from mutating ShapeEnv,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,129936,186364,,,1,63a5b4cfa23271694589bd378b59991c539c10d907550c1de1b520cf51480d64
thread:issue:131794:pr:186355,pr #186355: Fix AOTAutograd unused output backward pruning,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,131794,186355,,,1,51d52a195a445b80a3442869a94d2d5bf4fcafda706f7e0313e9f1f475a5640c
thread:issue:132635:pr:186349,pr #186349: Improve Dynamo logger method graph-break guidance,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,132635,186349,,,1,e59244820de159425578ca77474b2b2d977f31591b8160f5413b442d78663203
thread:issue:132923:pr:186340,pr #186340: Ensure explicit compile attempts skipped code,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,132923,186340,,,1,ce484a867d00b779d553c475dd8702224a2a8f5023779bfde3e6a81198bcfad8
thread:issue:135492:pr:184294,pr #184294: Avoid unused storage size guards for indexing,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,135492,184294,,,1,d7dcf4daaa8ef6bea466d86b5cd1b1f4f299eea8b7875ae2a4c79d38c0c56828
thread:issue:135759:pr:186287,pr #186287: Reject non-functional export decomposition entries,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,135759,186287,,,1,f71918209a60912026f1e4640d19542c3e892588fea9840e0fa9f7c72e395653
thread:issue:136404:pr:186280,pr #186280: Disable profiler Python tracing during compile,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,136404,186280,,,1,55b15ebecd8991eee696f2c37dbce40c4b1242a9178724ec4286c8a73da0bd5e
thread:issue:136642:pr:186263,pr #186263: Increase small fake tensor constant limit,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,136642,186263,,,4,bef8c27a172645db7ddb339a773d2f416bc7e2317e40122572b8038eec8fd42a
thread:issue:137068:pr:186249,pr #186249: [dynamo] Support time.time as a runtime scalar input,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,137068,186249,,,1,513cfcb5786c8347c7330c32d6394a0c7a19bd9087a3dc0113712a015f22543f
thread:issue:137096:pr:186248,pr #186248: Simplify Min/Max using ShapeEnv value ranges,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,137096,186248,,,1,a6359dcb1db508a7e3e25d12b94dba87b7b943f9157808355e78bcd1fbe43020
thread:issue:137388:pr:186222,pr #186222: Include epilogue guards in Dynamo code_parts,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,137388,186222,,,1,4b619376bbb678e85b424f09bc5440cb9e014aea5d3ecdc7a3e9afa3d069d437
thread:issue:137875:pr:186217,pr #186217: Move FX graph cache hash details out of debug logs,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,137875,186217,,,1,240e6db9e238deebce439b66b74d6784971e15b1113203977aa211f5965dea69
thread:issue:137996:pr:186214,pr #186214: Profile lazy backward compilation with cProfile,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,137996,186214,,,1,e9c50e9992af7a3125432192517d4dbfc3b4eb65c039bb41913163150ffe7fa4
thread:issue:138141:pr:186210,pr #186210: Fix cond branch-local shape assertions,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,138141,186210,,,1,799f878e92fd908bb4bf923dbb35e39d3e5551caf0f1d1456febc3d7f456b62a
thread:issue:138301:pr:184256,pr #184256: [inductor] Cap reduction XBLOCK to max block,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,138301,184256,,,1,0a4eaf15676c5005379fa6f8267c9dc61a10f4d26bc1f6b825fd607c6da1ec76
thread:issue:138961:pr:186192,pr #186192: Fix CompiledFxGraph callable state across threads,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,138961,186192,,,1,4aac14cbd9ff7638d0ef36f09e43adf847c4b43da6e6b580142b09d3efc21fab
thread:issue:139603:pr:186168,pr #186168: Fix Dynamo tracing of torch.random.fork_rng,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,139603,186168,,,1,ae96f6402bd71d8b828cb0ef29eee017025866b7eb10a54a24095f1dd5ebf8d9
thread:issue:139707:pr:186166,pr #186166: Respect fake/functional modes for backend autograd kernels,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,139707,186166,,,1,9c3c276ff6e80c3492d1c9441a6bfeb5d50575fc05b002a37a6fb34169f17c92
thread:issue:140171:pr:186150,pr #186150: Fix compile storage semantics for as_strided scatter,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,140171,186150,,,4,f12b08ab088e9ee91ceb2940858a8c64d2237930a9ddf7ce97b7d061f41a7926
thread:issue:140607:pr:186119,pr #186119: Improve Dynamo observed exception messages,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,140607,186119,,,1,0161d6fa6af4a910e312aba77cf235dccebb86f5f15c6d103f2e86e1635ded04
thread:issue:140615:pr:186900,pr #186900: Improve pytree missing registration error,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,140615,186900,,,1,ccd36a5db9008bafa469fc4f88415cac44deef38c0cdb665821ba2ebbfaf7cb8
thread:issue:141005:pr:186078,pr #186078: Fix OptimizedModule _parameters access,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,141005,186078,,,1,ae03df5c1a2a57bb9f453aef4a386dde2ec0f552ce4a87d75d399b8b9856c9b5
thread:issue:141115:pr:186898,pr #186898: Add always-on AOTI input rank checks,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,141115,186898,,,1,cdbe1014e372e9b9a907b41215514d3ad05514604b7e2c9216deec7fbb566179
thread:issue:141162:pr:186072,pr #186072: Make constant warnings reorderable in Dynamo by default,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,141162,186072,,,1,c1938ff9ca721337e11fc3ea77b88f878055ce9acb9c484d5dec3d0cfe5291a8
thread:issue:141473:pr:186063,pr #186063: [dynamo] Route OptimizedModule delegated methods through compiled call,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,141473,186063,,,1,42f5bb556bcda24a4e737a17687b2482f2486e4ffb19055fb055a10400f8d6a1
thread:issue:141640:pr:186058,pr #186058: Preserve export graph metadata during normalization,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,141640,186058,,,1,2b5a3cbea918ca095aa296c76a238a63d7c9b0922c4ba4a34f0336c980a5a4c4
thread:issue:141744:pr:184189,pr #184189: Make generated Triton code use runtime device properties,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,141744,184189,,,3,175441225c1f20376b852197b5bf9cefa5c391595512a5316e6a70085dba322a
thread:issue:141790:pr:186053,pr #186053: Fix AOTAutograd cache save for call-free lazy backwards,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,141790,186053,,,1,2ed31bef5e91041780b5c6fb96c5f480f1e04c43c82521a4c4310908d3ba0f00
thread:issue:142321:pr:186052,pr #186052: Fix Inductor C++ pybinding source literal escaping,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,142321,186052,,,1,dbfd1de243e19319831abba06f08cfc160ba0f7c2b7c753f1a12729515b5929a
thread:issue:142489:pr:186044,pr #186044: Stop Dynamo from importing optional libraries at import time,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,142489,186044,,,1,80d778e3bbd80f0a565a02b5e4d35de303a25b802c8779c6f84c216618ee1c40
thread:issue:143463:pr:186032,pr #186032: Avoid reshape metadata guards on symbolic size-one checks,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,143463,186032,,,1,d85a73decb7c8e39e073bb9b1a45525df3c61b66e968ff24a181a3c556129b1d
thread:issue:143495:pr:186027,pr #186027: Fix symbolic broadcast inference for unbacked sizes,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,143495,186027,,,1,da32554ddb6fee4a2c691d502e4904c0305d577d4dcc91ad3157a2de79e29a79
thread:issue:144820:pr:186003,pr #186003: Fix Dynamo condition-dependent frame skip caching,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,144820,186003,,,1,3ee74de150caa38cce436d770e863fb7ba9aff402c32db8658c68614d1dce74c
thread:issue:145383:pr:185972,pr #185972: Handle malformed MSVC help output,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,145383,185972,,,1,9aa2983dfc87db9cab4d786fc0286511b4b0d6b003e1ffa1d78a934b98b3ba27
thread:issue:145445:pr:185971,pr #185971: Fix Dynamo random.Random replay on cache hits,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,145445,185971,,,1,5e167043632d3916b9506dfcf590fd12833ff5c64d286706c2cf35b50a12a8c1
thread:issue:145574:pr:185952,pr #185952: Support compiled async all_reduce wait,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,145574,185952,,,1,361c11b0ec985fe71900094b71dea1d28ec0a3eeffa0c6f760abba2580e11443
thread:issue:145596:pr:185947,pr #185947: Fix Dynamo logit with tensor eps,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,145596,185947,,,1,a2a2ac946543044d260b2e6441d4cf63e7bb04d13249ce28f119b0cb3da27ad6
thread:issue:145899:pr:185829,pr #185829: Fix autograd.grad graph breaks under aot_eager,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,145899,185829,,,3,89e952cafbc4ddcda4f399dd0690e2d1543a19e048aa0eee703a279dbbc6a374
thread:issue:146628:pr:185920,pr #185920: [Dynamo] Detect tensor mutations when reconstructing generators,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,146628,185920,,,1,02a205caa5eb50dbdf958897221c79c42a0d21dd9b360f0900c28970908427f0
thread:issue:146673:pr:185823,pr #185823: [Dynamo] Handle class method_descriptor calls generically,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,146673,185823,,,1,6a26defcb0ba29c58bfd4237025540a3bbac470ab03dec70ab374b3e53985696
thread:issue:146896:pr:185916,pr #185916: Fix FlexAttention nested index backward,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,146896,185916,,,1,ec8d8e3dcff92a29a2348186cd8b17f9c74314eb9498d49ec9124f9dce9f63ec
thread:issue:147486:pr:185800,pr #185800: Fix TORCH_LOGS stderr handling at process exit,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,147486,185800,,,1,9c56e88d12d51463256ea68b48869e7942a1e355dc3e35a06337def28c02d8c7
thread:issue:147530:pr:185799,pr #185799: Avoid dataclass constructors in Dynamo and make_fx graphs,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,147530,185799,,,1,8cefc81bcc222797d88934b1fa939154ee4a072e6cba897c47a3c8360062e815
thread:issue:147843:pr:185795,pr #185795: Preserve overlapping input layouts when cloning Dynamo inputs,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,147843,185795,,,1,761192d3796f7630618449deb009c6d871f63ef24a6ef0c8e0962795a7d0596a
thread:issue:148112:pr:185784,pr #185784: Fix flex attention captured vector gradient scatters,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,148112,185784,,,1,29760a9f67e0075ea776aa81da59ab582848d314323b14d8f65ff2983d3bb1d8
thread:issue:148199:pr:185781,pr #185781: Avoid unbound symbols in insignificant output strides,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,148199,185781,,,1,710fbbb66225495eda53541631386c884490c6c1466720e07b2ed1cfffae2415
thread:issue:148475:pr:185776,pr #185776: Propagate fake RuntimeErrors to Dynamo handlers,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,148475,185776,,,1,8460af95c2e1823eff2bff26343b61ac306942b95c8bb8a73e05a2d3ffdcf6e6
thread:issue:148695:pr:185775,pr #185775: Avoid storage access for no-storage tensors in MetaConverter,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,148695,185775,,,2,a5e114e58f0885e792a054ba422f57b54ff6f14a480c0633b7bf26875e62969b
thread:issue:148751:pr:184124,pr #184124: Fix Inductor multi-output replacement insertion order,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,148751,184124,,,5,ed31fc5a02e837ead0baa03869414c63971bc70e198224e3907c8e2af8ab6830
thread:issue:148842:pr:185773,pr #185773: Use max dynamic bounds for Inductor autotuning,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,148842,185773,,,1,8b6da35a9542fc4800feb1f55e5137c56feed00a02744b776a458407aa4147d4
thread:issue:149010:pr:185771,pr #185771: Support Dynamo tracing through textwrap.indent string formatting,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,149010,185771,,,1,03be43d7e05e0a30559113270f13ceb9905e45b6e25f69b5599c630571b49f6f
thread:issue:149037:pr:185769,pr #185769: Preserve generator return values in Dynamo,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,149037,185769,,,1,cb2cb012f7045e92ee7491c973b1eed3c0677fda446e2558ab98cd636af88998
thread:issue:149047:pr:185768,pr #185768: Improve Dynamo error for dynamic module hook registration,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,149047,185768,,,1,54999abd063119035ebc642d6cb4b403f2465a916d81d995bb719584e74baf5f
thread:issue:149556:pr:185766,pr #185766: Dynamo: support torch.jit.isinstance without graph breaks,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,149556,185766,,,1,90e1b6945f064f4114a17d76bc938515a22c9cb81ee494ccf21589928e9e6eae
thread:issue:149566:pr:185763,pr #185763: Improve disable API graph break messages,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,149566,185763,,,1,03ba3f184432b65533cd67481b2ef189ebe5e95c8770da2c7a04dc06d9960c35
thread:issue:149583:pr:185908,pr #185908: [Dynamo] Preserve live dict view semantics around mappingproxy mutation,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,149583,185908,,,1,0486cefa2c56f2b155d906c2accec0cb1d9ec5e32c1ddc6f84a70099d0aa2fec
thread:issue:149586:pr:185762,pr #185762: [Dynamo] Support _AutoDispatchBelowAutograd context manager,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,149586,185762,,,1,d6170cdc857ed489f230a58d6ace7235d6605534f03659cd367b739bdd2b405e
thread:issue:149894:pr:185752,pr #185752: Make OptimizedModule state_dict use public keys,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,149894,185752,,,1,fd3710a5c556b6980ea4faa22dcf46f1cb3a5f7c9cb6ff3665685f3d62bf0673
thread:issue:149963:pr:185742,pr #185742: Support None leaves in nonstrict_trace,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,149963,185742,,,1,6c06c8a9b217d0824766dd3c549c4d4e2eccafd00fecd368e50037fc360c1950
thread:issue:149968:pr:185740,pr #185740: Support inspect.signature on nested functions in Dynamo,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,149968,185740,,,1,265152d202efdccda04309363cd5e31ebbc8010a4d960317aef236d1802821b9
thread:issue:150063:pr:185734,pr #185734: Preserve torch._check messages in runtime asserts,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,150063,185734,,,1,d8acd1ec12e5b30dfb3d7b2469bc04504b48008969d6058154569fe5b036d043
thread:issue:150262:pr:185732,pr #185732: Fix nested compile during subclass fake tracing,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,150262,185732,,,1,4b930a77a97b455cc771113c68940beb28beb3a66a65f5343ac8552c410de17a
thread:issue:150319:pr:185731,pr #185731: Clear stale prefix exception table entries in Dynamo,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,150319,185731,,,1,f45cab4f1aff85cbdea302a2c8ca0b85a04af8443d73bb7ab72c6309c03b6498
thread:issue:150465:pr:185726,pr #185726: Improve Dynamo stack for nonstrict_trace input errors,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,150465,185726,,,1,64494edf647c3a69219c2e9a341eba7b6a80ed925e1272640208d8fc42ef8d12
thread:issue:150613:pr:185724,pr #185724: Fix scalar searchsorted decomposition for unbacked values,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,150613,185724,,,1,05ce9f4ec95ba85bf3693eb97f57df3d39e9a5f505aa3be4a4d346363b2f95ed
thread:issue:150621:pr:184119,pr #184119: Avoid fp8 bf16 prologue fusion on pre-sm90,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,150621,184119,,,11,f515b5317042aa3d6e61d7ecc67988369c7bc3876f8d69e74f8ffb37ea46312b
thread:issue:150915:pr:185722,pr #185722: [dynamo] Wrap module.compile leaf module call frames,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,150915,185722,,,1,c8f4adc7945c9ce7d19b7f3d432b54f9eb52e2ed0d86baa44588d47d97b026be
thread:issue:151649:pr:184109,pr #184109: Fix Inductor replacement patterns across view-to-reshape,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,151649,184109,,,5,101a4e247ab4a20542deb805ca674aa351cf2048d3375fa9af2f1011131b6020
thread:issue:152183:pr:185632,pr #185632: Make AOTAutograd metadata repr subclass-safe,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,152183,185632,,,1,1f831d404dd3658b14fc1ad0c3125f730e61c6eb8667d4bf11af525f8775b51f
thread:issue:152343:pr:185624,pr #185624: Fix subclass static input remapping for SymInt args,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,152343,185624,,,3,2d05d3e9b3834f6864ea7b30e490fe38d4af3d822e9cbc42e2a96aaef8a5e44c
thread:issue:152548:pr:184103,pr #184103: Fix FakeTensorUpdater HOP dtype propagation,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,152548,184103,,,10,8670847a0804c6a3443e8b21c5c1d5620f910e8a5d08926f44ffb2def963cdf3
thread:issue:152656:pr:185604,pr #185604: Improve infer_size unbacked broadcast error,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,152656,185604,,,1,e67cd69cfef4283a9cba9fb75f7791b0a9e958c1f7f8c124e3b3451df35a6ace
thread:issue:153149:pr:185591,pr #185591: Canonicalize avg_pool2d singleton lists during export,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,153149,185591,,,1,df0204ba8ddf7f83c4443ca146552e8ec6fccf41120ac01e8fbc98bd041f46bb
thread:issue:153175:pr:185588,pr #185588: [inductor] Codegen backed SymInt input runtime asserts,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,153175,185588,,,1,167e2097e830ffcb1b8dc4e2f65c88d2cecfdf93685185186d8f5e52832e6e6a
thread:issue:153247:pr:185568,pr #185568: Fix export of arange with data-dependent float step,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,153247,185568,,,1,6886bf57ce830c2c4b8204631b58929f382aac3029dd65fa20263b74f63383bf
thread:issue:154137:pr:185545,pr #185545: Handle symbolic bounds in slice decomposition,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,154137,185545,,,1,480f87323e2ffc73f614c46ad4bbf5d0c4d4a7a51a46af515b7f9bb34c670e8d
thread:issue:154161:pr:185536,pr #185536: Teach Dynamo PGO about sparse tensor strides,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,154161,185536,,,1,d389f6a5837f7e4c460fae83f3efe29d0d4ba05155bd11ebb52abc563c8e5bf5
thread:issue:154454:pr:185508,pr #185508: [dynamo] Preserve observed exception stack for reraises,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,154454,185508,,,1,e5ddfe845b6fbe7e0cb05909d1d7fef8376fd21734b888f3636bcf9b2115657f
thread:issue:154559:pr:185504,pr #185504: Support unbacked SymInt closures in cond export,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,154559,185504,,,1,8966ae8976e7670e0a60886f71b9c3d164ed9c8dff41de3197327b79de736b17
thread:issue:154592:pr:184090,pr #184090: Preserve benchmark integer inputs for Inductor debug,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,154592,184090,,,3,11afdb2b29d42c55bbbc52c62c8708380578299ec3ee5164ee230d74526bfb81
thread:issue:154807:pr:184086,pr #184086: Fix Inductor scaled_mm scale_result lowering,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,154807,184086,,,3,bf9c7da5bf2ad5487804a5b24992aaca950bdb3d8bb3d1f201c2de4a1f3b4a8a
thread:issue:155421:pr:185474,pr #185474: Fix nested tensor tangent metadata coercion,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,155421,185474,,,1,7c8514cfab61fcf95693255a1d5e9fa2bc621344cd92c85cd40452c08ff475f8
thread:issue:155800:pr:185471,pr #185471: Handle Dynamo tracing of torch._C._infer_size,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,155800,185471,,,1,33497e7a32dec21c968c1fc60186bb349a8d832cf9430d47b2b28a0174414ace
thread:issue:155856:pr:185469,pr #185469: Fix stable TMA descriptor reconstruction in Dynamo,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,155856,185469,,,1,87f2382930ac0a7e64e96ef5792726479d2ad1b06c4aef1891123935a7964325
thread:issue:156059:pr:185467,pr #185467: Keep tensors alive for Dynamo pointer reconstruction,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,156059,185467,,,1,cc0f567fba8958720a72f893d899575d095193c3608e99122eef554dc180be02
thread:issue:156127:pr:185466,pr #185466: [dynamo] Show carets in graph break user stacks,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,156127,185466,,,1,aa7f9dc1f3da49d1c477d3205376ff35e074c14acc02ddc763a7470576ead0f1
thread:issue:156191:pr:185461,pr #185461: [dynamo] Handle module-level logging ignores,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,156191,185461,,,1,c75b664685d839e6fc9d3fcf5d86444143694de6f2a2c3d3d5b59da2f37ab6cb
thread:issue:156614:pr:185456,pr #185456: Fix Dynamo tracing for Tensor.random_,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,156614,185456,,,1,7166d526e7a0bb6ee6d1c1ba2bb3f0248f539a55dfda3dfdfb9bf17cec2aa3c9
thread:issue:156640:pr:185451,pr #185451: Wrap internal Dynamo assertion errors,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,156640,185451,,,1,d4fa90c68ba491397138d1f7422b2c0202b1cd901bbb6708d71db520ced0c136
thread:issue:157926:pr:185438,pr #185438: Scope distribution validation disabling to Dynamo tracing,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,157926,185438,,,1,6129a5903636ae380c07107ad33a1fede981e9fac646d30d4e99ca782173101e
thread:issue:158087:pr:185436,pr #185436: Preserve eager exception assertions for eager_noexcept,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,158087,185436,,,1,c6ee3c7415e73f624c699544dd567226efc1bf6dada086e74f93f4520c74ebed
thread:issue:158384:pr:185419,pr #185419: [dynamo] avoid stale verbose hint for backend failures,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,158384,185419,,,1,97f3509a181646ff21ecd89b57a574a7fb4cc8c0304f9f5dd6c3ae7144af1e63
thread:issue:159457:pr:185373,pr #185373: Fix assume_constant_result tensor constants after graph breaks,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,159457,185373,,,1,1ccd918070f9e2f932060557fccb4af2019a34547881ffe974077ea56dfc1bb3
thread:issue:159468:pr:185371,pr #185371: Fix symbolic antialiased resize export shapes,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,159468,185371,,,5,8f3b0a654899e6aa72be75108a90d025eb7399d19718c24eb9fc5f3da2cce864
thread:issue:159613:pr:185365,pr #185365: Fix FX subgraph rewrite of adjacent aggregate matches,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,159613,185365,,,3,d7c8643d0a1fc549ff57f55dae5d98b82d69cbeb28f0c951d190781855fff767
thread:issue:159831:pr:185357,pr #185357: Forward container methods on OptimizedModule,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,159831,185357,,,3,81e7c3cbcff61e582293f5fd62624f1e320fbdb8d86c8a85595dc3f8509cb67e
thread:issue:160083:pr:185336,pr #185336: [dynamo] Add regression test for prepended forward pre-hooks,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,160083,185336,,,1,639e263434ced682826fce20231066874d56b7fac48f5acaab21f4d4d206d07e
thread:issue:160124:pr:184059,pr #184059: Fix split reduction kernel metadata,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,160124,184059,,,1,e8acfb577070dcfab2414099c10ae0aa6454402dc8a8277c6cea47a3e5dd41fd
thread:issue:160247:pr:185333,pr #185333: Graph break existing list tensor assignments in Dynamo,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,160247,185333,,,9,5fcaba5f2de964d1712e0ed36c8e9c31169f51c30df90d27951f6094b0b54a2c
thread:issue:160271:pr:186893,pr #186893: Fix dynamic size-one tensor recompiles,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,160271,186893,,,5,81fca4e6cbe63eaf31676e46fed678ae9046945a8440e44e9dd76ce741a0d3d6
thread:issue:160544:pr:185320,pr #185320: Improve scan HOP direct call error,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,160544,185320,,,1,2d86cade94ea4905cd93df633374257d021212b0019c55a8519394473c11e89e
thread:issue:160939:pr:185304,pr #185304: Fix while_loop stride metadata preservation,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,160939,185304,,,1,dc3494229e1b16ab872d77b98d7350d7f5084282e26d5ccfa611ece1e1661c34
thread:issue:161132:pr:184048,pr #184048: Enable Inductor OpInfo CPU coverage,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,161132,184048,,,1,43563b44a1ba98e4fc177a5010572dfbb3c9c015f9611457435bd7acb779cd42
thread:issue:161138:pr:184047,pr #184047: Enable cooperative reductions by default,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,161138,184047,,,5,ba54460773e6f807ae2ee3a2a8c50747c9597988596ce7e98d10f7ecb1154959
thread:issue:161764:pr:185159,pr #185159: Inductor: gate max-autotune 1x1 conv GEMM lowering,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,161764,185159,,,3,9953edd9c1cf9d85e40767efbde861d57a2618d506350c49f1b4a8e06ff34ac0
thread:issue:162415:pr:185138,pr #185138: [dynamo] Enforce one compiled frame for fullgraph skipped frames,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,162415,185138,,,1,2928529e3a149c2ed22b4f0794fbce33254f7e30fea79aa34e52531d33aac8f0
thread:issue:163641:pr:185103,pr #185103: Fix unbacked scalar equality reasoning through div/mod,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,163641,185103,,,1,7657fdd6a13328cf29012be8b065df097e29c1b7322e7c333a483e2c56b402b3
thread:issue:163713:pr:185100,pr #185100: Keep fake tensor cache synthesis out of proxy traces,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,163713,185100,,,1,baf7973c619bcf114d8d538d9862c33ae35a33395b86424a6902a37fd905a445
thread:issue:164124:pr:186886,pr #186886: Fix keyed user Triton autotune in Inductor,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,164124,186886,,,1,b97c5a409edcc6e65662f4ee04b647c684aff7aee942a18c22872a5f20bde185
thread:issue:165032:pr:185084,pr #185084: Avoid contiguity guards on maybe-singleton symbolic dims,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,165032,185084,,,1,be0e18b21b8dc97ef483c57c424f115ac8258780096d6dcad124a61b62da1713
thread:issue:166525:pr:185063,pr #185063: Handle scalar carries in Inductor while_loop,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,166525,185063,,,1,4088aeaffb7da4f4028cfcc9da04dc64807474db79ee58e4d9a17ae75b25dcf1
thread:issue:167596:pr:185042,pr #185042: Avoid value guards for residual lazy constants,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,167596,185042,,,2,ca1d39d63b747da658fae6b728835830a1a7719a9ef2eb299bcc2c2a3d11cf15
thread:issue:167725:pr:184007,pr #184007: [inductor] Clear stale subprocess cache env vars,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,167725,184007,,,1,d6e0782779707994d77bb217ca46e231c47de93a11cb2eb273f2fcc3697d7f9c
thread:issue:169769:pr:184988,pr #184988: Allow HOPs to pass through unchanged input tensors,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,169769,184988,,,1,e896797ec779f6ffd8fa7f78cf6134ac806c23519297b57d39a2b3d29a3b5ff9
thread:issue:170696:pr:184977,pr #184977: Fix Dynamo wrapper method sources for nested compile,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,170696,184977,,,1,25218dcf910921e18732887a60a3327fe4b8014be06cfe8c9a49673f515f2d0c
thread:issue:172942:pr:183994,pr #183994: Specialize numpy scalar module attrs,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,172942,183994,,,5,7cb3a9dcfb0adae5ff9d783c68c30e8094b5b7d8c45da7f4661e67c7be62444b
thread:issue:173046:pr:184934,pr #184934: Validate packed RNN batch_sizes before fallback,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,173046,184934,,,1,8e7a89d46e82a120ba52ce2cf512dd7e0c00eceffe1a1da76957733be30b920f
thread:issue:173578:pr:184920,pr #184920: [dynamo] Store FunctionIdSet entries by identity weakrefs,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,173578,184920,,,1,b55a54ed828d80a69e9271c255323ca5e0145f329427711852509437e245631d
thread:issue:174183:pr:188327,pr #188327: [test] Add error_inputs for nn.Conv2d module,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,174183,188327,,,2,e85f29595159d489d9bb923a6f540ce1943b153c93b9d623c3daeb1f7be93197
thread:issue:174371:pr:184864,pr #184864: Error on overlapping as_strided mutations in compile,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,174371,184864,,,1,19cf4b715f5788b161300cfd58c0c8759bf68d5ce2fd215ebe0fa185f0103666
thread:issue:175264:pr:184838,pr #184838: Fix Dynamo raw descriptor building,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,175264,184838,,,1,fb26c677f0e552043c8bf7dd0338fbac577b16e49e770fbdbe24ca9635acc914
thread:issue:175292:pr:184820,pr #184820: [dynamo] respect custom metaclass __call__ during class construction,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,175292,184820,,,3,7c22b1b0ba4ef099991d535c98fbbee6eb87ef0fcd47217c7ddccff188456bb3
thread:issue:175477:pr:186845,pr #186845: [make_fx] Decompose detach by default for higher-order grads,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,175477,186845,,,1,f66d9b538117b8966e85868afb3fac619b22680c9573398af59b58568609bd27
thread:issue:175858:pr:184803,pr #184803: [dynamo] Avoid key-set guard for dict setitem,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,175858,184803,,,1,82e95c0a8073b66e0ad2a0c57df9d1b0bd63bdfe0dec0508e6e1e860ddfc10ee
thread:issue:176178:pr:186849,pr #186849: Fix CuTeDSL grouped_mm descriptor init shape,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,176178,186849,,,1,1baefd012c49533979009a744b47d81abe994c3d3187baac14ba2a193808d0e4
thread:issue:176211:pr:184793,pr #184793: Fix no-tensor-aliasing recompile reason for stale sources,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,176211,184793,,,1,f66917eb5a58144422083718f36b77e059b2351e586f19f9fadf370596bf9075
thread:issue:178248:pr:184713,pr #184713: Support Dynamo class setattr for hook handles,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,178248,184713,,,1,7d201a5d1b299b43213631415dacda518723e34518c1e2def4ca4537b2eabecc
thread:issue:178317:pr:183866,pr #183866: Preserve native matmul lowp output precision,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,178317,183866,,,2,47b4f275e6c7e5bc33c21ae4a403c07570a3e48beb058ac54d6dada3e327785b
thread:issue:178379:pr:178298,pr #178298: torch.linalg.solve_triangular: support for eligible non-dense args,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,178379,178298,,,2,de20ec2d4f47d28f66f768c1c850e927c0dc6e99bf933cddd19385038a92b0de
thread:issue:180447:pr:183744,pr #183744: Update ACL dependencies for BF16 inner product,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,180447,183744,,,7,ed46c068e8fb3d0d883812999678c78b5ad3fbd171737737c4de350ace81e9a7
thread:issue:180807:pr:183731,pr #183731: [inductor] Validate fallback output dtype metadata,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,180807,183731,,,1,17e2907fdb4398114edb231274a6384a38ac1e18a52495ecaceb25bdfcf7aae5
thread:issue:181177:pr:184656,pr #184656: Fix CPU SDPA higher-order grad compile,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,181177,184656,,,2,a700ecc73d1fe606d22bfb830d1c28ece92b26507720538aea6adcf8be3f88bf
thread:issue:181347:pr:181823,pr #181823: Docs: update torch.multinomial num_samples requirement note for bette…,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,181347,181823,,,1,31359910f004accaeccec448833a13cfe3ab6fdc2ae222eb4374084a68128e56
thread:issue:181474:pr:189317,pr #189317: [doc] Add accelerator test reuse guide (#181474),PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,181474,189317,,,2,0648bf8c5ef071b196f7a4ed7aecbb62c9ef113d9e03337cbd45d921bd11ca0d
thread:issue:183015:pr:184595,pr #184595: Allow functorch grad fake propagation in Dynamo,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,183015,184595,,,1,ac2374b4795a2e77bc853a733e8ee0de840142e5bc69150daf69022920bdf792
thread:issue:183259:pr:183652,pr #183652: Fix dynamic slice lowering for generic views,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,183259,183652,,,3,f208e05dd84a3a467f3da247f63078d8c6ef01d90412150a74582d0fb148100d
thread:issue:183266:pr:184590,pr #184590: Fix uncached NestedTensor seqlens under compile,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,183266,184590,,,1,d4d2696ab48780d545f5cc19a914df98b365a2ce7b95aade1915bd5833231f2a
thread:issue:183901:pr:186913,pr #186913: Fix int64 value_expr arithmetic in Triton codegen,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,183901,186913,,,3,05b7cc174f59fbf84bdeac50b6f0c7d35c8f8c4b14a3ec4a7d40cd7e2a93dd80
thread:issue:184002:pr:184367,pr #184367: Fix AotAutograd aliased-output replay for traceable wrapper subclasses,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,184002,184367,,,2,9a31f7675dd7fff4a3f502df7cb32357d5ef443bd58779e71f88cea67f1c0190
thread:issue:185141:pr:186923,pr #186923: Graph break on custom op fake tensor failures,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,185141,186923,,,1,e327456fe677697e55cfcaaecdefcd8d7e6de83e158ed4f452016f7a8087769f
thread:issue:185200:pr:185231,pr #185231: [C++] Ensure wrapped tensors can be printed,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,185200,185231,,,1,b92593ff423b6239e6d9b0fede6f789836ab694acae68757cbfa473b97723cce
thread:issue:185382:pr:186246,pr #186246: Fix stale scalar value in addcdiv graph-break decomposition,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,185382,186246,,,1,ab97799e253df1d120d36652875586d4f867a295f232bfcc2fd43385c615ca42
thread:issue:185396:pr:185501,pr #185501: Fix Issue #185396 inductor complex64 triton signature,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,185396,185501,,,2,4528c7e636c0c4d66e8a078b46681c58e45d209dc32b886f9b14db8396103c6f
thread:issue:185480:pr:185878,pr #185878: Fix Inductor expm1 on float subnormal inputs,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,185480,185878,,,1,c9bedeb1484dc1a3e989b24d3c46f1d551d9f752db48fbfe20ca9f2e4aabd06e
thread:issue:185497:pr:185876,pr #185876: Prevent saving mutated grad inputs in AOT min-cut,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,185497,185876,,,1,dc2b1052c97935cd2852bfc4bdde28c9a462f581fff9a70685e1570dc80df233
thread:issue:185509:pr:185873,pr #185873: Preserve mutation renames after scheduler recompute,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,185509,185873,,,1,d4b3f19e8e28a44dd589d8d96fafb247a63837a934099a42f993a4fd7c1b842f
thread:issue:185510:pr:185871,pr #185871: Fix symbolic stride-order sorting without explicit ShapeEnv,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,185510,185871,,,5,be1292f961d3ec5bf393a83d512001d92137d909c1329cec6025b31333c4af9a
thread:issue:185589:pr:185866,pr #185866: Fix CPU Inductor numerics for BatchNorm erf normalize,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,185589,185866,,,1,1a199067f9be8cab2b0b585b665d5ce3d2c2c52b435c16f8cb6f833505528fd0
thread:issue:185814:pr:186656,pr #186656: Autotune XBLOCK for mix-order reductions,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,185814,186656,,,5,b6dea32a87b61fa5db1d0fcf2fac260440844bc3ec37e919f0fffcae1239a840
thread:issue:186028:pr:187614,pr #187614: Use sym_numel in linalg_cond string overload,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,186028,187614,,,1,e5b904b63eb7bb76249b985756db14319d9236221506777007562274b8badf25
thread:issue:186354:pr:186388,pr #186388: Avoid native eager router graph breaks under Dynamo,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,186354,186388,,,1,75666dc020784af8cf9dc3e8f01484368bbca2325fd0de832f7bd8929b4e28dd
thread:issue:186875:pr:186942,pr #186942: [inductor] Round fmod/remainder scalars for bf16/fp16 CPU,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,186875,186942,,,2,53045fe48303918409e4ed86c94b80e27253a4c7d730bf2c396ffe8466f61d7c
thread:issue:187041:pr:187057,pr #187057: Fix value-opaque graph output replay,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,187041,187057,,,1,517aef5deaa42ba37c230f7e11d64b15dbea17279aba014ccb79407ae8eb25eb
thread:issue:188048:pr:188049,pr #188049: Fix Inductor div floor nonfinite CUDA results,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,188048,188049,,,3,70ebda0d2b74a82b2a50cc18d039e5751a5d6cf5a07d593f44ece9ff0446f5b9
thread:issue:188128:pr:188784,pr #188784: Fix `torch.asarray` for buffer-protocol objects with no dtype,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,188128,188784,,,3,7b53be0979091593f13b67db0194282c65e7337fd8e2222a595554e134e3d64a
thread:issue:188131:pr:189009,pr #189009: [xpu][fix] Enable test_fallback_kernel_with_symexpr_output on XPU with correct head_dim,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,188131,189009,,,2,4a8caae05d0dacfeedff0c4ad1aef332e705be68d6307b0a7d7fb4c7f00d1f82
thread:issue:188150:pr:188913,pr #188913: [inductor] Add triton.cudagraph_max_recorded_sizes to cap dynamic shape re-recording,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,188150,188913,,,2,abf6fad10e26f2aa754a429f384b27d7713db03b5af3f2d212b849bff450675c
thread:issue:188355:pr:188641,pr #188641: Mark linalg.eigh as cudagraph-unsafe in Inductor,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,188355,188641,,,1,fea743cabb30bd04b3a86e6094321eb583f6ae01ab07ddd5df7090668eef1fea
thread:issue:188680:pr:189141,"pr #189141: [test] Add opt-in sign-sensitive special-value parity check (copysign, atan2, reciprocal)",PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,188680,189141,,,2,b40046f570426635363f16beaf68bd6996fba8ca95b6106b16b3200828da727e
thread:issue:188757:pr:188828,pr #188828: Skip conv_transpose3d redispatch test on Windows CI - flaky test,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,188757,188828,,,2,58be2f8de3de74aabf9d9c8df3573636cd840deb3033f6e0bc78e068624e4721
thread:issue:188773:pr:188783,pr #188783: [xpu][fix]Fix bmm_outer_product override to support XPU tensors,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,188773,188783,,,2,6f4e349ee390d6e0d3d2389b9135a8853d82592a96a239043e5832fef3eab91b
thread:issue:188790:pr:188980,pr #188980: Fix inconsistent accelerator guard in checkpoint staging test,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,188790,188980,,,2,cae9fc33226e7159a5a9f3037f7c8590578d1a1478261633f7b977947b50f63d
thread:issue:90923:pr:186557,pr #186557: Resolve scalar overloads before AOT compilation,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,90923,186557,,,1,c71590bdb2810cdad34b8c33ab9aaf355cd51dd8f78f44ec53288668e31435da
thread:issue:91468:pr:186554,pr #186554: Preserve hooks for dependent AOTAutograd outputs,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,91468,186554,,,1,80a5af5fd1245014afba33c329e8bf9ce0647f8e6a8eeff70977843bd576aafc
thread:issue:93367:pr:186551,pr #186551: Preserve symbolic tensor metadata in after-AOT repros,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,93367,186551,,,3,f34c85bc3ccd32b6b5aa3afb1100d411764b65cea6653251390c0221e687c089
thread:issue:93437:pr:186564,pr #186564: Skip uninitialized outputs during accuracy minification,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,93437,186564,,,1,05c4391b130580eee1f6cee4034b68ebb8d3acbf77d7460fcf998e358bc71aea
thread:issue:93501:pr:186553,pr #186553: [dynamo] Graph break on packed sequence lower ops,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,93501,186553,,,1,a00415148fb6dc270360c756646b6fce35e7a65e3120e7e738a775234a45a723
thread:issue:97078:pr:186549,pr #186549: Fix fake deepcopy for custom module mappings,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,97078,186549,,,1,236348c073460f63b93073bd8a494cfdea3f383275c9f93e156ebb2b9c1c3f7e
thread:issue:97750:pr:186546,pr #186546: Preserve original Dynamo minifier backend failures,PR blocked,73,Review active PRs and linked issues as one maintainer work thread.,2,0,97750,186546,,,1,aa4bd72bfc1c2911d576d0d824edce6a9e49c2b51d1a552caeac5b3e69bfb48a
thread:pr:184114:pr:189076,pr #184114: Add cpp wrapper fallback alignment assertions,PR blocked,73,PR blocked,2,0,,"184114, 189076",,,1,3ab658014b9b78b216ccd81265243786310579f7db813e5de3142f06a3edd86c
thread:pr:188978:pr:188979,"pr #188978: swap isinstance(t, FakeTensor) to is_fake",PR blocked,73,PR blocked,2,0,,"188978, 188979",,,4,1ee3d468c8dd93e0d043487182f89ba05568ee07100a598de7e9859702960206
thread:pr:169241,pr #169241: [XPU] Migrate 10 dynamo test cases for XPU,PR blocked,73,PR blocked,1,0,,169241,,,0,c77e8e5a17513674c7f62b0070ac0aadd4a409be133a66c1986a20de1049d3cb
thread:pr:169798,"pr #169798: [XPU] Migrate 2 test cases autograd, creation_ops for XPU",PR blocked,73,PR blocked,1,0,,169798,,,0,82e0daa3cb0f32cee0668063aa1547ef1a22e44c55750784e23160b3f2d4444d
thread:pr:169937,pr #169937: Fix vectorized float32 to uint8 conversion on AArch64,PR blocked,73,PR blocked,1,0,,169937,,,0,199bbd1dffc1e4ed3dc44b0c1e6f3f7361cf47d44e0caf4569eaa96d9a422117
thread:pr:170538,pr #170538: [XPU] Migrate test_testing for xpu,PR blocked,73,PR blocked,1,0,,170538,,,0,1e62cd5c08b7f8ce7070ea5d43234ce04beefe045975ce0ff24075c6316f995f
thread:pr:171617,pr #171617: [FSDP] Preserve container object identity in _recursive_to function,PR blocked,73,PR blocked,1,0,,171617,,,0,0d01710b8a14e741af8fd2c951f5353fd5e32eefc05f7c94615fc5c6ddbb7bda
thread:pr:174241,pr #174241: [MPS] Fix conv backward to preserve input memory format,PR blocked,73,PR blocked,1,0,,174241,,,0,70b4dbc1093fdf410e2eff5295f5aa80d6bbe8013d52137f2969adcfb7d50a03
thread:pr:175081,pr #175081: Add shim config includes for linting,PR blocked,73,PR blocked,1,0,,175081,,,0,d5009358b71e253a217b241a3102231baed6a92f55920035b37bc4669a1ad480
thread:pr:175082,pr #175082: [pytree] make `torch.utils.pytree` as public API,PR blocked,73,PR blocked,1,0,,175082,,,0,a78068b731402b6fbdbcd573f3d9b525198f1d852ade48bd1a41fd4948d31f81
thread:pr:175246,pr #175246: Fix NLLLoss backward failure for non-contiguous 4D inputs,PR blocked,73,PR blocked,1,0,,175246,,,0,d911ede8d69451f7e596c521b9094c0a62864ac4cb7a8887f620a49e71dbbda2
thread:pr:175526,pr #175526: while_loop vmap batching rule,PR blocked,73,PR blocked,1,0,,175526,,,0,d163f9655ae16250b1d16ebbc73589838009d30335ad7c783d7245f342024edd
thread:pr:175566,pr #175566: Fix test_extension_backend race condition,PR blocked,73,PR blocked,1,0,,175566,,,0,f39225217fd741f22fd02b9b35f18fc8788e57d0514684dab6b0c74d8948ecac
thread:pr:175947,pr #175947: [dynamo] Remove the dict over guarding on dict setitem,PR blocked,73,PR blocked,1,0,,175947,,,0,c84d1a1b76884588cf4dd28179aa7aef328885a6ed00f6c7410aa76f51dd0478
thread:pr:176044,pr #176044: [opaque object] Disallow creation of reference-type opaque objects under custom ops in torch.compile,PR blocked,73,PR blocked,1,0,,176044,,,0,847b739b013c8e79b8f67432ee7fbe89c62c25a08bc7c8ffe351aa31619081e9
thread:pr:176084,pr #176084: ci: Cleanup dead centos code,PR blocked,73,PR blocked,1,0,,176084,,,0,33242b2db4446c4b8233b94450b874718ac97f6a79f391243868e8b230cca5db
thread:pr:176295,pr #176295: [inductor] Fix cudagraph output->input feedback crashes,PR blocked,73,PR blocked,1,0,,176295,,,0,f6de78bf82371210586f51fab8a089914752b357710e37b930614ab7a0cdc7da
thread:pr:176670,pr #176670: [CUDAGraph] Enable `graph_capture_record_stream_reuse` with conditional nodes,PR blocked,73,PR blocked,1,0,,176670,,,0,43124e4c63c485295bcb817ce4414e7ba32cd644ab04059fce11d82a2033d067
thread:pr:177170,pr #177170: [FSDP] Fix alias annotation for fsdp::all_gather_copy_in,PR blocked,73,PR blocked,1,0,,177170,,,0,781acb1eb236ecb22bf56013cea0fa3734c7e50bd1a37990cdb2a852f34a80bc
thread:pr:177546,pr #177546: [inductor] Fix pad_mm leaking padded strides to user-visible outputs,PR blocked,73,PR blocked,1,0,,177546,,,0,5af69c6380180b4fc61915a3b1e8ed591d92f71de45f100586f645bf848e816a
thread:pr:178106,pr #178106: Support Optional[list[torch.Tensor]] in infer_schema,PR blocked,73,PR blocked,1,0,,178106,,,0,cf0ba284a08150dfb5e9b3e7e71d6e7bd639cfa6c8015c92aa0b4036e91231ce
thread:pr:178209,pr #178209: Fix CUDA batch_norm eval returning populated save_mean/save_invstd,PR blocked,73,PR blocked,1,0,,178209,,,0,bdd0d89f0803262d5f7b3e87ffaca0a2a12c965739de3078d99a55cc160a387e
thread:pr:178584,pr #178584: Remove unused expected failure decorator from FakeTensorOperatorInvariants test and clean up condition in Module class,PR blocked,73,PR blocked,1,0,,178584,,,0,44c3d741ea166619e44d46d641b6b01056e5cce6c9488754979c14a48feb3192
thread:pr:178664,pr #178664: Implement igamma/igammac backward w.r.t. first argument,PR blocked,73,PR blocked,1,0,,178664,,,0,05815049e8a72ec080dc018a92d32f6997153937f985219ebe3e8f2326004460
thread:pr:178805,pr #178805: [openreg] Add comprehensive test documentation for OpenReg backend test suite,PR blocked,73,PR blocked,1,0,,178805,,,0,2f95e8105aafacfdafbcc51b675e45841608af80caf6d7ceab29e6130f245afc
thread:pr:179205,pr #179205: [functorch] Fix vmap/functionalize under inference_mode,PR blocked,73,PR blocked,1,0,,179205,,,0,e372be96c639c05f08bd905b546f13c6fc801471072dc92f06d1a60daeb81c1d
thread:pr:179235,pr #179235: [Inductor] Add _debug_assert_fused option to foreach_map,PR blocked,73,PR blocked,1,0,,179235,,,0,4d017349e737ac296b4f44240450d51c140a8397f3515f074382e56eba751c2a
thread:pr:179376,pr #179376: fix(export): add SDPA decomposition for contiguous output,PR blocked,73,PR blocked,1,0,,179376,,,0,f866113b69266075a096d7102bd5ee24de2da547d291fa5d60ba1803dcb2ca9c
thread:pr:179404,pr #179404: [DTensor] Replace __module__ hacks with @exposed_in decorators,PR blocked,73,PR blocked,1,0,,179404,,,0,4b3b06f7648c1f6340e80de22834e3ada1b23fce5859ec1c07bf04e28e0fac25
thread:pr:179526,pr #179526: inductor: handle matmuls gracefully in foreach_map,PR blocked,73,PR blocked,1,0,,179526,,,0,0a4eeb324d2eb2cd80451fb718ae6cfdd8211ecaaff1080c590f573a64e021b4
thread:pr:179584,pr #179584: fix inference_mode getitem compile error,PR blocked,73,PR blocked,1,0,,179584,,,0,da856e5cd4ddf1c7dac858da06709dc773fca0366f8f2562f0f32bd001e50b14
thread:pr:179592,pr #179592: Fix MPS SDPA correctness when score matrix exceeds 2^32 elements,PR blocked,73,PR blocked,1,0,,179592,,,0,6b4c23e9ee1446f36396d6a81a8cb1f86266d8c5d543196f88b902fffc8f1e17
thread:pr:179619,pr #179619: [PowerPC] Guard deviceTypeFromActivity with USE_KINETO check,PR blocked,73,PR blocked,1,0,,179619,,,0,cfc883c387e6cdf1571fc5fd4fba7d7bce47bf89a803c7589abea96ad3625349
thread:pr:179627,pr #179627: Fix all masked Transformer eager/compile inconsistency,PR blocked,73,PR blocked,1,0,,179627,,,0,b18b84cccf91aec2ac0b419991e82d2388016ec43129bd700bf586fbeff0db3d
thread:pr:179658,pr #179658: Add OpaqueGenerator to allow make_fx trace Generator,PR blocked,73,PR blocked,1,0,,179658,,,0,b14911feac977bca4015c1131db3ceb6901f9ad91f1752acdb83e30c3e3c7572
thread:pr:179882,pr #179882: [Devmate Tasks][Butterfly] Fix flaky fbcode//caffe2/test/distributed/tensor:dtensor_ops - test_single_dim_strategy_i0_cpu_float (#179882),PR blocked,73,PR blocked,1,0,,179882,,,0,8759526e8e7ecb6083341595015a8edca665a0bdf13305cbb43ea03b14bdf37c
thread:pr:180162,pr #180162: Add CPU tensor validation to FBGEMM operators to prevent segfault,PR blocked,73,PR blocked,1,0,,180162,,,0,7f676b7906fa028440044d2adb0ef3f1338819d4ea23e4504cfd8662592e8cf2
thread:pr:180315,pr #180315: Add _Float16 support for Tensor::item() C++ API,PR blocked,73,PR blocked,1,0,,180315,,,0,9987445e688341183f65a6382dccdd08685f06598109af285f048abaf972d6e1
thread:pr:180319,pr #180319: [ROCm] Handle profiler trace format differences in combo kernel tests,PR blocked,73,PR blocked,1,0,,180319,,,0,ee601c62c119cf0569571eea169fa48a28e8548c8358074bb1f09ff6414a8772
thread:pr:180557,pr #180557: [docs] Improve documentation of normalization layers,PR blocked,73,PR blocked,1,0,,180557,,,0,f5e039fc4d807c1dcbc7479c6067fe67bc74cbff974992c87ba2d1dbb5808e75
thread:pr:180561,pr #180561: [optim] Support ndim >= 2 in Muon and add param_groups_for_muon helper,PR blocked,73,PR blocked,1,0,,180561,,,0,b884c49a4055d5bcf5775f5fd6a51062b17f0018e8ed1b7dc0e2feb728683696
thread:pr:180570,pr #180570: Fix torch.cond crash with FunctionalTensor constants (#180354),PR blocked,73,PR blocked,1,0,,180570,,,0,1c8add5b79c5ff4fc49579aa82bff3c818a75543e9f2ea24f377bb9d52d3e58a
thread:pr:180737,pr #180737: [inductor] Support pin_memory for torch.rand and torch.tensor,PR blocked,73,PR blocked,1,0,,180737,,,0,bd55bc8a5f2240c90578d6e2c658cd04609a54f3962ac637db7f16dab03076ea
thread:pr:180782,pr #180782: docs: remove unsupported out= from torch.linalg._powsum docstring,PR blocked,73,PR blocked,1,0,,180782,,,0,771945618d21a4b549ff139b29b64cb772aa7c50ea2a338c99864c81dd068152
thread:pr:180783,"pr #180783: docs: fix index_add_ `index` description (points into self, not source)",PR blocked,73,PR blocked,1,0,,180783,,,0,f6737c4a7c7f9b7d536383f6704048a287a9a8fe0d6cac36cbb925b4737c3d67
thread:pr:180810,pr #180810: [inductor][static launcher] validate pointer arg dtypes before launch,PR blocked,73,PR blocked,1,0,,180810,,,0,21960ceba8707ac2d25f2dbc57bca65827f61b30efbad8e70b4da2d9ff1fa93e
thread:pr:180971,"pr #180971: [inductor] Using is_gpu() instead of hardcoded ""cuda"" check",PR blocked,73,PR blocked,1,0,,180971,,,0,a602b8a1133154e80fd25760b05e5b09e8b6c0316ebd9123a3f4b1e7a19b940b
thread:pr:181074,pr #181074: [inductor][coordesc] extract get_neighbor_configs and get_all_direction_configs (#181074),PR blocked,73,PR blocked,1,0,,181074,,,0,3466c26606605bd97d4d06cffde247ca2d5f6cf0b9fc53a64a295073314f5165
thread:pr:181252,pr #181252: [Codemod][MigrateInternmcUrls] fbcode/caffe2 (#181252),PR blocked,73,PR blocked,1,0,,181252,,,0,b8491815c924dedf04d10fb757dd502bed813f7cd02035020f652c01effb357a
thread:pr:181475,pr #181475: Add `tensor` and `from_numpy` to `torch.__all__`,PR blocked,73,PR blocked,1,0,,181475,,,0,f7d3ea47824b98533d02e93099aab8c80534f6e7a9a278e444271a2dd72dc517
thread:pr:181481,pr #181481: [Codemod][MigrateInternmcUrls] fbcode/caffe2 [A] (#181481),PR blocked,73,PR blocked,1,0,,181481,,,0,d38b622f0cfeb37c344c7875bdacbf2e7a6e2d7232b6cdff21489982b9977a1c
thread:pr:181546,pr #181546: Fix AssertionError: failed to set ranges (#181546),PR blocked,73,PR blocked,1,0,,181546,,,0,7685c02e239c3731310866753e9be3546de096d4806b536d04da91d6615b8942
thread:pr:181559,pr #181559: [XPU] Enable _flash_attention_forward/_flash_attention_backward,PR blocked,73,PR blocked,1,0,,181559,,,0,75d28848da8540fe596424eb1f4ee94c965f2290a434d3443214fcbd638f9bde
thread:pr:181561,pr #181561: [AI Codemod][DevmateCppModernizeFindToContains] fbcode/caffe2/torch/csrc/jit/backends/xnnpack/xnnpack_graph_builder.cpp (#181561),PR blocked,73,PR blocked,1,0,,181561,,,0,3f7c34eb35cfb696c9f2de25f828f8c9fe5362c6a195335b1545b82d82080263
thread:pr:181571,pr #181571: Generalize test_pooling.py to be device-agnostic and enable it on XPU,PR blocked,73,PR blocked,1,0,,181571,,,0,ffe9bb96e3245a99721c2c4f09fc416e9d354cd91c3fcb61a107a411a3ef3791
thread:pr:181582,pr #181582: [AI Codemod][RlAutotuneFbsourceWearablesMemory] torch::jit::mobile::Function::append_instruction (function.cpp) (#181582),PR blocked,73,PR blocked,1,0,,181582,,,0,e1e2f59f50ed4099ec4601807f2be9642b42dc8a3631bb48be9263d05758c240
thread:pr:181915,pr #181915: Support returning None from torch.cond() true_fn and false_fn,PR blocked,73,PR blocked,1,0,,181915,,,0,fd3818ed42a4f4fc3182cae12587c9efac3790435d05ff048107b3643b300b9e
thread:pr:182068,pr #182068: Make custom op aliasing check warn (not error) in CI,PR blocked,73,PR blocked,1,0,,182068,,,0,18bb58582086b910d826ceaeb9db1929e25e759ae03d987cd08ab06764befc07
thread:pr:182092,pr #182092: Allow torch.meshgrid() with empty input to return empty tuple,PR blocked,73,PR blocked,1,0,,182092,,,0,bf16ba6a699beea6e73894fa4a84c9360b814cf3d0d547e3eefb20e54194947b
thread:pr:182216,pr #182216: Add Half/BFloat16 copysign overloads to fix NaN sign bit loss on CUDA,PR blocked,73,PR blocked,1,0,,182216,,,0,8d49fdd7ad5d5d3e18db96881ff0ec2b862f76a628d69818cb674f9e2d8aa9f4
thread:pr:182342,pr #182342: [Inductor] Deal with choices that fail to compile in async pipelined autotuning path (#182342),PR blocked,73,PR blocked,1,0,,182342,,,0,11888adba4f207a2f1b07233b4c447650092405d0a7aad3af128cdaaaf9e27cf
thread:pr:182414,pr #182414: fix(decomp): cast padding to int in replication/reflection pad decomposition,PR blocked,73,PR blocked,1,0,,182414,,,0,311a720a9dc7600e29cce11654720e2f83f4a6416ff1c34993bf025ec49344a1
thread:pr:182431,pr #182431: [Inductor] Elide redundant .to(tl.float32) in Triton codegen when value already has target dtype,PR blocked,73,PR blocked,1,0,,182431,,,0,9a77d267bf9638b9394bee084817f7239a094a17fbf0bd1192fef7bcc3cee84d
thread:pr:183509,pr #183509: [distributed] Add fused scaled_mm_v2 + reduce_scatter for NVFP4 support,PR blocked,73,PR blocked,1,0,,183509,,,0,662af377657f9dd128ea4541302cb45fd59ffc2233ee305b0727bb35fcdcb0f1
thread:pr:183525,pr #183525: [MPS] truncate fractional start/end to integer type in linspace when …,PR blocked,73,PR blocked,1,0,,183525,,,0,104323dc1f8fe65537704d5f18b282d7b63dcb1982e4fcec11a9c4985f2319f9
thread:pr:183668,pr #183668: [ROCm] fix flaky test_honor_sm_carveout,PR blocked,73,PR blocked,1,0,,183668,,,0,f04bfe2e959d73d59d6c5163478eca575a25db69ec4534b8fa438c2a62b820f5
thread:pr:183681,pr #183681: Fix Triton compile-time autotune fallback,PR blocked,73,PR blocked,1,0,,183681,,,1,e64f20afee753370400ec664c3aa0c34670c9a7dc817e68192c3b5cf0d1f5cf7
thread:pr:183854,pr #183854: Filter C++-wrapper-incompatible extern autotune choices,PR blocked,73,PR blocked,1,0,,183854,,,2,bcaaf27cf601799ec084b9e1bbcc1aca38e1789f35940b9a521e32572e84a9b2
thread:pr:184133,pr #184133: [aot_autograd] Reduce runtime_wrapper Python overhead: 4 micro-optimizations,PR blocked,73,PR blocked,1,0,,184133,,,0,3dd1b64c6f2bac765f36e59943b919675a56f392db2c33fc5096fdb5086548ea
thread:pr:184143,pr #184143: Bound Inductor op count for high-order diff,PR blocked,73,PR blocked,1,0,,184143,,,0,7c4f28fcf784b058af8ebe12dbe5124a4a9d9a2802db7fdfdb9be1bb550b7870
thread:pr:184155,pr #184155: Fix Inductor bool vector mask casts,PR blocked,73,PR blocked,1,0,,184155,,,0,4266c9a198939fca127dab03669589e990ee63372a00f0091a5a6a5300f998c3
thread:pr:184164,pr #184164: Fix while_loop compile with empty carry and mutation-only state,PR blocked,73,PR blocked,1,0,,184164,,,0,6cc3ffdbfb8fa1f1d6b0fdb998dbfd357e7455277834fa94c675705b1e9768b5
thread:pr:184279,pr #184279: Avoid MultiOutput for single-output fallback kernels,PR blocked,73,PR blocked,1,0,,184279,,,1,b37355b59dbdee92a488b25ec8469eb6cc1551966358738a6e834edd7282ed00
thread:pr:184305,pr #184305: Fix zero-bias SDPA flash selection,PR blocked,73,PR blocked,1,0,,184305,,,0,908c1f682b283d63b22154e92ca6764e9bafa4520aa56b136d949fbe634e838d
thread:pr:184334,pr #184334: [CUDA] Enable bf16 grouped_mm CUTLASS fast path on SM12x and Windows,PR blocked,73,PR blocked,1,0,,184334,,,0,959199ed0f0f172f1311639e4830c025fcb88988e3586e6925828ebf60cff227
thread:pr:184409,pr #184409: Fix local_map deferred inlining to functionalize mutable operations,PR blocked,73,PR blocked,1,0,,184409,,,0,246c7c3d8c639b88c287295395d820d0b8908f74263dae34aca6e404b9070a72
thread:pr:184420,"pr #184420: Fix pow(-0.0, +/-0.5) sign on sqrt/rsqrt shortcut",PR blocked,73,PR blocked,1,0,,184420,,,0,8af72262f2c14327ad448ef39e20a35e8f13af9910324753dcd3b20dedad91b0
thread:pr:184459,pr #184459: Decompose supported aten._trilinear in Inductor,PR blocked,73,PR blocked,1,0,,184459,,,0,90f2a8fb9e9f17c4f47a960fa478625267f38967c99d20530ba1dd9285680f80
thread:pr:184631,pr #184631: [MPS] Skip strided NDArray path in _mps_linear_nograph when inputs are contiguous,PR blocked,73,PR blocked,1,0,,184631,,,0,87472dce7eafb62a26f16e2022f84811d401400894ca8ca2b7765f8c50eb281e
thread:pr:184696,pr #184696: Normalize export slice default bounds from getitem,PR blocked,73,PR blocked,1,0,,184696,,,0,44a55328be7a435350c45a726fcc9d719de327cdfd4584811dab5d6e5a0d7279
thread:pr:184704,pr #184704: Capturable param in optimizers restored using custom vts and cleanup hooks,PR blocked,73,PR blocked,1,0,,184704,,,0,5b974c686bbd34a0a36d727e0890ccd776453d49268be66ccc986aff59a8560e
thread:pr:184705,pr #184705: [dynamo] Preserve exception chaining in InternalTorchDynamoError rewrapping,PR blocked,73,PR blocked,1,0,,184705,,,0,552746d6eea9ce39da3f81082512b73cfaa84e515f8eee4b3c702e63beb79063
thread:pr:184822,pr #184822: Add Sphinx redirects for renamed autosummary Stream/stream pages,PR blocked,73,PR blocked,1,0,,184822,,,0,790f2c963413269229668b10ae5eae24ddbe59abec4ceab28ef0a50792358043
thread:pr:184968,pr #184968: Fix autocast state propagation for normalization layers in functorch batched autograd,PR blocked,73,PR blocked,1,0,,184968,,,0,00db59440c062c607d0af56c210ee4ac65951bc1d40349060e3294870987c6b0
thread:pr:184986,pr #184986: Handle unbatched input for non-zeros padding modes in Conv1d/2d/3d,PR blocked,73,PR blocked,1,0,,184986,,,0,b61d4108ff7eceedd066fd156795da2daaf067d93373a6b22fa0839ea265b276
thread:pr:185077,pr #185077: Avoid folding mutated buffer storages,PR blocked,73,PR blocked,1,0,,185077,,,0,7abaa8dac7a2d29e9a084d13ab729e08e4209fc445909e64d22f96cd13111992
thread:pr:185122,pr #185122: Validate parameter/buffer storage bounds at export time,PR blocked,73,PR blocked,1,0,,185122,,,0,2ce314a4e4aba5b08ceffa6a0e7cfde01b996a86a3a015987fc12002c70218b3
thread:pr:185177,pr #185177: Fix DTensor nll_loss weighted mean bug (closes #185167),PR blocked,73,PR blocked,1,0,,185177,,,0,c8736e37851797637bd86b857b40fcf61f41fb3fbaab9a13a0749a1ea9612b2d
thread:pr:185227,pr #185227: Error out when getting symmetric memory device pointers for NCCL < 2.28,PR blocked,73,PR blocked,1,0,,185227,,,0,2aaf31f212264bc373617228b8302f3b28de552a0fb7409d3b65404081ceea6f
thread:pr:185316,pr #185316: [dynamo] Resolve autograd.Function static attrs in getattr,PR blocked,73,PR blocked,1,0,,185316,,,0,339143036a89741e9ad94bf20cfd604931922244b298d0cac83a2b11a9d19ecf
thread:pr:185358,pr #185358: [PrivateUse1] Add dedicated LSTM dispatch branch for PU1 backends.,PR blocked,73,PR blocked,1,0,,185358,,,0,a5fedb8a718438cb3f783c49f201c4d632fdc73bea35e69e17a9b3618bf3becd
thread:pr:185415,pr #185415: [ROCm] Fix LayerNorm backward kernel for AMD Strix Halo GPUs,PR blocked,73,PR blocked,1,0,,185415,,,0,3fa2d6a6a657a2b88c4546b13e6f04e2e45014d3ed4f0e6c111790b3bfda75d6
thread:pr:185518,pr #185518: [inductor] Add probability checks for Bernoulli lowering,PR blocked,73,PR blocked,1,0,,185518,,,0,94f6fb42605c6c838072bca6fc99f9ff7801b78a5486370f6913dc18ee2cf3d0
thread:pr:185578,pr #185578: Fix addmm max-autotune SliceView bias guard,PR blocked,73,PR blocked,1,0,,185578,,,0,8b5a15040aa2ad566a9a4bc8fbbcff2959dc58c81f1cd7d3d8c38c94d707d8f5
thread:pr:185579,pr #185579: Improve error message in linalg.eigh for large matrices,PR blocked,73,PR blocked,1,0,,185579,,,0,18eb4a830470a7de938f2464b6223776a3c159148c70624c91bf66f0262351bc
thread:pr:185661,pr #185661: [fx] Fix pytree codegen for tuple/dict positional inputs,PR blocked,73,PR blocked,1,0,,185661,,,0,156994a1d2832b26028711d5b6eeab8ffb48764c05b47797e23402d5aee28134
thread:pr:185727,"pr #185727: Document bernoulli(input, p) overload in torch docs",PR blocked,73,PR blocked,1,0,,185727,,,0,2d75bf3abb916f0709c1a17babb626ff83ce1a7473cbc33c73e33c355505287a
thread:pr:185738,pr #185738: Make HOP alias and mutation support opt-in,PR blocked,73,PR blocked,1,0,,185738,,,0,a258216ce9b2a69f35e7bf262d88ba83da1b94462dfe8e1ad3e6fbd4202e47fe
thread:pr:185791,pr #185791: Fix cond branch shape guards leaking to export constraints,PR blocked,73,PR blocked,1,0,,185791,,,0,aae9b88f4bc9162442869a00a79154cea9ba0ca382ad109d3446df0e8324b8ee
thread:pr:185834,pr #185834: [inductor] Preserve signed zero in Triton neg codegen,PR blocked,73,PR blocked,1,0,,185834,,,0,2f6eefa99e991c18d282f2725b6344530be2eecbc277bba28329d5b97b4976d9
thread:pr:185838,pr #185838: [inductor] Propagate subgraph tensor constants to the root graph,PR blocked,73,PR blocked,1,0,,185838,,,0,24fa4937739421c390f6399f30e6af665b0c5981aa4d97c8098cf8f1fda517fd
thread:pr:185842,pr #185842: [decomp] Fix multilabel_margin_loss for padded targets,PR blocked,73,PR blocked,1,0,,185842,,,0,0d820aaa45184267748bd2f861d9316e972331b08372253bbdc9fe652f353555
thread:pr:185851,pr #185851: [autograd] Fall back to tensor backward for symbolic pow exponent,PR blocked,73,PR blocked,1,0,,185851,,,0,1a44d346a531acebbeef6893d03d5fe00ca6daa8abc014f568fca3dd90980321
thread:pr:185911,pr #185911: [dynamo] Keep relational guards out of dict-tag skips,PR blocked,73,PR blocked,1,0,,185911,,,0,8c91dd1de65f814d9bcc55e3c146ba3fc25db642a9171d457ac3774b9c8f3ce1
thread:pr:185973,pr #185973: [Accelerator] Enable Default Stream API,PR blocked,73,PR blocked,1,0,,185973,,,0,44e776fc86fdcfcffc143a156821c4defdc50cff8b17ff4033df0169cd1c2010
thread:pr:185988,pr #185988: Fix FlexAttention scalar captured grad backward,PR blocked,73,PR blocked,1,0,,185988,,,0,09e3be92d204a222b903f1038aabe71487436344026b591fc665d238941bbb23
thread:pr:186049,pr #186049: Fix bool argmax meta validation for Inductor,PR blocked,73,PR blocked,1,0,,186049,,,0,b1ad7c84c2906f572d10b9c230410f96e0738cba245ed9ca07733c84f306ff6b
thread:pr:186061,pr #186061: Fixing torch.pow crash with integral dtypes on CUDA by gating reciprocal/sqrt/rsqrt shortcuts,PR blocked,73,PR blocked,1,0,,186061,,,0,15738a0aaf5f1226c509124be7b187ff374aeabc8a2aff70e2a080fc2bfa70ac
thread:pr:186085,pr #186085: Fix transient dispatch-mode skips in Dynamo,PR blocked,73,PR blocked,1,0,,186085,,,0,5e6210c9a3934d01a7a0f1802a5c29d2dc162b829b720d0c3754bbbcc4c449fb
thread:pr:186127,pr #186127: [AI Codemod][PerfAICT-General] fbcode/caffe2/torch/csrc/jit/serialization/pickler.cpp (#186127),PR blocked,73,PR blocked,1,0,,186127,,,0,ab32f4ceb2610ad2a13224aa510490f392bbc568e9eb4b04ceadfacc29309494
thread:pr:186178,pr #186178: Fix matmul folding when an inner folded dim has size 1,PR blocked,73,PR blocked,1,0,,186178,,,0,5d81ea67a9dbdc1c90e4572885ea2d867f81d227b8995c2e6b712d7fba25be5d
thread:pr:186294,pr #186294: [dynamo] Proxy OptimizedModule __class__ to wrapped module,PR blocked,73,PR blocked,1,0,,186294,,,0,377375551cb5895a8e358170ab48446de83b499a08dcc7a40ad99a047306f6c6
thread:pr:186306,pr #186306: Support tensor torch._assert in export,PR blocked,73,PR blocked,1,0,,186306,,,0,bc2f5ae4908f3877df5512d46af284881aa95c0ce8e6a405756fc9cce4b012f4
thread:pr:186319,pr #186319: Avoid full simplify in vector-range index analysis,PR blocked,73,PR blocked,1,0,,186319,,,0,c786825d2e4a75b0b5acfb7fe24957a5f5d661f1b675c4f7066a1aa7aa026da1
thread:pr:186330,pr #186330: Fix compiled autograd.Function manual vmap rules,PR blocked,73,PR blocked,1,0,,186330,,,0,e156439598db0d82b763f714ddb4ce4dbdf895c3bea0d169ab6e7968b964eb3a
thread:pr:186336,pr #186336: Support multiword CXX commands in Inductor cpp_builder,PR blocked,73,PR blocked,1,0,,186336,,,0,378252b85415614bbff9cefe2e437967bb0805e0cf532922c7226c27505e410d
thread:pr:186479,pr #186479: [dynamo] Respect mark_dynamic on Parameter inputs,PR blocked,73,PR blocked,1,0,,186479,,,0,82772565f95c2c9c2b619cb2a1d0bb223fa0fee7dae003ac23f3f1a25bc51c34
thread:pr:186532,pr #186532: Fix index_add decomposition bounds checks,PR blocked,73,PR blocked,1,0,,186532,,,0,c3443b8379191d17af95f8342f5241e965e3a51f55e3300275517751da6f03bb
thread:pr:186573,pr #186573: [Inductor] Stabilize checkpointed reduction autotuning,PR blocked,73,PR blocked,1,0,,186573,,,0,b76361c41fa21c7eb812ae170870c033780a2c2040e881c6014bde0f297cf2cb
thread:pr:186619,pr #186619: [pytorch] Fix Inductor missing wait on AsyncCollectiveTensor nested inside DTensor,PR blocked,73,PR blocked,1,0,,186619,,,0,e14d77955f8104d55c901ed0940e46f2959af6da60e35c74c2a606c557b94ce9
thread:pr:186779,pr #186779: Error on unsupported batch norm third derivatives,PR blocked,73,PR blocked,1,0,,186779,,,0,5d2c99d3c23db7f9057ec15a200b23f96c99195890f629a8b5d9c8505df340e8
thread:pr:186874,pr #186874: Fix Dynamo resume lifetime for _EnableTorchFunction,PR blocked,73,PR blocked,1,0,,186874,,,0,50e205c14b18ad2af57072d1b7b6ad123a6ad4ac6b45ecc965837dccef743f3c
thread:pr:187032,pr #187032: [dynamo] Support member access on protobuf enums,PR blocked,73,PR blocked,1,0,,187032,,,0,184a0584f2e4c64445133a6330289681be4901684b3dbfea43ec15c53c80008b
thread:pr:187169,pr #187169: Support weights_only and safetensors format in torch.export.save and …,PR blocked,73,PR blocked,1,0,,187169,,,0,9fa6e6e400234763bbd244a892393eabd1036c6c339145c15bcd0be43e13228b
thread:pr:187207,pr #187207: Use abs() in lp_pool to avoid NaN for non-integer norm_type,PR blocked,73,PR blocked,1,0,,187207,,,0,5c320a9db95b60c7f416cd5ff4cde6a97e731a4f11f3a85adaccdab1c0eb0cf6
thread:pr:187250,pr #187250: Support NumPy 2 __array__ copy kwarg and __array_wrap__ context on Tensor,PR blocked,73,PR blocked,1,0,,187250,,,0,6eb7e53184bdcdaff098bef2fcbdf5a408879ed32b64f5115e3bd714a2a3f0fb
thread:pr:187368,pr #187368: [inductor] add input validation to normal and bernoulli decompositions,PR blocked,73,PR blocked,1,0,,187368,,,0,56058ddc3d2bfeee218ae5ffb6d07eeb12046938d48dceee5fb5653eae84ac7e
thread:pr:187405,pr #187405: [dtensor] Propagate `_scaled_mm` tensor meta without eager scale checks,PR blocked,73,PR blocked,1,0,,187405,,,0,c1c3b32675262e5d7091be1786cc237757c0da6707f1fc4a7f24891af8ebf08f
thread:pr:187596,pr #187596: Fall back on AOTAutograd version-check capture errors,PR blocked,73,PR blocked,1,0,,187596,,,0,9661c57fb25905f5e7ae625b5320fa42061644fd7da1620a879c799e913fca4c
thread:pr:187653,pr #187653: Add NanDetectMode for forward-pass NaN/Inf detection,PR blocked,73,PR blocked,1,0,,187653,,,0,25e9a894630fb9238fba389109855f8a77ae5169a5b9f1be20265b183169506e
thread:pr:187661,pr #187661: Return empty result for FFT with empty batch on MKL backend,PR blocked,73,PR blocked,1,0,,187661,,,0,d3cfb8efa9d2ccd73be7d66a0f86e2d821e0400474414ca506ac90a77a68d732
thread:pr:187664,pr #187664: [MPS] Fix `baddbmm` operation to handle zero-sized dimensions correctly,PR blocked,73,PR blocked,1,0,,187664,,,0,13667ff0525a9d0539ffec865ae4440d47b5605a88e6588edf1204b931960dcf
thread:pr:187714,pr #187714: test: Command to reproduce error is incorrect,PR blocked,73,PR blocked,1,0,,187714,,,0,659f4ff057f87e9ab29fdb9dcbd5ba24159d61f645a1633774db50be8e489f4e
thread:pr:187726,pr #187726: Expose public API for torch.cuda.clear_cublas_workspaces,PR blocked,73,PR blocked,1,0,,187726,,,0,397d7ac3a9a2dea78b0a59fa098ebe6f62eff9a2ef950e7a4ebfdda3be326b2e
thread:pr:187782,pr #187782: [dynamo] Skip guard creation for unused function inputs,PR blocked,73,PR blocked,1,0,,187782,,,0,be66830060a88686aea79ff9ad302b6c9f0e1d479531c59e638cef9781715ff1
thread:pr:187786,pr #187786: Support int step tensors in foreach optimizer implementations,PR blocked,73,PR blocked,1,0,,187786,,,0,27011417d71fa7639bd9d877f2cd92ece9c1b02871f79ef7c09c30d41322f6d9
thread:pr:187871,pr #187871: Fix scaled_dot_product_attention: explicitly check for both attn_mask…,PR blocked,73,PR blocked,1,0,,187871,,,0,c7f5cc5d814c746f27c4073abbc5b8d44c5d5c58d89c57a27f5f025f0c5ae007
thread:pr:187927,pr #187927: docs: document FlopCounterMode semantics,PR blocked,73,PR blocked,1,0,,187927,,,0,1340131ace5536e1e018c61e55bab23bf7042c6f82b841060656ea873e6b3b5b
thread:pr:188286,pr #188286: Fix out= variant recompilations for CompositeImplicitAutograd linalg ops,PR blocked,73,PR blocked,1,0,,188286,,,0,59d10d1ef24e6caae5c26956c8f7575a435824a33f740b14c358baa3c6943841
thread:pr:188291,"pr #188291: Fix dims docstring for torch.flip and torch.rot90 (axes, element type)",PR blocked,73,PR blocked,1,0,,188291,,,0,5c625409cfb1ecdcced9693997c70839f9a11046282baf3f1ab13899ab087131
thread:pr:188314,pr #188314: Fix tutorial link for GPU Direct Storage,PR blocked,73,PR blocked,1,0,,188314,,,0,673b56c3ad6ce6fdf34453f50a8865c7ed2dd47f6ae9bd91dd02f2a0bd035dca
thread:pr:188320,pr #188320: [c10d] Integrate ncclGather into PT,PR blocked,73,PR blocked,1,0,,188320,,,0,a0eb59d40041013de28f4ebf52b3027c0c3abb0ac08f62d5721fa2460d153c53
thread:pr:188328,pr #188328: [meta] Add output_padding and bias validation to transposed convolution meta path,PR blocked,73,PR blocked,1,0,,188328,,,0,0f525cfdc61f068d0221b06020273374a82f7989f1a89b29b40ffa728e6a0c45
thread:pr:188337,pr #188337: [inductor] Handle non-comparable arguments in index_propagation Min/Max,PR blocked,73,PR blocked,1,0,,188337,,,0,07e17adcb741523d8c1c8dd930d7afc39b4f8eb52b0415a5852d04adb4aa3e45
thread:pr:188357,pr #188357: [ROCm] Add MIOpen convolution ops to FORBIDDEN_CUDAGRAPH_OPS for cudagraph safety,PR blocked,73,PR blocked,1,0,,188357,,,0,280dc022fb0a16abde638adee652d1b97edcb6c448ca358cdb1fbdfd16df09d4
thread:pr:188368,pr #188368: [Inductor] Fix ValueRangeError in Inductor for zero-iteration loops,PR blocked,73,PR blocked,1,0,,188368,,,0,51012cc04b761581c7f7ef78bf9811616580cc2e826d6bcc8e45488c45eafac6
thread:pr:188380,pr #188380: docs: fix broken gpu_direct_storage tutorial URL in serialization notes,PR blocked,73,PR blocked,1,0,,188380,,,0,1e95300096c85a4cacf5be49b50ccd456ca3b0197bd699428960f58209c843b9
thread:pr:188394,pr #188394: Speed up CPU quantile/nanquantile with partial selection,PR blocked,73,PR blocked,1,0,,188394,,,0,b8ac74f6e093675fb6ddfe97349bc618cf0c46ea28d6490efef6f808d34c5a9c
thread:pr:188401,pr #188401: dtensor: include local tensor device in _stable_hash_for_caching,PR blocked,73,PR blocked,1,0,,188401,,,0,b2f2e4f3704f33f0a7686079f9de4c4418d27a34b82d641df3188ef9a923c587
thread:pr:188406,pr #188406: [dynamo] preserve Stream/Event subclass type across graph-break rec…,PR blocked,73,PR blocked,1,0,,188406,,,0,6349b60d86a69aa5e3828f024e35367bd01e9a90018e4a87d55a6421eb323903
thread:pr:188410,pr #188410: Use typing_extensions.TypeAliasType for better reexport of __module__,PR blocked,73,PR blocked,1,0,,188410,,,0,adba520122dd869a5c089985da850968259937dba5cef8fe84606b27995ad256
thread:pr:188426,pr #188426: Fix incorrect CPU bf16 conv3d when innermost output dim is 1 and strided,PR blocked,73,PR blocked,1,0,,188426,,,0,f37f3d5b9f41c7369232399ef6955cb5fda448bf11db8d3a9ac0e4fd81696439
thread:pr:188466,pr #188466: Fix inductor scatter crash with empty index,PR blocked,73,PR blocked,1,0,,188466,,,0,8dc7f86247ba10ec545600a9e5cc3f58824dfea5f6aa1a76a5c140f34e2c7c13
thread:pr:188511,pr #188511: [AI Codemod][PerfAICT-General] Replace cast with castRaw in TensorType::isSubtypeOfExt to avoid shared_ptr refcount overhead (#188511),PR blocked,73,PR blocked,1,0,,188511,,,0,21472c3253b7a4f02316b7c2cd35ea4b1e824e9df36a8cc22e6746744ddef893
thread:pr:188521,pr #188521: Fix OrderingBarrier additional_buffer_deps key mismatch in control_deps lowering,PR blocked,73,PR blocked,1,0,,188521,,,0,27a175bfd0446ecacf8eb5f9394364ad0cb21504f0b24fc24c366bdec21dacea
thread:pr:188549,pr #188549: [AI Codemod][PerfAICT-General] Vectorize index_select bounds check via fused unsigned compare (#188549),PR blocked,73,PR blocked,1,0,,188549,,,0,be117f0419d68728119711d59ffd149a5a8d8c75e3e65f033136d155e2e910d3
thread:pr:188853,pr #188853: Register DTensorSpec as an opaque reference type for torch.export,PR blocked,73,PR blocked,1,0,,188853,,,0,9815f487fda72aeb559961dc099314175e1636a5e6e1b5225e55e6ce013f98fe
thread:pr:188859,pr #188859: Raise a clear error when linalg.eigh/svd workspace exceeds 32-bit LAPACK limit,PR blocked,73,PR blocked,1,0,,188859,,,0,36b849c569654b11d4255bdc0b0db48f99b920496458cc3943e75ec054d1ce8b
thread:pr:188902,pr #188902: [ATen] Add reduction tag to missing base reduction ops,PR blocked,73,PR blocked,1,0,,188902,,,0,82021bf5a9843492bc1dab078db8328f427cc6b35eb6a25ba218d6c3abff4c88
thread:pr:189191,pr #189191: Cuda Graph Default Enable (#189191),PR blocked,73,PR blocked,1,0,,189191,,,0,ed26cf8bf16ef43eda84915125860c542b69ff7bb49908cd75761b5ed04a3894
thread:pr:189193,pr #189193: [XCCL] Allow XCCL to be set as default in `_new_process_group_helper`,PR blocked,73,PR blocked,1,0,,189193,,,0,5ad01e73c91ae2424f120de99533fbdbb9a6d90adee095a1771fc8b908236e5c
thread:pr:121445,pr #121445: [NJT] Actually inline NT torch function during dynamo,ready for maintainer decision,72,ready for maintainer decision,1,0,,121445,,,0,c209cff797c7934d6bba57917cec39a03efc23a10507daa24a66e957d0915431
thread:pr:141912,pr #141912: Add stdlib includes to c10d debug handlers,ready for maintainer decision,72,ready for maintainer decision,1,0,,141912,,,0,1eff1d73b01984bdc3163c78a7a9af44c34903a050b89e9431bd8a80183a628b
thread:pr:171101,pr #171101: workaround for transpose_mxn test failure on AArch64,ready for maintainer decision,72,ready for maintainer decision,1,0,,171101,,,0,98b7e5ced2b85a60d513eeb403f3a82b717532cf6e443b1ec05100052a3df515
thread:pr:183857,pr #183857: Support float8 dtypes in index_copy on CPU and CUDA,ready for maintainer decision,72,ready for maintainer decision,1,0,,183857,,,0,bc0ae703dcd67a72e12f05a6bcae4331b0f73142ecc5ada2dc27d8fcb61677a3
thread:issue:167007:issue:170672:issue:188153:pr:184978,pr #184978: Improve export error for untraceable lifted constants,ready for maintainer decision,71,Review active PRs and linked issues as one maintainer work thread.,5,0,"167007, 170672, 188153","184978, 185059",,,4,ed76495177452a9b0863044d2f4c571f6d3cefceabc7c16bbcc94f95facab358
thread:issue:128554:issue:161671:pr:185170:pr:186418,pr #185170: Fix ExportedProgram pickle after graph_module edits,ready for maintainer decision,71,Review active PRs and linked issues as one maintainer work thread.,4,0,"128554, 161671","185170, 186418",,,3,deb453abd5849fc6bdfa3b871c083837af5186c95d0036de6793a50e8ab96a66
thread:issue:157610:issue:162734:issue:163037:pr:185124,pr #185124: Fix prims.broadcast_in_dim functionalization,waiting on contributor,71,Wait for contributor update on PR #185124; keep related issue/PR context attached.,4,0,"157610, 162734, 163037",185124,,,11,24b54ed5f632b1f1674282255485db349494a28bd5278ec6a8238a41972d9c9b
thread:issue:101192:issue:147380:pr:185803,pr #185803: Fix joint export for tied parameter aliases,ready for maintainer decision,71,Review active PRs and linked issues as one maintainer work thread.,3,0,"101192, 147380",185803,,,2,c0be5d78582a9615cb1f2d3d819655c322e4dec1724a113101fc1cb7ece07e71
thread:issue:140683:pr:186118,pr #186118: Preserve guard eval defaulting without callback churn,ready for maintainer decision,71,Review active PRs and linked issues as one maintainer work thread.,2,0,140683,186118,,,1,a3502384c7d473decb3ce70ff25934a86fc60cae578cb5dfe20daf1ebd482a15
thread:issue:145785:pr:185926,pr #185926: Preserve tuple arguments in export serde,ready for maintainer decision,71,Review active PRs and linked issues as one maintainer work thread.,2,0,145785,185926,,,1,582837e55b12286ac42a2b386dd4197bae8104b1bb781b3b6ca52fd2a41dca23
thread:issue:153387:pr:185564,pr #185564: Handle pre-dispatch export of subclass tensor constants,ready for maintainer decision,71,Review active PRs and linked issues as one maintainer work thread.,2,0,153387,185564,,,1,8d7f32cc444594153a2e76c9ad6e0c71fa12aecc6fb7e5868332b1191e336ea1
thread:issue:154647:pr:185499,pr #185499: Fix unbacked memo reuse during decomposition retrace,waiting on contributor,71,Wait for contributor update on PR #185499; keep related issue/PR context attached.,2,0,154647,185499,,,3,03a874b598bad9822bc6da6f3f04792674aa11de1c3342ff50cd879c20ee2059
thread:issue:156511:pr:185460,pr #185460: Fix CPU perf smoketest baseline drift checks,draft/noise,71,Review active PRs and linked issues as one maintainer work thread.,2,0,156511,185460,,,1,52bd08a8d3a1f6593fc2fd951d72f6587fbaad4d0e6c6d03c91053718ff43283
thread:issue:156786:pr:185450,pr #185450: Fix reinplace alias checks for index_put,ready for maintainer decision,71,Review active PRs and linked issues as one maintainer work thread.,2,0,156786,185450,,,1,0d394c2d07a38cfc328ed570ea905333a036a3f907a49b39c6e2a0c99d033098
thread:issue:158540:pr:186894,pr #186894: Fix non-strict export of vmap tensor indexing,ready for maintainer decision,71,Review active PRs and linked issues as one maintainer work thread.,2,0,158540,186894,,,1,e4f13033a621d1de1caf7abe78ca306f1e13b8f727ddbbfb69bdf5c55b4f4064
thread:issue:159124:issue:165444,issue #159124: AOTInductor preallocate output tensors,stale/low urgency,71,stale/low urgency,2,0,"159124, 165444",,,,1,9ec19670e4e14318f19c945d8212b7855d38b81e5671ee59c602d4ad8b79c2d5
thread:issue:159346:pr:185404,pr #185404: [inductor] Fold small-row expanded bmm to mm in post-grad,ready for maintainer decision,71,Review active PRs and linked issues as one maintainer work thread.,2,0,159346,185404,,,3,9d2c708f513ff9f17f31e3767052b0d0b3adcf970c85b1a762583f48479adce2
thread:issue:159918:pr:185353,pr #185353: Normalize export state names for subclass decomposition,waiting on contributor,71,Wait for contributor update on PR #185353; keep related issue/PR context attached.,2,0,159918,185353,,,1,1361e2ff778731b80b964893205c7e05cbaf5430a079817364db7e50625781b3
thread:issue:161807:pr:185150,pr #185150: Preserve OrderedDict outputs in FX tracing,waiting on contributor,71,Wait for contributor update on PR #185150; keep related issue/PR context attached.,2,0,161807,185150,,,5,6d358ee3a753ff0959b7f2a905d5573151dc23880fa0e070c08cc1212a9de897
thread:issue:162386:pr:185140,pr #185140: Preserve reentrant checkpoint in non-strict export,ready for maintainer decision,71,Review active PRs and linked issues as one maintainer work thread.,2,0,162386,185140,,,1,49b4ffe049495fa1ce4212a838052387fac08edb7d42ce8514866796041b88bd
thread:issue:163303:pr:185106,pr #185106: Fix non-strict export state aliases,ready for maintainer decision,71,Review active PRs and linked issues as one maintainer work thread.,2,0,163303,185106,,,1,a00c5b7a9c90a63acdc94a7fc47027d0240ad43aa78aa948d06457d5902ea2d7
thread:issue:163323:pr:185105,pr #185105: Fix export move_to_device_pass device rewrites,ready for maintainer decision,71,Review active PRs and linked issues as one maintainer work thread.,2,0,163323,185105,,,1,4bbdfff5a002ae056e93e3043bdec23af25d0661463eb8ccd3a03b0bdeb2c928
thread:issue:164313:pr:186879,pr #186879: Fix export dynamic shapes for variadic kwargs,ready for maintainer decision,71,Review active PRs and linked issues as one maintainer work thread.,2,0,164313,186879,,,1,f88c7f0c2a575ee87c6ac85b29d44982db27df5d2ad6ad9535b1ddcd00528b53
thread:issue:165073:pr:185082,pr #185082: Preserve lifted SymInts during strict export capture,draft/noise,71,Review active PRs and linked issues as one maintainer work thread.,2,0,165073,185082,,,1,e682a0c3bb707bb62f6b25d95bc717b7333451d2bbf8b475fa07a67afdae9604
thread:issue:165641:pr:185073,pr #185073: [export] Treat free new-tracer state as constants,ready for maintainer decision,71,Review active PRs and linked issues as one maintainer work thread.,2,0,165641,185073,,,1,a0e01e0f83f78a5e6f5d8b0e3ec040f8e0bc020c4fe419c6b55d0b627de6b959
thread:issue:169991:pr:184985,pr #184985: Fix nested compile handling in AOT joint export,waiting on contributor,71,Wait for contributor update on PR #184985; keep related issue/PR context attached.,2,0,169991,184985,,,12,05947adcf961ac4eb7cb4f922f4681304ecec925bec96b11480250f073a38fdc
thread:issue:170539:pr:186833,pr #186833: Remove stale export OpInfo xfails,ready for maintainer decision,71,Review active PRs and linked issues as one maintainer work thread.,2,0,170539,186833,,,1,436afd0155f4e5d71a3bc7d3be40bd097e0dba034fd555f1729d025b8f78e847
thread:issue:149497,issue #149497: symbolic_trace failed on deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B,needs triage,71,needs triage,1,0,149497,,,,0,388c6fa15187dfb5769364fbb0fada0283e2811bdadf76b89f41c5a05c256b78
thread:issue:163312,issue #163312: [symbolic trace] fuction with decorator,needs triage,71,needs triage,1,0,163312,,,,0,f409ee0369f5936c08a5af99ea46ecce77d67bb81d4b5a02c8f9c6d59d7520f3
thread:issue:166121,issue #166121: [profile] profile_memory doesn't work if _ExperimentalConfig(profile_all_threads = True) is set,needs triage,71,needs triage,1,0,166121,,,,0,ae3e464eece68ecc84b5cfa5779bb84c1d23292e849f8df813d4cb7ac1aaa8f4
thread:issue:169210,issue #169210: Segmentation Fault when Converting QInt32Storage to bfloat16,needs triage,71,needs triage,1,0,169210,,,,0,9637eadb768ca35708971fca1878d05761a287398117f0a445ee8ec9492dcb79
thread:pr:157149:pr:187690:pr:187744:pr:188004,pr #157149: [generator] Close all open generators in compile_subgraph,waiting on contributor,70,Wait for contributor update on PR #188004; keep related issue/PR context attached.,10,0,,"157149, 187690, 187744, 188004, 188638, 188639, 188824, 188825, 188834, 189024",,,153,dd27ebef81adadb9921a7bf9c0c5ed438296a5a73674c432750988c53483b96f
thread:issue:152595:issue:165040:issue:165141:issue:169857,"issue #152595: ROCm, 7900 XTX (gfx1100): Pytorch FLASH_ATTENTION SDPA is 2.5x slower than MATH (fp16, head_dim 256, seqlen 4360, 12 heads)",stale/low urgency,70,stale/low urgency,6,0,"152595, 165040, 165141, 169857, 177834, 188564",,,,5,02b014d184616efd6a5340b361187e7b14eda686677b31f888fa14857aa31a27
thread:pr:185424:pr:189021:pr:189022:pr:189051,pr #185424: Add CPython Dynamo agentic loop plan,draft/noise,70,draft/noise,6,0,,"185424, 189021, 189022, 189051, 189052, 189053",,,37,672500944f6e99bc8006f5e3aa27fb1a04210e6fadb2569fe043a71134582e1e
thread:issue:85329:issue:87960:issue:87961:issue:88148,issue #85329: Crash in `torch.package.PackageExporter`,high-priority/blocker,70,high-priority/blocker,5,0,"85329, 87960, 87961, 88148, 94594",,,,4,c65e3bfb88aa0691ee5d363c951f780a807d421432b5e4a70181bedb467e5748
thread:pr:178393:pr:179094:pr:187631:pr:188029,pr #178393: [DO NOT MERGE] Dynamo 3.15 support,draft/noise,70,draft/noise,5,0,,"178393, 179094, 187631, 188029, 189306",,,25,8e623852500673715506d1fd6f65b6824130e1e34b4ed506f29ff01ee9172a85
thread:issue:112583:issue:141896:issue:169934:issue:65766,issue #112583: Nesting no_grad in autocast causes backwards graph to be (partially) lost outside of no_grad,high-priority/blocker,70,high-priority/blocker,4,0,"65766, 112583, 141896, 169934",,,,3,743ce94079741e5ccda0c2af9a79f99dd02ff0d2ba896cb58c0abad2f844a974
thread:issue:176662:pr:179286:pr:186854:pr:188440,pr #179286: Use C++20 concepts where it improves readability,PR blocked,70,Review active PRs and linked issues as one maintainer work thread.,4,0,176662,"179286, 186854, 188440",,,4,08cb81506d2c6da9823cfd131b52da115042dcff5a86780e9d370da0238676bc
thread:issue:178153:pr:183276:pr:183277:pr:183278,pr #183276: [Inductor][Refactor] 2/N Introduce shared heuristic registry with codegen_heuristic support,PR blocked,70,Review active PRs and linked issues as one maintainer work thread.,4,0,178153,"183276, 183277, 183278",,,3,717af04ec810b7e7ee34dc841243b54f6c10e28cad1d43a559e7254e8ea14887
thread:issue:182263:issue:78475:issue:78489:issue:82098,"issue #182263: libtorch_cpu.so BLAS symbols are interposable, causing crashes when another BLAS is present in the process",high-priority/blocker,70,high-priority/blocker,4,0,"78475, 78489, 82098, 182263",,,,3,d4e7f95920a7b04f5c92b1ce608b4106b0b44ab1b4950b5f3681aad5ea926731
thread:issue:188420:issue:188425:issue:188430:issue:188432,issue #188420: torch.compile produces inconsistent outputs for MultiHeadAttentionModel with scaled_dot_product_attention,needs maintainer decision,70,needs maintainer decision,4,0,"188420, 188425, 188430, 188432",,,,3,5e82eb8a5930376c76113fae715e5664f426d70c4997b586628bc2d58bd8d8d0
thread:pr:181780:pr:181781:pr:188175:pr:188176,pr #181780: [dynamo] set canonicalize_output_graph_node_order=True for export tests,PR blocked,70,PR blocked,4,0,,"181780, 181781, 188175, 188176",,,12,b3d294f87cada0944724faa4aabfe6c91c0e7cc2b75387e02e2df3f2e5a70003
thread:issue:135859:issue:187953:pr:187952,pr #187952: Fix out= variant recompilations for torch.max/min/topk,ready for maintainer decision,70,Review active PRs and linked issues as one maintainer work thread.,3,0,"135859, 187953",187952,,,2,9a8d0082ca3c80e4a2605398c1a0290bcd08ecb2b032b6c82b4c9a69c115e625
thread:issue:93880:issue:99640:issue:99918,"issue #93880: `PYTORCH_DEBUG_MODE`, better invalid index embedding lookup error message on cuda",high-priority/blocker,70,high-priority/blocker,3,0,"93880, 99640, 99918",,,,2,428b10b82c08c173d56610d8d869e2c46bb2c7e121d6ab2aa41110962bac742b
thread:issue:142515:issue:142836,issue #142515: [MPS] `out_channels` <= 2**16 guard for convolution is too broad,stale/low urgency,70,stale/low urgency,2,0,"142515, 142836",,,,1,96882b0b0a37fdffe9e108a2c3154a019515fdb9f1a23bde6bf15750979125fc
thread:issue:145702:issue:153961,issue #145702: PyTorch VS2022 official build Windows binary illegal instruction on AVX2(max ISA level) CPU,high-priority/blocker,70,high-priority/blocker,2,0,"145702, 153961",,,,1,a40b9a8b3778adcc6750232e060a18ed0ef43787d7bd98859f1b5154902984e0
thread:issue:152954:issue:152963,issue #152954: DTensor placement propagation for `as_strided` missing,high-priority/blocker,70,high-priority/blocker,2,0,"152954, 152963",,,,1,f70e304d2d3ac90f4258cd73974c994d8d2c10f6266c335698e3eecc25e04838
thread:issue:179891:issue:188953,"issue #179891: torch.xpu.get_device_properties() segfaults on Intel ARC PRO B70 (Battlemage/BMG, device 0xe223)",high-priority/blocker,70,high-priority/blocker,2,0,"179891, 188953",,,,1,d900da21798f5173b745465296f9517fd6f5262e49811e6fac1f20663e6e30c1
thread:issue:188557:issue:189077,issue #188557: operator_microbenchmark ROCm gfx950 job never completes -- runs 8h+ and is cancelled at the workflow timeout,high-priority/blocker,70,high-priority/blocker,2,0,"188557, 189077",,,,1,8d7f8a97eba232e2296ad4b5710671b4513807561de959a984f37a4de3bc2c94
thread:issue:188733:issue:45111,issue #188733: [vllm] [2.12.1 regression][Spec Decode] CUDA WARP_ILLEGAL_ADDRESS in FlashAttnFwdSm90 during CUDA graph capture (EAGLE),high-priority/blocker,70,high-priority/blocker,2,0,"45111, 188733",,,,1,bc4fae0b7244ecd2bc6a4cda1e436b4bd0e15964e121b88bcde52a8f77552da2
thread:issue:30774:issue:49372,issue #30774: Natively Declarable Fast-path Functions,high-priority/blocker,70,high-priority/blocker,2,0,"30774, 49372",,,,1,558f63608dcc253315a5e870144a44ca74cce32c8615a1964d14d752536c07c6
thread:issue:41486:issue:42815,"issue #41486: LSTMs leak memory in CPU PyTorch 1.5.1, 1.6, and 1.7 on Linux",high-priority/blocker,70,high-priority/blocker,2,0,"41486, 42815",,,,1,bd9721b41df09c8f739e43bcbad431d5c74c9f3c78e74dfb20717b8a85059743
thread:issue:47993:issue:51039,"issue #47993: build pytorch from source on ubuntu, building error from fbgemm::SparseAdaGradSignature",stale/low urgency,70,stale/low urgency,2,0,"47993, 51039",,,,1,6e28b46fae64dc2411f89bbe4695a06e1a47ead124389cfba7ba6ac15b7d3a6c
thread:issue:76865:issue:77200,issue #76865: Object-base collectives create tensors at unexpected devices,high-priority/blocker,70,high-priority/blocker,2,0,"76865, 77200",,,,1,3b9a039a0057889fcdd2bd6a74b0523bcc19a2e485d011192ef99e4275bffa4c
thread:issue:91570:issue:92226,issue #91570: Security policy impractical / lacks contact information?,high-priority/blocker,70,high-priority/blocker,2,0,"91570, 92226",,,,1,43978bafdc3c30c9b077791f619cb9bd3b62140a796885b1e959c78e577c2113
thread:pr:187508:pr:187509,pr #187508: c10d: register nccl2 backend,PR blocked,70,PR blocked,2,0,,"187508, 187509",,,6,f16035d2b5a695a0b18e7a390e958e26a033862c15ecc5da063c05e5c6be9ef7
thread:pr:188042:pr:189181,pr #188042: [CI] Split AI advisor garbage verdict into infra_issue + garbage,waiting on contributor,70,Wait for contributor update on PR #188042; keep related issue/PR context attached.,2,0,,"188042, 189181",,,4,ea1b84d629e5f9382a38d983479f94e03d693c04011cc78e8eb9747bdd1a6234
thread:issue:113586,issue #113586: mps bug: failed assertion `[MPSNDArrayDescriptor sliceDimension:withSubrange:] error: subRange.start (8192) is not less than length of dimension[1] (512)',high-priority/blocker,70,high-priority/blocker,1,0,113586,,,,0,5e80eba5dfb9311708363e3b1f817d6010e810600f05d7561d530459a860b20e
thread:issue:120601,issue #120601: Torchrun arguments shadowing script arguments,high-priority/blocker,70,high-priority/blocker,1,0,120601,,,,0,5497d63d86637044f4153c00300b303318f59684ec7a401cfcd79e052e542666
thread:issue:124484,issue #124484: HOP dispatch isn't faithful,high-priority/blocker,70,high-priority/blocker,1,0,124484,,,,0,14610e10ea72e660f554a23f89e020929653d241c9cc84173d76de71073ca9e9
thread:issue:139989,issue #139989: Support AC with graph break,high-priority/blocker,70,high-priority/blocker,1,0,139989,,,,0,061f78c1bb934844f5fc498a1f9048171a8e3cb44ff11ff75fd34018b8bc308b
thread:issue:142048,issue #142048: Device check missing in torch.linalg.solve_triangular leading to hard crash,high-priority/blocker,70,high-priority/blocker,1,0,142048,,,,0,3266cbb3636c747d73abd9028f5099210d9366efe4bbe04e17e12056209c2992
thread:issue:142457,issue #142457: Aborted (core dumped) in `slow_conv_transpose3d`,high-priority/blocker,70,high-priority/blocker,1,0,142457,,,,0,8676825bb2dd024c2a1ee2b4048193256e5588eb797440247a0d3d3a50ed262f
thread:issue:142459,issue #142459: Segmentation fault (core dumped) in `max_pool3d_with_indices_backward`,high-priority/blocker,70,high-priority/blocker,1,0,142459,,,,0,6c402c528fc2a9401330d03d88b4093abbf71cfbe8ecfbd1de70e119d9d375ff
thread:issue:142834,issue #142834: Segmentation fault in `replication_pad2d_backward`,high-priority/blocker,70,high-priority/blocker,1,0,142834,,,,0,6e5b75c422e6b56eee6360e79ec5930e11d27232ee0bad7a65a52eb6b8879d17
thread:issue:143484,issue #143484: Segmentation fault (core dumped) in `gru_cell`,high-priority/blocker,70,high-priority/blocker,1,0,143484,,,,0,82830e4c22917872e5ef3a3727347dafd6c970a5e34986b628102b91f0c1be04
thread:issue:143487,issue #143487: Aborted (core dumped) in `replication_pad1d`,high-priority/blocker,70,high-priority/blocker,1,0,143487,,,,0,28e2cf085d564ac8d73e12719902df2c9a706d43f01848629473ac3c1d8267fa
thread:issue:150098,issue #150098: Aborted (core dumped) raw pointer API is exposed directly to Python,high-priority/blocker,70,high-priority/blocker,1,0,150098,,,,0,baf192daef329dccbe6236b9b2e6dc9e0d558396bb3da03b4ac623ffd56a9f57
thread:issue:152302,issue #152302: NCCL out of memory error after updating to PyTorch 2.7,high-priority/blocker,70,high-priority/blocker,1,0,152302,,,,0,4de5f38cf54cd2b91ec3b67c98ba0662d47b2f6b30ee6afb6d8e90028d444ae6
thread:issue:156819,issue #156819: FakeTensorUpdater doesn't support HOPs,high-priority/blocker,70,high-priority/blocker,1,0,156819,,,,0,4b7c8c4c29ef6c0bb3ed3e6940c0242d05500f0f3a7dc1564b215dc8c944c4f3
thread:issue:157128,"issue #157128: Windows Source Build Fails with OSError: [WinError 126] on aoti_custom_ops.dll for RTX 5080 (sm_120), Pre-built PyTorch Works",high-priority/blocker,70,high-priority/blocker,1,0,157128,,,,0,3dca39e7cc87bf4c9dffee6791fa94657850d84861cab903cfc274bc071ea2de
thread:issue:161295,issue #161295: UNSTABLE inductor-periodic / cuda12.8-py3.10-gcc9-sm80 / test (inductor_torchbench_smoketest_perf),high-priority/blocker,70,high-priority/blocker,1,0,161295,,,,0,a5ddf87908e09c754df4b7720156289bd5fee4354a7f35fca362ee9a772b8486
thread:issue:163501,issue #163501: Improve Registration Error message with const SymIntArrayRef&in PyTorch 2.5+,high-priority/blocker,70,high-priority/blocker,1,0,163501,,,,0,c37191e632d996e1037b40520001bfed2232b6a54f0e12ea62d375fa9022b6a6
thread:issue:165448,issue #165448: `import torch` fails on 3.13 Linux nightly jobs with AST parsing errors,high-priority/blocker,70,high-priority/blocker,1,0,165448,,,,0,81e483b8c873afe0230492743e3f7d33206954c63ad36caf8e5e10ed3cf870ad
thread:issue:167373,issue #167373: Torch Tensor -> Nanobind ndarray breaks cuda graphs on torch 2.9,high-priority/blocker,70,high-priority/blocker,1,0,167373,,,,0,8bdd6d768782bfd3bcc0d9e27bd6c074005e02b70ad2b1cc7507b2a51b2b2c1d
thread:issue:168058,issue #168058: [DTensor] inconsistency of distributed RNG (Shard(0) != Shard(1) or 8 GPU != 4 GPU != 1 GPU),high-priority/blocker,70,high-priority/blocker,1,0,168058,,,,0,67978b469b5c83400d72d55c3989046143c6569d0ac9d4c532a093ddb4b298dd
thread:issue:168997,issue #168997: [Umbrella issue] Deployment of index updates to download.pytorch.org - Switch dependencies that are on pypi to utilize pypi's CDN,high-priority/blocker,70,high-priority/blocker,1,0,168997,,,,0,4b9bdfc35cc1fd47ce09fbc81e55801585850ae721e3896ea55628224bbc55c5
thread:issue:169741,issue #169741: ReplicationPad2dwith extremely large padding value causes segmentation fault,high-priority/blocker,70,high-priority/blocker,1,0,169741,,,,0,5d2daf5da3e0b153c5e902ef1a9b7a930f90562c97d49670bb9cbdb72ee13af7
thread:issue:170787,issue #170787: [ARM] Unit test regression AArch64 - int4mm kernels require bf16 hw support,high-priority/blocker,70,high-priority/blocker,1,0,170787,,,,0,e7929acd23c3618fdaf99b7d2912d8f025b27a2e46c6e3318ea74cb9addfb6eb
thread:issue:172537,issue #172537: [Regression] 20x slower vision model forward pass in PyTorch 2.9.1 vs 2.8.0,high-priority/blocker,70,high-priority/blocker,1,0,172537,,,,0,9dad52971bf8c4a9b6ccd50d4576e25d4f0dbb0f22e4a878845abc6e2e7d490f
thread:issue:172798,issue #172798: `backward` does not respect `with torch.device(device)`,high-priority/blocker,70,high-priority/blocker,1,0,172798,,,,0,9b7f2fa25b6771d3b3a945d7374f1f38249c173dfd9c624b7eb4c28adc9a9ef3
thread:issue:172907,issue #172907: [CI] Changes to install scripts in lintrunner.toml are not tested in CI,high-priority/blocker,70,high-priority/blocker,1,0,172907,,,,0,8ea854622acccc25f65a84bc381aaba2c7247cb4f62239a9ae3f94edec41d6c5
thread:issue:173313,issue #173313: Bad perf for torch.compile for merge_attention with transposed inputs,high-priority/blocker,70,high-priority/blocker,1,0,173313,,,,0,6d44ef3635f496435f84e561e0c8df68135070ad9f5c8cba082a5735a1caea28
thread:issue:173574,issue #173574: `torch.arange` triggers `SIGFPE` (Floating Point Exception) instead of RuntimeError when writing float results to `int64` out tensor,high-priority/blocker,70,high-priority/blocker,1,0,173574,,,,0,a54631b58e82fb988e9434eec3c3b1800aab7e5489ba27ba22833996339119d1
thread:issue:174704,issue #174704: Recompile time with torch.compile 2.10,high-priority/blocker,70,high-priority/blocker,1,0,174704,,,,0,dbc3332fcb77a650fd4b62e45f09cd5baff46b10cc27edae75a89da6a3b072c6
thread:issue:175436,issue #175436: Intermittent SIGILL in `torch.sin` on CPU (`torch==2.10.0+cpu`) on Xeon E5-2670; crash in `mkl_vml_kernel_dSin_*`,high-priority/blocker,70,high-priority/blocker,1,0,175436,,,,0,ad51f7ab3e6538c0d655bcdbc1255ab6cdda431ef3f9b1e9644a09eb4ca92aa5
thread:issue:175928,issue #175928: [release 2.11][vllm] Language Models Tests (Extra Standard) 1 test_models[float-jason9693/Qwen2.5-1.5B-apeach],high-priority/blocker,70,high-priority/blocker,1,0,175928,,,,0,b9e58476fe9ce38390cb23ccb6aa5977dffb8f01c72879395d9a64c98cfa1d5e
thread:issue:176285,"issue #176285: AArch64 test_cpu_repro - Expected to not find ""transpose_mxn"" but found it",high-priority/blocker,70,high-priority/blocker,1,0,176285,,,,0,30ee7861f696134eb77a1780128dcc1fac4fda77103b888baf464e3c93215122
thread:issue:176784,issue #176784: [CUDA] AdaptiveAvgPool2d: START_IND macro causes int32 overflow → OOB read,high-priority/blocker,70,high-priority/blocker,1,0,176784,,,,0,731bf0998f5f4d58c1a77abc20a3c8e379c34250246fffd24bcdf95b60d555bc
thread:issue:177714,issue #177714: XPU: Fatal Level Zero `RuntimeErrors` instead of recoverable `OutOfMemoryError`,high-priority/blocker,70,high-priority/blocker,1,0,177714,,,,0,2af5fa51d5bca701c55c4f1d331c488e9400812125e75bd918c1f6860ccdca01
thread:issue:178726,issue #178726: Greatly increased CPU memory use on aarch64 H100 node from torch 2.9.0 to torch 2.10.0,high-priority/blocker,70,high-priority/blocker,1,0,178726,,,,0,7a8208327335bb0fc9f101e7b691c54e436b619a3c3a0536be79c03100b8df07
thread:issue:178880,issue #178880: `torch.compile` produces different classification output for model using E8M0 bit-manipulation quantization (`view(int32) → bitshift → clamp → uint8`),high-priority/blocker,70,high-priority/blocker,1,0,178880,,,,0,c45c679e0e5c881b1bb4f40ddade847792494d98d880c03c9b4e96fd4847f9e6
thread:issue:178939,issue #178939: Pytorch needs Kineto even when using USE_KINETO=OFF,high-priority/blocker,70,high-priority/blocker,1,0,178939,,,,0,533ab272df06446790d06fd510a0ca6f894fdfe88f323c0dcb39f03d762f12ac
thread:issue:179030,issue #179030: XPU: Segmentation Fault when using newer drivers,high-priority/blocker,70,high-priority/blocker,1,0,179030,,,,0,c18b07c78b08784afd2cde33eb49234cf4ab3812337c86962fc599d1f71e33ad
thread:issue:179804,issue #179804: torch.compile + aot_eager: backward ignores inner autocast(enabled=False) vs eager,high-priority/blocker,70,high-priority/blocker,1,0,179804,,,,0,6ae01fecc042e83e62e801f7492544f4e567e3bdad7b3a4ab91d07feea662d1a
thread:issue:181024,issue #181024: [CUDA] `test_dlpack_tensor_on_different_device` fails with `torch.device` instead of `torch.cuda.device`,high-priority/blocker,70,high-priority/blocker,1,0,181024,,,,0,f757a9cd58b79dba2183f9d02432f0c3f0f9bbdff1dc1855287f37d1792e9863
thread:issue:181362,issue #181362: DISABLED deit_tiny_patch16_224.fb_in1k fail_accuracy on ROCm dynamic_inductor_timm_training,high-priority/blocker,70,high-priority/blocker,1,0,181362,,,,0,81722fd78367a264e137d9fc70098f63e2b0547c9ed8192b0647b78d079b6be1
thread:issue:181869,issue #181869: [vllm] [torch nightly] AOTAutogradCache.save: _pickle.PicklingError on triton launcher (DFlashDraftModel speculative-decode path),high-priority/blocker,70,high-priority/blocker,1,0,181869,,,,0,812c420c66d83123b701b439331acc1473012ff853a7e28d18ae26a474b945f8
thread:issue:182949,issue #182949: [CD] Drop CPython 3.13t from binary build matrix (manylinux upstream removed it 2026-05-07),high-priority/blocker,70,high-priority/blocker,1,0,182949,,,,0,ee5997fb8752e92a4582bae3d52ca17ab8aca8de3a5cda8fe32711f75078f250
thread:issue:183401,issue #183401: [inductor][cpu] resnet50_quantized_qat CPP wrapper torch._inductor.exc.InductorError: CppCompileError: C++ compile error in 2026-05-11 nightly release,high-priority/blocker,70,high-priority/blocker,1,0,183401,,,,0,2541fbcef294444ac7c90445571c577bb969f01258f396d305662bb5b1bcb25d
thread:issue:183890,issue #183890: torch.compile hits RecursionError on nn.Module with custom __getattr__,high-priority/blocker,70,high-priority/blocker,1,0,183890,,,,0,585d76d614c7d7e653b2477009ca290d4f7176d8f765fa89124065955ef9e93b
thread:issue:184027,issue #184027: [inductor] `torch.compile` crashes on convolutions when weight is a plain tensor — stride prediction mismatch,high-priority/blocker,70,high-priority/blocker,1,0,184027,,,,0,01a4f4d543a8539897a56b4eb5dffbd267b04f0736f83bdee83ccf4322700145
thread:issue:184431,issue #184431: [vllm] [2.12 regression][multimodal] Qwen2-Audio text-then-audio_embeds: prompt_embeds vs raw-text outputs diverge under --enforce-eager,high-priority/blocker,70,high-priority/blocker,1,0,184431,,,,0,0cdf553cf4924802d625a27f1785b15d1e315e0c642d833cebcf5d84918222be
thread:issue:184549,issue #184549: [vllm] [2.12 regression][Inductor] assert_size_stride (8192 vs 512) in BERT-with-rope AOT-compiled path for Nomic v2 MoE pooling model,high-priority/blocker,70,high-priority/blocker,1,0,184549,,,,0,e1d894a894443e6c56294add28ecd6a2300f95cd18223d771b4ccf86c138cd92
thread:issue:184880,issue #184880: [ROCm] RX 9060 XT 16GB hard-capped at ~8GB usable VRAM in PyTorch — segfault on allocation beyond cap,high-priority/blocker,70,high-priority/blocker,1,0,184880,,,,0,25e1c9c759e044d9c1e6b71b07f78656dd4823367e5c0fb09691e8b844567563
thread:issue:185346,issue #185346: [Inductor] Performance regression of phlippe_densenet model due to non-fused Cat kernels from new fusion mode,high-priority/blocker,70,high-priority/blocker,1,0,185346,,,,0,c29afec30a96b199bd2aa0777fc8416e9db360f1db8757ce8f9381f8620f8c23
thread:issue:185405,"issue #185405: [Inductor] AssertionError: len(index) == len(stride) in ir.py when using torch.compile(mode=""max-autotune"") on CPU BMM/GEMM",high-priority/blocker,70,high-priority/blocker,1,0,185405,,,,0,c9edc8cf18381f55c3cc0a8ea302a8ea94301af311f58da09901fb07f2ac3e3b
thread:issue:185494,"issue #185494: [dynamo] LazyBatchNorm2d fails under torch.compile(dynamic=True) — ""SymIntArrayRef expected to contain only concrete integers""",high-priority/blocker,70,high-priority/blocker,1,0,185494,,,,0,39e65ebf49bb06ff1ac94fbb780c85b26bf307c8e8a879886f167385a455003d
thread:issue:185839,"issue #185839: torch.linalg.qr (float64) repeated in a loop crashes the process (SIGSEGV, exit 139) on ROCm 7.x",high-priority/blocker,70,high-priority/blocker,1,0,185839,,,,0,5d952f314d13339dd8113396f4f90591466633db7e2fba9f909667e4f02ba8d9
thread:issue:186174,issue #186174: ROCm: Flash/mem-efficient SDPA hangs during batched autoregressive generation with left-padded sequences,high-priority/blocker,70,high-priority/blocker,1,0,186174,,,,0,e6d5397e06d892f4cbadd4424f2c7be21a39139c54e2117e031bb7c5daebbc33
thread:issue:186212,issue #186212: C++ API: at::batch_norm crashes with an internal assertion on invalid inputs,high-priority/blocker,70,high-priority/blocker,1,0,186212,,,,0,608e56f3dd045fae193903dd6bfd58ac1c45ff59078eeeb37e33ee5ca468fe2d
thread:issue:186516,issue #186516: torch.cuda.device_count() returns 0 on ROCm when amdsmi is installed (duplicate libamd_smi ODR clash); HIP reports 1,high-priority/blocker,70,high-priority/blocker,1,0,186516,,,,0,929a4683989025b5b622c7a83d7cebb36fbe1ec1cf6c5388fee2f3bc49c95a52
thread:issue:186952,issue #186952: GPU is not accessible on some MacOS-15 runners,high-priority/blocker,70,high-priority/blocker,1,0,186952,,,,0,43371962cabe7e280e9e3ecdeab895ea5609612497cfb81e88deb9e41959fcc6
thread:issue:187319,issue #187319: torch.compile FX graph does not distinguish custom ops with the same name but different mutates_args,high-priority/blocker,70,high-priority/blocker,1,0,187319,,,,0,6a4780519164734d8c463ba9a31edea1e7ce92a644a571e563627cb2d3425571
thread:issue:187320,issue #187320: Stable ABI isn't stable when PyTorch changes the default CUDA version,high-priority/blocker,70,high-priority/blocker,1,0,187320,,,,0,e9a89e2d911e8adc86544d86748a6601d9a1d11322eda740df45c6921eda7b8e
thread:issue:187343,issue #187343: expandable_segments:True crashes on ROCm,high-priority/blocker,70,high-priority/blocker,1,0,187343,,,,0,033f46b130c5dd8a7128ded1799bfac3177c099806ec998e34012a80be9a34b1
thread:issue:187444,issue #187444: `torch.compile` vs eager DDP divergence with ResNet50 + AdamW lr=1.0 on two-node NCCL,high-priority/blocker,70,high-priority/blocker,1,0,187444,,,,0,098ad571b24ea9be7e4f37bd932490fdd6b73aa2800c64ff9f80cf64f3aca6b6
thread:issue:187445,"issue #187445: `torch.compile` vs eager DDP loss/grad/logits divergence with ResNet34 + SGD lr=0.1, batch=2, TF32 allowed, and 1MB buckets",high-priority/blocker,70,high-priority/blocker,1,0,187445,,,,0,0f6e64a96e87ca3abc36d1e5e56e1e901fed34437d35088a4495bbe897760166
thread:issue:187446,"issue #187446: `torch.compile` vs eager DDP grad/parameter divergence with ResNet50 + Adam lr=0.05, batch=1, sqrt LR scaling and step scheduler",high-priority/blocker,70,high-priority/blocker,1,0,187446,,,,0,9f4d728ee58baac2be88806768b88c515bc48ddca81ae2f85231bae9bfb61426
thread:issue:187447,"issue #187447: `torch.compile` vs eager DDP parameter-fingerprint divergence with ResNet50 + Adam lr=0.1, grad accumulation, grad clipping, and 1MB buckets",high-priority/blocker,70,high-priority/blocker,1,0,187447,,,,0,35b9af6b79474c177fad97e15e6a9140eaed49a9d4aee93b9805c5f8f789309c
thread:issue:187448,"issue #187448: `torch.compile` vs eager DDP grad/logits divergence with ResNet50 + AdamW lr=0.01, warmup cosine schedule, grad accumulation, and DDP bucket views",high-priority/blocker,70,high-priority/blocker,1,0,187448,,,,0,45116df4ef3ee54948e25919a971ab598b7f58e06edff1a7d4cc6c5a98137a64
thread:issue:187449,issue #187449: `torch.compile` vs eager DDP logits-sum divergence with ResNet50 + AdamW lr=0.01 on two-node NCCL,high-priority/blocker,70,high-priority/blocker,1,0,187449,,,,0,913117e859e04b56a58f3b377df21e3c670ab34d94acd89c90af0d2819b8f125
thread:issue:187951,issue #187951: inductor_timm tests has been broken since 6/12,high-priority/blocker,70,high-priority/blocker,1,0,187951,,,,0,f9615786bd257cd92612b4ceb24ee3280aa4bb9759d310bb430da91c6a7199af
thread:issue:188127,issue #188127: torch.compile causes persistent RSS growth during repeated inference in PyTorch 2.9 with ROCM,high-priority/blocker,70,high-priority/blocker,1,0,188127,,,,0,74cb2a8a6566c0b74bea2023afe7384396f8bf6af218c35cd386bec241e11a67
thread:issue:188218,issue #188218: [inductor] `LoweringException: AssertionError: max not supported for zero-dimension tensors!`,high-priority/blocker,70,high-priority/blocker,1,0,188218,,,,0,826da800969225d143c523d93af409d396aef81aba2447357b37d9d904eb859f
thread:issue:188332,issue #188332: Inductor: allgather fusion creates rank-specific cache that crashes when loaded by other ranks,high-priority/blocker,70,high-priority/blocker,1,0,188332,,,,0,3e35f9869c76b091dc71c5584c601d718e6a6455e6ddf1ba8e8cf9e2e0c486d6
thread:issue:188336,issue #188336: AArch64 CPU - Inductor benchmark Regressions,high-priority/blocker,70,high-priority/blocker,1,0,188336,,,,0,1fad25683c24e501269bf0db47bdbea5f3d9966d9fb8072184bc6e8b4aeb5b4f
thread:issue:188674,issue #188674: DISABLED test_workspace_runtime_integration (__main__.TestNVUniversalGemmEpilogueFusion),high-priority/blocker,70,high-priority/blocker,1,0,188674,,,,0,c060cb4241a48d91813f739d01944a0872378e560145dea7cc4b6c173d915893
thread:issue:188678,issue #188678: DISABLED test_redispatch_nn_functional_conv_transpose3d_cuda_float16 (__main__.TestTorchFunctionRedispatchOpsCUDA),high-priority/blocker,70,high-priority/blocker,1,0,188678,,,,0,2fc51c3312b11de6780b00336f2509ca45ee51b4aa95973a90591c1d3a753b48
thread:issue:188691,issue #188691: DISABLED test_stream_invalid_device_index (__main__.TestCuda),high-priority/blocker,70,high-priority/blocker,1,0,188691,,,,0,1bdaae36ceb5fcb0e4b61697ac363d9ca29ef583275361d140e05140855f1ab1
thread:issue:188835,issue #188835: DistillGPT2 fails accuracy on A100 in inductor smoke test,high-priority/blocker,70,high-priority/blocker,1,0,188835,,,,0,035f8ace71af610bb189b4d1272ceddfda9b949ca50e43d0505f8d1396b601b7
thread:issue:188864,issue #188864: [MPS] torch.fft.hfft aborts on 5D complex64 input with dim=0,high-priority/blocker,70,high-priority/blocker,1,0,188864,,,,0,10557608952d162ce53994e1d49f5fa0046f5c0dd6253ae78ca68e8f6ad510b8
thread:issue:188912,issue #188912: vit_base_patch16_siglip_256 amp training accuracy regression on H100 (2.13.0),high-priority/blocker,70,high-priority/blocker,1,0,188912,,,,0,087ac5570f5673b54b99eab90753f843f55362a07e3f661a41421e1cd8feccbf
thread:issue:188955,issue #188955: rms_norm autocast fp32 policy (2.12) materializes standalone cast kernels under torch.compile when dtypes already match. Peak-memory and throughput regression for bf16 models,high-priority/blocker,70,high-priority/blocker,1,0,188955,,,,0,a655b3b1616b10e8e07a705bfff1c7516b121ef634dbeee9e469799a17c42bc9
thread:issue:188958,issue #188958: [CUDA] illegal memory read on batch_norm_backward_reduce_channels_last_kernel,high-priority/blocker,70,high-priority/blocker,1,0,188958,,,,0,266547c738e5d8720798ddf8a43a465eacb4a98d0c0213319c34d1302439e92d
thread:issue:188959,issue #188959: [CUDA] illegal memory read on col2im_batched_kernel,high-priority/blocker,70,high-priority/blocker,1,0,188959,,,,0,367d7f1bcbc373dae46bb696dbd6adc6243a2cedabc5fc955b8ddc1feb93e39c
thread:issue:189018,issue #189018: `_scaled_mm_v2` meta implementation checks for device-specific constraints,high-priority/blocker,70,high-priority/blocker,1,0,189018,,,,0,6cb2e09fae7002f5fd95c0afd9d36c5166c74bc0200075ba1d79996dd3a5c3f5
thread:issue:189283,"issue #189283: [CPU] operator_benchmark: isolated ~2x regressions (index_add, baddbmm) since ~Nov 2025",high-priority/blocker,70,high-priority/blocker,1,0,189283,,,,0,ed80fcf92c7a797d3fac2ee6fb463ee693a83a0f53f813692c88bf437aaee356
thread:issue:44428,issue #44428: Support for CUDA matrix multiplication on long (and other integer) tensors,high-priority/blocker,70,high-priority/blocker,1,0,44428,,,,0,4f08d32a26981320e8e991078b60f5888e1143802da05c50f92d4ee04b6c4ac1
thread:issue:73294,issue #73294: Internal assert failed at rref_context.cpp,high-priority/blocker,70,high-priority/blocker,1,0,73294,,,,0,5565773176cfb29e5388dd4239a8819d61364434164e2980b1715ca9c2516c1f
thread:issue:82293,issue #82293: pytorch's checkpoint_wrapper does not save memory while fairscale's checkpoint_wrapper saves huge memory,high-priority/blocker,70,high-priority/blocker,1,0,82293,,,,0,e49fa4ee641ccccdb8d4513400feea56a4d61524e4b966d1daeec183b7235d9b
thread:issue:94765,issue #94765: MPS internal error in `torch.gather` when last dimension is a singleton dimension,high-priority/blocker,70,high-priority/blocker,1,0,94765,,,,0,cb38c283817f46f0811ac2badc39c68c94d1bce699f9f8cae08387c3f8cd44c0
thread:pr:170051,pr #170051: Add pivoted QR decomposition to ATen and torch.linalg,ready for maintainer decision,70,ready for maintainer decision,1,0,,170051,,,0,f6a833851f4a853a0e6cea51bd5e785cb47993b2e1d6cc30368ceb968727e67b
thread:pr:171780,pr #171780: Add support shared mempool for accelerator Graph,ready for maintainer decision,70,ready for maintainer decision,1,0,,171780,,,0,ad3242bfd1b526b712b22510ea4c182b318c8f062fca038c1884ee891c3f72f7
thread:pr:174275,pr #174275: Adjust to use __riscv_v to detect for RISC-V Vector Extension,ready for maintainer decision,70,ready for maintainer decision,1,0,,174275,,,0,272f871df4f2c556dda226390c0a9d4677ebb37554d5aecfd49fbad638035538
thread:pr:174667,pr #174667: Update tolerances for tan* functions w/ complex dtypes on XPU,ready for maintainer decision,70,ready for maintainer decision,1,0,,174667,,,0,844f7ad71df42a15f72ebc1bfcb206697f0c5a475c49cde33f8d2b15cce82f85
thread:pr:174826,pr #174826: add mem_get_info usage notes to CUDA memory management docs,ready for maintainer decision,70,ready for maintainer decision,1,0,,174826,,,0,36cfc4273199601a2365658f5fdd2a302e1f6d2f5331def281d2cfc0703e2316
thread:pr:176321,pr #176321: test: replaces deprecated numpy.linalg.linalg module,ready for maintainer decision,70,ready for maintainer decision,1,0,,176321,,,0,0fed02f961b2b89e20c3cf963727c0de9e6f1b962277ab070cce48d3f1ddaace
thread:pr:176404,pr #176404: Add prioritized text linker optimization for lld.,ready for maintainer decision,70,ready for maintainer decision,1,0,,176404,,,0,b46688b3a15df633b436b67aa0aa419197f42530bb3571f224f2099a11665862
thread:pr:178051,pr #178051: [pytree][dynamo] add generic_pytree (torch.utils.pytree) to test_tree_map.py parametrization,ready for maintainer decision,70,ready for maintainer decision,1,0,,178051,,,0,93f297461783803c51b66eaa86386253983e1a58c3305c9970bb3bfcfca25a77
thread:pr:178258,pr #178258: [submodule] Upgrade to protobuf v33.6,ready for maintainer decision,70,ready for maintainer decision,1,0,,178258,,,0,c97e26ace73f49d109c001d2e426aee2260b45b56a0397af0d7c7b9dd9f9ddd9
thread:pr:180828,pr #180828: Reuse ACCELERATOR_TYPE in test UT files,ready for maintainer decision,70,ready for maintainer decision,1,0,,180828,,,0,d3bd092894ac0fe59e69c53b5023937d364b650fdc45740f5090a01102777aa7
thread:pr:181431,pr #181431: test split group,ready for maintainer decision,70,ready for maintainer decision,1,0,,181431,,,0,4ed005c8527cb092bf4afdda0cd8d33ff0de08c0a345e4a0f36b68b8e8a05382
thread:pr:181614,pr #181614: [TEST] Fix `_debug_handlers` import error in `test_stacks_handler_all_success`,ready for maintainer decision,70,ready for maintainer decision,1,0,,181614,,,0,fa1ce79d0d2ca5547d5ce468d575ce4bc53a7099310cc448a2c050bac6ba3f4b
thread:pr:182014,pr #182014: [inductor] Fix spmd_check deadlock in recursive post_grad_passes,ready for maintainer decision,70,ready for maintainer decision,1,0,,182014,,,0,dd60148fd3a2da016b0f0d30836aa9bec96022e6790f30ac4111e15049fd1b01
thread:pr:182033,pr #182033: Add a utility method for interrupting long running subprocess during run_test.py,ready for maintainer decision,70,ready for maintainer decision,1,0,,182033,,,0,094aed9614692950bd64191c1dd83d6f4ce5ee9fd5ca1f48789ef021f01dd92b
thread:pr:182062,pr #182062: [CUDA] Early-exit in `release_cached_blocks` if `captures_underway` isn't empty,ready for maintainer decision,70,ready for maintainer decision,1,0,,182062,,,0,ff03dff50e68f671ff75a9cab659634443e6b000cfd2c540c7c86adf1bdde85a
thread:pr:182784,"pr #182784: [xpu][test] Port distributed _tools,algorithms,launcher tests cases on Intel GPUs",ready for maintainer decision,70,ready for maintainer decision,1,0,,182784,,,0,eb7e038d7ceb1efafd40c8223fc3872fcce54630d97dea6a257c8aad5ffaa948
thread:pr:182906,pr #182906: Port 11 tests under test/distributed folder to XPU,ready for maintainer decision,70,ready for maintainer decision,1,0,,182906,,,0,723a9ae017d152f32de34154dd301ae663fd26d5043be2c5293157e029e9371f
thread:pr:182924,pr #182924: [Fix] DLPack toDLPackNonOwning: handle MPS storage_offset,ready for maintainer decision,70,ready for maintainer decision,1,0,,182924,,,0,698e88cec0b3dc0470b1fdc3dc3d08c675bdc819714a4eb5610fb944de675e55
thread:pr:183056,pr #183056: [lint] Fix PYI036 violations and re-enable the rule,ready for maintainer decision,70,ready for maintainer decision,1,0,,183056,,,0,aeae2e02927f93f154253c35d9bbec5c2635881ea216e1ff52a6eaa722eb9674
thread:pr:183272,pr #183272: [DTensor] Switch test_math_ops and test_matrix_ops to DTensorOpTestBase,ready for maintainer decision,70,ready for maintainer decision,1,0,,183272,,,0,b97ba58940f5d5ce7ece868a7a43c6d9ad6b39cacf1df5d53efbfa509bdf53c5
thread:pr:183389,pr #183389: Port 7 tests under test/distributed folder to XPU,ready for maintainer decision,70,ready for maintainer decision,1,0,,183389,,,0,6ca6ec13a82071912c249cd9d419357f3b6c3f8b13437166e666211a07bfa96e
thread:pr:184163,pr #184163: [Test]Refactor test/nn/test_dropout.py to be device-agnostic,ready for maintainer decision,70,ready for maintainer decision,1,0,,184163,,,0,4a173338d27d18171bb70f0fd803ecd4c38c4d4b8bcdd0cb05980b446c7d6f5d
thread:pr:184261,pr #184261: [Test] Make test_serialization.py device-agnostic for out-of-tree backends,ready for maintainer decision,70,ready for maintainer decision,1,0,,184261,,,0,d584db6a7b3244264fd9ad2e8bcb156d6d1b1e02e7493c2f13d8fcc549d16455
thread:pr:184545,pr #184545: Fix UntypedStorage.new null allocator crash,has linked issue,70,has linked issue,1,0,,184545,,,0,963118f85fc0af34428609f55f20fb5d30163bf96269c996d58fe4daad14e1da
thread:pr:184593,pr #184593: [Test] Make test_autograd.py device-agnostic for out-of-tree backends,ready for maintainer decision,70,ready for maintainer decision,1,0,,184593,,,0,ef51ee0431dc13dbdb9e2345c3eb48038c93a9a5f86c3e7bad8e4a26c8db72eb
thread:pr:184795,pr #184795: Autogenerated operator!= in C++20,ready for maintainer decision,70,ready for maintainer decision,1,0,,184795,,,0,f6f17dafc78459649002f6724d90a53bad48810e26e322eb2ddcf6a2f8c3f5ed
thread:pr:184947,pr #184947: [dynamo] fix pyright typing on torch.compiler.disable,ready for maintainer decision,70,ready for maintainer decision,1,0,,184947,,,0,c48b4b6e5af51d21a128db771ad77b4d6d5325a92179e8c22a0b07aae4a3a257
thread:pr:185123,pr #185123: [c10d] Add missing runCollectiveChecks for allgather_into_tensor_coalesced in ProcessGroupWrapper,ready for maintainer decision,70,ready for maintainer decision,1,0,,185123,,,0,070d6f2401af6acf6e9b8b536b7c3bf5ac14748721f44b73b2c917dc8f1d8a16
thread:pr:185189,pr #185189: [Test] Make binary_ufuncs file device agnostic for out-of-tree backends,ready for maintainer decision,70,ready for maintainer decision,1,0,,185189,,,0,12a1596b3cabd8fd1d1bc5f7a0ecaa593e5fa9fabbfd158165747619cf97cc34
thread:pr:185207,pr #185207: [Test] Make dynamo/test_aot_autograd_cache.py device-agnostic for out-of-tree backends,ready for maintainer decision,70,ready for maintainer decision,1,0,,185207,,,2,3f2fb0f9c9657e8f377c0d10c2a0a0aa2fa4da2dd11ede22b15cb19e106e794d
thread:pr:185309,pr #185309: Add sparse COO dispatch for linalg.vector_norm,ready for maintainer decision,70,ready for maintainer decision,1,0,,185309,,,1,9f956d080d64a6a7cffa07560706447d2b16d070eeff11bcd00825c20dac0b36
thread:pr:185586,pr #185586: Update README of OpenReg module to consistent with its structure and principle,ready for maintainer decision,70,ready for maintainer decision,1,0,,185586,,,1,e68e2e34cd887a69605c87eaac2b4650a5e1728c67aff85e1be39c6dee9e0280
thread:pr:185798,pr #185798: [Test] Refactor test/test_indexing.py,ready for maintainer decision,70,ready for maintainer decision,1,0,,185798,,,0,976c6c33905bf9a86c9a19cbd6e2b369463d27e1b9b78e828423a8db43c1dbcf
thread:pr:185802,pr #185802: [Test] Refactor test/test_out_dtype_op.py,ready for maintainer decision,70,ready for maintainer decision,1,0,,185802,,,0,e869d77693e0ed4d5c877eb9b230be66793e7829ddd2ca84e18d3088b3ee51ec
thread:pr:185821,pr #185821: [BE][Ez]: Add c10::SmallVector reserve calls and simplify log func calls,ready for maintainer decision,70,ready for maintainer decision,1,0,,185821,,,0,9d13c201b58693bc0f1b699e1bce353f48b6b9c5fb738f6c6b838b9450098bc3
thread:pr:185934,pr #185934: [NVSHMEM] Fix link_whole and -fPIC for nvshmem_device_link_target,ready for maintainer decision,70,ready for maintainer decision,1,0,,185934,,,0,0688bc996139c0dfde4a094d848af2c2f74ca4c452ce52811f8a333b0170ef8a
thread:pr:186012,pr #186012: Add ownership headers to GitHub workflows and a linter to enforce them,ready for maintainer decision,70,ready for maintainer decision,1,0,,186012,,,0,df7a6aa20fe28fae17aba81a262ea4b31ec3e0cb75d98bec87aed8b989f20f11
thread:pr:186038,pr #186038: add distortion benchmark,ready for maintainer decision,70,ready for maintainer decision,1,0,,186038,,,0,d24b0863c6759f294f5a133c898f22162dce88bdd7148c4647f90365aa5e1488
thread:pr:186268,pr #186268: [TEST][CUDA] Add some additional cleanup to `TestCudaallocator`and re-enable `test_memory_plots`,ready for maintainer decision,70,ready for maintainer decision,1,0,,186268,,,0,83f909867cf1db994c6563de77f20363cd6476a6517a1e801640826627a8f4c1
thread:pr:186351,pr #186351: [Test] Refactor test/test_scatter_gather_ops.py,ready for maintainer decision,70,ready for maintainer decision,1,0,,186351,,,0,a9520e7877fb0f032f580c046282514bd59bfcce66e635f723357b3d2e6f5790
thread:pr:186352,pr #186352: [Test] Refactor test/test_ops.py,ready for maintainer decision,70,ready for maintainer decision,1,0,,186352,,,0,5abf78945016fbba68b1d8c7a2bc858d6f63d16abd424351783549d30b0f7a76
thread:pr:186468,pr #186468: Fix incorrect LR curves in Chained/Sequential scheduler doc images,ready for maintainer decision,70,ready for maintainer decision,1,0,,186468,,,0,579293d92b902708ebf0e0a4dde5608ba42311d45b9dd69549ee695f32cfd0f8
thread:pr:186519,pr #186519: c10/util/intrusive_ptr.h: add [[nodiscard]] to query methods,ready for maintainer decision,70,ready for maintainer decision,1,0,,186519,,,0,6ee34305b8899820e912dfc495576218043427aa6bf7cbc3e78ce899ea961898
thread:pr:186575,pr #186575: Update FindMKLDNN CMake configuration,ready for maintainer decision,70,ready for maintainer decision,1,0,,186575,,,0,94d67dc53091541d137334dc29e1f4177c0a5641a0695a019673f683fc45db51
thread:pr:186635,pr #186635: c10/util/ArrayRef.h: specialize std::ranges::enable_borrowed_range,ready for maintainer decision,70,ready for maintainer decision,1,0,,186635,,,0,2d95ca65c2b80583917b70caf19b0419c327d38020aed6d965443734a7612fe0
thread:pr:186639,pr #186639: [CUDA graphs] Add resolve_and_remap convenience for kernel annotations,ready for maintainer decision,70,ready for maintainer decision,1,0,,186639,,,0,2bcf269f69fe18b1087b457a934b869b7f796f27f1ae03b809c2c0e98f928915
thread:pr:186809,pr #186809: [cudaMallocAsync] Count graph-mem pool in memory_reserved(),ready for maintainer decision,70,ready for maintainer decision,1,0,,186809,,,0,c750092ed518422b4276a8efa0eb8866a51a8c476d30d7bf23ce678bd2759698
thread:pr:187068,pr #187068: [torch][profiler] Fix refcount leaks in _TensorMetadata layout/dtype getters,ready for maintainer decision,70,ready for maintainer decision,1,0,,187068,,,0,f103aee12b6383a8ac25861ed7a0eee590a65896a8b21ca3cfd9f4d5b5931798
thread:pr:187130,pr #187130: [BE] Do not try to install CUDA-12.8 wheels in torchbench images,ready for maintainer decision,70,ready for maintainer decision,1,0,,187130,,,0,6b692da16dc9629538b181bca1ad84f60456b927d09a2df54bf889976267caf7
thread:pr:187236,pr #187236: Make Overlap Runtime Estimation Device Aware,ready for maintainer decision,70,ready for maintainer decision,1,0,,187236,,,1,6ee618db1a43f7a76881756afdb7f34b6ce4ec3981a3bce24db5b5c57c82585d
thread:pr:187625,pr #187625: Widen optimizer hyperparameter type hints to accept int and Tensor,ready for maintainer decision,70,ready for maintainer decision,1,0,,187625,,,0,5eb828de848b6f73912c04fd9e15d3b0e41cebdc6d18544c81db39cf58791128
thread:pr:187720,pr #187720: Fix stale_issues year cutoff clamping the day to 28 in every month,ready for maintainer decision,70,ready for maintainer decision,1,0,,187720,,,0,0e04f2cfe5719c942e741c97858c51645b25b12b44b0ab67e40751397cc28403
thread:pr:187731,pr #187731: Fix extract_scripts step-index zero-padding off-by-one,ready for maintainer decision,70,ready for maintainer decision,1,0,,187731,,,0,66cd0ca7c395ff3d1294e2820a4095f468d70fa23c6eec2d64f9e7bd065ab8d3
thread:pr:187779,pr #187779: Fix MPS fused RMSNorm: do the weight multiply in fp32 to match CPU/CUDA,ready for maintainer decision,70,ready for maintainer decision,1,0,,187779,,,0,7a7401fb5c6a6b70e595f065722c90a4079b91e71be2d3ef8833442d67796110
thread:pr:187887,pr #187887: [Inductor] Add decomposition for prims.uniform,ready for maintainer decision,70,ready for maintainer decision,1,0,,187887,,,0,07e8970baef17ea76d1163dd1ad8fbd8ae9d2487391ca53493dc8a07870b4394
thread:pr:187926,pr #187926: [Test] Refactor test/test_shape_ops.py,ready for maintainer decision,70,ready for maintainer decision,1,0,,187926,,,0,bb57a30be4d437d01b67ab6e1055a6c3cc9cf7f72af32b61a0bccc42488f8fbc
thread:pr:188044,pr #188044: [Test] Refactor test/test_view_ops.py,ready for maintainer decision,70,ready for maintainer decision,1,0,,188044,,,0,310ca9ae2a62cc79a0d4d8585bccddfb43099a82e0544a3413fc0bdb35ae5df0
thread:pr:188229,pr #188229: Fix 64-bit indexing in avg_pool3d backward atomic kernel,ready for maintainer decision,70,ready for maintainer decision,1,0,,188229,,,0,e211d9fda48e98bc3da997c4052c6e6dbc26056ca41cfb1865b0037c09ed20fb
thread:pr:188330,pr #188330: Dispatch `index_put_` to `masked_fill_.Scalar` when possible,ready for maintainer decision,70,ready for maintainer decision,1,0,,188330,,,0,0a76f8d19742259c101e7dcd7ce525345c02cf667a08210436a58c799d6c5de3
thread:pr:188396,pr #188396: [Test] Improve test reuse in test/test_tensor_iterator.py for out-of-tree backends,ready for maintainer decision,70,ready for maintainer decision,1,0,,188396,,,2,0b65dda98e7eb041f0aee0798fc323f8702d43931e7092709306bb288468b2f1
thread:pr:188472,pr #188472: [inductor] Bucket all_reduce in the manual overlap scheduler,ready for maintainer decision,70,ready for maintainer decision,1,0,,188472,,,1,7ff5384725ee783a57fc3a17a0da3853e561e8ac2a6fbc2475bd342300174418
thread:pr:188480,pr #188480: [Dynamo] Add UNSPECIFIED_BACKEND lint rule requiring explicit backend=,ready for maintainer decision,70,ready for maintainer decision,1,0,,188480,,,2,d63223a8704852bf5bb6f252aa9e57c4afda8f8aae9ea78d23b7f977fee6abd5
thread:pr:188656,pr #188656: [Test] Improve test reuse in test/test_unary_ufuncs.py for out-of-tree backends,ready for maintainer decision,70,ready for maintainer decision,1,0,,188656,,,0,50ee1ada9a722e72edac269372240feba31d3114c0072cd781adca9d89020575
thread:pr:188666,pr #188666: Enable xpu for TestRefs and TestDecomp.,ready for maintainer decision,70,ready for maintainer decision,1,0,,188666,,,0,b6af187edba6c4ac44ce9db2caaf5ff850baba4b49f3d3fbf22fe304bb4b829c
thread:pr:188785,pr #188785: Upgrade submodule oneDNN to v3.12.2,ready for maintainer decision,70,ready for maintainer decision,1,0,,188785,,,0,e52a6a83dcdcc929bdacec7cab005ea28e75da1266e1f1627755bf9a078cf0f8
thread:pr:188832,pr #188832: [pytorch][cuda] make_graphed_callables: expose capture_error_mode (#188832),ready for maintainer decision,70,ready for maintainer decision,1,0,,188832,,,0,a7f9303f7dd7bcdf412187fffff203e801336923cc07c28634e53c973c37a2fc
thread:pr:188901,pr #188901: Implement Double Backward for `cdist` and `pdist`,ready for maintainer decision,70,ready for maintainer decision,1,0,,188901,,,0,4abc313d1a8f472c6983505831b8c040733cc89291cc4d42a06371704cfc2abf
thread:pr:189004,pr #189004: Add priority&priority_range to torch.Stream,ready for maintainer decision,70,ready for maintainer decision,1,0,,189004,,,0,4b8103cc6a6af6849e4cab056d65b7c1dd366a7be3b621e3735d0c71f877e52f
thread:pr:189016,pr #189016: [Test] Improve test reuse in test_torch.py for out-of-tree backends,ready for maintainer decision,70,ready for maintainer decision,1,0,,189016,,,0,85b8c0aa74232112ea156047691b387ac9be05d435e48641fa3a569773541af6
thread:pr:189017,pr #189017: Respect the blocking parameter in torch.Event,ready for maintainer decision,70,ready for maintainer decision,1,0,,189017,,,0,f6833121d9aa0954d52a66f97e0dbea53ee21dc5927168c207723acc78b3f2b1
thread:pr:189033,pr #189033: Auto-triage open source PRs when a maintainer engages,ready for maintainer decision,70,ready for maintainer decision,1,0,,189033,,,0,5fc92aa352a74469f3386285df1a523f4e1047751ac96378eb795a71fa19bde6
thread:pr:189172,pr #189172: Fix protobuf resource-compiler build with clang-cl (RC1106 /bigobj),ready for maintainer decision,70,ready for maintainer decision,1,0,,189172,,,0,3ca2d939dd2f499f6e9a65c80938796518cd112a13b43fd4d2dc1e10989ec0ce
thread:pr:189187,pr #189187: Retune against triton 3.8,ready for maintainer decision,70,ready for maintainer decision,1,0,,189187,,,1,3eeca369585f09149ae99276225cba78140e8693e7a8c92e0f16cc70efa21b3d
thread:pr:189201,pr #189201: [TEST][Inductor] Fix some `test_buffer_mutation*` tests,ready for maintainer decision,70,ready for maintainer decision,1,0,,189201,,,0,22dc962ced0f84704b82b51e07e1485cfe417e2768b43e856696aaa24c378578
thread:pr:189292,pr #189292: Fix dead ExecuTorch XNNPACK doc link in mobile_optimizer,ready for maintainer decision,70,ready for maintainer decision,1,0,,189292,,,0,df24822b5a1f48013222096ed6d3d35468f61b8805d81a2dbe8bd80dc56ec087
thread:issue:114397:issue:114403:issue:114411:issue:114412,pr #186517: Detect aliased subclass inputs after desugaring,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,8,0,"114397, 114403, 114411, 114412, 114413, 114414, 114415",186517,,,7,77a3038c17b5c90c2433d4e4f4fc629057215906f19702ecf4333214a6dd8cca
thread:issue:167191:issue:173252:pr:173948:pr:185050,pr #173948: Fixes LazyConvTranspose modules when used with torch.compile(dynamic=True),waiting on contributor,67,Wait for contributor update on PR #173948; keep related issue/PR context attached.,4,0,"167191, 173252","173948, 185050",,,4,c687ec5963f7416f1ea40be5fc5f4dec2bb955954b93868e74f98a6681043026
thread:issue:134385:pr:167224:pr:187592,pr #167224: [FlopCounter] Add skip_unsupported parameter for graceful handling of unsupported HOPs,waiting on contributor,67,Wait for contributor update on PR #167224; keep related issue/PR context attached.,3,0,134385,"167224, 187592",,,6,864fdca7300933df9e94d3e56719c6783d5400ca75c908ad72f4bb47eb4cc6bc
thread:issue:136628:issue:148003:pr:185786,pr #185786: Handle non-natural SymPy powers in value range interp,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,3,0,"136628, 148003",185786,,,2,35fbb9a25327f10f0b57f0409e6a0167cba9de96e503550730805e79fdd1b61b
thread:issue:151196:issue:172428:pr:185718,pr #185718: Fix fake functorch-wrapped tensor handling in jacfwd vmap,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,3,0,"151196, 172428",185718,,,2,4dc92186d3363de83fa4a9a6d962e42e8be280b38bdd795aa2b994fb1452fef6
thread:issue:154146:issue:176770:pr:183908,pr #183908: Fix deferred runtime assert symbol availability,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,3,0,"154146, 176770",183908,,,2,a7ba9ba5c802a49873b81b1f0e7ef1f32d6e9c6d714dbb33f90d0719ac96eae9
thread:issue:159550:issue:185575:pr:185369,pr #185369: Fix dynamic output sizes for adaptive avg pool lowering,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,3,0,"159550, 185575",185369,,,2,9dead64f9073a33e9c45e1ee79e7d495384c2d4a9dfca978569c29addfb2f6b8
thread:issue:169779:issue:187463:pr:184987,pr #184987: Fix index_select refs bounds semantics,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,3,0,"169779, 187463",184987,,,4,b872bf5f4dcd8d4bd4703f69826cc063b113b2f30d0036804ccf8b83c8aa51fb
thread:issue:106135:pr:186533,pr #186533: Fix SymInt vector dispatcher signature detection,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,106135,186533,,,1,cfc0113ceb91b460d54fb52fe1e552937d9d362e195ac4947695f52c2d61e656
thread:issue:111320:pr:186531,pr #186531: Remove stale Dynamo skip for max_pool1d pooling test,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,111320,186531,,,1,53fbf494f680c8e7e58465d962c9120113b80d2b152d38147641b4035256ec6e
thread:issue:112788:pr:184436,pr #184436: Fix Triton async compile cache keys,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,112788,184436,,,1,6ec92207484edff8f8ed69d17f5fd447094cc30aa88ea2875be9bfbaa383430f
thread:issue:113809:pr:184432,pr #184432: Serialize cudagraph tree runtime,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,113809,184432,,,1,1874a015689425eb00bd1fa17621503a553eb88fe3a37b8c8df9e967230c810f
thread:issue:117555:pr:186506,pr #186506: Register missing CIA decomposition py_impls,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,117555,186506,,,1,480cf22ecefd7fb0a667ea1437f3a119f184b7e6a4ffcaa3c74625811ee247af
thread:issue:118739:pr:186503,pr #186503: Preserve input aliases for multi-output view returns,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,118739,186503,,,1,6cbb87eb06c98b8812a3177a3676ffad53b77a1e21a2330784ef2a145a55a74c
thread:issue:121318:pr:186495,pr #186495: Initialize compile context for non-strict export tracing,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,121318,186495,,,1,d0c3c92c8c59d98a54ae7ed9e9be60bff7bcc553b762b315c7ce824663302ea3
thread:issue:122129:pr:186488,pr #186488: Improve schema errors for symbolic int list mismatches,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,122129,186488,,,1,d13c3fbcb665a25abcaf4519e69f5a3cc525765d8087af25fe60a4bfea045d78
thread:issue:122386:pr:186478,pr #186478: Fix AOTAutograd resized out tensor mutation replay,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,122386,186478,,,1,a8e6e589990375fabbbfa6ff135479e55a875d551ad38364e6661a6055c150a1
thread:issue:122773:pr:186471,pr #186471: Handle symbolic tensors in overlap status,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,122773,186471,,,1,9a59d0bcf5e6b7baf32b471ac00544b3f04062a707a12d9095a5357ae3acc875
thread:issue:122860:pr:186470,pr #186470: Raise user error for joint parameter mutation export,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,122860,186470,,,1,15eb91dd36ba05ac2231c814bda0625e42c0b998448131a182409807eabf4889
thread:issue:123651:pr:186464,pr #186464: Fix singleton quotient divisibility replacement,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,123651,186464,,,1,bd2219a5f10391bbd251912ffdc6377dedfa42fe1f2205b5db2cea2c81d07aa7
thread:issue:123663:pr:186463,pr #186463: Split foreach bool inplace error samples,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,123663,186463,,,1,f370526bbcc1b7401e8b57991f9de129f84640232eba56725063819d182473f4
thread:issue:124181:pr:186459,pr #186459: Fix fake view strides for AOT group norm tracing,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,124181,186459,,,1,0f45637633196abdf259f0773908f3238fc468aaf567be73c64b6292a02e1c9b
thread:issue:125077:pr:184347,pr #184347: [inductor] Add third reduction tile for strided block pointers,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,125077,184347,,,3,9807f81319f642183336e6adf54452dc5ab02159ca8cb741bd24ee462d1e5a1a
thread:issue:126834:pr:186443,pr #186443: Validate random_ and uniform_ meta arguments,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,126834,186443,,,1,f03a7e46bd5c427383d7d396ac3e63154e22a55b2d13fa458efcc287b5c6a181
thread:issue:127112:pr:186431,pr #186431: Fix ModuleTracker with Dynamo runtime modules,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,127112,186431,,,1,b9f97e1aa7c703257ccf917d42e0f8abc1e3a6e23323237ac2173d8847b8ac67
thread:issue:128649:pr:186409,pr #186409: Fix subclass view replay with dynamic metadata,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,128649,186409,,,3,ae2e69bd8bca64358364f076de3790ab1f7a75614b92d77b81a492e3a0c417b5
thread:issue:135061:pr:186291,pr #186291: Fix export distribution validation assertions,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,135061,186291,,,1,7ea4fd3b40c714511ca3c59cb207f6921523053d39fed614ffc4c671e03fea02
thread:issue:137009:pr:186261,pr #186261: Fix pre-dispatch tracing for saved tensor detach,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,137009,186261,,,1,1b8e7221bb1dbc7f58a6642e493e2852797d561e891031084b4158b1fee79daf
thread:issue:137275:pr:186225,pr #186225: Preserve NJT nested ints in AOT subclass outputs,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,137275,186225,,,1,fdf892a461a86e1af8add93624f144e645d631ae9a8eb3c3e23cc2583c4d775d
thread:issue:138229:pr:186201,pr #186201: Deduplicate symbolic cond inputs after export passes,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,138229,186201,,,1,4e8ad1f87d137ecadb5e1d5a087bd623697f76ffba8da2bc39bc72132d536ec2
thread:issue:138264:pr:186199,pr #186199: Fix functionalization return mapping for mixed mutable out ops,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,138264,186199,,,1,9fff34418eb8401af582bbc5dc78d5f3aaf35524524a01cdf0ef438c5e1945d9
thread:issue:138926:pr:186194,pr #186194: Canonicalize trunc of symbolic integer division by one,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,138926,186194,,,1,e5663b4184b02e59cd58d3887073097cad60d2f7ea32a469435a7eae80d0943f
thread:issue:139092:pr:186188,pr #186188: Fix FakeTensor wrapping for meta tensor constructor,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,139092,186188,,,1,65a0adc34f49a70a68a581ebf7cddc26062248a84920a5bee33b4b1fbf4ddd10
thread:issue:140707:pr:186111,pr #186111: Fix stale mutated input history in AOTAutograd,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,140707,186111,,,1,48ac64f846e9a7a143dd21d6faed763cff1777f9a4f41f11a8619471975757a4
thread:issue:140795:pr:184207,pr #184207: Release max autotune benchmark artifacts,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,140795,184207,,,1,8b2c15d2c8be767103966b917eac563fb6aa7b87c83947763f9e3f359a2dd761
thread:issue:141017:pr:186074,pr #186074: Fix multilabel_margin_loss decomposition sentinel handling,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,141017,186074,,,1,f636ec971407e1dbf4b84e0b5f28e5b0fd22893e7ac673d1ec5402ada3c9cb10
thread:issue:143412:pr:184178,pr #184178: Fix masked Triton conv load pointers,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,143412,184178,,,1,72a7568c307c0e12f7bc7d2eb07912e29b335a023a226b7152d98701c5fac627
thread:issue:144376:pr:186007,pr #186007: Preserve eager post accumulate hook ordering in AOTAutograd,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,144376,186007,,,1,d8fde5a152c7848b0b71e0e2c59fdd82b8ce42990540b0455851bbf3ef05e978
thread:issue:145773:pr:185929,pr #185929: Fix Dynamo benchmark model state isolation,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,145773,185929,,,1,39ed52797b667bbdfea4ff4966c957320bca9607633f25d6bf6ccbb460f2506d
thread:issue:146274:pr:185922,pr #185922: Fix fake one_hot invalid class count checks,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,146274,185922,,,1,bdd088743a0be84442b8e6834190da8ed55cb8fa2fbac0801a37933204e52c1e
thread:issue:147135:pr:185914,pr #185914: Log aot_export fw_metadata instead of printing it,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,147135,185914,,,1,d8a3f3c1ee888fd638dd5d66d877eff235e65dbe1a9e9bf01aad2eac4f532649
thread:issue:148779:pr:185774,pr #185774: Fix non-strict export for Tensor.numpy,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,148779,185774,,,1,f320616094759fe901eaaa84d813036f47095a14d4f8b83cf6fb538f27385a1e
thread:issue:148859:pr:185772,pr #185772: Fix CPU small-output bilinear upsample precision in Inductor,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,148859,185772,,,1,4aa9d705a72443f258caa644b33e9543592a7090660c4c6fadc0fde035adf6f4
thread:issue:149857:pr:185755,pr #185755: Use gmem efficient attention kernel for long fp32 KV,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,149857,185755,,,1,d01d957cea10a77c518f11cfa1ed30dde3d09edb148cdf423e74f5f1201f283c
thread:issue:153227:pr:185569,pr #185569: Handle symbolic bool short-circuit with unbacked terms,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,153227,185569,,,1,bf6bbdc7a2d8f3fc28f7a6a27af54967bc38ea1a8b10f4b0ee9a218c16fc2310
thread:issue:154282:pr:185526,pr #185526: Fix Dynamo tracing of LazyGraphModule,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,154282,185526,,,3,035df51f7f953280ac69fa93621319cb63d135c9109af3da2f769ce76a779e6e
thread:issue:155238:pr:185483,pr #185483: Add fake support for pad_packed_sequence,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,155238,185483,,,1,04cbbc4b51133313e195fb6a2d2ca36d57f39be85b5ab24d81302267c25f1201
thread:issue:155584:pr:184083,pr #184083: Preserve user Triton allocator in Inductor,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,155584,184083,,,1,073ad42016ccc02f0df024c233999110d0a122404b8fc596a3d8f428e756f4d2
thread:issue:158521:pr:184065,pr #184065: Fix static launcher 16-bit float scalars,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,158521,184065,,,1,7cc925b8b0c4f39191f809e21404e9ef8ceaeeac2b5e923a4597d0e692b1e31b
thread:issue:159160:pr:185408,pr #185408: Fix torch logging to avoid closed pytest stderr,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,159160,185408,,,1,7b0dd56e2135e5209e320c68a98740e2122f8cf81e8394f891777112781b65cc
thread:issue:160094:pr:187012,pr #187012: Support optional Tensor list schemas,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,160094,187012,,,1,fa285ccd809aca0383690345ff313a4e2f9730aa848fc486307633c7aaea1367
thread:issue:161088:pr:185293,pr #185293: Require bool conditions for torch.where,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,161088,185293,,,1,6a7c49bb45514f22372979231b9e3b945e36b9dc1556a2c943746edab391c102
thread:issue:161456:pr:185191,pr #185191: Support wrapper subclass storages in HOP alias checks,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,161456,185191,,,1,852e58163a30e09605005867909c93b885427983c261f57d4aac3ad032cd3a24
thread:issue:161705:pr:185166,pr #185166: Fix export slice bounds across dynamic dim,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,161705,185166,,,1,679faffe99345501c151d0bddbcc3947223c278db46e39a4e5c460cca5cf0718
thread:issue:161796:pr:185155,pr #185155: Avoid warning spam for backend fallback graph breaks,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,161796,185155,,,1,a0729888e4ff36bea9063618675b5ad18973794e2a3aae4bdc503b904f4160a8
thread:issue:162287:pr:185147,pr #185147: Fix Z3 translation validation for torch.sym_not,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,162287,185147,,,1,8c4592d483e40180276f8970eb43f73864eebe743d952f54284e542617b301af
thread:issue:162973:pr:185126,pr #185126: Optimize GB_REGISTRY linter scanning,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,162973,185126,,,1,194e6ec3bd8ad74230cbc9aaba04db2a8625f1190484de8bde987b4c97686ac3
thread:issue:163040:pr:185121,pr #185121: Raise TypeError for invalid torch.export.load inputs,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,163040,185121,,,1,72769c7f442601ac6917f6c89e78779e08c03df9f8286ecea78f85ff4c8f256f
thread:issue:164559:pr:185088,pr #185088: Avoid graphsafe RNG inputs in descriptor compile,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,164559,185088,,,1,229fb9fc6ca5645d3cfe318beb1acdb9577dc64a48937a28f75806c0189f9348
thread:issue:166604:pr:184015,pr #184015: [inductor] Handle scalar workaround args in pattern checks,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,166604,184015,,,1,7f31f6c9b072f7686e8534d16e5558ae34f4fb86c3985f232564f510bfa88b05
thread:issue:167098:pr:185052,pr #185052: Fix Inductor reuse for disjoint strided slices,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,167098,185052,,,1,89a142981ed50f96de65e7ddfc0254547ef892e45de0bfdbd5ec0a7f8706a4b2
thread:issue:167117:pr:185051,pr #185051: Fix FX GraphModule serialization of string annotations,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,167117,185051,,,1,749388747dfcba0f9f8cd68c83a3f8a8d4b445835886bfde9492ae6bed1a5eff
thread:issue:168974:pr:185009,pr #185009: Add fuzzy DebugMode hash mismatch comparison,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,168974,185009,,,1,35b02940cdf8c4cbf09d0b011b2f5f74fdbb04d90cad359bffd1732aeb2420cd
thread:issue:168975:pr:185008,pr #185008: Add DebugMode tensor hash export,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,168975,185008,,,1,d77feb5704ac627f9a8e9f939afc86b096f47f95932384e5f6987c3d702ac3cc
thread:issue:169477:pr:184995,pr #184995: Fix AOT getitem replay under inference_mode,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,169477,184995,,,1,4057543ab3d42be95b0af7000dd1053fce67f157387c9a1b6c8483bda12c2ff3
thread:issue:170509:pr:184981,pr #184981: Fix torch._numpy bool numeric subtract,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,170509,184981,,,1,fcab1aa82c8e3a06852d2a21ef6d8a2ac69e6a0b7efc5d48a4c5907ff2b7a0a4
thread:issue:173052:pr:184928,pr #184928: Fix linalg_pinv tensor tolerance broadcasting,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,173052,184928,,,1,3ef1ff858f2f53035e03f77b3f6c5fbcca43b5edf494c270c1ebc65b303a4f6a
thread:issue:173816:pr:184917,pr #184917: Make FX traceback preserve state thread-local,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,173816,184917,,,1,309b30e05aef71025e00e1f36d35695c328ea64a39caa8f2d4108869b548ac1c
thread:issue:175833:pr:183958,pr #183958: Enable SDPA constraint cache by default,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,175833,183958,,,2,0fe830ece4a62a8475b1dcb1ee348192411749924a9b9134b32251efc62f6dc2
thread:issue:177327:pr:183879,pr #183879: [inductor][cpu] Enable int8 WOQ GEMM template on Arm,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,177327,183879,,,1,7f4035d60050c391fb3db7c87f2b57177f887b0b6ec5a26ba815b76f73d9a854
thread:issue:177712:pr:184747,pr #184747: Remove no-op SDPA masks before backend dispatch,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,177712,184747,,,1,ce98c4b484b5d48bd56c75d89e3bdfed86a544cfe8ff3779963d69cb54a9fd44
thread:issue:177750:pr:187601,pr #187601: [functorch] Fix dynamic layers under inference_mode,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,177750,187601,,,1,be05f7d4fe31dfd55dc8dc37b2bdf34fbdfe0d17c3ebe5fdc3ad016da9f74b87
thread:issue:178388:pr:184709,pr #184709: Fix Dynamo tensor subclass constructor side effects,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,178388,184709,,,1,f691646d2fe2f61a149d1027355cb621ff11888309d64f7eb4bd03c4fb975a94
thread:issue:178677:pr:184695,pr #184695: Fix all-masked TransformerEncoder compile consistency,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,178677,184695,,,1,5e0e2b25e01b7df076a6bba96710f1b75319931cecb882e92041d01e8b79f8c4
thread:issue:181304:pr:186859,pr #186859: Recover fake/proxy views from functional tensor bases,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,181304,186859,,,1,bf1686bbe2bc8946db40cd9c5d541e372e7a301281330a2a7f69b7ff8abb69ff
thread:issue:183249:pr:183628,pr #183628: Fix fractional max pool RNG in Inductor,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,183249,183628,,,1,cd56f3dbfe179e8f2a9f7d7b78aa270c5aeba1952bd84c534194f948df984de5
thread:issue:183957:pr:185897,pr #185897: Support concrete enum types in torch.library.define schemas,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,183957,185897,,,1,fc36c433e5dc8c5612a98cdc4ddc31fa92336a3d2b2fc0bb82616e5764406df7
thread:issue:185185:pr:187604,pr #187604: Fix fake slice pending unbacked replacements,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,185185,187604,,,1,088588eb08a38e7d7cb1122aaf155e0ae1fbf0988fce9a2dbdabe05360d9dc80
thread:issue:185630:pr:185861,pr #185861: Fix ViewMeta serialization for special views,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,185630,185861,,,1,514bc1748d642914cddedeb85659e9e36abb372b0447f7fbd15e8cdff5ba0aae
thread:issue:186799:pr:186953,pr #186953: Fix CPU amin/amax backward rematerialization mismatch,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,186799,186953,,,1,f260c1bac2de142d6a04cdcfcc0fd35a85a61401c0e95491be6720447c2afb1a
thread:issue:187945:pr:187956,pr #187956: Propagate FX subgraph replacement meta values,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,187945,187956,,,1,31c9bad60f0354600ad960bee30fc285c73cf422cbe82b2bae97965963be4695
thread:issue:90507:pr:186561,pr #186561: Preserve autograd metadata for functionalized inplace views,ready for maintainer decision,67,Review active PRs and linked issues as one maintainer work thread.,2,0,90507,186561,,,1,b312db58bdff1f53786721135f743e4174980b8af44c626346a9fcb244a8b99b
thread:pr:184079,pr #184079: Enable graph deduplication for Inductor compile,ready for maintainer decision,67,ready for maintainer decision,1,0,,184079,,,2,a49cdcef451fd834a46a3eec0fce946bf880534d5dc298ce54972dc473abff12
thread:pr:185000,pr #185000: debug_mode: avoid hashing async collectives before wait,ready for maintainer decision,67,ready for maintainer decision,1,0,,185000,,,0,8f08a25506e3618e125ca3a4f1f7276f1bc93cd886559b0bf8ce64462ddd8d6e
thread:pr:185079,pr #185079: Fix invalid Dynamo setattr exception timing,ready for maintainer decision,67,ready for maintainer decision,1,0,,185079,,,0,5319dda71ac92af0439da86079338ccf66ad36604ae2f6e2628e01aeb3190b54
thread:pr:185793,pr #185793: Add ONNX lowering for prims.remainder,ready for maintainer decision,67,ready for maintainer decision,1,0,,185793,,,0,646fb21742001b4389f645cc1c251b385e88c64fb8df912a7c95a2dd631d88b1
thread:pr:185995,pr #185995: Fix stale pending-backward state for dead CUDA graph paths,ready for maintainer decision,67,ready for maintainer decision,1,0,,185995,,,0,0f4cc85af4240b3702f68be4fd5536e0e08dab51240957b06ff8483cb3898850
thread:pr:186315,pr #186315: Preserve extra autocast keys on FakeTensor,ready for maintainer decision,67,ready for maintainer decision,1,0,,186315,,,0,0e1e72a14f6246547c522084ceb51a13a8b252cbbbe442ad960a826f23beae48
thread:pr:186321,pr #186321: Fix PythonMod axiom reuse after size refinement,ready for maintainer decision,67,ready for maintainer decision,1,0,,186321,,,0,989315fa8c8d337cb47be9b43b424c32fda249339088c42d1defc06342209982
thread:pr:186322,pr #186322: Add explicit schema registration for TorchBind methods,ready for maintainer decision,67,ready for maintainer decision,1,0,,186322,,,0,d6fba1fd31ed245e3a714b858205f4997f9a378a4d5f7b357ea2bff9e0b5ac66
thread:pr:186328,pr #186328: Preserve strides for functionalized in-place mutations,ready for maintainer decision,67,ready for maintainer decision,1,0,,186328,,,0,6c836b006a85c19e3dcb3a78a16427c12b3a1d8c67f765307155ddb176f8fce8
thread:pr:186339,pr #186339: Improve torch.compile sibling output grad error,ready for maintainer decision,67,ready for maintainer decision,1,0,,186339,,,0,ddd33dd1995588acbbd12db6302e83234def21db040d1256db3af68f02c33e21
thread:pr:186342,pr #186342: Preserve nested fake subclass inputs in AOTAutograd,ready for maintainer decision,67,ready for maintainer decision,1,0,,186342,,,0,1ea22b7c8d318e2421e954d77f92a52b11ac38210e831884b3d46f45527b755b
thread:pr:186527,pr #186527: Support export of packed LSTM with data-dependent lengths,ready for maintainer decision,67,ready for maintainer decision,1,0,,186527,,,0,0e49a04ab4c62d21eebce9a4f634649c086e715552d64c4fba2094ef6bab5310
thread:issue:108406:issue:119526:issue:157807:issue:160875,pr #180247: Migrate build system from setuptools to scikit-build-core,waiting on contributor,65,"Wait for contributor update on PR #180247, #180248, #187633; keep related issue/PR context attached.",17,0,"108406, 119526, 157807, 160875, 163311, 164469, 169479, 179374, 180624, 183913, 188898","180247, 180248, 180250, 185208, 187633, 189263",,,25,bf5102438cebf626d16d2b5bef1af7154210f738c4fb47658d6418befcb2c67e
thread:issue:147179:issue:153302:issue:158793:issue:159013,issue #147179: [FSDP2] OOM when use integer `reshard_after_forward` that smaller than DP size,stale/low urgency,65,stale/low urgency,10,0,"147179, 153302, 158793, 159013, 159014, 159015, 159017, 159018, 159019, 177815",,,,10,17201016dac0efc41e56d3eda52acc0551f1fd877426ca9a109789534a0d7af8
thread:issue:50444:issue:51135:issue:51136:issue:51137,issue #50444: [TorchScript] Rough edges discovered during type system rewrite,needs triage,65,needs triage,10,0,"50444, 51135, 51136, 51137, 51138, 51140, 51563, 52920, 53017, 53022",,,,15,82f8ae57bf80d82c0b71f4cd33e1a52845e26ef80ef73ca91e82b428521826e9
thread:issue:146326:issue:146328:issue:146329:issue:146330,issue #146326: Tracking issue for Improvements for DeepSeek,needs maintainer decision,65,needs maintainer decision,8,0,"146326, 146328, 146329, 146330, 146331, 146332, 146368, 146542",,,,8,ca03775edd415f197adb48037ec2bc15016b2b37b6dabc0828d7b96136c97090
thread:issue:113045:issue:114299:issue:120003:issue:147662,issue #113045: [Feature][DTensor] Manage additional `_padded_local_tensor` attribute,needs maintainer decision,65,needs maintainer decision,7,0,"113045, 114299, 120003, 147662, 158657, 174960, 187615",,,,7,29de69f2166e285e54acd054a77f34a7f6c35be66c186e59fbc1ea9ca64ea315
thread:issue:44511:issue:55585:issue:67589:issue:67590,"issue #44511: LR scheduler throws warning when using scaler.step instead of optimizer.step, and when saving optimizer state",stale/low urgency,65,stale/low urgency,5,0,"44511, 55585, 67589, 67590, 76282",,,,5,b8b193957bf34cf7f273f490a40f86dd24ef900ef3b118fb5910ffb02ca37c38
thread:issue:107298:issue:115388:issue:119215:issue:75097,issue #107298: dist.destroy_process_group did not destroy the process group well,stale/low urgency,65,stale/low urgency,4,0,"75097, 107298, 115388, 119215",,,,4,d5bcb2286d627748d00f6bbe5a77df7f3cb21bbf88da07705080ea64b1dcd09c
thread:issue:166291:issue:177427:issue:186601:issue:187239,"issue #166291: ""FSDP does not support non-contiguous parameters yet"" prevents using channels-last convolutions",needs maintainer decision,65,needs maintainer decision,4,0,"166291, 177427, 186601, 187239",,,,3,2f3ffa558056fcbe0fe6140f0dd649a5524ff921d2ee8f29925e306139462317
thread:issue:174469:issue:176877:issue:185142:issue:185590,issue #174469: [RFC] Overview of Test Case Refactoring in PyTorch,needs maintainer decision,65,needs maintainer decision,4,0,"174469, 176877, 185142, 185590",,,,3,8f78126ce87fc22a3d04e100f3fd077c872b57319dfcc39960c6a27c413574f4
thread:issue:145949:issue:147692:issue:165727,issue #145949: [CUDA][Blackwell] Blackwell Tracking Issue,stale/low urgency,65,stale/low urgency,3,0,"145949, 147692, 165727",,,,2,af5533562c8c05e4088b988c4ce1c79abca99f634fa8f2914ed345efe8b8ad23
thread:issue:165230:issue:165231:issue:187094,issue #165230: _int_mm is slow with row-major rhs matrix,needs triage,65,needs triage,3,0,"165230, 165231, 187094",,,,2,c344db31661fd1fb803f82df69bf5114e9ac7ec8b89b228245736f590984ca59
thread:issue:172568:issue:186107:issue:186465,issue #172568: Control flow operators issue tracker,needs triage,65,needs triage,3,0,"172568, 186107, 186465",,,,2,e1f3c7a520297305d5b755789b7a96b37caf6213616e59de3c4c688ceba6b435
thread:issue:180284:issue:182641:issue:182661,issue #180284: Unused jvp function on autograd function causes inscrutable shape error,needs maintainer decision,65,needs maintainer decision,3,0,"180284, 182641, 182661",,,,2,8b43bf51d237c6fdc3ca09f05d0a26b41713850d7dab8177e4d0eec58cc40b40
thread:issue:182382:pr:182377:pr:182379,pr #182377: [DTensor] SymmetricMemory-backed DTensors,ready for maintainer decision,65,Review active PRs and linked issues as one maintainer work thread.,3,0,182382,"182377, 182379",,,3,4875ecbaaeffba81affb6666f312da7aa24a24fd63722a2bbc202b7e6518e37a
thread:issue:167631:issue:167645,issue #167631: `jit.export` analoge for `torch.export`,needs triage,65,needs triage,2,0,"167631, 167645",,,,1,2582cf3976f6ccb687d3c9dc9664bb3e97c0e89a7b398d671601aecd7ec62250
thread:issue:173546:issue:174081,issue #173546: [vllm] [release 2.10] V1 Test e2e + engine : test_draft_model_correctness[False-args0],needs triage,65,needs triage,2,0,"173546, 174081",,,,1,a0e3f663433ea2680c6c290f5b9cf44ab3129129bb2468f1a3a14597b783777d
thread:issue:177254:issue:177258,issue #177254: AArch64 Unit Test Failure - test_aot_inductor_arrayref - RuntimeError: unknown architecure,needs triage,65,needs triage,2,0,"177254, 177258",,,,1,deafa6621a04a0474804c51df4d983bfda5591308ca31e416d711492efcff128
thread:issue:189098:pr:189097,pr #189097: [RFC] Expose Kineto custom logger registration,draft/noise,65,Review active PRs and linked issues as one maintainer work thread.,2,0,189098,189097,,,1,ac9dfd32cf63f68f99a9a3faa80ca37619210a745f935b5c3b2da0a8619b7b8e
thread:issue:49961:issue:50112,issue #49961: rfc: automating the switching of inputs to the device of the params,needs triage,65,needs triage,2,0,"49961, 50112",,,,1,70b113e534131c5dec0e4c59eb04a33a18e00b9af37cefd91abcb5ed8e408565
thread:issue:100310,issue #100310: Using ddp training with different machine,needs triage,65,needs triage,1,0,100310,,,,0,3abc4c24f62bacab5dfd62d02a33e485954b65deb1b9cf0a698d577ba4370f58
thread:issue:101962,issue #101962: SummaryWriter background thread holds the GIL for too long,needs triage,65,needs triage,1,0,101962,,,,0,2d0255a9940475cc16f9a1b88fa3ef91f7ba925619ce494ca454451051d96adc
thread:issue:102186,issue #102186: torch/distributed/_spmd/api.py should aot_module_export instead of make_fx directly,needs triage,65,needs triage,1,0,102186,,,,0,2c6905f3870fd58359cf8be5fa906168d1604f38570e9f9c6ecaf74c85a79b74
thread:issue:102596,issue #102596: Jetson NX with torch 1.12.0 :cannot import name 'ProcessGroup' from 'torch.distributed'.,needs triage,65,needs triage,1,0,102596,,,,0,8fc97d1ce53514baacf4279744ee01693c93d75eba082e287793ae886608d517
thread:issue:103221,issue #103221: [help] did torch.distributed.launch can be applied on k8s cluster with pytorch-operator,needs triage,65,needs triage,1,0,103221,,,,0,7dc9142439ef04b79d3336aa1833c26a377cf05101f767f016c69a029497be3e
thread:issue:106633,issue #106633: Refcount problem for torch.distributed.Store objects defined in Python,needs triage,65,needs triage,1,0,106633,,,,0,f83064a2dadd576a59b50385b2c1423d1719a87238375bd0975ccdd655de44bc
thread:issue:106690,issue #106690: DDP grads not synced when static_graph=True and module output is a dict subclass?,needs triage,65,needs triage,1,0,106690,,,,0,4ae1628f7048c0c3e10c3bb0ea8a91fe5c56a60ab34e9f0f2a21da7645d95979
thread:issue:106711,issue #106711: 'CUDA out of memory' when using a GPU services for reinforcement learning in Torch rpc tutorial,needs triage,65,needs triage,1,0,106711,,,,0,97ec192fc6ef7a8ff9d67ead0558bac017a49f7e7ecbb43cb4a8379e39c5cf49
thread:issue:109941,issue #109941: Need latest NCCL support to reduce GPU HBM consumption,needs triage,65,needs triage,1,0,109941,,,,0,0907bca47fb4464206d7e6809bfe999e3a00374129706c0e2b44a0e99e2be7ef
thread:issue:111138,issue #111138: Module states cannot be fully synchronized due to the DDP broadcast_buffers breaking change,needs triage,65,needs triage,1,0,111138,,,,0,ba60e6a9cbe1555597d2e74fbdbe3fa49147467c96db862311d40d1a2f7324a1
thread:issue:112152,issue #112152: DDP backpropagated gradients not the same across all gpus when forward inference not using all declared modules,needs triage,65,needs triage,1,0,112152,,,,0,30b62930219d57e06071574a49d755cf6c27634992e92108f2df8b2bad83b06f
thread:issue:114396,issue #114396: is PGNCCL abortCommsFromMap working correctly?,needs triage,65,needs triage,1,0,114396,,,,0,35b794948250bca314e58331ca9497f8cd1c4daec3edf5356b2c4e8e7d91ec9a
thread:issue:119346,issue #119346: DISABLED test_source_multithreaded_basic_work_in_main_thread_False (__main__.TestProfiler),needs triage,65,needs triage,1,0,119346,,,,0,201708d5cceef6336e975dbfecb99ea239ae7d2a10338723eab8f8a66b62fa9b
thread:issue:119357,issue #119357: DISABLED test_source_multithreaded_basic_work_in_main_thread_True (__main__.TestProfiler),needs triage,65,needs triage,1,0,119357,,,,0,445d24004dd7a1a424866f48b614844fa707466f64e1f16bbc89ca8210668882
thread:issue:119364,issue #119364: DISABLED test_source_multithreaded_close_in_scope_work_in_main_thread_False (__main__.TestProfiler),needs triage,65,needs triage,1,0,119364,,,,0,bbaa284a26f0035694155d1988334931b442e60d06e98ce16c8469dc3edb4831
thread:issue:119490,issue #119490: DISABLED test_source_multithreaded_complex_work_in_main_thread_False (__main__.TestProfiler),needs triage,65,needs triage,1,0,119490,,,,0,8b4f7162e058235f1f60c637771a149d163c78fbcc2f59a21a8206d358c45c30
thread:issue:119514,issue #119514: DISABLED test_source_multithreaded_close_in_scope_work_in_main_thread_True (__main__.TestProfiler),needs triage,65,needs triage,1,0,119514,,,,0,13bbebc2173285c04091911aab4ebd6b0cd9e357680f6d78221e0337d28b5112
thread:issue:119536,issue #119536: DISABLED test_source_multithreaded_complex_work_in_main_thread_True (__main__.TestProfiler),needs triage,65,needs triage,1,0,119536,,,,0,aa13bc4ace72673bb52d29324f5fae6a9cdc1e9da438120aab5958dddef10814
thread:issue:119537,issue #119537: DISABLED test_source_multithreaded_open_in_scope_work_in_main_thread_False (__main__.TestProfiler),needs triage,65,needs triage,1,0,119537,,,,0,54cbd2ded014948768e44a0d6ad1220242b246f448e66f963224cb168184f05e
thread:issue:119576,issue #119576: DISABLED test_source_multithreaded_multiple_preexisting_work_in_main_thread_True (__main__.TestProfiler),needs triage,65,needs triage,1,0,119576,,,,0,fb33aaf375925a917b70b7dc2c9283fee0e45c2911931d638c2c1c4f76028522
thread:issue:121471,issue #121471: [profiler]Not passed shapes for lists of tensors to kineto_result,needs triage,65,needs triage,1,0,121471,,,,0,60e5d5ecf18f74c2cfa32f6d2ad1938f4e3d3f9dab2dc2ed793e882399da3428
thread:issue:123983,issue #123983: Segfault in TCPStore and FileStore compare_set(),needs triage,65,needs triage,1,0,123983,,,,0,541d3340dae3ab44eac5d3d44bdedfcfe6ef41a276b804c8d41ab218ade5f960
thread:issue:124328,"issue #124328: PreDispatch tracing doesn't produce correct ""val"" for Sparse Fake tensors.",needs triage,65,needs triage,1,0,124328,,,,0,54243747a933e83b992b7437e64ca372fc70db5e510af27cfbd585cdd478d3ca
thread:issue:124666,"issue #124666: When using DDP, model outputs that are are custom dataclasses would make things silently wrong",needs triage,65,needs triage,1,0,124666,,,,0,0ce9148d7af018af72d133f02c663d8cbcb68ca3d4262bc4df34622973d4192b
thread:issue:129341,issue #129341: NCCL Blocking Send/Recv are Non-blocking in practice,needs triage,65,needs triage,1,0,129341,,,,0,b219fcb13301e91bea644692ff7d80891d71b815601091c1bb9e0a0c2cc7086b
thread:issue:130113,issue #130113: ncclUnhandledCudaError: Call to CUDA function failed.,needs triage,65,needs triage,1,0,130113,,,,0,dbecae49fa8cbebeea50067017dd0bd31c263070f10a5d11f7bc5d41308fa47c
thread:issue:130251,issue #130251: [funcol] numerical error when funcol.reduce-scatter a non-contiguous tensor,needs triage,65,needs triage,1,0,130251,,,,0,d351694bc5173b0a2b4d22f25dda2b71cb538c4981f82c74007e67993f22b39b
thread:issue:130345,issue #130345: Provide shutdown() method for ProcessGroupGloo distributed backend,needs triage,65,needs triage,1,0,130345,,,,0,206d70ff0c79f14b12c34561e3d5602b4f60c279e3e50d6c148125c9f35afb34
thread:issue:130515,issue #130515: [Bug Report] inconsistent global src and local rank in dist.nn.functional.broadcast,needs triage,65,needs triage,1,0,130515,,,,0,ea98b4fb98b3cd3dc3f3a5b0882d38e1e50033bae879ee1dc63d7d414f03b35c
thread:issue:130958,issue #130958: Document ProcessGroupOptions,needs triage,65,needs triage,1,0,130958,,,,0,6438c4e417e565247978d09edef16e8c053aa7bcca3a30660abddb5b6744209c
thread:issue:137007,issue #137007: [RFC] ProcessGroupNCCL uses non-blocking API by default,needs triage,65,needs triage,1,0,137007,,,,0,6f8ae405a6a474646756709777fd868df86628efcb67f65adbe1be06a0254bb3
thread:issue:137016,issue #137016: [RFC] Accelerate intra-node communication via NCCL buffer registration,needs triage,65,needs triage,1,0,137016,,,,0,31dc7d82a9d3dda7b95462124188322e968c5bff71ef0dbeb4ed16f8c155eb97
thread:issue:140183,issue #140183: FSDP Hybrid Shard worse loss than than Full Shard,needs triage,65,needs triage,1,0,140183,,,,0,78eadee545d095fb792ad5fb376ea5fc093960bf255d9d8e5f69bb24a3ce2002
thread:issue:141668,issue #141668: [inductor][cpu] AMP CPP wrapper performance regression in 2024-11-25 nightly release,needs triage,65,needs triage,1,0,141668,,,,0,d4e4cb17692ab29a86937e81d01a7188cce7e21d7db75b71301c4466bb1273b1
thread:issue:143536,issue #143536: FSDP learning hangs when the program tries to save the model,needs triage,65,needs triage,1,0,143536,,,,0,262de81a9428b11ee8157689062622164eeb0c605f8f07eca0c7819bb0e2ad39
thread:issue:143574,issue #143574: Torch elastic restart fails with torch-2.6.0 nightly build: NCCL unhandled system error,needs triage,65,needs triage,1,0,143574,,,,0,4146fc189ee462120f05d972f04da39941ecc6ab9bcf12de03d6327066c2b988
thread:issue:143593,issue #143593: TCPStore crash when initializing from multiple threads,needs triage,65,needs triage,1,0,143593,,,,0,dce9fc57e29671a2618f211a44441845afc2ead5805cd7b0b53fe0821753d3aa
thread:issue:144359,issue #144359: Incorrect Results with Tensor Parallelism,needs triage,65,needs triage,1,0,144359,,,,0,a0a78f52ab13ddbf1c3f205979a654d8a6cc95c39537e42c0828aaf550977414
thread:issue:144663,issue #144663: Support loading and executing a ExportedProgram from torch.export in C++ environment,needs triage,65,needs triage,1,0,144663,,,,0,adf5c8af1efcc9092569e1e320c67b907e2b39d2500db9eeea25eefcc73d5f6c
thread:issue:144683,issue #144683: torch.export treats two of the same parameters as the same node,needs triage,65,needs triage,1,0,144683,,,,0,3f00dc0349da05aac27cfa3aa387fb714eaa732247c166d4fd94f1a90a21dc0f
thread:issue:146152,issue #146152: [torch.export] Cannot export TorchVision fasterrcnn_mobilenet_v3_large_fpn,needs triage,65,needs triage,1,0,146152,,,,0,d42eeb9a4c975f8e816c97f322ca5c8629a7da7984f20c7c853a7b4bfed5d10d
thread:issue:146907,"issue #146907: INTERNAL ASSERT FAILED at ""/pytorch/torch/csrc/distributed/c10d/reducer.cpp"":2134, please report a bug to PyTorch.",needs triage,65,needs triage,1,0,146907,,,,0,f7234225a4800bbc4a48467826cd8040ef8ba4a3cddb8c620e823bfd3d88a06c
thread:issue:151525,issue #151525: [inductor] [cpu] [silent incorrectness] `nn.LazyConvTranspose2d-torch.randn-F.linear-torch.argmax` output incorrect results on CPU inductor,needs triage,65,needs triage,1,0,151525,,,,0,c490328130ca5a6ff801411eba0587956cb18915acd316122dff573abcaf4d8e
thread:issue:151654,issue #151654: [sparse][exported-graph] Wrong Exported Graph for the Sparse Tensor,needs triage,65,needs triage,1,0,151654,,,,0,4085493e51560ca5220755a9c11a1d73184fedc6e1ba3ae7cd9a4c69b171f96a
thread:issue:151726,issue #151726: Exported Module cannot call train() or eval(),needs triage,65,needs triage,1,0,151726,,,,0,0cd43c55f7153545dd3c7df4218e280ad3ba052e43b4943aebe5b4828977a1eb
thread:issue:152532,issue #152532: The 2.7.0 release tarball is missing `.ci/docker/ci_commit_pins/nccl-cu12.txt` required for building,needs triage,65,needs triage,1,0,152532,,,,0,f0851db53d64060800a5ca4473525f20ee146c2273dbc5fa74fe82b0e7294fa6
thread:issue:153705,issue #153705: Unable to export a model using scan with inplace modification,needs triage,65,needs triage,1,0,153705,,,,0,6d38209b76bb1454fa45eddb3791affbf7eb7459bc95c6d3cafd50ac735cac5f
thread:issue:153779,issue #153779: NCCL timeout while training on multi-gpu,needs triage,65,needs triage,1,0,153779,,,,0,4bf55ba8071701bd33b2ce1d30df270a2bb94526f83f280d0674085b027c95ad
thread:issue:153949,issue #153949: Profiler does not capture steps on Blackwell,needs triage,65,needs triage,1,0,153949,,,,0,996cf8c4faac90bbe2718940d432adf0ce63dff6dc962eabc16fcb7a55896c76
thread:issue:154354,issue #154354: integer convolution,needs triage,65,needs triage,1,0,154354,,,,0,aa341a853f381386954e38f0e85b311da8ead8da28000ea4d272ad0bd5bb077b
thread:issue:155924,"issue #155924: Activation Checkpointing breaks ""torch.distributed.checkpoint.state_dict._get_fqns""",needs triage,65,needs triage,1,0,155924,,,,0,cba984d63418adea56fdd461eddc41bbc2c477079d4d4fe734b485f8e72cd422
thread:issue:161147,issue #161147: RuntimeError in DTensor evenly shard view op,needs triage,65,needs triage,1,0,161147,,,,0,76b6302379faae02e92743b99f2fee53f758910f7684423f940957d395175e46
thread:issue:162878,issue #162878: CUDA 13.0 Segmentation Fault in test/test_numba_integration.py::TestNumbaIntegration::test_array_adaptor,needs triage,65,needs triage,1,0,162878,,,,0,639672c9dbe08aebc7db51be05b03c564e553e3cabf6319eba49966d6701663d
thread:issue:163193,issue #163193: Incorrect results when adding partial DTensor with a scalar,needs triage,65,needs triage,1,0,163193,,,,0,15530ee9e1fe2156ebb1c35d975ba4162ab21e8deb575ca463a091cf690be009
thread:issue:166205,issue #166205: Enable Graph Capture & Profiler Integration for PrivateUse1 Backends,needs triage,65,needs triage,1,0,166205,,,,0,a0ad756f37cc976391e89cba4e301a20785d52ceb923c9271cde9857656fcfb0
thread:issue:166301,issue #166301: Improve Docs: Reproducibility and Randomness in dist training with multi-process data loading and explicit RNG seed,needs triage,65,needs triage,1,0,166301,,,,0,0505d1984ead8a4e07f70ba2f039afc2ca0544f4bf21fdbbcfd7343e35c55a4b
thread:issue:166572,issue #166572: [Regression] torch.export.ExportedProgram.run_decompositions Starts to Break in Torch 2.8 with Mutable Buffer,needs triage,65,needs triage,1,0,166572,,,,0,5bb816bde4f5017f8551238530d855fa3b4d79811dff68d77f92d7ec3ed87dac
thread:issue:166577,issue #166577: Pipelining - cannot calculate dWeight if a custom torch.autograd.Function is used with a schedule that performs separate dWeight and dInput backward passes,needs triage,65,needs triage,1,0,166577,,,,0,8e88c87de2c9fccd22cb21cbb3e704614186c047d05d47f7f2e867807315476e
thread:issue:167304,issue #167304: RPC cannot run in jetson orin because of the specific uuid of orin,needs triage,65,needs triage,1,0,167304,,,,0,369b466fac0b5323903a34d231fe2ed59e1244c19c31e4ac79594a92005d4390
thread:issue:167623,issue #167623: torch.export: support external weights (.pt2 format),needs triage,65,needs triage,1,0,167623,,,,0,8bcf041152626f940934710aca0899ba4c9940dbd1c267e871aad239649eba3a
thread:issue:167966,issue #167966: Symmtric_memory: torch.ops.symm_mem.all_to_all_vdev_2d restricts the number of experts per rank (ne) to less than 16,needs triage,65,needs triage,1,0,167966,,,,0,e6222e69e4823d2b906da806b757f9081320400e2b5dee7a94e7bd5f25b95383
thread:issue:167975,"issue #167975: Multi-node distributed training hangs indefinitely after NCCL initialization completes, specifically blocking at `verify_param_shape_across_processes` in `torch.distributed.utils`.",needs triage,65,needs triage,1,0,167975,,,,0,0944d5d90ce1dfef0137d0a248c1e98b44a9d569331a67076380875d4aac0c48
thread:issue:169448,issue #169448: FSDP2 model saving broken unless ignoring frozen params,needs triage,65,needs triage,1,0,169448,,,,0,0975b4b08d881dff1dec1a533cca2857ad2eec0b70023677f20d48d130861799
thread:issue:169970,issue #169970: Does torch._grouped.mm work with cudagraphs over multiple nodes?,needs triage,65,needs triage,1,0,169970,,,,0,44f8d2fa5c053565e5ff9f1764ffca8975d51ad02ad0a9c22d750d910d9f4aa5
thread:issue:170045,issue #170045: ncclUnhandledCudaError: Call to CUDA function failed. Last error: Cuda failure 999 'unknown error',needs triage,65,needs triage,1,0,170045,,,,0,67c2c41e6420e853e01d3a8a2eebf6263ef39661353b7efc1fa9a6e7a5964034
thread:issue:170707,issue #170707: Update Rendezvous etcd backend to use etcd v3 API,needs triage,65,needs triage,1,0,170707,,,,0,4a4d1d5a7d7ed71592e88ab974f60de84db704f42eaeb26e9f619d1f73ed3d4c
thread:issue:171689,"issue #171689: After enabling XPU Graph, I cannot collect profile data during the replay phase.",needs triage,65,needs triage,1,0,171689,,,,0,17264c4bc5c61b44ef45215fb01d1d7383016ea77a1746054563cf9e70b00c32
thread:issue:172082,issue #172082: torch.ao.quantization.observer work on CPU (very slow!) due to using registered buffers on CPU,needs triage,65,needs triage,1,0,172082,,,,0,41b841d74b49e110dc3a57bc3919715a0ec59ebce87b4767059719291bf094f7
thread:issue:172218,issue #172218: `_expand_group` can get DeviceMesh ranks more directly,needs triage,65,needs triage,1,0,172218,,,,0,97f1bf0de2c1df10071031cd959b30c9c58d6f68dfb258b6ce422c994b579b00
thread:issue:172348,issue #172348: Build failure in NCCLSymmetricMemory.cu: undefined identifier `offset_`,needs triage,65,needs triage,1,0,172348,,,,0,4f64a688babc3ab03be8383555521425613c0c641d3111b04d1d1b3480a97689
thread:issue:172561,issue #172561: [export] Constraint expression suggested by torch export is not supported,needs triage,65,needs triage,1,0,172561,,,,0,ab584afed7c626809a08abde71aab3e1f00bca5191d4a32db0ba8b47424cec94
thread:issue:172617,issue #172617: Symmetric utils missing,needs triage,65,needs triage,1,0,172617,,,,0,2cc0bc04bb422cd0e45608c1d331b36776059fc780f944aab92c8806ce63e02a
thread:issue:172873,issue #172873: Need for all_reduce to initialize NCCL communicator,needs triage,65,needs triage,1,0,172873,,,,0,95000b07d27432b49f4c36831a5d7c6ca50daaa0706bae3c0cfad4179bdd092a
thread:issue:173075,issue #173075: UniformQuantizationObserverBase computes wrong scale and zero_point,needs triage,65,needs triage,1,0,173075,,,,0,f9a8e1cc521fd91515a831acc79b9febb72ac98555ced814fee41007b2b52f58
thread:issue:173311,issue #173311: Issue: Remove unnecessary restriction on num_thresholds in TensorBoard PR curve summary,needs triage,65,needs triage,1,0,173311,,,,0,5c469cf349aad8f33cbfd6effe38303942e1f82e98a08d2c80fdea9b161492fa
thread:issue:173492,issue #173492: torch.jit.script failed for inplace bitwise_and,needs triage,65,needs triage,1,0,173492,,,,0,564f07dd6564b3e51abfce7874e2bd1e8f5587533f1b38a4f9df4e9c1e0c8911
thread:issue:173680,issue #173680: Large CPU overhead with FSDP2,needs triage,65,needs triage,1,0,173680,,,,0,c4d7574ea6072c973ccd0196867cc9627bc2aa364192e99f9d92d879d52c566b
thread:issue:173815,issue #173815: [RFC] Abort a DeviceMesh,needs triage,65,needs triage,1,0,173815,,,,0,b67781e5e35dd84017daf638eeef797ec7d63c466318843357071cce7873416a
thread:issue:173988,issue #173988: [DTensor] Implement decomposed sharding propagation for SDPA operations,needs triage,65,needs triage,1,0,173988,,,,0,001cb059c353774eac123f46d7f6fd5725536fa3bca0c9cfa08eb0325b3601de
thread:issue:173990,issue #173990: [DTensor] Propagate 'grad_dtype' attribute in 'to_local' and 'from_local' operations,needs triage,65,needs triage,1,0,173990,,,,0,9c1840d0c09904b8c9cda397503e7c8e4fe0753a09546086a3befd8b41285b2c
thread:issue:174273,issue #174273: using profiler on jvp function with record_shapes=True causes memory leaks.,needs triage,65,needs triage,1,0,174273,,,,0,d74e162a5dc04c68007f8321323f5db96b3117552a56a5fce6c940ca06f9f51d
thread:issue:174276,issue #174276: [DTensor] Avoid forced redistribution in layer_norm strategy,needs triage,65,needs triage,1,0,174276,,,,0,2abee111c809d16787df0a3f8953313a9558b2b709114230547daea97d8a885a
thread:issue:174375,issue #174375: [CUDA] flamegraph.pl download fails on cross-device filesystems,needs triage,65,needs triage,1,0,174375,,,,0,29d6a9bf52b12a3edd7a8348c2b5f90cd2b0b78e25ae60e0a88ef0b237086b7e
thread:issue:174488,issue #174488: [cd][testing] Add almalinux/9-base test to CD pipeline,needs triage,65,needs triage,1,0,174488,,,,0,3df16197543d32c7a708a2beb5dd88641fa5ad825b27b986ea92b4cac1b84f12
thread:issue:174555,issue #174555: Profiler chrome traces express record_function and profiler step incorrectly when using with_stack=True,needs triage,65,needs triage,1,0,174555,,,,0,14759ef0cbe973dbf81ae49e2685df1c8168375736c9e26a4358494a01f75d8c
thread:issue:174669,issue #174669: `TypeError: cannot pickle code objects` with Python 3.13,needs triage,65,needs triage,1,0,174669,,,,0,af5d2a1e4d9a064137a348f599ad72c91e405ed0543975edaa92746f558873c4
thread:issue:174732,issue #174732: [DTensor][aot_eager] compiling sharded embedding with dynamic seqlen fails,needs triage,65,needs triage,1,0,174732,,,,0,693b06c644016018a98e90d0795e5d28cfab41c291dd2ada15df556dad7e356f
thread:issue:175045,issue #175045: Consolidate or retire pytorch/almalinux-builder Docker images after MAGMA deprecation,needs triage,65,needs triage,1,0,175045,,,,0,8041bb18d4e447ae20b87c8eb580696d03cbc773e55d5f29432775aa68e9d9da
thread:issue:175145,issue #175145: [torch.disitributed] Persistent Deadlock when overlapping NCCL P2P and Collectives from separate Python Threads,needs triage,65,needs triage,1,0,175145,,,,0,830feb88c8f34873ccd217ffa98c1f3114689b5fb509bd7c95805518c3d6aeed
thread:issue:175293,issue #175293: TL-parse does not work for export.,needs triage,65,needs triage,1,0,175293,,,,0,407df430b729c0e0f1ba824e51b39af39a9c59a8e15f07cd7d438ead4e666eb9
thread:issue:175469,issue #175469: [DTensor] ExportedProgram.run_decompositions() fails with AssertionError: out is not NotImplemented,needs triage,65,needs triage,1,0,175469,,,,0,11dfa724071941500c5b64ac46f7603f9af4eb5cacb919028265b9bf97d27923
thread:issue:175493,issue #175493: FX GraphModule saving with TorchScript submodules,needs triage,65,needs triage,1,0,175493,,,,0,3c6f52637b9cf8a7c124fa686735d4e3f3b32c5be3d3883220dbc8087edfa213
thread:issue:175520,issue #175520: Question about torch._dynamo.export and nn.GRU support,needs triage,65,needs triage,1,0,175520,,,,0,123958404787199224bbfd27b25d480a2457400badcb4904a334639f9d98afb6
thread:issue:175708,issue #175708: O(N * B) loop in _C._jit_pass_lint (potentially O(N^2)),needs triage,65,needs triage,1,0,175708,,,,0,7d5a06d1ad92aafec7ddc0205fd618482c215e3af852a9068715ea924fc890ff
thread:issue:175724,issue #175724: [Distributed][NCCL] Request for Backport: SIGSEGV in HeartbeatMonitor due to thread-unsafe getenv/setenv race (#167523),needs triage,65,needs triage,1,0,175724,,,,0,d6c3c5a822213c09a2f054f17ace172d38ae2920ef4189f78342989946fb4a77
thread:issue:175790,issue #175790: Only first CUDA profile of the process is correct?,needs triage,65,needs triage,1,0,175790,,,,0,2af8eedd90874e0516992f02a59dbaf2f8ffca4dd105475e42a27f500bb7986a
thread:issue:176148,"issue #176148: export runtime error: Tensor specified that the operator does not mutate the argument, but this seems to be empirically wrong.",needs triage,65,needs triage,1,0,176148,,,,0,064d6906e32cd349c8674bb9590267420d075ef09109b5fecc7e58aea3aa8354
thread:issue:176276,issue #176276: Path Traversal in torch.distributed.FileStore allows writing files outside intended directory,needs triage,65,needs triage,1,0,176276,,,,0,e547fb09628264fdd3da3beae3bbdf917d2ea628924d437c9e60f11164d9d5a4
thread:issue:176288,issue #176288: memory profiler data dump streaming-friendly format,needs triage,65,needs triage,1,0,176288,,,,0,bded0d1ad1840c97eb19272fa5118fd7fbfd75a99de856c82bcbfe8718df84a6
thread:issue:176439,issue #176439: [wip][RFC] Symmetric Memory pool explicit lending semantics,needs triage,65,needs triage,1,0,176439,,,,0,8b0535eb4fb6965c3d1611a35122bbef0bc0dd4d64c6d7f1ace3082adedfcbf1
thread:issue:177059,issue #177059: Way to have DTensor ops go through torch_function rather than go directly through torch_dispatch?,needs triage,65,needs triage,1,0,177059,,,,0,224ade8dac3d137254b4e67784dd28103e8a14532f53b11082fcc0a81b96ff5d
thread:issue:177239,issue #177239: [Question] JIT Path Deprecation — Maintenance Timeline and Removal Plans,needs triage,65,needs triage,1,0,177239,,,,0,97145ceeab6763547aa7fd766d3b6328120074606e33def8c4174c186d7396a8
thread:issue:177247,issue #177247: AArch64 Unit Test Failure - test_jit_autocast,needs triage,65,needs triage,1,0,177247,,,,0,502508550ad492066ecc55ad347f01d687e7fec95430a419b87bd1db955502db
thread:issue:177249,issue #177249: AArch64 Unit Test Failure - test_ops_jit.py test_variant_consistency_jit_linalg_lstsq_cpu_complex64,needs triage,65,needs triage,1,0,177249,,,,0,9d4fc4e30c7f13ae5013fc8224249ff0eac3f2559aafea0b171b5c5a29551006
thread:issue:177309,issue #177309: [ROCm][NCCL watchdog] Cross-thread stream-capture mode restrictions in hipEventQuery/hipEventSynchronize cause false watchdog failures,needs triage,65,needs triage,1,0,177309,,,,0,11980ea93aaf55ba5e5bb714f9a53e59d229e251ac58775b55b43e64e7a849c7
thread:issue:177371,issue #177371: LocalTensorMode: all_to_all_single crashes with uneven split sizes,needs triage,65,needs triage,1,0,177371,,,,0,7d0d950b163e34720d9c51593abd56c48d28b72258d5c433dcda94f77c6235ad
thread:issue:177476,issue #177476: Crash in torch.ao.nn.quantized.functional.interpolate when input tensor has a 0-sized dimension,needs triage,65,needs triage,1,0,177476,,,,0,78842b1b8f573054d7f290196193ec30a33c5396381cac41035d23ef5cada872
thread:issue:177508,issue #177508: [inductor][cpu] torch_multimodal_clip AMP multiple thread performance regression in 2026-03-16 nightly release,needs triage,65,needs triage,1,0,177508,,,,0,a3e1fa14c31ea7ec6854ac113c86d882c943c831e054acdf926a923518447972
thread:issue:177825,issue #177825: Internal Assert Failure in `torch.ao.nn.quantized.GroupNorm` with incorrectly sized bias tensor,needs triage,65,needs triage,1,0,177825,,,,0,18bec21df29f09967581b85e6347416ec2136c6f43bf920e41f785e83cb527e9
thread:issue:177959,"issue #177959: dist.new_group(..., use_local_synchronization=True) hangs with Gloo for overlapping groups, while NCCL succeeds on the same participant-consistent creation order",needs triage,65,needs triage,1,0,177959,,,,0,2265ced452850b097da49ad12ab6447b9048899432343f6ed3655e2924fe78d7
thread:issue:178044,issue #178044: Compilation fails with Apple Clang 17 due to ambiguous py::make_tuple return types,needs triage,65,needs triage,1,0,178044,,,,0,7973279730f62806ff233166e3919f255aeaa84c90b283e551aa284a93575961
thread:issue:178087,"issue #178087: [Profiler] Expose moduleHierarchy, activityType, and other KinetoEvent methods to Python",needs triage,65,needs triage,1,0,178087,,,,0,db329a87f8e88973cfa1fdca1881fe8823e5a8ab5eed416d0db1d88cf95da57b
thread:issue:178138,issue #178138: Conflict between PyTorch CUDACachingAllocator and NCCL symmetric memory (ncclCommWindowRegister) causing incorrect all-reduce results,needs triage,65,needs triage,1,0,178138,,,,0,22c4ab88eeb6ada17c0608824cd87cd28374b57868aaa36cf6a434509adc1e8a
thread:issue:178191,issue #178191: [distributed] ProcessGroupMPI: cudaAwareMpiCheck() does not support Cray MPICH GPU-aware MPI,needs triage,65,needs triage,1,0,178191,,,,0,8e57840ca0d2cc2554563201cd74a84a536bdb93ba70e7ed1f52e83a9495a103
thread:issue:178756,issue #178756: Case mismatch in `_new_process_group_helper` prevents default backend type from being set for custom out-of-tree backends,needs triage,65,needs triage,1,0,178756,,,,0,3a92d8f5490c8a8477b0c342670b5f571f3c4bc8eb5ba252d98f6c351e63c111
thread:issue:179109,issue #179109: Request to support fp8 x4 packed dtypes in PT,needs triage,65,needs triage,1,0,179109,,,,0,8b439f6ac74b6a78194d825c847f402bb230a430658fc11bc271f81b3f609520
thread:issue:179111,issue #179111: The smallest values of E4M3 for two-level quantization is incorrect,needs triage,65,needs triage,1,0,179111,,,,0,a1cf5bf79201260a79fd07ef5a25cbda721785d765dc7b3cbf3725d347181757
thread:issue:179176,issue #179176: Support unbacked in export Dim API.,needs triage,65,needs triage,1,0,179176,,,,0,44f2e3741e08156e597dd237647f0d75af10de2d59c54e654d6d83d199437ceb
thread:issue:179238,issue #179238: Shutdown race in Gloo backend,needs triage,65,needs triage,1,0,179238,,,,0,01f7b47d18e0b2a76ae2a5344a831b37dd243b0d7d5ab3f9ad66914b783f7bc9
thread:issue:179770,issue #179770: [symm_mem] small-M FP8 fused_scaled_matmul_reduce_scatter is slower than unfused baseline on B300/SM100,needs triage,65,needs triage,1,0,179770,,,,0,de745aa65232f468111a8f4428f4b1855e429ba8b3803f6b341c962aecc8b47d
thread:issue:179848,issue #179848: Race condition updating pgStatus_ in ProcessGroupGloo::runLoop,needs triage,65,needs triage,1,0,179848,,,,0,4b6b27fa7479e401ecb1618aa5a1cf99b11885cccedcee1a33c7a429875bcacb
thread:issue:180307,issue #180307: PackageImporter.load_pickle() race can silently deserialize the wrong tensor across concurrent loads,needs triage,65,needs triage,1,0,180307,,,,0,0a5828ac1de757e1dfd7f9800e0768c22c5856ce5364b64353f209cf36fadcac
thread:issue:180550,issue #180550: [Discussion] Parallelizing serial for-loops with free-threaded Python — and the `torch.distributed` challenge,needs triage,65,needs triage,1,0,180550,,,,0,a6977d8f09f019cbb5d2eaf40bbcfdf9f0dbe89f86ab8a68a1ef8884aced7a4e
thread:issue:180995,issue #180995: ConvTranspose + BatchNorm fusion broken for `groups != 1`,needs triage,65,needs triage,1,0,180995,,,,0,6f4cf6edfae55f5d2587c17a2bf34cdb50b05d3abf1bd311cb29219b6adb01c0
thread:issue:181086,issue #181086: [Symmetric Memory] symm buffer init and rendezvous cannot be assigned to a specific CUDA stream,needs triage,65,needs triage,1,0,181086,,,,0,7eb47fdcc237a5ec38315ac1c1c303d4842c1b4394aa4da6a5499fb357cb6741
thread:issue:181181,issue #181181: FP8 Per-Head Quantized Scaled Dot Product Attention,needs triage,65,needs triage,1,0,181181,,,,0,092c50c056435984e0326be77ef2df86296cb38894dc105600d1811554ed05e1
thread:issue:181231,issue #181231: `jit.script` sequence unpacking breaks when using square brackets,needs triage,65,needs triage,1,0,181231,,,,0,9201e76ca5b83555381b8f77b9a8a74cf65d1ff34261bd6700b69caa32e70404
thread:issue:181429,"issue #181429: Make torchrun options ( --nnodes, --nproc-per-node, etc.) same a mpirun/mpiexec",needs triage,65,needs triage,1,0,181429,,,,0,9cb6f9ef4984569a23da40333de449fbe7f3847be95fc7e8025db8f447dd20fa
thread:issue:181473,issue #181473: `torch.jit.ScriptModule.register_full_backward_hook` raises RuntimeError due to `_is_full_backward_hook` attribute type cast failure,needs triage,65,needs triage,1,0,181473,,,,0,ceb2a58bbfa68e8dc6d78dffc2abad48d4e1aab5fd7813719f5a750021416202
thread:issue:181480,issue #181480: DISABLED LiteInterpreterUpgraderTest.DivTensorV2,needs triage,65,needs triage,1,0,181480,,,,0,cd67d28fb491ab868a447b2bab44b826b3d4ebd4fedded2f87d91c864ed36e5a
thread:issue:181621,issue #181621: torch.compile (Inductor) generates invalid C++ kernel: tmp8 used outside its declared scope when torch.gather index is a self-referential expression,needs triage,65,needs triage,1,0,181621,,,,0,eef48ee86fa1e2328add1298a33ff339783818b5357ad59baee97653ed68e4a0
thread:issue:181638,issue #181638: torch.compile(backend='inductor') hangs indefinitely on LayerNorm → flatten → cumprod,needs triage,65,needs triage,1,0,181638,,,,0,381f586cfe3f8b15e1ee36647e62d473597bc19c7e5daf21538f3d6dd634b9d1
thread:issue:181865,issue #181865: CUDA trace callbacks leak across tests in a shared process,needs triage,65,needs triage,1,0,181865,,,,0,7571316aeace2efbeec8c111fd1a5187d0717a435f5d98a7c0a39b352d0e85ac
thread:issue:182373,"issue #182373: [ROCm] torch.profiler drops GPU activity in spawn child if parent calls torch.cuda.init() (regression in 2.12, 2.10/2.11 pass)",needs triage,65,needs triage,1,0,182373,,,,0,0ce1fe020593c7fbab738d458cf34dc6921159935644329423bb794f7c63e469
thread:issue:182653,issue #182653: CPU FPE crash in `torch.nn.quantized.functional.conv2d` (qnnpack),needs triage,65,needs triage,1,0,182653,,,,0,9ef14947feffbb5fb79aa1abe06a5168ef9bc601955339a13a90abb996ddf258
thread:issue:182988,issue #182988: ROCm MI300 (gfx942): native segfault in mixed-precision BatchNorm2d backward,needs triage,65,needs triage,1,0,182988,,,,0,23336439a683309d6e9a60db4d86034e25b69d272509da207d73f12ce9835213
thread:issue:183702,issue #183702: `grouped fully_shard(list) fails when a grouped module is reused later in the same iteration`,needs triage,65,needs triage,1,0,183702,,,,0,a7c2be1fb04e8ad7baaa90f94ea5799c814632102fdc059bb4c3b8ee120372c3
thread:issue:183842,issue #183842: docs.pytorch.org/memory_viz has Unhandled version 5 error,needs triage,65,needs triage,1,0,183842,,,,0,7225077231958b0e2847ec45616b88a3c1cca5b4988c86ee3d0fa188fbb8ef2a
thread:issue:183888,issue #183888: torch.compile Inductor fails on unfold followed by in-place abs_,needs triage,65,needs triage,1,0,183888,,,,0,32991d005eebf233931fcb6cf857e694e32247727b88bb9e346a052b2e3512cf
thread:issue:184010,issue #184010: [Feature Request] Support PrivateUse1 device in ProcessGroupGloo,needs triage,65,needs triage,1,0,184010,,,,0,221d53926dbdac70ba987a66188a666708381c70dd437f1cc820d1da9ee8e035
thread:issue:184078,issue #184078: add_params metric_dict values not shown in tensorboard hparams tab,needs triage,65,needs triage,1,0,184078,,,,0,0a3d96da0dafdb7b35e9108ad1caa49d80897f42f6a7f73a94924546dc481f68
thread:issue:184147,issue #184147: [AMD] MoRI support for MoE EP Combine/Dispatch torch.distributed.TokenSwitch Interface & AMD Pollara Support,needs triage,65,needs triage,1,0,184147,,,,0,b1516f29e35f244e19f7663e3cba2926faaad7ed4d376ef3f8eec861afd74778
thread:issue:184647,issue #184647: Replacing a submodule's `__call__` with `fx.Interpreter` on its exported `gm` causes the parent's output to contain `FakeTensor`,needs triage,65,needs triage,1,0,184647,,,,0,dcfea1fe49c23e037d324913f098eb09cc6dc05ff39835ebd691e26ba3472924
thread:issue:184982,"issue #184982: deprecation of derived quantized dtypes (quint8, qint8, qint32)",needs triage,65,needs triage,1,0,184982,,,,0,12ed372c3d55f952480651f620686c5ddbd0c59746cb0b80cc370327ad45ab0b
thread:issue:185267,issue #185267: DISABLED test_correctness_ne_cpu_float32 (__main__.TestLazyOpInfoCPU),needs triage,65,needs triage,1,0,185267,,,,0,d8b7bd8c9e01cafdc45eade6de438959e986eb1ed1613837ccdadee92c779a33
thread:issue:185268,issue #185268: DISABLED test_correctness_with_reusing_ir_ne_cpu_float32 (__main__.TestLazyOpInfoCPU),needs triage,65,needs triage,1,0,185268,,,,0,6fad3ffc2e926f4693bfd18f4c8c082d3afe7447bdd009de7ab159bc1fc71e51
thread:issue:185440,issue #185440: [JIT] Out-of-bounds read in MemoryReadAdapter::read() via crafted TorchScript model,needs triage,65,needs triage,1,0,185440,,,,0,71079eb31d8eeadad2364f824a432c002a9598712c9206c109842831c2e559a0
thread:issue:185442,issue #185442: [JIT] Use-after-free in Source::_M_dispose() during TorchScript error-path cleanup,needs triage,65,needs triage,1,0,185442,,,,0,0d8c9fa20df1c2a802d8f5e87dcb0eb5a04dd722f7329d499ec97828afa629d3
thread:issue:185443,issue #185443: [JIT] Uncontrolled memory allocation in Pickle Unpickler opcode handlers (setInput / readBytes),needs triage,65,needs triage,1,0,185443,,,,0,84232579e37e08da808e41b91e3140cc6b3bc86ef05b0cbbd4c7e416f51cbbb1
thread:issue:186026,issue #186026: Annotate Cuda Graph kernel nodes from profiler with the originating fully qualified layer names,needs triage,65,needs triage,1,0,186026,,,,0,6704c2a8b278db40ac278c174bd77b7525bba4fab149966e1309204eab2f6579
thread:issue:186344,issue #186344: DISABLED test_cupti_monitor_collection_raw_dump_smoke (__main__.TestProfilerCUDA),needs triage,65,needs triage,1,0,186344,,,,0,633cbd4d522fef6faf9fb62983a19fdd39a245f5cc63eb9591b913720ad61761
thread:issue:186363,issue #186363: test_quantization.py int8 tests fail on AArch64,needs triage,65,needs triage,1,0,186363,,,,0,9967e5b6e071813f651bf48364645f594f2f8fa75ace40fb3a8af8dd3c9e1c59
thread:issue:186400,issue #186400: DISABLED test_cupti_monitor_collection_repeated_lifecycle (__main__.TestProfilerCUDA),needs triage,65,needs triage,1,0,186400,,,,0,48cd5857f1d8cdf90ea2fbefd6af5642907fe5efbe25f0391082dc1325b551e5
thread:issue:186587,issue #186587: DISABLED test_cupti_monitor_matches_stock_op_and_kernel_names (__main__.TestProfilerCUDA),needs triage,65,needs triage,1,0,186587,,,,0,a8c1357731c1c747b267459fc933a95c243f08d8c8edb396d68752cc2e802d0b
thread:issue:186591,issue #186591: Importing Pytorch results in undefined symbol ncclCommResume,needs triage,65,needs triage,1,0,186591,,,,0,7f6dd4b301129741885d72834f003834b6a86c81337bb1434717e24bc5e8f9ba
thread:issue:186637,issue #186637: DISABLED test_cupti_monitor_multithread_runtime_thread_assignment (__main__.TestProfilerCUDA),needs triage,65,needs triage,1,0,186637,,,,0,126a236b3659d5a77b9be5278cad683787bef49dd582fd9de618a62deece3b84
thread:issue:186670,issue #186670: _load_mobile_module_from_bytes SIGSEGVs on malformed flatbuffer model (unchecked Function.class_type index),needs triage,65,needs triage,1,0,186670,,,,0,3d3415da96e23b1eefc920858b4c8865be525a5eb3a9aca7ae8059c36043e1c4
thread:issue:186773,issue #186773: DISABLED test_cupti_monitor_record_shapes (__main__.TestProfilerCUDA),needs triage,65,needs triage,1,0,186773,,,,0,d36dd17ad1016897e8b45e191596a4dfeb78af0d61ef8bcc2e4a9e49630c2276
thread:issue:186823,issue #186823: DISABLED test_cupti_monitor_trace_has_expected_events (__main__.TestProfilerCUDA),needs triage,65,needs triage,1,0,186823,,,,0,02ac7140c2b511055518a7fc1a7a29afe49de145477c7c85f5eaafe7ec9e43b7
thread:issue:187432,issue #187432: TypeError: can only join an iterable within `default_planner.py`,needs triage,65,needs triage,1,0,187432,,,,0,45ea498d013d4b2f805a3e66f90955f92b249989eaed570bc179eb9b0a16d076
thread:issue:187659,issue #187659: Segmentation fault in `TestQuantizedOps.test_empty_batch` on ARM,needs triage,65,needs triage,1,0,187659,,,,0,35d26c62dcfe504dac530d9917631b4f39f22a8350caf78927f983a1b3a7007a
thread:issue:188227,issue #188227: [inductor] generated kernel calls `empty_strided_cpu` with a negative stride → `Expected result >= 0`,needs triage,65,needs triage,1,0,188227,,,,0,8159e7a45725608366d8fb4fa04e238d5134dafca1111422489c6d78d583816b
thread:issue:188266,issue #188266: SIGABRT inside `gloo::transport::tcp::Address::fromPeerName(int)`,needs triage,65,needs triage,1,0,188266,,,,0,fc0e86430987fcee5089a2fa2fe380600c4d3244d335a108539e18cb64063957
thread:issue:188304,issue #188304: Update NVSHMEM to Resolve RoCE Failures,needs triage,65,needs triage,1,0,188304,,,,0,32a6defe166bcd26ba9e88f322fb644d6ca8a1df8326bbf89d98051f69e6780b
thread:issue:188481,issue #188481: export_chrome_trace produces invalid timestamps,needs triage,65,needs triage,1,0,188481,,,,0,1e4fe3aaa6b6f95efc4021c9117b81962531b720c44f50db7c1e6e2762c64557
thread:issue:188490,issue #188490: test_profiler_stack_trace_augmentation fails on Windows due to stale profiler golden,needs triage,65,needs triage,1,0,188490,,,,0,daaea2131b124f1a07f58785be1b62707dfae1b9029789aa762ba32ff6be81ac
thread:issue:188553,issue #188553: test_cholesky_solve_batched_many_batches_cuda_float64 fails on ROCm gfx942 (Tensor-likes are not close),needs triage,65,needs triage,1,0,188553,,,,0,3741fe71902b06e76b9ecd2c9c78f02170d88783cf7266ca27a10b2fa3ab7bd8
thread:issue:188567,"issue #188567: torch.distributed._symmetric_memory: NCCL-backend rendezvous() fails (""Failed to window register / invalid argument"") when it is the first NCCL op on the process group",needs triage,65,needs triage,1,0,188567,,,,0,2528ed1b59e442e85cf0a248e5059f925779042f4ae1a21392835dd92ef27ee9
thread:issue:188833,issue #188833: [pipelining] Interleaved1F1B warmup head-of-line bubble from front-loaded RECV,needs triage,65,needs triage,1,0,188833,,,,0,486aea4f811780d86628d5cac2309f0726952e5478c55346d9d03d7d872d61f3
thread:issue:188836,issue #188836: DISABLED test_dropout_parity_dynamic_shapes (__main__.TestDropoutAlignRandomEager),needs triage,65,needs triage,1,0,188836,,,,0,7f28439b3f32c648f93e04b912c91c58b0de9a1862333116eacdd06d33e4e480
thread:issue:188896,issue #188896: DISABLED test_launch_elastic_worker_raise_exception (__main__.ElasticLaunchTest),needs triage,65,needs triage,1,0,188896,,,,0,03f7dbe233264096bf742628204e1808545ec2fdd34a1977ab07af514516017d
thread:issue:189228,issue #189228: [symm_mem] one/two-shot all-reduce deadlocks when issued from multiple streams: ops lack channel plumbing,needs triage,65,needs triage,1,0,189228,,,,0,94b637911213a5449d24181ceedcda042ca0eeee4e9f8cf1ac5c95edc39e2237
thread:issue:189245,issue #189245: [cpp_extension] Stale .lock file from FileBaton causes permanent deadlock when compilation process is killed,needs triage,65,needs triage,1,0,189245,,,,0,499c36816fb8c843bba4267660eb291197dfee1ecdcd5707291a4aae1aa65b2a
thread:issue:189280,issue #189280: [CPU][quant] operator_benchmark: qcat ~3-4x slower (likely) and fake_quantize ~700x slower (possible) since ~Nov 2025,needs triage,65,needs triage,1,0,189280,,,,0,03059aeae8b635626abdec18c8e1f35e4aace8a01edb833e0e1e227e05103f4e
thread:issue:189293,issue #189293: AttributeError: Can't get local object 'WeakValueDictionary.__init__.<locals>.remove' during AOTAutograd graph caching on Python 3.13,needs triage,65,needs triage,1,0,189293,,,,0,ea776042bd30727f893c11d7eabddd75c35a4bbf20799176895a25b8f6c13aad
thread:issue:189298,issue #189298: torch.jit.trace fails with MaxUnpool in 2.13,needs triage,65,needs triage,1,0,189298,,,,0,4d5aa544daaa04cf735e114b953b7b3a6bb24d5bef3d002d06a798edaa33f509
thread:issue:189308,issue #189308: `torch.distributed.checkpoint`: `.metadata` loaded with raw `pickle.load` on the public `load()` path — same class you already warn about on `torch.load`,needs triage,65,needs triage,1,0,189308,,,,0,45908bd6ae4190b5186dbe41853ade934048fb6566a2364f29fda23f8fa728f4
thread:issue:189309,issue #189309: [Stable ABI] Add a has_side_effects op tag. DCE silently deletes side-effect-only op calls,needs triage,65,needs triage,1,0,189309,,,,0,a2611b8dedaf4f22b1ecd1334b5545c12f411e169eb6319321f6396dd1331f77
thread:issue:189322,issue #189322: DISABLED test_aoti_debug_printer_codegen_cuda (__main__.AOTInductorTestDualWrapper),needs triage,65,needs triage,1,0,189322,,,,0,dde84a363a01fbd8af2d3f6d7af0ddd014dbf8ed0b53aed0ec395d920577b776
thread:issue:189323,issue #189323: DISABLED test_fake_export_contiguous_cpu_float32 (__main__.TestExportOpInfoCPU),needs triage,65,needs triage,1,0,189323,,,,0,4a4171d84a08906edc309cdc2d0a076901fe06811b7fe2ec4a486440a1c281df
thread:issue:33692,issue #33692: JIT tracing check fails with boolean tensor modifications,needs triage,65,needs triage,1,0,33692,,,,0,3a47c682ebb35047b8d014e357543f105d9dbe198301b64465678bce05aeec87
thread:issue:45745,issue #45745: Native CUDA-Aware MPI support for MVAPICH2 and other MPI libraries,needs triage,65,needs triage,1,0,45745,,,,0,da31f2ccad6b4949d89c684b8c26f8d3cdd034ee2b7b7446be7bc134e7d55919
thread:issue:52741,issue #52741: torch.jit.script is swallowing a rfloordiv warning,needs triage,65,needs triage,1,0,52741,,,,0,96a0b0fdc33eb3e85227eb42f443edb443ced1fe95f5fe2ae4c72df1561e39a2
thread:issue:55577,issue #55577: Source location range issue for dictionary,needs triage,65,needs triage,1,0,55577,,,,0,72e0d875664652bce7b8b010bd7005342a5dbedd566645a34b33208ebb05b98d
thread:issue:61210,issue #61210: Jitted function cannot be pickled,needs triage,65,needs triage,1,0,61210,,,,0,5451e8d5390ec547929c3b2997aebc7a1f9c238e8fb803aa57253cd5831c98b8
thread:issue:71049,issue #71049: Distributed broadcast fails with simple GPU tensor on Windows + GLOO,needs triage,65,needs triage,1,0,71049,,,,0,5ff2614910d2c494510c82dd86a6860fd0788cd21b80e026af8330c3ca67c48c
thread:issue:72482,issue #72482: Improve torch::deploy documentation,needs triage,65,needs triage,1,0,72482,,,,0,827a0ea1521aacecd75be8d09ae7b87b1cb4b2f9446ea13934e414206d8934c7
thread:issue:73542,issue #73542: Investigate why tensor shapes are not populated when printing WorkNCCL for broadcast,needs triage,65,needs triage,1,0,73542,,,,0,36ad9598d3e041ae932d3b30d37ce5621aa278cf75ca3f954839218a978c4c91
thread:issue:74590,issue #74590: [CTA] Let's Stamp Out Flaky Tests!,needs triage,65,needs triage,1,0,74590,,,,0,6fbd67aa9d0ed8d2230ebfc942e9c3caaece4b2614d46a9b743627b762ef62aa
thread:issue:74809,issue #74809: Incorrect results for `torch.distributed.gather` for tensor created from permuted NumPy array,needs triage,65,needs triage,1,0,74809,,,,0,cce876ba920d404c3e3d80e50d5ae176844f3dd718c14e273b0c452e355f3a79
thread:issue:75171,issue #75171: torch.jit.load fails when path contains non-ascii characters,needs triage,65,needs triage,1,0,75171,,,,0,5b5a4c51e3dcbff4610a4f913c9eac047ee907f942ec01ccce69fdfdf58ed510
thread:issue:76039,issue #76039: [RFC] NPU device for PyTorch,needs triage,65,needs triage,1,0,76039,,,,0,4513914f2a0c3681044ba4edc879fc56f147cbf0d2a1dcca725581350b1116b5
thread:issue:82306,issue #82306: when distribute training load pretrain model error,needs triage,65,needs triage,1,0,82306,,,,0,bbe4db06e67659f90e5fb137bf59271673c148169bd0afcfd56742dc137b4d35
thread:issue:82308,issue #82308: torch.einsum gets wrong results randomly when training with multi-gpu,needs triage,65,needs triage,1,0,82308,,,,0,dec35dbe146c02740ee77242cb4d39b1e1e9fb598ba00a89fd417e815f07c2f8
thread:issue:83824,issue #83824: RuntimeError: Interrupted system call when doing distributed training,needs triage,65,needs triage,1,0,83824,,,,0,00f6c19822332036f084dc785f80fb8c38bb2f0b95ce62d4bb957ff94b12dd73
thread:issue:85234,issue #85234: c10d all_gather aborts with Signal 8 (SIGFPE) when tensor.numel() == 0,needs triage,65,needs triage,1,0,85234,,,,0,cc11f55d8419c1c96bb04f4a53bd07a4ffd8bce1cf6a13c76a7fed381cf77c71
thread:issue:85671,issue #85671: nn.Embedding weights are not synced across processes with DistributedDataParallel when other parameters are present,needs triage,65,needs triage,1,0,85671,,,,0,126752a526e701c75da2a489a05bbee097757926251ff9b57c7928ae36da97d8
thread:issue:86849,issue #86849: `torch.distributed.all_reduce` allocates excess GPU memory when using NCCL backend,needs triage,65,needs triage,1,0,86849,,,,0,d3ab15f52404a2e9c15c8df29791a31352db4550995991bec5c87dfa0cd3e82b
thread:issue:86962,issue #86962: PyTorch RPC crashed when using IB,needs triage,65,needs triage,1,0,86962,,,,0,d632dc4ad7e994c86e4e27b1e49578c36b94fef1527122f9ab8ed23d2ae7532e
thread:issue:87995,issue #87995: multi-node distributed training rank0 hang at dataloader after a few epochs,needs triage,65,needs triage,1,0,87995,,,,0,eaa052a07ddbbc48c82829ee209c881587ae463a81c917059065e3eb39f19d2c
thread:issue:91173,issue #91173: not able to import pipelines as torch.distributed is missing,needs triage,65,needs triage,1,0,91173,,,,0,9898fd75ce8b6e0e72565694a17b7b3dbf002d7858eb0f1e92cda5c421c33df8
thread:issue:91535,"issue #91535: When dist.broadcast float32 to int64, it will silently generate wrong results",needs triage,65,needs triage,1,0,91535,,,,0,8e3874d20b3d702a7cf6ab5eb3ae2e7c39deccb0c5dcbad9a91421b8c932411c
thread:pr:155672,pr #155672: Fixed NLLLoss 1D input crash with torch.compile,PR blocked,65,PR blocked,1,0,,155672,,,0,7dd3cc3c46b4e08a73b8f543490318f713e28290b25ae2af8ade02a26d7c7c0e
thread:pr:160883,pr #160883: Fallback to eager AC when backend=eager for TagActivationCheckpoint,PR blocked,65,PR blocked,1,0,,160883,,,0,3532345b2f839a57b3287b21c9469cdd8c9f65c65342d36e326c9bf0eb01015f
thread:pr:173459,pr #173459: Add dtype mismatch validation for mm/bmm in meta registrations (#171190),PR blocked,65,PR blocked,1,0,,173459,,,0,f258142bb17bff6864a2326a4574d788d9568756963348c907e02f5dbecb07c6
thread:pr:177511,pr #177511: [export] preserve original traceback when augmenting export error messages,PR blocked,65,PR blocked,1,0,,177511,,,0,572701969705770d70fe55d07a21ee2ce007878846283489b0f189743cc9e363
thread:pr:180118,pr #180118: [dynamo] Skip pad_packed_sequence during tracing,PR blocked,65,PR blocked,1,0,,180118,,,0,6f89ee562df67a6efb255649e5c291049f5888b362accb09e1578dab21952586
thread:pr:180501,pr #180501: Fix clip_grad_norm_ failure with mixed DeviceMesh DTensor norms,PR blocked,65,PR blocked,1,0,,180501,,,0,76adade36bb79e517f9d3cd8d66153ad2eb03ffbfb1a27691756e8855d88bd86
thread:pr:180602,pr #180602: Enable ruff rule PLW1510: require explicit check= in subprocess calls,PR blocked,65,PR blocked,1,0,,180602,,,0,0c5da94f114815f56531137b5a9f8b5645d1255f29e9451245aec51b6d3b5957
thread:pr:181060,pr #181060: [fft] Validate _fft_c2r last_dim_size to fix heap-buffer overflow,PR blocked,65,PR blocked,1,0,,181060,,,0,78a52451e97551afa63895d02e1808d7a9b1b34eaede3bfa1caff48ab24996e6
thread:pr:182055,pr #182055: [Inductor] Fix flaky test_allow_reuse_disable_if_exceed_peak by checking reuse marker,PR blocked,65,PR blocked,1,0,,182055,,,0,54ec1ce409aa9be6c0d8ac6164872836732ffa567e73e1a832aecfa0e8c6fd27
thread:pr:182406,pr #182406: Fix misleading error when caching_allocator_alloc size overflows ssize_t,PR blocked,65,PR blocked,1,0,,182406,,,0,64e0367d3937b617fb3e20dc9941a569b46f997c1c894c8d6ffb8118b0ae23c3
thread:pr:183087,pr #183087: [profiler] Synchronize all active devices on profiler exit,PR blocked,65,PR blocked,1,0,,183087,,,0,843a75d0e0d94bbd20f036781be690ad8f03e0644b735d1fa5622640dd8777ed
thread:pr:183922,pr #183922: Unskip test_vector_norm on AArch64,PR blocked,65,PR blocked,1,0,,183922,,,0,0bcc32f6890c8d52d259533c7a4e2df33a0dab09fca1da573b2af29fce98770c
thread:pr:184151,pr #184151: [CUDA] Fix float32 std/var M2 overflow by using double accumulation,PR blocked,65,PR blocked,1,0,,184151,,,0,5b9193c33e3732e7941aed63505a8857fb9b2dddf74d9bf81886a029a0b2051e
thread:pr:184587,pr #184587: Fix torch.compile addmm failure with beta=0 nonbroadcastable bias,PR blocked,65,PR blocked,1,0,,184587,,,0,a30c98e558cba913a0cbf3a474a7e9bd938c7031ed1204fe039b0e0c49ecad90
thread:issue:100249:issue:105325:issue:12672:issue:150989,"issue #100249: [discussion] ""TensorList"" as first-class python-world abstraction (to mirror existing dispatcher's `TensorList` and `torch::autograd::tensor_list`) and as key for dispatch for merging `torch._foreach_*` into regular `torch.*` functions",stale/low urgency,64,stale/low urgency,39,0,"1512, 4959, 12672, 22924, 24915, 24985, 25162, 25522, 25691, 25743, 26957, 27522, 27617, 28743, 31758, 32851, 33181, 33296, 35642, 35759, 36650, 38228, 38419, 39570, 41292, 42405, 43012, 43502, 46176, 65156, 70533, 83775, 85157, 91471, 91508, 96085, 100249, 105325, 150989",,,,40,f53f55df91641860c9e28cb328ff1ec96b5b30ddbeb3f5696cc0b47fc28a6a1c
thread:issue:102078:issue:123421:issue:126963:issue:13993,issue #102078: AddressSanitizer: heap-buffer-overflow in test_comprehensive_nn_functional_embedding_bag_cpu_bfloat16,needs reproduction,64,needs reproduction,15,0,"88, 13993, 14436, 19668, 23240, 48491, 66247, 70245, 89817, 95944, 99561, 102078, 123421, 126963, 147141",,,,14,17ade92bca2591277113f7f019636f0f63bb73af2b1e10cd27260ad3be689039
thread:issue:26165:issue:64412:issue:64535,issue #26165: Memory leak in multithreading environment when loading checkpoint,high-priority/blocker,64,high-priority/blocker,3,0,"26165, 64412, 64535",,,,3,478904cb6eb38328fecdc87578c0c39dc3219cf7d681d69a1e8d7d578156563b
thread:issue:109819:issue:110605,issue #109819: ValueError: only one element tensors can be converted to Python scalars,high-priority/blocker,64,high-priority/blocker,2,0,"109819, 110605",,,,1,1a1ccba379d54caa0e28cd874cdac63bd91e8200712013210949096164995e4e
thread:issue:71149:issue:99042,issue #71149: Memory leak in distributions.multivariate_normal.MultivariateNormal,high-priority/blocker,64,high-priority/blocker,2,0,"71149, 99042",,,,1,2084cb141f757c5e1a7da9967c5da03567372697bd9c4662e0dcc08ec491ae3d
thread:issue:105872,issue #105872: PyTorch 2.0.x `CUDA error: operation not supported` when `Tensor.to` a different device,high-priority/blocker,64,high-priority/blocker,1,0,105872,,,,0,bfc5f1d9f017317357a16bf00832fdef847e8dd3a277efcf7aeb2488bb25ff67
thread:issue:111901,issue #111901: Multiprocess. DataLoader worker is killed by signal: Segmentation fault.,high-priority/blocker,64,high-priority/blocker,1,0,111901,,,,0,45783223684a6c644acc866a9b248f8a8291f8913381d6d52675d87b3866d950
thread:issue:117494,issue #117494: Kernel crashed when importing torch,high-priority/blocker,64,high-priority/blocker,1,0,117494,,,,0,f319f171f34b5e97605863dff4372676244f8b7dd1b1dee919c766c963fc91a1
thread:issue:118088,issue #118088: Segmentation fault (SIGSEGV) in c10::detail::getNonDeterministicRandom(bool),high-priority/blocker,64,high-priority/blocker,1,0,118088,,,,0,80e165f5207c6405717894073849db46d94a2f94c7ca010ee112b70937c5a91f
thread:issue:121235,issue #121235: Issue with sentence_transformers pointing to torch tensors manipulations,high-priority/blocker,64,high-priority/blocker,1,0,121235,,,,0,037d94323fe01fb0d4de43eb332638efbc4282b7217221059e6a2274f1db4922
thread:issue:122308,issue #122308: Segmentation fault about torch::nn::Conv2d in `libtorch2.0.1+cuda11.8`,high-priority/blocker,64,high-priority/blocker,1,0,122308,,,,0,6f10a85b0900ae9ae8e0f6d55ef16dc2a78b81e0fc77a4118bc86a38b4e4ce99
thread:issue:122478,issue #122478: 6470 segmentation fault (core dumped),high-priority/blocker,64,high-priority/blocker,1,0,122478,,,,0,09168a3241e64c929928335c3fcd50eecb29f4c3c99662015e5bafa5f722cf06
thread:issue:122935,issue #122935: Pytorch 2.1 / 2.2 uses 4x more memory during data loading than Pytorch 2.4 or 1.13,high-priority/blocker,64,high-priority/blocker,1,0,122935,,,,0,136fbbba5659637cb9392d98f5f5b502dedb89cb52f3ca1f2bb2526818ed6bb0
thread:issue:124656,issue #124656: Importing polars before torch causes a segfault,high-priority/blocker,64,high-priority/blocker,1,0,124656,,,,0,6e2fcf31111940340f05d1f15e63a9163203598113ce6f1232b177ced677726b
thread:issue:125664,issue #125664: Aborted (core dumped) in torch.fft.irfftn/hfftn/ihfftn with malloc(): corrupted top size,high-priority/blocker,64,high-priority/blocker,1,0,125664,,,,0,bf2ce4082e59001f1df399032c6ed1badccf9bf352f2df1de941775dc1b2cc92
thread:issue:127275,"issue #127275: Python sometimes crashes inexplicably, with ""/var/log/kernel. log"" displaying index errors",high-priority/blocker,64,high-priority/blocker,1,0,127275,,,,0,f155d7d63f56fb82ad23f51447137c90076892d73e3890034517e463e218260e
thread:issue:129304,issue #129304: `python3 setup.py bdist_wheel` tries to write to /usr/local/... during build,high-priority/blocker,64,high-priority/blocker,1,0,129304,,,,0,1c619db131b2714295383d502aa958999e0b4a788a9861a2e2bb056bfca829a7
thread:issue:130708,"issue #130708: Converting a numpy array of size larger than 32,768 to a tensor causes a segmentation fault",high-priority/blocker,64,high-priority/blocker,1,0,130708,,,,0,462bd109bc1042a7b7a38b3389ea651466758015932907a7457bd8701b57aa9a
thread:issue:134198,issue #134198: [Intel CPU] Same CPU operations on Torch 2.3+ many times slower those on Torch 2.1,high-priority/blocker,64,high-priority/blocker,1,0,134198,,,,0,3eb02649197f460c7e82cd48c7fded7aa7b341ef379a6fcd3451a78890026c3f
thread:issue:140230,issue #140230: Inconsistent Model Loading with `weights_only=True` for Custom Models,high-priority/blocker,64,high-priority/blocker,1,0,140230,,,,0,f925045fb29e55f1fb056f93bf5e6738b4647efe20a7d7eac3b4764c90270dff
thread:issue:143780,issue #143780: Looking for valid compiling option for extension based on torch-2.1.0+cpu.cxx11.abi,high-priority/blocker,64,high-priority/blocker,1,0,143780,,,,0,b1ee15d3ee2015399ec7cde85ccc852406ce8dae749b477cb35ba01969ccd5d2
thread:issue:149941,issue #149941: SEGV in static destructors when Vulkan is enabled,high-priority/blocker,64,high-priority/blocker,1,0,149941,,,,0,c8d7cf72bf2bb09543c65c64584e3e4d2c92951722e7a214857eb46c73e66650
thread:issue:152102,issue #152102: Segmentation fault with # FIXME: copy.deepcopy() is not defined on nn.module,high-priority/blocker,64,high-priority/blocker,1,0,152102,,,,0,4feca6983a7102ef5c9955552efafb83c215795593e491386a2e2f70ff771790
thread:issue:153937,issue #153937: Dynamic compilation fails with torch 2.7,high-priority/blocker,64,high-priority/blocker,1,0,153937,,,,0,f8d1f450ef0576a36a91a14a10b56a0c88addb70ff7cf9721c4e32e84c60f96e
thread:issue:154014,issue #154014: Floating point exception when autocast is enabled,high-priority/blocker,64,high-priority/blocker,1,0,154014,,,,0,d6c4b1bc4129865932844ce063e78f8476bc8bf653b162ae0dc0bfaa8f03a776
thread:issue:155693,issue #155693: torch typing: register_buffer -> device,high-priority/blocker,64,high-priority/blocker,1,0,155693,,,,0,9303e2708d99966b235616a1256fed8d98cdf7a0a16e7ed6c5298cf620069226
thread:issue:156613,issue #156613: [CUDA 12.9 | sm_120] Build fails on RowwiseScaledMM.cu with PTXAS error (PyTorch 2.8.0a0),high-priority/blocker,64,high-priority/blocker,1,0,156613,,,,0,85a29a7db43730dfec4dd440d23363827a7d23c336ac4f30fd9bbcaabeab1486
thread:issue:156723,issue #156723: Segmentation fault on Apple Macbook Pro M4 24GB RAM and 1TB flash dasd during forward() with AutoModelForSequenceClassification using CPU,high-priority/blocker,64,high-priority/blocker,1,0,156723,,,,0,e637e6facc41f1fd73e3706124b1b2f04f5ff345a9ff5a753326bc508c9b0914
thread:issue:157704,issue #157704: Seg faults on macos / OSX,high-priority/blocker,64,high-priority/blocker,1,0,157704,,,,0,77aad3f5c76cad00c3ace4e9a50aea32ba1f038e5f47fcfe2de57ae0d97bbba3
thread:issue:159454,issue #159454: libtorch_cpu.so link libm.so issue:GLIBC_2.29 not found,high-priority/blocker,64,high-priority/blocker,1,0,159454,,,,0,bd73c15267569690a6f8b4f442a05d7c8b27abc56cea99949b07c66a6964c472
thread:issue:160876,issue #160876: F.silu(torch.tensor(float('-inf'))) returns nan instead of 0 in PyTorch 2.7.0+cu128,high-priority/blocker,64,high-priority/blocker,1,0,160876,,,,0,92dcfe4254ee52bc42d144104eb641503788e000a205ee97a4af978d5f588669
thread:issue:161865,issue #161865: SEGFAULT in libomp.dylib after torch.matmul on M4 Max with MPS backend,high-priority/blocker,64,high-priority/blocker,1,0,161865,,,,0,4bd7b98936cbbeab2d6e9a43a8cce8961e77d37f2d4605f318bb48b96e922c5f
thread:issue:162343,issue #162343: [BUG] `Aborted (core dumped)` on RTX 5090 with Nightly Build when running TTS inference,high-priority/blocker,64,high-priority/blocker,1,0,162343,,,,0,ac71be74e752bdc65838b5907d7fef96a40e1a645c7b3ba23e44508a48f9d3ee
thread:issue:162797,issue #162797: Segmentation fault in mkl_blas_avx512_dgemv_t_intrinsics,high-priority/blocker,64,high-priority/blocker,1,0,162797,,,,0,c73a7d5ec221e34d9eb7dc7a643870b64227fa114056bac360a0a443124c300e
thread:issue:22259,issue #22259: using multi thread lead to gpu stuck with GPU-util 100%,high-priority/blocker,64,high-priority/blocker,1,0,22259,,,,0,4c47620f95249e455a732c602555cbea5d12c3864ab6d2ce4aae06f5399fabae
thread:issue:26120,issue #26120: Process fails with assertion error in magma-cuda100,high-priority/blocker,64,high-priority/blocker,1,0,26120,,,,0,15ca02226563c9a822496e8a0c97bf72edcc9a4d08d4c9d1330aa64ec847845e
thread:issue:29498,"issue #29498: ""malloc(): memory corruption (fast)"", action=3) at malloc.c",high-priority/blocker,64,high-priority/blocker,1,0,29498,,,,0,1aa5dad473b0a09e9f2b561c64883e3888fe1efb754b7fe5c9d4660180ecc80a
thread:issue:31362,issue #31362: sklearn and pytorch incompatibility issue,high-priority/blocker,64,high-priority/blocker,1,0,31362,,,,0,cfdb5f263c2ead144b2813d13851fbcb4f6e489b437389af2b4983a9236db5e4
thread:issue:35792,issue #35792: Python 3 build PyTorch got core dump,high-priority/blocker,64,high-priority/blocker,1,0,35792,,,,0,ba2582b30fcd4de7a9b7283dd90ae0701dc1c055df00a495d20ff17c9a8bef55
thread:issue:36871,issue #36871: Stochasticity for DistributedDataParallel on CPU but not on GPU,high-priority/blocker,64,high-priority/blocker,1,0,36871,,,,0,162eddf121096592a8c109bebf096cf640bf2d79e7d69cf4c789d0e33d449e2c
thread:issue:40425,issue #40425: Segment Fault after use cusolverDnDestroy() with torch1.5,high-priority/blocker,64,high-priority/blocker,1,0,40425,,,,0,7ab8c495d094a34409a27e9b8837dbd445a7b5dcc43999b1f8d864785d0fae2a
thread:issue:45198,issue #45198: DataLoader with cv2 and some numpy/cv2 import order causes workers to not work,high-priority/blocker,64,high-priority/blocker,1,0,45198,,,,0,73f0337e9dfa23b211dd924351858af02c2a4c0ea05a43e68ac8fecb013264dd
thread:issue:48555,issue #48555: segment fault for Image model training by four GPUs,high-priority/blocker,64,high-priority/blocker,1,0,48555,,,,0,a8f1310a9dea09dbbb7d00317e0c83589dfc733470873c77efd8e18f8eec2aa1
thread:issue:53895,"issue #53895: terminate called after throwing an instance of 'c10::Error' what(): isTuple() INTERNAL ASSERT FAILED at ""/home/wenda/libtorch/include/ATen/core/ivalue_inl.h"":927, please report a bug to PyTorch. Expected Tuple but got GenericList",high-priority/blocker,64,high-priority/blocker,1,0,53895,,,,0,eefd0f661fc5fe31091d231d4845ad54a84843405af0c7cc33587ac808a6771a
thread:issue:54578,issue #54578: cublasSgemmStridedBatched failure when calling grad of grad,high-priority/blocker,64,high-priority/blocker,1,0,54578,,,,0,79f3eb9fd7434c2c07638178cb963b35ddbf70733c127959fbd1ccfc6cbca157
thread:issue:60835,"issue #60835: Pytorch V1.1 runs OK, but Pytorch v1.5 to 1.9 run wrong",high-priority/blocker,64,high-priority/blocker,1,0,60835,,,,0,b30d5f31c7fd6fcb8079bbb53d05209620421ca0ba6eb2c851bedf070a9dd264
thread:issue:62308,issue #62308: Error while running FINETUNING TORCHVISION MODELS,high-priority/blocker,64,high-priority/blocker,1,0,62308,,,,0,2f4c750191913e9c61dcdbeff85b2d3f7c5131a46432c1204815e7f3584db2fb
thread:issue:62331,issue #62331: broken pipe error,high-priority/blocker,64,high-priority/blocker,1,0,62331,,,,0,d1949a9e44c956419e8b1e9bc3291034cf8f66fc481988c5cddf8101f070d830
thread:issue:66033,issue #66033: Crashed with `terminating with uncaught exception of type std::__1::system_error: condition_variable wait failed: Invalid argument`,high-priority/blocker,64,high-priority/blocker,1,0,66033,,,,0,06583bc19d709893a87395ff4591e9e41d993f299dbf5bee84ed2178b77ecad4
thread:issue:68920,issue #68920: Breakes with -OO flag in script,high-priority/blocker,64,high-priority/blocker,1,0,68920,,,,0,9de5014efd4e4e76d1a6c34dae84ed14f76420fe7c14482afa5b0bf2f93d6796
thread:issue:76543,issue #76543: Segfault in ~PyFunctionPreHook,high-priority/blocker,64,high-priority/blocker,1,0,76543,,,,0,be78af696ca8b75999c853f58318b0651df15fda1bd6913d3ce9608788830286
thread:issue:81532,issue #81532: forward program terminated from __cxa_pure_virtual,high-priority/blocker,64,high-priority/blocker,1,0,81532,,,,0,08c82c7951e7367ac570741305e6db635fcab7b56bb0417d6bfd2b2bb6b82d8e
thread:issue:87209,"issue #87209: Reproducible ""CUDA error: an illegal memory access was encountered""",high-priority/blocker,64,high-priority/blocker,1,0,87209,,,,0,9843328c0b6b3988289a875334ad6f62f8adfc16c49e8e7297e54fba62d505c1
thread:issue:87446,issue #87446: ipykernel crash importing torch after scipy in .ipynb file,high-priority/blocker,64,high-priority/blocker,1,0,87446,,,,0,e205655d1faa21ecd9ff294cf11547b1988e8c1a5763e5d42c53f696798482f4
thread:issue:89714,issue #89714: Segfault on torch.nn.functional.one_hot with large tensor on Python 3.9,high-priority/blocker,64,high-priority/blocker,1,0,89714,,,,0,deea50ad41706dd271128b60011f7dd791306640a26ba338baab7d659f5ee42f
thread:issue:91533,issue #91533: Cannot cast float64 to float32,high-priority/blocker,64,high-priority/blocker,1,0,91533,,,,0,edaa13dfe93f89c3580d00007e6090fd6c5be965af6fce20ae3903038d92405f
thread:issue:91633,issue #91633: Segmentation fault when running torch.nn.functional.fractional_max_pool3d on torch 1.13.1,high-priority/blocker,64,high-priority/blocker,1,0,91633,,,,0,f40c592433ff3accf14ab05c414b5a4dbd55ebb16bdcc16ea0766c4617aa9da5
thread:issue:96779,issue #96779: Segmentation fault (core dumped) during Torch finetuning (at random step),high-priority/blocker,64,high-priority/blocker,1,0,96779,,,,0,4347e9278e9b3918c233b008dc64900eba7fcec5485ff6e0c37411c4bcaa71bc
thread:issue:99200,issue #99200: Unused `import torch` followed by `cuml.NearestNeighbors` leads to nondeterministic segfault (during Python process exit?),high-priority/blocker,64,high-priority/blocker,1,0,99200,,,,0,94ffd1969179335adb0da1740305d7eef0201f502faf2a2cc2010e795a5fa926
thread:issue:187455:pr:182109:pr:182731:pr:182736,pr #182109: [MPS] Replace MPSGraph LayerNorm backward with native Metal kernels,waiting on contributor,63,"Wait for contributor update on PR #187553, #187787, #187976; keep related issue/PR context attached.",19,0,187455,"182109, 182731, 182736, 182749, 182791, 183688, 187456, 187457, 187458, 187553, 187556, 187557, 187559, 187787, 187976, 188156, 188333, 189146",,,23,bba9abaa5159935cfde19082f163fcd56bd01de06bb16d33f31b6e0c46f25597
thread:pr:188112:pr:188468:pr:188469:pr:188470,pr #188112: flex_gemm: feed local reductions into epilogues,draft/noise,63,draft/noise,10,0,,"188112, 188468, 188469, 188470, 188739, 189188, 189190, 189314, 189315, 189316",,,120,6a59a5e85cda6f75196026ae54ce7a27cfb419ae45576bee7533ec82cebb69f2
thread:issue:174541:issue:187590:issue:187591:pr:170092,pr #170092: [dynamo] Support constant ops on unrealized LazyConstantVariables,waiting on contributor,63,Wait for contributor update on PR #184858; keep related issue/PR context attached.,6,0,"174541, 187590, 187591","170092, 170644, 184858",,,11,7354ac59942b35ec1107673f28facda6635db8ce21c1684f9ad2d0b78edad9ba
thread:issue:188987:pr:176688:pr:176689:pr:178849,pr #176688: Enable eager tests for inductor-unittest - enable test_dtypes and align XPU op dyptes with OpInfo,waiting on contributor,63,Wait for contributor update on PR #178849; keep related issue/PR context attached.,6,0,188987,"176688, 176689, 178849, 179549, 179550",,,17,5021d4c0d88e341bc44e6a7d62dca5d1290258ddcd71ddb7faf412b88ab4f478
thread:issue:76324:issue:78065:issue:80152:issue:80157,pr #183328: [special] Add arbitrary-order modified Bessel functions,PR blocked,63,Review active PRs and linked issues as one maintainer work thread.,5,0,"76324, 78065, 80152, 80157",183328,,,13,50363c9448d8e6eccd97e344707c9d015cb9b86aeef7721cf3d695f07ca23c6f
thread:issue:116396:pr:176830:pr:181357:pr:188966,pr #176830: feat: add polyfills for setitem/delitem op,waiting on contributor,63,Wait for contributor update on PR #188966; keep related issue/PR context attached.,4,0,116396,"176830, 181357, 188966",,,8,0ad65025708ad936d5e9f04f649bdfc210d7f57d32e94b4f418f07fea59b7f6f
thread:issue:177259:pr:184355:pr:184356:pr:184678,pr #184355: [3/N][Test] Migrate TestBwdGradients device-agnostic skips to `@skipOps`,PR blocked,63,Review active PRs and linked issues as one maintainer work thread.,4,0,177259,"184355, 184356, 184678",,,21,3eb318e8a2b3bdf8f7f549594c8391034d156a47c5fd4518f2c3bb306679cbd7
thread:pr:186918:pr:189240:pr:189253:pr:189289,pr #186918: [Test] Add test hardware classification for test classes,PR blocked,63,PR blocked,4,0,,"186918, 189240, 189253, 189289",,,3,88ed8eb6b2200c2979362113fb4429d4f1606f98f0cfdfae715a62dfdc918b36
thread:pr:188621:pr:189083:pr:189109:pr:189185,"pr #188621: [profiler][cupti] Generate CUPTI field-id catalogs from the ABI as Field(id, ctype)",waiting on contributor,63,Wait for contributor update on PR #189109; keep related issue/PR context attached.,4,0,,"188621, 189083, 189109, 189185",,,20,f6f628cdc2264a2acc6b7d21810e369b4f03c89cffb84e7b0e0bb08082a92c13
thread:issue:158917:issue:172318:pr:165837,pr #165837: [OpenReg][Random] Enrich OpenReg random number generator module implementation add focused documentation,PR blocked,63,Review active PRs and linked issues as one maintainer work thread.,3,0,"158917, 172318",165837,,,2,75f8a3c71f25e22604ae835b2a85dd2399f136ce852927bc629db52999ac4277
thread:issue:106571:pr:189319,pr #189319: Enable flake8-bugbear B007 (unused loop control variable),PR blocked,63,Review active PRs and linked issues as one maintainer work thread.,2,0,106571,189319,,,1,b41c3f07f0561da2ed1c6f27264fbce2f0e5eaae187652a459cef421f510f593
thread:issue:181801:pr:182376,pr #182376: [CUDA][CPU] Unify CPU/GPU relu/max behavior,PR blocked,63,Review active PRs and linked issues as one maintainer work thread.,2,0,181801,182376,,,2,abd5ab4fe71e4d29b66fa1151d7870ba87d4f1c6e29e42bc4d5141c7baad8ad4
thread:issue:187183:pr:180208,pr #180208: Update slow tests,PR blocked,63,Review active PRs and linked issues as one maintainer work thread.,2,0,187183,180208,,,1,0288239488ea7a96d1b79beb7ba0f621cf66cf4e2436191b31d2e737cb65a5b4
thread:issue:188230:pr:188862,pr #188862: [inductor] Fix bool minimum/maximum crash in index propagation (#188862) (#188862),PR blocked,63,Review active PRs and linked issues as one maintainer work thread.,2,0,188230,188862,,,1,4de2cfb8e06a33db345e30578a70fbe40505a9c95b6946aadb39c095040e395b
thread:issue:189270:pr:187328,pr #187328: Add fusion provenance to profiler timelines,PR blocked,63,Review active PRs and linked issues as one maintainer work thread.,2,0,189270,187328,,,1,a4096179440d8dde6e0f7e5d13cb0576398e22f3ef05119b0a58329a11d7e5d0
thread:pr:185917:pr:185918,pr #185917: [Test] Refactor test_autocast.py: make behavior/cache tests device-agnostic,ready for maintainer decision,63,ready for maintainer decision,2,0,,"185917, 185918",,,1,1a25524441e14d7fdcf37aa3384bb99154f851264943d134a0ba5309948d4dbf
thread:pr:186055:pr:186056,pr #186055: Add CUDA graph support for torch.while_loop,PR blocked,63,PR blocked,2,0,,"186055, 186056",,,4,9a521e83c14fd986c78a5e89220b40bfd4920e99a2834c2c81674eb3ea28bce5
thread:pr:186282:pr:188766,pr #186282: Fix DTensor view-unflatten when one output dim is sharded by 2+ mesh axes,draft/noise,63,draft/noise,2,0,,"186282, 188766",,,1,422c25cddd1b867dc353da106adac3b116c7b7cdd4826cd0867e27ec1329fd3b
thread:pr:187465:pr:189088,pr #187465: [Symm_mem] Support expandable segements inside PyTorch Symmetric memory for NCCL backend,PR blocked,63,PR blocked,2,0,,"187465, 189088",,,6,b467b38384c5ac70e56942f329073125722cb703c34dc8bc1dad5002b1341101
thread:pr:188105:pr:188181,pr #188105: Add support for custom attributes on Exceptions,waiting on contributor,63,Wait for contributor update on PR #188105; keep related issue/PR context attached.,2,0,,"188105, 188181",,,4,3aa8813d60b04c022456d21fcebd90a0e5e99f9b0ea7bc9272b469c0734ab70e
thread:pr:188242:pr:188597,pr #188242: [ROCm] Support USE_ASAN=1 build,draft/noise,63,draft/noise,2,0,,"188242, 188597",,,2,2f3dca082b928ec3c6c67d7b241d5b9ff4a398bf782d0944e30aef88a03f9e49
thread:pr:188867:pr:189131,pr #188867: Add USE_KLEIDIAI_SME option to skip SME/SME2 kernels on CPUs without SME,ready for maintainer decision,63,ready for maintainer decision,2,0,,"188867, 189131",,,3,71fb5cb3428882c42b3629ce29497ecb3b0ec22cec886312f217622e8358242c
thread:pr:189095:pr:189096,pr #189095: [inductor] Preserve stream annotations across pattern replacements,waiting on contributor,63,Wait for contributor update on PR #189096; keep related issue/PR context attached.,2,0,,"189095, 189096",,,3,2b8399715c23f295424bd665d6347b9cb2d9f91ad44cb7dc6cf0cb6b88f04fa0
thread:pr:106149,pr #106149: Automated submodule update: kineto,PR blocked,63,PR blocked,1,0,,106149,,,0,e6f76b6710e41aeba651f409b377a295e169b89fbedc26274523ace0628307e0
thread:pr:113258,pr #113258: [pytree] support PyStructSequence types for Python pytree,PR blocked,63,PR blocked,1,0,,113258,,,0,7ed2c85076c4e6ed9364ac6057e723b548132755449644aef7f250212013aa9d
thread:pr:115316,pr #115316: Automated submodule update: FBGEMM,PR blocked,63,PR blocked,1,0,,115316,,,0,8e472abee8a18cf1b57be77cbffab0904b96a27aeb333bffb04392e971bfe613
thread:pr:152806,pr #152806: [invoke_subgraph] Force the output stride to be same as eager,PR blocked,63,PR blocked,1,0,,152806,,,0,0f2ab1ade0d237a69490e3aa66fc111766f9fbe8810d5c86e35bc2382095efda
thread:pr:165766,pr #165766: Enable kineto XPU scope profiler extending existing XPU profiler plugin,PR blocked,63,PR blocked,1,0,,165766,,,0,626d93ca81d87ea10c911a44b628add3065d98c5102c7b177d0730cf18752210
thread:pr:170396,pr #170396: [typing] Fix return type annotations for various functions,PR blocked,63,PR blocked,1,0,,170396,,,0,638f293997aab2f48611ae5bd111a71ede87cd052bbc5bc58fa1a7c46145b0a2
thread:pr:170567,pr #170567: [dynamo] Support dict mutation without a guard,PR blocked,63,PR blocked,1,0,,170567,,,0,ac8bcd1c4e66243cad7c44af0deba6369a29633a9e46cfd2c453d7b10a996369
thread:pr:170571,pr #170571: [CUDA][cuBLASLt] addmm - Lt path improvements,PR blocked,63,PR blocked,1,0,,170571,,,0,7644e1e0081502d36c3e73c805b3c4794f31efe298527954c204fddf1ee45a1e
thread:pr:170609,pr #170609: Add auto-generated cuDNN depthwise conv heuristic,PR blocked,63,PR blocked,1,0,,170609,,,0,d1c4209d3668b5b6cabcdceec09b4965439e676c60dc85c139df126716d78ac9
thread:pr:170679,pr #170679: Fix FileBaton leaving dangling lock files that cause compilation hang…,PR blocked,63,PR blocked,1,0,,170679,,,0,4158f6f6acd3bed9b3e80f19b4c111e0c48ee167c7d6d9a9e1d7960ab28c7f3c
thread:pr:171023,pr #171023: [DTensor] Support Shard to _StridedShard through all-to-all,PR blocked,63,PR blocked,1,0,,171023,,,0,50e00de7a1e5afde1f518cd7fc7391cf45503d04c391417243b7b9b1afe66256
thread:pr:171178,pr #171178: [DTensor] Refactor padding/unpadding with CollectivePaddingContext,PR blocked,63,PR blocked,1,0,,171178,,,0,bae93108e4ca9f37b9cc887ac9d8bb3b57e3150b7c83f7fe55eea96e577155ce
thread:pr:171219,pr #171219: [DTensor] Support Partial to _StridedShard,PR blocked,63,PR blocked,1,0,,171219,,,0,155b179975f730daf058ce0b7c001656f73735960129fb9443b3e10e8d136fa6
thread:pr:171338,pr #171338: [DTensor] Fully support uneven _StridedShard redistribution with device order,PR blocked,63,PR blocked,1,0,,171338,,,0,d91d65772e813bd1af1b3c84324cdf6464b7670bd3d45eacc533cdd12db3bf2c
thread:pr:171445,pr #171445: Intorduce torch.accelerator.is_graph_available to check if Graph is available,PR blocked,63,PR blocked,1,0,,171445,,,0,27555c7c9f0d645af566c24262c57db4a66c670620720c99ee16dbcca5671ab9
thread:pr:173390,pr #173390: Add interactive glossary with hover tooltips,PR blocked,63,PR blocked,1,0,,173390,,,0,082abfe546c730de62d54492ae68be25e877b26c61d606906890c020f4c6e0fe
thread:pr:173914,pr #173914: [CI] [CUDA] Add CUDA ARM CI image,PR blocked,63,PR blocked,1,0,,173914,,,0,cbd45808f3d6be0375b230350a802bb831ed58e5e39f210d0f57146ac0c926cd
thread:pr:174323,pr #174323: [effectful ops] get tokens working with AC,PR blocked,63,PR blocked,1,0,,174323,,,0,65eb2efc8c4233dc217613dc8a0d147d6fafcaf7b0132e4a1287c5a2c59dccdd
thread:pr:174395,pr #174395: Add optional out argument to F.scaled_mm,PR blocked,63,PR blocked,1,0,,174395,,,0,950f75b94a72544828698fb40d234d87b1be817297ee3c7e43fc1e149e2f1fc7
thread:pr:174409,pr #174409: [inductor] Add logging to autotune output showing cache hit,PR blocked,63,PR blocked,1,0,,174409,,,0,e3421dffb55b4004101244c957eb718daff5b2a1fec48d6f0b976b99cb411090
thread:pr:174699,pr #174699: Add CuTeDSL-based scaled grouped MM kernel,PR blocked,63,PR blocked,1,0,,174699,,,0,4a58878004c01128e6205589c199c505ae29add747885dfa24a6b491f1a28756
thread:pr:174771,pr #174771: Fix `sample_kwargs` in `polygamma` op info,PR blocked,63,PR blocked,1,0,,174771,,,0,8fefb228f98e9698566ef438110329ba9c65091c62cde74e9822da7889cc3659
thread:pr:174940,pr #174940: [BE] Delete old Metal ops,PR blocked,63,PR blocked,1,0,,174940,,,0,c10794d659f021a27553664c6ca4c1a2e4f8fc98afcd1724592dc81c9a5cdc0a
thread:pr:175011,pr #175011: [S615805][RecSysCheckpointing][PyTorch] Add strides to TensorProperties (#175011),PR blocked,63,PR blocked,1,0,,175011,,,0,3a3dc4b451cd296fab9dc9248c399effe60d9f58c273978fcbd584f4f05d592a
thread:pr:175242,pr #175242: Improve FX partitioner complexity,PR blocked,63,PR blocked,1,0,,175242,,,0,52d40955088a2788fb995077727c3703d1e8cc1ac013521c373069609c1c13c5
thread:pr:175450,pr #175450: [inductor] CUDAGraph P2P pool handling for symm_mem,PR blocked,63,PR blocked,1,0,,175450,,,0,4b5bc35628f2db5c400c396de92ca5d57662b3a579f3a55d3191f32af69f0462
thread:pr:175468,pr #175468: [DO NOT MERGE] Test AMD Capacity.,PR blocked,63,PR blocked,1,0,,175468,,,0,29ef0920da2734692bc5376528880f4bd17b2cfe32659a32ef463dd6cbe9a277
thread:pr:175472,pr #175472: Add FA4 fp8 implementation to SDPA,PR blocked,63,PR blocked,1,0,,175472,,,0,e2e743e9e1e2fd5b01c9ded15cbdf78813660ce7e7a83277d8c19a2c1e122ea9
thread:pr:175584,pr #175584: fix aten.remainder in inductor,PR blocked,63,PR blocked,1,0,,175584,,,0,6b84926c91074c709f6d7c7918a8f58ff7a317ed4ace7fec0e3530fa5da3faa2
thread:pr:175746,pr #175746: RISC-V Vector Extension (RVV) support for ATen,PR blocked,63,PR blocked,1,0,,175746,,,0,a4447c35553f694572453f2e45e93863238f00ef3cdbfdd99660e5623a17ab49
thread:pr:175758,pr #175758: Create a DeviceModule protocol,PR blocked,63,PR blocked,1,0,,175758,,,0,ef7a588511f8ee687b86a987a59973c5d6b14fed7600390e7cb4d423acd26952
thread:pr:175809,pr #175809: [cuDNN] Migrate to new frontend API for batchnorm,PR blocked,63,PR blocked,1,0,,175809,,,0,b5f00887e202a9d31fb367ac7e744c29e850fcd4f5f3c1f58cd19f1a4f660a4d
thread:pr:175900,pr #175900: [Inductor] Fix Enum serialization in user-defined Triton kernel codegen (#175900),PR blocked,63,PR blocked,1,0,,175900,,,0,4349ffaa7174101c44c5a8348426e5cd2c213fcffa3d64f70bc985bfcc4ba4d2
thread:pr:175914,pr #175914: [Export] Fix unflatten crash when mutation intermediate has list args,PR blocked,63,PR blocked,1,0,,175914,,,0,d4a67791121d4e0626638f17a110b5fe20fb71b4aa0a57f88594e7806b93b2c3
thread:pr:176239,pr #176239: [dtensor][partial] just_knob error message for fixing .item() on partials,PR blocked,63,PR blocked,1,0,,176239,,,0,63427538012f005c5fa46fd09c1625efcfa6f6bdc2174146b86f70403d2b3a96
thread:pr:176293,pr #176293: [CI] aarch64 manywheel and cuda image build fixes,PR blocked,63,PR blocked,1,0,,176293,,,0,fec453de8220afe648cf2a028d4dab1dfdcb918913fa2bd0ac4dccbe0308a6f5
thread:pr:176574,pr #176574: Increase dynamo coverage of the `operator` module by adding polyfills,PR blocked,63,PR blocked,1,0,,176574,,,0,9e273c3c8d88a53c6b0b7c7bba33c537016a2d2e4e6aeaa6fd91036ce4702556
thread:pr:176600,pr #176600: Fix 1C1W tensor's shape and stride,PR blocked,63,PR blocked,1,0,,176600,,,0,bcfba23b8bcd84f09c02b7916fb567692d8d8b56641ff30585b2a440634c2476
thread:pr:176657,pr #176657: [CUDAGraph] Detect and error on freeing parent-capture tensors inside CUDA graph conditional nodes,PR blocked,63,PR blocked,1,0,,176657,,,0,034337622db15fb9a84af5056b0e25c43bcd3c5fb2bef0c9e2b74acec9ae3287
thread:pr:176754,pr #176754: [CUDA Graph] Thread-safe RNG state for concurrent graph captures,PR blocked,63,PR blocked,1,0,,176754,,,0,d74007db9dadbe347c0df2da2d97a8c544e75da81e8055ac02ae05d7d8bffd8c
thread:pr:176876,pr #176876: [inductor] Add device-type benchmarking dispatch registry,PR blocked,63,PR blocked,1,0,,176876,,,0,451e21e4a234f3a606083dd11424812c008e54dfaa8dbd586b54b7ebd39f4588
thread:pr:176894,pr #176894: [cuDNN][Convolution] Expose cuDNN `allow_reduced_precision_reduction` numerical note to fix `float16` tests,PR blocked,63,PR blocked,1,0,,176894,,,0,4d94adcefc9c98ee825eaf121c302d86a902a62eb406182490e6f4311ea91d46
thread:pr:176920,pr #176920: Add distributed runtime autotuning for Triton kernels,PR blocked,63,PR blocked,1,0,,176920,,,0,3f5f83dfcbaf229ae61d67f010cbeb2164b47fa9b230699a9213e65323382ac7
thread:pr:176989,pr #176989: Add supports for itertools.batched and itertools.combinations_with_re…,PR blocked,63,PR blocked,1,0,,176989,,,0,760509bf8eb33a1b84e3ea083fe6884e956085a77ff060fa181119a03fd975ba
thread:pr:177087,pr #177087: [torch][inductor] Get shape information for extern/Aten kernels with enable_kernel_profile (#177087),PR blocked,63,PR blocked,1,0,,177087,,,0,c248a0b9057b7c0ddd496c515281cb781a37eecde1bb136a005d44c6eecadedf
thread:pr:177263,pr #177263: Eliminate the unnecessary logic,PR blocked,63,PR blocked,1,0,,177263,,,0,bc8612381f7ea4ccd5bc26288fe2c1d69c197ce08a8182b35fb7cf89e2654625
thread:pr:177283,pr #177283: RFC : WIP warmup compile with fake tensor with dynamic dims then run with real,PR blocked,63,PR blocked,1,0,,177283,,,0,03c16372122ae2d6c8d3a2dbbdbd7cb5a058f420dd7e6c535b31e08f8973cca1
thread:pr:177568,pr #177568: [MTIA] Add dilation support for max_pool2d_with_indices_backward lowering (#177568),PR blocked,63,PR blocked,1,0,,177568,,,0,85883617e71cfe27b6a504dc4429384728ac5e84b2c29a694af9c93e37b9c823
thread:pr:177618,pr #177618: replicate-only ops,PR blocked,63,PR blocked,1,0,,177618,,,0,f56c269a9e3e1a3d88086cda4e068fae7a11df732c287838608fde7265b0fa2b
thread:pr:177619,pr #177619: more ops,PR blocked,63,PR blocked,1,0,,177619,,,0,4d563e0619b617826baaca094b18831db44782cf2fbce9913bf59e8cdf5ac3f7
thread:pr:177757,pr #177757: [MPS] support for SparseCSR tensor and its unary operators,PR blocked,63,PR blocked,1,0,,177757,,,0,62561effa9cd7e1ee56e15c770dfe68a5a9eb68e751012806f9d996b6e62d4c4
thread:pr:177854,pr #177854: [Inductor] Run first cudagraph warmup in eager pool (#177854),PR blocked,63,PR blocked,1,0,,177854,,,0,b47fdafedd553de8e0d88dd6eb4785f27612237c069340fad5557d727aacc832
thread:pr:177924,pr #177924: [triton hash update] update the pinned triton hash,PR blocked,63,PR blocked,1,0,,177924,,,0,6bd31fd229ace142e8f0dad6e0fcc1c8c35970603aa39900a18cde91cd4e5124
thread:pr:178004,pr #178004: [ROCm] enable compiler-driven loop unrolling,PR blocked,63,PR blocked,1,0,,178004,,,0,fdcbf8565b00c2dd50bc4b3fd98ee556fd9b136a55df16a90399ac07e45dd45e
thread:pr:178014,pr #178014: add missing to() operator which is called in benchmark (#178014),PR blocked,63,PR blocked,1,0,,178014,,,0,b72ba26caca74393acdb0e58e1bf8dec6385bc2b4620270854bdce59715346e6
thread:pr:178083,pr #178083: Update inductor expected accuracy files,PR blocked,63,PR blocked,1,0,,178083,,,0,d174c7bcc03250280585a44d3cd6c097cb768112227acffb5ffecff1ca071708
thread:pr:178150,pr #178150: [12/12] Enforce C++20 minimum in header guards (#178150),PR blocked,63,PR blocked,1,0,,178150,,,0,739c9c7002c8279ef58e3f6581e95de179c2a1f6f22a55ebcf984d47cf94ed7f
thread:pr:178167,pr #178167: triangular_solve_stub: expand leading dim support beyond dense and non-overlapping inputs,PR blocked,63,PR blocked,1,0,,178167,,,0,ea5150ed1883180330a49e56522f8f07e962128a12a1ac3c483b86572e11a7ad
thread:pr:178242,pr #178242: [XPU] Add grouped_mm dispatch for XPU via sycl-tla,PR blocked,63,PR blocked,1,0,,178242,,,0,9bd37b1a86bef4b20d7adce97cdf996d4d15de8fa6295a683ba6bb36f56eb27d
thread:pr:178283,pr #178283: Bump torch_tpu pin,PR blocked,63,PR blocked,1,0,,178283,,,0,8e06d5c0d89b662255340fb4795b8b85dddf64ba58a76e9c5f569858e322a79f
thread:pr:178300,pr #178300: [Inductor] Retry Triton compilation without pointer_range_32 on failure,PR blocked,63,PR blocked,1,0,,178300,,,0,892803b0b95cd11aa42ac8419413656994d6667a431fcfa60d54da7263ffdd73
thread:pr:178314,pr #178314: Separate tagging path for AC when non-strict tracing,PR blocked,63,PR blocked,1,0,,178314,,,0,6215251f5a8a90b68f6507ab7e18720335087e6f2701ea7d9783ba79e016346c
thread:pr:178318,pr #178318: [Inductor] Add xmask unswitch for dynamic pointwise kernels,PR blocked,63,PR blocked,1,0,,178318,,,0,4ea30b5a0312b9e40b9f4c752f4d7c6fe38e2cfb8b1a0d178b376f07ce0d45e8
thread:pr:178336,pr #178336: [Distributed] Make DDP tests and tensor parallel dependencies backend agnostic,PR blocked,63,PR blocked,1,0,,178336,,,0,0127291b572daf86bdf3c1025875db1d42aa18b4f5bd1df4dde6e0d8c3ef2428
thread:pr:178338,pr #178338: [inductor] add option to autotune at fusion time for mm,PR blocked,63,PR blocked,1,0,,178338,,,0,ade0461aff602ee1b3c418e97e7cc2801b972db6f38afb6087131aaede50b34d
thread:pr:178416,pr #178416: [dynamo] Migrate allow_in_graph → nonstrict_trace in torch/ production code,PR blocked,63,PR blocked,1,0,,178416,,,0,3aede0f6ad19fb56c78b838491bbbfad7a70e31a194835fe7c4bb576a1d448ff
thread:pr:178500,pr #178500: [autograd] Support Mapping and OrderedDict inputs for autograd.grad and autograd.backward,PR blocked,63,PR blocked,1,0,,178500,,,0,ff2009f1554fdef0f4f92bed86d8f77b73b705a31836dd03f555b6ce9f01e2f6
thread:pr:178710,pr #178710: Add native torch.cuda.gds.save/load for GPUDirect Storage,PR blocked,63,PR blocked,1,0,,178710,,,0,1eab6428d4251413c95d48ed8f5a034361ba6fe4c8eb7a1b44939761ba7c8961
thread:pr:178751,pr #178751: Delay common_utils import to pytest_sessionstart to prevent faulthandler race condition,PR blocked,63,PR blocked,1,0,,178751,,,0,f138c98c297ba8675610250581a98c2639f71483518c03b209972f01efb23ecb
thread:pr:178800,pr #178800: debug server tutorial,PR blocked,63,PR blocked,1,0,,178800,,,0,703d450d1eab80951d40afc8839644ffb8be0a1d5159e8189bc7ffa95336b629
thread:pr:178804,pr #178804: Merge clang18 Python-version Docker images into one multi-env container,PR blocked,63,PR blocked,1,0,,178804,,,0,0351874a86e2ecddd2a4276c8ad8b4bd5f08dead6939896e921f5a9e56b70c2e
thread:pr:178812,pr #178812: [BE][MPS] Delete stale skips from common_modules.py,PR blocked,63,PR blocked,1,0,,178812,,,0,d20c941514e85ff892cf00adda0380204a1fea5b59709ce203238f2f9405e841
thread:pr:178906,pr #178906: Fix batch_isend_irecv to support ProcessGroup subclasses,PR blocked,63,PR blocked,1,0,,178906,,,0,064b86ef2693a1afd8d63ae5b198a448c4f80cf7890c1bf634684a0220a96077
thread:pr:178922,pr #178922: Add CUDA support for Lanczos interpolation mode,PR blocked,63,PR blocked,1,0,,178922,,,0,81ec68fa4d0719fc6ff3e021ce2b91cb36b92a6b8a7b113901f061a2dd6c9585
thread:pr:179127,pr #179127: Skip FBGEMM on Arm64,PR blocked,63,PR blocked,1,0,,179127,,,0,b1e1eea8d5cbd78c437e2161e714a70e8223c8b2da56d2b3b62601b7f58e19f8
thread:pr:179142,pr #179142: [Inductor] Add runtime assertion support to FX wrapper backend (#179142),PR blocked,63,PR blocked,1,0,,179142,,,0,bb6262bb9cc301092885f875456986111e6978290d862df3a813047dd07d33af
thread:pr:179224,pr #179224: spaceship operator for comparisons in Dict.h,PR blocked,63,PR blocked,1,0,,179224,,,0,0ba8973ae7bacec697b30dc8d5baef3c98d99967951bb577cee3e575b37f6ef8
thread:pr:179257,pr #179257: [c10] Fix off-by-one issue in ApproximateClock.cpp,PR blocked,63,PR blocked,1,0,,179257,,,0,c6bec5badbb0d164f01c225d94e3eb4a0d3725fe4738473edd3cb052b5fb3cc8
thread:pr:179400,pr #179400: Use std::in_range for integer range checking,PR blocked,63,PR blocked,1,0,,179400,,,0,10a73eac2daa611ab2f678dc0dce6c287fadd954c0280bbd1a3518426436e6ef
thread:pr:179436,pr #179436: Run clang-tidy with readability-container-contains,PR blocked,63,PR blocked,1,0,,179436,,,0,233d931121225455b8668e69013661d64313b955eef5d5a3102210fb10d57749
thread:pr:179547,pr #179547: [inductor] Add thread-based compilation support for free-threaded Python (#179547) (#179547),PR blocked,63,PR blocked,1,0,,179547,,,0,8cf0e21958c295c398c382207dad2c8d231df23322bf75b8098ee12a7737022a
thread:pr:179548,pr #179548: [inductor] Add dedicated thread pool for nogil Triton compilation (#179548) (#179548),PR blocked,63,PR blocked,1,0,,179548,,,0,94d633086e27fa43c476aade543a49c17219f274b533127c7415d3ff1985d506
thread:pr:179587,pr #179587: [pytorch] Fix resource loading for PAR compatibility using importlib.resources (#179587),PR blocked,63,PR blocked,1,0,,179587,,,0,7a7b1d47de98831ab3a2ecb570bc7015ad8443068a658b228c7ef42a51427037
thread:pr:179642,pr #179642: Cache cuSPARSELt matmul plan/descriptors in _cslt_sparse_mm (#179642),PR blocked,63,PR blocked,1,0,,179642,,,0,84ab028102c56f2adb341abc91b4cc6f0e536763962d3d7ce66849c426b16112
thread:pr:179651,pr #179651: [torchtitan hash update] update the pinned torchtitan hash,PR blocked,63,PR blocked,1,0,,179651,,,0,33a467ab2cdeed7756bb7439f8260fd7394753c907a961faa0e8af64b84df6a6
thread:pr:179699,pr #179699: [torch][inductor] Add KernelContextGuard profiling to all kernel codegen paths (#179699),PR blocked,63,PR blocked,1,0,,179699,,,0,fe2ecc31cea49216b7ac437284bcc9babfa9047605de7fb7a9fb40d2cc098a61
thread:pr:179702,pr #179702: [torch][inductor] Add dump_python_module config for AOT single-pass mode (#179702),PR blocked,63,PR blocked,1,0,,179702,,,0,4e91e0fb68394aac00a9938360a133894225b9e584ce18d39e203a532bb8b3ea
thread:pr:179820,pr #179820: [ATen][Native][CUDA] Use dynamic loading in CUDNN FrontEnd,PR blocked,63,PR blocked,1,0,,179820,,,0,cc7b3636884b1ab9fd5d5cf3c81f8d9e83c3579f9ba5feb6544f29fba7e48217
thread:pr:180050,pr #180050: [FSDPV2] Make FSDP v2 and composable tests device-agnostic for PrivateUse1 backends,PR blocked,63,PR blocked,1,0,,180050,,,0,b37085a0d7abd1f636677504f73940d0beb8e69ea7210001e2328bf4f658f592
thread:pr:180081,pr #180081: [DTensor] Fix GIL-unsafe thread-local py::object in DTensor dispatch logging,PR blocked,63,PR blocked,1,0,,180081,,,0,b01a91ef25652995059353c8a8b3285b6f00e43b5dc216ed19068dcef5c4ff5f
thread:pr:180083,pr #180083: [FSDP] Make FSDP v1 tests device-agnostic for PrivateUse1 backends,PR blocked,63,PR blocked,1,0,,180083,,,0,5537e2c7958f5bdbbd5e8cb2fea2c3cb374b2a57aec9c9bb7d0390b616137457
thread:pr:180167,pr #180167: Clarify that spectral_norm does not compute the true operator norm for conv layers,PR blocked,63,PR blocked,1,0,,180167,,,0,05646f3a36a59f2fe2189d676940efd251b338529e24551aedd547e7773afe8b
thread:pr:180181,pr #180181: [CI] Split benchmark docker builds into three-phase build,PR blocked,63,PR blocked,1,0,,180181,,,0,0e07481380a75ab54672197146a8dfd4b0c49c289b5f5c71a979125407f3ad7f
thread:pr:180183,pr #180183: autograd: handle kMkldnn and kJagged layouts in record_stream_any_impl,PR blocked,63,PR blocked,1,0,,180183,,,0,b28b672eca322a5abf48ba9daff4f65d302a1c08aba58016da2b6c7f56c184cf
thread:pr:180274,pr #180274: [WIP][Triton] update triton pin,PR blocked,63,PR blocked,1,0,,180274,,,0,7206f6ec539c48ab930cfc9b314d7acd79379b1a682eb4fe7d49ca922ef99f73
thread:pr:180283,pr #180283: [cuBLAS][cuBLASLt] Thread-local cuBLAS(Lt) workspace maps,PR blocked,63,PR blocked,1,0,,180283,,,0,64e2a7aa6e72a86b5e2f3d9faf788b4d89943bdb48fdfcb69a5f32374dca6c5c
thread:pr:180442,pr #180442: fix(mps): skip nograph linear on AMD RDNA2 GPUs when input_features>=256,PR blocked,63,PR blocked,1,0,,180442,,,0,a1e0cfd5392e8c73cfb4d01c3e1d5c10a51b1fcdc7a7132b9e30fbec6b9deb88
thread:pr:180455,"pr #180455: [AARCH64] Enable MKLDNN backend for AArch64 for INT8 Matmul, using torch op _int_mm.dtype",PR blocked,63,PR blocked,1,0,,180455,,,0,8665c0427253cb2159ebb89c92071b1ad61cd6b8d9b15c513b72c380ef096c12
thread:pr:180528,pr #180528: Add dynamo_timed wrappers for AOT autograd compilation stages (#180528),PR blocked,63,PR blocked,1,0,,180528,,,0,e187dbf0e73788ffaa438d85d99b1ca5d2e293dc814e76b39c032ebb76ff9cf9
thread:pr:180529,pr #180529: [inductor] Add TORCHINDUCTOR_PERSISTENT_AUTOTUNE_DIR for persistent autotune cache (#180529),PR blocked,63,PR blocked,1,0,,180529,,,0,89dac2dd1e6f18195637ec900bd1c78a61063791845208b1962b947e9a84bc7a
thread:pr:180535,"pr #180535: [AC remat] Per-region independent recomputation, remove user phase annotations",PR blocked,63,PR blocked,1,0,,180535,,,0,69e6dc6bf6ce235a361e5c516f6acc2749dfa1718c37fd651a179e08caa5d63e
thread:pr:180608,pr #180608: [5.5/11][aoti] Add V2 runtime test via ctypes,PR blocked,63,PR blocked,1,0,,180608,,,0,05ace94accccafeda38939e931a1af5e4a803a8d3bca9dbfdb2992f65daa1da1
thread:pr:180641,pr #180641: Prune binary size from flash,PR blocked,63,PR blocked,1,0,,180641,,,0,70e709b9c9f5420482b818e3a1889f52b406399d9e17460d8f1e1425539d0d23
thread:pr:180644,pr #180644: Relax align_random_eager test tolerances for Philox float precision,PR blocked,63,PR blocked,1,0,,180644,,,0,ba8411d97f06b31428510de36bfdf4ebada0c0d8299bbdaed453804ba46276ea
thread:pr:180665,"pr #180665: [FSDP2] Fix HSDP + CPUOffloadPolicy grad corruption race, add stream-behavior tests",PR blocked,63,PR blocked,1,0,,180665,,,0,6de0358f1c1621d53a601c799c0abaa861a3faead0f25d612fb48a1b5beef9da
thread:pr:180722,pr #180722: Friendler error messages,PR blocked,63,PR blocked,1,0,,180722,,,0,5942b371ec0bce41aa9a493da554244e4da6c3388218bfb49f4583a56d00020f
thread:pr:180858,pr #180858: [inductor] symm memm ag/rs with cuStreamWrite/WaitValue32 sync,PR blocked,63,PR blocked,1,0,,180858,,,0,45cf06e2b22737dd7f11cce779cf3a6723c4a28239a71c7cbe36e801e99bed1e
thread:pr:180885,pr #180885: CUDA: unrolled kernels for upsample_linear1d forward/backward when grid is small,PR blocked,63,PR blocked,1,0,,180885,,,0,4283e3c9148da2e1130fcb04c21bcdbd228a2d2128390a79376eab86bfb2d5df
thread:pr:180907,pr #180907: [shard prop] tag more ops,PR blocked,63,PR blocked,1,0,,180907,,,0,6fd14570259f871931a411804b20dde54c16bd2974c65d8d670cd351f7d41538
thread:pr:180930,pr #180930: [Inductor] Lower MAX_COMPLEX_POINTWISE_CAT from 8 to 5,PR blocked,63,PR blocked,1,0,,180930,,,0,68413aad331c71d801d56f984a7fb30a389ed13f411828dd4db11cbec397262c
thread:pr:180947,pr #180947: [CUDACachingAllocator] Add test for expandable segment OOM stat,PR blocked,63,PR blocked,1,0,,180947,,,0,d9277821c1991320b3b5777eb83eb5f031d471340606cb0bd589e627576ce5bd
thread:pr:180990,pr #180990: [pytorch] Fix symmetric quantization scale to match C++ ChooseQuantizationParams (#180990),PR blocked,63,PR blocked,1,0,,180990,,,0,94610fa9231dd8aea5918b9fb15aabd7b076050767c8bcb21cc5bbe386f807e0
thread:pr:181000,pr #181000: [inductor] Dump Python stacks on CI test subprocess timeout,PR blocked,63,PR blocked,1,0,,181000,,,0,eff68968a73ef9d8887928fe824b18013892b75ec50dd1f2c46b8d6f3bdd1cae
thread:pr:181007,pr #181007: [inductor][comms] Optional use of process_group alloc_tensor for collectives,PR blocked,63,PR blocked,1,0,,181007,,,0,93fbf6c66407975cdf68cd33def790bf5ed6db7b189bb229dc3d4e35f57d291f
thread:pr:181008,pr #181008: [inductor][comms] memory budget for pg allocs,PR blocked,63,PR blocked,1,0,,181008,,,0,7421743fb8aaf51e8c34520c93a22af585ee38227429ea9bb793aff94f96afd5
thread:pr:181009,pr #181009: [inductor][comms] pg_alloc config for strategies,PR blocked,63,PR blocked,1,0,,181009,,,0,65b664cfea45065ffdce95909093172ea9206bc2bf0631e64bbfc25caafbadc2
thread:pr:181010,pr #181010: [inductor][comms] allow compute to borrow memory from pg_alloc mempool,PR blocked,63,PR blocked,1,0,,181010,,,0,c44a98e3ae6f6688b142e11a56c3e75dcc9d509d545d8397ee7819e6610aee60
thread:pr:181023,pr #181023: [PP] Support extra loss_fn kwargs in pipeline schedules,PR blocked,63,PR blocked,1,0,,181023,,,0,bf6244090b6f37e2839e02ea762b81939fa00c7d3c0d81f1796a27b8650e4760
thread:pr:181041,pr #181041: [hammer] Make HSTU manual recompute policies compatible with PT2 compile,PR blocked,63,PR blocked,1,0,,181041,,,0,e95a838a4d2224ebc301fe3a059b738815f87332bc319c5443b620124cb68dc8
thread:pr:181087,pr #181087: [MPS] RDMA for distributed MPS,PR blocked,63,PR blocked,1,0,,181087,,,0,3195b95b686228cb608b66df9d1cc5918748491408dd447df774ea3acf0eacd3
thread:pr:181150,pr #181150: update copy_misaligned to copy_if_misaligned,PR blocked,63,PR blocked,1,0,,181150,,,0,19e31753dc88078a64d75badfb79218fc40e4fc05d852b4c37f6c20f429d7781
thread:pr:181167,pr #181167: remove size oblivious reasoning core,PR blocked,63,PR blocked,1,0,,181167,,,0,7b5cdc5383ae0ac658d9357aa8d0076698e582d0abc1374875476cd27713e5f0
thread:pr:181178,pr #181178: [Inductor] Allow max_autotune_gemm on Jetson Orin AGX,PR blocked,63,PR blocked,1,0,,181178,,,0,863d69ed322f4eabca9f554cc3c6e7cfb818f29ad0d7c1aeb4fbd29043ab2ffa
thread:pr:181188,pr #181188: [AOTInductor] Outline per-kernel-call teardown into AOTI_NOINLINE helpers,PR blocked,63,PR blocked,1,0,,181188,,,0,904920798bc1e326ee3361fad5f4623aa53476deac8120a9f91c588c6a3d58c5
thread:pr:181221,pr #181221: [DCP] Fix flaky test_async_process_executor by resetting global state between tests (#181221),PR blocked,63,PR blocked,1,0,,181221,,,0,5dacc0478884aaecf206b57c5d97c7b9ad6a1a9a69d39e9d6af1df7d52f674bc
thread:pr:181265,pr #181265: TEMP,PR blocked,63,PR blocked,1,0,,181265,,,0,873dd7965b899eebc40b15b3f3fa4c60c7d00874b8370c91c5d78adb86570545
thread:pr:181307,pr #181307: Set device in forward input tests,PR blocked,63,PR blocked,1,0,,181307,,,0,8efc510ac60301179e558c6ce9ff22ff4ccd25a098a181f948647dced5f80ca8
thread:pr:181396,pr #181396: Print and validate full CUDA version in smoke test,PR blocked,63,PR blocked,1,0,,181396,,,0,27cb8e7176659030bdd25ef188133ee242184cce4dfd8735929aafb996db0941
thread:pr:181398,pr #181398: [spmd_types] local_map,PR blocked,63,PR blocked,1,0,,181398,,,0,e7adacfa0b6d782e28e0255f4d1f4a3ceff81f823176e482463a60613ecaf9a0
thread:pr:181414,pr #181414: Update FBGEMM submodule to v1.6.0,PR blocked,63,PR blocked,1,0,,181414,,,0,73ec1a7eb44744a2243d2af22645941c99bcd41540efc609f9b22e99acba2c61
thread:pr:181518,pr #181518: [Inductor] enable MSVC precompiled headers for cpp-wrapper,PR blocked,63,PR blocked,1,0,,181518,,,0,76d90984e7e2acf5159e1993841ebb91d258da19131ae004c00d126f66b11e14
thread:pr:181530,pr #181530: [caffe2] Fix HIP build: cccl_counting_iterator needs explicit type,PR blocked,63,PR blocked,1,0,,181530,,,0,43ef6a8b1db0f451c65406bf47a84f4e15df373e7caf8037cb159ea57b9b2fb6
thread:pr:181570,pr #181570: Fix batched LU factorization grain_size heuristic for large matrices (#181570),PR blocked,63,PR blocked,1,0,,181570,,,0,77061a1e4c97a922c2609ab7f8af7d9595cdd5dbc8cda7b68e020e69126de6a9
thread:pr:181623,pr #181623: [torch.library] Add pallas_op for JAX/Pallas kernel registration,PR blocked,63,PR blocked,1,0,,181623,,,0,c8443322c7ec8aee2498ca854a934f543bce65e4056c8cb727b02cb4a7dde54d
thread:pr:181724,pr #181724: S390X: update binaries labels,PR blocked,63,PR blocked,1,0,,181724,,,0,70784aa34814fba3102408e2a760c3c1ea7e3ab2886cf6450f6476725760bded
thread:pr:181743,pr #181743: Remove skipXPU from test_binary_ufuncs,PR blocked,63,PR blocked,1,0,,181743,,,0,a351c3beb374111e0d1852a053c88d77b92d91c15a4fcfda7ac4c18ad471d157
thread:pr:181748,pr #181748: [Inductor][Triton] Enable decompose_k and use Triton for inner BMM on HIP (#181748) (#181748),PR blocked,63,PR blocked,1,0,,181748,,,0,1bbbc74752fe5b69484d37f030987afa774cd775141cb0ca70910a57660f67f7
thread:pr:181767,"pr #181767: [shard prop] autogen infra for inplace,out variants",PR blocked,63,PR blocked,1,0,,181767,,,0,2adceba0c92423838d744164eb0602b3e13ba2a0b88661ed8ffffb7ff7fb4527
thread:pr:181826,pr #181826: [inductor][autotuning] extract _should_coordesc_tune cached property (#181826),PR blocked,63,PR blocked,1,0,,181826,,,1,53eec55b93381fa202df6d6bfb0ef10d633efe2672c923da5de22be95776a85d
thread:pr:181904,pr #181904: [lint][RAWTHROW][1/5] Small patterns sweep: annotate + fix 22 exclude groups,PR blocked,63,PR blocked,1,0,,181904,,,0,89f60d62b920b34f4c2d3a32bcc97585ca462923dcbe22b422d3bab8b600a906
thread:pr:181905,"pr #181905: [lint][RAWTHROW][2/5] torch/csrc autograd, utils, profiler, fx",PR blocked,63,PR blocked,1,0,,181905,,,0,5cd328b6afa6a4b646437bddd466113c5758ba3506596f26c8c6f33eb6d181c1
thread:pr:181906,"pr #181906: [lint][RAWTHROW][3/5] aten, c10/util, distributed, inductor",PR blocked,63,PR blocked,1,0,,181906,,,0,c096813ebe4ed5377145ea9e241b3b621cb31519af358e842995f7440d650286
thread:pr:181907,pr #181907: [lint][RAWTHROW][4/5] Live JIT: python + runtime,PR blocked,63,PR blocked,1,0,,181907,,,0,1ad7713c935ded3f6bbc8788050cde08927f77259f6b229acb596510f7c7d727
thread:pr:181944,pr #181944: [cuBLAS][cuBLASLt] Shared default workspace size across cuBLAS and Lt,PR blocked,63,PR blocked,1,0,,181944,,,0,a002cc332e58ed36adce67d2e1a788a13722abfa162d55f21eff405e9a294f95
thread:pr:182011,pr #182011: [WIP][PoC] Enable Arm Optimized Routines for AArch64+Linux,PR blocked,63,PR blocked,1,0,,182011,,,0,c174c8f993550be96b0c8352f67eb108d021c99731a68a77544e8ec2d30b2b97
thread:pr:182013,pr #182013: Rename vec256_tests to vec_tests and fix executable search,PR blocked,63,PR blocked,1,0,,182013,,,0,6baf8e865956bcbe0309f4b4905ddcc1292101cd3ef4ab830dfcb8e93f336a8a
thread:pr:182038,pr #182038: [MPS] Migrate backward to MPP on MacOS-26,PR blocked,63,PR blocked,1,0,,182038,,,0,f79fa4c87790a39ebc276ee2c5bb4a0f184a65297b102a3b29ce82a245de382c
thread:pr:182051,pr #182051: [docs] Link CUDA stream semantics from Tensor.to/copy_ non_blocking docstrings,PR blocked,63,PR blocked,1,0,,182051,,,0,712f0ca45c97bf02f41a9134ef9beebda394b58e49be2d4f0114428d3fffce96
thread:pr:182074,pr #182074: [TEST][CUDA] Bump reflection_pad_2d decomp test to use `bfloat16` threshold,PR blocked,63,PR blocked,1,0,,182074,,,0,e71c2cea62386d1bf8c1a933397fd31fd062b30acf4502a4dd9078aba47b4182
thread:pr:182134,pr #182134: [dynamo_wrapped] Remove stale dynamo_expected_failures for torch_np error-path tests,PR blocked,63,PR blocked,1,0,,182134,,,0,3fa6d98c2ad35ec5c3d9a3dd022edbccc6e9686bef1e215865dd36b55d5e5f9e
thread:pr:182138,pr #182138: Update torchcomms.txt commit hash,PR blocked,63,PR blocked,1,0,,182138,,,0,e82a80a318da5d70330cfab0244d686eb690f6e8ca4aea49769b15fe131355c7
thread:pr:182145,pr #182145: [ROCm] Reland #181955: Redirect index_add_ to scatter_add_ for large tensors,PR blocked,63,PR blocked,1,0,,182145,,,0,e8b2d8d7e7097c2a49ee7b6011a73a3d84298107e07c2f741c8d086c0bbd5dab
thread:pr:182146,pr #182146: [caffe2] Add defensive fallback in get_logging_handler for unknown destinations (#182146),PR blocked,63,PR blocked,1,0,,182146,,,0,9ef9c310c964896d74ec69b3dcf61334a5c836635595d50c557cc1e25348001d
thread:pr:182183,pr #182183: [PT2][export] Add unsafe_allow_callable_serialization for scoped callable serialization,PR blocked,63,PR blocked,1,0,,182183,,,0,35a568275512a0ecdcfeab74a554df86a60ecb40d8b272f7f8529ffefc420531
thread:pr:182193,pr #182193: Tighten any_index_expr_overflows_int32 overflow check,PR blocked,63,PR blocked,1,0,,182193,,,0,aed653a701aa1ae76a8e8b69647405d645df98d6005d5c56ef7578c47cd3cb6d
thread:pr:182199,pr #182199: [xpu][test] Port distributed _shard tests cases on Intel GPUs,PR blocked,63,PR blocked,1,0,,182199,,,0,21fa6d44efb8b50cd506959d8fba61436f661be2ccd0d724798d42cd71d1cd99
thread:pr:182256,pr #182256: [MPS] enable metal performance primitives attention,PR blocked,63,PR blocked,1,0,,182256,,,0,1b326a82b22d2d96c7f26ecd751174d668e9437abf73bab59cce8c9eb34a9b4a
thread:pr:182310,pr #182310: Add Triton scaled-mm v2 blockwise recipes,PR blocked,63,PR blocked,1,0,,182310,,,0,ecfa20b5a41071fe1e52ee1fa0ae3b23591371cbb9d2f18a18150e759b481e9c
thread:pr:182362,pr #182362: [PyTorch][AOTI] Add timing instrumentation to AOTI model loading pipeline,PR blocked,63,PR blocked,1,0,,182362,,,0,cb9651f851025d0c8a72f55fde7d3acab23eb1611550aab8bb62ec02c636159c
thread:pr:182526,pr #182526: [DTensor] register missing pooling backward sharding strategies,PR blocked,63,PR blocked,1,0,,182526,,,0,96ef5db389ae2860ce2bb215662e1afe7052a50333ffa30bd9783474dc6541b1
thread:pr:182529,pr #182529: Use NCCLCommProvider for TorchComm + NCCL symmetric memory,PR blocked,63,PR blocked,1,0,,182529,,,0,dd33c22b4e6b4907567d67091f7b27f681801cc747630d01cdc3eff8cc2951ee
thread:pr:182530,pr #182530: index_add_/index_reduce_: add env-var kill switch for VecSize=4 vectorized path,PR blocked,63,PR blocked,1,0,,182530,,,0,cb53569dd4eda7100efc145250a28285ecd0f9c709cf97c0fabd86240791f7e8
thread:pr:182577,pr #182577: [Inductor] Enable combo_kernel_per_subkernel_blocks by default,PR blocked,63,PR blocked,1,0,,182577,,,0,56d6da75021f8995addab844aa3a67cb2b317cf0470a3865cd8d31bd7a566619
thread:pr:182655,pr #182655: Update ACL/OpenBLAS/manywheel build scripts,PR blocked,63,PR blocked,1,0,,182655,,,0,cfa197a2c36c77d309b55be62e740a4a5c8adfa8746fe9609e39047173a57a44
thread:pr:182727,pr #182727: [shard prop] P(*).clone()->P(*),PR blocked,63,PR blocked,1,0,,182727,,,0,b39e4cbfd71fb3d05927cdea39fe3f65584dad39c058d838a0be05d99a7515a9
thread:pr:182729,pr #182729: [DTensor] Preserve placement for aten.clone.default,PR blocked,63,PR blocked,1,0,,182729,,,0,4b9257fb380186b303ad55b0fde8449a302b8922f53bcd52790229436f2ddf9e
thread:pr:182851,pr #182851: [dynamo] Fix free-threading race in dict recursive-tag watcher callback,PR blocked,63,PR blocked,1,0,,182851,,,0,e91cb29e3005dfdb76ddd0a849f93874a51ce69c986562059ce5580781fe4d15
thread:pr:182920,pr #182920: Add backward pass for cuDNN FP8 per-tensor attention,PR blocked,63,PR blocked,1,0,,182920,,,0,f87b7f4baabc9b189703cf3c462a6857ee8d2691dc6f957d276a60d17a295fda
thread:pr:182948,pr #182948: [CUDA] Fix reentrant deadlock in torch.cuda._lazy_call,PR blocked,63,PR blocked,1,0,,182948,,,0,5efe10224b924bac55221df137a689aa0844b8c9578014708a0ecc52096a6979
thread:pr:182963,pr #182963: [symm_mem] Add _low_contention_all_reduce to symmetric memory,PR blocked,63,PR blocked,1,0,,182963,,,0,45918cd5c36fc249eb61e80b8185078f4ebdbf11eee81fabadf4c5a7db19a7b8
thread:pr:182964,pr #182964: [inductor][overlap] Wire _low_contention_all_reduce into LC FX pass,PR blocked,63,PR blocked,1,0,,182964,,,0,40b1c7988627bec703c36dad16a710a75137d6d0ca86d930b5aee94795777cf5
thread:pr:182986,pr #182986: [Eager] [Bitwise-Equivalence] Implement an INNER_TREE sum reduction (#182986),PR blocked,63,PR blocked,1,0,,182986,,,0,1204836241d6c16106555a63405cfa751b7935c243d04db6e2baafbd52e84594
thread:pr:183306,pr #183306: [AI Codemod][ClaudeCodeFbandroidWarningsAsErrors] [3] Bulk run of CodemodConfigClaudeCodeFbandroidWarningsAsErrors (10 units) (#183306),PR blocked,63,PR blocked,1,0,,183306,,,0,08cb447637836b2ce9b70e9172a6ceff43596a3b295bfca2452b978c1ac0b1b5
thread:pr:183323,pr #183323: Adding conversion from PyObject to torch::stable::tensor,PR blocked,63,PR blocked,1,0,,183323,,,0,6466e180a05b324fad489faa80845e51008ebe5db9c49ff1283cb5ac41f07ef6
thread:pr:183390,pr #183390: [SymmMem] Add NCCL implementation of all_to_all_vdev,PR blocked,63,PR blocked,1,0,,183390,,,0,cbd2ab02ae37006434a1007c3e056d4063d8161bfc04c230541189f768a7f8e2
thread:pr:183468,pr #183468: [Inductor] Fix linter from tf32 fix,PR blocked,63,PR blocked,1,0,,183468,,,0,f8a8918a84817283f7a770453529b99b06b0b9ff94b78d2864f0bb5646e35a14
thread:pr:183617,pr #183617: [fsdp2] fix mem accumulation of all reduce input buffers for hsdp with mixed precision,PR blocked,63,PR blocked,1,0,,183617,,,0,5d52fa902c2b01b1b98d99c2161b19616f3a3de55adffa5fc44faf8a4928da5f
thread:pr:183620,pr #183620: Port 2 tests under test/distributed folder to XPU,PR blocked,63,PR blocked,1,0,,183620,,,0,71259b58fba8cbcd2eddac7245c7843d62290e9a8653c418a5c54ae25675e250
thread:pr:183638,pr #183638: [inductor] Fuse NVFP4 nested-reduction packing,PR blocked,63,PR blocked,1,0,,183638,,,0,7fe1048f444f4c9d513d5bdcf933922afb56bbd9ea0eb9d4fb62f93bb010eca2
thread:pr:183682,"pr #183682: [inductor] extract _bundle_compile_result, _all_failed_fallback, _maybe_put_static_autotuner",PR blocked,63,PR blocked,1,0,,183682,,,0,6fa882c3ea305cdb55ed85c0a8c934a8b85049e75624cc04f17045ff4e2f0bb1
thread:pr:183683,pr #183683: [inductor] extract _iter_compile_results generator with optional thread pool,PR blocked,63,PR blocked,1,0,,183683,,,0,08a067c08d7ec1f987345d1cac089293efa394087f4a70ad6837361c92819bd2
thread:pr:183684,pr #183684: [inductor] extract _bench_launchers and _finalize_autotune_winner helpers,PR blocked,63,PR blocked,1,0,,183684,,,0,1221b0fb0efbd53451b8df4688e4fbc484cfc01e297f9e014dfceabda1974c24
thread:pr:183705,pr #183705: [SymmMem] Normalize device index in empty_strided_p2p (#183705),PR blocked,63,PR blocked,1,0,,183705,,,0,042d43190ea2ae9a9dc694130f14fae688a7e10daee073e2da06d051e9a97396
thread:pr:183759,pr #183759: [Testing] Expansion of non-contiguous tensor ops testing to cover more complicated and real-world applications,PR blocked,63,PR blocked,1,0,,183759,,,0,367eb8ddafebad2f60c1ba38ad2eb4d2d7c17c45bf03dd5973ba596c3bfe5b60
thread:pr:183776,pr #183776: C++ FakeTensor,PR blocked,63,PR blocked,1,0,,183776,,,0,48221f631ce6b60114a0032d147c6a6e3c8b2fd45cc84b9450240f266733f97f
thread:pr:183847,pr #183847: [distributed][docs] Fix group_src docstring in irecv,PR blocked,63,PR blocked,1,0,,183847,,,0,5fcd8c1988d7fde5c680792d63f2b80da3c8e9aa6a2aff8b6c7a6e3d8034e155
thread:pr:183910,pr #183910: [Test] Make test/nn/test_embedding.py device-agnostic for out-of-tree backends,PR blocked,63,PR blocked,1,0,,183910,,,0,78fb09b3c11e125e603f90d447aa97a708e644e71fc62d3bbf8f9bb54757e067
thread:pr:184192,pr #184192: [Test]Refactor test/nn/test_lazy_modules.py to be device- generic by replacing hardcoded device selection with torch.accelerator API,PR blocked,63,PR blocked,1,0,,184192,,,0,eca1e94efde86acc31827ee04f81dc96bbfbacd638daa4b0b2ed5bcdf96c8e7f
thread:pr:184218,pr #184218: Support decorator/curried calling convention for torch.utils.checkpoint.checkpoint,PR blocked,63,PR blocked,1,0,,184218,,,0,7a83ea376f92b3f5e2d2fabc709a440024b261e33c477097a200abf81e2195d0
thread:pr:184226,pr #184226: [Inductor] Allow mixed pointwise/reduction combo kernels,PR blocked,63,PR blocked,1,0,,184226,,,0,5fcab6d67b2bbf26d604e39ff76fc2358cbf9eec3711b4bfe33ebc4c3a3c180a
thread:pr:184253,pr #184253: [DO NOT MERGE][ROCm][CI] Test k8s dpx,PR blocked,63,PR blocked,1,0,,184253,,,0,45d7678ffcc8ec785a63134b63ef141b49f42e9ba88a2b973711180f00035587
thread:pr:184254,"pr #184254: Fix `replication_pad{2d,3d}` meta accepting negative output dims",PR blocked,63,PR blocked,1,0,,184254,,,0,15b03102c2b1cc4e23dd15457916348445ca32297618e6ef1e496fdc7841b0cc
thread:pr:184315,pr #184315: [Test] Make autograd/test_functional.py device-agnostic for out-of-tree backends.,PR blocked,63,PR blocked,1,0,,184315,,,0,1d9284fe41bcb9f23c3cd1e519e546a3884092e5e8453231d7b75af98462f02e
thread:pr:184365,pr #184365: Memory aware partition selection for partitioned scatter fx pass,PR blocked,63,PR blocked,1,0,,184365,,,0,121848c8bc0ea4dd389d27c40409d8707837ffc7909b30b7fb399ebf3947d37b
thread:pr:184404,pr #184404: [BE] Exclude `c10/metal` from HEADER_ONLY wraps,PR blocked,63,PR blocked,1,0,,184404,,,0,b1b0dd4198370a56c7ac36f738b98b19692cedfae8c3cbf8a9cfe78c2d3e470a
thread:pr:184447,pr #184447: [inductor][autotuning] add incremental autotune foundation,PR blocked,63,PR blocked,1,0,,184447,,,0,a0f230b604e0b150e7d10663099eba0c91a8ed1933d0a4ef7ef0dd65a219b8fe
thread:pr:184487,pr #184487: dynamo: round-trip torch.cuda.stream ctx mgr across graph breaks,PR blocked,63,PR blocked,1,0,,184487,,,0,c436480f667fbdffa62a66f113dccd0a880cd05e35d8902a70fb7c168487ab89
thread:pr:184509,pr #184509: [torch/csrc] Add bulk storage helpers; remove per-byte loops,PR blocked,63,PR blocked,1,0,,184509,,,0,9352c0827d1283cfe105373cc50e4da4209ff13df904d7b63202be66f7464e94
thread:pr:184568,pr #184568: Always reset Dynamo if it was imported,PR blocked,63,PR blocked,1,0,,184568,,,0,adccf1fdc8d5c5a8df715f7c730231abd75d40d3ae84c57b2ded49ccabcd5010
thread:pr:184607,"pr #184607: Speed up inductor test infrastructure (~4x collection, ~1.7x execution)",PR blocked,63,PR blocked,1,0,,184607,,,0,0af03938a75a6d526d4cab0c0ead014d9320964796bfac7c2b48efe0419b6ca6
thread:pr:184657,pr #184657: [MPS] Tolerate lenient mixed-dtype batch_norm in InstanceNorm test,PR blocked,63,PR blocked,1,0,,184657,,,0,dae13d3a107bb7a8a33ff11d4d5b99aad38a21fa1e65649a8a88ba017f841843
thread:pr:184824,pr #184824: [Reland] Port D104346887/PR 182675 for index_add fast path (#184824),PR blocked,63,PR blocked,1,0,,184824,,,0,7c09f4c8f33d3868707058f9ace3dd1dec13a028c697e746ca92e5bc74cb087d
thread:pr:184825,pr #184825: [DTensor] add sharding strategy for group norm backward,PR blocked,63,PR blocked,1,0,,184825,,,0,0c608a5378b335dbc2d5889ad46b732c047bd66cfcc2160f71b125469e5715af
thread:pr:184867,pr #184867: [symbolic shapes] Avoid direct replacement cycles,PR blocked,63,PR blocked,1,0,,184867,,,0,04a61d2b09bd21bc9cd3a1eb0ff16c18b40b284445d78f9281d0581c4ce219c6
thread:pr:184924,pr #184924: [Test] [CUDA] Skip test_variant_consistency_jit for grid_sampler_2d,PR blocked,63,PR blocked,1,0,,184924,,,0,f3280179955be8f5a4fdda4f504567adb66b672e74c42733d1382851256bf5b2
thread:pr:184946,pr #184946: [Inductor] Fix split reduction codegen and skip redundant masking for block ptr backends (#184946),PR blocked,63,PR blocked,1,0,,184946,,,1,cb583b4227c4f079d9dbdb4107f72682a695f4631594c2ae03a2584f436cff06
thread:pr:184955,pr #184955: [Inductor] Exclude indirect-indexing nodes from combo kernels,PR blocked,63,PR blocked,1,0,,184955,,,0,3436beaca0ed40676dbd25f0e3c5c193d9fe80b4fedca82b4c8d1ed7faf942f6
thread:pr:184965,pr #184965: [Inductor] Cap combo-kernel seed autotune configs by size,PR blocked,63,PR blocked,1,0,,184965,,,0,dcc0ef482438d06f420a4238197b1141c3ba4299a8c590cc26aec41fac0ba573
thread:pr:185053,pr #185053: [Test]Refactor test_custom_ops.py to be device agnostic,PR blocked,63,PR blocked,1,0,,185053,,,0,c9835e7cc7c10ff210aa815df80facf7d5279febb8d87e87be62a471fa51c374
thread:pr:185091,pr #185091: Prefer .view() and drop redundant .contiguous() in tensor-creator chains,PR blocked,63,PR blocked,1,0,,185091,,,0,311a323a06cba1bda7138901b7eee186c43cdbc904c38a9d3db000d36d38aab4
thread:pr:185097,pr #185097: Migrate deprecated torch.norm calls to torch.linalg.*,PR blocked,63,PR blocked,1,0,,185097,,,0,a136af4b8d5a1b57dcfbc1bffdc39293c3ae14c8c68f6b44350c148f76142c57
thread:pr:185101,pr #185101: Drop redundant tuple/list/int wrappers around tensor shape and numel,PR blocked,63,PR blocked,1,0,,185101,,,0,054a87f0e92f8f058a721cfa9da0a9bbfa56e7d9622c4f1bbacf82cdf88f9147
thread:pr:185109,pr #185109: Remove custom FindCUDAToolkit.cmake module,PR blocked,63,PR blocked,1,0,,185109,,,0,c910e287c30a85109dd99f309b22f514cf5a4692f1fc11f2507f4858515582c7
thread:pr:185113,pr #185113: [Test] Decouple test_transformers.py from specific hardware backends,PR blocked,63,PR blocked,1,0,,185113,,,0,d672c2d840f33875c0c22b07ec8fafdf3f4d3e940876e4f6f54be8ef880b0e97
thread:pr:185114,pr #185114: [Test] Refractor TestSDPACudaOnly into TestSDPAGpuOnly with CUDA-only assumptions removed,PR blocked,63,PR blocked,1,0,,185114,,,0,8465037be9abf982028f60a7d469e0e54ef94dc3af913acbb944a34ab503ccdd
thread:pr:185115,pr #185115: [Test] Enable TestSDPAGpuOnly on XPU and skip cases unsupported by torch-xpu-ops,PR blocked,63,PR blocked,1,0,,185115,,,0,df094f0e5cb64feb2c2bf294e0fef3338d66e7225ec268d4c657ebbd75c85837
thread:pr:185173,pr #185173: Run cpp/test_jit and cpp/test_lazy sequentially in test_libtorch_jit,PR blocked,63,PR blocked,1,0,,185173,,,0,65f4de5053d07057ad8ba0002378cc1b6a93ee4b58a3a3b53ee05420565eeef0
thread:pr:185183,pr #185183: [Inductor] Fix combo-kernel num_warps regression on mixed reductions,PR blocked,63,PR blocked,1,0,,185183,,,0,2265f86aad07da4016b9f2963116bf8f37df18a47b7be4a9891fc1ef6750fc06
thread:pr:185213,pr #185213: [Test] Make dynamo test_streams.py device-agnostic for out-of-tree backends,PR blocked,63,PR blocked,1,0,,185213,,,0,1437c318a84507e269e032b49638d8fe3a758580107ec88ae1d1436a38a37b4c
thread:pr:185220,pr #185220: [Test] Make test_wrap_inductor_compiled_regions.py device-agnostic for out-of-tree backends,PR blocked,63,PR blocked,1,0,,185220,,,0,920a3f87636b589b09f969c0163843da523843f362d5befa2c91cf2c20666d56
thread:pr:185236,pr #185236: Detect fp32 precision flag leaks in TestCase and fix offenders,PR blocked,63,PR blocked,1,0,,185236,,,0,ff01d0c471f3c92cf85b711f147200ce19e9f756955934d8ebba0ac7c9187aae
thread:pr:185251,pr #185251: [Test] Move device-agnostic grouped_gemm tests to new test_matmul.py,PR blocked,63,PR blocked,1,0,,185251,,,1,194355a19891aaee97be4ec6423b0ee1ae4ba35d4b369921fb025118f0fed78f
thread:pr:185261,pr #185261: add meta registration for scaled_grouped_mm_v2,PR blocked,63,PR blocked,1,0,,185261,,,0,490b6d78e5c100833217ee2522a94985d7e0d2937f399e9dc11360f4fc503ce7
thread:pr:185359,pr #185359: Add CE multicast low-contention all-gather,PR blocked,63,PR blocked,1,0,,185359,,,0,0068069b39024e4f1c14b2e09e4cc91745c366ef986e0204549d06ea03644e8c
thread:pr:185366,pr #185366: [ROCm][Distributed] Add synthesized test for sac_milp to ensure numerical correctness,PR blocked,63,PR blocked,1,0,,185366,,,0,5d767d9e6325a0be94d93b3427578d0dd34ab3a614141b3f204ceb477dabeaf9
thread:pr:185432,pr #185432: Upgrade pyrefly 1.0.0,PR blocked,63,PR blocked,1,0,,185432,,,0,9885e9352bdb0f4d4e661c4348e138e8cfcb14d7be298e992b96ade4439385f9
thread:pr:185491,pr #185491: Cleanup unused tensor variables,PR blocked,63,PR blocked,1,0,,185491,,,0,2f28f9217d5aa650acd8804da421a7f49d0a2cb7219ea6210f4b24830124ce2a
thread:pr:185570,pr #185570: tmp,PR blocked,63,PR blocked,1,0,,185570,,,0,3c733cab69960c3483b874c41788a5f241df1e1330685f18e0341f8b7a50a37d
thread:pr:185592,pr #185592: Remove android directory,PR blocked,63,PR blocked,1,0,,185592,,,0,7d453a8bad9606404a691ed4d9887b2ab11ce96c19c3906583da591f4a93f426
thread:pr:185606,pr #185606: Enable TF32 fpmath mode for XPU deconvolution,PR blocked,63,PR blocked,1,0,,185606,,,0,553be284093362096e481fee2d4c6d942991ebe55e88bf24eb17a1f0c151a227
thread:pr:185646,pr #185646: [AI Codemod][AsyncioGetEventLoopMigration] [asyncio-codemod] xplat/caffe2/test,PR blocked,63,PR blocked,1,0,,185646,,,0,952061453bbda30737bd939d71396c72464d19ab9eea441bfa6b0d5b68a381ae
thread:pr:185649,pr #185649: [Test] Make test_multihead_attention device generic for out-of-tree accelerators,PR blocked,63,PR blocked,1,0,,185649,,,0,d26eb5bb7aa161f454a21812df31e12bb5809f4917b8a7ee4849311b037be61b
thread:pr:185666,pr #185666: Add ROCm CI workflows for mi210,PR blocked,63,PR blocked,1,0,,185666,,,0,7b2ea32c40d286abbe1990472dba2829718f2ad29090f2250581325cdd0b9bad
thread:pr:185668,pr #185668: Fix typos in comments and docstrings across torch,PR blocked,63,PR blocked,1,0,,185668,,,0,d3b188c9a0eba04a5530eb79624f059b67828b2fbd6f25a41635e2ed5885aa06
thread:pr:185689,pr #185689: Upload docs preview directly to S3 from the OSDC pod when authenticated,PR blocked,63,PR blocked,1,0,,185689,,,0,9d8bd5f71d17a45a78fa60c3429453ea985f68f89992e6eebf0ce23c45975a4f
thread:pr:185700,pr #185700: [OpenReg] Add compiler integration doc for OOT accelerators,PR blocked,63,PR blocked,1,0,,185700,,,0,f083e811068809a5262798c479dd048cebac667e4ca14542b203e41816002657
thread:pr:185797,pr #185797: [Test] Refactor test/test_bmm_outer_product.py,PR blocked,63,PR blocked,1,0,,185797,,,0,8301ced1134722f8d40676e8ca25061a9fe812e171e61dcc9e748bc346c19d22
thread:pr:185830,"pr #185830: Fix typos in comments, docstrings, and error messages",PR blocked,63,PR blocked,1,0,,185830,,,0,6e6e799f655beb6f39c7eaf586779d7ab7d6f2d8d555267a96938f51bd899b7c
thread:pr:185844,pr #185844: [CuTeDSL] Just use tensor strides directly in `_to_fake_cute_tensor`,PR blocked,63,PR blocked,1,0,,185844,,,0,0e5bff54d3539a2f6afbdef815adf4140e77232583559ec26b4b3b971abca4b9
thread:pr:185859,pr #185859: ATen/core/TensorBase.h: replace symint enable_if aliases with C++20 requires clauses,PR blocked,63,PR blocked,1,0,,185859,,,0,3bf55361ed6f8714a3d825e09dd106681018167745bd12e6d8f8111583006fbe
thread:pr:185875,pr #185875: c10/core/DispatchKeySet.h: add noexcept to non-constexpr query methods,PR blocked,63,PR blocked,1,0,,185875,,,0,b5499a663fd98123086541402c618253bdc1b8eca676e15bf1a51eab59dc86ac
thread:pr:185932,pr #185932: [MPS] Add `lcm` Metal kernel,PR blocked,63,PR blocked,1,0,,185932,,,0,f290467c82d041109003b94c7c2301d7739f66043f7c49378d86a6e2b428b216
thread:pr:185956,pr #185956: torch/headeronly/core/TensorAccessor.h: replace typedef with using,PR blocked,63,PR blocked,1,0,,185956,,,0,7e22b6870de37fc47b0b625df7c699044b4889854d5af4b36a31573771baf680
thread:pr:185958,pr #185958: torch/headeronly/util/TypeList.h: simplify contains and find_if with C++20 fold expressions and if constexpr,PR blocked,63,PR blocked,1,0,,185958,,,0,3224d4c95ef0b51ec7a915ca5ffc76ecf571da03c23616a0281d6d57a05e3a62
thread:pr:186010,pr #186010: Surface ROCm GPU device faults in TorchElastic signal-failure messages (#186010),PR blocked,63,PR blocked,1,0,,186010,,,0,05ad71ca160eeadef47bb59bf543f85ca81ee57d4633397898117620107b99a3
thread:pr:186020,pr #186020: [WIP] Reintroduce ptxas bundling for CUDA 13.0 binaries (revert #174716),PR blocked,63,PR blocked,1,0,,186020,,,0,b4d49960b89fd4af42aebd98ae1857ffcaa7ef519a5a3c729bde2135f13c9ddb
thread:pr:186048,pr #186048: Dedupe namespaced codegen and annotation parsing in torchgen,PR blocked,63,PR blocked,1,0,,186048,,,0,f1508a1e68fa96680399d4e73ffe2a6ae09dcd205a01ed29b6e5cf909b617482
thread:pr:186086,pr #186086: [PyTorch] Add missing meta kernel for aten::_sample_dirichlet,PR blocked,63,PR blocked,1,0,,186086,,,0,a414e2cbfc5ce61511edc8b0e195e9b479ce8ce362b7645e894ef960401d87ca
thread:pr:186095,pr #186095: [inductor] Cache per-node tiling in fusion candidate evaluation,PR blocked,63,PR blocked,1,0,,186095,,,0,ae4b5bb461f50e94ab4b54cfce5c4e8f085368f575c51a91bab66733708e27e2
thread:pr:186096,pr #186096: [inductor] Skip combined tiling computation for <=2D pointwise fusion,PR blocked,63,PR blocked,1,0,,186096,,,0,6500341c1ac9b8fc319d60ca4a2c9d1bc6f8373a849c1dd97c53b8f3de87734b
thread:pr:186097,pr #186097: [inductor] Add early exits to can_fusion_increase_peak_memory,PR blocked,63,PR blocked,1,0,,186097,,,0,2419cd8ae7a04ff0c668ef9559a0ecdc40e39d6e0de1a6f482d28c0176f424e8
thread:pr:186101,pr #186101: [inductor] Avoid O(n) set union in decide_inplace_update,PR blocked,63,PR blocked,1,0,,186101,,,0,fae4de911f7a9ecf818f5f289ae614c0efc614f6e7ddc4b91f06e3547385676e
thread:pr:186102,pr #186102: [inductor] Fast paths in sizevars for concrete integer comparisons,PR blocked,63,PR blocked,1,0,,186102,,,0,c6773970b0d966979f5d58b6aaf4cf25769c1870da9e3378512b870b4957a1a4
thread:pr:186106,pr #186106: [inductor] Fast paths in sizevars for concrete integer comparisons,PR blocked,63,PR blocked,1,0,,186106,,,0,87d26735dcfaa248eba69d37d0677413e333d77eb587ee0cfd05f64c3573ff5b
thread:pr:186122,pr #186122: [ROCm] Enable test_inverse_errors_large & test_lu_solve_batched_many_batches linalg tests,PR blocked,63,PR blocked,1,0,,186122,,,0,bd928a5ae501d7306dd7855dafe0831a7b1d623d1ea4ebb451a59f126e0f9d2b
thread:pr:186129,pr #186129: [dtensor] fix conv1d output placement and group scaling,PR blocked,63,PR blocked,1,0,,186129,,,0,a359d9dde2187594a6a62710c54d839daa972bcd6d4dc035682a3b2c8f3b4f34
thread:pr:186130,pr #186130: Add _single ProcessGroup methods and deprecate the _base aliases,PR blocked,63,PR blocked,1,0,,186130,,,0,833b9b7346a0ba65ad36db7d7ed98dbff38e8913b3a1702eccfbcb5a81811dd1
thread:pr:186131,pr #186131: Migrate ProcessGroup method usages to the _single names,PR blocked,63,PR blocked,1,0,,186131,,,0,082f36aa2aaf81d5fe0d6ae35bfdf0e52439d3a23a7a23d33f49b70bd576c249
thread:pr:186137,pr #186137: [dynamo] Swap @disable / @allow_in_graph order on invoke_subgraph_wrapper_unboxed,PR blocked,63,PR blocked,1,0,,186137,,,0,9c875b0fecabb5ca4804bc07589279180c3521eba57ecb5e3aa919cb1a80f0e7
thread:pr:186139,pr #186139: Support closed-over data-dependent indexing in scan (forward and autograd),PR blocked,63,PR blocked,1,0,,186139,,,0,cd40ee40bc6f703b1102858c4f8d7836475da322fcf3b8be2347328afcae94b9
thread:pr:186190,pr #186190: [spmd_types] typecheck FlexAttention,PR blocked,63,PR blocked,1,0,,186190,,,0,8fb300b54c4c9110e3b7cdc73dec58bdfb6c074ddac5b4db5ecc6be68876f61a
thread:pr:186211,pr #186211: [Inductor] Skip nonfinite preservation for safe scaled softmax,PR blocked,63,PR blocked,1,0,,186211,,,0,9fdd6bda913a89e0a4b766e013346f7881cf791a6eb96915862d8164e688ccfa
thread:pr:186218,pr #186218: Limit expandable IPC receiver VA reservation,PR blocked,63,PR blocked,1,0,,186218,,,0,6af0a6e06fca665ef28bea0aa9cf5d39017b2e8e003143a8a33ffd7b09e34043
thread:pr:186243,"pr #186243: [dynamo, nested graph breaks] enable NGB on dynamo-wrapped tests",PR blocked,63,PR blocked,1,0,,186243,,,0,98da4f98454cae202283919dac189d46912711422a17f50088c7df36df3722c5
thread:pr:186253,pr #186253: [fsdp2] Add SPMD no-typecheck scopes to hooks,PR blocked,63,PR blocked,1,0,,186253,,,0,ec3a6569588668ecfade4de5149451e281c4c6f89ad3b45af275342057fea97d
thread:pr:186304,"pr #186304: c10/core/DispatchKeySet: promote Full, FullAfter, Raw to enum class",PR blocked,63,PR blocked,1,0,,186304,,,0,de0d281d44be29154758f8c90a57f20e390978af46ca28feee28e153268ae1b1
thread:pr:186332,pr #186332: Improved common device test module for xpu,PR blocked,63,PR blocked,1,0,,186332,,,0,823f1bcf926ac8b99ca4b5951a67d64600c8fb3919396424877ac22a08614639
thread:pr:186334,pr #186334: Fix torch.kron crash on non-contiguous inputs,PR blocked,63,PR blocked,1,0,,186334,,,0,e09c62940dd1542af74283ddd91275911bddb25b3917c7b7f457bfb007b7f69d
thread:pr:186370,pr #186370: inductor: validate square mix-order reductions by common reads,PR blocked,63,PR blocked,1,0,,186370,,,0,344393f61b9b74d4f4097d9dd4d0fce153cd8c19cda7eaa3e27a0c2073fa92f0
thread:pr:186389,pr #186389: [BE][Ez]: Update string_util APIs to use std::string_view more,PR blocked,63,PR blocked,1,0,,186389,,,0,19cb3aad5432a55212afdc03137218e12cc3367bdd56325623e524fa68ade9db
thread:pr:186414,"pr #186414: native_group_norm: Handle non-contiguous tensors, rather than throwing",PR blocked,63,PR blocked,1,0,,186414,,,0,a3911a863bc90e825b465b7c3891162e94f7838041161ddb67d7f569dd1cd1b4
thread:pr:186458,pr #186458: Fix SDPA backend detection for autograd cache,PR blocked,63,PR blocked,1,0,,186458,,,0,bde47eb8fc8b21a99861827dd1da2cfb10e172788066976b6dec8f5df4e2269c
thread:pr:186586,pr #186586: Propagate override_cudagraphs annotation across CM-lifecycle gaps,PR blocked,63,PR blocked,1,0,,186586,,,0,602efeb3519552ed83f2d7f8c0e3f6b112386c58928a01da518edd075dec1931
thread:pr:186599,pr #186599: Introduce isInitialized&set_rng_state in torch.accelerator,PR blocked,63,PR blocked,1,0,,186599,,,0,175260a478ef4f1ea78ddd6b65fdc04953920a45ac7376b5267cfa89b255f8ce
thread:pr:186600,pr #186600: Introduce manual_seed&seed to torch.accelerator,PR blocked,63,PR blocked,1,0,,186600,,,0,aa85e4e7cd11d5eba1242094da37e2cf488053430e88beebaea13df7820eccbe
thread:pr:186608,pr #186608: [user-streams][aoti] Add minimal AOTI user stream support (#186608),PR blocked,63,PR blocked,1,0,,186608,,,0,be18e4df22818f756944d30bc5a75bfbd642e49551efb6476dc15ef0ec573591
thread:pr:186609,pr #186609: [user-streams][aoti] Add minimal AOTI user stream support,PR blocked,63,PR blocked,1,0,,186609,,,0,69310f78256c4af6345d8d194ee5d3bbfec0958b492c5ba86908b5235b51026b
thread:pr:186634,pr #186634: c10/util/intrusive_ptr.h: replace hand-written comparison operators with operator<=>,PR blocked,63,PR blocked,1,0,,186634,,,0,8e3366c3d74559089c0d09027a6ded67aa0b65d68ece34b6360ba417bfdb78d5
thread:pr:186636,pr #186636: c10/core/impl/SizesAndStrides.h: add [[nodiscard]] to size() and isInline(),PR blocked,63,PR blocked,1,0,,186636,,,0,a303777583b6185f56f3c56a1707976f43bd8efebb9231628f7168ff1bd4b52a
thread:pr:186653,pr #186653: Make topk deterministic under deterministic algorithms,PR blocked,63,PR blocked,1,0,,186653,,,0,0cb8c1b54f99f2138b48a9f8d7361763fc2686ed70c71dd2ccae562875f0143a
thread:pr:186655,pr #186655: [profiler][cupti] Benchmarks: columnar-decode roofline + monitor distortion,PR blocked,63,PR blocked,1,0,,186655,,,0,41b75af544abdbfd8bc7479a99093207cf5d8add1c0faf4da24f62401f2172bf
thread:pr:186762,pr #186762: Fix typo in quantization documentation,PR blocked,63,PR blocked,1,0,,186762,,,0,984f4ad8023be54b0185fd3a9283c29b13cbc5b94de3bcfc436b046a62e0e7ad
thread:pr:186771,pr #186771: Introduce _isInBadFork and refine torch.random,PR blocked,63,PR blocked,1,0,,186771,,,0,b6837713abe07ef772664f3a80b07c698db65646e9f68716bb7751177becb8eb
thread:pr:186777,pr #186777: [1/N] Replace module-level RNG-related APIs with torch.accelerator,PR blocked,63,PR blocked,1,0,,186777,,,0,b3c8bd4c420746d10dc573e73efd14e65cee00175f55c27a12277484b333aaef
thread:pr:186790,pr #186790: Remove c10/util/Array.h,PR blocked,63,PR blocked,1,0,,186790,,,2,dc07dbdbd5079b5ffa9214cddae0595245d842fdb17999520b0052ec2c142024
thread:pr:186815,pr #186815: Remove deprecated torch.qr,PR blocked,63,PR blocked,1,0,,186815,,,0,2dd56ba614ca38c8de513b88130e5a194ae3ee4be71d48e5351b5b6e708b283c
thread:pr:186816,pr #186816: Remove deprecated torch.chain_matmul,PR blocked,63,PR blocked,1,0,,186816,,,0,2618684af995f3952d9037da009017a3190a9c89e963c30ac8f150fec904518d
thread:pr:186828,pr #186828: Add CUDA graph output and input copy annotations,PR blocked,63,PR blocked,1,0,,186828,,,0,d0b68a38eb24636359ccad5509efcb7b2edb75ecc7cd9410dfdbf96acf3bb2c6
thread:pr:186829,pr #186829: Auto-promote CUDA graph static input copies,PR blocked,63,PR blocked,1,0,,186829,,,0,fbf2ee3853cf29fd322c74e2291891077b8fb3fab48232307d915210c880d822
thread:pr:186830,pr #186830: Add CUDA graph disable regions,PR blocked,63,PR blocked,1,0,,186830,,,0,bd162c2b764bcb36531d9362b6270cd45648098150c85c1ca952bc7ccfaf96d5
thread:pr:186831,pr #186831: Improve cudagraph stack trace coverage and assertions,PR blocked,63,PR blocked,1,0,,186831,,,0,97bdcf6f5152e8621ce5378857029df51061c8a3784acfcc95aa071c05038a23
thread:pr:186844,pr #186844: Add owner headers to docs/lint/benchmark/vllm workflows,PR blocked,63,PR blocked,1,0,,186844,,,0,ef2623c1c603f9d563d1c6fb13305a62e1db71e204f5e75527747377353cb541
thread:pr:186848,pr #186848: Use C++20 concepts (1 of N),PR blocked,63,PR blocked,1,0,,186848,,,0,a30d204916d885b2cca7edcaea087e5b7078294effbb10b6b92ca71c56f23d13
thread:pr:186877,pr #186877: Use C++20 std::numbers in c10 MathConstants,PR blocked,63,PR blocked,1,0,,186877,,,0,876e4182cd3c85652b9371360cf20ad62d660f70a7aa4e3ff65dedeb93eae479
thread:pr:186880,pr #186880: Add FlexAttention shared memory compile hints,PR blocked,63,PR blocked,1,0,,186880,,,0,62286ff0e3859f28d6dcf749ca3042818b62bab582258652ebfca6ee7bae323a
thread:pr:186882,pr #186882: RFC Change error order,PR blocked,63,PR blocked,1,0,,186882,,,0,5adeeeaf1b5a289582aa0dfc56753254dbbdb9ca70495308e42df016a099c9b4
thread:pr:186899,pr #186899: [2/N] Replace module-level RNG-related APIs with torch.accelerator,PR blocked,63,PR blocked,1,0,,186899,,,0,43e2b00b3a155aa45efe418924dcf90429e7e4f699eb6c64d14ca38bb060bc85
thread:pr:186902,pr #186902: Generalize dynamo RNG-related part to be device-agnostic,PR blocked,63,PR blocked,1,0,,186902,,,0,af174a04ccc71f5d47d91e58c54dd74b3a33dbd84b9c898fbfd882708d8f573f
thread:pr:186906,pr #186906: Update trace_rules for torch.accelerator,PR blocked,63,PR blocked,1,0,,186906,,,0,0eaae39759bd68f0715818c059f4ef7844f7b23dc68aab6cd7592a8281e81660
thread:pr:186915,pr #186915: [WIP][XPU]Enable inductor cases when issues are fixed,PR blocked,63,PR blocked,1,0,,186915,,,0,1cbc2bc1aeb9c4dedb0f82fad438fb28f89093f9cbb774f614b679a94c137be4
thread:pr:186926,pr #186926: Remove sleef from test_aoti_abi_check linkage,PR blocked,63,PR blocked,1,0,,186926,,,0,aca6d22050bf409a6b47ed7b31ca958ef7fdd6408277baf70275003a9e7b7d91
thread:pr:186946,pr #186946: [mps] Add kernel for native_group_norm_backward.multiple_grads,PR blocked,63,PR blocked,1,0,,186946,,,0,03a405255b416222b80ee1d20185319777b9ad32289466fdf700a6ab1783e377
thread:pr:186947,pr #186947: [functorch] Add native_group_norm_backward.multiple_grads plumbing,PR blocked,63,PR blocked,1,0,,186947,,,0,b24964a72beaccc389044dee10ba21364c6e66abff62af5db32f22607686b380
thread:pr:186949,pr #186949: Fix for scan autograd issue,PR blocked,63,PR blocked,1,0,,186949,,,0,d76de7e336e1fbbd8bdfa9d868e1322abb0fbda80236cb6e33be8b2b53634118
thread:pr:186968,pr #186968: dynamo: enforce target_values is None|Sequence.,PR blocked,63,PR blocked,1,0,,186968,,,0,8952444b16428d24bc93fa4be76bc011a4faeed8aa22cc73ca5acd3bc1971e87
thread:pr:187002,pr #187002: Use C++20 <bit> instead of c10::llvm bit helpers in c10,PR blocked,63,PR blocked,1,0,,187002,,,0,531d85eb7f5bb2e56af9171c58e53a839cb4a86989ad3831a61e0724e7f48d44
thread:pr:187007,pr #187007: Use distinct cache variables for linker option checks,PR blocked,63,PR blocked,1,0,,187007,,,0,12fa9c9f181abb1bfdfa1e452c50d92cf85d82ad9deedaef5064478af5b18889
thread:pr:187009,pr #187009: Apply libnop clang warning suppression to TensorPipe subtargets,PR blocked,63,PR blocked,1,0,,187009,,,0,802c8fa8d5ba567c2a6b04bc0e5b687a74bad1c0b550572cf1ecd5db343aa50e
thread:pr:187015,pr #187015: remove spmd_types dependency from nightly wheel,PR blocked,63,PR blocked,1,0,,187015,,,0,56e7eacaa5eef1fe439ba06651b064fdc4e40a57bd7994c7f3d982fb7ce06e11
thread:pr:187061,pr #187061: [cuda graphs] Automatic nn.Module FQN annotation in Inductor cudagraph trees,PR blocked,63,PR blocked,1,0,,187061,,,0,bd1f9ff6c2a265722ec32c575ac0c169e100ad11a91eb598329c06f228d85169
thread:pr:187062,pr #187062: [inductor] Expose compile_worker_malloc_conf to tune the compile-worker allocator (#187062),PR blocked,63,PR blocked,1,0,,187062,,,0,6451dd754e87ff202eaf15712fcbd3579bb6319ecd51bab85e29fc746788f5ea
thread:pr:187069,pr #187069: [torch][autograd] Fix Python refcount leaks in autograd C++ glue,PR blocked,63,PR blocked,1,0,,187069,,,0,0af31cd0d9cb38689e6147d7870713402df8c8a44190ba4ad2b253272febf807
thread:pr:187131,pr #187131: [muon] add multi-tensor (foreach) standard Muon path (#187131),PR blocked,63,PR blocked,1,0,,187131,,,0,4ffefde432664487f0ba6fddbe71a4e6474c1101499ae97fd638bf0869ac90b8
thread:pr:187132,pr #187132: [muon] add Gram Newton-Schulz algorithm + batched standard NS for square groups (#187132),PR blocked,63,PR blocked,1,0,,187132,,,0,10d87b813547b3cb582eea798c0a373edd373a24ee07c71db4b126a97102b41b
thread:pr:187133,pr #187133: [muon] add CUDA graph capture for the NS computation (#187133),PR blocked,63,PR blocked,1,0,,187133,,,0,89b4c55ea28e1d23bf79cf8fa16b24f15a0724035c3724158c0d18fd87f4a18d
thread:pr:187134,pr #187134: [dynamo] Reuse ACT-traced graphs for resolved tensor inputs,PR blocked,63,PR blocked,1,0,,187134,,,0,fd51aeb07ec9dfebeff376021523ae5591d00a7042d0aa11a19b28f8876f0861
thread:pr:187142,pr #187142: Fully type-annotate torch/utils/data/graph_settings.py,PR blocked,63,PR blocked,1,0,,187142,,,0,74be9d7b1cfaf7fddf0d7851fe778e7e9dc122ac6ad6f0e55a7dd77288517a06
thread:pr:187150,pr #187150: Remove unused CUDA 12.4 docker build paths,PR blocked,63,PR blocked,1,0,,187150,,,0,e607bd0c8c7309aea286c53d4d6ea625e2396835195327f0eefb7bff062a2776
thread:pr:187164,pr #187164: [ROCm] unskip export tests on ROCm,PR blocked,63,PR blocked,1,0,,187164,,,0,0fb48d8ba9a2fb39a52949188be2fce1bd2c8de57871b1a5f8d34451c598235c
thread:pr:187168,pr #187168: Fix accuracy degradation for baddbmm on CPU for float tensors,PR blocked,63,PR blocked,1,0,,187168,,,0,3d6da1f10d83cfd569db13f858244425567501fa80498787d65a968786b027db
thread:pr:187189,pr #187189: [DO NOT MERGE][ROCm] Bump default trunk ROCm test shards to 20,PR blocked,63,PR blocked,1,0,,187189,,,0,f29b294209f1280ddbc374f192778f7a7a6e3fcdc94d90cf49f0103ae206df33
thread:pr:187205,pr #187205: Add get_device_name and memory snapshot accelerator APIs,PR blocked,63,PR blocked,1,0,,187205,,,0,d7719ff4fce53d67c0c4db7ee5c0b96bf8c30e0b5f9af77265d9d04adaad8ce1
thread:pr:187208,pr #187208: [BE][Ez]: Use CPP20 rvalue overload for ostringstream,PR blocked,63,PR blocked,1,0,,187208,,,0,6e48bc2174b5d4a3de2b1b3b0fdbb047c8ded3b3085774f682ec04023587036a
thread:pr:187224,pr #187224: Duplicate concurrent alignment fixups,PR blocked,63,PR blocked,1,0,,187224,,,0,de1226f38c616851a66c0f317328047f8d0309d0e980be083484495e01664ff9
thread:pr:187288,pr #187288: [vllm hash update] update the pinned vllm hash,PR blocked,63,PR blocked,1,0,,187288,,,0,ca609e3d06f04d34f879d1daf2fdc4c6d56e7bc3b3e1b4541a0ab25c09998b75
thread:pr:187307,pr #187307: Fix the potential global range overflow issue,PR blocked,63,PR blocked,1,0,,187307,,,0,b7bbf4834597d2ade90f1a42fc368b476d05f160f77096913ff70ab131e54c6f
thread:pr:187357,pr #187357: Convert _scaled_grouped_mm_v2 to a structured operator,PR blocked,63,PR blocked,1,0,,187357,,,0,58b5ec90eb5c88f21c472aee885b4fbd26d0e9f833cd8fac4ee608845bd15258
thread:pr:187367,pr #187367: Fix typos in comments and docstrings,PR blocked,63,PR blocked,1,0,,187367,,,0,cab751c3c5cf21fa1f6cde9680112a6432bb5b2b252cc7fcf8eb1a3167f3e252
thread:pr:187369,pr #187369: [Test] Add test hardware classification for test_complex.py and test_kernel_launch_checks.py,PR blocked,63,PR blocked,1,0,,187369,,,0,6bf81c9ced7f375d8985ce0eec417d64d2ca8c6141786b4fc065ac635ad270ba
thread:pr:187390,pr #187390: Widen aot_autograd annotations to match runtime values,PR blocked,63,PR blocked,1,0,,187390,,,0,124b216670531e8f4e635f6efad5feb04f9473cd59e7c9cb5b5e25241c09b68f
thread:pr:187391,pr #187391: [export][serde] Serialize SymBoolArgument in graph input/output specs (#187391),PR blocked,63,PR blocked,1,0,,187391,,,0,60884911e6c0e03b57ecbbf522bed4e4a5016e2bfd4f2dbccb7e5c2dbcb6c7c7
thread:pr:187419,pr #187419: Add numerics-debugging skill,PR blocked,63,PR blocked,1,0,,187419,,,0,5cc4b3bc26dd99e987be496e62f82e60bc13dbe87834548df629a5cc6e7c3313
thread:pr:187430,pr #187430: Add Kaiser-Bessel Derived window,PR blocked,63,PR blocked,1,0,,187430,,,0,ad449664a76efffb992c71da9aead5b8ad7b63455dbd661707dc5e83272eab22
thread:pr:187471,pr #187471: [inductor][mtia] Strip AMD-only autotune kwargs from MTIA Triton compile configs (#187471),PR blocked,63,PR blocked,1,0,,187471,,,0,162cc372b7343c7c8e2ee2d5013765ce2090bd3bdb4e04005a44c99e073aea78
thread:pr:187517,pr #187517: [profiler][cupti] CUDA-graph correlation utils: node-id remap + side-stream events,PR blocked,63,PR blocked,1,0,,187517,,,0,e6d4010673531607adfcb4807791139f9991fd822364c9de299b44a801b4a5af
thread:pr:187518,pr #187518: [profiler][cupti] In-process NCCL metadata plugin (ncclProfiler_v6),PR blocked,63,PR blocked,1,0,,187518,,,0,29c913b827895f5dfab699226ed0af5699f8f3467324673739a271a459e56e27
thread:pr:187519,pr #187519: [profiler][cupti] Comms monitoring: CommsObserver + comm hook + plugins (CollTrace replacement),PR blocked,63,PR blocked,1,0,,187519,,,0,cacd06cda98aa98a1c51464781c966498cd8d86522e61d616a77abeb780956d4
thread:pr:187520,pr #187520: [profiler][cupti] Comms monitoring tests,PR blocked,63,PR blocked,1,0,,187520,,,0,1deadfdc31ab44905d0bd444ad8fa628351f26c6a5803c723838250d51842770
thread:pr:187529,pr #187529: perf(dynamo): add guard lookup token plan,PR blocked,63,PR blocked,1,0,,187529,,,0,fb0602665d4e4b8db46bdce5e58d767e85210032c34cdf2870f073be174f41f7
thread:pr:187545,pr #187545: [BE] Use `std::numbers` in `c10/util/MathConstants.h`,PR blocked,63,PR blocked,1,0,,187545,,,0,25acd4a0f759b4250d20aa7717b50ce4545e6eed3152e0af192fcdee142da716
thread:pr:187548,pr #187548: Update ck aiter gfx1250,PR blocked,63,PR blocked,1,0,,187548,,,0,e22218a52083f7469407ba649d3d32208d72a86feebfce9584a168099da3a1e2
thread:pr:187549,pr #187549: [xpu][test]testing: enable tf32_on_and_off() for XPU devices,PR blocked,63,PR blocked,1,0,,187549,,,0,c16eeff7e17a07539220261c673092d22b080ce665d66e7bfd5ccca0c44e3725
thread:pr:187550,pr #187550: [test][inductor] Fix stale `xfail` and add SM12 skip.,PR blocked,63,PR blocked,1,0,,187550,,,0,96b50888434bd54f39e87efb4e984bc0bfd37ffc45708ccd7172b969d9787824
thread:pr:187619,pr #187619: Remove Dynamo guard from custom op kernels,PR blocked,63,PR blocked,1,0,,187619,,,0,352c3586a41a16c30b1d2a68f50e9e23b82c98cfb374e7c09e291cc093fe715c
thread:pr:187642,pr #187642: Add parameter-contiguous all-gather (symm_mem.all_gather_offset),PR blocked,63,PR blocked,1,0,,187642,,,0,ed6e3dabaf4fe59f1378469bd927e18443bd019941fa0257dfcffd5a27d7aef8
thread:pr:187675,"pr #187675: [inductor] Add fuse_region to set fusions,memory reuse boundaries",PR blocked,63,PR blocked,1,0,,187675,,,0,3deb1ab84f7bb72bb78da2b781b37399f67a04a19a46d46aa606a620b80380a6
thread:pr:187691,pr #187691: [BE]: Update fmtlib to 12.2.0,PR blocked,63,PR blocked,1,0,,187691,,,0,cb5ef9372af94513293c64f426d3f779180795aa4ccf468d0f5e6bbf900e1f07
thread:pr:187692,pr #187692: Skip mobilenetv2_100 eager-nondeterministic accuracy check,PR blocked,63,PR blocked,1,0,,187692,,,0,c88c57c0d1138a673d1394caaa2459a75ac32411ea1b4dfee5d0127d8bc64288
thread:pr:187699,pr #187699: Add a debug log when we skip sync_decision_cross_ranks (#187699),PR blocked,63,PR blocked,1,0,,187699,,,0,be28c7db43d0f4b217f9e70489aee4c02dc2e708ed2b9e50496bb22336baadff
thread:pr:187711,pr #187711: Strict pyrefly checking: round 1 (2/N),PR blocked,63,PR blocked,1,0,,187711,,,0,ab0378de3fe712ed88a3e7c805cbe92de4a714bcfc22b1901aa9d3fc3c43e213
thread:pr:187712,pr #187712: Strict pyrefly checking: round 1 (3/N),PR blocked,63,PR blocked,1,0,,187712,,,0,3eb3a60108d4e9f5556d4381d1f3bd883811fbe14c3f0c2641bab40017d334c3
thread:pr:187732,pr #187732: [DDP] Fix static_graph=True + no_sync() gradient accumulation regression,PR blocked,63,PR blocked,1,0,,187732,,,0,23ea9a26af54444a03d684e3d8b35da786dede3b6d50a3dd7277decb7181445f
thread:pr:187750,pr #187750: Tune all_gather_offset for skewed buckets,PR blocked,63,PR blocked,1,0,,187750,,,0,530caf945b91f99d6047811f96c0fca385278d50dd7014a17a5eefa774ddd725
thread:pr:187776,pr #187776: Speed up all_gather_offset writes with a shared unrolled copy helper,PR blocked,63,PR blocked,1,0,,187776,,,0,ce1e768565ac5157423baadddfce5ff14f670e399ff5ad92becc3b867d67d3a2
thread:pr:187778,pr #187778: Flatten all_to_all_nd copy to fix narrow-row throughput,PR blocked,63,PR blocked,1,0,,187778,,,0,709f737ffeff7d78826cd96b0a2db9bb082a12fd6cb24f8cb068364d63ffb321
thread:pr:187784,pr #187784: [TokenSwitch] Add expert-major dispatch layout,PR blocked,63,PR blocked,1,0,,187784,,,0,7894000bc011578e8edb7412fa6bc045574e843c54577c96a0195383474be264
thread:pr:187785,pr #187785: [TokenSwitch] Add symmetric-memory zero-copy dispatch/combine,PR blocked,63,PR blocked,1,0,,187785,,,0,8ab7d60d78b61ea9e32dba106f1307b763a73ceb91fc9679f36a910558800b19
thread:pr:187797,pr #187797: Add linalg SVD operator benchmarks,PR blocked,63,PR blocked,1,0,,187797,,,0,3689ece8b4c33ed065c21c046f2f5197192e2fac1448baddf4478c80562d68bd
thread:pr:187809,pr #187809: Make DTensor logging and export tests accelerator-agnostic,PR blocked,63,PR blocked,1,0,,187809,,,0,ed72e4c4f994ba421fd2a0f95a952e93e711bdf8d5e05cbde6185a6fcf901809
thread:pr:187823,pr #187823: Fix regex for extracting gist URLs from logs,PR blocked,63,PR blocked,1,0,,187823,,,0,5efdf2a537fb56983e0f5e9c5f57dd341157651e8c0be7f52a9e0749d4ab4320
thread:pr:187832,pr #187832: [FX][Dynamo] Add tests for nn_module_stack and source_fn_stack metadata contracts,PR blocked,63,PR blocked,1,0,,187832,,,0,53af2ca5745a9224c0010c28588f3a6d0dca1b8fbb4c4828b7bc954850cb6daf
thread:pr:187842,pr #187842: [inductor] Avoid cross-extern fusion when allow_peak_memory_increasing_fusion is off,PR blocked,63,PR blocked,1,0,,187842,,,0,c5ec9638a68805883d023f6b5c8737304c8f70d0fbbc7f32530f31446f5b9776
thread:pr:187860,pr #187860: fix DDE in meta__transformer_encoder_layer_fwd,PR blocked,63,PR blocked,1,0,,187860,,,0,59859e064a1d1a2044ec3ff639f112793844df24e94790dae19d68802d619bcb
thread:pr:187875,pr #187875: [AOTI] Refactor model container checkout ownership (#187875),PR blocked,63,PR blocked,1,0,,187875,,,0,3103ac4f5c97f83f80fcc6b8ec740acaef10aba72cecf434e098b8f433842f9c
thread:pr:187876,pr #187876: Fix mark dynamic half bounded range,PR blocked,63,PR blocked,1,0,,187876,,,0,f423665d5603dbdaca7798301de103368053082fa83816a46aca0a49767d9655
thread:pr:187891,pr #187891: device_mesh: fall back to new_group when parent backend lacks split s…,PR blocked,63,PR blocked,1,0,,187891,,,0,005d1f3451b8763be7e1235fe4958698f760d61957a11546fa8fc531ec53e60d
thread:pr:187897,pr #187897: fix: Add explicit null-buffer guards and shared-memory fallback for Intel Macs with discrete GPU (MPS),PR blocked,63,PR blocked,1,0,,187897,,,0,b2eb9c5bc370ffe67d421153fe9a264a3a6f782f508cc03d6d4b61c14729f897
thread:pr:187922,pr #187922: [TEST] Refactor test_embedding.py with hw_classification,PR blocked,63,PR blocked,1,0,,187922,,,0,6b508b327556e194f54bec3aac6f218dcba912e6d3f5ecf805e7159d9544872f
thread:pr:187931,pr #187931: [XPU Graph] Refine XPU graph capture detection in XPUCachingAllocator,PR blocked,63,PR blocked,1,0,,187931,,,1,d380206dd8a975c51ffbc9f4eb8941a1d126b030e1c117d1991bafabe1d1596c
thread:pr:187938,pr #187938: [Inductor] Support non-Triton fused kernels in the FX backend via a high-order-operator,PR blocked,63,PR blocked,1,0,,187938,,,0,cbd04578b3936810fb3b4ba593ba2a2f729fc1c0e8aac86b9c629771848286db
thread:pr:187961,pr #187961: [inductor] Add compile-time per-subkernel autotuning for combo kernels,PR blocked,63,PR blocked,1,0,,187961,,,0,d7bea22432190076d8c3b8e400b9e0a733602c0fa5cfd7b6b330c8cfc5d63e04
thread:pr:187962,pr #187962: [inductor] Evaluate compile-time ProcessGroup attrs in overlap estimates,PR blocked,63,PR blocked,1,0,,187962,,,0,489b1fa749972b375c3fab22ce0d973b5a41b100fe0f601de28907b9df9e17b1
thread:pr:187964,pr #187964: [inductor] Evaluate compile-time ProcessGroup attrs in overlap estimates,PR blocked,63,PR blocked,1,0,,187964,,,0,0bd1b2243ce4c01355d61ff7c06b5d448afcc8963e7b2de1e9da5257c0b51a39
thread:pr:187965,pr #187965: [inductor] Avoid disjoint horizontal fusion when allow_peak_memory_increasing_fusion is off,PR blocked,63,PR blocked,1,0,,187965,,,0,c30e7e5117c914d1ae556e2d1855a93391e0fddba9a6cf41c08b31aae8533f33
thread:pr:188000,pr #188000: Vectorize LayerNormSecondPass null affine cases,PR blocked,63,PR blocked,1,0,,188000,,,0,fd3a52a0c9b8cf8bb8253076925bef47dcc23dc7a2372d27cb21ac2262c91f6f
thread:pr:188043,pr #188043: [Test] Make test/test_sort_and_select.py device-agnostic for out-of-tree backends,PR blocked,63,PR blocked,1,0,,188043,,,0,02774b06983b2fd681bd66f124af777ff469df175217ac8fdd8f121d06f59870
thread:pr:188050,pr #188050: Add FX graph cache custom guard replay test,PR blocked,63,PR blocked,1,0,,188050,,,0,fe5604c9fefbda0e4d385c5b798dd1656fb6313e64fc7340c3dc670dc84d847d
thread:pr:188055,pr #188055: [inductor] Evaluate compile-time ProcessGroup attrs in overlap estimates,PR blocked,63,PR blocked,1,0,,188055,,,0,97826f0321b549c04cffad91163518258fb1938b6968ae5aafb0118a95d08682
thread:pr:188056,pr #188056: [inductor] Avoid disjoint horizontal fusion when allow_peak_memory_increasing_fusion is off,PR blocked,63,PR blocked,1,0,,188056,,,0,a594eb077c853586744206e828f14f0c1f1a083ea713c3f7920df6684cc2b2f3
thread:pr:188057,pr #188057: [inductor] Evaluate compile-time ProcessGroup attrs in overlap estimates,PR blocked,63,PR blocked,1,0,,188057,,,0,94cf3b575d6e5ac69fb0d0d5130ae5ad8dd2dc84c2e6618e38389dfb4c738a36
thread:pr:188058,pr #188058: [inductor] Avoid disjoint horizontal fusion when allow_peak_memory_increasing_fusion is off,PR blocked,63,PR blocked,1,0,,188058,,,0,5e71b5bfe60486e902883e9805c2574cad4faf9b8fc5fb401aac45d70512db30
thread:pr:188059,pr #188059: [inductor] Block only peak-crossing fusions in the peak-memory guards,PR blocked,63,PR blocked,1,0,,188059,,,0,e0ba100f000cafd1a060de2dc330260a471a39c2633fc281f9f629aec13210a9
thread:pr:188063,pr #188063: Enable Blackwell FP8 grouped GEMM,PR blocked,63,PR blocked,1,0,,188063,,,0,bbb8a72ed3c9640c531118cecaa01c91407264dd19c09734546f92223db50ba6
thread:pr:188067,pr #188067: [ROCm] fix wait instructions,PR blocked,63,PR blocked,1,0,,188067,,,0,2711542887974669b9efc0927d6fc0bfecd383af3d0a3caf06781791e34c8974
thread:pr:188093,pr #188093: Fix record_stack_traces: filter fx-internal frames from recorded trac…,PR blocked,63,PR blocked,1,0,,188093,,,0,3cbd118b11eda20d02fceba4f9d8fe80d7dff45312d5cd9b4f48d467bbb1395a
thread:pr:188094,pr #188094: [profiler][cupti] Remove unused Python reference decoder (CuptiMonitorBuffer),PR blocked,63,PR blocked,1,0,,188094,,,0,781c5a37a90aaf11780cdb931832094996462ad8fa296664d06319667b2ed8e2
thread:pr:188101,pr #188101: Require an installed dict watcher for the recursive dict-tag fast path on 3.12+,PR blocked,63,PR blocked,1,0,,188101,,,0,577fa1a737dd1a10e641649326dc385a77895d28f85554c59f19bbbfd6fc90f1
thread:pr:188103,pr #188103: Enable recursive dict-tag guard optimization by default on 3.12+ non-free-threaded,PR blocked,63,PR blocked,1,0,,188103,,,0,2ed6f237aea4adcac94df4dfa5cbfda4c0f3c4635f1a8f92c9f042dacdf3f94d
thread:pr:188130,pr #188130: [Test] Make test_autocast device-agnostic for Out-of-Tree backends,PR blocked,63,PR blocked,1,0,,188130,,,0,5685691896fce032131b523dc256d08ca6118bc113300ae5a7f7b3526c638c7b
thread:pr:188137,pr #188137: [inductor] Decompose shard_dim_alltoall via regular decomposition,PR blocked,63,PR blocked,1,0,,188137,,,2,24e943d09d9eb825cfaf019e87cda997d3ae9fdc7685a2b8e394a518eb996a7a
thread:pr:188161,pr #188161: Demote changing CUDAGraph-managed inputs to copied inputs,PR blocked,63,PR blocked,1,0,,188161,,,0,1044fb81b5e20e0504818bd9f4f47d3bb32da941bbebd1b57af9a97adf0ef284
thread:pr:188165,"pr #188165: [cutedsl] Pass current compute capability e.g., SM 10.3 -> `sm_103a` to CuTeDSL compile subprocess",PR blocked,63,PR blocked,1,0,,188165,,,0,10ec401694f8f206568c8ed6e715ce148f2048e01c20589c6cc70f3517c92ad0
thread:pr:188179,pr #188179: Use tiling-aware hints for persistent reductions,PR blocked,63,PR blocked,1,0,,188179,,,0,4b02c6e4c550b046b07c6af1f77f62a7a32cfe22f5aa844cb1c5dbf9a54df899
thread:pr:188180,pr #188180: Reuse full-range splits for persistent CSE,PR blocked,63,PR blocked,1,0,,188180,,,0,6b79701be912d6f58d7712654a18c83842115adbc5705110d4d4ba738979efdd
thread:pr:188188,pr #188188: [DTensor] local_map normalizes shard dim checking in_placements,PR blocked,63,PR blocked,1,0,,188188,,,0,da7b4c1f4edcae9ff752a89952a913e6ca5ca8c5a6b2ac2e9e32963a9a00324e
thread:pr:188215,pr #188215: [dtensor] Resolve ProcessGroup from mesh in-graph under compile_on_one_rank,PR blocked,63,PR blocked,1,0,,188215,,,0,1637be6faf061a6140a8f1b137aab8790384752a602b78e6ee4f4a0ea608bf19
thread:pr:188262,pr #188262: [CI] fix OpenBLAS build for non-AArch64,PR blocked,63,PR blocked,1,0,,188262,,,0,40064f0156fc0601478863d0ca3220459424b331b4f490b3821b53890ac937b7
thread:pr:188276,pr #188276: [FSDP] Add MORI SDMA all-gather backend and zero-copy output capability,PR blocked,63,PR blocked,1,0,,188276,,,0,8a6cd7c83c943cb03fdb89a3212176fac1546de90d07560420f4c99a21510774
thread:pr:188289,pr #188289: [mtia] Make torch.distributions test suite portable across acelerators,PR blocked,63,PR blocked,1,0,,188289,,,0,3aae2f74d77eda95021a3d4bfc7fa247048e4fdfdb807f7662cb0b6ef7ad2c40
thread:pr:188294,pr #188294: Fix GradientEdge None-grad error message formatting,PR blocked,63,PR blocked,1,0,,188294,,,0,fb83bd53fd64f8b96175b3ad9df22f5c95d6682f2ce1e4786b64fda1d130af6c
thread:pr:188295,pr #188295: Validate queue_callback input is callable,PR blocked,63,PR blocked,1,0,,188295,,,0,89bfa1ffabc90fa7b59e08235edd75ad2f83dbae16bd2072d945438c2d5b3be4
thread:pr:188296,pr #188296: linear_cross_entropy: support probability targets with reduction='none',PR blocked,63,PR blocked,1,0,,188296,,,0,776fad87c747e2797ed025c6eaf637751937363005011390348789398585cf25
thread:pr:188298,pr #188298: [aot_autograd] Add PySourceBuilder for runtime-wrapper codegen,PR blocked,63,PR blocked,1,0,,188298,,,0,40702d07a3a8e7cd972b3a061962403e350b6f7a0adc427111dd2b1b79feb654
thread:pr:188311,pr #188311: [cpu] Address clang-tidy warnings in vec(256/512)_int.h,PR blocked,63,PR blocked,1,0,,188311,,,0,013aa79ecc5ac1bd77d738f25fb9039302eb2c6786e39606e4daac8d61e7d072
thread:pr:188312,pr #188312: [cpu] Address clang-tidy warnings in vec(256/512)_qint.h,PR blocked,63,PR blocked,1,0,,188312,,,0,04b873bd0dccb3d262f729f835ea7d56cf8b1b963f13056083e220f59b44d39b
thread:pr:188316,pr #188316: [ROCm] Cap HIP stream pool per priority; add reservable streams,PR blocked,63,PR blocked,1,0,,188316,,,0,efa88ffa9b3223e59c4a6af985b2ee818b2c40b4ca3a549290bfc9971917fd49
thread:pr:188339,pr #188339: Fix/typo overridden applier,PR blocked,63,PR blocked,1,0,,188339,,,0,2f0ba13204222d6d89177fe10c6a4af2080c61e0561671ceb1e21718ff80a43b
thread:pr:188340,pr #188340: Fix/indices or section typo,PR blocked,63,PR blocked,1,0,,188340,,,0,ee18bf8b781be14dccc32b0c517488309cec4b4ea4eb76e6bd13c42bc0cacf54
thread:pr:188341,pr #188341: Fix/184845 pad docs,PR blocked,63,PR blocked,1,0,,188341,,,0,a471b9f04ec1561c37099b719cf086e4c3fdcb8fb4fb27bdda9b0e8ba41cbd5c
thread:pr:188385,pr #188385: [fx] _iterate_exprs: treat device/dtype/layout/memory_format as symbol-free,PR blocked,63,PR blocked,1,0,,188385,,,0,fada74f4df855f85de45bc74d73b8915d95ba527b1273dd9b5cb95c7d0b007cf
thread:pr:188439,pr #188439: [CUDA] [CI] fix potential unit test failure after cuda graph xfail (resource leak),PR blocked,63,PR blocked,1,0,,188439,,,0,dd02c71d1296376e7d2b7cece8c6901ca57296a22f4945fbd19ab84f11dee694
thread:pr:188444,pr #188444: [Test] Refactor tensor_creation_ops and foreach.py to be device-agnostic.,PR blocked,63,PR blocked,1,0,,188444,,,0,77592070426881f2ff8aad028eff554cd7ea2f07f0a9f44d845147362e97df9e
thread:pr:188451,pr #188451: [cpu] Address clang-tidy warnings in vec128_float_neon.h,PR blocked,63,PR blocked,1,0,,188451,,,0,fc26d2ccbac68a57f59866294854a5aed5218f19151210705aeb9798adb8b0c2
thread:pr:188465,pr #188465: [Doc] Document GIL and allocation constraints for PU1 custom allocators,PR blocked,63,PR blocked,1,0,,188465,,,0,f9e09f861574375ceb68311f5fdc96c43e250b77b74f66810d95a73e3b447db7
thread:pr:188498,pr #188498: Fix/grammar with the same parameters,PR blocked,63,PR blocked,1,0,,188498,,,0,6ca27f9fd140c470aca7da83493e11c48e8c8c63d510a2d1361e9afe73d76b5d
thread:pr:188499,pr #188499: Fix/typo paramater,PR blocked,63,PR blocked,1,0,,188499,,,0,965d24cc4878a1403b047b2a6d7e36251928920b5f2a93d38cd42a14560a39dd
thread:pr:188501,pr #188501: Fix/147170 deterministic partition ordering,PR blocked,63,PR blocked,1,0,,188501,,,0,34c62acf24ca898504ffe1817609a93221677f5f8e4cc5a98cf2e2a01312ff47
thread:pr:188502,pr #188502: Fix/188034 autograd setup context leak,PR blocked,63,PR blocked,1,0,,188502,,,0,d00453dbd74d18b131779a9af5e597af3735a5af0ae5ec5840d8180b94ac6756
thread:pr:188516,pr #188516: Fix/187177 localresponsenorm validation,PR blocked,63,PR blocked,1,0,,188516,,,0,4fef82b955440391b46d87c8c993cdd940a87f8304d0e6d4a044aea2ec2e6245
thread:pr:188517,pr #188517: Fix/180346 dtensor clip grad mixed mesh,PR blocked,63,PR blocked,1,0,,188517,,,0,4ee4507634792af234d1d01dcb247c61d9af6f9f46148384788f66023ea7e8f1
thread:pr:188532,pr #188532: Fix/threshold inplace validation,PR blocked,63,PR blocked,1,0,,188532,,,0,9774142298ca6fbc7ba1da44ef5b43433b22953a4f47b4955f883041ba40f39f
thread:pr:188535,pr #188535: [test stats] Sanitize empty-file pytest rows at the upload boundary,PR blocked,63,PR blocked,1,0,,188535,,,0,ef23b73f7eda53ce44e38579b3ad4a577228a44202903a548d321c8bff4c5fb9
thread:pr:188563,pr #188563: [AI Codemod][PerfAICT-General] Enable auto-vectorization of at::vec::convert via __restrict and indexed access,PR blocked,63,PR blocked,1,0,,188563,,,0,f899d69d3e412f08233d80c4b6250ff618f13df7daff793bcb10d6faa12bf9be
thread:pr:188571,pr #188571: C++ meta kernel for torch.cat (symint aware),PR blocked,63,PR blocked,1,0,,188571,,,0,4711faa52a38fd2044db2d1feee76d400287e424ef6675f571efa4f46c0e54ce
thread:pr:188581,pr #188581: [dynamo] Add polyfills for operator.contains and operator.call,PR blocked,63,PR blocked,1,0,,188581,,,0,8e656869a138404de7c1874fac19ce2f1816ca91c20a839b7a956a34dece8ee9
thread:pr:188616,pr #188616: [AI Codemod][PerfAICT-General] Skip per-operand has_storage scan for non-PrivateUse1 devices in build,PR blocked,63,PR blocked,1,0,,188616,,,0,72b793c4efd6d0e412c34ed6c6badc67805ac6a919901b712d5afa7fdb6c7a4b
thread:pr:188643,pr #188643: [AOTI] Add external constants API for zero-copy weight sharing (#188643),PR blocked,63,PR blocked,1,0,,188643,,,2,7bbfc90e5c5beac4ecbd41e16f50417265f9252ffdd397a4b7d8045085e6f7f2
thread:pr:188646,pr #188646: [inductor] Add tile_n=192 and missing cluster configs to NVGEMM supplement set,PR blocked,63,PR blocked,1,0,,188646,,,0,90017547e5e6015fe1e0b738934c52d4d03ca4e0aad4879d887fd23702f3dc68
thread:pr:188650,pr #188650: [SymmMem] Allocate symmetric memory through the CUDACachingAllocator v2,PR blocked,63,PR blocked,1,0,,188650,,,0,7529dcc92505b561fec49ce35da3721874f6bd833627324c8c8bf23db449d7ef
thread:pr:188736,pr #188736: [xpu][inductor] update Intel Triton commit pin,PR blocked,63,PR blocked,1,0,,188736,,,0,6bf5f19cbc2b275c8ec4574cb0a2926a21f0ae308029e23b6cbdb4922ed68a0f
thread:pr:188744,pr #188744: [dynamo] fix lazy module initialization with symbolic shapes,PR blocked,63,PR blocked,1,0,,188744,,,0,e1d881595ee0210d858b2b1ac7c7a74202491ad983442590fbccfc2b9b3c490f
thread:pr:188745,"pr #188745: [dynamo, nested graph breaks] fix empty nn.Module hook dict reconstruction",PR blocked,63,PR blocked,1,0,,188745,,,0,b74546b936a1dc1ceb259eb4a02d6a0b20c7d545e4c79d8c28c73191d1d473ce
thread:pr:188771,pr #188771: [inductor] Rewind block_ptr across outer loop in nested reductions (#188771),PR blocked,63,PR blocked,1,0,,188771,,,0,d5b6922604b2c318661e0d6bd90fe54e83e04b1ba4a0f52270e4f73e7c450920
thread:pr:188789,pr #188789: [xpu] Add IPC memory handle sharing support to XPUCachingAllocator,PR blocked,63,PR blocked,1,0,,188789,,,0,ef76d070926ce5a6d4d0a929bbde277c5fc8bc5c64b0b33f1e48209c1697cc8d
thread:pr:188822,pr #188822: [Inductor][Host-side TMA] Support tensordesc<> in the static Triton launcher,PR blocked,63,PR blocked,1,0,,188822,,,0,68b21ae358f6c903fe99cc4fa78443ec9b7b8d5d6271db3e6a3b046f4b4fcf9b
thread:pr:188826,pr #188826: [AI Codemod][PerfAICT-General] Use SymDimVector to avoid heap allocs in select_symint,PR blocked,63,PR blocked,1,0,,188826,,,0,42ad92808ccbd50a7b9f11d0c518d6e8ddd5ca810beb22b8a2432d5eb5f2f174
thread:pr:188827,pr #188827: [AI Codemod][PerfAICT-General] Use SymDimVector to avoid heap allocs in select_symint,PR blocked,63,PR blocked,1,0,,188827,,,0,e00a67ccd5ced0bcd48099b8a87cdee681803a963bfa3367b760ae4236907082
thread:pr:189005,pr #189005: fix(fx): filter internal frames in record_stack_traces to resolve broken stack traces (#130861),PR blocked,63,PR blocked,1,0,,189005,,,0,7da921d26a736c8a479b504721f33e47fdc2c0ee05258df9f23da966f6afa17a
thread:pr:189027,pr #189027: [dynamo] Guard os.environ reads with an ambient env-var guard,PR blocked,63,PR blocked,1,0,,189027,,,0,ba9d45088d1747036e19c2ecad9ec803c5b052a225759329f31edfffd02a3401
thread:pr:189042,pr #189042: [dynamo] Add specialize_args option to assume_constant_result,PR blocked,63,PR blocked,1,0,,189042,,,0,b6b6a6ab310b16e9dcd100ffaa6b785882b2f525ce122261eb8dd37f71764300
thread:pr:189044,pr #189044: [inductor] Handle symbolic float scalars in FX wrapper metadata hook and reference analysis (#189044),PR blocked,63,PR blocked,1,0,,189044,,,0,2c29df9a6cef63c6ff9e7b9a5755b830ebb108a1df35dfdac15fc110e673f794
thread:pr:189062,pr #189062: [profiler][cupti] Make CUPTI monitor buffer_size / flush_period_s config-only,PR blocked,63,PR blocked,1,0,,189062,,,0,aea0a40537551c7ead7ee9f7e322bbba90c286e375ab81bf2d7a6705b203e447
thread:pr:189080,pr #189080: [inductor] Generalize reinplacing to functional ATen ops with an in-place variant,PR blocked,63,PR blocked,1,0,,189080,,,0,48880bd8ae3089e20c7230a73232a8875900dedb2ddb865252ca9ef9b7cf1cb2
thread:pr:189103,pr #189103: Support dist.split_group on CPU-only (gloo) process groups,PR blocked,63,PR blocked,1,0,,189103,,,0,0834adfcb29a853680b3e0a3d907cbc42b4ade85b27f12cdbf9a6e78ab104a69
thread:pr:189107,pr #189107: Introduce Inductor lowering for _philox_uniform,PR blocked,63,PR blocked,1,0,,189107,,,0,d7cf231056f59687e17de73ac625987670fad3fa7bdaad8f7c8c4a2221944e71
thread:pr:189111,pr #189111: [torchcomms hash update] update the pinned torchcomms hash,PR blocked,63,PR blocked,1,0,,189111,,,0,4a82bd598bfb243d526f060f38057cd618f4bdd46c77e9a3ff90430a9d0122a2
thread:pr:189153,pr #189153: Bump onnx from 1.21.0 to 1.22.0 in /.ci/docker,PR blocked,63,PR blocked,1,0,,189153,,,0,fe7a50480f45d886da669ce24ba3fbcd080ae3ada61a050de95cd49ea0bfa79d
thread:pr:189178,pr #189178: Add PyTorch AI policy,PR blocked,63,PR blocked,1,0,,189178,,,0,227553073eebc9daddc9c24f3866269bd625fbd6dddca6e162a2fde6694ff178
thread:pr:189192,pr #189192: [MPS] Add `geqrf` and refactor `linalg_qr`,PR blocked,63,PR blocked,1,0,,189192,,,0,b33eee3a5457f117111046668ba1090b3d2301de5d23304cde4786dc2982e1bb
thread:pr:189195,pr #189195: Cuda Graph Default Enable (#189195),PR blocked,63,PR blocked,1,0,,189195,,,0,08a6911e5145b9f7a6fe1c0ee36dae72122061ecaf0ab8826ff9a688887cd2cc
thread:pr:189197,pr #189197: make post-processing in codegen mandatory,PR blocked,63,PR blocked,1,0,,189197,,,1,bebfeffbe1d3456d2f0bfc7701102d18f14a9d793380b7fde65d27aa8129dbf8
thread:pr:189202,pr #189202: [CuteDSL][RMSNorm] use const_data_ptr for alignment check,PR blocked,63,PR blocked,1,0,,189202,,,0,76ee2c97c9bc0ccd85b104b80d8aac27c8aa4ec37181faa24f1c740fbddd2ea1
thread:pr:189206,pr #189206: [Inductor] register `target_code` in `input_codes` tracker,PR blocked,63,PR blocked,1,0,,189206,,,0,763387faed673946e22d7212cc0bef80844fcdde8e4bdc1768667fb6c1abbb68
thread:pr:189217,pr #189217: [vision hash update] update the pinned vision hash,PR blocked,63,PR blocked,1,0,,189217,,,0,8b8274322cf0fc62b6abb49b3cb88728030d71e3d8059eab63e8ed4bb9718a52
thread:pr:189220,pr #189220: Strip the arc experiment from workflow check_experiments,PR blocked,63,PR blocked,1,0,,189220,,,0,7d2438dfec3ed24d0bbaaede9add377966702a877a5b2eaee93c839295972d39
thread:pr:189246,pr #189246: [DO NOT MERGE] Test L3 CRCR,PR blocked,63,PR blocked,1,0,,189246,,,0,21ed46b4a1f0a7987caa35724e92141405dbc0fc40724dfe0c94056da88328c1
thread:pr:189247,pr #189247: Register _foreach_clone XPU dispatch,PR blocked,63,PR blocked,1,0,,189247,,,1,af979e0f4202a6e2733269b7c345f397f00480627284e1066fac74773237ca96
thread:pr:189249,pr #189249: [Dynamo] Remove dead code in tvm and common backends,PR blocked,63,PR blocked,1,0,,189249,,,0,bad8ee85842e6fa5194441a8e1379957f486b559ac2d5b7150720c94657516a0
thread:pr:189257,pr #189257: [rocm] Re-enable conv2d backward parametrized test on ROCm by using fp16 on NAVI,PR blocked,63,PR blocked,1,0,,189257,,,0,430508baa264637293d0db90cabf933ee3081428aa6ec50927c40c0069f51138
thread:pr:189264,"pr #189264: [ROCm][CI] Make rocm-nightly workflow fully mi300-specific (name, ciflow tag, filename)",PR blocked,63,PR blocked,1,0,,189264,,,0,fea62ae3f37a6fa9e3354c24fb93de466fff9acc0a84fe394135739b2b78a279
thread:pr:189266,pr #189266: [ROCm][CI] Fix py3.15 ROCm manywheel numpy source build (pkg-config python3),PR blocked,63,PR blocked,1,0,,189266,,,0,8039d2ba7008288082188fbfae72828ca01c0ecc940ed0d24b320c8c9a9a6669
thread:pr:189276,pr #189276: Update third_party/kineto submodule to cb6e5f7,PR blocked,63,PR blocked,1,0,,189276,,,0,51668dd18e51e9022235cbf1bca9f814255012f0be6006ceb77118204c71de53
thread:pr:189277,pr #189277: Skip redundant contiguity rescan in empty_tensor_restride (robust) (#189277),PR blocked,63,PR blocked,1,0,,189277,,,0,4f490b87bec696cf6fb2fcd11c02095b8fb4e9dc7b7219119afa29494d9e2654
thread:pr:189302,pr #189302: [ROCm][CI] Make ROCM_SO_FILES a single shared source of truth for wheel bundling + preload,PR blocked,63,PR blocked,1,0,,189302,,,0,744e5e80d23ef1d56599e42408e2686944f1c583b044b1ffb4ec3b2e77c82b0b
thread:pr:189305,"pr #189305: Reapply ""Remove CUDA sync in torch.combinations to improve `torch.compile` support (#186595)""",PR blocked,63,PR blocked,1,0,,189305,,,0,6e59bd15ec01c30967aa9654ab5d619192d19b317bed5ef9220611cfaef38c67
thread:pr:189311,pr #189311: [wip] Reduce hackery for Dynamo appeasement,PR blocked,63,PR blocked,1,0,,189311,,,0,28d91b76a00a95cd62627815c78eb118349927e44bad861652bb30413f1dada8
thread:pr:189318,pr #189318: Bump pip from 26.0.1 to 26.1.2 in /.ci/docker,PR blocked,63,PR blocked,1,0,,189318,,,0,165898d69823c76d4d33ed3421311c6c8a8ac920db87e3fc7904077d4019ae1e
thread:pr:189325,pr #189325: [Dynamo] Fix self-referential bw_compiler on backend re-entry,PR blocked,63,PR blocked,1,0,,189325,,,0,674a4f5d35dea8f58136feed1870b7ab331ab5d63e62acaefc472d5b19c0b29e
thread:pr:189328,pr #189328: Add retry with backoff to download_mnist.py,PR blocked,63,PR blocked,1,0,,189328,,,0,7feb026d446f98596e3a22637326a98471ef2d9b8257237c236a629e0b527c7b
thread:issue:124931:issue:149534:issue:163408:issue:21876,"issue #124931: Missing description of Transformer argument ""memory_mask"" shape in 3D (including the batch dimension) case",stale/low urgency,62,stale/low urgency,12,0,"21876, 24826, 24930, 25132, 28657, 32529, 32590, 34573, 74092, 124931, 149534, 163408",,,,11,ea4ea6fad4e3a76360b63ef6705b40b9fffc97cc1c160094ecc0d7c6032bc2b1
thread:issue:113040:pr:185782,pr #185782: fix: [PT2] [Hardening] Track recompiles alongside graph breaks in our actu...,has linked issue,62,Review active PRs and linked issues as one maintainer work thread.,2,0,113040,185782,,,3,0ec6f11b2a743ebdb90235d85a8ff8f9591a89eedf32c275ac6ae460b7edc53e
thread:issue:179250:pr:179273,pr #179273: [optim] support of dynamic coefficients and normalization in Muon,has linked issue,62,Review active PRs and linked issues as one maintainer work thread.,2,0,179250,179273,,,1,01a20551e8eb5317ae4b9b22c7e4ab297368023ca30e5d0a8737e0ac4e196eae
thread:issue:108744,issue #108744: switch more test cases to use MultithreadTestCase,has linked PR,62,has linked PR,1,0,108744,,,,0,59d9dffd0d26d106f87142a2e771c9caea982e872b73566ed84ce7f575540df9
thread:issue:115016,issue #115016: Enable ruff rule PLW1510 codebase wide,has linked PR,62,has linked PR,1,0,115016,,,,0,002d02415f0fad354f9951bf987b2a504940b2975f628d42236909ab01ae399c
thread:issue:133443,issue #133443: [BE] Deduplicate auto_functionalized and triton_kernel_wrapper_functional,has linked PR,62,has linked PR,1,0,133443,,,,0,bffe75689b0e1c1140c0a5c6e1835da58f1db831fa265e4290ee6fe7de83e3a6
thread:issue:187781,issue #187781: [RFC] Topology-aware nn.Module primitives for memory optimization in torch.compile,has linked PR,62,has linked PR,1,0,187781,,,,0,18cfc82c9d8dd48fab27bd5bedc9d85134c7f2b3664467ed1ff23876590c655e
thread:issue:48251,issue #48251: torch.eye(d) is slow and hogs cpu for d >= 182,has linked PR,62,has linked PR,1,0,48251,,,,0,8128f4a325d36297b9890b065a2637bd637d2f5c28e10890ff2f12b4de0c1201
thread:pr:111673,pr #111673: [quant][bc-breaking] Remove deprecated QConfigDynamic,ready for maintainer decision,62,ready for maintainer decision,1,0,,111673,,,0,e8a6063a86c02dd19a9ed772752191aaaf8d33e0dd5bb967cfd39943c1d8a1d4
thread:pr:120076,pr #120076: Change ATEN generator argument type to const std::optional<Generator>&,ready for maintainer decision,62,ready for maintainer decision,1,0,,120076,,,0,37bd83220d0fce988461e37a001ff8c15c1781e453fee491d912bc0595503a91
thread:pr:124490,pr #124490: [dynamo] Support ndarray.dtype attribute access,ready for maintainer decision,62,ready for maintainer decision,1,0,,124490,,,0,7993a81991c8af40893f333d417f00750d4171f82ddfc938e82eccac83924d3c
thread:pr:132814,pr #132814: Update TwoTensor to no longer inherit requires_grad from inner tensor,ready for maintainer decision,62,ready for maintainer decision,1,0,,132814,,,0,5c491b2d6dd9a66d010922d03fe25198b2d91a06d63d61695db4ac81673fadfb
thread:pr:138436,"pr #138436: Replace use of PyTorch 2.0 with torch.compile, and minor edits",ready for maintainer decision,62,ready for maintainer decision,1,0,,138436,,,0,3f1fb43e0d82469f591e6574190d1aa1635cac9d86afdfecebcad4cf81809c2f
thread:pr:138471,pr #138471: Add torch.compiler.config.cache,ready for maintainer decision,62,ready for maintainer decision,1,0,,138471,,,0,29369c5f8171e4974bb924d3a98b8d4cb5b436bcd4f9820559ff4047323721b2
thread:pr:138626,pr #138626: Fix global namespace pollution in ATen/Dispatch.h,ready for maintainer decision,62,ready for maintainer decision,1,0,,138626,,,0,095782b02532ea63f033da03c9ea1339edefa9307c99cb59b5bc6b64e887aadb
thread:pr:146172,pr #146172: Factory function support for NestedTensor,ready for maintainer decision,62,ready for maintainer decision,1,0,,146172,,,0,9ddd4fc81f27738b8d6d6ff785574ae78e2c00f0e0f77fc77d270565743e40e6
thread:pr:161044,pr #161044: Thread based dataloading workers,ready for maintainer decision,62,ready for maintainer decision,1,0,,161044,,,0,8f9b242fa81e923e3cf32891f19508256d52b8a5d7b35d9a7c56dd2264a13891
thread:pr:166813,pr #166813: Add min/max support for barebones uint types,ready for maintainer decision,62,ready for maintainer decision,1,0,,166813,,,0,06458dd27a13cef427e5365c604fdedb02eb33401a848fea8311282c3eec7989
thread:pr:176001,pr #176001: [test] Resolve undefined GRAPH_EXECUTOR in test_jit_profiling.py,ready for maintainer decision,62,ready for maintainer decision,1,0,,176001,,,0,308276ee1316e73a8ac07282d268b8d588c017f0eb6b5b0ce80886bf7f799f4f
thread:pr:183335,pr #183335: Internal,ready for maintainer decision,62,ready for maintainer decision,1,0,,183335,,,0,4dd474320dc789bb368a7425d5075582e51fb372ed23dd5680c92e8157224d5a
thread:issue:102948:pr:189313,pr #189313: Support aten::cumprod ONNX export via CumProd (opset 26),draft/noise,61,Review active PRs and linked issues as one maintainer work thread.,2,0,102948,189313,,,1,f4854131840f37d8fde538a0f3b83521394c64d4129963fff7e7063857991858
thread:issue:129456:pr:186373,pr #186373: Disable duck sizing by default,draft/noise,61,Review active PRs and linked issues as one maintainer work thread.,2,0,129456,186373,,,1,1b825476e612704515bd15261e743c471d31c6c60bd46cad41c21cc842003c86
thread:issue:136271:pr:186283,pr #186283: Trace data_ptr through tensor constructors,draft/noise,61,Review active PRs and linked issues as one maintainer work thread.,2,0,136271,186283,,,1,f9b765392910533d736a32ae8781f306b2ddb9a5dc0cdb5cac30ff4ca679a3e7
thread:issue:139080:pr:186189,pr #186189: Report dynamo expected-failure XPASSes,draft/noise,61,Review active PRs and linked issues as one maintainer work thread.,2,0,139080,186189,,,1,aae0d1da2a4feaeff25e5f6b69995513c1d5a9286a8dcb08cd6240d0248d92be
thread:issue:140960:pr:186082,pr #186082: Improve SymIntArrayRef concrete-int diagnostics,draft/noise,61,Review active PRs and linked issues as one maintainer work thread.,2,0,140960,186082,,,1,136d19fb99caacd7ba03b9523a300dbab3c244a19b9d1bb5f5ec5368d1612335
thread:issue:149895:pr:185749,pr #185749: Fix Dynamo construction for classes with __getattribute__,draft/noise,61,Review active PRs and linked issues as one maintainer work thread.,2,0,149895,185749,,,1,adffba064fad719d0ad3a8d61290ad51d292d8906e92085bff6e6713c6b8e29b
thread:issue:149921:pr:185748,pr #185748: [dynamo] Model RemovableHandle hook registration side effects,draft/noise,61,Review active PRs and linked issues as one maintainer work thread.,2,0,149921,185748,,,1,80377bc349b4367c7670093be3f3e5ef4fc36169350af597a53137c99e00989a
thread:issue:163836:pr:181860,pr #181860: [c10] Add InlineEventBase to unify event APIs in DeviceGuardImplInter…,draft/noise,61,Review active PRs and linked issues as one maintainer work thread.,2,0,163836,181860,,,2,b4267badde386fa6cd35f46061fcc6fe9ed6a8dd555f9dae81bdebd000dd9893
thread:issue:187451:pr:187452,pr #187452: Polyhedral Optimization Pass for PyTorch Inductor (Phase 1),draft/noise,61,Review active PRs and linked issues as one maintainer work thread.,2,0,187451,187452,,,2,b756245e5408e5fc44e55e5549811fdd2f05d4cdb3fb70389ec43ec618e9a752
thread:pr:133289,pr #133289: Update native/mkldnn to native/onednn,draft/noise,61,draft/noise,1,0,,133289,,,0,06f190b93c15639990f6147d1125a43fc9a9a1b9c3fd8426dbee89b665d8d632
thread:pr:133297,pr #133297: Update cmake,draft/noise,61,draft/noise,1,0,,133297,,,0,0a5bb472e5288b6856fc8f1c92b7a32830b441b949929415dde68490285c31a1
thread:pr:177115,pr #177115: Fix opcheck to detect stride mismatches on CPU tensors,draft/noise,61,draft/noise,1,0,,177115,,,0,9d61fd1feb2f61d40b0b46ec23125107801a3762630a68d0802060b533c34ca9
thread:pr:177205,pr #177205: Fix all_gather_into_tensor with stack-shaped output under torch.compile + dynamic shapes,draft/noise,61,draft/noise,1,0,,177205,,,0,3c151696b57c7b958e4a6af6b6b3d63326fd86450db5875588849b0f7189e10a
thread:pr:177287,"pr #177287: Default backward_pass_autocast to ""off"" to match eager autocast semantics",draft/noise,61,draft/noise,1,0,,177287,,,0,e316923a22fd06e5e7d664fcb97b2937b253ee477eccc58669eab56b7c9adf0a
thread:pr:179443,pr #179443: [FSDP2] Fix HSDP O(n_layers) memory accumulation of reduce-scatter output buffers,draft/noise,61,draft/noise,1,0,,179443,,,0,7f7c535e435cedcc95a8c93c9f641ec126785c660ca39c8c78a6280f6da2bc46
thread:pr:179625,pr #179625: Enable ruff rule PLW1510 codebase wide,draft/noise,61,draft/noise,1,0,,179625,,,0,933fb491b5337960fe17f11b2e57125122eb4da3cce2da476fec3b2a2ddd378b
thread:pr:180379,pr #180379: [Dynamo] Inline root module forward hooks when compiling modules,draft/noise,61,draft/noise,1,0,,180379,,,0,cb38fd069565880cb70fecb1308992f0dee248584de2d523bce2792a1c7afba8
thread:pr:180707,pr #180707: Improve cpp_extension and drop setuptools runtime requirement,draft/noise,61,draft/noise,1,0,,180707,,,0,3e427b79fceaccae04c2f5a5a8ec19c5877dd13d3f12176b91c647009fe27a35
thread:pr:180716,pr #180716: Fix torch.compile addmm with beta=0 and mismatched bias,draft/noise,61,draft/noise,1,0,,180716,,,0,a83309e6340d7d0b372adc4aa84615761c8702008cc23ce5919518bcc3af619c
thread:pr:180955,pr #180955: [ROCm] Normalize AMDSMI UUID matching,draft/noise,61,draft/noise,1,0,,180955,,,0,2e0b7cfed62380bb1defc21b355b2b104c400365fa9e91ce2906056b19134288
thread:pr:182768,pr #182768: fix(resize): rollback tensor metadata when storage resize fails,draft/noise,61,draft/noise,1,0,,182768,,,0,5ca0015cda442be1655fe2403a986cdf96d62d97dfe42bbc976208f153635cbb
thread:pr:183909,pr #183909: changing Sleef_asinhf8_u10 to std impl,draft/noise,61,draft/noise,1,0,,183909,,,0,4e866b99f0ab546884985f0c221e4f6a17a0b5201db15f9ba4a6e43fc0ca45d2
thread:pr:185308,pr #185308: Align subclass output SymInt unwrapping with metadata,draft/noise,61,draft/noise,1,0,,185308,,,0,3fb37a1db58846277139688598e5098bb1e8ae4b7318230f4830fb42d8cf0f59
thread:pr:186797,"pr #186797: Fix mm_plus_mm XPU: guard broken Triton template, enable all mm_plus_mm tests",draft/noise,61,draft/noise,1,0,,186797,,,0,1cfeb173e2f4bb50e10b47c5b9586e33f69a7552672e34a8007731f44356330b
thread:pr:187461,pr #187461: Do Not Review: Polyhedral Optimization,draft/noise,61,draft/noise,1,0,,187461,,,0,d4b2c35547fc83448628fd01c9accebfe58739aa970d4336356b1a1e144d427a
thread:pr:187490,pr #187490: [Dynamo] Trace raw unbacked SymInt inputs,draft/noise,61,draft/noise,1,0,,187490,,,0,dbf4dc3f8f40ac1f4ae50aaf599b421729231796208ddfd68ae2631b27a79b1d
thread:pr:187883,pr #187883: Avoid per-symbol tuple allocation in the C++ symbolic shape guard,draft/noise,61,draft/noise,1,0,,187883,,,0,44f45f2d683b6d00d0ed40ff47d2b31489e750bafb0ab96bd119b8a37afbeee8
thread:pr:188037,pr #188037: Fix current stream selection,draft/noise,61,draft/noise,1,0,,188037,,,0,1c8abd8293550d72be0d3b9b4873eb6f0730389a244e58c06d319696a84f9c06
thread:pr:188178,pr #188178: Guard score_mod oob indexing,draft/noise,61,draft/noise,1,0,,188178,,,0,f6d4c576362b3d2053fb7a641454dd64be9663b005dd9f91050468a30379e83d
thread:pr:188253,pr #188253: Add XPU dispatch for linalg.polar,draft/noise,61,draft/noise,1,0,,188253,,,0,ae1c1ac4c5c60bd33511250d5b126ed61ccd7c8ac74999cbcc13d494cd36407d
thread:pr:188350,pr #188350: Fix Dynamo stream/event variable python_type,draft/noise,61,draft/noise,1,0,,188350,,,0,07d9dcc22c6c6c4110daef098fe2d8edc975b092f9f0afdae17d081a2ee5b005
thread:pr:188894,pr #188894: Fix HAS_XPU_AND_TRITON to require triton XPU backend,draft/noise,61,draft/noise,1,0,,188894,,,0,8347e7d92cfb06552a4a939af7b699a92a10c2e75fbeee18170a2302fc09652d
thread:pr:188992,pr #188992: [inductor][XPU] Fix AMP autocast round-trip fp16 accuracy (#188261),draft/noise,61,draft/noise,1,0,,188992,,,0,266404404de212780228af0048fea88fe5d85fb7b6e278c157915a0a25dc4cf2
thread:pr:188993,pr #188993: [dynamo] Preserve Event/Stream subclass identity across graph breaks (#188536),draft/noise,61,draft/noise,1,0,,188993,,,0,645eaf4a9083fadf46cf578341382d325778f118de2e8e9f819dbddf1a3d6c6e
thread:pr:188995,pr #188995: [dynamo] Allow raw unbacked SymInts in strict mode (FlexAttention BlockMask) (#187547),draft/noise,61,draft/noise,1,0,,188995,,,0,d375d270e1a29627a3b3193a9c64ae9e0858eb11ccc5ffea8eb1d0fb71ee1e3b
thread:pr:189147,pr #189147: Fix unload utility for newer Triton pyds,draft/noise,61,draft/noise,1,0,,189147,,,0,ed95131adeeb103eb4113a63e2960bf6fd0af9f2e4fef913547b4504483ad44b
thread:issue:106846,issue #106846: untimeError: The following operation failed in the TorchScript interpreter. Traceback of TorchScript (most recent call last): RuntimeError: nvrtc: error: invalid value for --gpu-architecture (-arch),needs triage,59,needs triage,1,0,106846,,,,0,9e73b0eb03da6c2586140dd0ab4e37ae106fcd9cb53e6e2192b9c322138d623b
thread:issue:133227,issue #133227: Issue Exporting PyTorch Model with pytorch3d.ops.ball_query to ONNX: Bug or Compatibility Problem?,needs triage,59,needs triage,1,0,133227,,,,0,5cff260e9b5658beb4fcca0a44658a3a2ef97b436689332911ed830fa9f3e4b1
thread:issue:139745,"issue #139745: cmake libtorch_jni.so, bug: 'fp16/psimd.h' file not found",needs triage,59,needs triage,1,0,139745,,,,0,600f85af704657744f933463556fa81d05a0d1da99390a86c5e20ed4823fb57f
thread:issue:15342,issue #15342: [JIT] Batch op. batch_sum return a mask which is on cpu,needs triage,59,needs triage,1,0,15342,,,,0,84b4ef801e42a15663ef088f8824dff2ce9f186eb456b8529bc67a8ce8a163a4
thread:issue:154708,issue #154708: torch.quantized_batch_norm API return different output when input is nan/inf or undefined mathematical calculation method,needs triage,59,needs triage,1,0,154708,,,,0,8ca7602b8fc32292799e45de11c2e6e9915d58614dbc5e6896df6532271f549e
thread:issue:158701,"issue #158701: RuntimeError: optimized_plan_ INTERNAL ASSERT FAILED at ""/opt/conda/envs/bld/conda-bld/libtorch_1741644046984/work/torch/csrc/jit/runtime/profiling_graph_executor_impl.cpp"":725, please report a bug to PyTorch.",needs triage,59,needs triage,1,0,158701,,,,0,3bcc48a7f287fd367cd5172799d81b13a0ab066f34e320b367eef4b86a6f2d75
thread:issue:16592,issue #16592: [JIT] NVRTC unknown error,needs triage,59,needs triage,1,0,16592,,,,0,21f02a0c4d3a9868baa3094c5e368bb9753e3e190c4484336d53abf930bf6c4b
thread:issue:167605,issue #167605: [Profiler] About adding the line number of calling stack from parent,needs triage,59,needs triage,1,0,167605,,,,0,ced1d7cf23d3f8a0a33180964375bbced2b7a825527acd885ec529a800f4ddff
thread:issue:19106,"issue #19106: Performance issue with torch.jit.trace(), slow prediction in C++ (CPU)",needs triage,59,needs triage,1,0,19106,,,,0,b055cded612bb3fc8bd73144b0f0583814704203edc990c84da2bc252e38bfec
thread:issue:30525,issue #30525: [Tensorboard] nothing appears in the projector tab,needs triage,59,needs triage,1,0,30525,,,,0,cdf681dbf996ae47d6721ccb483109f383c756f4eef3afd4519a3fe6bfef8e4e
thread:issue:30651,issue #30651: [TensorBoard] The different order of import SummaryWriter may cause Segmentation fault,needs triage,59,needs triage,1,0,30651,,,,0,d35d40b86c7585cb642102db422025d094bd67e3c8b6e8912b8f7e3f465cf921
thread:issue:49157,issue #49157: softmax fail,needs triage,59,needs triage,1,0,49157,,,,0,a77c3d72e91fc2b3b3c1856d773a029d5281f850f1c91689815e5d7c566c8b94
thread:issue:60930,issue #60930: error for using faster rcnn on c++ & GPU,needs triage,59,needs triage,1,0,60930,,,,0,7bf7874252c577ad29b7caaa108436c101716fb0dda605415980e780fc859a82
thread:issue:97876,issue #97876: Missing torch import in _contextlib.py when using torch.jit._recursive,needs triage,59,needs triage,1,0,97876,,,,0,0b158cf88a7d57aec8f290075eabb14a724bab7cda13b35a39e92df2e5c21e7e
thread:issue:132756,issue #132756: JITerator: Torch.prod produces RuntimeError: CUDA driver error: invalid argument,has linked PR,56,has linked PR,1,0,132756,,,,0,929251b9840e218bbc2dc1796ff46e91f388e9216e6469c0dc28dc981e156305
thread:issue:169688,issue #169688: [CPU] Stride 3 bfloat16 convolutions appear incorrect,has linked PR,56,has linked PR,1,0,169688,,,,0,04dc437c2fa6227f16f99354a07ff1b6fc93c009f5045fa799096d9ee82d7fe0
thread:issue:181650,"issue #181650: MPS: `add_dense_scalar_cast_float` shader fails with ""Read-only bytes bound to shader argument with write access enabled"" assertion during pyannote.audio inference (M4 Pro, torch 2.11)",has linked PR,56,has linked PR,1,0,181650,,,,0,263da1a545f1119fb7e689c27ee5a663852ed1df6b7d8ba6badc1f663e22a3f2
thread:issue:177833:pr:168223:pr:169107:pr:174025,pr #168223: Add failing test for negative epoch crash in LR schedulers (ZeroDivisionError),waiting on contributor,55,Wait for contributor update on PR #169107; keep related issue/PR context attached.,4,0,177833,"168223, 169107, 174025",,,3,4871a28013dd1403e964f49c7b91a88bb56b632a16b9cb1249fef7f7b9b4699a
thread:issue:162179:pr:162706,pr #162706: Fix accelerator-related skip-decorators for tests,PR blocked,55,Review active PRs and linked issues as one maintainer work thread.,2,0,162179,162706,,,2,f70bee5684375e7476e98cd213f8efcb7706c36234a3581e579ac70b952a1ceb
thread:issue:183795:pr:183487,pr #183487: Update XNNPACK submodule to 0b66c9f,PR blocked,55,Review active PRs and linked issues as one maintainer work thread.,2,0,183795,183487,,,1,923b9fc1ca2995d5fc8e828cd0a251e42ba07b3adef96068fbe2cfe9b19c0fa2
thread:pr:112441,pr #112441: [RFC] Tensordict integration,PR blocked,55,PR blocked,1,0,,112441,,,0,1282ed8c44fc4a8c0943d61fc3255615986448a8cbefec8f1e4688dbff6d8323
thread:pr:130556,pr #130556: Allow pipeline stage to take non-tensor args and kwargs,PR blocked,55,PR blocked,1,0,,130556,,,0,3d7adbc7b03acdd7a65d6fb5cb1eb92d64f991ed86db410fe77612b7201faeaf
thread:pr:140613,"pr #140613: Do not merge, do not close",PR blocked,55,PR blocked,1,0,,140613,,,0,7ec508d7623841026aa2a9f069003da9c90db69865c70c8fb516d5df34b37176
thread:pr:148668,pr #148668: Fix CUPTI lookup to include target directory,PR blocked,55,PR blocked,1,0,,148668,,,0,4f11aac4e6e7097d9a4de9f74f7d15ce126f96433562e9281ed3d03dd9164076
thread:pr:148900,pr #148900: [RFC][BE] assume error checking is on by default (#141914),PR blocked,55,PR blocked,1,0,,148900,,,0,21bd61cd73ca8a6f5963b40b7fa40aacfcfa7d75cae1afd3c7727f7f001047cc
thread:pr:157193,pr #157193: [c10d] Call destroy_process_group upon interpreter exit,PR blocked,55,PR blocked,1,0,,157193,,,0,bf529bd4c4422c5b187552331c98a58baf66bf736bc379e75c7a37fba73b9d75
thread:pr:160182,pr #160182: [WIP] [1/N] Introduce a generic CachingDeviceAllocatorImpl for cross backend use,PR blocked,55,PR blocked,1,0,,160182,,,0,e4008a74b6f9ee61e2e6aa3f3438dea74065966a159a1797ac3823f3d16be891
thread:pr:160638,pr #160638: [WIP] [2/N] Introduce a generic CachingDeviceAllocatorImpl for cross backend use,PR blocked,55,PR blocked,1,0,,160638,,,0,dc7871ce078c8f1a4841c11eb9161eb3fb246284b6cce3e019eeb1545ada4987
thread:pr:160872,pr #160872: [WIP] [3/N] Introduce a generic CachingDeviceAllocatorImpl for cross backend use,PR blocked,55,PR blocked,1,0,,160872,,,0,02b340b96a2126abc66d016441bb46fbf0147e87e1922a4ab3fb37e2305bb829
thread:pr:160878,pr #160878: [WIP] Introduce CachingDeviceAllocatorInterface as a base impl,PR blocked,55,PR blocked,1,0,,160878,,,0,7e882cbfc776d0754d39bc3ed1a046811499cfc7b970715946292e6d2cdd926b
thread:pr:160888,pr #160888: Allow branching on whether you are in AC region with is_checkpoint_enabled,PR blocked,55,PR blocked,1,0,,160888,,,0,fe4a121be120894659ed89b438c87555e05573e1a020c2183b31283b989045c7
thread:pr:161254,pr #161254: Add `zero_grad` param for `optimizer.step`,PR blocked,55,PR blocked,1,0,,161254,,,0,af0576cd41a962900cd6bd36d840972064203e9f8b196ac7ec85a5ceafdbc8b6
thread:pr:162033,pr #162033: [RFC] Switch all of c10d to use .h instead of .hpp,PR blocked,55,PR blocked,1,0,,162033,,,0,3e55f810e8930d21f0a5909fbef4633dc4c3d5dcb125ae205ea45876c760808d
thread:pr:164128,pr #164128: Consistently use c10_ovrsource in arvr mode everywhere,PR blocked,55,PR blocked,1,0,,164128,,,0,a2c0d1bd3e73d8d53ca640ce246d5499e79255284c4f952c6f3413f3196f5083
thread:pr:165800,pr #165800: [flex_attention]: adding a heuristic for USE_TMA,PR blocked,55,PR blocked,1,0,,165800,,,0,57c1bd38e9db46a92ba429be68d501b71689bec0e2aee9c885dc7912830a4f91
thread:pr:166732,pr #166732: Introducing statefulness for single process dataloader,PR blocked,55,PR blocked,1,0,,166732,,,0,721ca8cec6393c8dc248f641b2601e206781d81cd6208a568d240a81522c6a2e
thread:pr:167894,pr #167894: [inductor] split reduction even if all reads are broadcasted,PR blocked,55,PR blocked,1,0,,167894,,,0,5be7b7da821b626c01f16532490209d8345906653cbc32c571a077379601c5ee
thread:pr:170249,pr #170249: [do not review] Expose API to create ownership token for autograd.Function grad_fn,PR blocked,55,PR blocked,1,0,,170249,,,0,bcc8e61117e1223b7f1afd1d724c657483af6658f8f0e44b3282c2f6b5da42bd
thread:pr:170250,pr #170250: [do not review] Use new GradientEdge creation in PP,PR blocked,55,PR blocked,1,0,,170250,,,0,4ff1db40356168868e3ea85e4e9a1a63bdfa4dc05a35b3fe014b70f691837c4b
thread:pr:170807,pr #170807: Adding Glossary,PR blocked,55,PR blocked,1,0,,170807,,,0,6f837f63848aaa578b95344bcacb2c3d7b2a4bd18a2ef539591bd139f5b071d0
thread:pr:174893,pr #174893: [Inductor][NVGEMM] Enable Epilogue Fusions,PR blocked,55,PR blocked,1,0,,174893,,,0,f74ec7976cad9c38cb87f731826714d9371356e3f4b3530060b93fee4f764352
thread:pr:176857,pr #176857: [XPU] test_indexing test_reductions: Remove obsolete XPU skip guards after verification,PR blocked,55,PR blocked,1,0,,176857,,,0,d2fd51a3dc9877e7fc2f808fb69e83a810dcdd045c6e6bdd4750fa85ecf4d368
thread:pr:177726,pr #177726: Introduce test skip markers for Sandcastle,PR blocked,55,PR blocked,1,0,,177726,,,0,4ee8543ef76109c5792a89782c717de9f1a52f1d8fa1c8099fbdbb6c58e9102f
thread:pr:177867,pr #177867: Add ASIMD_BF16 Vectorized class specialisation,PR blocked,55,PR blocked,1,0,,177867,,,0,3670b2e51bd7a487e26270d5412a16be6ea70c107eee60be467bfd8668417228
thread:pr:178176,pr #178176: [DO NOT MERGE]Test New Landing Pg,PR blocked,55,PR blocked,1,0,,178176,,,0,b6df3d8c865db4423e0d1b8a9973549ee555af3d8b90711b855a469628997501
thread:pr:178245,pr #178245: Replace c10::call_once with static initialization,PR blocked,55,PR blocked,1,0,,178245,,,0,739740e82432f2cc771642e2b93b8ddd96e7e3f4cabe5557bffac5b3f2f0f24d
thread:pr:178360,pr #178360: Add device/backend agnostic support for distributed c10d/common tests,PR blocked,55,PR blocked,1,0,,178360,,,0,c0c35a39b565f34b3da7366428df4bb6bcb64d6c5abd1ad57860e2c6436d000f
thread:pr:179096,pr #179096: [Inductor] Add _scaled_mm_v2 lowering,PR blocked,55,PR blocked,1,0,,179096,,,0,9c94e5c6f922abdef0dd4e51067a98a11dcaee5ee44ea6d4193e0a43cb39155c
thread:pr:179139,pr #179139: Report ACL version info,PR blocked,55,PR blocked,1,0,,179139,,,0,e742f9aac55181a82718b14dcd2c139a3e4f6bee2e62ad0a7ece7e1d2b8d0bcf
thread:pr:179397,pr #179397: Modernize map usage in ATen,PR blocked,55,PR blocked,1,0,,179397,,,0,d30a9434147e255330eec6b01fb75389bb7f4036024ada1a2c42722510c844d1
thread:pr:179538,pr #179538: Fix incorrect type annotation in Future.set_exception (#179538),PR blocked,55,PR blocked,1,0,,179538,,,0,38dd142f512bfec1348c145f7456d8f64e11c91805b17a94460f576e619e43d5
thread:pr:179863,pr #179863: Teach CSAN to detect caching allocator memory reuse races,PR blocked,55,PR blocked,1,0,,179863,,,0,3cb8a03d3d7b8bb5fd516192bc5a865440820cbfd0889342ce8879d2e2589e79
thread:pr:180545,pr #180545: fix AOTInductor fails to compile dynamic-shape CUDA gather pattern with ValueError: The argument '((0)) + 48' is not comparable,PR blocked,55,PR blocked,1,0,,180545,,,0,fb83b30a12cf63fa846b6c43d9d6ef12d79a8423517ec28006e154f60fab5cee
thread:pr:181077,pr #181077: Enable fbgemm on MacOS,PR blocked,55,PR blocked,1,0,,181077,,,0,7946fb528db15c91e8244505f1d0f411731dc7b3d637ec04a542d04a440becb6
thread:pr:181281,pr #181281: [dynamo] Refactor IsInstanceBuiltinVariable,PR blocked,55,PR blocked,1,0,,181281,,,0,349bd8676c2668c61ccd6aca4edad5a1ab40ef84b495053d746bbe7d2f911ba5
thread:pr:181337,pr #181337: [dynamo] Refactor StrBuiltinVariable,PR blocked,55,PR blocked,1,0,,181337,,,0,fd0ed362c77e13a160ce9585498886a8dff302a7c8504e3c55e939ac5e485fbf
thread:pr:181338,pr #181338: [dynamo] Refactor TypeBuiltinVariable,PR blocked,55,PR blocked,1,0,,181338,,,0,a0b92058e7665f6e6fd0b3ff80017472b81f3b5e23b2a388d98632a0ee5e8a86
thread:pr:181359,pr #181359: S390x: fixes for docker image build,PR blocked,55,PR blocked,1,0,,181359,,,0,3ec2c7d993666534660de50a71cadf55683e1460b79e38c725998ec749e28f24
thread:pr:181435,pr #181435: Add operator benchmark monitoring skill for daily regression detection,PR blocked,55,PR blocked,1,0,,181435,,,0,120786af488556b8cb52ae7a6355011c603dc28f785fa36898cfec9d6f5c2582
thread:pr:181617,"pr #181617: Speed up inductor test infrastructure (~4x collection, ~1.7x execution)",PR blocked,55,PR blocked,1,0,,181617,,,0,117e0fb2c729ff98408f4d0b5b316d467ad2df4c66f1b9bc36875cd321f05e5f
thread:pr:181875,pr #181875: [ROCm][release/2.11] Fix test_transformers.py test suite for ROCm,PR blocked,55,PR blocked,1,0,,181875,,,0,1cbd866b4fb5694730b0d6ff12aa152260be50a99ddc0354718bd2205645ce41
thread:pr:181909,pr #181909: [Inductor][Triton] Forward hint_override in fused Triton compile,PR blocked,55,PR blocked,1,0,,181909,,,0,7ee43a3c1a2e0be6754366c3a6305964a88bd45743141abc79e9fad9c369ff6f
thread:pr:182179,pr #182179: [CI] Floor known-slow file times when stale per-config data under-shards,PR blocked,55,PR blocked,1,0,,182179,,,0,1fb6b73d1a9f0cabe3e18a4b5d84e4ff7d10571b31ebcc0a796829ccf6c8f98c
thread:pr:182196,pr #182196: [dynamo][dicts] Fix delitem for dicts,PR blocked,55,PR blocked,1,0,,182196,,,0,1e36aa506154eb317f871c70e969b29f1d37894cccdfe5af7283be260339331a
thread:pr:182274,pr #182274: [docs] Generate doc for values in ScalingType and SwizzleType enum,PR blocked,55,PR blocked,1,0,,182274,,,0,22384adf862a4b0d39b79007250bb80d406f02370957fad760033538116c3116
thread:pr:182398,"pr #182398: Revert ""[dynamo-wrapped] Skip all Dynamo wrapped tests for numpy (#182240)""",PR blocked,55,PR blocked,1,0,,182398,,,0,41de60834419e6b1c43c14cb629bf8865fa6be1af2eb88bf70fc0658d4f29b57
thread:pr:182598,pr #182598: [dynamo] Refactor builtins init static type check using _fn,PR blocked,55,PR blocked,1,0,,182598,,,0,f577b893057dcaffc9881f05b9c1f7d128149e3dc058687274af01975a8d7a3c
thread:pr:182730,pr #182730: [DTensor] Preserve placement for aten.clone.default,PR blocked,55,PR blocked,1,0,,182730,,,0,3b7c40cc37e3f258d8ce87a0e3f096e86c7c5296a6061d932bb2d69507e90c0e
thread:pr:182781,pr #182781: [1/N] Use const pointers and references,PR blocked,55,PR blocked,1,0,,182781,,,0,7e507cb1892bb67692ee6016d2d8b4de825e06f96b2e56258a19ebb3656a0e3e
thread:pr:182845,pr #182845: [AI Codemod][RlAutotuneFbsourceWearablesMemory] PackedLinearWeightsQnnp::apply_dynamic (qlinear_dynamic.cpp),PR blocked,55,PR blocked,1,0,,182845,,,0,4af5ebb3340a47bbb8f93918f4c551b96db59b51ee1f753774dd65159d4dfc7d
thread:pr:182846,pr #182846: fix: safely handle non-standard boolean values in the CUDA kernel,PR blocked,55,PR blocked,1,0,,182846,,,0,d9d02599c0e1e94082ff979eca3dbb81df5e958adad0b530070a8abd07603bd6
thread:pr:182900,pr #182900: test: relax tolerance of test_Embedding_discontiguous_cuda,PR blocked,55,PR blocked,1,0,,182900,,,0,ad0b18ee87970566d230c96e5e314808341d843e3c8f3744bf3ae617c4cdb5d4
thread:pr:183035,pr #183035: Fix arange int64 fractional-step zero handling on CPU and CUDA.,PR blocked,55,PR blocked,1,0,,183035,,,0,e5e4f5a638f91ee4cca75a88cbba0609e465427c92bfbdfa7f6abf19f6ed48c8
thread:pr:183086,pr #183086: Remove unused 'type: ignore' suppressions in dynamo/inductor/functorch,PR blocked,55,PR blocked,1,0,,183086,,,0,8bb305860cf03266b04cff1dc5cdb40208af87b3a2caec3bfcf913dfb4190e89
thread:pr:183105,pr #183105: Fix formatting in extensions.json,PR blocked,55,PR blocked,1,0,,183105,,,0,6a1a6baafc0050fda8c573b9bd15ed911dca0cb0639b5a98338e0776db375161
thread:pr:183254,pr #183254: Add RISC-V native fp16 conversion paths,PR blocked,55,PR blocked,1,0,,183254,,,0,393adf9fd49fb073a7b6cc690b6e052212ff305605da4dc593a3bc31354933ab
thread:pr:183262,pr #183262: [CUDA] Expose cudaHostGetDevicePointer in cudart bindings,PR blocked,55,PR blocked,1,0,,183262,,,0,c548141db754bc4d1df423c90f2b2ad547d244863cffc7ea5bce20694035150f
thread:pr:183384,pr #183384: [inductor] preserve scan carry strides with shape padding,PR blocked,55,PR blocked,1,0,,183384,,,0,eefaaa399dbaf1b6cf651bd516597091c76750de2bb66379213d83d59763f64b
thread:pr:183394,pr #183394: lint: ignore B904 for vendored code,PR blocked,55,PR blocked,1,0,,183394,,,0,fd1b669499ec241943f912d6f4d2eeea8d92d1a30415c35ba96522673608ee5f
thread:pr:183404,pr #183404: test/dynamo: add CPython coverage planning docs,PR blocked,55,PR blocked,1,0,,183404,,,0,45b87c43e905b55a2ede8a80a58683c946bbe609bf95590661e2777ba2475863
thread:pr:183418,pr #183418: [Avocado][MI350X] Non-overlay compatibility for new AITER/CK overlays (#183418),PR blocked,55,PR blocked,1,0,,183418,,,0,0439563a537caa7d28f703cb2021a1cf8721ce58899f8d57cfab3a028ac92009
thread:pr:183455,pr #183455: Port D104346887/PR 182675 for index large select (#183455),PR blocked,55,PR blocked,1,0,,183455,,,0,7b5c4f7c0e8436775afc7ab0b2d1c774d35c7d12ae9aac5b73234ad586a07484
thread:pr:183550,pr #183550: [AI Codemod][RlAutotuneFbsourceWearablesMemory] PackedLinearWeightsQnnp::apply_dynamic (qlinear_dynamic.cpp),PR blocked,55,PR blocked,1,0,,183550,,,0,69d95f77e3bcb00ccbe3ad8d2b5062fdeaae706b5ef6e67c8a19ced2f9955f72
thread:pr:183592,pr #183592: [ROCm] op benchmark: use noble nightly image,PR blocked,55,PR blocked,1,0,,183592,,,0,62d264b23270009fcf385d548c0ba830de83cf5771b706bfbd61958e8caeb8c0
thread:pr:183611,pr #183611: [inductor] Record output alias for mutates_and_returns_first_arg fallbacks,PR blocked,55,PR blocked,1,0,,183611,,,0,3d825e624000a0d9ba6d9455ac304024800961ee4925b59c0051748896507ddd
thread:pr:183651,pr #183651: [OSDC] Migrate h100-distributed.yml to OSDC (ARC) via dial-up pattern,PR blocked,55,PR blocked,1,0,,183651,,,0,9ea09f21e0292e1387c132cb061f53b4024d7ec63eed68118f05658fcbec8502
thread:pr:183747,pr #183747: Adding new test_c10d_ops_privateuse1.py variant for distributed testing,PR blocked,55,PR blocked,1,0,,183747,,,0,fa83da59bd93922d58096229b29f357ac067386502eb55749171cff995f08559
thread:pr:183940,pr #183940: [Native DSL] Vendor kernelagent-oink library,PR blocked,55,PR blocked,1,0,,183940,,,0,091788766852215e117c1c4b8500ed579f367569a9763b620700cb96729f60d6
thread:pr:183941,pr #183941: [Native DSL] Oink-based cuteDSL RMSNorm,PR blocked,55,PR blocked,1,0,,183941,,,0,4ec4436bf08c7cfe45ffc2066065d6a055e3ddf81c2a5b970da9ce318919bd76
thread:pr:183953,pr #183953: [DO NOT MERGE] Point CI workflows to test-infra IPv6 branch,PR blocked,55,PR blocked,1,0,,183953,,,0,8372557967ac8f1cb5894cc69c6f6f97bf1344bf27564c95033be15f501b302e
thread:pr:183978,"pr #183978: [docs] Add .. deprecated:: directives to torch.{cuda,cpu}.amp public APIs",PR blocked,55,PR blocked,1,0,,183978,,,0,6506c81c42f193b5c312fd4638e3f525e5769fbbc44454ff4eb4f19f8526c23d
thread:pr:183993,"pr #183993: Back out ""Port D104346887/PR 182675 for index large select""",PR blocked,55,PR blocked,1,0,,183993,,,0,72c4eb3aec576c3c711832d3aabe550fe3a8520508dbb0d68a15ebbd4a7c634c
thread:pr:184194,pr #184194: [inductor] Add simple_overlap pass for defensive comm/compute overlap,PR blocked,55,PR blocked,1,0,,184194,,,0,ffd49d39162dfdb29f0ec78920778facd2b12195ea2076f1cef111bd9fe6dc31
thread:pr:184236,pr #184236: [inductor][overlap] Enable simple_overlap by default,PR blocked,55,PR blocked,1,0,,184236,,,0,5e3819fb84d47715a958a90c71c7aa4232ecc797ef1c5894c622361ef689fb9b
thread:pr:184257,pr #184257: [scatter_add] Reject misaligned data_ptr / row stride in CuTeDSL override cond,PR blocked,55,PR blocked,1,0,,184257,,,0,4f3feb7b9a784f613739ca9b6c2641185ac54592c3bfcefe0e19b4f0b70c5bc1
thread:pr:184300,pr #184300: [xplat/caffe2] Add define_aten_lite macros for selective CUDA registration composed on top of pt_operator_registry,PR blocked,55,PR blocked,1,0,,184300,,,0,de4491a985d9c65bcdd7340c25c7a03b1cb91785209c01ffc09ec740ac0a2803
thread:pr:184307,pr #184307: Update SDPA error diagnostics,PR blocked,55,PR blocked,1,0,,184307,,,0,25d9c650cad42314c36407056677c2bb23a2d06c7bdcaf42e62b8edd09adf689
thread:pr:184389,pr #184389: [ROCm] Fix test_origami autotune comparison after PR #181617 cap,PR blocked,55,PR blocked,1,0,,184389,,,0,d7d50f72510cb9e8820283bd4d7a3a4b1aa94abf2b1e561a4798093f8457aec9
thread:pr:184401,pr #184401: Revert D105603961,PR blocked,55,PR blocked,1,0,,184401,,,0,a426c58dc3c6ab494ab1d5c25493063dec3e62949798240a579bdb93994f7622
thread:pr:184489,pr #184489: Forward CXX compiler to nvcc as host compiler for GCC too,PR blocked,55,PR blocked,1,0,,184489,,,0,ce4292a7c18093d56a33e498066b1752fd26c05f723436cc2660568fe9ccb398
thread:pr:184536,pr #184536: Skip `TestFullyShardSymmMem` for NCCL < 2.28,PR blocked,55,PR blocked,1,0,,184536,,,0,551c146ad3b870b30e853eda2f47998277e6660e7d09ebde3b982c379db02ad0
thread:pr:184616,pr #184616: Fix Dynamo list constructor iteration protocol,PR blocked,55,PR blocked,1,0,,184616,,,0,7059a128ac2f4c9cdea0d7f11f17ca83a25eb8c41f37668d36fde2305a4dc68e
thread:pr:184617,pr #184617: Fix Dynamo sequence repeat identity semantics,PR blocked,55,PR blocked,1,0,,184617,,,0,bfcbae050aca5c2d1f11233cfd2869d87c2d8b7e73c57b2541fff71a41e09c0d
thread:pr:184622,pr #184622: Fix Dynamo tuple constructor semantics,PR blocked,55,PR blocked,1,0,,184622,,,0,9f89edb6ce1058278aa5e6c7400b9a74c34c5c216c82d3fee463d2814104b4a9
thread:pr:184627,pr #184627: Fix Dynamo dict update semantics,PR blocked,55,PR blocked,1,0,,184627,,,0,c746a38d0d35d1d84c2698f395547efd4ed0399166bf6cd1840b708f18337720
thread:pr:184635,pr #184635: Fix Dynamo cmp_to_key list sort semantics,PR blocked,55,PR blocked,1,0,,184635,,,0,c0f3c5512731212184da5e82e81bd16ad11379cd5cd67afe9bdba8be98ad08f9
thread:pr:184645,pr #184645: Fix Dynamo defaultdict inplace union semantics,PR blocked,55,PR blocked,1,0,,184645,,,0,aa7c81c44a14e9faadff41ded6acc168377e3e71908ccc00521e30cdb3f43012
thread:pr:184669,pr #184669: Fix Dynamo frozenset copy identity semantics,PR blocked,55,PR blocked,1,0,,184669,,,0,5bd4a7642e61458b757cc89c0353cc4dc34f46c76809e94a7a01fc5ce107459b
thread:pr:184670,pr #184670: Fix Dynamo reversed iterator consumption,PR blocked,55,PR blocked,1,0,,184670,,,0,9e8280292b128bbde24a0d1dfd4ad04297358898c50c7f29e06d8b79e667dc72
thread:pr:184671,pr #184671: Fix Dynamo sequence search semantics,PR blocked,55,PR blocked,1,0,,184671,,,0,73a0f3b05ff5618385e96c5b36089398990c976b976f3c3d7c386aa96a65bb76
thread:pr:184672,pr #184672: Fix Dynamo dict equality semantics,PR blocked,55,PR blocked,1,0,,184672,,,0,9b28349dfac2baf385f6b444cc9918ebb47feb02927d8ec607123162322e5189
thread:pr:184673,pr #184673: Fix Dynamo dict fromkeys semantics,PR blocked,55,PR blocked,1,0,,184673,,,0,0182ac6459d9707850a96a9e87c2e29f3e4c6355fba9cc75c5dde623253b2027
thread:pr:184674,pr #184674: fixup! Add CPython Dynamo agentic loop plan,PR blocked,55,PR blocked,1,0,,184674,,,0,77093013cb328eef28e6037c793cdf3afba15983098fe034fae1e7d4d84e0fd7
thread:pr:184754,pr #184754: Fix Dynamo reversed iterator consumption,PR blocked,55,PR blocked,1,0,,184754,,,0,63efd7fb908578c534103bcbf93ea74e266dfb1e30a1a7408b200dd2767915e0
thread:pr:184755,pr #184755: Fix Dynamo sequence search semantics,PR blocked,55,PR blocked,1,0,,184755,,,0,a5b772fe9543429f0360ad431d84cdb2d0d8991ba6776bb18e681564ff0c48ae
thread:pr:184756,pr #184756: Fix Dynamo dict equality semantics,PR blocked,55,PR blocked,1,0,,184756,,,0,ae58fc22b34b2785702ed1db3c8e3deaf5892f2e1d4b8950a4bc3dd08511a7f5
thread:pr:184757,pr #184757: Fix Dynamo dict fromkeys semantics,PR blocked,55,PR blocked,1,0,,184757,,,0,ff35d77572ba8a9bbff8d81c74d9864c57b67273eaa1d58e963c41e9613bc8aa
thread:pr:184758,pr #184758: Fix Dynamo defaultdict shallow copy semantics,PR blocked,55,PR blocked,1,0,,184758,,,0,f366e2e76b2e5ecc9c25540b1ea1abdd0bd6fc0300712b6d970c21bf0e16b368
thread:pr:184759,pr #184759: Fix Dynamo frozenset dict-key hashing semantics,PR blocked,55,PR blocked,1,0,,184759,,,0,f35d2ea49c92406d600d78a28ea68242949e432db6fa468d2e8660d72beac887
thread:pr:184760,pr #184760: Fix Dynamo frozenset subclass dict-key hashing semantics,PR blocked,55,PR blocked,1,0,,184760,,,0,cd20e68b348826ce6b8306f1e110788de1c96194c1a5a30161121aa107e9dda9
thread:pr:184761,pr #184761: Fix Dynamo mutable set dict-key hashing semantics,PR blocked,55,PR blocked,1,0,,184761,,,0,4f3d39ecfca3dfb1067d52b3176f810077fa97af50337f8c8327b46c31041a17
thread:pr:184762,pr #184762: Fix Dynamo list extend iterator semantics,PR blocked,55,PR blocked,1,0,,184762,,,0,a898593c02f9a85b4a7e98b8a75f949ee5d5b998942e946d6e6d275b7396edbb
thread:pr:184763,pr #184763: Fix Dynamo dict copy GC tracking semantics,PR blocked,55,PR blocked,1,0,,184763,,,0,3b474a67f49ca64c430f915ad10b0ea53532b499c3b8783f2861fe076574df5e
thread:pr:184764,pr #184764: Fix Dynamo iterator reduce semantics,PR blocked,55,PR blocked,1,0,,184764,,,0,322f798f6540fd5ffb83e066eac0d2fd003f0adb2e8ef1bb982057a091ffc627
thread:pr:184765,pr #184765: Fix Dynamo ordered contains semantics,PR blocked,55,PR blocked,1,0,,184765,,,0,2b88a5df33459fdc730a93118332f272556f0af5a88499b83d18882b44d44a09
thread:pr:184766,pr #184766: Fix Dynamo set discard and remove semantics,PR blocked,55,PR blocked,1,0,,184766,,,0,49c5bf5234d0c9579640e8fc5e3f256213d07e6cc62591d74e31a11bbed4a87c
thread:pr:184801,pr #184801: Gate baddbmm expectedFailureDynamic on TEST_Z3,PR blocked,55,PR blocked,1,0,,184801,,,0,629996a942bb0fd17583fb93355facf68e66a37a38681717a7d338bdd945af06
thread:pr:184812,pr #184812: Improve Inductor index simplification,PR blocked,55,PR blocked,1,0,,184812,,,0,ac7970e53ad4caeb714d1526613fe8405a8bdc8520a929ab8af026ebf4a07926
thread:pr:184813,pr #184813: Use custom Min Max in Inductor symbolic indexing,PR blocked,55,PR blocked,1,0,,184813,,,0,e4ea3be8e0cbcbc9c553323a6488db0736c4acab55259b57adf410bb0d842717
thread:pr:185044,pr #185044: [AI SLOP] Support SymInt list custom op returns,PR blocked,55,PR blocked,1,0,,185044,,,0,64ed8cdee06cb1ca13f4a200226240a6d78f495f887bf3f2d4dd75f43d189c03
thread:pr:57772,pr #57772: Deprecate torch.svd and change svd -> linalg_svd,PR blocked,55,PR blocked,1,0,,57772,,,0,53963d2c5b1438612817aef5f292150e37ab4ed6af0eeb353c42ddfcc5ac0e70
thread:pr:88221,pr #88221: Add torch.tensor replacement and int_tensor prim,PR blocked,55,PR blocked,1,0,,88221,,,0,a86e17cdd1ee5a5ef012205284a326ecdca9d5182df4eb63e0bfef869f8bba44
thread:issue:169111:issue:169188:issue:188866:pr:184997,pr #184997: Fix TVM backend scalar tensor inputs,waiting on contributor,53,Wait for contributor update on PR #184997; keep related issue/PR context attached.,6,0,"169111, 169188, 188866","184997, 189010, 189125",,,12,3913d7b890c9b4aa111dacb7f0906a6fd895c36360ee0d3c13921cda17ab8eb0
thread:issue:188670:pr:169694:pr:183691:pr:184531,pr #169694: Increase precision for golden solution in transformer tests.,waiting on contributor,53,"Wait for contributor update on PR #169694, #183691, #184531; keep related issue/PR context attached.",5,0,188670,"169694, 183691, 184531, 187050",,,4,196814c4fac546475e4fc954b4c4e8aa311490a3a4a4036110c291b01dffc685
thread:issue:136446:issue:136586:issue:160057:pr:185347,pr #185347: Support torch.compile under user FakeTensorMode,waiting on contributor,53,Wait for contributor update on PR #185347; keep related issue/PR context attached.,4,0,"136446, 136586, 160057",185347,,,9,a991425cb8f7574e97c4440249e0587e384a962612df9e64dd80bfeadc1504d9
thread:issue:168861:issue:168863:issue:168868:pr:188006,pr #188006: [ROCm] Fix torch.pdist correctness on ROCm; re-enable test_pdist_norm_large,waiting on contributor,53,Wait for contributor update on PR #188006; keep related issue/PR context attached.,4,0,"168861, 168863, 168868",188006,,,5,fe7490c85c25f43aec3f01f75de7e7426eef16d040957fa87b2e6c6b727d60d2
thread:issue:129845:issue:133493:pr:186366,pr #186366: Support vmapped compiled autograd.Functions,waiting on contributor,53,Wait for contributor update on PR #186366; keep related issue/PR context attached.,3,0,"129845, 133493",186366,,,9,63513f37dd3b126d0866ebaeed72ce03231cf543ebb9e8b49df2c03f5285b6dc
thread:issue:139521:issue:139629:pr:184223,pr #184223: Run Inductor fallbacks below autograd,waiting on contributor,53,Wait for contributor update on PR #184223; keep related issue/PR context attached.,3,0,"139521, 139629",184223,,,2,c75ae4e61bf95e42c0a791c097c778198c406f18371f837db31d58af727a89b2
thread:issue:148682:issue:149982:pr:184121,pr #184121: Improve persistent tiled reduction defaults,waiting on contributor,53,Wait for contributor update on PR #184121; keep related issue/PR context attached.,3,0,"148682, 149982",184121,,,12,4bba903a518216e2ad5c0636b8fe68c2e54d6f759154065dd9dd243b97aa44a7
thread:issue:166009:issue:166014:pr:184016,pr #184016: Fix Inductor DCE for fallback alias mutations,waiting on contributor,53,Wait for contributor update on PR #184016; keep related issue/PR context attached.,3,0,"166009, 166014",184016,,,5,3f181fef7141a4eb6a6dbf6d76686e33d79fc325c2b548030204ee39f3920efa
thread:issue:167568:issue:182403:pr:185043,pr #185043: Functionalize mutations in deferred local_map,waiting on contributor,53,Wait for contributor update on PR #185043; keep related issue/PR context attached.,3,0,"167568, 182403",185043,,,2,2fe904cd6d32ad9bd8c42ee2bd2a074793ba9a37019575d26d7f0d5a3ac5ec3f
thread:issue:172026:issue:179510:pr:184958,pr #184958: Prevent escaped vjp TensorWrappers from reaching AOTAutograd,waiting on contributor,53,Wait for contributor update on PR #184958; keep related issue/PR context attached.,3,0,"172026, 179510",184958,,,6,6095c79a4621cc02dae0b761dfa457809e52ef1d54fc835306ad43dcd84c82b7
thread:issue:180156:issue:181807:pr:181895,pr #181895: [CUDA] Fix linspace integer dtype precision mismatch with CPU,waiting on contributor,53,Wait for contributor update on PR #181895; keep related issue/PR context attached.,3,0,"180156, 181807",181895,,,4,d58c5559ddf2c5ac8b4dcf54a490e0837a2274518173981b7a9bf069771926c6
thread:issue:181891:issue:182036:pr:184639,pr #184639: Support None returns from torch.cond,waiting on contributor,53,Wait for contributor update on PR #184639; keep related issue/PR context attached.,3,0,"181891, 182036",184639,,,4,7534eac2e5ca389714de2d8f53e519f5c467fadf556119af77693980bca9356a
thread:issue:186605:issue:186867:pr:189215,pr #189215: [MPS] Add `largeTensorTest` to `test_group_norm_backward_large_input`,waiting on contributor,53,Wait for contributor update on PR #189215; keep related issue/PR context attached.,3,0,"186605, 186867",189215,,,2,4dee666c34858b50fa0fe58d6a359b335e4d369f99b29c6bea1862f94edea100
thread:issue:113899:pr:186522,pr #186522: Trace Tensor.requires_grad_ in pre-dispatch make_fx,waiting on contributor,53,Wait for contributor update on PR #186522; keep related issue/PR context attached.,2,0,113899,186522,,,5,6fc654f6cfaa898abf10ba14bf69b258508cad6c6239e556e16281c6836adfab
thread:issue:116202:pr:186510,pr #186510: Fix Dynamo low-precision addc opmath,waiting on contributor,53,Wait for contributor update on PR #186510; keep related issue/PR context attached.,2,0,116202,186510,,,3,b5cba8bb85da9ca4a961a667cddd213ba7430e802605a6d8c1c1e5642e30d703
thread:issue:120381:pr:184380,pr #184380: Avoid fp32 materialization in checkpoint recompute,waiting on contributor,53,Wait for contributor update on PR #184380; keep related issue/PR context attached.,2,0,120381,184380,,,5,0e15d89bb5062d531fa3625f411d44461307367d6305bcc373e27990df7af932
thread:issue:125236:pr:184345,pr #184345: Use freezing compiler for traced inference graphs,waiting on contributor,53,Wait for contributor update on PR #184345; keep related issue/PR context attached.,2,0,125236,184345,,,18,d18fc4ddf8faaceb7bb6d5ae037305674be681b3ad7756d104de4448247e2161
thread:issue:126514:pr:186444,pr #186444: Fix Adam capturable zero-lr NaNs,waiting on contributor,53,Wait for contributor update on PR #186444; keep related issue/PR context attached.,2,0,126514,186444,,,1,835ad0a82b9f3778f0fbd21b65fcf61d90004b9ffa42a14204a034a46b4952a9
thread:issue:128063:pr:184319,pr #184319: [inductor] Align split reductions for tiled pointwise fusion,waiting on contributor,53,Wait for contributor update on PR #184319; keep related issue/PR context attached.,2,0,128063,184319,,,8,cf6e863fc946df7759a78019250531f620ea2efe5d67e76dafd62d6729acbfd1
thread:issue:128531:pr:186420,pr #186420: Run torchrec tests in Dynamo CI,waiting on contributor,53,Wait for contributor update on PR #186420; keep related issue/PR context attached.,2,0,128531,186420,,,5,13f3c28ca29bb92e35af1d6a47c9cced4bfb1d71e6e9adc5a7786309fbadaf46
thread:issue:129327:pr:186397,pr #186397: Keep constant-false torch._check in compiled graph,waiting on contributor,53,Wait for contributor update on PR #186397; keep related issue/PR context attached.,2,0,129327,186397,,,5,b39a8907dc4c1c5b6f29a93ede62c8bb4ab9bcc72498e22e10891fc51241f4bd
thread:issue:139756:pr:186160,pr #186160: Fix FakeTensor forward AD batch_norm meta rewrapping,waiting on contributor,53,Wait for contributor update on PR #186160; keep related issue/PR context attached.,2,0,139756,186160,,,5,314bb21b9db53b6c134ad6584a926af0766c89752ff18e7bac16dbc660783ae3
thread:issue:148651:pr:184130,pr #184130: Avoid torch imports in Inductor compile sidecar,waiting on contributor,53,Wait for contributor update on PR #184130; keep related issue/PR context attached.,2,0,148651,184130,,,1,65f0e77afb23006f7e9eda344fe2a1eb7b3f1dcd790b5abc97a3e40c4b571405
thread:issue:149565:pr:185764,pr #185764: Recommend nonstrict_trace for skipped C extensions,waiting on contributor,53,Wait for contributor update on PR #185764; keep related issue/PR context attached.,2,0,149565,185764,,,3,781b44cdfefed4a5f946d39d8297cd74d2c1b1963e87bfdc1bfd575f5eccd9b7
thread:issue:149767:pr:185759,pr #185759: Improve float32 flex attention backward tiling,waiting on contributor,53,Wait for contributor update on PR #185759; keep related issue/PR context attached.,2,0,149767,185759,,,6,c995b8a9f202bc3470b498d02b20ba13c35517adf434c98e5e94e22b44d8e294
thread:issue:149911:pr:185899,pr #185899: Fix Dynamo data mutation on functorch tensors,waiting on contributor,53,Wait for contributor update on PR #185899; keep related issue/PR context attached.,2,0,149911,185899,,,3,2ae80a744b1513b86b83cf3c0a6594e5f525730d4dde08517f0cdbe532d53a25
thread:issue:152346:pr:185690,pr #185690: Optimize AOT input view mutations,waiting on contributor,53,Wait for contributor update on PR #185690; keep related issue/PR context attached.,2,0,152346,185690,,,3,e6760bad01f0427d0cef67ac986b57271c2499f427dfc567b22bb0753b5b4c1b
thread:issue:153204:pr:185585,pr #185585: Fix mark_unbacked stride recompiles,waiting on contributor,53,Wait for contributor update on PR #185585; keep related issue/PR context attached.,2,0,153204,185585,,,5,3eafb9526726b112cc829a45372c4a2d10f2ea7f216bad4cb83eef5bf1b94722
thread:issue:154301:pr:184092,pr #184092: Skip tiny cudagraph partitions in max-autotune,waiting on contributor,53,Wait for contributor update on PR #184092; keep related issue/PR context attached.,2,0,154301,184092,,,5,620461a7a61e66bec49b630b0dccb03f6c5965875adb2d045d6c3823fb67a83c
thread:issue:154306:pr:184091,pr #184091: Handle cudagraph liveness changes during replay,waiting on contributor,53,Wait for contributor update on PR #184091; keep related issue/PR context attached.,2,0,154306,184091,,,3,28230b37a90cea4281a8dac8afd50269b77b7ffd5b2f919ba581324162bc8a8c
thread:issue:154851:pr:186897,pr #186897: Warn when custom op int schema specializes SymInts,waiting on contributor,53,Wait for contributor update on PR #186897; keep related issue/PR context attached.,2,0,154851,186897,,,5,b4b1226cfe9c29819fbfd6026bca20a94e9c901a7b6f67a1f4febbed4df6b4ec
thread:issue:159192:pr:184061,pr #184061: Fix fused moving avg observer functional meta,waiting on contributor,53,Wait for contributor update on PR #184061; keep related issue/PR context attached.,2,0,159192,184061,,,5,a2782432c60ad01781ba811266937e6dd83289a4bcd0613a0096bce856b2dc97
thread:issue:160388:pr:184052,pr #184052: Preserve CIA ops in AOT joint export,waiting on contributor,53,Wait for contributor update on PR #184052; keep related issue/PR context attached.,2,0,160388,184052,,,5,101c1293141769d57f7847e1490f2a27609734b7404a21df0b81608c1d28785f
thread:issue:160744:pr:189038,pr #189038: [MPS] Fix uint8 casting in MPS `copy_` kernels,waiting on contributor,53,Wait for contributor update on PR #189038; keep related issue/PR context attached.,2,0,160744,189038,,,3,83b4e83a76098ce3b2a0498d6db5ea8a7c7c40ade1dcd20b30dd28cd720f4e84
thread:issue:161067:pr:185299,pr #185299: Run CSE in AOT inference graphs,waiting on contributor,53,Wait for contributor update on PR #185299; keep related issue/PR context attached.,2,0,161067,185299,,,5,6c674911951f7b35b937e794a1b24a7df03afc577f149577e96ed3b480a42872
thread:issue:161289:pr:185274,pr #185274: Fix ci expected accuracy updates without ClickHouse credentials,waiting on contributor,53,Wait for contributor update on PR #185274; keep related issue/PR context attached.,2,0,161289,185274,,,3,e0f480238d4257043e09b46653c95ce30d25b782cd936e31ff142f5fd24f9b9f
thread:issue:161650:pr:185171,pr #185171: Document activation memory budget config,waiting on contributor,53,Wait for contributor update on PR #185171; keep related issue/PR context attached.,2,0,161650,185171,,,3,c1007244b93ae13213c37c0e20bc7e0d5ee88a93dee9c8aca31d18105bb5df6a
thread:issue:165547:pr:185075,pr #185075: Improve FakeTensorMode deepcopy error for real tensors,waiting on contributor,53,Wait for contributor update on PR #185075; keep related issue/PR context attached.,2,0,165547,185075,,,3,d74ffe782cc03d259f9a6f4a9a0811d1bcf8dc1e4542efe2a1406cca85031312
thread:issue:166516:pr:185064,pr #185064: Add unroll option to scan,waiting on contributor,53,Wait for contributor update on PR #185064; keep related issue/PR context attached.,2,0,166516,185064,,,6,50232912d5988fa05aaeb9a08c4643a81605ab9543276dfdb4887d81051695f4
thread:issue:167070:pr:184009,pr #184009: Handle SymPy boolean Inductor graph inputs,waiting on contributor,53,Wait for contributor update on PR #184009; keep related issue/PR context attached.,2,0,167070,184009,,,3,3b5c64b70d9bab73dfaebbed0498ede38227865e9282e5562a2ae77479a35aab
thread:issue:167719:pr:185035,pr #185035: Make symbolic formatting guard data-dependent scalars,waiting on contributor,53,Wait for contributor update on PR #185035; keep related issue/PR context attached.,2,0,167719,185035,,,5,81417123e9d22cac7c3cf7f6658a499295be137b09e16d1a5e21406d39e4e7ab
thread:issue:168136:pr:185017,pr #185017: Fix OpOverloadPacket tracing during initialization,waiting on contributor,53,Wait for contributor update on PR #185017; keep related issue/PR context attached.,2,0,168136,185017,,,3,df19b07f2108f9923dd6d61d97a32bd7f897fa8641f1e0e522f601012113a2fd
thread:issue:170648:pr:172074,pr #172074: Fixes grad_dtype property being lost in fsdp,waiting on contributor,53,Wait for contributor update on PR #172074; keep related issue/PR context attached.,2,0,170648,172074,,,2,3289091b899d32849386b247adfede8cc7edf1d4af6dad4ccb1179b5817db2d2
thread:issue:171905:pr:186764,pr #186764: fix: Use typing_extensions.TypeAliasType for better reexport of `__module__`,waiting on contributor,53,Wait for contributor update on PR #186764; keep related issue/PR context attached.,2,0,171905,186764,,,3,a1d4ceded87618b92864c10d63d13c0daea7c76470a2ac11633986ca888871f8
thread:issue:172869:pr:184944,pr #184944: Make OpaqueBase usable as a pybind base,waiting on contributor,53,Wait for contributor update on PR #184944; keep related issue/PR context attached.,2,0,172869,184944,,,3,c6ef29d29fa074f9144ef30a18d5834a4a7f5fdcde02627180c018d0f17b1ce9
thread:issue:172956:pr:184938,pr #184938: Fix AOTAutograd forward output overwritten by backward mutation,waiting on contributor,53,Wait for contributor update on PR #184938; keep related issue/PR context attached.,2,0,172956,184938,,,4,227e690a9b27fe042746440d7968f31de9731d8301a04dd31a4565d59fdd7642
thread:issue:174176:pr:184910,pr #184910: Reject direct aten.set in Dynamo,waiting on contributor,53,Wait for contributor update on PR #184910; keep related issue/PR context attached.,2,0,174176,184910,,,1,c21a3e6da577b118463fc2c09c42a1da724f356eff6509c1151618fc0afec3bb
thread:issue:174472:pr:183969,pr #183969: Fix CSE across unrelated mutation regions,waiting on contributor,53,Wait for contributor update on PR #183969; keep related issue/PR context attached.,2,0,174472,183969,,,1,c7d78be7e35d5025bb99202ea5f2dc7ac59a65fa66881a7ae4834bcdab0f26dd
thread:issue:175608:pr:182972,pr #182972: added graph break on exception based on if user code would catch it,waiting on contributor,53,Wait for contributor update on PR #182972; keep related issue/PR context attached.,2,0,175608,182972,,,3,956d868bcc7a5fb470a367cc1a0fdce7daf7b9743db578af614a1bcb116784b0
thread:issue:175968:pr:183954,pr #183954: Disallow mid-trace reference opaque creation,waiting on contributor,53,Wait for contributor update on PR #183954; keep related issue/PR context attached.,2,0,175968,183954,,,16,b095c8bfed015ee6bd90fde162f0bb95c6714bc010777567e74ee740b8b47185
thread:issue:176929:pr:183904,pr #183904: Fix CPU addcmul tensor-value decomposition numerics,waiting on contributor,53,Wait for contributor update on PR #183904; keep related issue/PR context attached.,2,0,176929,183904,,,1,2a020d005869d6f02b4c3315b5b21c60e467f80d60f86061c05c53716e8a9327
thread:issue:178096:pr:183871,pr #183871: [inductor] Match eval BatchNorm CUDA numerics,waiting on contributor,53,Wait for contributor update on PR #183871; keep related issue/PR context attached.,2,0,178096,183871,,,8,2d899bddc019fdec55f80fd3e1be3c73628d0d220f32b584356123160d494099
thread:issue:180833:pr:180353,pr #180353: [MPS] Replace isin MPSGraph implementation with native Metal kernels,waiting on contributor,53,Wait for contributor update on PR #180353; keep related issue/PR context attached.,2,0,180833,180353,,,2,5c0198e1cd26a7282d33caec1f02e9489640bb1969f2dc48442928dd098ac57f
thread:issue:182940:pr:184602,pr #184602: Preserve frozen ShapeEnv during fake tensor guards,waiting on contributor,53,Wait for contributor update on PR #184602; keep related issue/PR context attached.,2,0,182940,184602,,,3,909f9ce6d3d342833ea3704ed8f39d747e23dc558f21cad3c5d21942e841953e
thread:issue:183396:pr:184573,pr #184573: Fix compile support for CTC loss tensor lengths,waiting on contributor,53,Wait for contributor update on PR #184573; keep related issue/PR context attached.,2,0,183396,184573,,,5,6542914d602683b4966febdb0cf847b0bdb7217027bf178c3881a615f9db6273
thread:issue:183971:pr:184336,pr #184336: fix RPATH in libtorch extraction from wheel,waiting on contributor,53,Wait for contributor update on PR #184336; keep related issue/PR context attached.,2,0,183971,184336,,,3,88c74c9c21f1f5b59a695a309c5c164312baf71bb924676cf350279673c039f9
thread:issue:187155:pr:187161,pr #187161: [FSDP] Cache post-forward DeviceMesh to avoid duplicate NCCL communicators per layer,waiting on contributor,53,Wait for contributor update on PR #187161; keep related issue/PR context attached.,2,0,187155,187161,,,4,2720dc524266a76569916dd938b4aabed4065173b61c04d45d52eb1d036a8ec7
thread:issue:188723:pr:188994,pr #188994: [dynamic shapes] Preserve algebraic relationships in foreign unbacked SymInt transfer (#188723),waiting on contributor,53,Wait for contributor update on PR #188994; keep related issue/PR context attached.,2,0,188723,188994,,,3,16eaa38ffc5bdb4f913d3647e873ed79cee8324903c18316cc8338099cb99b30
thread:pr:172727,pr #172727: Fix torch.topk out variant recompilations by adding meta function for…,waiting on contributor,53,Wait for contributor update on PR #172727; keep related issue/PR context attached.,1,0,,172727,,,0,a2aac6c016744eed82e61c8f8fcaca974207b102c73406d7f027916289fccc4f
thread:pr:174718,pr #174718: [inductor] fix allocation with deterministic guard,waiting on contributor,53,Wait for contributor update on PR #174718; keep related issue/PR context attached.,1,0,,174718,,,0,b5fd5fcdda49c53da7445d2bbabc1793e3ad63766e0bf769f5108ea50574b918
thread:pr:174862,pr #174862: [FSDP2] Fix to differing gradient computation graphs in different ranks,waiting on contributor,53,Wait for contributor update on PR #174862; keep related issue/PR context attached.,1,0,,174862,,,0,6a311731c887fa21dab9e7e7a3dc9e9710d30ace8158ee9d3301d09a3cfcd9d7
thread:pr:175490,pr #175490: Fix CPU `NotImplementedError` for bf16/fp16 antialiased upsampling (`interpolate`),waiting on contributor,53,Wait for contributor update on PR #175490; keep related issue/PR context attached.,1,0,,175490,,,0,e54f57b79929553b4604b626e130b4aafc7b552b2e2667fff55cf156334b61f4
thread:pr:175964,pr #175964: Add validation for embedding_bag offsets to prevent illegal memory access,waiting on contributor,53,Wait for contributor update on PR #175964; keep related issue/PR context attached.,1,0,,175964,,,0,9cdc2d35a5467fa3c85d2e11ae23ec6a2163e0445880a9296aed5fd9befb159d
thread:pr:175969,pr #175969: [inductor][xpu] Fix convolution_backward accuracy for mixed channels-last / contiguous inputs,draft/noise,53,draft/noise,1,0,,175969,,,0,d38d49f74b3bd25dc536a4f9efd38cb2ecd9ebb767002cccc2577fc8c618e252
thread:pr:176072,pr #176072: Fix Triton launcher arg mismatch silent failure,waiting on contributor,53,Wait for contributor update on PR #176072; keep related issue/PR context attached.,1,0,,176072,,,0,512204c4c39e2994f48e60a54c2e8b4747a7e4a6c7705ab37d157fd914b2ccad
thread:pr:177457,pr #177457: [DTensor] Coalesce collectives for multi-arg redistribution,waiting on contributor,53,Wait for contributor update on PR #177457; keep related issue/PR context attached.,1,0,,177457,,,0,d792a749094f722d725dd3e16eaacf74593e8d39bfd90eb39a02dec7e43ce0f1
thread:pr:177823,pr #177823: [Inductor] Cap num_warps for non-1x1 Triton conv kernels to fix incorrect results,waiting on contributor,53,Wait for contributor update on PR #177823; keep related issue/PR context attached.,1,0,,177823,,,0,737d5f175bc7e6223b396261f67fca563d6c9869231e8842d1f5b4226c22a988
thread:pr:178493,pr #178493: [c10d][NCCL] avoid reusing scalable init store keys across communicat…,waiting on contributor,53,Wait for contributor update on PR #178493; keep related issue/PR context attached.,1,0,,178493,,,0,c1e3cd82507e5ab16a1b353c3f2f0d8f1193e19e01d26e86fab42b9ca2509d46
thread:pr:178591,pr #178591: fixes:division by zero check for inductor,waiting on contributor,53,Wait for contributor update on PR #178591; keep related issue/PR context attached.,1,0,,178591,,,0,66b5740e09086b2097c6953500870e2f57a924a2e8da059fd7085b9acff177ac
thread:pr:179830,pr #179830: Fix upsample_nearest3d stride mismatch under torch.compile (#179272),waiting on contributor,53,Wait for contributor update on PR #179830; keep related issue/PR context attached.,1,0,,179830,,,0,c7bf180e4cca9fc564b14ff20a82f53e0fc5eaba3d86dcc42a98524656f81b14
thread:pr:180675,pr #180675: Add test for autograd.Function fused fwd/bwd pattern (issue #180642),waiting on contributor,53,Wait for contributor update on PR #180675; keep related issue/PR context attached.,1,0,,180675,,,0,b24b4278d77bb77280f4afbf3c76f1d1d5cf7b714380d68e992a16c89117e2ea
thread:pr:180689,pr #180689: [ROCm] Handle profiler trace format differences in combo kernel tests,waiting on contributor,53,Wait for contributor update on PR #180689; keep related issue/PR context attached.,1,0,,180689,,,0,73fba7a9f9f779690ed95f546898f37ef108ccb8299aec28d88d013c49574762
thread:pr:181106,pr #181106: Fix torch.compile crash with inference_mode and view ops in forward (#180823),waiting on contributor,53,Wait for contributor update on PR #181106; keep related issue/PR context attached.,1,0,,181106,,,0,f8912c01987d9ae508629220babb92329badf44cff722188ce642e2cbdbc8024
thread:pr:181465,pr #181465: Fix torch.fx.Tracer.record_stack_traces regression (user frames filtered by internal PyTorch frames),waiting on contributor,53,Wait for contributor update on PR #181465; keep related issue/PR context attached.,1,0,,181465,,,0,153a7b8b5e49f2ee2475c5d1c7d0e558e138c4a2e00b4a091d6b47624b11babe
thread:pr:181526,pr #181526: Replace obsolete TORCHDYNAMO_DYNAMIC_SHAPES env var in torch.compile FAQ,waiting on contributor,53,Wait for contributor update on PR #181526; keep related issue/PR context attached.,1,0,,181526,,,0,212f0b10e81233f6aa3d9e569c3f16d56352662fac45b2e4504d50081a7d8972
thread:pr:181665,pr #181665: Fix UBSan in weight_norm CPU kernel for empty input (#181510),waiting on contributor,53,Wait for contributor update on PR #181665; keep related issue/PR context attached.,1,0,,181665,,,0,35bdf07ffe38b7a3d89ce1c69b1f259947cf267f4dac26186a90f71657486948
thread:pr:182276,pr #182276: [inductor] Unwrap ShapeAsConstantBuffer in tensor_constructor size,waiting on contributor,53,Wait for contributor update on PR #182276; keep related issue/PR context attached.,1,0,,182276,,,0,a3e55d0daef6beea3209a71e20693bbb534577fd2b9cb3de77486d9729b3dc7b
thread:pr:182618,pr #182618: [Profiler] Forward declare nlohmann::json in public header,waiting on contributor,53,Wait for contributor update on PR #182618; keep related issue/PR context attached.,1,0,,182618,,,0,766801e4060f39afcc32be5f204b945c78998e76a8f844a3bb21637b54fd853b
thread:pr:182794,pr #182794: [inductor][rocm] Fix AOTInductor autotune int64 Triton kernel test and enable FP8 platform-aware handling,waiting on contributor,53,Wait for contributor update on PR #182794; keep related issue/PR context attached.,1,0,,182794,,,0,130e6c88cb27bf115056642a8a0d65990487783fe5d0ec62ab6e431f08440f98
thread:pr:182980,pr #182980: Clear exception tracebacks to avoid reference cycles,waiting on contributor,53,Wait for contributor update on PR #182980; keep related issue/PR context attached.,1,0,,182980,,,0,7baa6f2c6f97c439b8c5b5366284b8281ef8c4c2bab56d1ac72be19935be38f4
thread:pr:183023,pr #183023: [Inductor] Filter triton configs by launch geometry in triton heuristics,waiting on contributor,53,Wait for contributor update on PR #183023; keep related issue/PR context attached.,1,0,,183023,,,0,b4339221e4c821142ba0d0402c7333891e977bb199af713173d8e36ad5daa77f
thread:pr:183029,pr #183029: [make_fx] Fix wrong 2nd-order gradients by decomposing detach to alias,draft/noise,53,draft/noise,1,0,,183029,,,0,f3eefc09c56e82a40d7557280d70b826dcf101c6efe6856e6292b00dd17ba4c6
thread:pr:183508,pr #183508: Chain-based dispatch for custom_op fast path,waiting on contributor,53,Wait for contributor update on PR #183508; keep related issue/PR context attached.,1,0,,183508,,,0,650dfb6b1e03308eb13344dafc317f8ec47fcbf0aec955c375124401dc581841
thread:pr:183535,pr #183535: [MPS] fix older gpu gen mac matmul failures,waiting on contributor,53,Wait for contributor update on PR #183535; keep related issue/PR context attached.,1,0,,183535,,,0,17a40e9295c7eda4464231e26d2032c16f6e0a580e80a819f3832be41f06c432
thread:pr:183944,pr #183944: Fix compiled autograd clear saved tensors semantics,waiting on contributor,53,Wait for contributor update on PR #183944; keep related issue/PR context attached.,1,0,,183944,,,1,e9dfbd137a2bca5a376add3f0f9fd91d316641cbda4b5b2dbe00623988719305
thread:pr:183980,pr #183980: [cpu][cuda] Extend native_group_norm_backward to handle grad from all outputs,waiting on contributor,53,Wait for contributor update on PR #183980; keep related issue/PR context attached.,1,0,,183980,,,0,c21566458621c64d98bf7251baad619f3b3b9d669ee1519be31fa8584ecefda9
thread:pr:184249,pr #184249: Fix extern autotune tensor kwargs,waiting on contributor,53,Wait for contributor update on PR #184249; keep related issue/PR context attached.,1,0,,184249,,,0,cf9cfae43f4b0ce8a1eef0a3aa213d694ddfd4b8ba1512c447bc8327cfd818f4
thread:pr:184303,pr #184303: Fix pattern matcher replacement graph cleanup safety,waiting on contributor,53,Wait for contributor update on PR #184303; keep related issue/PR context attached.,1,0,,184303,,,0,29077192d460edacd1c550252c8c6cd0cff920f0a2df08c99550eebe3fe3f0c5
thread:pr:184310,pr #184310: Fix flex attention shared memory config fallback,waiting on contributor,53,Wait for contributor update on PR #184310; keep related issue/PR context attached.,1,0,,184310,,,0,d24b6ce772acf85268041b0223973a99d6d9c062bca6267b35219c6924645e21
thread:pr:184312,pr #184312: Fix efficient zero tensor handling in torch.compile,waiting on contributor,53,Wait for contributor update on PR #184312; keep related issue/PR context attached.,1,0,,184312,,,0,0fd0a3b9cc65a733a062ce104c0f2625f7f01aef343048877577e956fb9ad243
thread:pr:184483,pr #184483: Make DTensor.local_map pytree-aware on in/out_placements,draft/noise,53,draft/noise,1,0,,184483,,,0,9201eca8d0fd4792373899257ecad7a081f571477328cd990efb3e4293962097
thread:pr:184680,pr #184680: Inductor Lowering of Trunc Div should use div_rn,waiting on contributor,53,Wait for contributor update on PR #184680; keep related issue/PR context attached.,1,0,,184680,,,0,0daf0cf3107ad3320ac5ebf3d55b3c4bc8f7e4e350b825e2b39db78f5451c854
thread:pr:185048,pr #185048: [Bugfix] fix `torch.library._register_fake` crashes when any unrelated `LazyLoader` module in `sys.modules` fails to import,waiting on contributor,53,Wait for contributor update on PR #185048; keep related issue/PR context attached.,1,0,,185048,,,0,ef61b374ea530dc502f131b85acddf11902f7af45f0265a7276b759c0457dc98
thread:pr:185157,pr #185157: [Inductor] Fix CPU bool argmin/argmax incorrect indices,waiting on contributor,53,Wait for contributor update on PR #185157; keep related issue/PR context attached.,1,0,,185157,,,0,5a4b0a5a207881146054f4167e97eb9d9dc0ac1d743eb333250b9e9c9734a23a
thread:pr:185163,pr #185163: [aot_autograd] Handle None in saved tensors for LSTM/RNN backward; enable allow_rnn by default,waiting on contributor,53,Wait for contributor update on PR #185163; keep related issue/PR context attached.,1,0,,185163,,,0,a20c2e52ca4952511ab7882184ae36f1206aa5164be4d69ea5baae54b5bad797
thread:pr:185256,pr #185256: Fix PendingUnbackedSymbolNotFound for tensor-derived slice bounds,waiting on contributor,53,Wait for contributor update on PR #185256; keep related issue/PR context attached.,1,0,,185256,,,0,5bf682293ba5e920994a54eab91a4124fbf6c7514d07b6a1abe302f43ad5e1ff
thread:pr:185263,pr #185263: Fix torch.compile normal tensor std validation,waiting on contributor,53,Wait for contributor update on PR #185263; keep related issue/PR context attached.,1,0,,185263,,,0,500549e075526831d9913bb5a67949682853cbb1e970c5686c75686a9cd0c802
thread:pr:185355,pr #185355: [pipelining] Fix backward hook accumulation on recv buffer activations,waiting on contributor,53,Wait for contributor update on PR #185355; keep related issue/PR context attached.,1,0,,185355,,,0,4979be9eeb9a7d9bfb32ae5e7a6291448bd28ab58790f1f32e027ea867c6e24b
thread:pr:185361,pr #185361: [pooling] Reject negative divisor_override in avg_pool2d/3d,waiting on contributor,53,Wait for contributor update on PR #185361; keep related issue/PR context attached.,1,0,,185361,,,0,54bedd3706060d9713370b8768c9496443336b9abf26ee84b484d6ed028e1889
thread:pr:185519,pr #185519: Fix ZeroTensor meta redispatch for nested hessian compile,waiting on contributor,53,Wait for contributor update on PR #185519; keep related issue/PR context attached.,1,0,,185519,,,0,0929d07fd5b031e8dd435908d2dd3f3be550f4c96f0732ddb2f5be6307e8024f
thread:pr:185645,pr #185645: Add batched small-matrix CUDA kernel for linalg.eig,waiting on contributor,53,Wait for contributor update on PR #185645; keep related issue/PR context attached.,1,0,,185645,,,0,7cf0d8372417983b56ff4820cb7f7aefd25b04b12b7191d3bddc3afa8ca445b3
thread:pr:185765,pr #185765: [Inductor] Fix correctness discrepancy in torch.minimum and torch.maximum with signed zeros (-0.0 vs 0.0),waiting on contributor,53,Wait for contributor update on PR #185765; keep related issue/PR context attached.,1,0,,185765,,,0,e14c30eb9fd5fee0253806ad91ce3ea16298a690c059c79139b4cd9e2b201aad
thread:pr:185831,pr #185831: Use minimum-norm subgradient for clamp/clamp_min/clamp_max at boundary,waiting on contributor,53,Wait for contributor update on PR #185831; keep related issue/PR context attached.,1,0,,185831,,,0,4aee2313dc50d2da24f0139718c05f0fa83c0134d5c315131e06cee77ccbf71e
thread:pr:185927,pr #185927: Fix clamp boundary subgradients,waiting on contributor,53,Wait for contributor update on PR #185927; keep related issue/PR context attached.,1,0,,185927,,,0,c2124c869e94548861ed386b91fdaca271e11ccd9a3bfe163132bdd276c7cd3e
thread:pr:185977,pr #185977: Fix index_add out-of-bounds parity between eager and inductor,waiting on contributor,53,Wait for contributor update on PR #185977; keep related issue/PR context attached.,1,0,,185977,,,0,491244b646e75aca11beac24829e6fca81c63fd3f2f6617716c51f5ae8269c50
thread:pr:186079,pr #186079: Add test cases for 0-sized tensor padding,waiting on contributor,53,Wait for contributor update on PR #186079; keep related issue/PR context attached.,1,0,,186079,,,0,4f31f6188d20bd5715e5d1bfa50c372dca904f3207b96cfa821f0433fa3e16c6
thread:pr:186445,pr #186445: [AOTI] Avoid serializing full storage for shared-storage constants,waiting on contributor,53,Wait for contributor update on PR #186445; keep related issue/PR context attached.,1,0,,186445,,,0,f2015afe6023c707521aab1f1ab09c1609ad6e7273f1c8b493b3e856da049ec9
thread:pr:186542,pr #186542: Support joint export with unused parameter grads,waiting on contributor,53,Wait for contributor update on PR #186542; keep related issue/PR context attached.,1,0,,186542,,,0,e2c9d988bac31d5aa633ab9b7379ba4d1f3429ac5c019d8906a6d7f3654a175e
thread:pr:186543,pr #186543: Fix Windows debug build issues,waiting on contributor,53,Wait for contributor update on PR #186543; keep related issue/PR context attached.,1,0,,186543,,,0,9932d0794a71be2ddba313a26972981415a854ae6926918ba9813f6c426dbb92
thread:pr:186579,pr #186579: Fix mse_loss/smooth_l1_loss allocating oversized storage for scalar output,waiting on contributor,53,Wait for contributor update on PR #186579; keep related issue/PR context attached.,1,0,,186579,,,0,cb77d68aa156e917ec6312b180b8cf9d74ea30fdf551fef6fc744a4854963a66
thread:pr:186766,pr #186766: fix: [BE] Deduplicate auto_functionalized and triton_kernel_wrapper_functi...,waiting on contributor,53,Wait for contributor update on PR #186766; keep related issue/PR context attached.,1,0,,186766,,,0,b974ec98812809cf179d98df28ba56bde1689106de4fb3f78190c8e426886f94
thread:pr:186776,pr #186776: Fix __qualname__ for OpOverload and OpOverloadPacket,waiting on contributor,53,Wait for contributor update on PR #186776; keep related issue/PR context attached.,1,0,,186776,,,0,0ff00f1b60e8a65200265c654f051800918f365309ff065cadbbe0ddaafade27
thread:pr:187106,pr #187106: [inductor] decompose broadcast-bias baddbmm into bmm + pointwise add,waiting on contributor,53,Wait for contributor update on PR #187106; keep related issue/PR context attached.,1,0,,187106,,,0,d11abc36840fef98b7f617d52015c12a85d7702000549c1ffa6bb080cad56af4
thread:pr:187154,pr #187154: Use TORCH_INSTALL_LIB_DIR instead of hardcoded path 'lib',waiting on contributor,53,Wait for contributor update on PR #187154; keep related issue/PR context attached.,1,0,,187154,,,0,67fa7e6d39c2c661694819aefdac9e9de17aa64c081ba17394dcb4e1c73b0407
thread:pr:187247,pr #187247: Type _add_docstr so type checkers see spectral ops as callable,waiting on contributor,53,Wait for contributor update on PR #187247; keep related issue/PR context attached.,1,0,,187247,,,0,24d1a49bc938dfe82fe95c558f2ecf81c7c82ae979d2db9d1db54f907796035a
thread:pr:187260,pr #187260: Fix static cudnn building for cudnn v9,waiting on contributor,53,Wait for contributor update on PR #187260; keep related issue/PR context attached.,1,0,,187260,,,0,c7c26e1d7428fbaee9b22135086cfaa13abf5d02722623b5105f3111e1de5b95
thread:pr:187316,pr #187316: Support save_on_cpu offloading under first-order torch.func transforms,waiting on contributor,53,Wait for contributor update on PR #187316; keep related issue/PR context attached.,1,0,,187316,,,0,d78c41f8f642fc184569dfd3072b89e9943714974585d2defd6ea9ca96e25b50
thread:pr:187323,pr #187323: Fix Tensor.to() ignoring memory_format for non-contiguous tensors,waiting on contributor,53,Wait for contributor update on PR #187323; keep related issue/PR context attached.,1,0,,187323,,,0,31e9084ce81411e9520b2c24c5a8a16b1ab8f53fa3f1d05ff5e7b27b0fce2abc
thread:pr:187477,pr #187477: Add context manager support to torch.dtype,waiting on contributor,53,Wait for contributor update on PR #187477; keep related issue/PR context attached.,1,0,,187477,,,0,30f0a26bbadf22e243da5b50ce1c570a0b80d745d1a669a85f971de1ad714a3f
thread:pr:187558,pr #187558: Fix torch.sign to propagate NaN instead of returning 0,waiting on contributor,53,Wait for contributor update on PR #187558; keep related issue/PR context attached.,1,0,,187558,,,0,941ac2d83ccc58f957478cddb96a3132360c43de7869a51d4f036325044ce1b3
thread:pr:187572,pr #187572: [Metal][ops] add EmbeddingBag offsets validation (#170370),waiting on contributor,53,Wait for contributor update on PR #187572; keep related issue/PR context attached.,1,0,,187572,,,0,29dacac73a50c46226e8ce39788ad62d15ed7a6d90fbf441b62c77d21dce7302
thread:pr:187663,pr #187663: Avoid oversized storage for reduced mse_loss / smooth_l1_loss,waiting on contributor,53,Wait for contributor update on PR #187663; keep related issue/PR context attached.,1,0,,187663,,,0,890ce40b1d7baaebaf94e895321c2b521bbf9fa013f4d021378aa9b213323250
thread:pr:188092,pr #188092: Fix cartesian_prod returning inconsistent shape for single tensor input,waiting on contributor,53,Wait for contributor update on PR #188092; keep related issue/PR context attached.,1,0,,188092,,,0,4d0490d08a5ff9133d6773c543d91e3ed4b5379aaeaf876e0e257b293b37194a
thread:pr:188249,pr #188249: [inductor] Fix bool equality value range analysis,waiting on contributor,53,Wait for contributor update on PR #188249; keep related issue/PR context attached.,1,0,,188249,,,0,05d240bd984e062ddedf019a96ca46920215383cb4f6838fe065155121dae369
thread:pr:188297,"pr #188297: Validate target range [0, 1] in binary_cross_entropy_with_logits",waiting on contributor,53,Wait for contributor update on PR #188297; keep related issue/PR context attached.,1,0,,188297,,,0,395be681666c1f27a42222cd17d5057ccd8e2eb82e44c92c93797e58e67dc091
thread:pr:188473,pr #188473: [MPS] Add complex64 support to linalg.lu_factor (unblocks complex det/slogdet/lu),waiting on contributor,53,Wait for contributor update on PR #188473; keep related issue/PR context attached.,1,0,,188473,,,0,77a0bdda51168fab3a5bd79996d81eb7e47dbe9105265faaf26127f2ae5eadfe
thread:pr:188748,pr #188748: [MPS] Make F.linear honor PYTORCH_MPS_PREFER_METAL,waiting on contributor,53,Wait for contributor update on PR #188748; keep related issue/PR context attached.,1,0,,188748,,,0,d6b542dbfdb6ee595a03084d52696aa6372c263f508b07db8e9b9c15945298c1
thread:pr:188952,pr #188952: Fix C2872 'std' ambiguous symbol when nvcc compiles compiled_autograd.h on Windows extension builds,waiting on contributor,53,Wait for contributor update on PR #188952; keep related issue/PR context attached.,1,0,,188952,,,0,bbff38a7a55625efd56795f9b93e1da197c0d6246c8108d12ab607f4e62b53eb
thread:pr:189019,pr #189019: [MPS] Add torch.mps.MetalGraph capture/replay API,waiting on contributor,53,Wait for contributor update on PR #189019; keep related issue/PR context attached.,1,0,,189019,,,0,41c2650c1f8a0c1be6cd0181b3fef07c663b7f829705e1fff65d6c94311f6ff1
thread:pr:189132,pr #189132: Register sspaddmm.out for XPU backend,waiting on contributor,53,Wait for contributor update on PR #189132; keep related issue/PR context attached.,1,0,,189132,,,0,418317f802dce42aa4cf8688c8e0f7d714f882e09717f71a6d1692c25f012b13
thread:issue:15386:issue:19826:issue:21457:issue:24470,issue #15386: F.grid_sample doesn't respect padding_mode when height of inputs is 1,stale/low urgency,52,stale/low urgency,25,0,"2732, 4085, 5565, 15386, 19826, 21457, 24470, 24823, 24870, 25014, 25039, 30565, 36107, 36108, 48389, 50334, 50335, 50336, 50337, 50338, 50339, 50340, 50341, 61528, 81868",,,,40,7854c406dede935770cd7958df42cbb74fd043e200e49f9a7cc157e14d938fad
thread:issue:102852:issue:111173:issue:113062:issue:130430,"pr #184779: [MPS] Rewrite _bincount_mps with native Metal atomic-add (~18 s → 2.1 ms, 8,500× speedup)",waiting on contributor,52,"Wait for contributor update on PR #184779, #186540; keep related issue/PR context attached.",23,0,"77764, 83264, 83968, 84489, 97310, 98169, 102852, 111173, 113062, 130430, 136692, 140570, 140718, 141287, 149261, 149325, 154052, 157648, 178492","184779, 184780, 186540, 188816",,,31,8cc911a3653ae1b8943000f072b438266ea73e58669e1093cdcc7b0177c4ac6d
thread:issue:106959:issue:154613:issue:18095:issue:31458,"issue #106959: `tensor.repeat` quirks: has no `torch.` variant, no `out=` variant, no inplace variant | `torch.tile` also does not have `out=` variant and uses `dims=` instead of `dim=`",stale/low urgency,52,stale/low urgency,8,0,"18095, 31458, 39007, 61585, 68616, 71210, 106959, 154613",,,,7,862cf9a8fbedd48e5415b5b35244f3266229185e1664b7d13e92fcf99515e90a
thread:issue:118507:issue:134323:issue:142392:issue:70073,issue #118507: RuntimeError in torch.istft with center=False: Window Overlap Add Issue,stale/low urgency,52,stale/low urgency,7,0,"70073, 81428, 91309, 94718, 118507, 134323, 142392",,,,6,df4d2e83e93d86511cad060a36bd546b845d1e65877406d9481d63d6ce5d7a2d
thread:issue:113713:issue:58414:issue:58858:issue:58859,issue #113713: [cuDNN][cuDNN V8 API] cuDNN Flash-Attention Upstreaming RFC/tracking issue,stale/low urgency,52,stale/low urgency,6,0,"58414, 58858, 58859, 58860, 58862, 113713",,,,5,8cad541321db06a8bce45be6f12b53306abef00d5d1bbfa57df8655a48270ddc
thread:issue:134570:issue:14095:issue:30443:issue:69519,issue #134570: histogramdd 'Cuda' backend,stale/low urgency,52,stale/low urgency,5,0,"14095, 30443, 69519, 99719, 134570",,,,4,94e034f876d970300addac4320e42b161e069d18f64e3bbfdfdecdbc5f0e0412
thread:issue:155992:issue:171412:issue:68320:issue:91965,issue #155992: [feature request] Native checkpointing to/from `s3://` (DCP / `torch.load` / `torch.save`),stale/low urgency,52,stale/low urgency,5,0,"68320, 91965, 96412, 155992, 171412",,,,6,d00379e6a804d00ad64e98abcc9200584232445ebec59cb30f950dd97ce2d4c4
thread:issue:41950:issue:52675:issue:62931:issue:92250,issue #41950: [docs] Clarify behavior of torch.cuda.device_count() when torch compiled without CUDA (cpu-only) or when CUDA is not available,stale/low urgency,52,stale/low urgency,5,0,"41950, 52675, 62931, 92250, 99982",,,,5,e727ae9ed5d2c9c861737113194cb2852ef6bc3791ae5e5f7ce650dcbb35bf74
thread:issue:59552:issue:63486:issue:64688:issue:68866,issue #59552: test_batch_isend_irecv_nccl fails with NCCL 2.8.3,needs triage,52,needs triage,5,0,"59552, 63486, 64688, 68866, 73960",,,,4,ec9cdaea2ae20e8e072ab0339b4f1cdb1251ce403188536eba87618137fb9c78
thread:issue:102517:issue:105943:issue:10950:issue:23490,issue #102517: [feature request] PyTorch support for sub-interpreters with PEP 684 accepted and release in Python 3.12,stale/low urgency,52,stale/low urgency,4,0,"10950, 23490, 102517, 105943",,,,4,ae2708de64d4ce0262af4eff7e555b264f4b5b09598d50f281ab7bbe04b2cc44
thread:issue:10615:issue:108858:issue:44380:issue:48559,issue #10615: Request for better memory management,stale/low urgency,52,stale/low urgency,4,0,"10615, 44380, 48559, 108858",,,,3,cb51465500e2bef3f91edefd47b7c46e62a1856f641b4e2ca6eda7256cf59217
thread:issue:132020:issue:41226:issue:62027:issue:86558,issue #132020: `torch.Tensor.to` ignores `memory_format` kwarg,has linked PR,52,has linked PR,4,0,"41226, 62027, 86558, 132020",,,,3,b6b10e8b422cabb295f2c62264cafaa5474748c09178219f54b34ff3fbfa0540
thread:issue:23940:issue:47953:issue:53879:issue:55122,issue #23940: Improve the performance of linear algebra operations in CUDA for small problem sizes,stale/low urgency,52,stale/low urgency,4,0,"23940, 47953, 53879, 55122",,,,4,06cedae86b4c5c3f9d827fcabd0bbacb2dc4bd8f75bce94f29460bfe1468b572
thread:issue:34058:issue:90608:issue:93081:issue:93955,issue #34058: libtorch.so file size is very large,high-priority/blocker,52,high-priority/blocker,4,0,"34058, 90608, 93081, 93955",,,,3,04ce38ef3ee376d341b10c2e391c0ba602bffa1f6c862836caab58318c6cc5a1
thread:issue:43567:issue:52633:issue:71911:issue:72106,issue #43567: `torch.svd()` CUDA gives incorrect results when input contains `nan`,stale/low urgency,52,stale/low urgency,4,0,"43567, 52633, 71911, 72106",,,,3,0b1f83524b224f1766bc129b91ca72bb42cd4954b06fe086d4ef3461998fa55f
thread:issue:44282:issue:91989:issue:98836:issue:99625,issue #44282: PyTorch wheel's own OpenMP library clashing with system-wide OpenMP library at runtime,needs maintainer decision,52,needs maintainer decision,4,0,"44282, 91989, 98836, 99625",,,,3,7b80226e420de8a87152d2e3cf59f75003fdfe1744ffd11d00110fc348bd91e5
thread:issue:102911:issue:102971:issue:94691,issue #102911: PackedSequences on MPS accelerator yields `grad_y` missing or crashes the kernel.,high-priority/blocker,52,high-priority/blocker,3,0,"94691, 102911, 102971",,,,2,21c3cb3b5cfc29306dbe567eba4cde7b56505b65665e6832842f4faef990b71a
thread:issue:126523:issue:144563:issue:153938,issue #126523: Summary of tests in machine readable format,stale/low urgency,52,stale/low urgency,3,0,"126523, 144563, 153938",,,,2,9486c89630af4a1d1a75ef053d426a8f06e0639818c56e3e9ded49bdfdca1b7c
thread:issue:21018:issue:50804:issue:75912,issue #21018: free(): invalid pointer Aborted (core dumped),needs reproduction,52,needs reproduction,3,0,"21018, 50804, 75912",,,,2,c2ff943ad0eece75e3c8ded60ad3d1393a5a9b9cf0f1d2122f3d873cf88cc209
thread:issue:35600:issue:86537:issue:96726,issue #35600: Increased memory usage in repetitive torch.jit.trace calls,high-priority/blocker,52,high-priority/blocker,3,0,"35600, 86537, 96726",,,,3,592b55f33ed9588fabb87282b18d907d3328887de516e8991873084e006d10d2
thread:issue:50012:issue:5212:issue:68174,issue #50012: torch.split is divergent from np.split,high-priority/blocker,52,high-priority/blocker,3,0,"5212, 50012, 68174",,,,2,7d98b05b095301de6957a72ff988b2be6df0724059b53a499374f5f53e8186a9
thread:issue:64812:issue:68349:issue:70350,issue #64812: PyTorch inference on tensors of a particular size cause Illegal Instruction (core dumped) on Jetson Nano,stale/low urgency,52,stale/low urgency,3,0,"64812, 68349, 70350",,,,3,ae9e1d951830710d5704af86835132f5dd3f5141e2ac82b5eac7ac8939e1281e
thread:issue:80017:issue:80104:issue:85607,issue #80017: RPC init fails and crashes when world_size is greater than 18,stale/low urgency,52,stale/low urgency,3,0,"80017, 80104, 85607",,,,3,a98392ee3e2882dedd81ed781087c1eb67fe2d9a8e4a7009ff4feddcf93fe641
thread:issue:102533:issue:103756,issue #102533: Segfault when running vulkan program linked against libtorch,stale/low urgency,52,stale/low urgency,2,0,"102533, 103756",,,,1,8b1e9a138fe4b0a547794653acdef9a83e76c70af863aafaba9388a085d9b413
thread:issue:102963:issue:166004,issue #102963: torch.svd fails on large matrices,stale/low urgency,52,stale/low urgency,2,0,"102963, 166004",,,,1,4a72f9c6b88154bfbdd1920310f566571c9c35339fac43b8788b2763f3f216ed
thread:issue:105494:issue:124423,issue #105494: workaround for using vmap when .item() is being used internally,stale/low urgency,52,stale/low urgency,2,0,"105494, 124423",,,,1,b5f8b4c83a18f3b92aeaa763de4439f7e2aaf86a8610c5d13f6fc7f400f70f69
thread:issue:109923:issue:2575,issue #109923: Import order issue with torch and pybind11 Library Statically Linked to libstdc++,stale/low urgency,52,stale/low urgency,2,0,"2575, 109923",,,,1,98778fc6b34afc9febb0aa6248770d599fec06477de849b5593144f05f36ade0
thread:issue:111693:issue:112043,"issue #111693: [export] 14k models: AssertionError: graph-captured input # 2, of type <class 'torch.nn.parameter.Parameter'>, is not among original inputs of types",needs reproduction,52,needs reproduction,2,0,"111693, 112043",,,,1,8851851d7fcd7c29825a3d8b700c935d7686ee0d81450227a53976e8e7a73002
thread:issue:115347:issue:124291,issue #115347: [Functorch] vmap over index_select expands the output,high-priority/blocker,52,high-priority/blocker,2,0,"115347, 124291",,,,1,c5fbe8784813a485eeb820f64ec2b0910dec91587348264f3fc606b34d4a82e6
thread:issue:129118:issue:141831,issue #129118: interpolate nearest get values zero when outputs over 4G elements,stale/low urgency,52,stale/low urgency,2,0,"129118, 141831",,,,1,3c89674df08160aba0eea43e35292da28f28d9b9458d095f65597183dcf02c9e
thread:issue:13402:issue:33867,issue #13402: batch_norm doesn't bump version counter of running stats,high-priority/blocker,52,high-priority/blocker,2,0,"13402, 33867",,,,1,cafabc4270af7d161432c3ef534598e3031ab2fbd1633bff163ddc43ae8d6b22
thread:issue:135437:issue:82627,"issue #135437: Cannot build libtorch from source with NVHPC compilers - ""Could NOT find Threads""",stale/low urgency,52,stale/low urgency,2,0,"82627, 135437",,,,1,2711716534915acd5ac76579de061ff0ce5d4d72c07b305b4238a848dbb6ddc9
thread:issue:13918:issue:38703,issue #13918: Pytorch very slow to convert list of numpy arrays into tensors,high-priority/blocker,52,high-priority/blocker,2,0,"13918, 38703",,,,1,3da71f7ca8279c400464e7f30aef694bd0a8d04dcc3c4564a1905983a70b9c52
thread:issue:139298:issue:141133,issue #139298: CUDNN sdp attention causes loss explosion,high-priority/blocker,52,high-priority/blocker,2,0,"139298, 141133",,,,1,44223c5bce1033fb48f6fec723db0fe6ea25983c388cc99dae85ffb8d513dbd2
thread:issue:141218:issue:141219,issue #141218: Floating point exception (core dumped) in `scaled_dot_product_attention`,high-priority/blocker,52,high-priority/blocker,2,0,"141218, 141219",,,,1,4ce9e809bb6174f9905d4ac16b6ef30817c033dbd87065af641ae0646670511f
thread:issue:146828:issue:32351,"issue #146828: On Linux, passing torch.Generator to multiprocessing.Process crashes for forkserver and spawn start method",high-priority/blocker,52,high-priority/blocker,2,0,"32351, 146828",,,,1,db0341d21bd7925a48b6f2209a2e459cbdb511016bc6bdb81d0fae510a3d55db
thread:issue:15849:issue:39443,issue #15849: DataLoader with option to re-use worker processes,high-priority/blocker,52,high-priority/blocker,2,0,"15849, 39443",,,,1,a474ce86e5422c107be4b3af8b6b62f9d8a782e63ff9f9674ff52c03192db657
thread:issue:164402:issue:164787,issue #164402: NCCL-2.28.3 build locally is unusable on H100,high-priority/blocker,52,high-priority/blocker,2,0,"164402, 164787",,,,1,134e9bbdf191fb505370077c5871b71db051e3dca76844a2d6b9d908346a3a89
thread:issue:16897:issue:82761,issue #16897: Implement `numpy.random.choice` equivalent,high-priority/blocker,52,high-priority/blocker,2,0,"16897, 82761",,,,1,b5d46ad71c659f551411f81002dd84bd48a1e9648615be6ce0f05ef0c9cc30f2
thread:issue:17199:issue:58962,issue #17199: Deadlock with multiprocessing (using fork) and OpenMP / PyTorch should warn after OMP and fork that multithreading may be broken,high-priority/blocker,52,high-priority/blocker,2,0,"17199, 58962",,,,1,8b82574bf8cd1394ab725a7695fb1e4d13e1ef92a93d1d77561313940315f16f
thread:issue:2576:issue:43115,issue #2576: CUDA multinomial is limited to 2^24 categories,high-priority/blocker,52,high-priority/blocker,2,0,"2576, 43115",,,,1,45c44d68e40d552dd6bbcc21dc1b554f492c9305933012cd4ff2a1d7d009fefb
thread:issue:32918:issue:34544,issue #32918: Interpolate in the “bicubic” mode with the same shape outputs zeros from second sample onwards,high-priority/blocker,52,high-priority/blocker,2,0,"32918, 34544",,,,1,38be943a742364a6efc084a3f7257a08c6f5b1d2fe4d4c9c3dc6726b093aa16c
thread:issue:34257:issue:85538,issue #34257: CustomFromMask pruning stores a copy of the user-provided mask,stale/low urgency,52,stale/low urgency,2,0,"34257, 85538",,,,1,6fe92982c9fcafc103eb598873fd469c1063e2a70202bf5e56a455c224364a14
thread:issue:40137:issue:56297,issue #40137: Improve RPC test debugability,high-priority/blocker,52,high-priority/blocker,2,0,"40137, 56297",,,,1,baedad3483d596e271c03690b4c0b0793b020999b9d2b3f7e30caa4a5538a61f
thread:issue:48945:issue:49459,issue #48945: CUDA error: illegal memory access Conv3d,stale/low urgency,52,stale/low urgency,2,0,"48945, 49459",,,,1,71062f20e910e56aac877b7494cad0721cc06be450c6ef7a558bdecf53530dd1
thread:issue:49758:issue:49909,issue #49758: c10::scalar_to_tensor(...) uses should be audited for performance and type promotion impact,high-priority/blocker,52,high-priority/blocker,2,0,"49758, 49909",,,,1,a987a377ea97bdc77146784238ff2105ba9d64bfbd6bcd1446342d2d6a05d0d4
thread:issue:51872:issue:80337,issue #51872: Tensor.nonzero tries to allocate huge amount of memory for tensors on GPU with num_elements close to INT_MAX,stale/low urgency,52,stale/low urgency,2,0,"51872, 80337",,,,1,d90f900c777e3403a6333db08767fc57c20b19d3ae8be5e34483f41e644e05af
thread:issue:55375:issue:57534,issue #55375: `DataParallel` (`broadcast_coalesced`) with complex tensors yield real views,high-priority/blocker,52,high-priority/blocker,2,0,"55375, 57534",,,,2,ea817eb4a3dd3ce0c289dce0a281365db1df466b6e73b0639f48b3d83351d315
thread:issue:67448:issue:67712,issue #67448: Deadlock in test_multiprocessing_spawn.py,high-priority/blocker,52,high-priority/blocker,2,0,"67448, 67712",,,,1,8e9e4519c35bfb8c5d9d5b7cb7b9e52b45e303f6cf96dd2358629148201fd357
thread:issue:69801:issue:69960,"issue #69801: Dropout2d doesn't drop channels for (C, H, W)",high-priority/blocker,52,high-priority/blocker,2,0,"69801, 69960",,,,1,e71491804b190cc45a3a640e4538c2cc216ff00b40ddf9eb356e3628925f69c0
thread:issue:81541:issue:82072,issue #81541: [vulkan]compiling VulkanOpContext.cpp with some errors,stale/low urgency,52,stale/low urgency,2,0,"81541, 82072",,,,1,d9aea55df9b52b1150c9139d9ee7ec1e9849dc9b67ec840171ef7dfc4fc3e453
thread:issue:94668:issue:94808,issue #94668: A segment fault can be triggered in torch.lstm with edge cases,high-priority/blocker,52,high-priority/blocker,2,0,"94668, 94808",,,,1,c227956065f0d1b0e30591f0103a9ae3731dbed5c1d77dba0fcdb7387c1313b0
thread:issue:96275:issue:96276,issue #96275: A Segment Fault can be triggered in torch.pinverse,high-priority/blocker,52,high-priority/blocker,2,0,"96275, 96276",,,,1,1e33ba0ff0b72330e0940a7ded4128614870d5be2e138b8d63905124c3d3ec0f
thread:issue:98537:issue:98541,issue #98537: Segfault when using torch.ops.* to access de-registered op,high-priority/blocker,52,high-priority/blocker,2,0,"98537, 98541",,,,1,081d93177c3913adadb6ca57101424bf63bde13c6509f342adebfc978d0b16c3
thread:issue:100085,issue #100085: [regression] torch.norm with out dtype bfloat16 cause runtime error,high-priority/blocker,52,high-priority/blocker,1,0,100085,,,,0,12cc8948350664d9e12f35aa8a59a1f9bf432f52b74df9339cbda2e136d99b6d
thread:issue:101529,issue #101529: Inconsistent performance degradation of 3x3 convolution (torch 2.0.1+cu118),high-priority/blocker,52,high-priority/blocker,1,0,101529,,,,0,b0758cb1af17695f51526c585fb132cb30a315d5635bbd5f5222001772a74dbe
thread:issue:102904,issue #102904: Unable to checkpoint model and optimizer state when using Hybrid Sharding Strategy,high-priority/blocker,52,high-priority/blocker,1,0,102904,,,,0,cf8f854e598df3e440d4868e31f00e67e7dc0a63b4add11f33441d85b185f704
thread:issue:103117,issue #103117: Observed regress in DataLoader spawn from PyTorch1.13 to PyTorch2.0,high-priority/blocker,52,high-priority/blocker,1,0,103117,,,,0,577ff0d7be1b745b76262241d41bfc9615f217ec152d11c52264bf90189bdba0
thread:issue:103966,issue #103966: 'MPS' Issue Running HuggingFace Transformer Pix2Struct Model,high-priority/blocker,52,high-priority/blocker,1,0,103966,,,,0,95f174bb3ac3d1f547f7800b12245841cdae29b0549a4d136d3c0429ef23b6b1
thread:issue:104342,"issue #104342: Segmentation error while using F.cross_entropy with mps(for code that works fine with device= ""cpu"")",high-priority/blocker,52,high-priority/blocker,1,0,104342,,,,0,237737a53bb3d263f28fad16239b1f156dc65ec33bff3a70262b97d9835d79c5
thread:issue:106362,issue #106362: Calling ops.aten.embedding_bag() function got silent crash,high-priority/blocker,52,high-priority/blocker,1,0,106362,,,,0,ea150257230e9274da2bf69744ecec19d05f6a4e5ee5c67de0551d19338f4248
thread:issue:106432,issue #106432: inconsistent dtype of scale and zero_point in observers,high-priority/blocker,52,high-priority/blocker,1,0,106432,,,,0,32c920572d2bbd4582ac4aa4f4c6f86d887ea0d05833006e86705a2c337b184c
thread:issue:106700,issue #106700: C++ API `torch::nn::MultiheadAttention` Crashes by division by zero,high-priority/blocker,52,high-priority/blocker,1,0,106700,,,,0,ee82024f7bed2022b930b2101b8cec2c427a9e3aee9099b30d733bb595e96ba0
thread:issue:107363,issue #107363: Regression in text encoding,high-priority/blocker,52,high-priority/blocker,1,0,107363,,,,0,ed1fc223723daf43db0d70618433ba4d806eb2b6c56b02bc5a7897b0bfeb57f7
thread:issue:107429,issue #107429: Abort when running torch.set_num_interop_threads,high-priority/blocker,52,high-priority/blocker,1,0,107429,,,,0,c0ccf0761bc9c147b3878e8b308f19fed0dc499c0bb6e11d23fcc1019daa08f1
thread:issue:107668,issue #107668: torch.dot gives wrong result on Macos,high-priority/blocker,52,high-priority/blocker,1,0,107668,,,,0,3ea07138b855bb2fa5df6e54328c2ffaf4c191d40ec231735e001117eeb3465c
thread:issue:107699,issue #107699: index.Tensor_out & index_put.out errors or segfaults with indices list containing only null tensors,high-priority/blocker,52,high-priority/blocker,1,0,107699,,,,0,c86ea9fdf7eef2d35d72a7c8f411cc538f6bd33e5e59a2c8cc7b7fcc87dd853a
thread:issue:108532,"issue #108532: Breaking incompatibility with Cuda 12.2, pytorch stable, torchvision",high-priority/blocker,52,high-priority/blocker,1,0,108532,,,,0,7862053da63cd71842559e3b8fea0f3fe4cf930d2b2a7556c259bd18ac2b889d
thread:issue:109014,issue #109014: Cannot install torchmetrics - ERROR 403,high-priority/blocker,52,high-priority/blocker,1,0,109014,,,,0,076a69684d09a300a7300e1c8ce086068db38abec01dcac3813ae659fb3838ef
thread:issue:110205,issue #110205: RuntimeError: Expected packed scalar Tensor to be of dimension 1. Got 0 instead.,high-priority/blocker,52,high-priority/blocker,1,0,110205,,,,0,059273c923f658bba92deddbf97451c4c9f5ae8425c1c9ab3172cf68ecfb7eb4
thread:issue:113037,"issue #113037: `TensorIteratorBase::is_scalar` return `false` for empty numpy tensors, but true for empty Torch ones",high-priority/blocker,52,high-priority/blocker,1,0,113037,,,,0,da46ffdcc6dd718f17a1fd5889f3e8f7fbc19cfcc6440c63e1384e99d681f034
thread:issue:113663,issue #113663: FP8 types should not participate in type promotion and should have no math ops defined on them,high-priority/blocker,52,high-priority/blocker,1,0,113663,,,,0,0c1988e75a640f23a1ef18a1b58fd24de9a3b3f3705ba8aa0e6fe651ad9802b1
thread:issue:114529,issue #114529: Bug in element-wise multiplication of `torch.sparse_csr_tensor`s on GPU - 0's in result considered significant - PyTorch 2.1.1,high-priority/blocker,52,high-priority/blocker,1,0,114529,,,,0,b52779d4642a543269f6a76868315f11781cc17943ed88f940cd2e650b77fb68
thread:issue:115416,issue #115416: torch.nn.functional.embedding segmentation fault on large negative ids,high-priority/blocker,52,high-priority/blocker,1,0,115416,,,,0,a74d01dcec868abada8117e90ab52e7b060b050ba0bbed424e7c5cdc8f60f842
thread:issue:116928,"issue #116928: Mistral works on (2.0.1+cu117) but ""CUDA out of memory"" on (2.1.2+cu121)",high-priority/blocker,52,high-priority/blocker,1,0,116928,,,,0,f9dbd959c2f9d62c94352c0cca943eaacf3af968e97c58fa27faa8a6baf2074f
thread:issue:117254,issue #117254: ```at::mse_loss```: Internal assertion failed,high-priority/blocker,52,high-priority/blocker,1,0,117254,,,,0,5a4067ec7c3706c263d2452af9d41c4fc2276a08a1b819a8603f92ed6e264b3f
thread:issue:117431,issue #117431: Program killed by SIGFPE in the backward pass of torch.sparse.sampled_addmm when input has specific shape,high-priority/blocker,52,high-priority/blocker,1,0,117431,,,,0,251019b884925529a6cf78c802e6df55ed6c106f389c9a7a92f6c717fadf7c80
thread:issue:119000,issue #119000: Performance regression in torch.nn.Linear,high-priority/blocker,52,high-priority/blocker,1,0,119000,,,,0,e5dd5e4e4389459ad6b8c0d885c60ae197cad75063fa9996c0c7c473a44361d7
thread:issue:119261,issue #119261: Performance Concern about CrossEntropyLoss,high-priority/blocker,52,high-priority/blocker,1,0,119261,,,,0,58bcfd9965bdb892ef566d1b7a8d9a8594e1857c330770db77e5f036357d9298
thread:issue:119367,issue #119367: Updating a slice on CPU with a slice from MPS results in an error,high-priority/blocker,52,high-priority/blocker,1,0,119367,,,,0,b0af653cacd320953836f3f158fa9a8ee37ffbf69fd6da5b2b786bf02bd3e385
thread:issue:119547,issue #119547: CUDA Memory error while having a lot of free memory,high-priority/blocker,52,high-priority/blocker,1,0,119547,,,,0,bfe64b3f382b8c9f1ac4ae15bb0899fa6ad77ff3d9a661dc2f076c850037a75e
thread:issue:120622,issue #120622: _native_batch_norm_legit doesn't bump version counter of buffers,high-priority/blocker,52,high-priority/blocker,1,0,120622,,,,0,696579305aa4addf3407ad9554d07b1810dc79b0416cdbf93499c6d24d2e3194
thread:issue:120995,issue #120995: AttributeError: module 'mpmath' has no attribute 'rational',high-priority/blocker,52,high-priority/blocker,1,0,120995,,,,0,40d726ec2b9a22f212c835b4e520f39ac0e0791dc71cda9dc84a4a0299c1edcd
thread:issue:121150,issue #121150: ☂️ oneDNN update functional/regression,high-priority/blocker,52,high-priority/blocker,1,0,121150,,,,0,096ff43331d30586245aa2c6d68a62c5f711d4915c19b504b991ee28020d829c
thread:issue:121711,issue #121711: torch.func.grad hangs on nested differentiation,high-priority/blocker,52,high-priority/blocker,1,0,121711,,,,0,d2c5fa098433eaf90e0eb8384fc6b40013106761cb24ed61e08fae619524cfec
thread:issue:122394,issue #122394: [MPS] Severe Performance Degradation with Deterministic Algorithm on macOS 14.4,high-priority/blocker,52,high-priority/blocker,1,0,122394,,,,0,4ab19703fc50805688ee797a31af61ee6cf0c0f3fd4f59c7abf6748995c75bf0
thread:issue:122617,issue #122617: [Regression PyTorch 2.1 & 2.2] `torch._transformer_encoder_layer_fwd` outputs on CUDA/NestedTensorCUDA backends significantly diverge from truth for torch>=2.1 (SDP math)),high-priority/blocker,52,high-priority/blocker,1,0,122617,,,,0,0a4fbf0f33afc092ccc35f447d57e50bb3dd7355e5bc78e1adce0acc50ddc783
thread:issue:123219,issue #123219: [CI] vec_test_all tests are not run in CI,high-priority/blocker,52,high-priority/blocker,1,0,123219,,,,0,1b26f50dd40f6d4592ea86399e67f5bf7f2d3ee5fbdeb30f3e11ee49551b4649
thread:issue:124644,issue #124644: torch.Tensor.remainder raises a floating point exception when divisor is -1,high-priority/blocker,52,high-priority/blocker,1,0,124644,,,,0,02bf2cfd7090586407783fcfbb77e650c889e9638699527b21ce591d88bf0568
thread:issue:125247,issue #125247: RuntimeError: could not create a primitive descriptor for a matmul primitive,high-priority/blocker,52,high-priority/blocker,1,0,125247,,,,0,4c4f6e929b0e480ff044dc351cddb8195801401294447e2778583e0dbe7cdf73
thread:issue:126385,issue #126385: Segmentation Faults loading Models pytorch v2.3.0 Apple M2,high-priority/blocker,52,high-priority/blocker,1,0,126385,,,,0,ba0de31302ee7353b36eba268596719ef7502ed64c8c88b11aeb062d2f3677ea
thread:issue:127102,issue #127102: segfault when registering op with mismatched schema,high-priority/blocker,52,high-priority/blocker,1,0,127102,,,,0,ca33344f98696a08e429678da8c4544a8cefda69a4173a104d040fe06ad6521c
thread:issue:128503,issue #128503: crash@sleef_tryVXE2 () while trying to run torch.compile() BERT model,high-priority/blocker,52,high-priority/blocker,1,0,128503,,,,0,3d0917998c6f1b33623dd4a413d566b638070cae1a5b8bac341e061a47457451
thread:issue:128690,issue #128690: `torch._transform_bias_rescale_qkv`:FPE,high-priority/blocker,52,high-priority/blocker,1,0,128690,,,,0,1906c937f12297052ec8fb4bcc342fc3c6d3cede590fd661c6f6dd1e474215b5
thread:issue:128691,issue #128691: Segmentation fault (core dumped) in `torch._remove_batch_dim`,high-priority/blocker,52,high-priority/blocker,1,0,128691,,,,0,9d1e4b2fccf9c7e1aa314abc1c0a467349842a578e7b2f45f83d7a25b060d7b4
thread:issue:128692,issue #128692: Segmentation fault (core dumped) in `torch._weight_int4pack_mm`,high-priority/blocker,52,high-priority/blocker,1,0,128692,,,,0,7c133ff4c5aab984d7f7522bb174c229402d620f75d0d365a1d4e1cd774c65b2
thread:issue:128695,issue #128695: Segmentation fault (core dumped) in `torch._weight_norm_interface`,high-priority/blocker,52,high-priority/blocker,1,0,128695,,,,0,689a96496d6a7a31d43c54339103ed74d05653d3061f324676f6c46ef46aa809
thread:issue:128819,issue #128819: Environment gating of CUDA_VISIBLE_DEVICES returns a CUDA initialization error.,high-priority/blocker,52,high-priority/blocker,1,0,128819,,,,0,695b00fba9a15050f0c87216d8070d8ef7f58059d35b1760b8fcf1e4990e8eb7
thread:issue:132086,issue #132086: Result of MPS memory operation can be corrupted if memory pressure is high,high-priority/blocker,52,high-priority/blocker,1,0,132086,,,,0,db1decc90d3a0bbc392e32071f719d2ee552b0ca4d112ad63b92bdd245d63aa1
thread:issue:133334,issue #133334: torch._scaled_mm row-wise hits CUDA invalid memory access when M % 256 != 0,high-priority/blocker,52,high-priority/blocker,1,0,133334,,,,0,0836eda42f696c11e1d2c8378fe1e02a00112b120cb64e1d797b273946eb6d73
thread:issue:133625,issue #133625: Binary Cross Entropy crashes kernel on apple silicon with singleton dimensions,high-priority/blocker,52,high-priority/blocker,1,0,133625,,,,0,816325a8519c8e5a7cb3820fc356926317e7c8f7eabae7f1de6a686413a7218e
thread:issue:134804,issue #134804: FlopsProfiler double counts sdpa flops,high-priority/blocker,52,high-priority/blocker,1,0,134804,,,,0,6a918fb0a4602313af9d5eb7cec17776301eaeeb8c9a73cafb0108d8cdacadb2
thread:issue:135637,issue #135637: MPS code contains references to undocumented APIs,high-priority/blocker,52,high-priority/blocker,1,0,135637,,,,0,662898531fdb7599b5b636aeebc3e7fd48cf646c583911a44c7428cf7d41598f
thread:issue:135885,issue #135885: ☂️ 150+ MacOS tests were marked flaky recently,high-priority/blocker,52,high-priority/blocker,1,0,135885,,,,0,35dd126445fdd3a4f3830eb3f8c06853bdf8c5ef1ad9d8c5d25c524dde1b5088
thread:issue:136217,issue #136217: Segmentaion fault on Apple M1 caused by libomp on v2.4.1,high-priority/blocker,52,high-priority/blocker,1,0,136217,,,,0,dc21c733896c6ab9b0e30c6560bbd89cb5a288647f76bbfb98c4f29d8c775591
thread:issue:136299,issue #136299: The results of baddbmm on CPU seem to have issues under certain conditions,high-priority/blocker,52,high-priority/blocker,1,0,136299,,,,0,2ad243e9d7a878dfc351a2dd326c0f77cb33baa61d8d2804c0bb52fb1b93c76d
thread:issue:136374,issue #136374: Runbook for GCP A100 runners should be updated,high-priority/blocker,52,high-priority/blocker,1,0,136374,,,,0,8f19a4ca340f041f60d4be006f990a3666110f038e27d40ad2b8ec3cd4acca65
thread:issue:136704,issue #136704: Segmentation fault (core dumped) in `torch._fft_r2c`/`torch._fft_c2`/`torch._fft_c2r`,high-priority/blocker,52,high-priority/blocker,1,0,136704,,,,0,0de12a64428e73bc733c1b08245e506ee0da0c8f2aecaf8c9e3820aa7ccc229f
thread:issue:136709,issue #136709: Segmentation fault (core dumped) in `torch._weight_norm`/`torch._weight_int8pack_mm`,high-priority/blocker,52,high-priority/blocker,1,0,136709,,,,0,deec673934816cd697da7c34bb333ad9f1cd5b549d4972510953b0ea5ed0f558
thread:issue:136714,issue #136714: Aborted (core dumped) in `torch.linalg.ldl_solve` with double free or corruption (out),high-priority/blocker,52,high-priority/blocker,1,0,136714,,,,0,11b83f8bf47cd5d82cf265c3c1e8bff29916e3a01b4b832f02070aa7a177ab43
thread:issue:136719,issue #136719: Segmentation fault (core dumped) in `torch.nn.functional.max_pool1d`,high-priority/blocker,52,high-priority/blocker,1,0,136719,,,,0,2f93db4a20ee19f16c2062a47269b7e5a92a3326e5c9ff79fd9bb25e309aa051
thread:issue:136819,issue #136819: Aborted (core dumped) in `torch.smm`/`torch.hspmm`/`torch.hsmm`/`torch.sspaddmm`,high-priority/blocker,52,high-priority/blocker,1,0,136819,,,,0,c449295f1a16ca7838c7d91553cf7c6242932a80b4aa3b015480688d5a5d4846
thread:issue:137098,issue #137098: [CI] Multiple tests fail with `Host and device pointer dont match with cudaHostRegister` and are being disabled,high-priority/blocker,52,high-priority/blocker,1,0,137098,,,,0,0affe85e9ee53429155ac3b63b0128aec050010926ffacfb66dc62571733b9c1
thread:issue:137347,issue #137347: cuDNN dBias error starting on 10/2 nightly,high-priority/blocker,52,high-priority/blocker,1,0,137347,,,,0,b0ca7e8ef5ec6afdc85e5818cb3f86dbfe2edac965e5527ac3ff29c9885dbc4d
thread:issue:138425,issue #138425: torch.div throws a floating point exception when rounding_mode is `trunc` or `floor`,high-priority/blocker,52,high-priority/blocker,1,0,138425,,,,0,7c31e2e194cef9af61ad69cca0e0a7775689ef227f37eb4383dbda78174ec1ac
thread:issue:139242,issue #139242: Tensor.index_reduce produces incorrect result,high-priority/blocker,52,high-priority/blocker,1,0,139242,,,,0,98d10d4aacd8d0c4f326b7e7adfaa05d6c4cb5355232e06d688433335734c0d7
thread:issue:139987,issue #139987: FPE in torch.nn.functional.pixel_shuffle,high-priority/blocker,52,high-priority/blocker,1,0,139987,,,,0,9c7c978a328623ba22dacb21e2ce6fbd0f853bdf1cf6abdf116a80b9e04c1eb6
thread:issue:140272,issue #140272: There is probably a destructor ordering accident going on with registerPythonModule,high-priority/blocker,52,high-priority/blocker,1,0,140272,,,,0,6c97ebf159497e24a09a112e3feb7999477c3a1f38cbc6e0c079b154ac33a342
thread:issue:140945,issue #140945: Segmentation fault (core dumped) in `replication_pad1d_backward`,high-priority/blocker,52,high-priority/blocker,1,0,140945,,,,0,a7c8a2f17387f0922837ba2b149d3e2e7907b2015f528173e35c22e3cf2602bb
thread:issue:141083,issue #141083: ARM build failed with recent XNNPACK update: third_party/XNNPACK/src/reference/unary-elementwise.cc:125:14: error: invalid ‘static_cast’ from type ‘xnn_bfloat16’ to type ‘_Float16’,high-priority/blocker,52,high-priority/blocker,1,0,141083,,,,0,0e60310cf13c66aece24c8fe94681035d817cb029ddec41f3460cd9d352cc187
thread:issue:141229,issue #141229: FPE in `torch._scaled_dot_product_attention_math`,high-priority/blocker,52,high-priority/blocker,1,0,141229,,,,0,8c0be008d681d7db94b3d73f79ea15299f31d663ea51fe8cefaebf76f0f9bf74
thread:issue:141358,issue #141358: PyTorch crashes with `mkl_lapack_dgetrf` if `from_numpy and set_num_threads` are used,high-priority/blocker,52,high-priority/blocker,1,0,141358,,,,0,eb4bf3093fd61398be0d9fb443e818fdf9d50acf637ef77746e9580fb08c695f
thread:issue:141481,issue #141481: Segmentation fault (core dumped) in `segment_reduce`,high-priority/blocker,52,high-priority/blocker,1,0,141481,,,,0,6748d1502e04bf8b675e7e9bf343016d7028757a6a97ac578af7eb136b2629d2
thread:issue:141836,issue #141836: StrideAPI caused regression in channels-last logic,high-priority/blocker,52,high-priority/blocker,1,0,141836,,,,0,d7dc895eb88b8b2009fec4c9270feca93437660b440f6922a003b23f11e1a3f2
thread:issue:141847,issue #141847: Floating point exception (core dumped) in `slow_conv3d_forward`,high-priority/blocker,52,high-priority/blocker,1,0,141847,,,,0,113c202e6545aa375e8523ebe4262ec1238a34a0c813d221c26b5634d47408df
thread:issue:142032,"issue #142032: Python 3.12 ""from functorch.einops import rearrange"" fails with ""RuntimeError: First class dim doesn't work with python 3.12""",high-priority/blocker,52,high-priority/blocker,1,0,142032,,,,0,2d40ac4aec88defab9b971fc8f0f0c03a6ef14cb28577c86be72135b61f8ce24
thread:issue:142209,issue #142209: Segmentation fault (core dumped) in `convolution`,high-priority/blocker,52,high-priority/blocker,1,0,142209,,,,0,01416acb59cde03fdc0bde64f08138823d4f2d8bab6c6950122d48b4d93b79d6
thread:issue:142221,issue #142221: [MPS] nansum crashes with scalar input,high-priority/blocker,52,high-priority/blocker,1,0,142221,,,,0,ad2b28a639107e830618bb457de8f33895333798c4e42e8627ffa848bcdbd348
thread:issue:142837,issue #142837: Segmentation fault (core dumped) in `embedding_dense_backward`,high-priority/blocker,52,high-priority/blocker,1,0,142837,,,,0,b3ef608d19bf3980777b9a9e62be233ebe0cdb0e95c2b51c3eea59e27d53f39a
thread:issue:142838,issue #142838: Segmentation fault (core dumped) in `fractional_max_pool3d_backward`,high-priority/blocker,52,high-priority/blocker,1,0,142838,,,,0,1f56b0c0262e769cbf1a111c28dad74a4e2d5a721a181163e40eb076d1013ee8
thread:issue:143123,issue #143123: MPS incompatibility: Calls into the C++ engine to run the backward pass,high-priority/blocker,52,high-priority/blocker,1,0,143123,,,,0,36d728e1a842b6e2c85d087c90755e355390dd7c5ae7f7d141e68ff1c83904db
thread:issue:143480,issue #143480: Segmentation fault (core dumped) in `conv1d`,high-priority/blocker,52,high-priority/blocker,1,0,143480,,,,0,f2cbc4ff184860bcb3ce9818ae90137e25961bd2b5e390b4cc77ab377a1c0050
thread:issue:143481,issue #143481: Segmentation fault (core dumped) in `conv3d`,high-priority/blocker,52,high-priority/blocker,1,0,143481,,,,0,68293e6a0d073eb8df25abd92516a7e781dbebd47d418c4610e2384ec5c9ad64
thread:issue:143482,issue #143482: Segmentation fault (core dumped) in `embedding_backward`,high-priority/blocker,52,high-priority/blocker,1,0,143482,,,,0,cbe32422e9fc5efb12ffaea21a06d87aad12eea83903394f05a47b3f8111c222
thread:issue:143483,issue #143483: Segmentation fault (core dumped) in `embedding_bag.padding_idx`,high-priority/blocker,52,high-priority/blocker,1,0,143483,,,,0,e54c169778574c83fd032ce8716df8b91d9d7c02228895b91616263f3685a870
thread:issue:143485,issue #143485: Aborted (core dumped) in `mkldnn_rnn_layer`,high-priority/blocker,52,high-priority/blocker,1,0,143485,,,,0,699ba277e3e18fdce9fe17b012288fdeeb886688bd2410d4bc00b9769a2874f7
thread:issue:143490,issue #143490: Segmentation fault (core dumped) in `replication_pad2d`,high-priority/blocker,52,high-priority/blocker,1,0,143490,,,,0,4c9fb54a283ac88b6b5c78fdf5da974cd7d340981d5b6a47c8184ecbeaeb6a6a
thread:issue:145065,issue #145065: SIGFPE error when passing very large kernel_size to `avg_pool1d`,high-priority/blocker,52,high-priority/blocker,1,0,145065,,,,0,f3ef074c078009b003ee50b54c394255fbb2ec9773e13514f02461a747a34324
thread:issue:145071,issue #145071: Illegal memory access and segmentation fault due to large `storage_offset` in `as_strided`,high-priority/blocker,52,high-priority/blocker,1,0,145071,,,,0,d135c8ff2aee8fa19323880fdfe68bc4151ee26edb89638d5d12143b538c40a7
thread:issue:145259,issue #145259: No Range Check for `storage_offset` in `as_strided` Function,high-priority/blocker,52,high-priority/blocker,1,0,145259,,,,0,71c696152bd2b304e6e4feefd4c9a93622737d256059e592c833304ee078cf2f
thread:issue:145267,issue #145267: `torch.ops.aten.embedding_dense_backward` Crashes with Out-of-Bounds Indices On CPU,high-priority/blocker,52,high-priority/blocker,1,0,145267,,,,0,1f65baf258f1b02810181a742ff65c031ed93137696121240d006b8b866beefb
thread:issue:145604,issue #145604: `torch.ops.aten.copy` causes SIGSEGV when handling sparse CSR tensors with invalid metadata,high-priority/blocker,52,high-priority/blocker,1,0,145604,,,,0,2e52713b14685f2251cb09002cae5a31de881a8ce604fed7e1c21ae153948189
thread:issue:146286,issue #146286: Non-blocking GPU to CPU copy of complex numbers with the different conj status produces wrong results,high-priority/blocker,52,high-priority/blocker,1,0,146286,,,,0,47ed5af825099aa068abab5e8db232f3f624445773f0326ef01d9c20f96f6a53
thread:issue:146551,issue #146551: libtorch_cuda_linalg.so: undefined symbol: mkl_lapack_dsbrdbn on a source built PyTorch 2.6.0 with USE_STATIC_MKL=1 on CUDA platform,high-priority/blocker,52,high-priority/blocker,1,0,146551,,,,0,2b883e31c8685061176b655cdcc2604501ef94f6e3214a6107699bffe81f1a1e
thread:issue:146787,issue #146787: Floating Point Exception in `torch.ops.aten.native_channel_shuffle` with `groups=0`,high-priority/blocker,52,high-priority/blocker,1,0,146787,,,,0,1d09f6ccc799f3c03b6da3dd05857947029b0cd9edf576d767351f27130dd3a5
thread:issue:146789,issue #146789: Segmentation Fault in `torch.ops.aten.linalg_eigvals` After Invalid `unfold_copy`,high-priority/blocker,52,high-priority/blocker,1,0,146789,,,,0,7db8668e055544aca3dd10d8d36c0c046dccc028faf6bf547ffc3a3b1e1595ce
thread:issue:146791,issue #146791: Floating Point Exception in `torch.ops.aten.unfold_backward` with Specific Input,high-priority/blocker,52,high-priority/blocker,1,0,146791,,,,0,7ea16d156e07b9f394854ffab09e4bebc798583e754d10800e9e821b80dcfd03
thread:issue:146793,issue #146793: Segmentation Fault in `torch.ops.aten.as_strided` with Large `storage_offset`,high-priority/blocker,52,high-priority/blocker,1,0,146793,,,,0,b834af086c2659a0a1b845e14f44c4e091d5d11ca8579121beb9547ee49e0b78
thread:issue:146794,issue #146794: Segmentation Fault in `torch.as_strided_copy` with Large `storage_offset`,high-priority/blocker,52,high-priority/blocker,1,0,146794,,,,0,81bc326d264c5b64da12ec9fb45ac0270b2987bda024c88a88f508d03a96f66c
thread:issue:146798,issue #146798: Torch 2.6 Unexpected Graph Break with SubConfigProxy,high-priority/blocker,52,high-priority/blocker,1,0,146798,,,,0,491df8cf4b76d768c9bdf763dce3c32f06fb0735cfcbd38f9109daa6f0bd77ed
thread:issue:146857,issue #146857: Segmentation Fault nn/test_convolution.py test_ConvTranspose2d_output_size_downsample_upsample on Aarch64,high-priority/blocker,52,high-priority/blocker,1,0,146857,,,,0,87ec464e7e62279b786791af3055e4a84706b010ec81de438733270e747655b7
thread:issue:147362,issue #147362: Floating point exception (core dumped) in torch.nn.functional.conv3d,high-priority/blocker,52,high-priority/blocker,1,0,147362,,,,0,3be9a3fc4536185e89fdc2c949d0992b8302e7f57d28df37e0cfc8a4b20b8958
thread:issue:148699,issue #148699: CUDA 12.6 Inductor accuracy test failures,high-priority/blocker,52,high-priority/blocker,1,0,148699,,,,0,a42bdfb2dd55bdb4b662de5cbccdf5651e72494c09eede894dc6f579aa424a90
thread:issue:149201,issue #149201: Segment fault when used with FAISS on arm,high-priority/blocker,52,high-priority/blocker,1,0,149201,,,,0,841f8ff84ea7dacaf0f2461fa1ee9b1e7768c9f3efbb08c2bc95a90def0915b8
thread:issue:149658,issue #149658: unexpected kwarg 'no_python_abi_suffix' when installing stable version of pytorch with sam2,high-priority/blocker,52,high-priority/blocker,1,0,149658,,,,0,2613a567b6c540926ea2de65d804d998dee0a09edac43bbe9e711cad443515df
thread:issue:149820,issue #149820: `Segmentation fault` when using `torch.sparse.mm` with `torch.sparse_csr_tensor`,high-priority/blocker,52,high-priority/blocker,1,0,149820,,,,0,140bf1e7999fcf295fb6ff5b48de3c3d6d4cbb2ace852412d19e1216893de728
thread:issue:150187,issue #150187: at::BlasBackend::Ck does not handle all ROCm BLAS gpus,high-priority/blocker,52,high-priority/blocker,1,0,150187,,,,0,75b4e3330771198b432b3dad4ff60a4936dc42ac1ee4f2a6289ae86519786bc4
thread:issue:150388,issue #150388: double free or corruption (out) in torch.as_strided_scatter,high-priority/blocker,52,high-priority/blocker,1,0,150388,,,,0,337dea428130f9ba72f10e4b583bd5ce56c3f68e2f4a4780579b2a6e60811a88
thread:issue:151680,"issue #151680: undefined reference to `at::native::blas_impl::fp16_gemv_notrans(int, int, float, c10::Half const*, int, c10::Half const*, int, float, c10::Half*, int)'",high-priority/blocker,52,high-priority/blocker,1,0,151680,,,,0,cdf6bef9eb3d7ff5332ff520d441be0bb9f91c0bef262c2ef1aeb299567e718a
thread:issue:152292,issue #152292: `torch==2.6` broke `nn.Module.dtype` typing,high-priority/blocker,52,high-priority/blocker,1,0,152292,,,,0,7b93708f5f1294ab43019d31443812eb5cb64f670213d0feb6553bea0ec5a70a
thread:issue:152535,issue #152535: optree package status in PyTorch,high-priority/blocker,52,high-priority/blocker,1,0,152535,,,,0,307518a78629c22f3f598efd9cd51813d559cf31a448d53d1e1a9beae779d5dc
thread:issue:152747,issue #152747: torch.compile causes stride mismatch in SDPA with non-contiguous query in torch 2.7,high-priority/blocker,52,high-priority/blocker,1,0,152747,,,,0,e2b24be902afad02d27a87cf42b80842f12d218c7bb1d450a0d6be3e89981043
thread:issue:152777,issue #152777: Segmentation fault (core dumped) in torch.nn.functional.alpha_dropout,high-priority/blocker,52,high-priority/blocker,1,0,152777,,,,0,e5523ba59628c1cc177fc663cdf4373267486da7421159d2823ff009a2e6beb6
thread:issue:152824,issue #152824: `mypy` stage of `lintrunner -a` has intermittent but continuing crashes,high-priority/blocker,52,high-priority/blocker,1,0,152824,,,,0,294cccd90e9cdebe44b65c82010b70ffff6b3f40e813fba0eedfa81999dd5149
thread:issue:152913,issue #152913: Segmentation fault (core dumped) in torch.nn.functional.max_unpool2d,high-priority/blocker,52,high-priority/blocker,1,0,152913,,,,0,06fe994aaf9946bbea09b54dc6b022cf8fe29370b5928eaf07d1b27f6def2053
thread:issue:152979,issue #152979: FPE when using `torch.lcm_` with int32 tensor and int16 scalar,high-priority/blocker,52,high-priority/blocker,1,0,152979,,,,0,dd8f11408ef6ae449d28fc0ea8b3bbf1e90e9197e23990893fe9762d3fb6ec54
thread:issue:153050,issue #153050: Process never ends when sending tensors through multiprocessing queues in Python 3.12+ with filesystem strategy,high-priority/blocker,52,high-priority/blocker,1,0,153050,,,,0,3c2eaf8a267a76cfe7c83e91ae2a7f826e0548b2163225708174f52d9506a0ee
thread:issue:153203,issue #153203: FPE Crash in torch.fmod with large negative integer modulo value,high-priority/blocker,52,high-priority/blocker,1,0,153203,,,,0,874ea2f5d2a37e0e9f6bcf76575ee9ab76bcb7b62ad37372cae06f17c2d7929f
thread:issue:153231,issue #153231: `torch.native_channel_shuffle` crashes with Floating Point Exception when given large integer parameter,high-priority/blocker,52,high-priority/blocker,1,0,153231,,,,0,48bc3cb40f5d1bf1b5d9c2ae582a878b2484954eddbeedc77cda6043e2f1a191
thread:issue:153733,issue #153733: [SOURCE] links for patch releases lead to 2.7.1,high-priority/blocker,52,high-priority/blocker,1,0,153733,,,,0,1d77dc0e14233e856c4e7c193b54d245dd83e2e04d926a9e2d8a260a03ea7b5a
thread:issue:154312,issue #154312: torch.logdet produces incorrect results for singular matrices on CUDA vs CPU,high-priority/blocker,52,high-priority/blocker,1,0,154312,,,,0,f6ae735908daf8ba5172727dd9023dc0037a77af413643b13538c7193b7677ea
thread:issue:154422,issue #154422: `Segmentation fault` in `torch.fx.experimental.partitioner_utils.map_arg`,high-priority/blocker,52,high-priority/blocker,1,0,154422,,,,0,49bd9fb2e56b63c71a55a157976420bae4175505761d0e649aa4dc8d08255715
thread:issue:154424,issue #154424: `Segmentation fault` in `torch.matmul` and `torch.sparse.addmm`,high-priority/blocker,52,high-priority/blocker,1,0,154424,,,,0,6618267cfc3d38ab5d954f41965b3c305e258db8b75639ea7f5f13cf635ce6c2
thread:issue:154882,issue #154882: MPS max_pool2d_with_indices failure: destination values and indices length mismatch along axis,high-priority/blocker,52,high-priority/blocker,1,0,154882,,,,0,556264b6bb194dc6bd5f05c97c849439e4df3911f53051845114fdd8776afe70
thread:issue:154887,issue #154887: MPS batch_norm mixed dtype failure,high-priority/blocker,52,high-priority/blocker,1,0,154887,,,,0,3239c65f6046522373d66fde7e44ab4d6a0d2d22ae612551d25a9eaa9bc90446
thread:issue:155201,issue #155201: Importing xgboost before torch + openmp causes seg fault,high-priority/blocker,52,high-priority/blocker,1,0,155201,,,,0,198764b66559f4094eb3dc9d45c7d3e6615af541257d17390368276d25c8d47c
thread:issue:155293,issue #155293: [MPS] 5D 4Mln rng attempts fail with internal error,high-priority/blocker,52,high-priority/blocker,1,0,155293,,,,0,13eacced55e9eb2fe32d4c429e5cdba6bb425eea69f0ed88be8e118f8fd312f0
thread:issue:155393,issue #155393: crash in torch.histc,high-priority/blocker,52,high-priority/blocker,1,0,155393,,,,0,89b77b1f7cb5298e88f294f2578a91e95037cacf751c9a09a228532c2714062a
thread:issue:155607,issue #155607: CUDA 12.6->12.8 slow and periodic failures,high-priority/blocker,52,high-priority/blocker,1,0,155607,,,,0,8daca3567b6ef33b2ea056bf00ea7e4572e4c39f8dbc2dd7c14ed9b2332fef56
thread:issue:155981,issue #155981: Floating point exception (core dumped) in torch.nn.functional.fold,high-priority/blocker,52,high-priority/blocker,1,0,155981,,,,0,d11c1e66c823ebafee5105783d92d80e9cc7c053f8545c457ce3009d1d3df4d3
thread:issue:157097,issue #157097: Segmentation fault in torch.repeat_interleave,high-priority/blocker,52,high-priority/blocker,1,0,157097,,,,0,b572cb24e9607ec6a1d72fb8fdcf6ceeec32d8f02d000bb3ac993e11e9c4313d
thread:issue:157098,issue #157098: Floating point exception in torch.nn.functional.conv_transpose3d,high-priority/blocker,52,high-priority/blocker,1,0,157098,,,,0,3d8da1eadfa6acb2f50a58956cc67cf69f1069d3c8b7407cbb35fb82301968e4
thread:issue:157461,issue #157461: vLLM tests failing in torch 2.8rc but passing with torch 2.7,high-priority/blocker,52,high-priority/blocker,1,0,157461,,,,0,6a5dddba7f3007dfded28b0d81d1d032a3a74e3f4b5695254ea709b7e4a7bdf8
thread:issue:158154,issue #158154: `torch.native_channel_shuffle` throws FPE when `groups` > size of the second dimension of the input tensor,high-priority/blocker,52,high-priority/blocker,1,0,158154,,,,0,d7bfa39e8cee0e30187e0d7b82d3dbab15bcae887aa10202cff64b555d8d9e26
thread:issue:158728,issue #158728: [CI] Inductor-perf-nighly-macos started to fail to install torchvision,high-priority/blocker,52,high-priority/blocker,1,0,158728,,,,0,13046950f15d1bdd0ef903b8688a68b552dd672e2d1c854662048280b385d49a
thread:issue:159713,issue #159713: [MPS Crash] Assertion failed in _getLSTMGradKernelDAGObject using LSTM on macOS with MPS backend,high-priority/blocker,52,high-priority/blocker,1,0,159713,,,,0,e154a0bf2b160faba9e7b45680ce1c83a8d92d7803fcf58d464cfda8812990ed
thread:issue:160296,issue #160296: [inductor][cpu] Performance regression in 2025-08-02 nightly release,high-priority/blocker,52,high-priority/blocker,1,0,160296,,,,0,95b3a9dd67e6e4e22f8cd02525af3cb0d6cbacc1cec2982bbc7a7461e9b6706d
thread:issue:161358,issue #161358: periodic tests are failing to build torchvision,high-priority/blocker,52,high-priority/blocker,1,0,161358,,,,0,a8bd4b318ca6deafaa83b3b8d1fa2c02b359896771b2842bf1528ede708a27ce
thread:issue:161419,issue #161419: mobilenetv3_large_100 aot_eager accuracy failure,high-priority/blocker,52,high-priority/blocker,1,0,161419,,,,0,3c3728ce1265b689ac5cd0f3cf33e7a04ad408179ffe21804cd61e608e25a2d5
thread:issue:161760,issue #161760: Invalid __shared__ read of size 8 bytes in torch.histc,high-priority/blocker,52,high-priority/blocker,1,0,161760,,,,0,3df5a776041737c5e9759fa073eb4ce4a7c3e5b75787954ef73cefed43d2dcd9
thread:issue:161874,issue #161874: Floating point exception in torch.nn.quantized.Conv2d.register_full_backward_hook,high-priority/blocker,52,high-priority/blocker,1,0,161874,,,,0,338a2fbb0cea884ab8b0efe68edc8776f8c0bd9f0a857a029205dc90ae8686a4
thread:issue:162884,issue #162884: setup.py develop --cmake no longer works,high-priority/blocker,52,high-priority/blocker,1,0,162884,,,,0,de30b12472bd403b4dea15164a6d944a0e44800914dddae70f5bf31cc5153444
thread:issue:163539,issue #163539: conv3d bfloat16 wrong result,high-priority/blocker,52,high-priority/blocker,1,0,163539,,,,0,c5779b1d20067250be3e2790049574c0a00499bdef6a275eb8d5bdbedf3d99a8
thread:issue:163675,issue #163675: Torch CPU wheels no longer available for download from PyTorch website,high-priority/blocker,52,high-priority/blocker,1,0,163675,,,,0,27c0805d005464cbfabb3685bf6a8e0b28004b61ba1cfd305780138576327635
thread:issue:164562,issue #164562: Update FlashInfer version used by vLLM tests on PyTorch CI to v0.4.1,high-priority/blocker,52,high-priority/blocker,1,0,164562,,,,0,c07386f6bebc9c6355d1a12b292c7eb96971c5e2a0adebbcaedff853e519c8d5
thread:issue:164957,issue #164957: Something broke `from functorch import make_fx` on main,high-priority/blocker,52,high-priority/blocker,1,0,164957,,,,0,58b41d8a2e1c36ec647c785a3abe2da00d4d5d68d3bee03b98ccb1af45619291
thread:issue:165123,issue #165123: https://github.com/pytorch/pytorch/pull/128406 caused ISA-specific symbols leak,high-priority/blocker,52,high-priority/blocker,1,0,165123,,,,0,4b0c3a15b321c2990d3f96dcbf784561cc13c57f87963f8e700d80dc8f7c618f
thread:issue:16953,issue #16953: Seg fault with test_rnn_retain_variables on ppc64le,high-priority/blocker,52,high-priority/blocker,1,0,16953,,,,0,fb21f13d890c23153627fbf9594e00cce2ebf59d8ff65515677af42e1815738e
thread:issue:171284,issue #171284: GRU: Segmentation fault (core dumped),high-priority/blocker,52,high-priority/blocker,1,0,171284,,,,0,00c189123a9db265f0b37406eedaca53dbc06bd5f1d535bd2b7b5a21d3712b08
thread:issue:172011,issue #172011: torch.pixel_shuffle CPU-GPU crash with massive upscale factor,high-priority/blocker,52,high-priority/blocker,1,0,172011,,,,0,d6c0ddffc7487b97e8f6a1329c9e08e4ee23c494385ab4fb2baf23ed868d048a
thread:issue:18300,issue #18300: NCCL backend fails when calling broadcast from different threads,high-priority/blocker,52,high-priority/blocker,1,0,18300,,,,0,88f9d9b7ee879ead3edcec60743de60e7cf35eac3ee8896f343a39711c69d5aa
thread:issue:184878,issue #184878: Division by Zero Floating Point Exception in monitor aggregation,high-priority/blocker,52,high-priority/blocker,1,0,184878,,,,0,3afb9f99f4c8492ed3e151c8d0f84c264266958ccc96843e46ee8f4102b3b204
thread:issue:19739,issue #19739: Importing open3d after PyTorch causes free(): invalid pointer,high-priority/blocker,52,high-priority/blocker,1,0,19739,,,,0,770be22732be1dc0af06c66f63324384c217db4448666d6afe3aa0c20fddadb5
thread:issue:20497,issue #20497: Statically make `__setstate__` set all attributes/parameters,high-priority/blocker,52,high-priority/blocker,1,0,20497,,,,0,b163f10c0b6324241866549887a74a0713bc4dc9c4b86528fafcbfc538760c68
thread:issue:20748,issue #20748: Segmentation fault when use torch::from_blob,high-priority/blocker,52,high-priority/blocker,1,0,20748,,,,0,842841f0627b54269fe97d1e981dff9e6c14c322ee508a431705a201a6720cff
thread:issue:25518,issue #25518: Assign torch.cuda.FloatTensor to List tensor,high-priority/blocker,52,high-priority/blocker,1,0,25518,,,,0,52fd5659fad22c0a0613dc50202649488f2707a0deb237c09a8df94ac120af01
thread:issue:28980,issue #28980: Unsafe use of `at::parallel_for` in current codebase,high-priority/blocker,52,high-priority/blocker,1,0,28980,,,,0,6ec557fe7226df43ac4486d7035617c2ba02ea6d4eee09a86d5d7addd5635281
thread:issue:31558,issue #31558: More dynamic PyTorch APIs,high-priority/blocker,52,high-priority/blocker,1,0,31558,,,,0,570300dcea63785a181ceac0e0cfd9ae66e2ad5d866c8170ae0a165601ec3c7f
thread:issue:35151,issue #35151: Segfault when using misaligned data pointer (from joblib),high-priority/blocker,52,high-priority/blocker,1,0,35151,,,,0,4a570d7ee600437295c9f17ee7d4814164c2754b717fbcf2eeea5b750b05a513
thread:issue:35418,issue #35418: PyTorch / libtorch executables fail when built against libcuda stub library,high-priority/blocker,52,high-priority/blocker,1,0,35418,,,,0,8993d8c93b982c1fc6e891edd6ab45055d37bc8d87700f1ad4e65bff145016df
thread:issue:35597,issue #35597: libtorch_global_deps.so not found.,high-priority/blocker,52,high-priority/blocker,1,0,35597,,,,0,36539f2841cf9bd31ba10ddb65dbe4b720b8d9c582e2d2166e645a3872d9982e
thread:issue:36040,issue #36040: [JIT] Huge delay (1274s vs 0.031s) when running scripted model,high-priority/blocker,52,high-priority/blocker,1,0,36040,,,,0,ad683b648584212468a96b16a1a3de823dd945f241486a4469ada60f81dbe250
thread:issue:36152,issue #36152: Seg Fault: import vaex with torch,high-priority/blocker,52,high-priority/blocker,1,0,36152,,,,0,a833b0958b8b4d4f74c2284eb6a5bf1f5914c8a8d01157ae2b5dc054fcd52bd5
thread:issue:37449,issue #37449: RuntimeError: CUDA error: an illegal memory access was encountered with channels_last,high-priority/blocker,52,high-priority/blocker,1,0,37449,,,,0,cd8cc1f0cc52e8461d3c14b36032442c876616597a08fc5ff1d3cc1f1237ac82
thread:issue:38122,issue #38122: Pybind11 cpp extensions broken with pytorch v1.5.0,high-priority/blocker,52,high-priority/blocker,1,0,38122,,,,0,f6f4b04765b63aa47bde556ad447e8d9091b3e7a0fa335468fbca64bc5887d25
thread:issue:38231,issue #38231: Python autograd engine threads never terminate in Python 3.5-3.8,high-priority/blocker,52,high-priority/blocker,1,0,38231,,,,0,630dae62345358ff56f26ecac555d5822356696afd612333959b32a190c82ae7
thread:issue:38623,issue #38623: Providing CUDA tensor to model on CPU causes a crash,high-priority/blocker,52,high-priority/blocker,1,0,38623,,,,0,d913e5e3deacacf9437fb299718b6afb895cb4b3fdfde037ae94266f112c6c86
thread:issue:38910,issue #38910: Memory Leak with Docker GPU,high-priority/blocker,52,high-priority/blocker,1,0,38910,,,,0,b3614fd54fe5761b163169a37e3317b5faf53e42aba2f51b2c63d716a41e8cc0
thread:issue:39061,issue #39061: Some @slowtests are never run in CI,high-priority/blocker,52,high-priority/blocker,1,0,39061,,,,0,0bace5684aa5103909fa38696d12022f3400cf53df417b00aeee9b470cc8f104
thread:issue:39388,"issue #39388: RuntimeError: NCCL error in: /pytorch/torch/lib/c10d/ProcessGroupNCCL.cpp:514, unhandled system error, NCCL version 2.4.8",high-priority/blocker,52,high-priority/blocker,1,0,39388,,,,0,2e3268bb078b360cc8bb93372fca449d8d6e0e9c30edf63289c425c3aebe9e2c
thread:issue:40457,issue #40457: DataParallel with Torch 1.5,high-priority/blocker,52,high-priority/blocker,1,0,40457,,,,0,8003d9ea3263ae6c89bb9e1ab954ab63ca5f0acebc9e4411f66080089ae99c22
thread:issue:41661,issue #41661: libgcc_s.so.1 must be installed for pthread_cancel to work,high-priority/blocker,52,high-priority/blocker,1,0,41661,,,,0,179680ba2d4cce2245610cab91e86f7af4a3c3c502e70c664e0b629eebfcd023
thread:issue:42452,issue #42452: `device-side assert triggered` - when probability > 1.0,high-priority/blocker,52,high-priority/blocker,1,0,42452,,,,0,c380c1f26c15b9e1c5ca1ebea56e56fe667ca88677399ab0e6fbbab18a68e4b6
thread:issue:42487,issue #42487: Support recursive data type in TorchScript,high-priority/blocker,52,high-priority/blocker,1,0,42487,,,,0,713c7b5490e202bd804e0469ee234c64aac1fc77424df19ab355e78af00d7f55
thread:issue:42784,issue #42784: libtorch 1.5 crashes when used on macs when using torch::max without AVX support,high-priority/blocker,52,high-priority/blocker,1,0,42784,,,,0,6565c695d5f802e0316e4c9264690ca8fa538b7335462e3fca5df8d32d7f6aa7
thread:issue:42793,issue #42793: distributed training with c10d does not work with layerdrop in pytorch > 1.4,high-priority/blocker,52,high-priority/blocker,1,0,42793,,,,0,4f176ed9448503c2e967ce43ec70bf6b909057b7449461937a5c12b6541afd24
thread:issue:43005,issue #43005: `torch.distributions.Categorical` crashes with illegal instruction,high-priority/blocker,52,high-priority/blocker,1,0,43005,,,,0,50447877c2463232c16feb8fda7cae1cf998d134a7d65e361cbb588f31a8c750
thread:issue:43484,issue #43484: pybind11_object_dealloc error,high-priority/blocker,52,high-priority/blocker,1,0,43484,,,,0,dc1a1f2a8e016d25b8d79d12fe4402f59e020c8b06ad55f55c1518865a1e5749
thread:issue:45136,issue #45136: Multiple torch.load in one file,high-priority/blocker,52,high-priority/blocker,1,0,45136,,,,0,3e991cedb351acfbadd47a96dc965fa32ed1173815d9d5effd288fe74a333597
thread:issue:46155,issue #46155: Complex backward returns NaN values,high-priority/blocker,52,high-priority/blocker,1,0,46155,,,,0,4e84db3aeec903b94b2056e7a4769e47c0a8515e9064fbe6a7ee67c16c1bccb6
thread:issue:46491,issue #46491: cannot call rpc.init_rpc twice within a single process,high-priority/blocker,52,high-priority/blocker,1,0,46491,,,,0,707bc213aafb505b86f149279a00566b89203058f2825164538e45d20133c189
thread:issue:47056,issue #47056: Training slows down and memory usage increases when upgrading from PyTorch 1.6 to 1.7,high-priority/blocker,52,high-priority/blocker,1,0,47056,,,,0,e3a5fddc1a7b435b731cfff9a0b0758586aaca85305acfddde7f16af64c31789
thread:issue:47116,issue #47116: trying to Installing pytorch on python3.9 via pip results in an non-descript error,high-priority/blocker,52,high-priority/blocker,1,0,47116,,,,0,597bfc3f3f2521ca2ca8ec3420f6bd58cbbba2263a7e25bbea50cfacb80a8ad3
thread:issue:47161,issue #47161: segfault during shutdown with torch1.7,high-priority/blocker,52,high-priority/blocker,1,0,47161,,,,0,ba571de7b6802c4b33610383cd16833ef12b2d23d5874b8a5c123fa2b7a2ae3e
thread:issue:47844,issue #47844: Incorrect output loss value under specific CUDA version,high-priority/blocker,52,high-priority/blocker,1,0,47844,,,,0,cbba5c05b531502d75c3dcf46d7cfe8a8c3216aa82b0e3c7d1a3d1d54e0ede63
thread:issue:48365,issue #48365: Custom exception for out of memory,high-priority/blocker,52,high-priority/blocker,1,0,48365,,,,0,9fe9b4bbd511ab8acd1796771d382b21ddd5acbeed407abfd2af6db6fa327b19
thread:issue:48382,issue #48382: hangs indefinitely at os.waitpid(),high-priority/blocker,52,high-priority/blocker,1,0,48382,,,,0,4c721136aa81d954ed78c65577d6ed540538f2f29f8b7467530390afcb445dc1
thread:issue:48400,issue #48400: conv_transpose3d returns different result when the input and kernel are mkldnn tensors,high-priority/blocker,52,high-priority/blocker,1,0,48400,,,,0,11e05e5d44c096434beb552d9b8e30d1802f43aea5d5863d498a9429e8e3a466
thread:issue:48558,issue #48558: [FR] tensor ctors should support directly creating in shared memory,high-priority/blocker,52,high-priority/blocker,1,0,48558,,,,0,bdf6efaa88fa54fb09b2d5929f6e5a6a97481218e40e55ec86d9945a9eb10f29
thread:issue:49392,issue #49392: Uninitialized variable was not detected in ASAN CI config,high-priority/blocker,52,high-priority/blocker,1,0,49392,,,,0,d5098827ce1aac7f434b1bee22afecbdb169c0d529e0fe920b844599a1e6b65b
thread:issue:49473,issue #49473: Pytorch 1.5+: Kernel dies / Segmentation Fault with torch.cuda.mermory_allocated() and torch.cuda.mermory_reserved(),high-priority/blocker,52,high-priority/blocker,1,0,49473,,,,0,66cc8a8cf1dc312eb17b89e3460a3b2816e47fb2e3ad5e1cec2418dc3c5047c5
thread:issue:49520,issue #49520: Segfault in torch.bincount,high-priority/blocker,52,high-priority/blocker,1,0,49520,,,,0,24ac7540244929b80bb0c161052facdcab57f312c797b44c19369517e395be44
thread:issue:49877,issue #49877: The same tensor requires more memory on RTX3090,high-priority/blocker,52,high-priority/blocker,1,0,49877,,,,0,cad1a0cd3dd1cc3a2193b011dffc53d9a220b8360901ecd33220c2a6ed0d789c
thread:issue:50013,issue #50013: torch.Tensor.repeat is divergent from np.repeat,high-priority/blocker,52,high-priority/blocker,1,0,50013,,,,0,62f69396faea74420843c7053d5aa006fe80fedd31b52ee5a7297d90af7d23a6
thread:issue:50037,issue #50037: Torch quantized_lstm_cell op out-of-bounds access,high-priority/blocker,52,high-priority/blocker,1,0,50037,,,,0,ed8b00ab895a68dc0a6b2c7e586810e7f02e0dc87153a991b46e5d0b7bbc049e
thread:issue:50038,issue #50038: Torch _remove_batch_dim OP out-of-bounds access,high-priority/blocker,52,high-priority/blocker,1,0,50038,,,,0,856b7179253bb128a45d2d8e11dab85bf3c1cd002c2f8f834556aa5a89f77bec
thread:issue:50879,issue #50879: Unexpected slow dropout in stacked RNN/LSTM/GRU,high-priority/blocker,52,high-priority/blocker,1,0,50879,,,,0,aff29eada3fa15c6aed1799b20acca3778d8ea0764c44e390453946cf4a0b1ce
thread:issue:51083,issue #51083: apex internal assert failed,high-priority/blocker,52,high-priority/blocker,1,0,51083,,,,0,88475c5c71b532cf04a657b093578ff1f403fc1c5891e492d6cf6b34e361f136
thread:issue:51134,issue #51134: segmentation fault in torch.nn.ReplicationPad3d/2d when padding is large,high-priority/blocker,52,high-priority/blocker,1,0,51134,,,,0,e8d1bea5979c3c809bbd98f9d812f92a0cb1eb22ed00d48baf3f4db051fbdac2
thread:issue:51332,issue #51332: Cant compile from source 1.8.0a0 because 'magma_v2.h' file not found and undefined references to `magma_*',high-priority/blocker,52,high-priority/blocker,1,0,51332,,,,0,33cdbba270ab4a7b8048089e4f9777a3708963f4d8f61b4930c606a184fa19d5
thread:issue:51732,issue #51732: torch.nn.functional.ctc_loss crash(segfault),high-priority/blocker,52,high-priority/blocker,1,0,51732,,,,0,5ef081400bfd2b406e7908bdab43020caf9301ffda9c16f2267a6a8276a518ec
thread:issue:51863,issue #51863: TestVectorizedMemoryAccess.CopyKernel starting to fail after driver to 460.39,high-priority/blocker,52,high-priority/blocker,1,0,51863,,,,0,a210af1c40dd663c60ac7bd960ebef82358e6c36f7929d13abeff95f5eebb9e4
thread:issue:52663,issue #52663: code linking to libtorch cannot use thrust/cub functions,high-priority/blocker,52,high-priority/blocker,1,0,52663,,,,0,c6abc318453d0c487f486381c4731d4b86118e4293d5b864ba548166fcb91b08
thread:issue:5353,issue #5353: Bugs: Score Function approach in REINFORCE for PONG,high-priority/blocker,52,high-priority/blocker,1,0,5353,,,,0,b9240f277c797ecfcfd75afe4d890c73ef8a6c340c4585f111c6443fc99b9f97
thread:issue:53625,issue #53625: torch.arange() issue,high-priority/blocker,52,high-priority/blocker,1,0,53625,,,,0,8a36f11043f276a624cbb8df547c52c9caf30a57188dd6d5528c0116c15dc189
thread:issue:53796,issue #53796: foreach ops in autograd,high-priority/blocker,52,high-priority/blocker,1,0,53796,,,,0,2e2d4af1650e41f4c8a1e9bf44c4abc2243f91d5f5249bf9805ae05b077e5abe
thread:issue:54418,issue #54418: Some Numba tests are failing,high-priority/blocker,52,high-priority/blocker,1,0,54418,,,,0,c90e5aa7eb9024dc3a2fdf7726ac70c8ec7b370428f690689396542b0a99a6b5
thread:issue:55669,issue #55669: Segmentation fault when loss.backward(),high-priority/blocker,52,high-priority/blocker,1,0,55669,,,,0,46b9af559b5e9fe12ae871ad09a1cb4921bde3ed16bde47dd39fb29088c1abd5
thread:issue:57178,issue #57178: MaxPool2D Returns Wrong Shape With Ceil_Mode,high-priority/blocker,52,high-priority/blocker,1,0,57178,,,,0,d9dfac930bddb2535009789af1cfe2d8ab22ca02b3ffd8083aa4bfa80d9b05a3
thread:issue:58005,issue #58005: torch.distributed.nn.all_reduce incorrectly scales the gradient,high-priority/blocker,52,high-priority/blocker,1,0,58005,,,,0,74f177813def100b5e2955d35ac13e53bc86104123206a253cff90f617aaad96
thread:issue:58370,issue #58370: Add better python operators annotations for IDE type checking,high-priority/blocker,52,high-priority/blocker,1,0,58370,,,,0,f92626d8665a185bb921c1bdd821d28d2de80ea95a82fce07a00619449311dd7
thread:issue:59196,issue #59196: torch.zeros memory leak : cpu,high-priority/blocker,52,high-priority/blocker,1,0,59196,,,,0,c9b5c145f3fb8cab6b09c76887ae7d7eaacc24115d5b69d340e3b04315eaf15d
thread:issue:59237,issue #59237: Multiplication of `torch.tensor` with `np.array` does the operation with numpy.,high-priority/blocker,52,high-priority/blocker,1,0,59237,,,,0,70c2d9aab4f4d70adce2176da666567e9e47314069f1e64e55d4da507268a6e6
thread:issue:59645,issue #59645: Tensor.type() does not work with meta tensors,high-priority/blocker,52,high-priority/blocker,1,0,59645,,,,0,c7db03178e2c7137f71e6e974b1bae53ca969730faa5fbd6b434d89753bfcffa
thread:issue:59745,issue #59745: [c10d] Work objects should have a general operator<<,high-priority/blocker,52,high-priority/blocker,1,0,59745,,,,0,fdc54f730d13adfddc04ce5b230d3488b3881c2fd11653477ca1ce6e446ac4d4
thread:issue:60081,issue #60081: [Bug] numpy is no longer a required dependency,high-priority/blocker,52,high-priority/blocker,1,0,60081,,,,0,2593690747dc2a3eded57081d459dc9d277795f7c76abde0ee1546b076347fb9
thread:issue:60626,issue #60626: Impossible to raise the limit on number of shared memory tensors,high-priority/blocker,52,high-priority/blocker,1,0,60626,,,,0,15d91b67488316e5e858d3f0747418b4e29fb27e3b243cbf1faf99e73347c786
thread:issue:60628,issue #60628: pip install nightly torch torchaudio and torchvision together will install 0.3.0 version of torchvsion on Ubuntu,high-priority/blocker,52,high-priority/blocker,1,0,60628,,,,0,8deb10df4f0c7e629ab6660e15f015c9146231e913b3091b3b59ec461a837ef8
thread:issue:61223,issue #61223: optimize_for_inference does not handle MKLDNN for BatchNorm1d,high-priority/blocker,52,high-priority/blocker,1,0,61223,,,,0,af0e5d07fbda8aac087a8137cea7b972f79a19aa8036df8a2cc9cefcf7d97769
thread:issue:63749,issue #63749: Segmentation fault when using C++/pybind11 module without also importing torch,high-priority/blocker,52,high-priority/blocker,1,0,63749,,,,0,5ad58be25c87d05314aee6bf8f5f5f8ed2b9ebb08d05009a6aa1a3223ff75c6c
thread:issue:64794,issue #64794: linalg.lstsq out variant fails internal assert because it uses non-inplace view op for some inputs,high-priority/blocker,52,high-priority/blocker,1,0,64794,,,,0,25c162c05ae76523f90da131e71dcb545e5d787a802a5446f16357c15ab675c7
thread:issue:65366,issue #65366: Exception thrown in final autograd callback (queue_callback) not caught if not on CPU thread: terminate called after throwing an instance of 'python_error',high-priority/blocker,52,high-priority/blocker,1,0,65366,,,,0,280b9a3e93371dc29201c7592202c644964803c21795ae3e0c6396967317e090
thread:issue:65815,issue #65815: Overflow error in torch.linalg uncaught,high-priority/blocker,52,high-priority/blocker,1,0,65815,,,,0,0fd4a18bfd938ab51ca196593f91a33279785f0ef5acb43efbd2c7b6e20e62b0
thread:issue:66782,"issue #66782: Segmentation Fault when importing Torch, 2021 version",high-priority/blocker,52,high-priority/blocker,1,0,66782,,,,0,e5de6753a2a308ce0233201a1a6f327382900a3a06232d6e632e170266c422e9
thread:issue:67237,issue #67237: [docs] Google finds docs pages that give 404,high-priority/blocker,52,high-priority/blocker,1,0,67237,,,,0,bc7850fad8d1577a5483032be10ee875729587c37aac27a621959335762ec19f
thread:issue:67247,issue #67247: pytorch_linux_xenial_py3_6_gcc5_4_test may timeout during test_multiprocessing_spawn,high-priority/blocker,52,high-priority/blocker,1,0,67247,,,,0,1714542e36e054902a738deb47226026cf31857be63741d45a65bfe78530023c
thread:issue:67797,issue #67797: Conv2d kernel performance regression on CPU since PyTorch 1.9,high-priority/blocker,52,high-priority/blocker,1,0,67797,,,,0,a27aea94a6df289a3574b6b4ef693c8b5c484c54ef5b35329ddb47067135adb2
thread:issue:69512,issue #69512: Inconsistent behavior of cosine_similarity between fp16 and fp32 inputs,high-priority/blocker,52,high-priority/blocker,1,0,69512,,,,0,6aa0e3d2cacbd12916e99b30370e39c62a9ba47e001b4c00fb4b0c88cf008cc8
thread:issue:6998,issue #6998: [pytorch] Not handling python reload properly,high-priority/blocker,52,high-priority/blocker,1,0,6998,,,,0,cfa2808b682a0c1cbc2b10cb61234350959a07095d5e711946ae33575432720e
thread:issue:70108,issue #70108: Derivative for _ctc_loss_backward,high-priority/blocker,52,high-priority/blocker,1,0,70108,,,,0,badcc1542a7674781a1dc85a92be2366ea4b90d8050f4462b57f95d1a2ee5a79
thread:issue:71227,issue #71227: Move torch::deploy tests to their own workflow job,high-priority/blocker,52,high-priority/blocker,1,0,71227,,,,0,4b19b96d2556db588d86c60bdb3e665e1c7259eb33b78bfc94a8a1e3fe1c0cfb
thread:issue:71391,issue #71391: Calling .backward() inside of an LBFGS closure function throws an exception in Libtorch v1.6.0+,high-priority/blocker,52,high-priority/blocker,1,0,71391,,,,0,7a7439e8153b6a3b6dcb34ece47a72568570c4054ccebb8f0528e4b70aae2b2e
thread:issue:72182,issue #72182: test_fn_fwgrad_bwgrad_special_ndtr_cuda_float64 fails,high-priority/blocker,52,high-priority/blocker,1,0,72182,,,,0,e688045dae956e006043e2b45c18cdc32576d4e6a379ea3397526edf9a234e71
thread:issue:72189,issue #72189: test_fn_fwgrad_bwgrad_[trapezoid|trapz]_cuda_complex128 causes CUDA memory exception,high-priority/blocker,52,high-priority/blocker,1,0,72189,,,,0,b979b6a8a9f56f88d4e4993d839c144eb493cfdbaaac24c09814ea03a22cce81
thread:issue:73174,issue #73174: `LayerNorm` triggers INTERNAL ASSERT,high-priority/blocker,52,high-priority/blocker,1,0,73174,,,,0,eba17c73470fcd4ccaf615df4054c8d5f16428e928e3934bc8bb1f61280241fc
thread:issue:73179,issue #73179: Floating point exception in _nnpack_spatial_convolution,high-priority/blocker,52,high-priority/blocker,1,0,73179,,,,0,39d5a58e3045de95c1c383b744586d723871e137455b653542484b8734b59835
thread:issue:73185,issue #73185: Segmentation fault in fractional_max_pool2d,high-priority/blocker,52,high-priority/blocker,1,0,73185,,,,0,8d87aa94709f63fb4256eb87b9fec48eaa02d9ca207f42f45317643ccb31c11e
thread:issue:73186,issue #73186: Segmentation fault in fractional_max_pool3d,high-priority/blocker,52,high-priority/blocker,1,0,73186,,,,0,c0b7c2f1c1d8e535fdad8830182ad21f12a253fda8bcecc60ee5b93602af8056
thread:issue:73349,issue #73349: Large performance difference of loss.backward() between torch-1.9.0 and torch-1.8.0,high-priority/blocker,52,high-priority/blocker,1,0,73349,,,,0,f2b7f3ed83af4880f30c6a26b5e0bb721a0058651d4bb6874a777520244a62ab
thread:issue:73488,issue #73488: After XNNPack update `TestXNNPACKSerDes.test_linear` started to fail,high-priority/blocker,52,high-priority/blocker,1,0,73488,,,,0,12671be6cad5fff7185809f2db5277d4b24878004bfbb8c97803c5643dccd67b
thread:issue:73792,issue #73792: Bug: torch.distributions.mixture_same_distribution._pad_mixture_dimension,high-priority/blocker,52,high-priority/blocker,1,0,73792,,,,0,7c08b0eb6091dabd38d94f652557a749e0307c40555410fe9c44008e2b77328a
thread:issue:74422,issue #74422: Embedding Pytorch in C++ using pybind fails on interpreter shutdown,high-priority/blocker,52,high-priority/blocker,1,0,74422,,,,0,a0134c7a838e4d85aafd4216244cc85f04822c181a91e94a8352db435ec0f7a7
thread:issue:74547,issue #74547: importing open3d before pytorch causes matmul to produce a segfault,high-priority/blocker,52,high-priority/blocker,1,0,74547,,,,0,cbeb904efdc740032f25a32b136b474e3fea1adb737557bd79b4d8b57c1eae1b
thread:issue:76191,"issue #76191: [Feature request] Exclusive prefix sum, `torch.cumsum(input, dim=0, exclusive=True)`",high-priority/blocker,52,high-priority/blocker,1,0,76191,,,,0,0da517cab6cbba990b1ca04b2fac58be9ea2d65138506a518ede59c9c3aa1efb
thread:issue:77284,issue #77284: non-rentrant checkpointing uses same memory as non-checkpointed code,high-priority/blocker,52,high-priority/blocker,1,0,77284,,,,0,35731e0fca1398d3602969418059f782bf94eb79cd7272f7fa62080d0da5ac3d
thread:issue:78829,issue #78829: Certain import order triggers segmentation fault,high-priority/blocker,52,high-priority/blocker,1,0,78829,,,,0,363725b8847fae12052cc79ef2720bbbab345de8b498ed054eccdc41c9b6c805
thread:issue:78924,"issue #78924: If large enough tensor is being cloned, parallel dataloading hangs on M1 Mac",high-priority/blocker,52,high-priority/blocker,1,0,78924,,,,0,fbbaf4b796a420ea7b5445c4746df19d2d1cff10187ccecf9175a6a7a652385b
thread:issue:79014,issue #79014: RuntimeError: cuDNN error: CUDNN_STATUS_INTERNAL_ERROR,high-priority/blocker,52,high-priority/blocker,1,0,79014,,,,0,41af6e7aaa92c3447e7aeb9590edd5ab25b44e1eaa20d9446bd2367179d13846
thread:issue:80012,issue #80012: static builds are broken by MKL_DNN,high-priority/blocker,52,high-priority/blocker,1,0,80012,,,,0,2f2a6453d61cd9891548c124fc81c2257029f4569cceaad4d2713c19c78bab5a
thread:issue:80431,issue #80431: CMake Error: File /opt/pytorch/build_variables.bzl does not exist.,high-priority/blocker,52,high-priority/blocker,1,0,80431,,,,0,a503769b709af25fbfba7b608103446a1f0e152dd417584f5a90e0f051d2f3d2
thread:issue:80488,issue #80488: Negative values still produced by torch.nn.functional.kl_div,high-priority/blocker,52,high-priority/blocker,1,0,80488,,,,0,058d510ecbe814f702244d9571f1b3caf83dba7fbf56ff3d8a7939c097bc5177
thread:issue:80946,issue #80946: torch.nn.functional.linear fails for multi-dimensional bias from torch 1.12,high-priority/blocker,52,high-priority/blocker,1,0,80946,,,,0,73e4b3baae35d2bc45141e8d7f506b0f8f049aa7aabfe5f364177777a4707884
thread:issue:81127,issue #81127: Docker updates cause subsequent builds to fail,high-priority/blocker,52,high-priority/blocker,1,0,81127,,,,0,55f2852065752ce4bde110e4179f690482ae4d290edc60b1a620e1d1a45d7e69
thread:issue:81180,issue #81180: MPS: Placeholder tensor is empty!,high-priority/blocker,52,high-priority/blocker,1,0,81180,,,,0,8a678c04df7fbe37d2a63530a3d21d646608719b26f403e1599383727f924f5d
thread:issue:81195,issue #81195: torch._weight_norm with specified dim returns wrong output,high-priority/blocker,52,high-priority/blocker,1,0,81195,,,,0,b3b370bdc853da18810410ea01eb0895e8fcbc3f36ce3dd5aa5abbfbf4d29d40
thread:issue:82068,issue #82068: upsample_bilinear2d() received an invalid combination of arguments,high-priority/blocker,52,high-priority/blocker,1,0,82068,,,,0,141cd16ebb816c2c7cb4f6c82a242d305a525fe95fe57f5219da16d9edf6735a
thread:issue:82813,issue #82813: functionalize and make_fx are not composable resulting in segfault and cuda error,high-priority/blocker,52,high-priority/blocker,1,0,82813,,,,0,3e5577af99f7f2e8e3b5fad7cea2ec0f2f5d3316199cee57156ee59d3d53bb81
thread:issue:82894,issue #82894: linalg and lu tests fail when run in parallel on linux cuda,high-priority/blocker,52,high-priority/blocker,1,0,82894,,,,0,51a592d3d8c7c8b800b7c4844c6dfa2c7124668c0b59da5a19a192017c2baf8b
thread:issue:83070,issue #83070: make_fx + aot_autograd segfaults,high-priority/blocker,52,high-priority/blocker,1,0,83070,,,,0,4e2635fb535cbfc0baa9996f77f9dd9e9914e25b4574fc98cfbd1688d003d413
thread:issue:83243,issue #83243: RPC: wait method of Future object return 0 sometimes in rpc framework,high-priority/blocker,52,high-priority/blocker,1,0,83243,,,,0,cba02ea7d0e7a2eeffd498c84cfc43f97546863afcd491d7896127634868736d
thread:issue:83393,issue #83393: Test public bindings in CI gives weird output on error,high-priority/blocker,52,high-priority/blocker,1,0,83393,,,,0,a3e44ab52427bca8cf6a6860c189159f91eb0a5f758417e6d7f8561947308f6d
thread:issue:83577,issue #83577: libtorch malloc cause coredump,high-priority/blocker,52,high-priority/blocker,1,0,83577,,,,0,5cae1fd9ed4a730335dce1a66c3cc330c56f6b60034d8430a72fb832953bac0d
thread:issue:84051,issue #84051: model.load_state_dict won't work in Child process if a sufficiently large tensor was padded in the Parent (even if empty padded),high-priority/blocker,52,high-priority/blocker,1,0,84051,,,,0,e7e702556b6ba92a419759285424eabbf3d96cb8477a792329b2424cce8a50ba
thread:issue:84181,issue #84181: Session of Google Colab crashes when `torch.utils::SummaryWriter` is called after importing `torchaudio`,high-priority/blocker,52,high-priority/blocker,1,0,84181,,,,0,b82243fd67eed92e7890e921ef7f9c428ac10371eb29438a1810e938cc0e1158
thread:issue:84990,issue #84990: Segmentation fault in `torch.futures.collect_all`,high-priority/blocker,52,high-priority/blocker,1,0,84990,,,,0,0a168168750590bab8f9c7fcce71fa608dbb3170cd633c114f9af95d57d03c29
thread:issue:85214,issue #85214: Segmentation fault in mkldnn_reorder_conv2d_weight and mkldnn_reorder_conv3d_weight,high-priority/blocker,52,high-priority/blocker,1,0,85214,,,,0,a329c16a79991e7044c513400b151f38e3e36ce83e2c65e4bcd0fba0e0b51095
thread:issue:85215,issue #85215: Floating point exception in gather gradient computation.,high-priority/blocker,52,high-priority/blocker,1,0,85215,,,,0,e042b88326730fc173d380d5f7ed60ac5d6d921d70a376713dc62e7bae92ac42
thread:issue:85217,issue #85217: Segmentation fault in native_batch_norm,high-priority/blocker,52,high-priority/blocker,1,0,85217,,,,0,46105d7dee19fe933d43772500a18ecf3588c4d6e8ef9798ed2cebc4d7b62cad
thread:issue:85909,issue #85909: Steam Deck Core Dump,high-priority/blocker,52,high-priority/blocker,1,0,85909,,,,0,0b6899781d6adbf2ba7e6274800201270f7f9a1c1859f66ff33800b399147aa5
thread:issue:86120,issue #86120: TransformerEncoder src_key_padding_mask does not work in eval(),high-priority/blocker,52,high-priority/blocker,1,0,86120,,,,0,1c96144cdadd4a54614ff9e025619219e24517b5ef65ebe37a85ce31e1a0ce31
thread:issue:86704,issue #86704: Performance tests mnist_hogwild-cpu_memory CPU memory increase by 30%,high-priority/blocker,52,high-priority/blocker,1,0,86704,,,,0,c36fcb46cf1c656c1a8ba7f98d75b398de4f78f74b073b64eead8df4acd5bfc5
thread:issue:87283,issue #87283: Missing string parsing for some parameter types in python arg parsing logic,high-priority/blocker,52,high-priority/blocker,1,0,87283,,,,0,76a81b43cd6be0a90416ef8f614bb5270cc19ed4f3af566a399bf1cd46e4b0ee
thread:issue:88036,issue #88036: A segment fault can be triggered in fbgemm_pack_gemm_matrix_fp16,high-priority/blocker,52,high-priority/blocker,1,0,88036,,,,0,2129177f4ffea6b9f58bab92f02111e410fec551c0fc6f66dd1215834fd24dec
thread:issue:88842,issue #88842: test/test_ops.py is segfaulting on master build with DEBUG assets,high-priority/blocker,52,high-priority/blocker,1,0,88842,,,,0,8d3270a905c9644de54858fa7e085de34e22d0396607151fbd8969128610001c
thread:issue:88882,issue #88882: Can't import torch --> OSError related to libcublasLt.so.11,high-priority/blocker,52,high-priority/blocker,1,0,88882,,,,0,0c86235499e32e685c93ce8090754f493635aa749c89d804e52aa4272cee2139
thread:issue:89006,issue #89006: Add smoke-tests for CPP extensions compilations,high-priority/blocker,52,high-priority/blocker,1,0,89006,,,,0,c31ad77eb65930076ed585b5c1f1bce97787f37447ffd601e3bc61877999943c
thread:issue:89718,issue #89718: `torch.Tensor.flatten` Trigger Segmentation Fault when trying to provide and output named dim,high-priority/blocker,52,high-priority/blocker,1,0,89718,,,,0,9d17c81135be658b448d09d15ceedce0f1156a2ebd605f4f8bd3610498cc410f
thread:issue:90535,issue #90535: Illegal hardware instruction using torch.nn.Conv2d on aarch64 (Raspberry Pi 4),high-priority/blocker,52,high-priority/blocker,1,0,90535,,,,0,751c3f8e21fcc36300fa98578c454f46665f4d9f5d13ef1bde4fa7033f8f78bb
thread:issue:90676,issue #90676: RuntimeError: Error in dlopen: libnvJitLink.so.12: cannot open shared object file: No such file or directory,high-priority/blocker,52,high-priority/blocker,1,0,90676,,,,0,2334298bf8e56c75794c770e94730bec4991948b6254de957be3a5a7a449125a
thread:issue:91590,issue #91590: Crashes of linalg.ldl_solve on different edge cases not coming from linalg.ldl_factor,high-priority/blocker,52,high-priority/blocker,1,0,91590,,,,0,f678b44edabd7078ddfe40af04449e1e541b9b615f886528c6c67d2dbee3e0fc
thread:issue:91754,issue #91754: CUDA error: initialization error,high-priority/blocker,52,high-priority/blocker,1,0,91754,,,,0,2e0010be93aa3f4c6068267b80c001c04675d6987e405b4a2a060a3fbfea6df3
thread:issue:92033,issue #92033: Unable to use vmap atop torch.distribution functionality,high-priority/blocker,52,high-priority/blocker,1,0,92033,,,,0,61d9364f53c66c88a5588fb95246c49f874507188dc776c359fa686f2bd3179b
thread:issue:92776,issue #92776: segfault when running torch.lu_unpack,high-priority/blocker,52,high-priority/blocker,1,0,92776,,,,0,f2990e7803b058d3c7911d156270c665bd437ce286835b8dd8a1db3e122d4e03
thread:issue:93846,issue #93846: large number of temporary files generated when using dataloader with num_workers>0,high-priority/blocker,52,high-priority/blocker,1,0,93846,,,,0,b641364c0d8d3762822ffa7e06c14c66cf528378b8584f4afdb1f02679439689
thread:issue:94125,issue #94125: A Floating Point Exception can be trigerred in torch._C._nn.slow_conv3d,high-priority/blocker,52,high-priority/blocker,1,0,94125,,,,0,442f135eda07117d5300f0747a7104f2d36d789262f19aa9cb5bbd1c4d84f6ce
thread:issue:94544,issue #94544: Abort Caused by Virtual Function,high-priority/blocker,52,high-priority/blocker,1,0,94544,,,,0,f1ff134d968dc0f76497a4cac352b3f8ce1ea262ecde6d0af15e14ee12acce4f
thread:issue:94669,issue #94669: A segment fault can be triggered in torch.svd,high-priority/blocker,52,high-priority/blocker,1,0,94669,,,,0,0a48278e2bf57640aa17da8028bc79a881ab252242fbeed01b6b1d32095d1579
thread:issue:94696,issue #94696: A segment fault can be triggered in torch.max_pool1d_with_indices,high-priority/blocker,52,high-priority/blocker,1,0,94696,,,,0,acc4e25da644edca1fd3a19d266bd6368ab94da244922482ba705a7af90e5819
thread:issue:94698,issue #94698: A segment fault can be triggered in torch.avg_pool1d,high-priority/blocker,52,high-priority/blocker,1,0,94698,,,,0,16317710ad4012d50204fa08c275503d90d06c09a8f484dbfaf3d0ed4c655ac8
thread:issue:95432,issue #95432: Regression bug in `torch.nn.ReLU6` and `torch.nn.Hardtanh` that `inplace=True` doesn't work in PyTorch 1.10.0~1.13.1,high-priority/blocker,52,high-priority/blocker,1,0,95432,,,,0,28f495f4b63ed02c6279b2853f806cf0c4807e7f3dc04d2f877a642b62dead6b
thread:issue:96056,issue #96056: MPS device throws error for `F.adaptive_avg_pool2d`,high-priority/blocker,52,high-priority/blocker,1,0,96056,,,,0,a81b8db1c1635b358366a4cb0100beba35b0b5665328aa5f25cd980049737a9e
thread:issue:96277,issue #96277: A Segment Fault can be triggered in torch.adaptive_max_pool1d with an edge case,high-priority/blocker,52,high-priority/blocker,1,0,96277,,,,0,034deb2d161f2a1394ad7049f34bdd38134db13620e93968e4d860400c9171d6
thread:issue:97086,issue #97086: A crash due to Floating Point Exception can be triggered in torch.index_select,high-priority/blocker,52,high-priority/blocker,1,0,97086,,,,0,d737fa78ba64200c08dd43b0971580823f49df45f33af3c60a6fb03488cbfa51
thread:issue:97109,"issue #97109: ""Adaptive pool MPS: input sizes must be divisible by output sizes"", I keep getting this error even when I try to adjust for size",high-priority/blocker,52,high-priority/blocker,1,0,97109,,,,0,66300f0211798d7f4bb9aa9b8c38c381e57db61b784f529f29efe129ec701692
thread:issue:98136,issue #98136: A Segment Fault can be triggered in torch._grid_sampler_2d_cpu_fallback,high-priority/blocker,52,high-priority/blocker,1,0,98136,,,,0,16aa226433daa93beba3b38e6ade1001a02580b60d721fb8c3dc3c280284c55a
thread:issue:98503,issue #98503: Power VSX vectorization support disabled,high-priority/blocker,52,high-priority/blocker,1,0,98503,,,,0,ca39897f2c2383bda640d1c6c652db03658e8959c27dc8b1e8e09acac2d1c14c
thread:issue:99509,issue #99509: 'Illegal instruction (core dumped)' for gpt-j bf16 generation task using greedy search,high-priority/blocker,52,high-priority/blocker,1,0,99509,,,,0,b4df9bcb97de3bbce178fa316ad4c941335e9fa1a30d5fccfa661ef90cd4d49d
thread:pr:189068:pr:189069:pr:189070:pr:189071,pr #189068: Register nccl-lazy TorchComms backend,draft/noise,51,draft/noise,7,0,,"189068, 189069, 189070, 189071, 189072, 189073, 189074",,,30,c8e281dc6588317d8f6fe3ce9787401838d21df20885399285776580a7763436
thread:pr:187859:pr:188366:pr:188367:pr:188376,pr #187859: [functorch] Add torch._functorch.aot_autograd.compile_to_python,draft/noise,51,draft/noise,5,0,,"187859, 188366, 188367, 188376, 189165",,,20,ed3235a40b88886d096d00e4f6b97661644a77740eba02a6a7b18567e8a1bb85
thread:issue:181647:issue:181648:issue:181649:pr:188720,pr #188720: [ROCm] Enable cholesky_ex via hipSOLVER xgeev (ROCm >= 7.14),draft/noise,51,Review active PRs and linked issues as one maintainer work thread.,4,0,"181647, 181648, 181649",188720,,,3,9e89c696b73c06bc4d2430640c6071ea13d43bb5878982e596816a491191bd79
thread:issue:186350:pr:186209,pr #186209: Unify triton-xpu CD nightly build into 'triton' package with version suffix,draft/noise,51,Review active PRs and linked issues as one maintainer work thread.,2,0,186350,186209,,,1,f102c8236303ffce12841c22a374900aa7d7dcc65b9d10bd187ebace3e2438d7
thread:issue:186837:pr:186835,pr #186835: [RFC]: Auto-select and run Operator Performance Benchmarks from PR Diffs,draft/noise,51,Review active PRs and linked issues as one maintainer work thread.,2,0,186837,186835,,,1,2f35befad8d4681d97f8c06e04f0c49f36e642ead3b156c34a0b0af05c5be7c5
thread:pr:187479:pr:188747,pr #187479: Add Python c10d Backend trampoline,waiting on contributor,51,Wait for contributor update on PR #188747; keep related issue/PR context attached.,2,0,,"187479, 188747",,,1,9acdd3c945af77d11eb94f9ad610be04ee0bee14a6a166e6451dfd02fe21583c
thread:pr:189320:pr:189321,pr #189320: [inductor] Enable nested Inductor configs for invoke_subgraph regions,draft/noise,51,draft/noise,2,0,,"189320, 189321",,,2,d4fc22a12f46affb1942face4bbda01357c597e3df13bb584aa252afb5d81afc
thread:pr:159718,pr #159718: [Don't Review] Test XPU CI,draft/noise,51,draft/noise,1,0,,159718,,,0,a95d815a34a3c4460ac2e9bfdbbece6c327e097b64f390dd8f9a8cd1c080406b
thread:pr:172682,pr #172682: [NOT FOR MERGE] Initial test of features disabled when doing cooperative reduction,draft/noise,51,draft/noise,1,0,,172682,,,0,b68aa5fb51c58c9e1b33cca2ed9a0910b40c36d1545afb73119d7b29582bd9ea
thread:pr:174503,pr #174503: [Draft]Trying to see if cpp modification would work,draft/noise,51,draft/noise,1,0,,174503,,,0,e460aa6ca9d1b257e25a75cb0899f28c599867ea089966a99becf496719e91d9
thread:pr:175486,pr #175486: [WIP][inductor] Layout allocator approach for symm_mem graph inputs,draft/noise,51,draft/noise,1,0,,175486,,,0,34aefc093ebc5aec370a38f9cd94c23dd80723ad52b3e1bf6f8945c66bd3b012
thread:pr:177038,pr #177038: Add cublaslt backend for scaled grouped GEMM,draft/noise,51,draft/noise,1,0,,177038,,,0,0cca2ef85e44109a65976e952e96dc38c29936737c7d40a571f68201dacd7ea2
thread:pr:177401,pr #177401: Allow non-CL images to go through NEON path in `F.interpolate()` for bilinear and bicubic modes when antialias=True.,draft/noise,51,draft/noise,1,0,,177401,,,0,1b19d2201e4195c211c4f5d280d3ca1e8e2ee47df5a1a768279ae55226a7d4bf
thread:pr:177438,pr #177438: Infer device_type in fork_rng from devices argument,draft/noise,51,draft/noise,1,0,,177438,,,0,c80576e6c784da2fd154eb68bb0b932f4cd6727d7031092d18fba2cc6d87c904
thread:pr:177445,pr #177445: Add SymInt and SymFloat to Number,draft/noise,51,draft/noise,1,0,,177445,,,0,6cfaa85550f7a23196a7b9ea896c7430a96ac43cee96874d4a474a474fe42245
thread:pr:177471,pr #177471: tag on ops,draft/noise,51,draft/noise,1,0,,177471,,,0,057342b1b6d8956a00a8b83ec3c36d1e789c1b5ad6fbc7c5cf409e63c55e72b8
thread:pr:177612,pr #177612: [inductor] Add pattern match for fused gate+up GEMM + SiLU on XPU,draft/noise,51,draft/noise,1,0,,177612,,,0,9daaeb32a8e2cfdae785f4769d5d06dec5be13b8e0ee24d5d0c564d8a070a501
thread:pr:177648,pr #177648: LinAlgUtils -- update the toolkit to address the MathLibs dispatch with redundant memory allocations/movements.,draft/noise,51,draft/noise,1,0,,177648,,,0,5a6dfec994bbb78c8aae938144ed1cf0b220bcc3c22021bf09b33108d0323da3
thread:pr:178277,pr #178277: Skip fake autocast tests for `linalg.pinv` and `pinverse` on XPU,draft/noise,51,draft/noise,1,0,,178277,,,0,c48f61a8fc0615d2e6c72959eef7adbf91bc915825d99cfe4d5754dde252fb18
thread:pr:178485,pr #178485: [DO NOT MERGE][Inductor] Migrate oneDNN Inner Product to oneDNN MatMul for MKLDNN linear ops,draft/noise,51,draft/noise,1,0,,178485,,,0,e05c729d9efc2097b7e5a05fc748030419e1669260a5008c4822fac22fdd2d96
thread:pr:178862,pr #178862: [FSDP] enable async H2D prefetching for world_size=1,draft/noise,51,draft/noise,1,0,,178862,,,0,4fc8e3b88983914391df99a137ef2868b18f805966d76f374494216c7619ceba
thread:pr:179149,pr #179149: [Inductor][RFC] Symbolic Analysis of User-Defined Triton Kernels,draft/noise,51,draft/noise,1,0,,179149,,,0,a56e0db10b47b3d4a96181ffeabdbe58f400304aeaf6f6a12243f307dfcd5abb
thread:pr:179675,pr #179675: [WIP]port dynamo cases to Intel GPU,draft/noise,51,draft/noise,1,0,,179675,,,0,4b8fe1565d10286ac33d7ff3a60b447ed2e130caab347c928a0d174423225c0c
thread:pr:179769,pr #179769: [cuda graphs] Integrate kernel annotations into inductor cudagraph trees,draft/noise,51,draft/noise,1,0,,179769,,,0,3639f3203dbf6e593b3208366a163b88b7ae07ba3a736a3567b750987b5e30d3
thread:pr:180186,pr #180186: Try to enable LTO in OSS,draft/noise,51,draft/noise,1,0,,180186,,,0,a859f092d6191f886d466b8e0fd407696b0c1c4939f7d52ae3ba0c6b8268ab1f
thread:pr:180192,pr #180192: [DO NOT REVIEW] fix torchtitan CI error,draft/noise,51,draft/noise,1,0,,180192,,,0,fad0a57a53eb903cc5236594ee8bd3fb3b38abb3d55f3e2a6bc8e0274a3b5089
thread:pr:180207,pr #180207: [TEST] Refactor test_structured_trace.py to be device-generic,draft/noise,51,draft/noise,1,0,,180207,,,0,6368608ec9b3de85eb44be891c029ee15737e839b1c09a221518f370f5b39853
thread:pr:180407,pr #180407: [nocommit] symmem nccl ib,draft/noise,51,draft/noise,1,0,,180407,,,0,b9eff2972c6162f60850beb53ef778dc22fdeee3d4b935252152f12a719fca6e
thread:pr:180409,pr #180409: stateless ghstack,draft/noise,51,draft/noise,1,0,,180409,,,0,06a06335ee7d259da3f786bd5a80486c4a18dba19f3b1e0410c13d6bdc13d953
thread:pr:180410,pr #180410: debug,draft/noise,51,draft/noise,1,0,,180410,,,0,d8e28f8ddddb1d67b908f0da9e4a82a3a3febb53be237ff466037ba82e4e3fac
thread:pr:180411,pr #180411: stateless dtensor,draft/noise,51,draft/noise,1,0,,180411,,,0,27995b8d930d34f67350c923470272ce71e6071f2d6a8062f933306882faeebc
thread:pr:180412,pr #180412: stateless spmd,draft/noise,51,draft/noise,1,0,,180412,,,0,c597c96c053f3d7b34598f9c421a3e07ddf2f574682e5ea39130b2dda9432e80
thread:pr:180423,pr #180423: Fix checkpoint DeviceContext leak onto TorchFunction mode stack,draft/noise,51,draft/noise,1,0,,180423,,,0,88e8314e1e4a49b0d825963263b46d8deb82bcbd600ca9af5065aea8cd3d2676
thread:pr:180597,pr #180597: [nocommit] symmem custreamwaitvalue32,draft/noise,51,draft/noise,1,0,,180597,,,0,db251983166071683f0814d6bb7dedac9c0ffaadcc1d3bdd052882f44551a558
thread:pr:180682,pr #180682: AOTI: Only clone mutated buffers in _unlift_graph,draft/noise,51,draft/noise,1,0,,180682,,,0,ec861ab188165c08e2645c96739ba4ee37efa0330878020508e5bc202113b8ec
thread:pr:180683,pr #180683: AOTI: Stream external weights to disk instead of in-memory blob,draft/noise,51,draft/noise,1,0,,180683,,,0,7ed9ce2b28fdd50a7e29a3277f58dc6d4c51c04e2f14288c3e6be032122aa2f6
thread:pr:180684,pr #180684: AOTI: Write tensor data directly to file without Python bytes copy,draft/noise,51,draft/noise,1,0,,180684,,,0,f239ac0074c17929dbcca9696965ae1e4b5a8dd4f61b3702370630b8e4cf407f
thread:pr:180754,pr #180754: [ROCm] Fix decorator for skipping FP8 tests,draft/noise,51,draft/noise,1,0,,180754,,,0,386a7c6032f621c1789b164d063eba0398fcd9054ae346bb2a9f494090d85d21
thread:pr:180866,pr #180866: Add save_tensor() API for selective activation checkpointing,draft/noise,51,draft/noise,1,0,,180866,,,0,4aed79df8aa0a39efa7eee4eedc2f35109ee91fd057ffaef278e8c3c05f15f56
thread:pr:180867,pr #180867: Use SavedTensor for SAC cached tensors to enable saved tensor hooks,draft/noise,51,draft/noise,1,0,,180867,,,0,3bb52102b6c6f343befddf979ba46d871cb10c4009c6487e75ea4ef00cdfff2b
thread:pr:180875,pr #180875: [CUDA][cuBLASLt] torch.addmm - improve numerics of bias fusion with reduced precision SplitK algorithms,draft/noise,51,draft/noise,1,0,,180875,,,0,59464115305ad3a51e7e19b9577d34893ae305b6e21cb9df31bbc574e875c7fc
thread:pr:180924,pr #180924: Remove allow_cache_entry_mutation from SAC,draft/noise,51,draft/noise,1,0,,180924,,,0,65685b37ded777a21ff7f3c3e402f91cdb3021ef9efd8f5c558df2c83094420b
thread:pr:180925,pr #180925: [dynamo] Extract dynamic graph break hints,draft/noise,51,draft/noise,1,0,,180925,,,0,ab6dfd9274a4ec1209a350b8db0bdb8a028c3950c091e4d0d0bcb73064221148
thread:pr:180933,pr #180933: [spmd_types] fully_shard,draft/noise,51,draft/noise,1,0,,180933,,,0,12379b09bbec9084b2b1fdf577de76e1022a0b3a725472e837328695a8f19ce9
thread:pr:180986,pr #180986: Enable nested_compile_region for AOTInductor + tensor-subclass params,draft/noise,51,draft/noise,1,0,,180986,,,0,0855dae6fc0ae517ee37eb9c5fe8756899b211edd2824e15254d5abedb25cdcc
thread:pr:181047,pr #181047: [FSDP2] Introduce StreamHandoff; migrate AR keep-alive,draft/noise,51,draft/noise,1,0,,181047,,,0,6c9b28692cabaf40a82185c35cca8ec10c5e665bcbd72fa4d05ea05c92413f7d
thread:pr:181189,pr #181189: Tensor.record_use: precise cross-stream lifetime for the CUDA caching allocator,draft/noise,51,draft/noise,1,0,,181189,,,0,fc4782061bbab99cefd34294b4013bf244ee10dfa3667a815d1815b35087bb0e
thread:pr:181199,pr #181199: [spmd types] fully_shard in spmd_types,draft/noise,51,draft/noise,1,0,,181199,,,0,4641caf3a46e0bc54c9d08e9d20257d90f0005e9e8599bf81fc56896e53a2ce1
thread:pr:181205,"pr #181205: NOT MERGE, JUST FOR TEST",draft/noise,51,draft/noise,1,0,,181205,,,0,4dafdacc613638e8ca09803894bb2e8d6088126147c12241b319039690338f07
thread:pr:181215,pr #181215: Add cooperative GPU-sharing infrastructure via CUDA process-checkpoint,draft/noise,51,draft/noise,1,0,,181215,,,0,13d77a3790391b731a279ce85a4e599746b32c98a9163f608cf05a2788782575
thread:pr:181218,pr #181218: FSDP2 chunked-loss ROCm race: stream-level wait_event at post_backward exit,draft/noise,51,draft/noise,1,0,,181218,,,0,6a13206dbba4195aff2f749c95b45a5ba661d22d5855d7c06afd591d010bd55e
thread:pr:181269,pr #181269: [draft] refactor linux workflows,draft/noise,51,draft/noise,1,0,,181269,,,0,530d069b496656cc5aca77f2e560b2afee707249adca2057be9b15188389e2d4
thread:pr:181278,pr #181278: [spmd_types][unused] local_map for spmd_types tensors,draft/noise,51,draft/noise,1,0,,181278,,,0,268e276e8ec88150bd77cb733243356d8777792c466fdf513f78eb2d30bbba07
thread:pr:181324,"pr #181324: [dynamo] Remove TypingVariable, trace more of typing.py",draft/noise,51,draft/noise,1,0,,181324,,,0,df4eb54dc01e474f60939cc698878b094e527d23c7ffee0f8d07d75f16a6e908
thread:pr:181392,pr #181392: [AC] Add example_inputs param to remat pass and mark recomputed nodes as backward,draft/noise,51,draft/noise,1,0,,181392,,,0,2b7b64bdad9424bef06a99a737d43b086d3ebcf6d1b2cff111044102d99897d7
thread:pr:181412,pr #181412: Add Claude-powered operator benchmark triage,draft/noise,51,draft/noise,1,0,,181412,,,0,67ab6145783159f7d679bb2e48b127c720e62ee2c851c21913d6241151231fb6
thread:pr:181443,pr #181443: Cublas grouped gemm nvfp4,draft/noise,51,draft/noise,1,0,,181443,,,0,00e4e747de55f34e943d4c33bb4c316505f0bc369a7ba0297214dc31a43e0668
thread:pr:181444,pr #181444: Scaled cublas grouped gemm sm90,draft/noise,51,draft/noise,1,0,,181444,,,0,26a3bea48a33f60569cb7f156bb1e0699497122450cb4b2f890c6f4053c44e8a
thread:pr:181449,pr #181449: AOTI: Add use_fake_constants mode for zero-weight compilation,draft/noise,51,draft/noise,1,0,,181449,,,0,2e260a52bffe363c1732e16ebb404ef7ce99e928740048e7927eac42cc3a57ab
thread:pr:181578,pr #181578: [Draft] [04/27/26] Dashboard result update for ROCm,draft/noise,51,draft/noise,1,0,,181578,,,0,7242417d99ff4cd726c71e5a75b481520f8a0ff4ebac4cc23482e39072005663
thread:pr:181611,pr #181611: [don't merge] test combo kernels,draft/noise,51,draft/noise,1,0,,181611,,,0,a60b6692077c98fdbd403f0cd30900f6857652da2383eb9dae870892d6d973ce
thread:pr:181615,pr #181615: [ROCm] CI: skip operator matmul microbenchmark,draft/noise,51,draft/noise,1,0,,181615,,,0,b11646e72bec01e12ef53178338545ae72b139dca3303cda00949258d709ca04
thread:pr:181627,pr #181627: [ ROCm] [ DRAFT ] 181288 + 177219,draft/noise,51,draft/noise,1,0,,181627,,,0,7a0fc21abffb0817c367951cfe7994bbb97e4e5eb9a30a832735bc14979c159f
thread:pr:181628,pr #181628: [ ROCm] [ DRAFT ] 181288 + 177219 + 181615,draft/noise,51,draft/noise,1,0,,181628,,,0,dfa649da76d3bc59410125802f95973fe6e1bf6cdcea0b01abc164e8e10b19fd
thread:pr:181664,pr #181664: [FSDP2] Design notes: options for RS/AR back-edge in HSDP,draft/noise,51,draft/noise,1,0,,181664,,,0,b5b789ae27c2c4de99b2a91eff9e4f10c71e0143213c9c52ecf0ef3fb9383d4c
thread:pr:181678,pr #181678: [spmd_types] CI CPU job w/o spmd_types,draft/noise,51,draft/noise,1,0,,181678,,,0,c32ca1af45b0ea7aebaf19d3c85f3eb069382960aea56bcbd90f8dd2287fb725
thread:pr:181956,pr #181956: [do-not-review] cudagraph while node support,draft/noise,51,draft/noise,1,0,,181956,,,0,55ffb63802c91abf4e88d128612c9b05915b1eaf9275140780a0c453eaa03d2f
thread:pr:182143,pr #182143: [OSDC] Migrate periodic.yml to OSDC via dial-up pattern,draft/noise,51,draft/noise,1,0,,182143,,,0,9278135a64a8b2fe9f4e97ad1aea148eeb2964636599f269fddae2dd06943194
thread:pr:182226,pr #182226: [CI] Install git-core PPA without add-apt-repository,draft/noise,51,draft/noise,1,0,,182226,,,0,de94e08df176868799136aca24ff1761d752b542211c778eb3ce16d42eff244e
thread:pr:182271,pr #182271: [ci][WIP] Cut redundant before-test fanout + trim over-eager periodic crons,draft/noise,51,draft/noise,1,0,,182271,,,0,96b8aa30c92e23ce5e0cf78d16adc238f2f7d33f647711e4c9daae4918f85e47
thread:pr:182314,pr #182314: [OSDC] Migrate dynamo-unittest.yml to OSDC via dial-up pattern,draft/noise,51,draft/noise,1,0,,182314,,,0,dda9cc2a2f711469e78ccccf4e3b6cbfb88826739ec6ce70962e4fb8f857882c
thread:pr:182334,pr #182334: [inductor][autotuning] pipeline AsyncCompile + CachingAutotuner (#182334),draft/noise,51,draft/noise,1,0,,182334,,,0,33a329cffd16b1c51b10f208fcb2818c2319a790d75f4c6e06e44f930a654287
thread:pr:182387,pr #182387: Allocator-level memory attribution primitives,draft/noise,51,draft/noise,1,0,,182387,,,0,ad13511d9cf0ac272db3608cec1fd1499df553a0fe58016b2f2bef9ae9f1d0dd
thread:pr:182388,pr #182388: Training hooks and public API for memory component tracking,draft/noise,51,draft/noise,1,0,,182388,,,0,3309f64f30ce6d3021b883c4155c0321337130f754c14363a7e2c438674af7de
thread:pr:182389,pr #182389: Collective buffer detection for DDP and FSDP,draft/noise,51,draft/noise,1,0,,182389,,,0,63e9a739f0bab6dde36ded5d33462f35ae8008769b2e4325394a81e34ca4a1aa
thread:pr:182415,pr #182415: [c10d] Apply record_comm names to NCCL launch annotations,draft/noise,51,draft/noise,1,0,,182415,,,0,452663a39d86fe3abb4346968180bcf32ba4543b9631b38a078837c08f375436
thread:pr:182631,pr #182631: [aten] Add aten._foreach_mm operator,draft/noise,51,draft/noise,1,0,,182631,,,0,66c12645b4c43a09a29f94035528e85469b534acebddad2c33171966a0db714b
thread:pr:182691,pr #182691: Parameterized cuda graph launch 3,draft/noise,51,draft/noise,1,0,,182691,,,0,66146f09d11396aecf7ea2277042765e9ece3dd19fd7df78e76dfb700656d559
thread:pr:182810,pr #182810: [ROCm] Implement TORCH_HIP_VERSION,draft/noise,51,draft/noise,1,0,,182810,,,0,4196fa735c1403036fc6f758c54b5c515ac6c3b85f4d2b9b1dfe4c01cf294194
thread:pr:182844,pr #182844: [CI] Pin miniforge to 26.1.1-3 to fix docker builds,draft/noise,51,draft/noise,1,0,,182844,,,0,9134b80d050715dd778a883cfb25595092594a7d4628a30abc5eab9fba73d6e7
thread:pr:182864,pr #182864: [donotrvw][broken][WIP] spmd_types + PP: _SpmdTypesMeta and nullcontext decorator fix,draft/noise,51,draft/noise,1,0,,182864,,,0,fda55d8ab792bcda94c98d66dcbc97f53dc63084abaf402dba0ded8c526e0237
thread:pr:183270,pr #183270: [inductor][overlap] Post-scheduling foreach_mm for optimizer mm batching,draft/noise,51,draft/noise,1,0,,183270,,,0,6754d58ca6b43b675d8119bc2bcd90c0c3bdf4797fa23461aa60540a50396cc2
thread:pr:183579,pr #183579: [OpenReg][Quantization] Register dequantize_tensor_per_tensor_affine stub for PrivateUse1,draft/noise,51,draft/noise,1,0,,183579,,,0,2df5ba2384e35672e7fedcb0a6dcebf409e808b04640efdf4d5832d58450df67
thread:pr:183613,pr #183613: [CD] Tighten ROCm manywheel build pipeline,draft/noise,51,draft/noise,1,0,,183613,,,0,3d02e1993a2f6be0cb6e1b2924cc787b0f41e5d6b7033dc7b19c71d5a9513016
thread:pr:183626,pr #183626: [Inductor] Seed combo-kernel autotune from standalone subkernels,draft/noise,51,draft/noise,1,0,,183626,,,0,b46b396e30e750dc95707b6bc7d8e6ac47bd8c5743a8ff7e44f8cfb30147771b
thread:pr:184302,pr #184302: Add Stream-K MM backend support to Inductor,draft/noise,51,draft/noise,1,0,,184302,,,0,708f83918ec6a04743bf5e6bb7671ba45f3a69bb26e9f44c78bdd412755906a5
thread:pr:184886,pr #184886: [inductor] Diagonal-symmetric mm TritonTemplate for Gram matrices,draft/noise,51,draft/noise,1,0,,184886,,,0,2a29ef270f5023d9934e4c487b29ca196f2d22d9b1ccdf5b9a9ca687c00fa73f
thread:pr:184892,pr #184892: xpu: use stride-based conv descriptors and preserve grouped layout mapping,draft/noise,51,draft/noise,1,0,,184892,,,0,bf17afc874a6caf29d9ce6a5a00dd4ecedcbd4e7927d0dbae884ae10e7f1f226
thread:pr:184909,pr #184909: [Inductor][Draft] A Helion backend for Inductor,draft/noise,51,draft/noise,1,0,,184909,,,0,22a6bd3506c18f19ad22d93ee6aeaa0d40e2cb4e0f0c89b028d7da1596ab621d
thread:pr:184991,pr #184991: [inductor] Add memory-aware graph-level CSE pass,draft/noise,51,draft/noise,1,0,,184991,,,0,74ab77ea63372484eb9ac79bb31a4cbb309e9b210f0ae19045518dc977c466e9
thread:pr:185172,pr #185172: Add support to custom devices to operator_benchmark.,draft/noise,51,draft/noise,1,0,,185172,,,0,6741f2ac522a4851c5ba5bb923d43dc570ad77b71fc549935c34903db74b23d0
thread:pr:185218,pr #185218: [not-for-land][squash] Regional inductor cache fix,draft/noise,51,draft/noise,1,0,,185218,,,0,0a1d075b700dedb147cbe19fd4ddfea5ecf4a6deeef0e40e914c0c921401f300
thread:pr:185219,pr #185219: [cherry-pick] no topological sort in manual overlaps,draft/noise,51,draft/noise,1,0,,185219,,,0,4870233b5a2d2d463288dee09b2a5afc21e82fab6278baffa49a40165fac5955
thread:pr:185253,pr #185253: [c10d] Make FakeProcessGroup reduce collectives numerically consistent,draft/noise,51,draft/noise,1,0,,185253,,,0,f05e8c797a7a6296d8103b30225aff634e733c78728496d6303a7fce3803fffe
thread:pr:185258,pr #185258: [ROCm][DO NOT MERGE][DEBUGGING] MI355 hang reproduce and dump subprocess logs,draft/noise,51,draft/noise,1,0,,185258,,,0,1a3026a572eaf86c1b25accecfa2fec159456faf78b6f78384e1869f6ff2c2ec
thread:pr:185284,pr #185284: Add checkpoint_name() API for explicit tensor naming in SAC,draft/noise,51,draft/noise,1,0,,185284,,,0,7ff2c982c4f714242dd9f7643026a29300d50ca2205f85bd1fdaa6fb577ef389
thread:pr:185285,pr #185285: Allow debug mode with context_fn in checkpoint(),draft/noise,51,draft/noise,1,0,,185285,,,0,465d16312df578b89f22080a1ba87b402e373293fa72ed95c31cbba63ae4be03
thread:pr:185286,pr #185286: Add version counter check for gradient tensors in the autograd engine,draft/noise,51,draft/noise,1,0,,185286,,,0,59eee5c4495a28a6dee3968e4ede672275f4cc2498b28b9652ad2376f26791e9
thread:pr:185287,pr #185287: Add memory_budget() context manager for per-region activation memory budgets,draft/noise,51,draft/noise,1,0,,185287,,,0,c0e231e74ea1eeaf955d0e9f67266bd53a065e9f4d8caf77f7f08dca1f76be9e
thread:pr:185339,pr #185339: [not-for-land][aten] Add cuBLAS/cublasLt grouped GEMM backends for _foreach_mm,draft/noise,51,draft/noise,1,0,,185339,,,0,37b412fd877a3fce82e7ae05d23c519694882fb0d4ad258c0884df531431aa16
thread:pr:185340,pr #185340: [not-for-land][native_dsl] Add Python _foreach_mm override via _grouped_mm,draft/noise,51,draft/noise,1,0,,185340,,,0,f69d96090a9e1759e05f5b5e486be857efef5b41fc00cfb07a83a361a82246ae
thread:pr:185363,pr #185363: [not-for-land][native_dsl] Python _foreach_mm override via stack + _grouped_mm,draft/noise,51,draft/noise,1,0,,185363,,,0,d20185e47b084515c89f98894277879219c909871a69b4a8c7a7f35e41eaa707
thread:pr:185364,pr #185364: [not-for-land][native_dsl] Add _grouped_mm_from_ptrs pointer-packing path,draft/noise,51,draft/noise,1,0,,185364,,,0,501858867ea8d2833e2f21b0675d6776401b4533ca57736e35f13d532db0309e
thread:pr:185376,pr #185376: [not-for-land][native_dsl] Add _foreach_mm_from_ptrs pointer-bridge and py_ptrs override,draft/noise,51,draft/noise,1,0,,185376,,,0,26a237c297577f5b5dedf023add6e25ceba6a30018e1f2bcdf1c35c308b21c11
thread:pr:185457,pr #185457: [Inductor][xpu] Enable grouped_mm Triton template on XPU with TMA disabled,draft/noise,51,draft/noise,1,0,,185457,,,0,fa6500a7aa31f083276251d4b280107709603052457d62d12c34213637351219
thread:pr:185548,pr #185548: [test] Migrate in-test OpInfo skip to OpInfo skip DB,draft/noise,51,draft/noise,1,0,,185548,,,0,e3ae4e4c25a1a3805aafeb4ac2750249a31cb527684e415b268eb3a825485e78
thread:pr:185555,pr #185555: Add torch._rust Rust extension and build plumbing (#185555),draft/noise,51,draft/noise,1,0,,185555,,,0,c018fa8ff2193133202b49a65ea5dd0950b614dee2b6355e5852614ecff941e7
thread:pr:185565,pr #185565: upstream torch.comms,draft/noise,51,draft/noise,1,0,,185565,,,0,d4b7be4bb9ea290a9dd36ba1ffcb2cce679566d66a5aa8f6651bead45a552acb
thread:pr:185612,pr #185612: [WIP][CI][CRCR] Update mergebot rule to adapt CRCR L3,draft/noise,51,draft/noise,1,0,,185612,,,0,cbd67e04e8eab32c1f06d372f55e03a645b61e5f77ad2446ef1fa68d57a30e8c
thread:pr:185705,"pr #185705: Fix typos in comments, docstrings, and error messages",draft/noise,51,draft/noise,1,0,,185705,,,0,6d36b764b7cb11883a01c790bd852a0defaeced0a4900dd86bf68fafbd9fb152
thread:pr:185706,pr #185706: Fix typos across comments and docstrings in torch,draft/noise,51,draft/noise,1,0,,185706,,,0,b9cdb3b9110597ca63b3b1cf11d922c15ddc8a9b723619bbbf4dfd6bd6aa2812
thread:pr:185712,"pr #185712: Fix typos in comments, docstrings, and strings across the codebase",draft/noise,51,draft/noise,1,0,,185712,,,0,c563458842f08035cd70d7e77a3217f1d3baed9375e093e9385e126adb235525
thread:pr:185713,"pr #185713: Fix typos in comments, docstrings, and error messages",draft/noise,51,draft/noise,1,0,,185713,,,0,26ae9157b57f5230c20b0bd9a72bbae223ff872df7aa36eeb9d46ce809525c5c
thread:pr:185787,pr #185787: [WIP] Migrate XPU test runners to linux.xpu.test and disable legacy jobs,draft/noise,51,draft/noise,1,0,,185787,,,0,bca763ec7e949b4c6c91ce6a186662d1166654d15b612f8ff57bad173ed45505
thread:pr:185822,pr #185822: [DO NOT MERGE] ci: add manywheel build telemetry (Feb baseline 0cd9ecaf),draft/noise,51,draft/noise,1,0,,185822,,,0,afb979e69e07a6c1d4ab02b5dd6bee3e5d50d2d90e01434ccf193f404b6593fe
thread:pr:185835,pr #185835: [c10d] Add delegate backend to FakeProcessGroup for numerics testing,draft/noise,51,draft/noise,1,0,,185835,,,0,896c51f373f11e3ee72488da09ca6f3380aae7fffddddcf47c605ac789220b7b
thread:pr:185935,pr #185935: [DO NOT MERGE] ci: add manywheel build telemetry (May 6f7bb01c),draft/noise,51,draft/noise,1,0,,185935,,,0,0013144e1a4362905e4fcbe93e54430a94dd635bb77337306af1c87f7bec6556
thread:pr:185936,pr #185936: [DO NOT MERGE] ci: add manywheel build telemetry (Apr midpoint 977c5623),draft/noise,51,draft/noise,1,0,,185936,,,0,77ffc561626f71027afd5d45224dfd64402293ec03fb30a26e869dffb37fc757
thread:pr:186002,"pr #186002: Temporary Commit at 6/2/2026, 2:55:48 PM",draft/noise,51,draft/noise,1,0,,186002,,,0,c22f14ce321b1e442bfc4385c0ea5b8735a3a27b0a01fd5ef64dec660efdd7bb
thread:pr:186033,pr #186033: [CD] Enable Python 3.15 & 3.15t Windows wheel builds and Triton XPU Windows build,draft/noise,51,draft/noise,1,0,,186033,,,0,cc4ed0e2880303f44dfb94c2e64b42fb406a653f519c6846829bbdead52359f8
thread:pr:186271,pr #186271: [CUDA] Add TunableOp support for scaled GEMM,draft/noise,51,draft/noise,1,0,,186271,,,0,a07f971faa94f2ef5a75669369bfb9991763b70ff00de303239c6e7543fe5680
thread:pr:186277,pr #186277: [activation_checkpointing] Add policy= kwarg and name() to torch.utils.checkpoint,draft/noise,51,draft/noise,1,0,,186277,,,0,d09555057377100c19424e1e5348fe6b1ab37fc3908e75fba6ba18ff655da564
thread:pr:186343,pr #186343: Enable FA build for SDPA on Windows+CUDA,draft/noise,51,draft/noise,1,0,,186343,,,0,853b5e94f02d5a8c32837daa0b38e40e5e24dee56d096a0f35cbf42aa7b5106d
thread:pr:186378,pr #186378: [activation_checkpointing] Support AutoNamingMode + policy name keys under torch.compile,draft/noise,51,draft/noise,1,0,,186378,,,0,1fa66c69ce22bd2d1bb593083e29b757c7b9688940fd0939c781fa45a94e8774
thread:pr:186382,pr #186382: [activation_checkpointing] AutoNamingMode + name/module-FQN keys in checkpoint policy= (eager),draft/noise,51,draft/noise,1,0,,186382,,,0,2d7d895c140f4696aff7132d27c57736cba6839fb84c2de3546234ddc85955ad
thread:pr:186541,pr #186541: [MPS] [WIP] Port all MPSGraph matmuls to Metal,draft/noise,51,draft/noise,1,0,,186541,,,0,948abcb97ff578d88c670c24d79028a96e8c679f6743105d450c04298db7827a
thread:pr:186590,pr #186590: [Draft][xpu] Add NVFP4 support on Intel GPU,draft/noise,51,draft/noise,1,0,,186590,,,0,4fd25dc9f4695773485af2e3fcba2af088818f498cbea381bda68b556192beb3
thread:pr:186752,pr #186752: [Inductor][Triton] Add host-side TMA support for B200 GEMM templates,draft/noise,51,draft/noise,1,0,,186752,,,0,79a5f600f08bc78d0779dc118644f3e22f1a28baad448e548d377621c099aa02
thread:pr:186753,pr #186753: [fx] Promote raw-SymInt arg warning to RuntimeError,draft/noise,51,draft/noise,1,0,,186753,,,0,de5801f4a77b6d486bfba3bd6189f1f598824cec7806d15faa1396ad6cffc603
thread:pr:186756,pr #186756: [Inductor][Host-side TMA GEMM] Enable host-side TMA for TMA epilogue store,draft/noise,51,draft/noise,1,0,,186756,,,0,8ce3c97b673f61822bdc98c6a5ca0a906909e16437c4433eb0065c1edc34be25
thread:pr:186798,pr #186798: [TESTING ONLY] [ROCm] testing heuristic config approach,draft/noise,51,draft/noise,1,0,,186798,,,0,196f6c29a62304bd29a01a652d1b86df4800fc8ce3bcb7edc5becc7a602597e3
thread:pr:186945,pr #186945: Enable nested Inductor configs for invoke_subgraph regions,draft/noise,51,draft/noise,1,0,,186945,,,0,2c646e3451adbf319feeccd7e8fa1de44f4702e7c4c828b20f213e6005e6fc1d
thread:pr:186984,pr #186984: Move header-only c10/util container/storage utils to torch/headeronly,draft/noise,51,draft/noise,1,0,,186984,,,0,20682134e23cb0d4c948d55b143ae6c1bbcbd5f230ba6541cc1c54daa35052e8
thread:pr:186986,pr #186986: Move header-only c10/util string/filesystem utils to torch/headeronly,draft/noise,51,draft/noise,1,0,,186986,,,0,1dcc1a2bdf590fb821df030d358768be11bbcf08f4291ebdb80011886fa2c754
thread:pr:186987,pr #186987: Move header-only c10/util large/vendored headers to torch/headeronly,draft/noise,51,draft/noise,1,0,,186987,,,0,f6bcd21965cab55394728380af82010cb94d18f8cd04fe2858d2a6797dfeab0e
thread:pr:186988,pr #186988: Move header-only c10/util Layer 1 utils to torch/headeronly,draft/noise,51,draft/noise,1,0,,186988,,,0,d3378367f79f190d760f873b3f31f357140df204c40534b5dc4b526bd86166d4
thread:pr:187004,pr #187004: perf(dynamo): add guard lookup fast paths,draft/noise,51,draft/noise,1,0,,187004,,,0,ab549f9bea4462e89b92ac15ccd982cf727dfa229ebd3b43244fd067066cadf8
thread:pr:187056,pr #187056: [TESTING] Separated ROCm triton commit pin,draft/noise,51,draft/noise,1,0,,187056,,,0,3d237bbb4748f602935a191020ae953c77ae060f9e237031eac42cbcafc2297b
thread:pr:187116,pr #187116: Add stable tie handling to topk,draft/noise,51,draft/noise,1,0,,187116,,,0,1f4125f6e3bf7c62edfb4568330c661abc2bd71e6c3bc43323e4b1770456df41
thread:pr:187206,pr #187206: [WIP][no-ci] auto_custom_op,draft/noise,51,draft/noise,1,0,,187206,,,0,0c483502b3587448375613e1009fe9807eebdfc2b465ec1512d4915edd5f92d1
thread:pr:187249,pr #187249: perf(dynamo): add guard partial memo,draft/noise,51,draft/noise,1,0,,187249,,,0,092629f9f67357d13401d28c6dfe67b5063d8337bbfbb05de73bae8c3d6eb9cf
thread:pr:187258,pr #187258: [aot_compile] Add torch.compiler.precompile,draft/noise,51,draft/noise,1,0,,187258,,,0,37cdbb170b07718687bbcd083a46247d35e0384ccd05f3708d87b8e11b2560b1
thread:pr:187435,pr #187435: [inductor] Support control_deps as fusion barriers,draft/noise,51,draft/noise,1,0,,187435,,,0,5ae69dbc5c3bcc6951881b80c2802e69cc160732b7c83dfce11a2222a52ac4d7
thread:pr:187464,pr #187464: [ROCm] [Operator Benchmark] Add gfx950 ROCm nightly CI lane and docker image,draft/noise,51,draft/noise,1,0,,187464,,,0,d6142f82351544a3b6e80bf9f2b96665954528e5649d2bd2a52621036e41b7ef
thread:pr:187472,pr #187472: [dynamo] don't reconstruct live torch.Stream values as StreamContext across graph breaks,draft/noise,51,draft/noise,1,0,,187472,,,0,23db5478cdd872090b60cbb3c8c004f0a13897abaf6527d96048afb5b8f2af22
thread:pr:187606,"pr #187606: [inductor] Per-arch ROCm channels_last layout-opt multipliers (gfx942, gfx950)",draft/noise,51,draft/noise,1,0,,187606,,,0,816cccecd73c97a089b4ab0d1e8911e46aba8f257bafa775fa511b9c813d4182
thread:pr:187624,pr #187624: [DRAFT]codeowner updates finer grained groups,draft/noise,51,draft/noise,1,0,,187624,,,0,97093c591e1fa43b54894cbfb6f6f35d07b3614f9e878e81521cce94b87eb79e
thread:pr:187636,pr #187636: [AOTI] Flip the default value of autotune_at_compile_time to False,draft/noise,51,draft/noise,1,0,,187636,,,0,eecf676d8d7aa3d1e99ecc9ad7092210fb69ccb1cfdc5344378d14f984056716
thread:pr:187640,pr #187640: Support in-tree XNNPACK build on Windows ARM64,draft/noise,51,draft/noise,1,0,,187640,,,0,9bfa0a445bc1bc0727340f4bf4fd69ea8d5ed859491469e36afa165c7dc2f726
thread:pr:187650,pr #187650: [Pytorch Test] Add skip_if_lt_x_devices and migrate FSDP autograd tests to be device-agnostic,draft/noise,51,draft/noise,1,0,,187650,,,0,90974ab610cddf88fdcc0f4d2602fb49c116f2046b255d7dbf02a57e1faebfea
thread:pr:187674,pr #187674: Preserve boxed lifetimes for regional compiled calls,draft/noise,51,draft/noise,1,0,,187674,,,0,bc1c493d7e55bda72e91d1c9645e60c7ce6fe8b0f54fc67dcf62d2d0c9a9a685
thread:pr:187739,pr #187739: [ROCm][CI] Misc updates,draft/noise,51,draft/noise,1,0,,187739,,,0,c7828a6a0da8970270a2e4cb2f9fe864ccc22b4c3c15965fed049c5f4695f841
thread:pr:187756,pr #187756: [testintro][DEMO - DO NOT LAND] exercise the test-diff CI comment,draft/noise,51,draft/noise,1,0,,187756,,,0,14ab1ad797e001917e48a54619c5475d236e31c81a6327bdf321c7459e06ddb5
thread:pr:187815,pr #187815: [Inductor][XPU] Unfuse addmm for all half-dtype inputs on XPU,draft/noise,51,draft/noise,1,0,,187815,,,0,41b91d75e71d03dcc0efd1cfdac9ac7c6db91207b0192f37604b5560bdca9fc5
thread:pr:187818,pr #187818: [WIP] generalize unitttests,draft/noise,51,draft/noise,1,0,,187818,,,0,ee708506bdbc06c5c380dbd0d2eaefd92977e112a1d6b1bd821251ae0fbbf4b2
thread:pr:187839,pr #187839: [inductor] Resolve ProcessGroup attrs in overlap estimates,draft/noise,51,draft/noise,1,0,,187839,,,0,f6e66e4efd4f92769544397843e19c8e0868a9f4ff5c1894a753a067abd812f0
thread:pr:187840,pr #187840: [inductor] Add allow_peak_memory_increasing_fusion config,draft/noise,51,draft/noise,1,0,,187840,,,0,497b6f081a7cab75935ece89b387486ef596a117d47c34a534bd18f4cd81a0d9
thread:pr:187841,pr #187841: [inductor] Avoid disjoint horizontal fusion when allow_peak_memory_increasing_fusion is off,draft/noise,51,draft/noise,1,0,,187841,,,0,7a5ff074489039da5c2abfd5acee8519ca25131418451ded36697b1b2c3ea198
thread:pr:187843,pr #187843: [inductor] Avoid late-output vertical fusion when allow_peak_memory_increasing_fusion is off,draft/noise,51,draft/noise,1,0,,187843,,,0,0f2585c9dd21eca3ecc5288196166ea2120f7dbaf2400a0094a8cd1ea49ba050
thread:pr:187844,pr #187844: [inductor] Realize reused pointwise add when allow_peak_memory_increasing_fusion is off,draft/noise,51,draft/noise,1,0,,187844,,,0,dc8848076dc6cd22dbaa0019ea17ba45c6e0040a1b757280f06d385270e7d82b
thread:pr:187845,pr #187845: [inductor] Prefer successor-unblocking LPMF when allow_peak_memory_increasing_fusion is off,draft/noise,51,draft/noise,1,0,,187845,,,0,766e098fbb6921dda6aca3b9f59ef46d62c1fd4142af05f461ef3f497001f8ae
thread:pr:187846,pr #187846: [inductor] Pick low-pressure buffer reuse when allow_peak_memory_increasing_fusion is off,draft/noise,51,draft/noise,1,0,,187846,,,0,7666cc4c3e6a7e8a2ce8e09d1f0837148df8e8bb5cd00300f1db6cb3226ee80b
thread:pr:187870,pr #187870: compile-on-one-rank: device-agnostic inductor codegen and launcher,draft/noise,51,draft/noise,1,0,,187870,,,0,b95a0223a623f7f884942396bf5af01805296ed67cfc385f8c9e5e6db7ee7f7e
thread:pr:187878,pr #187878: Add c10d Watchdog: a libuv-backed timeout service,draft/noise,51,draft/noise,1,0,,187878,,,0,62f99c418ed2b57442d53afceaa6e1cde5eeba4bd6785e685cf4dcb73a844e20
thread:pr:187882,pr #187882: [RFC] Specialize float attributes of nn.Modules,draft/noise,51,draft/noise,1,0,,187882,,,0,ad9c0a20827aaac0a764c400cdc22c033ccacbf11be307b1d5e256dfee9f3210
thread:pr:187892,"pr #187892: [profiler][cupti] Perfetto-native (.pftrace) export, full chrome-args parity",draft/noise,51,draft/noise,1,0,,187892,,,0,4dade59b95ea6abb3e4ecd8363c25c456bedb43a56bae4818312b9f94ca36d54
thread:pr:187900,pr #187900: [XPU] Migrate CI Docker to Intel OMIX for oneAPI + GPU driver installation,draft/noise,51,draft/noise,1,0,,187900,,,0,eb32f92e7d586f51bc3f9780da4a341e5956512ab56c4db6f2559a681326ef26
thread:pr:187909,pr #187909: [DONT MERGE] Test new xpu runners,draft/noise,51,draft/noise,1,0,,187909,,,0,e3da4dda76842b032cf49892bf9f5c35eac730837213fdfc9d66849b8c9af1c4
thread:pr:187971,pr #187971: [inductor] Add strict_numerics config to emit INNER_TREE reduction ordering for sum,draft/noise,51,draft/noise,1,0,,187971,,,0,41f99ee1079f64d29405e657c885af7eb595dfda19524c62d522664d931b14e4
thread:pr:187981,pr #187981: [not for land] CODA backward epilogue fusion prototype,draft/noise,51,draft/noise,1,0,,187981,,,0,23ec7f6ddd3f141fc6f900f7ac839d397df6e5d7974f2f69ff824d1e98380e97
thread:pr:188003,pr #188003: [AMD][hipBLASLt] Enable streak-K work stealing for GEMM operations,draft/noise,51,draft/noise,1,0,,188003,,,0,5d2b56b8e1c66b80651d4e4ae9d6aeea003931abc0a13f6cb04e940f60dfcf97
thread:pr:188016,pr #188016: Update torch-xpu-ops commit pin,draft/noise,51,draft/noise,1,0,,188016,,,0,989ac9daff5cc40e74baf55b263cd51c16fc1c3dbeca82eca490b4a91bc559f4
thread:pr:188119,pr #188119: [inductor] Enable combo kernel compile time autotuning by default,draft/noise,51,draft/noise,1,0,,188119,,,0,7772a928832d0dde4f05070050c5082221baf9f3a3ed34c794b7288e908537a1
thread:pr:188136,pr #188136: [BE] Build all XPU wheels on a single runner,draft/noise,51,draft/noise,1,0,,188136,,,0,0b09f8c8bbfc4e8d6218e1e7fcb65c28ce4dd131bbc18974218b69ebde757bdb
thread:pr:188168,pr #188168: Fix hipBLASLt FP32 precision degradation on MI300X (ROCM-3139),draft/noise,51,draft/noise,1,0,,188168,,,0,5961d0bbd9698ccd96ad9440156627510de38b2c2da0cc8dafc4c04ddae2966f
thread:pr:188184,pr #188184: Update definition of ROCM_VERSION in cpp_extension.py,draft/noise,51,draft/noise,1,0,,188184,,,0,9679fd63e396a3a0dea91bdaa5114db8dc5cd70f39ea0e6c612d9adaf1d230bb
thread:pr:188189,pr #188189: Fix split_group non-member handling regression for mixed backends,draft/noise,51,draft/noise,1,0,,188189,,,0,8766e58da9e60f8ba585565bc9d9dbcab53c82ab22a7c5a83663d882d1ab22a3
thread:pr:188203,pr #188203: [inductor] Fix combo-kernel peak-memory KeyError when reorder_for_peak_memory is disabled,draft/noise,51,draft/noise,1,0,,188203,,,0,8d63555054d6f31a15a3af548c9afa89d8931788926b93a808196631d0d497d8
thread:pr:188236,pr #188236: [inductor]: Include convolution_backward in decide_layout_opt conv node detection,draft/noise,51,draft/noise,1,0,,188236,,,0,fcc23484d01dc1367f762396b36e882a80cf2e07451d9aac1e268eca19c6f6f3
thread:pr:188259,pr #188259: Add XPU caching allocator checkpoint pool state bindings,draft/noise,51,draft/noise,1,0,,188259,,,0,a3d759ecc249f1bd1438291a49f7e0bab2950136967d5e32830c30372b145d0d
thread:pr:188284,pr #188284: [Draft] Fix BUILD_SEPARATE_OPS cyclic dependency — register with xpu_hal.dll,draft/noise,51,draft/noise,1,0,,188284,,,0,67f1513a5f40d21545c58a8079d0920ad208cf6ac935cb36fbc87508f68ed638
thread:pr:188326,pr #188326: feat(dynamo): backport recursive dict tags,draft/noise,51,draft/noise,1,0,,188326,,,0,4bd65a6f6c112875cf7a766924e533d8f5dfc766ba9ebe94fbc487aaf6a8c0b3
thread:pr:188404,pr #188404: Inductor: Add metadata overlap dependency implementation,draft/noise,51,draft/noise,1,0,,188404,,,0,ce40a648ec9bd256ca0a78c5fa868e1e65b0a0216f6dacc91d600a955fd36cf8
thread:pr:188429,pr #188429: [ROCm][CI] Switch rocm-preview to TheRock wheels,draft/noise,51,draft/noise,1,0,,188429,,,0,2fb9aba3b1f4117ae2f8636d74b4c049379005419b87e45ed8e7781ca9df1ef5
thread:pr:188435,pr #188435: [dynamo] Model uninitialized closure cells in __build_class__,draft/noise,51,draft/noise,1,0,,188435,,,0,d2c45149a96607437c2dba3cefc5d2ae9125fcaa136ff31cad9bca0cfc54f44b
thread:pr:188448,pr #188448: Add DTensor sharding strategy for searchsorted,draft/noise,51,draft/noise,1,0,,188448,,,0,d5f3ca5036646dd4c9075d0c38c4b9062a97daf9a4396dbc0f5bc243b838ef60
thread:pr:188463,pr #188463: [PU1] Add output_size parameter to torch.bincount for eager-meta support.,draft/noise,51,draft/noise,1,0,,188463,,,0,1e5852c2b58a0a8748f80a2c5fcb8b7d81882e8697941c4839c550d2031fdd34
thread:pr:188464,pr #188464: [spmd_types] infer R@FSDP axes from active mesh for empty params,draft/noise,51,draft/noise,1,0,,188464,,,0,00fbe751524e85aa2ee40bfc4ffb3939e6c8e77e030adc5dc9b93caa3562eb31
thread:pr:188539,pr #188539: Add support for win_arm64 CUDA builds,draft/noise,51,draft/noise,1,0,,188539,,,0,52f20f53940107caf7453c1d87dc4b22f68b661c61ef98be8537fc6e4b26e3df
thread:pr:188577,pr #188577: Migrate Build Triton Wheel to OSDC,draft/noise,51,draft/noise,1,0,,188577,,,0,e5a1055dab5a63cfa269629986050bca012f132c0daa383f97684633c4716360
thread:pr:188592,pr #188592: vllm wheel 0630,draft/noise,51,draft/noise,1,0,,188592,,,0,4df699b497b39cbdb73b739f4344e03cee294a82136d205627dd473e0fc84cac
thread:pr:188629,pr #188629: docs: document S3-compatible object storage for distributed checkpoint,draft/noise,51,draft/noise,1,0,,188629,,,0,34b603c9ba92b077f9b73dd6a2882668da84f5486213ae04ddc83a69b580147b
thread:pr:188662,pr #188662: [ROCm] Updated mi300 runners to have 8 shards,draft/noise,51,draft/noise,1,0,,188662,,,0,576038b606006457d1c3fb1091ff312d7f4af1e77c686f9bb11837d62b1fd94f
thread:pr:188734,pr #188734: Add tracer kwarg to torch.compile for make_fx tracing,draft/noise,51,draft/noise,1,0,,188734,,,0,3eecb6c7b61ec72dbd9090aad9451032b2f841aba32f7321f296d3aadd03cd61
thread:pr:188787,pr #188787: OneDNN 3.12.2 acceptance test,draft/noise,51,draft/noise,1,0,,188787,,,0,cb4e6f9bdbe76a40d73da911838103cd495164c2da9334eb9f4718b1be62f322
thread:pr:188858,pr #188858: [DO NOT MERGE] Complex Wrapper Test,draft/noise,51,draft/noise,1,0,,188858,,,0,64ef92fd88c204d72ed5860ab72f41ccd6a0b89eb6af80fea3e9d16332025012
thread:pr:188886,pr #188886: [AOTI] Support PDL launch attributes via `cuLaunchKernelEx` in CUDA wrappers,draft/noise,51,draft/noise,1,0,,188886,,,0,78bb16230eb9f7c9db9c0ab96db82a76a7726b02efaa1de0c027c9e85d0f8e4c
thread:pr:188887,pr #188887: Enable cpp_wrapper for mixed order reduction,draft/noise,51,draft/noise,1,0,,188887,,,0,e75c9b4fc5cc92a87153cbbda2fbbc62c895f2ca0d1fe170919f34dab78f921e
thread:pr:188963,pr #188963: [XPU] Enable XPU testing for 50 test classes across 40 test files,draft/noise,51,draft/noise,1,0,,188963,,,0,8f3ef3f2b931491d2abf06bf9e72158fd73f4ccdd804b5f02edfe9cb6244437b
thread:pr:189001,pr #189001: [inductor] Fix flaky combo-kernel peak-memory test,draft/noise,51,draft/noise,1,0,,189001,,,0,1489db9e189142ee398c90bded08d03c7405a2eb52a3626b4d1dd57997be6db8
thread:pr:189002,pr #189002: [inductor] Don't combo-fuse data-independent producers feeding attention,draft/noise,51,draft/noise,1,0,,189002,,,0,5d19a405b10563f4674bbf31a4b5e5a8f16f337a13e5131b2f14f1f948e27ba9
thread:pr:189003,pr #189003: [inductor] Coordinate-descent tune compile-time combo kernel blocks,draft/noise,51,draft/noise,1,0,,189003,,,0,0545a0a1e12c6d1d3cb8847030b239823cbeae8949a191fd3923082509193a9c
thread:pr:189084,pr #189084: [DO NOT MERGE] Testing Rocm 7.14,draft/noise,51,draft/noise,1,0,,189084,,,0,6ce1e1d1e2f6e94b90b04c0d7ca381168ce14508c26eb726a534053edb29fda1
thread:pr:189128,pr #189128: Gil acquire bool conversion,draft/noise,51,draft/noise,1,0,,189128,,,0,3744f26d835f89f06b0210875a8a26a000d38a0aa073ac19904b6ae49eac2dd7
thread:pr:189198,pr #189198: [DO NOT MERGE][ROCm] Validate 7.14 skip set on ROCm preview lane with TheRock ROCm deps preloading,draft/noise,51,draft/noise,1,0,,189198,,,0,fa85107ad7188e3915c3c0b5b48dc142c9fef73fa3db8d3822ce47964b27f206
thread:pr:189203,pr #189203: [DO NOT MERGE] This is a reference branch to test 7.14,draft/noise,51,draft/noise,1,0,,189203,,,0,cb3a66b8ea60cb6747d94a61617fc86f90c1678ba85043f3183d3d57ffb04733
thread:pr:189205,pr #189205: [DO NOT MERGE] Harmless comment just to enable CI,draft/noise,51,draft/noise,1,0,,189205,,,0,3a44297920b0db8bc64e1aeb037e795f2af868921efe96ee9390f7564970a100
thread:pr:189209,pr #189209: [inductor] Fix with_effects star deps keyed by buffer name instead of operation name,draft/noise,51,draft/noise,1,0,,189209,,,0,6b9dd21a4fe7979a26acdcca495b4699e5a74b95dfe015ecc016bb64b5bff8b1
thread:pr:189216,pr #189216: [Dynamo] Implement DELETE_DEREF,draft/noise,51,draft/noise,1,0,,189216,,,0,a7b56bdedfaa9e36a6aa0623a31ed1f2ad892d5a9946c2d186a810e43119b02d
thread:pr:189218,pr #189218: [DO NOT MERGE][ROCm] Validate 7.14 skip set on TheRock preview lane,draft/noise,51,draft/noise,1,0,,189218,,,0,9d6310b0e4ea773b3fdebdb7c3e3870894e9295391945d3b742c64efa3a27ddb
thread:pr:189227,pr #189227: [xpu][feature]Extend SYCL device-specific math functions to NumericUtils,draft/noise,51,draft/noise,1,0,,189227,,,0,1391a3be9ba782b00f0288b8007f3a188b1f21f6f6f874056e509e930e91b9bf
thread:pr:189229,pr #189229: [DO NOT MERGE][ROCm] Use torch.version.hip to detect ROCm when /opt/rocm is absent,draft/noise,51,draft/noise,1,0,,189229,,,0,cf200944bbf4ba3f609437aa80b1cf38d1a3c3fbb4d809ee07911984f8ce7204
thread:pr:189237,pr #189237: Add c-shims for aten.unsqueeze and aten.expand,draft/noise,51,draft/noise,1,0,,189237,,,0,c2295204178ed4dbfcf94d1e62674cb9731322243f8e8ebb6aaee9f40237f60d
thread:pr:189250,pr #189250: [Test][Refactor] Decouple test_dataloader and test_comparison_utils according to the accelerator relevance.,draft/noise,51,draft/noise,1,0,,189250,,,0,91096b3be9225cb452428afe81cc0b8eec294f62fdcbc2bac57efbc765e86396
thread:pr:189275,pr #189275: Consolidate CD wheel build pipelines under .ci/wheel/ and unify the numpy pin,draft/noise,51,draft/noise,1,0,,189275,,,0,c61e564ee3a7cc58bf0c43c7fa26e4e0af8688b99efb42fbd422ec0d940a7816
thread:pr:189299,pr #189299: [MemoryViz] Decimate over-long stacked-area polygons so the summarize…,draft/noise,51,draft/noise,1,0,,189299,,,0,0a65fdfb5c8385444f8f1655ce777ba78bf118c9f9de1295ac3bece43ef719b7
thread:pr:189301,pr #189301: [ROCm] Bump AOTriton to 0.13b,draft/noise,51,draft/noise,1,0,,189301,,,0,f9257e83e1f645f33419699d91ef8f16d1a7ea1b2ae5b55b89361a008989324c
thread:pr:189329,pr #189329: [DEBUG] Profile the XPU SYCL device-link bottleneck,draft/noise,51,draft/noise,1,0,,189329,,,0,893a9cb5b6c08fb9b4a60c5d90c03a999579f21f9f8b8f88dc941a6c87024dba
thread:issue:104247:issue:1529:issue:16668:issue:167656,"issue #104247: [proposal] ""Name"" string attribute for modules, parameters, buffers, tensors for more pleasant debugging (including distributed FSDP2 logging)",needs design decision,47,needs design decision,15,0,"1529, 16668, 25267, 27479, 30049, 31252, 31353, 35633, 84321, 85652, 85698, 91692, 104247, 167656, 168976",,,,15,9feec9d2498c4c35e831f61829fedf05db59fae36877d086ab349ae0074f29b7
thread:issue:103475:issue:106596:issue:110175:issue:26889,issue #103475: [Inductor] Optimize More Cases of Int32 -> Int64,stale/low urgency,47,stale/low urgency,10,0,"26889, 40373, 43501, 44026, 46948, 54982, 68204, 103475, 106596, 110175",,,,12,b45247230ca632bbb47c5ba4bb3cc19c0eca4d84ad2d16899df499d4e4643414
thread:issue:16293:issue:25066:issue:29092:issue:29177,issue #16293: Unexpected behavior of jit.trace when PYTORCH_JIT=0,needs triage,47,needs triage,6,0,"16293, 25066, 29092, 29177, 40914, 48108",,,,5,d2b334270ed43a8b15fe9bc8b7f37245dd421c626925e4d099b78a045c2a4b1d
thread:issue:104702:issue:127062:issue:127521:issue:69364,"issue #104702: [feature request] Specialized memory layouts and wide blocked/tiled dtypes for cublasLt (`CUBLASLT_ORDER_COL32`, `CUBLASLT_ORDER_COL4_4R2_8C`, `CUBLASLT_ORDER_COL32_2R_4R4`)/onednn: e.g. torch.float16x32 / torch.int8x32 / torch.bits1x512 (akin to torch.quint2x4)",stale/low urgency,47,stale/low urgency,4,0,"69364, 104702, 127062, 127521",,,,3,0791df8a22a3220f9981532a20d9b4e60d6c854ac56dc31ccbfbf40ee3952a95
thread:issue:126714:issue:151579:issue:24422:issue:64492,issue #126714: move label_to_label.yml to an issue?,stale/low urgency,47,stale/low urgency,4,0,"24422, 64492, 126714, 151579",,,,3,775ddfbadb8f26536cf8bc6c5c35f6ffa7b9d80405a7bf7b0dfd0e8bee24c32f
thread:issue:24904:issue:28206:issue:30459:issue:38614,issue #24904: tensorboard add_graph error,stale/low urgency,47,stale/low urgency,4,0,"24904, 28206, 30459, 38614",,,,3,cad83dc63e6532a2743c21cca986f69fd1d6a8ca895ad7e9d6dee23a501fcc6a
thread:issue:100741:issue:36577:issue:67902,issue #100741: Reading `state_dict()` in C++,stale/low urgency,47,stale/low urgency,3,0,"36577, 67902, 100741",,,,2,fec106a81702474991ba9083c50dababf248dd815f3bbb5a7140ad18d3bd5463
thread:issue:124572:issue:41625:issue:43125,issue #124572: Variant of slice that doesn't perform clamp behavior,stale/low urgency,47,stale/low urgency,3,0,"41625, 43125, 124572",,,,2,de9736fa881c1ef0df5dc92b6b4281162d30c07f25a28d9d029c5cc2d6b933c9
thread:issue:24243:issue:30138:issue:54133,issue #24243: hasSideEffects INTERNAL ASSERT FAILED when using .split method with JIT,stale/low urgency,47,stale/low urgency,3,0,"24243, 30138, 54133",,,,2,ef225dafd81394a788515ea179c462091adff6a0b113fe9c2690529a85c97ca5
thread:issue:30812:issue:33670:issue:33691,issue #30812: Tensorboard graph failing to nest modules,needs triage,47,needs triage,3,0,"30812, 33670, 33691",,,,2,c928f678077abeced7c1a9e8854d05faecd6045c87f47f95880d55c1ec731c20
thread:issue:32976:issue:51835:issue:68559,issue #32976: [JIT] pytorch 1.4 breaks torch.jit.script(LSTM/GRU),needs triage,47,needs triage,3,0,"32976, 51835, 68559",,,,2,8fb326768d52232aa6757a89fa9eb02dee985f68b01030676c31099c29ae9963
thread:issue:44710:issue:72295:issue:89241,issue #44710: AMP support for libtorch/c++,stale/low urgency,47,stale/low urgency,3,0,"44710, 72295, 89241",,,,2,fdbe71195be997e53207b52f225ac5420f141674c70ab588c23092da049a69bd
thread:issue:60953:issue:76659:issue:77869,issue #60953: Add support for Python's optimized mode to torchscript,needs triage,47,needs triage,3,0,"60953, 76659, 77869",,,,2,6fcab272de72d5cdd15c70f96a2d4d1cd94f84829c5bca6a5729f3a68dc8c35a
thread:issue:63740:issue:63812:issue:63929,issue #63740: DDP Gradient reduction is not triggered after calling autograd.backward(),needs triage,47,needs triage,3,0,"63740, 63812, 63929",,,,3,122d50f04ddcf3a27a29c3ef9438e7ea64a8a0b3e96dbe53d7196bb260d26618
thread:issue:10471:issue:68104,issue #10471: [Caffe2] Multithreading in Caffe2,stale/low urgency,47,stale/low urgency,2,0,"10471, 68104",,,,1,e486c77115a8e5a5bed1e7127e0837db771ea421fc686ca773d87cb700d5ac5f
thread:issue:105582:issue:49444,issue #105582: RFC: Integrating oneDNN Graph Compiler into Inductor C++/OpenMP Backend for Enhanced Graph Fusion and Performance,stale/low urgency,47,stale/low urgency,2,0,"49444, 105582",,,,1,248c3da50e01159fc640669bb3263deab080ff2c2d7d376ddb311a236f1c38b1
thread:issue:106128:issue:106129,issue #106128: Torchscript optimizer incorrectly applies constant propagation to convert prim::ListConstruct() into prim::Constant,needs triage,47,needs triage,2,0,"106128, 106129",,,,1,82ce1025a9cc4b1c07c6f8714751317994d35b4953e57f9f6343280dac299a61
thread:issue:111676:issue:114967,issue #111676: [export] self.buffer += 1 raises error,stale/low urgency,47,stale/low urgency,2,0,"111676, 114967",,,,1,ec8488049b2d4a447728187e556dcf5aac960de0db84ba93fad9e2f12c285456
thread:issue:118462:issue:118463,issue #118462: Inconsistency in Output for `torch.nn.functional.poisson_nll_loss` in JIT Compiled Model,needs triage,47,needs triage,2,0,"118462, 118463",,,,1,84d2c5b3ee71b8e015608c8b1fe9a7aca134ee56bdebb83b2fef94f740882369
thread:issue:11967:issue:63145,issue #11967: Jit cannot trace autograd for certain operator,needs triage,47,needs triage,2,0,"11967, 63145",,,,1,3a2c860ec142462776e0034b8a0b44efaf02330a968cd66b2f24c666012ad61d
thread:issue:124275:issue:76410,issue #124275: Performance Impact of `torch.autograd.profiler.record_function` in DataLoader's `_BaseDataLoaderIter` Iteration,needs maintainer decision,47,needs maintainer decision,2,0,"76410, 124275",,,,1,e7f7fdf770b31eac9275bb70f20076892e95875157e1b966d89b5a019c0808c6
thread:issue:124461:issue:52312,issue #124461: Support constant value in torch.jit.script,needs triage,47,needs triage,2,0,"52312, 124461",,,,1,72406119365d50e7e82c40681ed952c7b00de80342bbb339eccda6c99742a48a
thread:issue:125245:issue:174481,issue #125245: [export] Add customization API for unflattener to control restored module types.,needs triage,47,needs triage,2,0,"125245, 174481",,,,1,16359293d06832cbd9ab841154887742004a373c435ceb5b5de50194af1a0e22
thread:issue:125678:issue:60252,issue #125678: Unnecessary warning when numpy not installed,needs maintainer decision,47,needs maintainer decision,2,0,"60252, 125678",,,,1,a3779b15e5197224f159bc56c692c6e12cc88670e7b24d5960e89ffd8922d48f
thread:issue:126276:issue:38034,issue #126276: torch script not support ModuleList/Sequential ?,needs triage,47,needs triage,2,0,"38034, 126276",,,,1,b057870d695ec46320c915b7d33004ea78e9bf48cef460dbf98b04019fcccad8
thread:issue:131452:issue:137252,"issue #131452: JIT load traced model with ""Expression of type"" error",needs triage,47,needs triage,2,0,"131452, 137252",,,,1,d1410cfb5e8a5f59fd63d419b9e22339dc8ef753f6fafe27e865778ea9ceec6a
thread:issue:134191:issue:159169,"issue #134191: AdamW is CPU-bottlenecked with FSDP2, with slow foreach kernels",stale/low urgency,47,stale/low urgency,2,0,"134191, 159169",,,,1,7a96580466774850d5612c6459e63cf4f880cf110eb82b49b04b0a94d7abe1bb
thread:issue:138220:issue:138221,issue #138220: Update auto_functionalized_v2 to pass view subgraphs,needs maintainer decision,47,needs maintainer decision,2,0,"138220, 138221",,,,1,77d0f34263b7ba2cdbc4037a88c4dc399506f90182eb681bbad6b3a60fad7526
thread:issue:143258:issue:75002,issue #143258: Proper support for optionals in TorchScript,needs triage,47,needs triage,2,0,"75002, 143258",,,,1,5bd9a2dd2c13b2ae2ea951926003c107e7adb3a4905ddf9430724819ec9725aa
thread:issue:148748:issue:148946,issue #148748: export deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B failed,stale/low urgency,47,stale/low urgency,2,0,"148748, 148946",,,,1,201ed855bea6c70e864ccec82c0c30abf6189ef89828ba79cd0641abf1bae373
thread:issue:15421:issue:30635,issue #15421: JIT is not compatible with data parallel,needs triage,47,needs triage,2,0,"15421, 30635",,,,1,f478ba9869d74a4974a15ff4e9c6cb5fa3f028d39d22f4e68c4f70ab43cb2e53
thread:issue:162365:issue:162451,issue #162365: [Windows][CPU] 3 UT of test/inductor/test_torchinductor.py failed with 0903 nightly whl on windows,needs triage,47,needs triage,2,0,"162365, 162451",,,,1,49453f8a1988f5a88dda6785926af1866f2b927fde7f9c674da780d941c29d06
thread:issue:19053:issue:42350,issue #19053: RNN module weights not compacted,needs triage,47,needs triage,2,0,"19053, 42350",,,,1,fd382a7d01d5b083431bd72e16b2aa65f2a160f92d4b1678b5d9cff90adc38f5
thread:issue:22155:issue:44077,issue #22155: [jit] Optional type refinement on non-named expressions,stale/low urgency,47,stale/low urgency,2,0,"22155, 44077",,,,1,d16fd47784e027804a54a03dd0187030b2b50ca9310bb3c2183519175723dd89
thread:issue:27504:issue:29843,issue #27504: [JIT] List Comprehensions With Ifs not Supported,stale/low urgency,47,stale/low urgency,2,0,"27504, 29843",,,,1,141b0dd1414c9e1e4bbab19d7e1ebe8f44d37e4a5618387b7fc157252f30fc4f
thread:issue:30786:issue:48549,issue #30786: C++ / JIT Parity for ops in `torch/functional.py` and `torch/tensor.py`,stale/low urgency,47,stale/low urgency,2,0,"30786, 48549",,,,1,e27d28d9b5e13467a083c7fb069d78d5ae36a94354a1926996ff4a28579981cd
thread:issue:31772:issue:71548,issue #31772: Support Python builtins on iterators in JIT,stale/low urgency,47,stale/low urgency,2,0,"31772, 71548",,,,1,4393b5f0659df797417b7d795e4c1c76c8f69620cb9d11f4b6306034d316b0da
thread:issue:32651:issue:63725,issue #32651: [Tensorboard] Problem with subfolders from SummaryWriter,stale/low urgency,47,stale/low urgency,2,0,"32651, 63725",,,,1,ceadaa8d71e72f53077a78ad1abb4c10e5c3d14a04ce7bf3daa9285f5a083024
thread:issue:42355:issue:42427,issue #42355: nn::Sequential Link error in Android version,needs triage,47,needs triage,2,0,"42355, 42427",,,,1,ef247505be3030253f80fc4375139371d8597dbb3db8534586991cb117084957
thread:issue:42847:issue:63085,issue #42847: Get test_jit.py below 10k lines,needs triage,47,needs triage,2,0,"42847, 63085",,,,1,558d074c263bfe80223d8924bf879803cf7fa7e3e17435f76948a7721211cf2e
thread:issue:44768:issue:45208,issue #44768: [JIT] function decorated with @torch.no_grad() can not be exported when another function is called inside it.,needs triage,47,needs triage,2,0,"44768, 45208",,,,1,e4fbeaba2c43812d9396f3f6a8f55422e5160e555ca83f2f9047958b3d752c77
thread:issue:45465:issue:46749,issue #45465: torchscript model size increases when loading and saving it again,needs triage,47,needs triage,2,0,"45465, 46749",,,,1,e1e7453b9e6b9e0d62e9712d3a8c02a665145b039219d291dd94f5050daed8c0
thread:issue:47964:issue:90365,issue #47964: TorchScript is unable to distinguish int and ScalarType,needs triage,47,needs triage,2,0,"47964, 90365",,,,1,e3215067edfe5e1111d6cf06b1d8b5d1f8ef0305ddec5dd53ea1b719c0fda320
thread:issue:48932:issue:71359,issue #48932: Unknown builtin op: torchvision::nms when loading scripted FasterRCNN,needs triage,47,needs triage,2,0,"48932, 71359",,,,1,e63541e3bd814c3e90bca9f83cb3e49d2f73f1685aadf45d2c1c7b23f627bcb3
thread:issue:50533:issue:50678,issue #50533: Non-avairable nightly version for converting models to NNAPI,needs triage,47,needs triage,2,0,"50533, 50678",,,,1,dde32d23c8b4d3fc499eca0c1484577ac56fc7c5ac1fa826f3292233021cc125
thread:issue:51018:issue:60333,issue #51018: len of ModuleList is incorrect for jitted models,needs triage,47,needs triage,2,0,"51018, 60333",,,,1,0bc00ce14e4ad827db8180d6e444aebec3d9d61c108497bde649ae78ae815efe
thread:issue:52788:issue:52800,issue #52788: distributed/optim/test_zero_redundancy_optimizer uses unittest.main() instead of common_utils.run_tests(),needs triage,47,needs triage,2,0,"52788, 52800",,,,2,fbae02c43c0df54637412c0a7aa0c359a0b41be560f9ff0ddeca14a8c4fca325
thread:issue:55571:issue:66627,issue #55571: Update TCPStore.wait() error message to be more friendly,stale/low urgency,47,stale/low urgency,2,0,"55571, 66627",,,,1,9c602e10dd28dbc2f3756c113da4d4061176cc33fcd8d29166ce703e1e032cc3
thread:issue:55757:issue:56244,issue #55757: Synchronize RRef.to_here() CUDA Streams properly when the profiler is enabled,stale/low urgency,47,stale/low urgency,2,0,"55757, 56244",,,,2,30e2d729185141353df6ab39134267c206b5c6f8551559cdb61db1d651153697
thread:issue:61693:issue:68871,"issue #61693: Pytorch Mobile on Android, LAPACK library not found in compilation",needs triage,47,needs triage,2,0,"61693, 68871",,,,1,9b486f20b27940b8c25fa243a4f88f82f735b56f21ed24524eac75cc382970d1
thread:issue:67570:issue:71595,issue #67570: RFC: Overlap optimizer computation with DDP/FSDP backward,needs triage,47,needs triage,2,0,"67570, 71595",,,,1,6db7d7b0d73f6bcec8beb9d3508f99083b3a29ac709b36e46092f4c6d9e11bad
thread:issue:67955:issue:68340,issue #67955: Torch 1.9.1.post3 - Error in magma_getdevice_arch: MAGMA not initialized (call magma_init() first) or bad device,stale/low urgency,47,stale/low urgency,2,0,"67955, 68340",,,,1,ee9c07616f2d5f63c0337192943dadc405f5edf3a318e87b0bf60f2d1ac612ed
thread:issue:68385:issue:83271,issue #68385: ImportError: cannot import name 'ProcessGroup' from 'torch.distributed',needs triage,47,needs triage,2,0,"68385, 83271",,,,1,a9b119a4d732d851b85ec958d8448a18e51dc8f56f4051e44d0e516fcdf5d733
thread:issue:69033:issue:72112,issue #69033: Lazy Tensor Core Documentation Out-of-Date,stale/low urgency,47,stale/low urgency,2,0,"69033, 72112",,,,1,d81850ae69f31dd14369d49ec5252f25d469db2eba65a6768399182b1e7aa5da
thread:issue:69741:issue:69744,issue #69741: JIT: Support for `torch.autograd.functional.jacobian` in TorchScript,needs triage,47,needs triage,2,0,"69741, 69744",,,,1,94119f29671c26d496561e66d342d2b3816c39a7d53d2ed3fee0585831aa8e59
thread:issue:71470:issue:92073,issue #71470: torch.jit.script failed to compile nn.MultiheadAttention when specifying the kdim and vdim parameters.,needs triage,47,needs triage,2,0,"71470, 92073",,,,1,dcb68c0f19efba2f967eb67af54a79b4fb9656ead16d03f77a428cca2e5fc6fb
thread:issue:75336:issue:96060,issue #75336: Libtorch Crash after Attempting to Evaluate Model after Opening an OpenGL Context,needs triage,47,needs triage,2,0,"75336, 96060",,,,1,ecb21d73fc383c4e38f95c65de97442466862a1917991e58cac931441cf7c206
thread:issue:76578:issue:88448,issue #76578: [JIT][Autocast] Batchnorm folding pass during freezing doesn't preserve types,needs triage,47,needs triage,2,0,"76578, 88448",,,,1,d077b85a32ea2eb03c9c87dca70069a9a355fda3a24052e445e83e96aa4bf7ca
thread:issue:79715:issue:91293,issue #79715: [FX] TypeError when tracing cat taking split's output as input,stale/low urgency,47,stale/low urgency,2,0,"79715, 91293",,,,1,0a857e7b0fa51b8175399a83ee4336643332a844807d1599c634d822f68b2282
thread:issue:80561:issue:85792,issue #80561: Cannot run scripted BERT_Pytorch,needs triage,47,needs triage,2,0,"80561, 85792",,,,1,0f6ac7a27baa1734659a4c544f65d099a8b82dbdc01535e42e5a327ec938a104
thread:issue:85877:issue:86804,issue #85877: JIT model could return 'NaN' gradient after the first execution,needs triage,47,needs triage,2,0,"85877, 86804",,,,1,d3cb390c505e9a9abf4b4710a00473de220f9d69d308d6595269fabc6cda0004
thread:issue:86791:issue:95124,issue #86791: We don't have an op for vulkan_prepack::conv2d_clamp_prepack but it isn't a special case.,needs triage,47,needs triage,2,0,"86791, 95124",,,,1,7ef721e4d003c095c6a4134d439c984a13fc6115658458045b389e1108359b3b
thread:issue:90366:issue:90367,issue #90366: [JIT] INTERNAL ASSERT FAILED `torch.mul` with boolean primitive constant,needs triage,47,needs triage,2,0,"90366, 90367",,,,1,4dd0c1f3fd46d813940b595bfe75af76112efb56938485b70740e7c0a9a226b7
thread:issue:92910:issue:92987,issue #92910: torch.jit.save() generates different contents in a file among different endian machines,needs triage,47,needs triage,2,0,"92910, 92987",,,,1,2b06ee21ac7f73998347504b1576c6709d26f8826ca21eb7168518308d799dec
thread:issue:100636,issue #100636: tracing does not work when torch.distributions is involved,needs triage,47,needs triage,1,0,100636,,,,0,157f0098393021deb360fa4555ef51dbc10ba0c21104b68a65eca78a6217e8c6
thread:issue:100730,issue #100730: `torch::jit::EliminateExceptions` lowering pass never completes on specific model,needs triage,47,needs triage,1,0,100730,,,,0,f35b14b14cec445993d2f9e7223d87d79a212474d6ed24118a4e324c7174471e
thread:issue:101055,issue #101055: DISABLE libtorch-2.0.0+cu117 destructor exception,needs triage,47,needs triage,1,0,101055,,,,0,1709cfb80490af64e65342ec0d65822746876867378b095b68c5fd872d8369b5
thread:issue:101855,issue #101855: torch.jit.trace() Floating point exception,needs triage,47,needs triage,1,0,101855,,,,0,9494bed360cfad3732dcdc1438850807a352e83efc1271a095023dab7c5c357e
thread:issue:102653,issue #102653: torchscript dataclasses have bad support for class types as fields,needs triage,47,needs triage,1,0,102653,,,,0,946e4255b4496369c2a98d48a1ae33feea7fd1aaff208dceadb8958b5201be52
thread:issue:103070,issue #103070: Unexpected failure in LLVM JIT when running TorchScript model in C++,needs triage,47,needs triage,1,0,103070,,,,0,313b52cf14e04753945187176c30708dc26992f07dfb15b2d778d74a29e171d3
thread:issue:103250,issue #103250: torch.jit.script mean(keepdim=True) segfaults on GPU,needs triage,47,needs triage,1,0,103250,,,,0,f3dac1db0f70d01dd3d6e52db7095d8c1da17dc23fea9dea508c00777fbf5c5b
thread:issue:103788,"issue #103788: [Torch Mlir] avg_pool1d function padding init value should be (0,)",needs triage,47,needs triage,1,0,103788,,,,0,cc2378bb79f5180f859e0bcbbab973807fb266fe012bd7881269c093810d1065
thread:issue:104041,issue #104041: Scripted model is loaded on GPU but the inference seems to utilize the CPU with zero GPU utilization,needs triage,47,needs triage,1,0,104041,,,,0,7a32b585aab11a28b039a64f23387b39cb0cab5f57b5875525eb6932d85b384d
thread:issue:104405,issue #104405: Detailed error: Tensor-likes are not close! When use torch.jit.trace,needs triage,47,needs triage,1,0,104405,,,,0,56eed17944361a9376912f8d22438f797ffc56c2f3799baf5c7ffe90fadfcf86
thread:issue:104450,"issue #104450: Numpy/scipy module works fine with Torch modules, but not TorchScript. How to torchscript a numpy/scipy module?",needs triage,47,needs triage,1,0,104450,,,,0,eb1dc536d08b21a3b4057e6861b0ddcde1cf4112303d2e92ddd3630392bf1b45
thread:issue:104458,issue #104458: [Feature Request] Add a new overload of torch::jit::load to restore traced shape and type,needs triage,47,needs triage,1,0,104458,,,,0,06da57811afab713a5c6bf1c6bb019303b8f6af6f81a237fe55a05591f7d5504
thread:issue:104712,issue #104712: torch.jit slicing error (styleganv2),needs triage,47,needs triage,1,0,104712,,,,0,faa60c7f8b72b4dcef7c759a9abeacae393206e48a9c321063e86a817dd180f5
thread:issue:104732,"issue #104732: torch.jit.trace says ""Arguments for call are invalid"" on torch.ops.aten.sub(3, x, alpha=3)",needs triage,47,needs triage,1,0,104732,,,,0,5c3f27eba4d19bd068742a9cba3a32153af94a8f9b2c7d2a2c21eb348be3bc1b
thread:issue:104906,issue #104906: torch.jit.frontend.NotSupportedError: keyword-arg expansion is not supported: for dgl.nn.HeteroGraphConv(),needs triage,47,needs triage,1,0,104906,,,,0,39b10e4b1b0c6c3ec746294a605eeba24bc203c5234933d638421d6eefae292c
thread:issue:104959,issue #104959: Saving a LightningModule torch.jit.ScriptModule is incompatible with torch.amp.autocast,needs triage,47,needs triage,1,0,104959,,,,0,9279cf850064bd3786beb198433ad987b328212ead00dc70efb6760bef0d93f2
thread:issue:105196,issue #105196: Error loading TorchScript model with torchvision::nms operation in libtorch,needs triage,47,needs triage,1,0,105196,,,,0,1e8262de4da35c9d832ff376cac618ebdbfa26ba3dff3aafe5a74b651eb4d41f
thread:issue:105821,issue #105821: JIT input aliasing does not support aten::fill_,needs triage,47,needs triage,1,0,105821,,,,0,ea8a8832f26cbb6dccbaa7e930972b1d22c7f591eeac67db2779495541dcd483
thread:issue:105900,issue #105900: torch.jit.frontend.NotSupported when compiling stable-diffusion,needs triage,47,needs triage,1,0,105900,,,,0,64952521271a3ba0aecee5ae96f9a2236f4e9583c6babe383021de34c2cc164c
thread:issue:105917,issue #105917: Torch.jit.frontend.NotSupportedError: not supporting functions with variable number of arguments.,needs triage,47,needs triage,1,0,105917,,,,0,28ba16a69019b702acefe35caee114a3fe12eebaf5ed6f93285034110c6da67a
thread:issue:105918,issue #105918: Torch.jit : RuntimeError: Unable to extract string literal index for ModuleDict,needs triage,47,needs triage,1,0,105918,,,,0,55030b65baa2c286c00d99a158928098631c7aedccde32501b1415b6177215eb
thread:issue:105997,issue #105997: Flatbuffer torchscipt files don't load in PyTorch Android Lite 1.13.1,needs triage,47,needs triage,1,0,105997,,,,0,8e80b4ef880c4d6ebba1e401d7f364e9e76f2ec3e52c4f44605f4200bcbf4f56
thread:issue:106435,"issue #106435: [JIT] .item() dict keys cause `RuntimeError: Cannot create dict for key type 'Scalar', only int, float, complex, Tensor, device and string keys are supported`",needs triage,47,needs triage,1,0,106435,,,,0,7ec65c5e59b2b365df95719432228e0ad57f4ba223022cbea99c8b9fd4bb538e
thread:issue:106699,issue #106699: torch.jit.script: scripting doesn't work with wraps,needs triage,47,needs triage,1,0,106699,,,,0,cf30f09ef01b6a386e5e3251be49625ef0fc131011fac2ad4208eae1f2fd30ae
thread:issue:106764,issue #106764: [Feature request] Add new API Tensor.device_as,needs triage,47,needs triage,1,0,106764,,,,0,ae8d0d40b16bae752febeaf91081d15679e951ddcd9e2994ec9f1fdc261044e9
thread:issue:107568,issue #107568: RuntimeError: Unsupported value kind: Tensor while torch.jit.script nn.Module,needs triage,47,needs triage,1,0,107568,,,,0,2f4521f3e59c4971a03531900055491a27892f079a045652a4336cfe17e7e2bb
thread:issue:107771,issue #107771: libtorch infer error : CUDNN_STATUS_INTERNAL_ERROR,needs triage,47,needs triage,1,0,107771,,,,0,ae1491831c4567e18f496c8310d6ead2b9afe722660d85ffcce2e8adf94515b9
thread:issue:108226,issue #108226: torch.nn.functional.pad() with value type bool,needs triage,47,needs triage,1,0,108226,,,,0,518e7256a6215c70cae522e578e7811839390fa5fcc651fb04cfee888491b3b7
thread:issue:108332,issue #108332: RuntimeError: dims.value().size() == self->getMaybeRFactorDomain().size(),needs triage,47,needs triage,1,0,108332,,,,0,c7e58c80fd0e8ebc36eca7b223a022833f4d9d142e2e4c8e96929b2ca035b822
thread:issue:108665,issue #108665: INTERNAL ASSERT FAILED in `shape_type_inference.cpp`,needs triage,47,needs triage,1,0,108665,,,,0,5fa2985c6b7c2f25e33f6d4088bf6289d09149303ccafed06a3f4863157f25fe
thread:issue:108670,issue #108670: torch.jit.script produces incorrect gradients,needs triage,47,needs triage,1,0,108670,,,,0,50cce457b4273271505bc617ad5116cf18c58a36e72300f49454ab753755f399
thread:issue:108865,issue #108865: Heap buffer overflow with `torch::load` on fuzzy data,needs triage,47,needs triage,1,0,108865,,,,0,c3360cfe2f9fd888ee77aaed9242cd6545986cf7001dce6fe027675fb5690690
thread:issue:109113,"issue #109113: The API ""torch::jit::_load_for_mobile"" is limited to create an object living on the stack.",needs triage,47,needs triage,1,0,109113,,,,0,0f2a4c6ff8e853be17f2c7b6a16da8aeb0b0deb1ae745b3751b5a2b73d08112c
thread:issue:109488,issue #109488: [bug] FALLBACK path has been taken inside: runCudaFusionGroup,needs triage,47,needs triage,1,0,109488,,,,0,c228882e142adfaab936ae751009afade1d8bb7cd924460492acb4980e4cc6de
thread:issue:109579,issue #109579: [Android: React Native] couldn't find DSO to load: libtorch-code-gen.so when loading model,needs triage,47,needs triage,1,0,109579,,,,0,30fdcc6fe8c73788f903bdb5df47d77dffbcbd7d4a2b4ba07829aeba20044fc4
thread:issue:109699,issue #109699: [TorchScript] Support ScriptFunction arguments in torch.jit.script calls.,needs triage,47,needs triage,1,0,109699,,,,0,d13792978173ffc74164598a66a6b58d026fbe309b266153b5d7b02f54ac5c3b
thread:issue:109719,issue #109719: Make torch.cuda.graphs.is_current_stream_capturing() available in TorchScript,needs triage,47,needs triage,1,0,109719,,,,0,c16b328be46b3f44558072277dfeab2c4404bbe46ce74b5fb4fe9deb02449de4
thread:issue:110259,issue #110259: Cannot avoid kineto_LIBRARY-NOTFOUND error when using pre-built pytorch,needs triage,47,needs triage,1,0,110259,,,,0,a0f84d44d800dcd112263003e6c2dc49acdeaad017fcf71b0ea896b3a0ee1989
thread:issue:110332,issue #110332: `torch.jit.load()` might unresponsive in IBM s390x when loading some certain torchscript saved by x86 machine.,needs triage,47,needs triage,1,0,110332,,,,0,d785a28bfe2a7edfb0b583df00291d5ff82e33b89830be1a3c1f94c3ae608df8
thread:issue:110379,issue #110379: Pytorch LoadNativeLibrary issue,needs triage,47,needs triage,1,0,110379,,,,0,ce4194bb44f08835f97101a98d79180b54fd29432ecf98c4110c78e645c15733
thread:issue:111135,"issue #111135: No op for aten::where with argument types: Tensor, Tensor, bool.",needs triage,47,needs triage,1,0,111135,,,,0,e8a8e573806bc6a19019af75bdf21d245b0681368533e99b697ff1a630b305d1
thread:issue:111142,issue #111142: 'torch._C.Node' object has no attribute 'cs',needs triage,47,needs triage,1,0,111142,,,,0,595a69612011572f0db7500a1a2e2da57661fa2fa30c6dd8b3b4108697fca7f2
thread:issue:111480,issue #111480: torch.jit.script persistently changes default from utf-8 to ascii,needs triage,47,needs triage,1,0,111480,,,,0,09a683681f767088c5f72bf0a3358a90aa97dfd9df7ef1af7684ef0dd6fa121a
thread:issue:111792,issue #111792: torch.jit.trace is not able to trace torch extension,needs triage,47,needs triage,1,0,111792,,,,0,45d74c4417a42a5e0e30a5b648c3e4c86b090ebe111e0bc9c15f5ebe5a0e8da4
thread:issue:111863,issue #111863: Jit scripting support for `|` and mixing `typing`.,needs triage,47,needs triage,1,0,111863,,,,0,c753d07fd8755890cfc225757736aaee1de9c2e085e051a84a767c886ab1ee26
thread:issue:112372,issue #112372: [Tracer] RuntimeError: _Map_base::at when tracing using autograd.Function,needs triage,47,needs triage,1,0,112372,,,,0,f51f060b1d33c9be99be0d6c31ad2d82101069822132e17fb925a68d3f2fb393
thread:issue:112425,issue #112425: torch.jit.trace output changes when I add comments to the file,needs triage,47,needs triage,1,0,112425,,,,0,b87561c8bcf48811ab22e26756e10a1fd19026cd1d832dde217f1debea2b0e21
thread:issue:112844,issue #112844: torch.export does not support torchaudio.transforms.Spectrogram,needs triage,47,needs triage,1,0,112844,,,,0,3fc079133cfeea842907bc3a6ea0655e08025e9921fa6d6a0e38c6e5fb7ac562
thread:issue:113945,issue #113945: torch.jit.trace has incorrect execution for += operation during compilation,needs triage,47,needs triage,1,0,113945,,,,0,c909267787a50e6f6a03d627d6aad8312df9dca9b93e759073504e87f48cb90c
thread:issue:114102,"issue #114102: ""aten::prelu"" can not be used in the metal backend",needs triage,47,needs triage,1,0,114102,,,,0,2c24bebaa1db628bfab7661c211720db323c4dad4f388ee9815ba6e3e95d6d35
thread:issue:114112,issue #114112: [JIT] - torch.script - 'Optional[Tensor]' object has no attribute or method 'size',needs triage,47,needs triage,1,0,114112,,,,0,2c57e9a8099ede2f7df9be49c8310874e45526cc321235070c42a9a92287a415
thread:issue:114283,issue #114283: CppException in Android Studio with faster_rcnn_fbnetv3g_fpn.yaml Configuration,needs triage,47,needs triage,1,0,114283,,,,0,ba8d750e15298cf2af3a0dba5e5af1c41a86b5ceabe15c2ef64072572408d189
thread:issue:114543,issue #114543: [torch.jit.script] Torchscript produces incorrect result when argmax result is used in indexing,needs triage,47,needs triage,1,0,114543,,,,0,9909cccc97ec86e1acf92f6e222dede64a5c47cbde9e6a57d0a81e42b79bbc8e
thread:issue:115055,issue #115055: Better test suite for torch.export.export,needs triage,47,needs triage,1,0,115055,,,,0,e3264c2af4f75617817fa5592083ac11661b4cc6deafb476c3291bbdd156d43c
thread:issue:11532,issue #11532: [JIT][tracer] Slicing shape is specialized to tensor rank,needs triage,47,needs triage,1,0,11532,,,,0,16060c11f7ff689d12249fc1cf333eb4596b5b9d56266efbc8b2551070a2ebde
thread:issue:115488,issue #115488: torch.export.export is overriding source_fn inside higher order op submodules,needs triage,47,needs triage,1,0,115488,,,,0,c0a99d76d93a6ef7b66bb35635b902e1a0d14fdebb49835d8f5ae932d3a542ce
thread:issue:115511,issue #115511: [Export] aten::_transformer_encoder_layer_fwd is not core aten and has no decomposition.,needs triage,47,needs triage,1,0,115511,,,,0,6866e4cd48089d58c1bc1a02102a5e1fecfe4e9b5deb7b6ab95643bd63746234
thread:issue:115622,issue #115622: Torch jit load model with enums,needs triage,47,needs triage,1,0,115622,,,,0,2eb0f5deb1c5aef7010c5e52613e5c909c2749e487b817ff3c2afe26c1ae5598
thread:issue:115639,"issue #115639: `torch.jit.save` fails for static function from C++ TorchScript extension with ""Failed to downcast a Function to a GraphFunction""",needs triage,47,needs triage,1,0,115639,,,,0,6f007bfca157cb941cd5452fa6576f0442b8e4adb22bdb83d6d17e18ebbfb12a
thread:issue:115733,issue #115733: RuntimeError: thread_predicates_.find(tv_inp) != thread_predicates_.end(),needs triage,47,needs triage,1,0,115733,,,,0,9b514affda409fa39ce3f19b773dbd85ee9e5df1f97cabf96dfb902ab08f3555
thread:issue:115736,issue #115736: Precision Loss in JIT ScriptModule at float64 Tensor Init,needs triage,47,needs triage,1,0,115736,,,,0,d84caabdc8f27e0bb29556587ec559ed44fbe807458c6e4b8543a335dc0dd9b6
thread:issue:115914,issue #115914: torch.jit.script fails when parsing torch.SymInt,needs triage,47,needs triage,1,0,115914,,,,0,befff06d5a52e94c8cf48c584f4d73e6b811ecda1281d5e03e2fd7bec31b1480
thread:issue:115985,issue #115985: Faster Pytorch dequantize() + matmul for quantized models,needs triage,47,needs triage,1,0,115985,,,,0,cd70b2ea8b581f7963521a60b630f40b5aa8ae15e7311f0b1b975c39d4077a0c
thread:issue:116010,issue #116010: Why has no torch::jit::load() in Pytorch Android's libtorch_cpu.so,needs triage,47,needs triage,1,0,116010,,,,0,2ac9b43a9f656af334dc33184371e38042d2d2c67c54068ff41f284a7a137091
thread:issue:116240,issue #116240: [inductor][cpu] RuntimeError: quantized_resize_cpu_ does not have a deterministic implementation,needs triage,47,needs triage,1,0,116240,,,,0,b4c582f0e7bad9464dd1c687a2aba85a57014b41c72a5c9a90fc60f1c27d49cf
thread:issue:116250,issue #116250: `torch.Tensor.rot90`: Heap-buffer-overflow,needs triage,47,needs triage,1,0,116250,,,,0,269d396aafdcadcc9ac1a9b58f842ada2932354880fcef0ef9b7881e05219e5a
thread:issue:116251,issue #116251: `torch.Tensor.flip`: Heap-buffer-overflow,needs triage,47,needs triage,1,0,116251,,,,0,9963833785f75a6058a8294d5b7e4215fbbfbe3cc8ed95b51ca53bfb41e871b4
thread:issue:116252,issue #116252: `torch.Tensor.fliplr`: Heap-buffer-overflow,needs triage,47,needs triage,1,0,116252,,,,0,d45671468808023134d1dce741e695f43a7fbbaa23969cfd1f95c3d97be235c2
thread:issue:116253,issue #116253: `torch.Tensor.flipud`: Heap-buffer-overflow,needs triage,47,needs triage,1,0,116253,,,,0,a107eed0a0ca76cb279331e743bdbe34fa8b9d2dd5b53a0d1a40b69eb8e48359
thread:issue:116498,issue #116498: torch 1.8 cannot torch.jit.load for script model,needs triage,47,needs triage,1,0,116498,,,,0,c315bc57d0e3a3c992280b1b9ff7e4c3a368ff918564187f188322e27f8c0429
thread:issue:116544,issue #116544: [export] namedtuple inputs can't be retraced,needs triage,47,needs triage,1,0,116544,,,,0,2433e15851e47b516adce2d25719cfa0f77329ab268b2d7bc3d13336725e42ff
thread:issue:116612,issue #116612: Loaded torch.jit traced module treats dtypes incorrectly in autograd,needs triage,47,needs triage,1,0,116612,,,,0,c07b18a7def5ca87c656808d74b5fba471d5c13b1121611f451ccf7335976f1b
thread:issue:116662,issue #116662: test/test_vulkan.py fails with vulkan_prepack::conv2d_clamp_prepack not found,needs triage,47,needs triage,1,0,116662,,,,0,912548b78cf918b0642590cafdd52a8168ea08b7a8b1abf031955d5a79bab7e3
thread:issue:116795,issue #116795: multiple output support of jit mobile model?,needs triage,47,needs triage,1,0,116795,,,,0,db7875d4a57b42163ec88f89df323dbdf16a17aae0abe9cf5f3310fee104b70e
thread:issue:116865,issue #116865: Optimizing Per-Channel Quantization for Improved Inference Performance,needs triage,47,needs triage,1,0,116865,,,,0,65ca35e06196cac530708d7c29c69dfa7695a9fea984b926f5d9ce86aaaf0094
thread:issue:116992,issue #116992: torch.fake_quantize_per_tensor_affine with inf input provides wrong output value,needs triage,47,needs triage,1,0,116992,,,,0,a9b5ec0e48a98650387d22d074f627eb9f27eea3482ed961b9b248337458184f
thread:issue:117102,issue #117102: conver_fx report len(device) > 1 error,needs triage,47,needs triage,1,0,117102,,,,0,d701269d5ff50964e3c83c2e70b9fa0dc91dfadbbd0d893759a926e12dc38842
thread:issue:117229,issue #117229: ```torch.nn.functional.pad```: Internal assert failed,needs triage,47,needs triage,1,0,117229,,,,0,3d6068cbedb6d1025434b19fbffe7c6f06d6b0bc571f55c1e3b4c2f440204680
thread:issue:117239,issue #117239: ```torch.nn.Upsample``` class: Internal assert failed,needs triage,47,needs triage,1,0,117239,,,,0,9565d19abd62f0f74e581e29b1e882784cd8769f740453fe309deb39264b0352
thread:issue:117541,issue #117541: Jit model running on cuda:1 uses memory on cuda:0,needs triage,47,needs triage,1,0,117541,,,,0,ae07fbaca664644d984d60cd910312519c9f371783893bc36b0bb94035d62bce
thread:issue:117920,issue #117920: Could not run 'quantized::embedding_byte' with arguments from the 'QuantizedCPU' backend.,needs triage,47,needs triage,1,0,117920,,,,0,812781fe46166b5441e3e9f821b25722cdcdd93f3df0797ef83326d8ca1cf2e9
thread:issue:118069,issue #118069: Quantising a multimodal large language PyTorch model to run on Android,needs triage,47,needs triage,1,0,118069,,,,0,bf0e94a6dc96d388212c3ed42285ace79320f11589e15cbeeea6e32070e55b74
thread:issue:118086,issue #118086: Different behaviors with torch.isfinite and torch.tril between eagermode and JIT,needs triage,47,needs triage,1,0,118086,,,,0,016b83bbb36ad3d3a7046818b8c6bcae615391b0554c2f1fabe23e8de07e5cca
thread:issue:118165,issue #118165: [Quantization-aware training] Can not quantize nn.MultiheadAttention module,needs triage,47,needs triage,1,0,118165,,,,0,a5e50acb7a7e115ea49131169712c19a04117d1b1d3a1b511fcb77aca9a5ccb8
thread:issue:118189,issue #118189: Cannot load a jit-traced model if it uses nan_to_num with too large a constant.,needs triage,47,needs triage,1,0,118189,,,,0,3831faa3391744ba0d84c6aade63e50ab9b20e16930dcc5430998c992fd5cce3
thread:issue:118297,issue #118297: Inconsistent outputs of `cosine_similarity` and `linalg_svdvals` between eagermode and torch.jit.trace(),needs triage,47,needs triage,1,0,118297,,,,0,03b0ac09d6c89f191cfe051caceb354ac9ceae0cffe1a326d19912312a0b4257
thread:issue:118300,issue #118300: Inconsistent outputs in `linalg_ldl_factor_ex` and `special_spheriacal_bessel_j0` between eagermode and torch.jit.trace,needs triage,47,needs triage,1,0,118300,,,,0,b7ed8a046d02801e3ac62e17a7b6e18cc141b26b2eb07a1fbc4e660c80e2d7d7
thread:issue:118302,issue #118302: Inconsistent outputs of `torch.adaptive_avg_pool1d` and `torch.threshold_` between eagermode and torch.jit.trace,needs triage,47,needs triage,1,0,118302,,,,0,5567c23f401bed3e560a3e96ada61aa826fe8938ce5481e730703cf374831d36
thread:issue:118303,issue #118303: Inconsistent outputs of `mish` and `log10` between eagermode and torch.jit.trace,needs triage,47,needs triage,1,0,118303,,,,0,a00975b4e0bed223b9d0e5cb2e8ac417f2162068e2989d09d3280831741c1ad7
thread:issue:118400,issue #118400: Inconsistent output of `torch.lodget` with JIT compilation,needs triage,47,needs triage,1,0,118400,,,,0,217472b1b4bac61c798439d26a3b6651bd3c842a0c072568b67812969dbe5d0f
thread:issue:118401,issue #118401: Inconsistency in Output When Using `torch.nn.functional.selu` with JIT Compilation,needs triage,47,needs triage,1,0,118401,,,,0,5ad21bd162855b6e30bb4e399e0b741dd4d642645d27205b20600a2c723b1c1a
thread:issue:118410,issue #118410: [export] Error loudly when tracing over a module not in the module hierarchy,needs triage,47,needs triage,1,0,118410,,,,0,ec70864ddb7dd75fa4969e347d02ae1cac5968fb38c2df54aeb72b64e85f2429
thread:issue:119192,issue #119192: [JIT] Deserialized `RecursiveScriptFunction` forgets default values for arguments.,needs triage,47,needs triage,1,0,119192,,,,0,da226c616f4f8ea9f997b2ff81de05275b0b2803afea361ed31382981ff65922
thread:issue:119454,issue #119454: Python 3.12 CPU test_fx.py::TestFunctionalTracing::test_nn_functional_adaptive_max_pool is failing,needs triage,47,needs triage,1,0,119454,,,,0,00aa76ad242cded82fd6d0288e449fab2cb560bd6af4efe77e7d328739c09204
thread:issue:119484,issue #119484: Follow-up #119356: Preserve node names for torch.export.export,needs triage,47,needs triage,1,0,119484,,,,0,89062a80a2312f4fcf991144dabee606b78799a16686152ca0834a075a92c3f3
thread:issue:120166,issue #120166: QAT using pytorch-quantization cause accuracy loss after exporting to onnx,needs triage,47,needs triage,1,0,120166,,,,0,8274171feadebe4bb6ef6a477fbdfb7ce68defe94bb56708cc6c2c476676d45a
thread:issue:120437,issue #120437: fx::GraphModule::to_folder doesn't seem to handle _PyTreeCodeGen correctly.,needs triage,47,needs triage,1,0,120437,,,,0,97a6af3f9d480b848a3ca1ec61be28b3c9836b26fb499473efb5fd9656b2c2f2
thread:issue:120485,issue #120485: Try to make `jit.trace` working with dynamic control flow but it does not work,needs triage,47,needs triage,1,0,120485,,,,0,30053002e4aea764843c7f933565a62fe5f6425c6c7d0e2b9ccd7c533a60eb32
thread:issue:120705,issue #120705: Trying doing static_quantization with Pytorch 2.1.0 torch.ao.quantization,needs triage,47,needs triage,1,0,120705,,,,0,f739310c2f929ff9bc9881915998e10c3e112f01e65c0bd5e95d2f931143cfb3
thread:issue:120976,issue #120976: RuntimeError: xnnpack::convolution not available！,needs triage,47,needs triage,1,0,120976,,,,0,677698f9338c4906bf222ec4c382a9daef61f28e99c39ff89cd624113fada6a0
thread:issue:121191,issue #121191: Export-based GraphModules fail with deepcopy,needs triage,47,needs triage,1,0,121191,,,,0,99433f656bf8492f084a6764c102d3fdd1560f388aaed89dc530cca9737863fe
thread:issue:121230,issue #121230: INTERNAL ASSERT FAILED in torch.pow,needs triage,47,needs triage,1,0,121230,,,,0,c49f079bc0abe5c3f415db401b404b157f0110df440480e4f86aecbdef72df47
thread:issue:121581,issue #121581: The results of PyTorch and TorchScript models are inconsistent.,needs triage,47,needs triage,1,0,121581,,,,0,e27c04b9f9b37c55441ae27157c12a5df98cb164b2c962f32dd768840814cb6e
thread:issue:121597,issue #121597: LibTorch-Lite 2.1.0 Libs missing x86_64,needs triage,47,needs triage,1,0,121597,,,,0,be26f43d7cc5de14131af3a7100bb87df2df6779a37e60e69b46b7e25475b833
thread:issue:122020,issue #122020: torch.jit.script(model) and torch.jit.trace(model) performance degradation.,needs triage,47,needs triage,1,0,122020,,,,0,6a83b4e6a2b8f67fb10a91033fbcb04bf858e3c9b062e73116a382cca3630e15
thread:issue:122183,issue #122183: [FX Graph Mode Quantization] Prepare model fails with Linear-BN when `affine=False`,needs triage,47,needs triage,1,0,122183,,,,0,80290b516788155f865a6178438f1d83f80cd7bf1d9bd448935ffc34076cfd91
thread:issue:122224,issue #122224: support QAT fusion for conv transpose + bn + relu,needs triage,47,needs triage,1,0,122224,,,,0,baaff8859f81e11448abffc2387ec2a0b5330f335634e9ca4512d62a8331c99c
thread:issue:122303,issue #122303: How to exclude some modules from quantization?,needs triage,47,needs triage,1,0,122303,,,,0,96a773f4af32949c418dd2dd5ec3f6a38fecd3877bf98bb5d06b0914145952ba
thread:issue:122813,issue #122813: cpu performance for int4mm kernels,needs triage,47,needs triage,1,0,122813,,,,0,d02a5c50a58b01ae5fb0795e3d2a4bf2c104e61855261b737467dd36bd206cd3
thread:issue:122934,issue #122934: Cannot export DETR model with torch.export,needs triage,47,needs triage,1,0,122934,,,,0,f49b4f81f99a4205460b9eab667ac54594a200f6a35b3368bf291407b5e095c2
thread:issue:123066,issue #123066: [JIT archive] Add a flag to not include debug files,needs triage,47,needs triage,1,0,123066,,,,0,07cdf732bb45d8ca18841bdca5459457198d00a10c6cbfc7982f8e1bb37022b6
thread:issue:123168,issue #123168: TorchScript: Union cannot be used as a tuple,needs triage,47,needs triage,1,0,123168,,,,0,1da481b658599987fb67bcf020d3b2ce4d42a1f40c16c46009bc2af25e79be52
thread:issue:123345,issue #123345: torch.export does not support Sparse CSR tensors,needs triage,47,needs triage,1,0,123345,,,,0,7c17a9be0399900466789d9a311f8802b36573105d405b65fc6257eefb3ee730
thread:issue:123464,issue #123464: torch._C.JitType annotation is busted,needs triage,47,needs triage,1,0,123464,,,,0,d5a05733a893adafcde1dd16342c10f5da8547c6c4ed4f42df12d6c5f2c57275
thread:issue:123499,issue #123499: Support Exceptions in Pytorch Export,needs triage,47,needs triage,1,0,123499,,,,0,2805f9846bd690c58a1cbacba9a5dba873e8c62efcf0082c1cf47115c6df267b
thread:issue:123824,issue #123824: Precision problem in `quantize_fx.convert_fx` and question about the `example_inputs` for `quantize_fx.prepare_fx`.,needs triage,47,needs triage,1,0,123824,,,,0,36d27ac15e0895dc5ac7576223c61ec3e9ae227eef6eeace8cb246291dcf5577
thread:issue:123961,issue #123961: JIT reports RuntimeError when decorating forward with torch.jit.script,needs triage,47,needs triage,1,0,123961,,,,0,e31ae80b8c8825d9bf6bdb1c5fb271b2afa70b4714e58aa2ca3220b1335b5f9c
thread:issue:124190,issue #124190: Re-enable skipped 3.12 dynamo-wrapped profiler tests,needs triage,47,needs triage,1,0,124190,,,,0,aab09b35c6812a6e768cc4e01b44181fd2d2197d24615101fb633f7b63e2e524
thread:issue:124458,issue #124458: Understand the oneDNN graph fusion with torch script,needs triage,47,needs triage,1,0,124458,,,,0,7354be74670f3005f60df7d78662aee279c7588005200ef40c69c38dd23f5f74
thread:issue:124507,issue #124507: torch.export of ResNet with dynamic height fails due to constraint violation,needs triage,47,needs triage,1,0,124507,,,,0,d78eef2ee4ae2e00a81e09a80e35e742eb0899d34a1d9b53f23382cb0bed2127
thread:issue:124674,issue #124674: Improving format of communication metadata in PyTorch Execution Trace.,needs triage,47,needs triage,1,0,124674,,,,0,e3a28945c41265c87c992ae12d649da192c459ccf6b6992ce0dc7b00db7aea9e
thread:issue:124697,issue #124697: [INDUCTOR] [CPU] [GPT-FAST-MOE] large perf regression with coordinate_descent_tuning disabled,needs triage,47,needs triage,1,0,124697,,,,0,735e95e6e0533caae99a5847937ce82e8ba9418ad93cd8cf013d55f00b703346
thread:issue:124796,issue #124796: Dynamo Export Support for Qwen/Qwen-7B-Chat: Mutating module attribute _ntk_alpha_cached_list during export,needs triage,47,needs triage,1,0,124796,,,,0,49de2092eee4cd5f1f40bc38d09906336640d9f56766d102e719b57f9f951f8d
thread:issue:125003,issue #125003: bool as scalar in jit ir API not working as expected.,needs triage,47,needs triage,1,0,125003,,,,0,f04b4ae1cfa195670627f557e04f940b398b691d8ef6a49c4f35c981a058feb6
thread:issue:125004,issue #125004: RuntimeError: Unsupported qscheme: per_channel_affine. When using quantise_fx(,needs triage,47,needs triage,1,0,125004,,,,0,f924857452d38db76ac86bd6d34ab401e545d96795c2df27a5bf767494794698
thread:issue:125007,issue #125007: Support exporting the compute graph JiT compiled by inductor?,needs triage,47,needs triage,1,0,125007,,,,0,c8a447c4910223e052527b9c6a657c7ef93ec891d9661beec80452a68452817b
thread:issue:125033,issue #125033: Tensorboard's SummaryWriter.add_graph() doesn't work with packedSequence,needs triage,47,needs triage,1,0,125033,,,,0,2257a5c52992519b5c7bab1f16b0e6f8ecfae823938eba7a45584a6355603a24
thread:issue:125266,issue #125266: [export] linalg ops after core aten decomposition,needs triage,47,needs triage,1,0,125266,,,,0,7e0f3ecc7bc96821fb094143cb1ee611bcf392c96c999bcf0f05736054653cfe
thread:issue:125272,issue #125272: Profiler Kineto Config: Warning: Failed to recover relationship between all profiler and kineto events: 51361 vs. 0 reassociated. (function reassociate),needs triage,47,needs triage,1,0,125272,,,,0,85cf0d7525e366179308e432c174a1bfbf1190b17a17a1b26a46d2f90eb1b17f
thread:issue:125293,issue #125293: Need a C++ way to convert PyTorch model to TorchScript model,needs triage,47,needs triage,1,0,125293,,,,0,832784c30eef2a62012983d51a9787105e8254985009aed372ce176cd495133b
thread:issue:125564,issue #125564: Issues in loading quantized weights with Version==None,needs triage,47,needs triage,1,0,125564,,,,0,5f3297ba856e8f264eb64e112f3789737e5b9e4bc6bd3fcd1ef50b6634f6f0e6
thread:issue:125734,issue #125734: ☂️ `torch.compile` generates slower code for LLMs than eager on ARM platform (M1/AARCH64),needs triage,47,needs triage,1,0,125734,,,,0,0b63c03aa6d0adca74fa22606a90d3bb3f75ac1742357a9f04a2094b97749b58
thread:issue:125749,issue #125749: 'MultiHeadAttention.attention' is being compiled since it was called from 'MultiHeadAttention.forward',needs triage,47,needs triage,1,0,125749,,,,0,f5659800c9f8d4dabb16b112eec21cbc783835159ce15d1f22bc0b2a305d896b
thread:issue:126127,issue #126127: torch.export.export() fails with dynamic shapes when more than one shape is dynamic,needs triage,47,needs triage,1,0,126127,,,,0,11ad08c47c2e9874229e210ae351ab8308c5d270a5758fe3a3a7be702d895634
thread:issue:126140,issue #126140: [PT2E Quantization] _fuse_conv_bn_ | RuntimeError: !argument.default_value() INTERNAL ASSERT FAILED,needs triage,47,needs triage,1,0,126140,,,,0,a34cf5ddb39e15d685c000acb44b35faf00eac6d051f5c08e197dfb45bab8d9b
thread:issue:126167,issue #126167: `aten::transpose.int` is acting errouneously,needs triage,47,needs triage,1,0,126167,,,,0,08274a651b7cf4622c9a0ec9955e8d39b7366f6deac39a40526c6d0c9372bab6
thread:issue:126266,issue #126266: Quantization occurs with RuntimeError: `zero_point` must be between `quant_min` and `quant_max`.,needs triage,47,needs triage,1,0,126266,,,,0,ec29b53ecd9752b84a6dd5973bba3c90046f1232f777570b42e4d20844a334c9
thread:issue:126277,issue #126277: [JIT] Unexpected change in requires_grad attribute after calling JIT traced function with the profiling executor disabled,needs triage,47,needs triage,1,0,126277,,,,0,b062594b71ccd58e5ed36a4d80afe00903a6cd6c4ec274bf4cfb67f03a424ea4
thread:issue:126291,issue #126291: Pytorch -> ExecuTorch conversion fails with BatchNorm2d layer,needs triage,47,needs triage,1,0,126291,,,,0,a7a409f319afe644b5fdd81132984875685f35a6c96475ba82790af0caa9a5df
thread:issue:126703,issue #126703: PyTorch's Execution Trace Observer Producing Invalid JSON,needs triage,47,needs triage,1,0,126703,,,,0,12d59273427cfb3b510af9da01a6efd756cecef00973d8a8ed017bb7592464a6
thread:issue:126770,issue #126770: _convert_input_to_fake doesn't work when there are no inputs,needs triage,47,needs triage,1,0,126770,,,,0,6fe45813516960693116d5172a0920a8b5b58e541f0bd52bbaee9014cf011751
thread:issue:126843,issue #126843: Torchscript export of `Salesforce/blip2-opt-2.7b` model fails with Conv shape error,needs triage,47,needs triage,1,0,126843,,,,0,b3a01ff99daf45a64831f26227f4477261b9939175026ad7c0769a615f47c696
thread:issue:126974,issue #126974: TypeError: slice indices must be integers or None or have an __index__ method,needs triage,47,needs triage,1,0,126974,,,,0,38eb4d06bb5e1efb8e3b4783bde1594e64813806f652f90c80a16ecc50ea4c61
thread:issue:127013,issue #127013: [CI] Download android and ios dependencies from pip instead of conda,needs triage,47,needs triage,1,0,127013,,,,0,064d5379cc93ecd965282348595a32a08c1cddd76d08f5e9d7e57b890ac2f339
thread:issue:127076,issue #127076: [PT2E Quantization] `prepare_pt2e` produces inconsistent data types for primitive int,needs triage,47,needs triage,1,0,127076,,,,0,888482dbc6d9b05e703169da2522d47d23f08160027d79297c3f3541c6d8dfe8
thread:issue:127215,issue #127215: No way to config profiling scope in torch.autograd.profile,needs triage,47,needs triage,1,0,127215,,,,0,d457a701b70648f9b9b95848c2944c12ba0041519933feb23c60e2ee0d4d299f
thread:issue:127283,issue #127283: [BUG][JIT] `torch.jit.script` is not compatible with `DeprecationWarning` and `FutureWarning`,needs triage,47,needs triage,1,0,127283,,,,0,b956316d18ce6abfb152931597fc8195da31205384d1112c312674c759be6a02
thread:issue:127357,issue #127357: torch.export.export() throws out an error when dealing weighttying model.,needs triage,47,needs triage,1,0,127357,,,,0,3e45f9244af84b30b5797340ca21bb18008217a0f4a3f074984c620beb3f67ab
thread:issue:127571,issue #127571: [export] Cannot mutate tensors with frozen storage,needs triage,47,needs triage,1,0,127571,,,,0,844f469947ec217fd7ba08326a88a0cab605c16da96b11747ca93878b60dd4a7
thread:issue:127643,issue #127643: [export] Errors out when unflattening TorchTitan,needs triage,47,needs triage,1,0,127643,,,,0,78d18ea30209220668e0be09127d411b0fe384a22b33da51816e9906c09b984c
thread:issue:127679,issue #127679: RuntimeError: `jit.freeze` fails to find externally assigned attributes,needs triage,47,needs triage,1,0,127679,,,,0,7f0bd0da26d09aaa3a45211c7e9645cdc4e5d6ffd13e7251c27b126a0ca0767a
thread:issue:128019,"issue #128019: RuntimeError: r INTERNAL ASSERT FAILED at ""../aten/src/ATen/core/jit_type_base.h"":544 when using torch.jit.trace and torch._C._jit_interpret_graph",needs triage,47,needs triage,1,0,128019,,,,0,cc03c3d5de01359116caaf2e0e32025acdc0611aaa2ea52b5c0a8284b01d94a4
thread:issue:128114,issue #128114: Import Error: cannot import name 'XNNPACKQuantizer' from 'torch.ao.quantization.quantizer',needs triage,47,needs triage,1,0,128114,,,,0,8ce356e28e4f5f605a73ce9a4435695e0fd55f74dee0beae1f6cde5640371176
thread:issue:128125,issue #128125: optional memory leak with torch.jit.script on models with fft,needs triage,47,needs triage,1,0,128125,,,,0,f44a24e34eebeb335e6b4e69117076e035d2369e784c76280d90895dbae290d6
thread:issue:128261,issue #128261: JIT script can cause unneeded `requires_grad` inside `no_grad` blocks,needs triage,47,needs triage,1,0,128261,,,,0,d888d4e571aad83957c98a6437d19f58ece596dc13ed32d20fbaa928db966213
thread:issue:128292,issue #128292: put profiling results from different ranks into one webpage,needs triage,47,needs triage,1,0,128292,,,,0,640ddf0e3f9aeaa3bce173d68b5ee925402ac162ce01041f80fb85c3d36ab6d7
thread:issue:128385,issue #128385: [export] Llama2 export with dynamic shapes fails using Torch 2.4 nightly,needs triage,47,needs triage,1,0,128385,,,,0,b171290d53cc937febcce08fe9b2735705021bbe07651e2a6dcb06cf6829b828
thread:issue:128491,issue #128491: Improve type inference in torchscript,needs triage,47,needs triage,1,0,128491,,,,0,ae15ec060a5bb065a1917a8a7c3b053b6fb4abebc1193cee36e3bc80c86db2b0
thread:issue:128496,issue #128496: torch.jit.script not work for ParameterDict.items,needs triage,47,needs triage,1,0,128496,,,,0,7d56470f368a9a7f798475aa28b19a96d89c1c668c745de38d9949d9783697bd
thread:issue:128585,issue #128585: TorchScript JIT UnicodeDecodeError: ‘utf-8’ codec can’t decode byte 0xe0 in position 0: unexpected end of data,needs triage,47,needs triage,1,0,128585,,,,0,9ea8988f2ac6311d16bf1dfeb48372bc039a660582d08382ce9a0e49d72294f5
thread:issue:129282,issue #129282: Support for torch.Generator with JIT,needs triage,47,needs triage,1,0,129282,,,,0,bc804036455e4bb77e4eab7c1fd861faada1a1f2e2b5e355d8b2b4ceaccc41b6
thread:issue:129347,issue #129347: Error when loading a module file by calling torch::jit::load(..),needs triage,47,needs triage,1,0,129347,,,,0,76bfcfdc06563c065c74499bc4ca02cecbdf2694ba8c708a16ae574cbc85e55c
thread:issue:129633,issue #129633: pytorch_android-2.1.0.aar/jni/arm64-v8a/libpytorch_jni.so: invalid local symbol '__bss_end__' in global part of symbol table,needs triage,47,needs triage,1,0,129633,,,,0,0acafd6d44af731e4ada1370ddda86f03a01f785b9494c5445c579a291fcda1a
thread:issue:130229,"issue #130229: RuntimeError: 0 INTERNAL ASSERT FAILED at ""../torch/csrc/jit/ir/alias_analysis.cpp"":615, please report a bug to PyTorch. We don't have an op for aten::full but it isn't a special case. Argument types: int[], bool, int, NoneType, Device, bool,",needs triage,47,needs triage,1,0,130229,,,,0,b89821e02ad09d1c0262efd83ff6968a5348fd208e172fe606c72847951e6f3e
thread:issue:130309,"issue #130309: 【Pytorch mobile android】torch jit forward fail,Calling torch.geqrf on a CPU tensor requires compiling PyTorch with LAPACK. Please use PyTorch built with LAPACK support.",needs triage,47,needs triage,1,0,130309,,,,0,feb1a130cda6a0bae25bf6623845123742278443129722766546e1757041df60
thread:issue:130407,issue #130407: [RFC][c10d] a new Pytorch API (split_group) to create a process group through ncclCommSplit,needs triage,47,needs triage,1,0,130407,,,,0,23c645e09bbcc043b61b18437440502fcf4dd34cd91ae9ee3381d32b69c537fa
thread:issue:130555,issue #130555: torch.export.export re-export preserving metadata for output node,needs triage,47,needs triage,1,0,130555,,,,0,71c86971b4637d968c1c3c2356bf0bc2be8594c5b8cec734cd2b916ed2b1bc16
thread:issue:130726,issue #130726: TorchDispatchMode fails on jit.trace,needs triage,47,needs triage,1,0,130726,,,,0,eedbbbf92cf61ce2699485e57b9ba13edcc41647634596b15d25cc46ed70a335
thread:issue:130735,"issue #130735: Problem with combination of jacref, vjp, tanh in torchscript",needs triage,47,needs triage,1,0,130735,,,,0,5f6c2e7849d7fe694844283fe2b0e798ae961681eaa2dc3c772d51f135d6460d
thread:issue:130820,issue #130820: [Export] Mutating module attribute error during export on YOLOX,needs triage,47,needs triage,1,0,130820,,,,0,72ad7d9ea6c883b2e50de7a25f2fec7f557be3a5eefc04fcfb2a9cff0c4acdce
thread:issue:131022,issue #131022: Code bug in call_prepare_scriptable_func_impl function.,needs triage,47,needs triage,1,0,131022,,,,0,8700b7053a591ae9bf704d064f18d931e12a55ef3199dfdeb36f18f8e9e5c581
thread:issue:131055,issue #131055: PT2E XNNPACKQuantizer has unused operator_type_config and set_operator_type,needs triage,47,needs triage,1,0,131055,,,,0,9bb1bcdee0f59bd244c082d987201f6ec74b41fd838b8b715360cc719d76c24c
thread:issue:13164,issue #13164: Tracing custom ops,needs triage,47,needs triage,1,0,13164,,,,0,e863cabb10f5757d8d1c39a947e532252c236dd8cf22dc0a20c12b015182e464
thread:issue:13170,issue #13170: Allow traced modules to return dictionaries,needs triage,47,needs triage,1,0,13170,,,,0,85279d07df3df8b824c82ed90d37a6fe72f7393e1ed21d1a927313dc4500024c
thread:issue:132009,issue #132009: 2:4 semi structured sparsity matmul not supported on hopper,needs triage,47,needs triage,1,0,132009,,,,0,932bfcf6fec4c9f86af6519ebe6f51e2d995204742164371f4be882dea9c71a8
thread:issue:132029,issue #132029: torch.QUInt4x2Storage: RuntimeError cuda_dispatch_ptr INTERNAL ASSERT FAILED on torch 2.5.0.dev20240708+cu121,needs triage,47,needs triage,1,0,132029,,,,0,6c8ca992735abeef284d06ab224df2d2b0995c2df4c09b794959c9eef3c8b79f
thread:issue:132240,issue #132240: torch.nn.functional.interpolate completely broken with torch.jit.script and torch.fx,needs triage,47,needs triage,1,0,132240,,,,0,699e65744b3f6c2faa066955bf1da4a3c76c111d745e383a7d06eda9722f498d
thread:issue:132315,issue #132315: Support Enum for dictionaries in TorchScript,needs triage,47,needs triage,1,0,132315,,,,0,1fabd620dfe937e44177d7d0b1fca73a3c01ee97f4c78dd0289d75d833a5c6f0
thread:issue:132385,issue #132385: crnn model from pth to torchscript problem,needs triage,47,needs triage,1,0,132385,,,,0,aed87121291ff1f03fc7d846576ca04b7e823b1a8483d2c754c727a5eb2fd077
thread:issue:132459,"issue #132459: [export] Implement a common ""convert dtype to int"" function",needs triage,47,needs triage,1,0,132459,,,,0,7b783bc90d33f346a33b4753c68aeba43dae2b9872937a4a0e479816fb31da90
thread:issue:132738,issue #132738: Profiler getProfilerConfig interface can not support Ascend npu platform,needs triage,47,needs triage,1,0,132738,,,,0,fec65249fab4d489c7494c64c5530a30b4857408c44162a426b3cfbebb7522ed
thread:issue:132930,issue #132930: Treating SymNode compute more fungibly for the purpose of export flattener (or not?),needs triage,47,needs triage,1,0,132930,,,,0,2ddcf62bd6fc3514623b11b655afe2bb9cc266fa887a2c2db6bc52af710857c0
thread:issue:133255,issue #133255: torch.export ExportedProgram from GPU cannot be loaded on a CPU only machine,needs triage,47,needs triage,1,0,133255,,,,0,35b1b5c0ad6e051a509df21cd7d2f40277518d2bc235cbef22def1b833fc5886
thread:issue:133444,issue #133444: Collect tensor shapes only when using record_shapes in profiler,needs triage,47,needs triage,1,0,133444,,,,0,61cdeb441ca51fffa71416d16fd68e8609f0a65183e5a88b58ef4d468c19401c
thread:issue:133537,issue #133537: `torch.jit.script` fails on a simple function,needs triage,47,needs triage,1,0,133537,,,,0,632f3711c0251e5c62d3e9a8925b151128a13ce1084e583c939bf8318618c625
thread:issue:133552,issue #133552: fx qat symmetric config zero_point not zero,needs triage,47,needs triage,1,0,133552,,,,0,93ee1a0ffe2f3ed6025072c73ad2e5fa3aea753c048de80bf48d4279132e49c9
thread:issue:133557,issue #133557: torch quantize error,needs triage,47,needs triage,1,0,133557,,,,0,f8731f8dabb687ade17106b875417d619014f4be13d1f02f176258df830c28c4
thread:issue:133675,"issue #133675: [torch.jit] RuntimeError: false INTERNAL ASSERT FAILED at ""../torch/csrc/jit/ir/node_hashing.cpp"":148, please report a bug to PyTorch.",needs triage,47,needs triage,1,0,133675,,,,0,0d180b6d0fcdcb04fd63dadd8a312ed14abbdcca3e430afe1b07cbcdb59bebdf
thread:issue:133678,issue #133678: [torch.jit.trace] error encountered when tracing model with weight clamping during forward,needs triage,47,needs triage,1,0,133678,,,,0,6bf2979fd4df3e956afe1f553b2e2efd006e3415ef47f2a9400e5e5a1c52f1fe
thread:issue:133795,"issue #133795: RuntimeError: createStatus == pytorch_qnnp_status_success INTERNAL ASSERT FAILED at ""../aten/src/ATen/native/quantized/cpu/BinaryOps.cpp"":204, please report a bug to PyTorch. failed to create QNNPACK Add operator",needs triage,47,needs triage,1,0,133795,,,,0,011c3c360df42d32aae00e998b4212713765ee352a590347893a8322161dc77f
thread:issue:133840,issue #133840: Pytorch Mobile - Different output on Android and PC with the same model,needs triage,47,needs triage,1,0,133840,,,,0,42a96c512e2fabf4762dba133d1f0191e9d877bcfbc75e26aab53ce99a98eb44
thread:issue:133847,"issue #133847: [Bug Report] when using % in torch.jit.script function, the output conforms to the form of C++ % operation, rather than Python % operation.",needs triage,47,needs triage,1,0,133847,,,,0,a316d444655092d21de20cbef16114b150c1a969fcba5f8b83eaf578ab6910d6
thread:issue:134220,issue #134220: [Operator Benchmark] [CPU]OP: weight_norm_sparsifier_step functionality failed with parts of long config shapes,needs triage,47,needs triage,1,0,134220,,,,0,515a7f9589915faa3522256c5c419e58bc6b1efdd7adfb1604ef62a1769871e9
thread:issue:134303,issue #134303: [FakeTensor] Error: SymIntArrayRef expected to contain only concrete integers in UpSample2D block,needs triage,47,needs triage,1,0,134303,,,,0,aa6d4424e23df99a645b56b3ddbe1ac233a60309c16aa6f53fcf4dccdeb28cd5
thread:issue:134559,issue #134559: [JIT][foreach_add] Arguments for call are not valid,needs triage,47,needs triage,1,0,134559,,,,0,f31a4fb2c13a556790bb007bc41847fc3e61763ea184bf3c7d26c2f3d213ba9e
thread:issue:134562,"issue #134562: ScriptModel Save ERROR, when I convert part of the model to TRT",needs triage,47,needs triage,1,0,134562,,,,0,7d0579f019c9cd26a505be9b12f7197c0514c4dddbff279eaf9ca0298669030b
thread:issue:134679,issue #134679: [CPU] jx_nest_base float32 both inductor and eager performance regression in 2024-08-26 nightly release,needs triage,47,needs triage,1,0,134679,,,,0,2f7d9261c02a8e11a3041bb42b075219ed556509e903f3b7d98ce5b1eb9357ca
thread:issue:13477,issue #13477: [jit] restrict promotion of single-element arguments to lists,needs triage,47,needs triage,1,0,13477,,,,0,dce896935622c0df9da8900ed85e75231ed2469b0ba29e29e2d22f442dbfc1e1
thread:issue:134849,issue #134849: torch::jit::load invalid_path error,needs triage,47,needs triage,1,0,134849,,,,0,d21f4fe528e506d827ecf0f0401f13a21c7f6c547f672b146338f063a646bed3
thread:issue:134898,issue #134898: Segmentation fault would be triggered when using `torch.jit.script` and `torch._C._jit_shape_compute_graph_for_node`.,needs triage,47,needs triage,1,0,134898,,,,0,1e8b1dcdf555d6f13e28768e79bcad8e0036b68af2a0f42255839569ee75cae8
thread:issue:134903,issue #134903: [torch.jit] Crash would be raised when using TorchScript modules,needs triage,47,needs triage,1,0,134903,,,,0,6323b9ee7f7d6da5857aad6d841d8a1f20cdb0bf46ebfe2d6e3907892b9cac5b
thread:issue:134918,"issue #134918: DYNAMIC_TRT_MODEL_CONVERSION 0 INTERNAL ASSERT FAILED or, comfyui made me come here",needs triage,47,needs triage,1,0,134918,,,,0,6209e3e9225d86feef44587ab0e470bc0d536d0dad10487c2dd3f096268897cc
thread:issue:134922,issue #134922: [torch.jit.script] An INTERNAL ASSERT error will be raised when permuting the inputs of a jit scripted graph with mismatched indices.,needs triage,47,needs triage,1,0,134922,,,,0,a0702fcd9ab808400b50b23ed805d1c350cdb8ab1bd0056845b0e6aa100e7bdd
thread:issue:134923,issue #134923: A crash would be raised when using `._StackString` with empty lists,needs triage,47,needs triage,1,0,134923,,,,0,88d4b951b4f7c25f59916d3e03eca09f4b82da0a9c33fae0df3d2a5d10bdc2b4
thread:issue:134924,issue #134924: `TorchScriptTesting._PickleTester` caused a segmentation fault with empty input lists in a jit traced function,needs triage,47,needs triage,1,0,134924,,,,0,6b18ee95a1a563c847d2e0edd6d65b178f06de1a13de50f243b73c9844f564ac
thread:issue:134937,"issue #134937: [torch.jit.script] INTERNAL ASSERT FAILED at ""../aten/src/ATen/core/jit_type_base.h"":556, please report a bug to PyTorch",needs triage,47,needs triage,1,0,134937,,,,0,2c17524895d6ce12ec7246b076bd19a0aee4b302343c01439d17653cf977eed8
thread:issue:134971,"issue #134971: [torch.jit] INTERNAL ASSERT FAILED at ""../torch/csrc/jit/passes/symbolic_shape_analysis.cpp"":741, please report a bug to PyTorch. input_shapes size: 1 n inputs size: 2",needs triage,47,needs triage,1,0,134971,,,,0,7228ccd47f99675c511419a7b604b8b63bcbe9bc4bbbe3e6926ae18e84fd94fc
thread:issue:135003,"issue #135003: [BUG]: torch.profiler is blocked in ""_disable_profiler()"" when profiling cuda graph on single GPU",needs triage,47,needs triage,1,0,135003,,,,0,683f54028f5703bc2114287599c294e73807a8cf6af44164fbcdb1d9e0b18144
thread:issue:135569,issue #135569: [inductor][cpu]AOT inductor performance regression and accuracy failure in 2024-09-04 nightly release,needs triage,47,needs triage,1,0,135569,,,,0,3510b8df3741680ca2714ff952559661e53d5a8d4edf43b5abc8334459baf66a
thread:issue:135580,issue #135580: Inconsistent results between libtorch and pytorch for model with shared weights,needs triage,47,needs triage,1,0,135580,,,,0,4f9fd79b8f5b3c5eb4607cdad11acc31c35e479acb58d50ca87155b6c99ea8eb
thread:issue:135689,issue #135689: [inductor][cpu]GPTNeoForSequenceClassification AMP single/multiple thread static/dynamic shape default/cpp accuracy failure,needs triage,47,needs triage,1,0,135689,,,,0,0d4a11ba6590656fd9ee23025788deee5215de30a9d0e76b83394ff4794115ae
thread:issue:135690,issue #135690: [inductor][cpu] hf_BigBird float32 multiple threads performance regression in 2024-09-08 nightly release,needs triage,47,needs triage,1,0,135690,,,,0,4fb92c0a4b9bb113aa8fc645b5d6b48c9d6496f0b8eae70ed6b7542f2295e82e
thread:issue:136094,issue #136094: Quantized model is way slower than regular model.,needs triage,47,needs triage,1,0,136094,,,,0,27db7f4e9fe3841dc03c21aa9ab1d78b47bec545cbffd87da377f27481d62d42
thread:issue:136127,issue #136127: A division-by-zero operation in a scripted function can cause the core dumped,needs triage,47,needs triage,1,0,136127,,,,0,31b217957126afb042b1c15903312a58a672daf4cd62f74ae0fa2a9877b81158
thread:issue:136256,issue #136256: TorchScript raises an OSError when using frozen dataclass,needs triage,47,needs triage,1,0,136256,,,,0,53a2a57074fb9f39f2351bcfc9b4468d079cee0bf42f7b4dbc7144cf07c7133d
thread:issue:136618,issue #136618: [inductor][cpu]pyhpc_isoneutral_mixing singel thread AMP cpp wrapper performance regression in 2024-09-22 nightly release,needs triage,47,needs triage,1,0,136618,,,,0,114c36780101dea3c41f2b85d0f1d4b9618a33bc189518e83b32afc57cb65fbb
thread:issue:136654,issue #136654: [TorchScript] typing_extensions.deprecated doesn't work,needs triage,47,needs triage,1,0,136654,,,,0,30b793fb5ed8e41021c0c6dcead669115c9fc25fdc1ecc8d3c9e1c02085a3272
thread:issue:136828,issue #136828: false INTERNAL ASSERT FAILED in `torch.empty`/`torch.ones`,needs triage,47,needs triage,1,0,136828,,,,0,b62934674c722c377d20f16ff1a2a1a011a9504abfc497a3d18ccb1d1eed3927
thread:issue:136829,issue #136829: false INTERNAL ASSERT FAILED in `torch.jit.set_fusion_strategy`,needs triage,47,needs triage,1,0,136829,,,,0,4a87d43673d262aa7a38001be62101f0eb45f8c8673821d6d997c38cc63a58e2
thread:issue:137202,"issue #137202: Met error ""we don't have an op for aten::full but it isn't a special case."" when running microsoft/Florence-2-base",needs triage,47,needs triage,1,0,137202,,,,0,3a9b3e44c91ee2316d0f7051292be5ab22597b32f7409b25bfe539c7aeaea988
thread:issue:137561,issue #137561: Segmentation fault would be triggered when using torch.jit.trace and torch.jit._fork,needs triage,47,needs triage,1,0,137561,,,,0,5257306706d1fdbd6b8d63b71a9cfabd9c1e728e6ed4930298c564fafdebfb0a
thread:issue:137691,issue #137691: [inductor][cpu]float32 and AMP performance regression in 2024-09-30 nightly release,needs triage,47,needs triage,1,0,137691,,,,0,20eaa52e85c105e070ada3f0a61e0e3788fe6efffd5b2dcc7a11931fe48dedd6
thread:issue:138193,issue #138193: [profiler] are CUDA events mutually exclusive with _ExperimentalConfig.profiler_metrics?,needs triage,47,needs triage,1,0,138193,,,,0,971ae6f141eb3eb61da82ee4e0fa93932410700ec5237de2a2fabf3ab0095b60
thread:issue:138341,issue #138341: `torch.package` bug,needs triage,47,needs triage,1,0,138341,,,,0,ccdebd1e10ca90a7bdd1b005e76603d9404a6fc7cf351fdc29b8e1cc9beff29d
thread:issue:138820,issue #138820: RuntimeError: HalfTensor is not supported,needs triage,47,needs triage,1,0,138820,,,,0,173ee0589b4e75c3d7c65c393b21d642611c6a411147818ab9cca3febbfac27d
thread:issue:138973,"issue #138973: When using torch.jit.trace for inference, the element of the shape attribute of tensor and the return value of numel() are tensor",needs triage,47,needs triage,1,0,138973,,,,0,4ef308d4a40ee9cc007ce8ba9836e5896cbda758932b3f2c0cecbb26c4a65d9b
thread:issue:139063,issue #139063: Quantization API examples in the docs are outdated,needs triage,47,needs triage,1,0,139063,,,,0,5e01abbc8731b2d3b1434cc43be5ae2b2d74d2da526849fa570ff3066b7d7773
thread:issue:139297,issue #139297: `torch.package` warning -- `TypedStorage` is deprecated,needs triage,47,needs triage,1,0,139297,,,,0,072d65aeeda950c850cb4481a04de3ff162f55c8ed050c37214d335d17b4c549
thread:issue:139464,issue #139464: Maybe there is a precision issue with torch.quantize_per_channel?,needs triage,47,needs triage,1,0,139464,,,,0,be072a1d825c2f11dfbe9af719e127d26bb1b57cf2c2aea54a0be949cad4239a
thread:issue:139470,issue #139470: serialization of PT2E model impacts torch.fx.passes.utils.source_matcher_utils.get_source_partitions,needs triage,47,needs triage,1,0,139470,,,,0,8566117ca5158a18215ddffeafaed8dfb7953dfdd66a40afef95ab4ea37124e3
thread:issue:139556,issue #139556: [JIT] ModuleList getitem error when saving and loading with torchscript,needs triage,47,needs triage,1,0,139556,,,,0,3f3030c737a81623358b8d6aba6f619085f2403904fe201aba718faa7c7f1d00
thread:issue:140101,issue #140101: Inconsistent gradients in torch script function,needs triage,47,needs triage,1,0,140101,,,,0,be807d0ed42e39a055c7c87150b150833d940b4946ea71d23c15565eb30c36df
thread:issue:140468,issue #140468: [TorchScript] Failure if you script a wrapper module and then an interface-implementing submodule.,needs triage,47,needs triage,1,0,140468,,,,0,ddf28afcc6ac4c8ea3fc66507797064382dce37a724d53417b0f708734d5a825
thread:issue:140539,issue #140539: torch.jit.trace doesn't trace all operations completely (no data-dependent control flow),needs triage,47,needs triage,1,0,140539,,,,0,87ec816589715143f4daad015a14cbbbab2ad803f0767464bc1e5b21f0563c72
thread:issue:140684,issue #140684: Segmentation fault (core dumped) in `torch.autograd.profiler.profile`,needs triage,47,needs triage,1,0,140684,,,,0,5bd6a0c5c04a0e272fd3822857b2797f0fffb9d00a8869a89ea0aa0808e7ef4f
thread:issue:140689,issue #140689: [inductor][cpu]opacus_cifar10 and functorch_dp_cifar10 AMP performance regression in 2024-11-11 nightly release,needs triage,47,needs triage,1,0,140689,,,,0,64fa772000c43deb6b4907552e7ff4f4cd6e65c7304445df7eddd3023efd7d7d
thread:issue:140702,issue #140702: `Tensor.prod` gives a jit warning,needs triage,47,needs triage,1,0,140702,,,,0,50d3972c63301a7855b663c02b1671796556ed047e33f77e44c3311fb23dc38a
thread:issue:140805,issue #140805: [AOTI] Test case `test_aot_inductor.py::test_multi_device_cpu` is Running GPU AOT Inductor in CPU container runner.,needs triage,47,needs triage,1,0,140805,,,,0,28f6e1c96a06b980caaae047652954ecf4e045bd0ceca88795d25ea259d53ad2
thread:issue:141116,issue #141116: How to fuse batchnorm to conv2d in the graph exported by torch.export,needs triage,47,needs triage,1,0,141116,,,,0,0fa601e3cb89a01e5a47c7261a4f7f657bba334ffac4c187ed9ae7db848613a0
thread:issue:141854,issue #141854: RuntimeError: quantized::conv2d_relu (ONEDNN): data type of input should be QUint8.,needs triage,47,needs triage,1,0,141854,,,,0,670e543ae5e84624e5e6e635875cc823feff361e8facbe879fb840250bf7856e
thread:issue:141855,issue #141855: Can not convert Llama-3.2-11B-Vision-Instruct by torch.jit.script / torch.jit.trace,needs triage,47,needs triage,1,0,141855,,,,0,250b6d9702901f8ec84d60e1882508f9582b7a8cda1cb4554d1b89f12b924b81
thread:issue:141893,issue #141893: Forward from pytorch/executorch: export error with Macbook pro M4,needs triage,47,needs triage,1,0,141893,,,,0,6c7bed69160e8f99392b22c0896717c301a8b3a44610a2d1d3a5790479a424a6
thread:issue:141996,issue #141996: [export] Move pytree serialization into export serializer,needs triage,47,needs triage,1,0,141996,,,,0,5ff66595c5c9e1ba1256093c5c36b27326a15888531215282755e5efd425b5e1
thread:issue:142170,issue #142170: [3.13] Fix failing torch.package dynamo-wrapped test,needs triage,47,needs triage,1,0,142170,,,,0,5f52eae39029de7e9566f8ab9ff2ad3ae10339b5b5cf8e37514e97df3b643c30
thread:issue:14270,issue #14270: Pytorch C++ API with cuda : Expected object of backend CPU but got backend CUDA for sequence element 1 in sequence argument at position #1 'tensors',needs triage,47,needs triage,1,0,14270,,,,0,c2061422c3c80d2c478a6e73386f7897a9f3284e0e28249e614ae555f05c4829
thread:issue:143289,issue #143289: RFC: Dynamically Quantized 4 bit matmul API and usage,needs triage,47,needs triage,1,0,143289,,,,0,a59aa4ca919f877e4146ac8dbf3be1a32f748dde9330ead3435a3968b234e704
thread:issue:14332,issue #14332: [jit][script] support slicing with tensor literals,needs triage,47,needs triage,1,0,14332,,,,0,76837247b54bbd0d4ac42ce7569fdf31498fb846dcb38e2cbdbce54053785403
thread:issue:143379,issue #143379: Running decompositions over torch.export program adds one more output to the graph result,needs triage,47,needs triage,1,0,143379,,,,0,e2170a5225bb37bc237986b968748300b74a03e7a12f617688ea189ec123ee88
thread:issue:143565,issue #143565: Failed to load TorchScript of SSDLite in android,needs triage,47,needs triage,1,0,143565,,,,0,84ade48c73fc4235b4d374055238e26e3de3d63bcbf7b0abedeee6b0d602d6ba
thread:issue:143831,"issue #143831: After pth is converted into ptl, the prediction result is very different from pth",needs triage,47,needs triage,1,0,143831,,,,0,aa869889e5648b965aa4c7642ebcbe547ea5a8765c1e814639bc45894de17b87
thread:issue:143854,issue #143854: Can't script a tensorrt model,needs triage,47,needs triage,1,0,143854,,,,0,59eb561bfde084b9143abca8d78c12bdd3044daea3e9a651b679217232025ce8
thread:issue:143857,issue #143857: `@torch.jit.script` causes `pytest-cov` to miss function body,needs triage,47,needs triage,1,0,143857,,,,0,eb66e46fee10e6c382b3acb4f4ae2750cb2adea7c41f6b250ea37d0cafe7f485
thread:issue:143886,issue #143886: [RFC] Identifying dynamic int8 symmetric vs asymmetric quantization of activation/input in Inductor-CPU,needs triage,47,needs triage,1,0,143886,,,,0,e5f437cfdac1c23684451d2c4718f6511f37737af099b608e7adc942d8d68718
thread:issue:144039,issue #144039: [Mac/M1] torch.compile() -- expm1 returns an inaccurate result compared to the interpreted version,needs triage,47,needs triage,1,0,144039,,,,0,52a8166b985bbe60a495a757537e00b3c607eeca15c90d298496e640fde109dd
thread:issue:144148,"issue #144148: Unable to parse and visualize ""trace view"" with tensorboard",needs triage,47,needs triage,1,0,144148,,,,0,842fe9dca0ec64f9f33d399b1f14e68f3b14e2aee97ccf834478b9438f516cda
thread:issue:144331,issue #144331: [Export] UserWarning: Attempted to insert a get_attr Node with no underlying reference in the owning GraphModule,needs triage,47,needs triage,1,0,144331,,,,0,50a4d18861814f80472670f4d7f7fa7056c1aa0d4897329ded2e16aaf7370904
thread:issue:14455,issue #14455: [JIT] Tracing a script function/module where not all args are Tensors,needs triage,47,needs triage,1,0,14455,,,,0,191951961eb771349011fb14d4c6c20d803440e2cb08e3509ff0668fe1f0cda7
thread:issue:144613,issue #144613: The label marked by torch.profiler.profile.record_function() appears twice in the output,needs triage,47,needs triage,1,0,144613,,,,0,230431819f931703e36cffe12b789282b9c2c4b5f4af585fb93e82306104c680
thread:issue:144770,issue #144770: [ARM] multiple test failures in TestQuantizedConv on Aarch64,needs triage,47,needs triage,1,0,144770,,,,0,f41932608151de7a66c0de87276221d60772064679d83410441350b5139fa112
thread:issue:144778,issue #144778: `torch.profiler.record_function` doesn't register kernels from `backward` function,needs triage,47,needs triage,1,0,144778,,,,0,9ab0302ea1afee6ae6fa285d7bd02c3210154e91b601c4df5905ccba05ee55cb
thread:issue:14484,issue #14484: prim::ConstantChunk derivative formula doesn't handle undefined inputs,needs triage,47,needs triage,1,0,14484,,,,0,63d61c5ffb1fc1b6433caf28bcd846ec2a3deb5e5a72c144da931135db1f2f7f
thread:issue:145394,issue #145394: [EXPORT AOTI] `aoti_compile_and_package` custom_ops dependecies,needs triage,47,needs triage,1,0,145394,,,,0,46c13b04faf7a176b02ccd2f827dbd0d4326c01a10b63056f06e6763149866db
thread:issue:145456,issue #145456: torch._dynamo.exc.Unsupported: Graph break due to unsupported builtin torch._C._dynamo.eval_frame.set_eval_frame.,needs triage,47,needs triage,1,0,145456,,,,0,81edbc3c2d43c742785b1c00fb647b4657bf40ab657521e4ea27591d2cfe0568
thread:issue:145743,issue #145743: TorchScript run_method fails from 2.5.0 onward on Ubuntu,needs triage,47,needs triage,1,0,145743,,,,0,2a750cbde092b93cafa7ccdca0216265339b5329a94943a3776640acff72b73f
thread:issue:145995,issue #145995: torch.jit.trace does not move tensors to GPU.,needs triage,47,needs triage,1,0,145995,,,,0,c3ea50a0edecc5a9fb32590b1670550c249b0b19303b24b52eb964c58ad8d2e1
thread:issue:146127,issue #146127: When I try to run rt-detr model in C++ libtorch i face the given error,needs triage,47,needs triage,1,0,146127,,,,0,0942a82aadf2fb7cad091d156a019966602f70fdf3b7ae6cc8ff39cfbd4526de
thread:issue:146315,issue #146315: [export] dynamic_shapes with `Dim` fails when `DYNAMIC` succeeds,needs triage,47,needs triage,1,0,146315,,,,0,facbb3d15a08646d108cab6739cbba8eedd199e9be66944835d4559efc3a8287
thread:issue:146483,issue #146483: [ARM] Unit test failure on Neoverse-N1 - CPUReproTests.test_lstm_packed,needs triage,47,needs triage,1,0,146483,,,,0,b1c56a624c2fd3da6db420ba701b4bd7611e5b6aadd3933a9d1e4fdb28771ce4
thread:issue:146576,issue #146576: Torch Dynamo Export Failed on RetinaNet from Torchvison,needs triage,47,needs triage,1,0,146576,,,,0,92bbad1859a953189c514493b0f1c18ab8a008eb78b9f7907d1aa296576abce8
thread:issue:146698,issue #146698: Possible memory leak when running test_model_exports_to_core_aten.py test,needs triage,47,needs triage,1,0,146698,,,,0,0478268ff330af66c37bc9cf991b2c98666f905131fab0b2043a707af02186fe
thread:issue:146757,"issue #146757: [torch.jit.script] INTERNAL ASSERT FAILED at ""./torch/csrc/jit/ir/ir.h"":505, please report a bug to PyTorch",needs triage,47,needs triage,1,0,146757,,,,0,dae8cc192911a919a6355603f6103388fce83bbbd422b50dc2665c40b96ce0ab
thread:issue:146758,"issue #146758: INTERNAL ASSERT FAILED at ""/pytorch/torch/csrc/autograd/functions/utils.h"":74, please report a bug to PyTorch",needs triage,47,needs triage,1,0,146758,,,,0,5d598df9e290d02e09aaa2529795f40cb9354bd14a02d2efabea3c5345fe309e
thread:issue:146759,"issue #146759: INTERNAL ASSERT FAILED at ""../torch/csrc/jit/ir/alias_analysis.cpp"":617, please report a bug to PyTorch.",needs triage,47,needs triage,1,0,146759,,,,0,57c7b3412191045bcc0aab2fb4790566de5a22199851681607e5dd33bb50c25f
thread:issue:146760,"issue #146760: [torch.jit] INTERNAL ASSERT FAILED at ""../aten/src/ATen/core/ivalue_inl.h"":1967, please report a bug to PyTorch.",needs triage,47,needs triage,1,0,146760,,,,0,c747b4869093fcf3ea12cd6636c20522c57637041a743d83ba92aa8eb7ac313b
thread:issue:146773,"issue #146773: INTERNAL ASSERT FAILED at ""/pytorch/aten/src/ATen/native/quantized/cpu/qsigmoid.cpp"":65, please report a bug to PyTorch.",needs triage,47,needs triage,1,0,146773,,,,0,30f5d991281118140eef6ba7543959a5cf870c0409cf4d1a0268529c4c2ac003
thread:issue:146775,issue #146775: [torch.jit] Crash would be raised when using torch.jit.script,needs triage,47,needs triage,1,0,146775,,,,0,afb3a6e5351f64b3812a68370e195f8947246a2833b195a554361cb8b20321d3
thread:issue:146776,"issue #146776: INTERNAL ASSERT FAILED at ""/pytorch/torch/csrc/jit/testing/file_check.cpp"":607, please report a bug to PyTorch",needs triage,47,needs triage,1,0,146776,,,,0,2a44ada9a6d9d9940400ab96dce2339b8693a9617df89fd75d83711150595736
thread:issue:146778,"issue #146778: [torch.jit] INTERNAL ASSERT FAILED at ""/pytorch/torch/csrc/jit/mobile/register_ops_common_utils.cpp"":34, please report a bug to PyTorch.",needs triage,47,needs triage,1,0,146778,,,,0,c63d78a00016a0ce4a10e27783b4e0f059eb4d08c2840fdabed8518ef3531efb
thread:issue:146824,issue #146824: Inductor-CPU might load (and store) fewer elements than the vector-width,needs triage,47,needs triage,1,0,146824,,,,0,d21f97e189aefda4d652efaf9a193a4127341ec84bd9b70e0b85004a78b05f91
thread:issue:146914,issue #146914: TestSelectAlgorithmCPU.test_int8_woq_mm fails on Aarch64,needs triage,47,needs triage,1,0,146914,,,,0,1fe78082038ac0b5ceb7496dee4da7c5dd0b2fa851c6a96a817770b49a95b091
thread:issue:146915,issue #146915: Unit test failure on Aarch64 - TestSelectAlgorithmCPU.test_linear_with_in_out_buffer,needs triage,47,needs triage,1,0,146915,,,,0,c5ce1bee5d338958cfee0bcabd0d023b02d0b6baa38686b26a4fdb6cbc29c621
thread:issue:146974,issue #146974: [inductor][cpu]detectron2_fcos_r_50_fpn multiple thread accuracy failure in 2025-02-10 nightly release,needs triage,47,needs triage,1,0,146974,,,,0,bedb7f5d7b26fa4bf37581b79fd6ba56fd387eaa960d2e2ddbe5c4b1b7347eaf
thread:issue:147055,issue #147055: INTERNAL ASSERT FAILED or SEGFAULT when JITting a function that can return different types,needs triage,47,needs triage,1,0,147055,,,,0,4f21ceed8b2efd036c32fb6e0fb30f263189e594369347ec749ffc0e1f3c1753
thread:issue:147112,issue #147112: Add quantized version of BatchNorm1d module,needs triage,47,needs triage,1,0,147112,,,,0,de3f420d882306b8884a49cc1c44b859e2179484f390f79326b180dcf1bd73f6
thread:issue:147187,issue #147187: [torch.export] How to export a model with kv cache,needs triage,47,needs triage,1,0,147187,,,,0,73e39ad160e1c4ff181ac72b2c1a98e13443a3f676a6b7082c7576c62c7f2972
thread:issue:147444,issue #147444: NaN values originating from the normalize_weight_jit function during the backward pass,needs triage,47,needs triage,1,0,147444,,,,0,bca32577a4554aa3dd1d473c9e59294c411c4e91619f995b30098992b0f97808
thread:issue:14750,issue #14750: [caffe2] ConvTranspose with group attribute,needs triage,47,needs triage,1,0,14750,,,,0,3c08ae6eed3408b7ef1f0589b04af99174467b610b1ff5b2d9c7261a14137611
thread:issue:147604,issue #147604: RuntimeError when running profiler in a loop,needs triage,47,needs triage,1,0,147604,,,,0,a638243c7abc3e79213148bae790be48f655af7885b4314e922ba08bedd5895c
thread:issue:147722,issue #147722: Segmentation fault in `torch.ops.profiler._call_end_callbacks_on_jit_fut`,needs triage,47,needs triage,1,0,147722,,,,0,a7a952010fbbb4305f16277a8f8938563dbd45c414cecfad49bce0bcb3974eb0
thread:issue:147739,issue #147739: torchvision export model error:: torchvision.models.detection.retinanet_resnet50_fpn_v2,needs triage,47,needs triage,1,0,147739,,,,0,80aa1df2fd72240d9095b47253edd710a8a2d05c99fc56707105d9e397059689
thread:issue:147818,issue #147818: Bug in `torch.ao.nn.quantized.Sigmoid` Parameter Restoration after `state_dict ` Loading,needs triage,47,needs triage,1,0,147818,,,,0,2152604da02b472229464d645adcf047fca298bd663a5884aa265d5ff585491d
thread:issue:147844,issue #147844: Python warnings are printed multiple times,needs triage,47,needs triage,1,0,147844,,,,0,ecc99dcf07455ce70a04d4a2ee6f35b9f5cfb74989ba304f036d433801a7f5fd
thread:issue:147919,issue #147919: [inductor][cpu]AOT inductor AMP static shape default wrapper occupied almost 3x disk than before,needs triage,47,needs triage,1,0,147919,,,,0,2e0e7059ed2b92dccd0ae0f71ca13c20bfbf7f69c05290f9a7a70d3982eb08ea
thread:issue:148071,issue #148071: [inductor][cpu] maml performance regression in 2025-02-24 nightly release,needs triage,47,needs triage,1,0,148071,,,,0,932731c663c6f7a33ff9a5b7dd83d7b5d0ea0ffdb1f58e405c83ff24c2087c03
thread:issue:148171,issue #148171: Inference llama after Export PTQ,needs triage,47,needs triage,1,0,148171,,,,0,480074ff26e84d88b8687f06280880eeba1782b3632d5c35b41eb731ba155acd
thread:issue:148382,issue #148382: [Inductor-CPU] Debug util request: fine-grained mechanism to disable out-of-template epilogues,needs triage,47,needs triage,1,0,148382,,,,0,428bcee9c8277c0478191b84fca1aaaf769a36fc03df6f2d9243f9608fddc246
thread:issue:148447,"issue #148447: Union type raise error when running python with argument ""-O"" for torch script.",needs triage,47,needs triage,1,0,148447,,,,0,5590f843895e870cc700c3152627539cb6559ffed93ff64bf76b4de3d5025095
thread:issue:148543,"issue #148543: When Ues torch::jit::Module in UE env, there are some structured exception caused by calling module.forward()",needs triage,47,needs triage,1,0,148543,,,,0,e024ecd67f23bf6b75133159c33a306a498f0f408e115f70a6c796c1e769350b
thread:issue:148557,issue #148557: AOTI is OOM-ing when eager doesn't,needs triage,47,needs triage,1,0,148557,,,,0,fc1cf0f604c2cc3cc21ae40904712c362665bdbaef7cabd46251379c59b8af9c
thread:issue:14867,issue #14867: [C++/Pytorch] Get input shapes,needs triage,47,needs triage,1,0,14867,,,,0,f68f6cc144a5cb7f1987544bbb8cde60226a000e7c6c89087778b36d5b2ec72d
thread:issue:148902,issue #148902: Remove Direct Arm Compute Libray (ACL) Integration for Quantized Matmuls: `qlinear`/`qlinear_dynamic`,needs triage,47,needs triage,1,0,148902,,,,0,299a38f446c4c6932c4debbba7475bffbc9c727f8f7453363d4872afad904134
thread:issue:149186,issue #149186: torch.fx.symbolic_trace failed on deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B,needs triage,47,needs triage,1,0,149186,,,,0,2107f2e42aae72d02db5d657cf3c3f7c87351382f6e66f98b79db2c32cec1858
thread:issue:149223,issue #149223: Inconsistent results in using torch.jit.script API from API documentation.,needs triage,47,needs triage,1,0,149223,,,,0,06df21e1c694fec459fea0ae477d0dbfe8b5d8d2dfd756f5eb51ac74a2e9f334
thread:issue:149251,issue #149251: false INTERNAL ASSERT FAILED in `torch.jit.trace`,needs triage,47,needs triage,1,0,149251,,,,0,a4003899ad43f4ba06a78f54b546673f3ee4907e88659edcdff9340c644fbffa
thread:issue:149391,issue #149391: Cannot torch.jit.script nn.MultiheadAttention when bias is set to False,needs triage,47,needs triage,1,0,149391,,,,0,54525e369c683d3a1ea8021df68e1af7ee9e8c1b8944fcdfc96baeda7da084f2
thread:issue:149500,issue #149500: [RFC] : Dynamically Quantized 8-bit Matrix Multiplication support,needs triage,47,needs triage,1,0,149500,,,,0,435ec5a007a95bb2f22b5a1905702c251d65a5b1339ae1a9cae19c26ba6b1f3c
thread:issue:149501,issue #149501: Inductor produce significantly different inference results with the originl original model,needs triage,47,needs triage,1,0,149501,,,,0,549bc6577da2fb0212ad028a914c8635c37767dc99558d3599d2d7c380d199ac
thread:issue:149619,issue #149619: Unexpected return type from `torch.split` under `@torch.jit.script` decorator,needs triage,47,needs triage,1,0,149619,,,,0,85942b4d6296996504c9cee893ff3cf5f633fe542add89198f4956281ef31930
thread:issue:149736,issue #149736: `INTERNAL ASSERT FAILED` in `torch.jit.script`,needs triage,47,needs triage,1,0,149736,,,,0,f8f0d068ccbb1a1587d5b00c5f4fd07dcef26278d5b42e011d0dbb0e9e7e2383
thread:issue:149737,issue #149737: `INTERNAL ASSERT FAILED` in `interpolate` and `torch.import_ir_module`,needs triage,47,needs triage,1,0,149737,,,,0,08e799908bd2c8cd5c9746fdab9f4c3c87ae594a0d6a4e0eae531318a61060d4
thread:issue:149800,issue #149800: `Segmentation fault` in `torch.jit.jit_module_from_flatbuffer`,needs triage,47,needs triage,1,0,149800,,,,0,61851a88659eaf91adfe8f61f1054064c821a0ee17814bbc198493cd7ac8038c
thread:issue:150342,issue #150342: Torch trace doesn't respect @torch.jit.ignore on torch.nn.Module forward method,needs triage,47,needs triage,1,0,150342,,,,0,f55986640dd09d455bda2e90412d088ffb224969b51c2d5a4f9db440945510b5
thread:issue:15043,issue #15043: [JIT] jit.trace fails with custom GRUs and CUDA when the sequence is longer,needs triage,47,needs triage,1,0,15043,,,,0,aa3071a69b4dc025af82202e822879afcb15b3189b15ebdb3b0965bbffdc628c
thread:issue:150526,issue #150526: Consider context when tuning kernels for max-autotune to more accurately reflect the performance of real workloads,needs triage,47,needs triage,1,0,150526,,,,0,2add0fbc373ce290f715065ccda4918e46677d98ca48de63c392f8a317cec9c6
thread:issue:150577,issue #150577: [TorchScript] Enum scripting failures in python 3.11+,needs triage,47,needs triage,1,0,150577,,,,0,57461697575044eaaad9734b3ab0922995203c35977b73a40387655503538682
thread:issue:150607,issue #150607: `torch.jit.script` does not respect `torch.set_default_dtype`,needs triage,47,needs triage,1,0,150607,,,,0,df67f47835e77cfeaee477faebc7a81a60c42f821074fae0ff7a06ecc60b4498
thread:issue:151016,issue #151016: [Dynamo][Export] Untraceable issues when exporting the Stable Diffusion 3.5 model,needs triage,47,needs triage,1,0,151016,,,,0,af0aad360abdee321e6b64b2133ec3ae975bc631a766c8deb2aaff9c5cafc4bd
thread:issue:15116,issue #15116: torch.save does not work if nn.Module has partial JIT.,needs triage,47,needs triage,1,0,15116,,,,0,1a8b1cf4d9acc04910bff3451d621d672a4784aed6ee095c6c7bcc3ede6816b8
thread:issue:151165,issue #151165: Bug on running TorchScript on H100,needs triage,47,needs triage,1,0,151165,,,,0,02814509420dfa0f43101dd67afe8f2f7a13bdeae1363a1c45539f67162305c1
thread:issue:151233,issue #151233: Code Injection via `torch.jit.annotations.parse_type_line()`,needs triage,47,needs triage,1,0,151233,,,,0,e54dc16dd274fd73a6c629a4d08cfc6ddbb34bf5acdfa133cbcc2832404f370c
thread:issue:151428,issue #151428: TorchScript Model Saved on x86 Returns NaNs When Loaded on s390x,needs triage,47,needs triage,1,0,151428,,,,0,e0d10342fffb6d217963928e1f9a1d4a0cb105ae43eb1908149f0e09c1f0e6b4
thread:issue:151468,"issue #151468: Bug At `d\\dependencies\\libtorch\\include\\ATen\\core\\jit_type_base.h"":289`",needs triage,47,needs triage,1,0,151468,,,,0,ae25543363935d8507759b50330d9882ed2a246891e24ba78b97472f1bb7ffca
thread:issue:151536,issue #151536: [inductor][cpu] pytorch_CycleGAN_and_pix2pix AMP/AMP_FP16 multiple thread performance regression in 2025-04-07 nightly release,needs triage,47,needs triage,1,0,151536,,,,0,d80bc61853440c25445aca0795229d34d4d0982782958abf99cbdb34de978e68
thread:issue:151746,issue #151746: [AotInductor][Export][Triton] how to export custom triton kernels when use torch.export.export,needs triage,47,needs triage,1,0,151746,,,,0,88b571d0efa1d7901db27df1405e1dc19aba61c3b1431020c5bc111f59cc7ba5
thread:issue:151969,issue #151969: Unrecognized in PyTorch traces w/ Cuda 12.6,needs triage,47,needs triage,1,0,151969,,,,0,fa56adb5bab22b4d6973301749a8db9b645a34826cb73a5023ab4f0dc68a3731
thread:issue:152283,issue #152283: Forward compatibility in torch.export,needs triage,47,needs triage,1,0,152283,,,,0,521f8f44c06298863aff3891aa187960b20f621c2374229e807d4487de47804d
thread:issue:152465,issue #152465: torch.export with dynamic shapes on Static Cache HF LLama model fails,needs triage,47,needs triage,1,0,152465,,,,0,0e65e85c10e54760dfb17fabd215ad2ace235a40ca7b31df11db7fe925bd1c0a
thread:issue:152612,issue #152612: Enable AOTI for Metal inductor,needs triage,47,needs triage,1,0,152612,,,,0,ef678d5d8c199874ba1a7f246a18ea2a4c95a080f1e5f5f6c005ff0ab70e3ddb
thread:issue:152617,issue #152617: Pytorch Profiler crashes while using it with Pytorch Lightning module,needs triage,47,needs triage,1,0,152617,,,,0,a433efff787b7968319565913c8e685310efebf45da65a8cfd4d0e9eb9768245
thread:issue:152813,issue #152813: Mismatch in dynamic quantization performance for torchao and torch.quantization,needs triage,47,needs triage,1,0,152813,,,,0,6d862ae14fa2daf081a71f14a46febdc2dad30912092cd8634d300328ef36716
thread:issue:152921,issue #152921: [inductor][cpu] pytorch_CycleGAN_and_pix2pix AMP multiple thread performance regression in 2025-04-27 nightly release,needs triage,47,needs triage,1,0,152921,,,,0,5b859edcf0461c7cc0387376673c6873c37dbebe7d323e727fea3015032c2254
thread:issue:153179,issue #153179: export: No error raised on data-dependent assertion.,needs triage,47,needs triage,1,0,153179,,,,0,046d4509281ad5a9732cab6c07f908b797c81ec18cf4412e713f68248fc146e5
thread:issue:153224,issue #153224: [JIT] Inconsistent results between eager and script modes for instance_norm with custom running stats,needs triage,47,needs triage,1,0,153224,,,,0,433cf4122b8f81d15789068ea55f21da6d538e3854716c8f78fee9a9ce4c51a3
thread:issue:153315,issue #153315: [JIT] Compilation-induced discrepancy in F.instance_norm when passing input as running stats,needs triage,47,needs triage,1,0,153315,,,,0,65b157d2d5beafe350924c876d1f5d8510574850a2ff24283c9439c1829a3bba
thread:issue:153358,issue #153358: torch.dequantize result inconsistent on CPU and GPU,needs triage,47,needs triage,1,0,153358,,,,0,1adeb0b9d7fb73df5174b6834b24687544f52414a7a2fb237b12fbd798441741
thread:issue:153429,issue #153429: [Export] Non-strict mode can't handle conditionals on tensor subclass types,needs triage,47,needs triage,1,0,153429,,,,0,aba8e3cfccb90bdbb30f97f98448308326e2484af9f8f42463277aa46dfdc507
thread:issue:153722,issue #153722: JIT Script doesn't handle LLONG_MIN correctly,needs triage,47,needs triage,1,0,153722,,,,0,b05e49c22fb02bbc70ec2df33ab9a9d8d61ff59d41210069d322ed7300490fd7
thread:issue:153901,issue #153901: torch.profiler support for tf32,needs triage,47,needs triage,1,0,153901,,,,0,881f0e56b297f99c2278b9cbbb1bfb944f904d81c17081c695c6b0feb01b1186
thread:issue:153931,issue #153931: TorchScript Fails to Handle type() for Any Input Type (Tensor/Non-Tensor),needs triage,47,needs triage,1,0,153931,,,,0,35399374842846e87efafce592988d93ee0653552312d58fbb1fb67e1090d388
thread:issue:154094,"issue #154094: [inductor][cpu] aoti shared memory run on multiple cores, performance will drop",needs triage,47,needs triage,1,0,154094,,,,0,ca440afd49ca347c1fe8d2b067db78273599aaab02f8d7ddf38c378ddbaf51b1
thread:issue:154101,"issue #154101: INTERNAL ASSERT FAILED at ""/pytorch/torch/csrc/jit/ir/ir.cpp"":1307",needs triage,47,needs triage,1,0,154101,,,,0,494c25812fe9db92bf147fffab758d645df914d23d2899181cd77f15c0447aac
thread:issue:154323,"issue #154323: when pass a scale generated from a registered_buff of a nn.Module to fake_quantize_per_channel_affine, onnx runntime will raise Exception： “For per axis quantization, scale must be 1D tensor with”",needs triage,47,needs triage,1,0,154323,,,,0,edbf8fd5fd6df721de28eca44ad54820d72df6ea6675623836323c9a54460afa
thread:issue:154324,issue #154324: torch.jit.script gives false results for autograd if complex data types are involved,needs triage,47,needs triage,1,0,154324,,,,0,7d5d2d7401123ad6466514e515aff0aa31953daafc8ab7a033f33d1c89101c07
thread:issue:154407,issue #154407: [JIT] torch.jit.script raises an exception with view(dtype),needs triage,47,needs triage,1,0,154407,,,,0,fb4c9e97ccfe37e07a68d7e33e0a523a28fa2d59d6deaec4b7d6204ad7684836
thread:issue:154423,issue #154423: `Segmentation fault` in `torch.jit.ignore`,needs triage,47,needs triage,1,0,154423,,,,0,f4fb46025c6c0528c423565d52df6e7f6e3d8b6de8fb1cfebd872a9cec670aec
thread:issue:154478,issue #154478: `__jit_ignored_attributes__` is not respected in `torch.jit.trace_module`,needs triage,47,needs triage,1,0,154478,,,,0,2e11bb278369d050c477d6589c1b72615e8b6c171427faec51b816919da6aab0
thread:issue:154502,issue #154502: `typing.get_type_hints` fails on TorchScript model after loading with `torch.jit.load`,needs triage,47,needs triage,1,0,154502,,,,0,0d6ec0b040def2a44a88a978513c991f7d5b8d1d131bb438d8f036c6f726c41e
thread:issue:154607,issue #154607: Internal Assertion Failure with torch.cuda.Stream in @script_method on CPU-only Systems,needs triage,47,needs triage,1,0,154607,,,,0,623f102f98a19a680a9c6501153d6ae9289f144dabbc21aea0aa4a3cfff9fe81
thread:issue:154853,"issue #154853: Add node.meta[""stack_trace""] to make_fx",needs triage,47,needs triage,1,0,154853,,,,0,ec18af85aeae4d03d6772a67caeb1cb4ca019cce9df12cc0a6326557f2203a2e
thread:issue:155003,issue #155003: Internal Assertion Failure with Invalid Arguments in max_unpool2d under TorchScript,needs triage,47,needs triage,1,0,155003,,,,0,b79b87c78b9665f56adaec31e9464d2db182bdb0ec3186d44a239abe86769504
thread:issue:155004,issue #155004: Internal Assertion Failure with Invalid Arguments in max_pool2d under TorchScript,needs triage,47,needs triage,1,0,155004,,,,0,6dab9a520b8e5d7d93148fc5eda5a897a1e75a2de179aa41d6fa8b2afc090da4
thread:issue:155007,issue #155007: Internal Assertion Failure with Invalid Arguments in max_pool1d under TorchScript,needs triage,47,needs triage,1,0,155007,,,,0,798f1de85e531acd62b0ca8caea19b56b8a383eeb60ab42406dcd6a0877bc9d2
thread:issue:155009,issue #155009: Internal Assertion Failure with Invalid Arguments in max_unpool1d under TorchScript,needs triage,47,needs triage,1,0,155009,,,,0,faeb18f957a9a20079efd6b51f7c366215bf9426b456aadfb8cc25543155b096
thread:issue:155321,issue #155321: mishandling torch.package.PackageExporter raises Aborted when given a tensor instead of an importer,needs triage,47,needs triage,1,0,155321,,,,0,d2b1d02e8e21500f5fc32bcf0e6a0f91aa9c93aa8173c1281e625c21092924a8
thread:issue:155550,"issue #155550: [torch.export] Cannot export TorchVision raft_small, raft_large",needs triage,47,needs triage,1,0,155550,,,,0,9994718494b7fc1680cef518ea1b54744bd9d36aae9ce11ca7fbc63834a900df
thread:issue:155862,issue #155862: Export Huggingface models with StaticCache,needs triage,47,needs triage,1,0,155862,,,,0,53e182486bd56e731612c6716b02bf5bf58ed167d07626e30ebb0d5eb7a6b0a3
thread:issue:156544,issue #156544: `TorchScript` does not allow accessing methods of nested tensors,needs triage,47,needs triage,1,0,156544,,,,0,01070289c8e23a836aebf4531efb2e0042809bd5623511bc47108c2c08df4c77
thread:issue:156557,issue #156557: [JIT] TorchScript fails to compile model using random.choice() with confusing error message,needs triage,47,needs triage,1,0,156557,,,,0,e71f55802a4eefa00186240a8249d69c417e2e7b5b7974d022fb5182d78f68d6
thread:issue:156558,issue #156558: Using collections. namedtuple in the forward method of the model resulted in compilation failure,needs triage,47,needs triage,1,0,156558,,,,0,d604f4c5d3c5e2dfbe272e159ff7f8221bdb81eba9379d57d5597c4e23cb8cd2
thread:issue:156814,issue #156814: torchscript: RuntimeError Unsupported dtype complex,needs triage,47,needs triage,1,0,156814,,,,0,6fed7755337a17644352523a730172ec9d23ff4bec779dcec6dbe9d8ec56cc9d
thread:issue:156882,issue #156882: Export always give a value range with max length - 1,needs triage,47,needs triage,1,0,156882,,,,0,fc1217bf1f2d3a1c09859b3df425c4f5218d74a25329f3831f225288beac7029
thread:issue:156915,issue #156915: Unable to compile with Vulkan support,needs triage,47,needs triage,1,0,156915,,,,0,b82160e8ef13642bcff49185dd30360a584aa41e682f841f6fbecba1b05b5234
thread:issue:156955,issue #156955: `torch.jit.load()` crashes with C++ `c10::Error` if file system is full (no graceful handling),needs triage,47,needs triage,1,0,156955,,,,0,f26848f8264a191a5906016e34b62f1eb890b3caf14bb0eebadfd7276f761247
thread:issue:157103,issue #157103: Issue with super() resolution in TorchScript when inheriting from custom class and nn.Module,needs triage,47,needs triage,1,0,157103,,,,0,88563051049f2cf496f242128b9aad4270e1fbcc78abd6e8df9d915f68b1ce55
thread:issue:157603,issue #157603: Incorrect inference of the groups parameter type for channel_stuffle (int misclassified as Tensor),needs triage,47,needs triage,1,0,157603,,,,0,3cb9475f69ef2b42234e7890cce4b4c08c980f680a079efbc9f747b5b189b4bd
thread:issue:157625,issue #157625: Is there some official method to extract the featuremap of each node in pt2 graph like the function torchvision.models.feature_extraction.create_feature_extractor(),needs triage,47,needs triage,1,0,157625,,,,0,4cd23bb5565a7414da71b4f0e614c8a85f2d9e98549e3ce5bad5e8985313af69
thread:issue:157728,"issue #157728: fill_ overflows on uint64 in range [2**63, 2**64) when profiler is engaged",needs triage,47,needs triage,1,0,157728,,,,0,8f7f109a19f1608dad9f1f4764fd3a62fd5955f044092cdcab3b26f6a53219d5
thread:issue:158227,issue #158227: Degraded JIT execution performance on 5090 Windows,needs triage,47,needs triage,1,0,158227,,,,0,340eb0d2342c2cc05b6e26bcbe2d45ce01fbc5394f142523bfebf662da168084
thread:issue:158331,issue #158331: DISABLED test_conv2d_unary_cpu_cpp_wrapper (__main__.TestCppWrapper),needs triage,47,needs triage,1,0,158331,,,,0,f01de8bd152186726926a6858c16e04b3f65eac32a76223bc0804009c6f0c57f
thread:issue:159256,issue #159256: can kineto profile only one thread with corresponding gpu tasks among multithreads,needs triage,47,needs triage,1,0,159256,,,,0,85272f5bce4cc3268de895d7c23d760e7fa43f40cc18c9d7e3b30a2e9a3fc36b
thread:issue:159265,issue #159265: Unpin dependency version when possible,needs triage,47,needs triage,1,0,159265,,,,0,cea93c8773b59c79b0b581585a7f6aff95d8958da670e648fa8a627fa320c732
thread:issue:159269,issue #159269: Feature Request: Allow torch::inductor::ModelPackageLoader to load from an in-memory buffer,needs triage,47,needs triage,1,0,159269,,,,0,ea7087cf236bac4f45c4b2aff255e5110aabedca162f8162348c052b4bb50fc1
thread:issue:159316,issue #159316: Support for non-scalar valued Loss Tensor in `_export_forward_backward()`,needs triage,47,needs triage,1,0,159316,,,,0,988e288aa1bd619982bf4dbaee8717e9b8402c885f26c7c8d3288172231659a4
thread:issue:159505,issue #159505: Cannot pass through None for example_inputs in prepare_fx,needs triage,47,needs triage,1,0,159505,,,,0,fd2050ad072ab38a37554586cb9625632059ff57fe512bd260a80e22b44c5c3f
thread:issue:160005,issue #160005: [BE][download.pytorch.org] Index fix user navigation,needs triage,47,needs triage,1,0,160005,,,,0,fc094fd8669b543d386f87e045e605e5dff0939b9eb91bbe5beb773552a79779
thread:issue:160391,issue #160391: `torch.compile` backward pass fails with `AssertionError` in Inductor C++ codegen when model returns tuple outputs,needs triage,47,needs triage,1,0,160391,,,,0,c1cbdce88330aba73675cbd6842a67092373d491df84ee7ce0861b2472aaad69
thread:issue:160392,issue #160392: bug in libtorch optimize_for_inference,needs triage,47,needs triage,1,0,160392,,,,0,a31e41eec2e3e80a5b2546b16698ef120d649551ab0d3333b1bde30b8435f936
thread:issue:160631,issue #160631: torch.quantize_per_tensor not implemented for SparseCPU tensors,needs triage,47,needs triage,1,0,160631,,,,0,c2747ef6614c1d56f3fb4dfcedac6b7086a48b1d958e8faa79dd05e5fd644b24
thread:issue:161179,issue #161179: Consolidate magma-cuda build for Linux and Windows,needs triage,47,needs triage,1,0,161179,,,,0,61aa407f24359e3f4991bafa8665b68045829bbce6a63a705c027776707c3f79
thread:issue:161815,issue #161815: Windows AMI release issues and improvements,needs triage,47,needs triage,1,0,161815,,,,0,80d5eac3af2b4ddb15384adc700608ca1e136c3e6ab59e49cb94c676c4667e30
thread:issue:161854,issue #161854: torch.ScriptObjectProperty.name.deleter occurs Segmentation fault (core dumped),needs triage,47,needs triage,1,0,161854,,,,0,9c7d4bd52c4ab60d5cb8905f314692a4453a38f8cac61c77661d1bbe12e6476a
thread:issue:161935,issue #161935: torch.jit.trace_module.register_parameter :cannot assign 'torch.cuda.FloatTensor' object to parameter 'new_param' (torch.nn.Parameter or None required),needs triage,47,needs triage,1,0,161935,,,,0,a151d12b3b4b105b10db6a336e8995fcb0abbf5bf6412f605f1b6aa13c3353aa
thread:issue:162135,issue #162135: [inductor][cpu]dla102 AMP default wrapper hang issue,needs triage,47,needs triage,1,0,162135,,,,0,ecab4643a76c86398450fc0499edff9e344af893c492c26554f8f3b9d22e6dcc
thread:issue:162481,issue #162481: Incosistent tracking of device activities when calling profiler.step() in torch profiler,needs triage,47,needs triage,1,0,162481,,,,0,abf8ec752fae29e4cc3118470ab18016de7ae76e29488099b3950415b7ab5d9d
thread:issue:16263,issue #16263: [JIT] Support C++ front end module and JIT interop,needs triage,47,needs triage,1,0,16263,,,,0,2cd8544950c93a23c51a8c9db9d8f9b288291cc0a055319eb170e99a5863fa17
thread:issue:162948,issue #162948: Metadata object loses arbitrary attributes added dynamically,needs triage,47,needs triage,1,0,162948,,,,0,2bc81de6df207178b766b32fc41e0b13c6f65aee92715272f99f058e79423f60
thread:issue:163791,issue #163791: torch.export fails on Phi-3.5 model,needs triage,47,needs triage,1,0,163791,,,,0,cdfd566247593c17cf7c3380f20284721ff5f7268dd46d395e42817ff5fc97de
thread:issue:163945,issue #163945: Failed to run a backward op in TorchScript interpreter,needs triage,47,needs triage,1,0,163945,,,,0,5ff46a29cf18d98a49a456b81fe1858f9dffd0f903da2f559107bc90dbb69025
thread:issue:164083,issue #164083: Broaden support of unflatten to split graphs,needs triage,47,needs triage,1,0,164083,,,,0,4dc4909c520b5e96d1086c526d975cabbc1ae5b12f22e028ac9b9261f14105d0
thread:issue:164145,issue #164145: Improvements to profiler for bitwise equivalence use case,needs triage,47,needs triage,1,0,164145,,,,0,31db713ef014f47848c488fecd74c1c98912f168ebb88a42137f264cd95fe169
thread:issue:164208,issue #164208: CUDA Illegal Memory Access in Multi-threaded LibTorch Applications,needs triage,47,needs triage,1,0,164208,,,,0,d7341fcfd5b351269cecf924951c183e9a7da0abc9fe7a13092d6c1ac5fa2f34
thread:issue:164534,issue #164534: DISABLED RecordDebugHandles.ScopedCallbacks (__main__.test_jit),needs triage,47,needs triage,1,0,164534,,,,0,8628e926534125bd8feef2cf088d8e414a2a6268db68f34c47050702600a9ee4
thread:issue:165395,issue #165395: [export] run_decompositions fails with index error,needs triage,47,needs triage,1,0,165395,,,,0,f12895a934dd4d07bb9e56c79be8ae725d53107e4ea59cf079af3e82ea799ff6
thread:issue:165400,issue #165400: aten::bitwise_xor,needs triage,47,needs triage,1,0,165400,,,,0,b519ee979b39d7e97d3a314b2e1ffbe239546f1df3677952936133c9dbffbae6
thread:issue:165829,issue #165829: Floating Point Exception (Core Dumped) with Dynamic Quantization on nn.Linear with Output Dimension 0,needs triage,47,needs triage,1,0,165829,,,,0,14f1fb64feab20572ad71e8212678c76d6f550fbc54b91bc4bdcf966827481fc
thread:issue:165900,issue #165900: Converting weights `.pt` content between `dict` and `RecursiveScriptModule`,needs triage,47,needs triage,1,0,165900,,,,0,49e7a3cc44fb246eaec96eb8d7d94c774e2924512f097f2ed5c00f05714214c5
thread:issue:165929,issue #165929: Numerical mismatch b/w eager & aot_eager_decomp_partition backend in SegFormer layer (2.9.0),needs triage,47,needs triage,1,0,165929,,,,0,d9b08363385f1b53c0d5521c3ee0eca4d81b54ab8a0d930c3575153eac82c896
thread:issue:16602,issue #16602: Error while installing pytorch,needs triage,47,needs triage,1,0,16602,,,,0,12c577a56118681c0cf69a17705cfc5123edd03d785ca3d80eb7cac916ba260e
thread:issue:166311,issue #166311: Unable to export profiler trace,needs triage,47,needs triage,1,0,166311,,,,0,d96701f8b3a222654a39640a9992f93b201fc43858d18c0e0723a63599f4daec
thread:issue:166939,issue #166939: Reduce aten::empty Calls for Models with Constant Input Shapes in AOTInductor,needs triage,47,needs triage,1,0,166939,,,,0,2cd6e39dbe81f1cf435952993e19754676d76d3462966976fc9ac966acb4222e
thread:issue:167199,issue #167199: [CPU] torch.compile consumes too much extra memory,needs triage,47,needs triage,1,0,167199,,,,0,783174fb22303303cfedd64ca5b572fef8f75131195e724055287efe9032aaa5
thread:issue:167412,issue #167412: How can I train in C++ using a Pytorch torchscript model,needs triage,47,needs triage,1,0,167412,,,,0,792f20bdc20e8728ebbec7180f70811ac5cc148db29f22a517e611d4505e3a83
thread:issue:167730,issue #167730: Feedback about ExportDB,needs triage,47,needs triage,1,0,167730,,,,0,0769b81ace8bc00c3f9bd182a0f14658cfed78cfa83480c5b0ede8686596bdbf
thread:issue:168077,issue #168077: scripted models wrongly cast integers to floats,needs triage,47,needs triage,1,0,168077,,,,0,157794a533b5120137811c1c7c4783e311be6509921ca5c2316bb2fc8a1ab67d
thread:issue:169009,issue #169009: Dynamo export: Qwen3 Omni model running into data-dependent expression errors,needs triage,47,needs triage,1,0,169009,,,,0,7ff601b8aa1a6260de478fd305984c613e88ce17062da5c1bd08b757b1838205
thread:issue:169069,issue #169069: Triton Pin update 3.6 related - All periodic linux-jammy-cuda12.4-py3.10-gcc11 are in broken state. Some linux-jammy-cuda12.8-py3.10-gcc11-debug are broken as well,needs triage,47,needs triage,1,0,169069,,,,0,d54769b6df5f3fdc8bbdf71d560f49fa32d26ab3587d6fa7afbc4a74c0f7eac1
thread:issue:169357,"issue #169357: prepareProfiler() skips collecting info of ""privateuseone""",needs triage,47,needs triage,1,0,169357,,,,0,c6abc3cdb7738a8f7dd90f4cac873e9e688eaf299d2125363a0053e7ed5cfa18
thread:issue:169666,issue #169666: Inconsistent`flop_counter.get_total_flops()` results for a custom decoder model,needs triage,47,needs triage,1,0,169666,,,,0,3087bec60ec103003aa51cd8f9bac3a559b209e833daed0266031394a3e9dafb
thread:issue:170169,issue #170169: questions about default_weight_fake_quant config with pytorch quantizations,needs triage,47,needs triage,1,0,170169,,,,0,636a19fa66e654592273c4f5502e170c5c9189b089095cce853f7624d6c16c2e
thread:issue:170319,issue #170319: torch.profiler shows GPU ops (Triton / FlashAttention) to have `cpu_op` category for some reason,needs triage,47,needs triage,1,0,170319,,,,0,94a07b2b81df7287cdb1ad19b5b43f1c9f76e1e52201f481f12d05be9232a46a
thread:issue:170548,issue #170548: unable to export a model with bucketize,needs triage,47,needs triage,1,0,170548,,,,0,899b8154b46315c3eb7a88ae03bcdc14f1987cc357c1fe0be67e4df1bc5c4c17
thread:issue:170673,issue #170673: `torch.export` Fails with RuntimeError when Accessing Parameter via `.data` in `forward`,needs triage,47,needs triage,1,0,170673,,,,0,011450032f2613d243d7da986bde27931ab988b7da9bb8bd785d2648ab88022d
thread:issue:170680,issue #170680: Error when composing jit.script and jacrev,needs triage,47,needs triage,1,0,170680,,,,0,a412d2f57aa71a900cb1db2ed628a44eef37538b5dffb6f534c117b2df0e6a95
thread:issue:170774,issue #170774: CUPTI metrics not recorded in profiler,needs triage,47,needs triage,1,0,170774,,,,0,4b8d5502b3a3697d09808e9923e48b76a50593439df4f1a0b6d0217bc20b25e1
thread:issue:170783,issue #170783: New higher order ops to handle loops,needs triage,47,needs triage,1,0,170783,,,,0,b46555bbcf814b1113b3ccbaba8ecfac9f7b23447a01bd7a8a4f01e57481c0dc
thread:issue:170867,issue #170867: Operator benchmark: option to measure GPU execution time only (less CPU noise),needs triage,47,needs triage,1,0,170867,,,,0,694e4571e7e58bad377ed662e7c6a30c86ba5ab2b2fe7f58785a2f515dab8ed6
thread:issue:17126,issue #17126: Support callables in scripted functions,needs triage,47,needs triage,1,0,17126,,,,0,fa238b0fae8a3b25aa151b39da7daee1d3bd0cf1c16ba866bddbe629a3f1ef0c
thread:issue:171377,issue #171377: PyTorch JIT `get_debug_state()` triggers INTERNAL ASSERT FAILED after `torch.jit.trace`,needs triage,47,needs triage,1,0,171377,,,,0,bb1f7c2102a0ccb89451b820eee787365f2683ff078a397affad148102d507fa
thread:issue:171516,issue #171516: How to verify that default_decompositions successfully reduce operators to the Core ATen IR set?,needs triage,47,needs triage,1,0,171516,,,,0,f96bad70fd22258d8fcc5a7859ba093d04b202a1ec1fbbe1e3dfffa0b26f36d6
thread:issue:171675,issue #171675: PyTorch quantization `QConfig` triggers INTERNAL ASSERT FAILED when printing config with invalid observer `dtype`,needs triage,47,needs triage,1,0,171675,,,,0,6fae565fe0254bc668065583ab819b4c20693b9467891cebc2d2527f07e0f150
thread:issue:17932,issue #17932: JIT torch.ones_like with dtype starts failing on master,needs triage,47,needs triage,1,0,17932,,,,0,3bc88118d48c8c4bb002b6fff2ecf809a0f064ac0c9b4ee6cf46d56f65b2777a
thread:issue:18312,issue #18312: [jit] Trace perform weirdly on BatchNorm,needs triage,47,needs triage,1,0,18312,,,,0,909da13206e71d14025a9243c70b92ed57fc0961dd8b865b2afac98d93e18248
thread:issue:18376,issue #18376: Cleanup code in register_c10_ops.cpp,needs triage,47,needs triage,1,0,18376,,,,0,b05dfc4cf791dbd160462035cacdef9ee36dbf0dba3218c03eae42af5ec2db9b
thread:issue:18380,"issue #18380: C++ inference (CPU-only) stalls in Android, and crashes on Mac/Linux",needs triage,47,needs triage,1,0,18380,,,,0,e51785e63576ee6a4033a2c6fea987482a21dc2ddf5907668ec72aeaff63c397
thread:issue:18476,issue #18476: [CPP] Allow binding config structs into the Python front end,needs triage,47,needs triage,1,0,18476,,,,0,1149db2680b5d3c4e8b61e41a5238d262a5bd835f9cfa41ee8476d9d02029970
thread:issue:18520,issue #18520: Jit fail with TracingCheckError with tracing model with layers created after init.,needs triage,47,needs triage,1,0,18520,,,,0,66df9ec6172e269960a8b3fef89b125f7fa324f1b3f3193c4bd06731600d3d07
thread:issue:18975,issue #18975: Tracing and Scripting nn.Conv2d shows different op in the graph,needs triage,47,needs triage,1,0,18975,,,,0,08d1c0f3446a404bd90c8469ec3b391b6430c6e28eb46b4942b1373bdc4c8281
thread:issue:21064,issue #21064: [JIT] Expose subgraph execution for intermediate output extraction,needs triage,47,needs triage,1,0,21064,,,,0,62b4960ae4bb08fffd5961c4d8b682cd4580657054401fc5a61e2ba2775ed0b7
thread:issue:21255,issue #21255: IsType<T>() ASSERT FAILED [Detectron e2e_mask_rcnn_R-50-C4_1x.yaml],needs triage,47,needs triage,1,0,21255,,,,0,5dc1a1577a01637f6abff0cddbe19feee83709a40d711241b3989bf1356eb7cb
thread:issue:28191,issue #28191: [JIT] Allow user to provide aliasing information on input tensors and model parameters,needs triage,47,needs triage,1,0,28191,,,,0,2f8c182a73262ec955099f5ea2d9f6999557381a4267fc762ce09d19a8ef846c
thread:issue:28267,issue #28267: Problem with jit TorchScript while copying data between GRUs,needs triage,47,needs triage,1,0,28267,,,,0,d8f0cdf14827c808ea187720712751fa8fbce2f4ebdfafadca4f69c826a0cbe8
thread:issue:29125,issue #29125: _compared_saved_loaded doesn't work with torch.tensor constants,needs triage,47,needs triage,1,0,29125,,,,0,5ba6f0b5b5063d23a7905ef16722527773232458fd1900748f767d1ce070d46a
thread:issue:29128,issue #29128: JIT should respect SKIP_PYTHON_BINDINGS and SKIP_PYTHON_BINDINGS_SIGNATURES,needs triage,47,needs triage,1,0,29128,,,,0,274351b31bc4c07b63cf6e2c31d08710e8054b0b568233b9c6fde2cae50bcb1b
thread:issue:30365,issue #30365: TorchScript Performance: 150x gap between TorchScript and Native Python,needs triage,47,needs triage,1,0,30365,,,,0,ecfab8ad9d0ce37b0a868deb62a3c473a7ba377829e01e72cbffca147ef0569b
thread:issue:30863,"issue #30863: [TensorBoard] Installed tensorboard 1.14.0, TestTensorBoardWriter.test_writer failed.",needs triage,47,needs triage,1,0,30863,,,,0,c504ae8fda8ceb5cc72a2f1d8fd2c1207ab42dd71e0db950caf40c4c944425ef
thread:issue:31603,issue #31603: Typo in `torch.utils.tensorboard.add_image`,needs triage,47,needs triage,1,0,31603,,,,0,625b6b41005caa1be9242a91f672f92401434a011e5129c6a35f12b58d0894c1
thread:issue:32901,issue #32901: QNNPACK linear doesn't preserve dimensions,needs triage,47,needs triage,1,0,32901,,,,0,1fe9a3ad33641c6c86def63d83c22905179ea1e518f168fddd27848791837b0c
thread:issue:33110,issue #33110: InlineAutodiffSubgraphs in JIT inlines non-differentiable custom groups unexpectedly.,needs triage,47,needs triage,1,0,33110,,,,0,c484921c871e58a92a64e317a36b335701a861e6e1cd5a9aa4dfd59c6a9fd6f1
thread:issue:33124,issue #33124: QNNPACK: GNU aarch64 assembler does not support 4s on neon mov,needs triage,47,needs triage,1,0,33124,,,,0,f27704a7e8bb26764a13902cbb795e48c5756152a093d6058846619fe0b418fa
thread:issue:33281,"issue #33281: Uninitialised value was created by a stack allocation, reported by valgrind",needs triage,47,needs triage,1,0,33281,,,,0,8710536704d9e6607eb98d559d4e31c48ebd5fbe436112d5382d616384723938
thread:issue:33285,issue #33285: iOS libtorch superpoint model bug,needs triage,47,needs triage,1,0,33285,,,,0,a45435531c5420493422ce860ba9218886fbce2e8fbc36f691b2c091070a5ac3
thread:issue:33491,issue #33491: jit.trace checker fails for LSTM,needs triage,47,needs triage,1,0,33491,,,,0,1031de5faf540e9092fb99ee2c743a3a0a6ddb2c765046e00a651e68f95f24cb
thread:issue:33631,issue #33631: bitmapToFloat32Tensor() 1 channel Tensor [feature] [mobile],needs triage,47,needs triage,1,0,33631,,,,0,52d165f8de4ede402e8ab8ee1d7cf5e9b23c56407c9812cda0e8b159eb64d220
thread:issue:33810,issue #33810: Wrongly detected: Division of ints in TorchScript uses Python 3 true division semantics,needs triage,47,needs triage,1,0,33810,,,,0,3b788082c07a923016b66efd581ae6922b3b7bd2a429b76a20c725eadb87dd35
thread:issue:33993,issue #33993: ModuleList doesn't support slicing in TorchScript,needs triage,47,needs triage,1,0,33993,,,,0,6ae529b6598643d67e95b2a2c5e459371bb54f235be5182abc3a0948bed4b24e
thread:issue:34682,issue #34682: Pytorch not compatible with react native android,needs triage,47,needs triage,1,0,34682,,,,0,7bd120c4f806657a39f977269c79c5a70d1cb23139079898ff837a0e194b9ff3
thread:issue:34713,issue #34713: torch.jit.script report error when using index to subscript nn.ModuleList,needs triage,47,needs triage,1,0,34713,,,,0,f54a0b11a55c637806b0e8abc0eb6d8dfbf5f062f12f479340417a385f9f11c7
thread:issue:34851,issue #34851: [jit] `__hash__` magic method is missing,needs triage,47,needs triage,1,0,34851,,,,0,770431100032bb824e59b06c82534307a30850379cbfec55a65bc494853374a3
thread:issue:35095,"issue #35095: tensorboard add_graph 's ""operator_export_type""",needs triage,47,needs triage,1,0,35095,,,,0,5673b4b4f1c145d54fc7ff0f347a2ee39618e10aae38e376636d5e6072252eaa
thread:issue:35215,issue #35215: Get hold of Backward graph from a C++ module.,needs triage,47,needs triage,1,0,35215,,,,0,40abcd6a56604679f029dd1c484018e9842228bb62d4bdfb458811295169bec4
thread:issue:35372,issue #35372: How to support single-process-multiple-devices in DistributedDataParallel other than CUDA device,needs triage,47,needs triage,1,0,35372,,,,0,f21d92b0d54b0fd866192711f704cc1a9e1466214f78128a0e078974b4ba0537
thread:issue:35526,issue #35526: RuntimeError during converting Reformer model to TorchScript,needs triage,47,needs triage,1,0,35526,,,,0,b02db7b522b1bced7e1085a728ea5c5799c32e8d64cb09e038b0a42c5eff2bc6
thread:issue:35606,"issue #35606: Could not find any similar ops to ""foo..."" in the `Libtorch`",needs triage,47,needs triage,1,0,35606,,,,0,7b78b71bf629850aef52dc01580a58df0fe6c6f1587bcf1a27200047bacf4cac
thread:issue:36061,issue #36061: [JIT] support list of nn.Module in torchscript,needs triage,47,needs triage,1,0,36061,,,,0,354cc9ef6e38ff540e3e6e576a6b7f459113b5d1c3eae20b4585089d3c6e9443
thread:issue:36182,issue #36182: ## 🐛 Bug: QNNPACK tests failing on master on Nexus 6,needs triage,47,needs triage,1,0,36182,,,,0,a39ec226286aa7e9f9470c5e1b80212e0891b3761bce57447fe0d2ffce18918b
thread:issue:36211,issue #36211: [JIT] support self.named_buffers and self.named_parameters in TorchScript,needs triage,47,needs triage,1,0,36211,,,,0,73673b2c4e0ed33a68b7397dded207ac8908ac22bfa767f2d91ddd74aab773d0
thread:issue:36603,issue #36603: TorchScript Support for Named Tensors,needs triage,47,needs triage,1,0,36603,,,,0,22feecb6d03a7a3ee591595d2e54c518038c0c9586b196e8ca5614bb3e462f40
thread:issue:36718,issue #36718: Support alpha channel in tensorboard.add_figure,needs triage,47,needs triage,1,0,36718,,,,0,ca2afd433c9e17e658fcb8fc3b242697b1fc7de4e3a41e9e297940a604e52c37
thread:issue:36910,issue #36910: jit.script leads to RuntimeError: attribute lookup is not defined on python value of type in some cases,needs triage,47,needs triage,1,0,36910,,,,0,f8a95f6a901f3d8b444ae5ff522cac627f91c453de844e955041b2e94fa99050
thread:issue:36926,issue #36926: JIT string ops and other miscellaneous ops probably shouldn't be in aten namespace,needs triage,47,needs triage,1,0,36926,,,,0,78c0426e07a114cda6836a80bc1c8a04a82be3dad5cadd413254fb2601052ce1
thread:issue:37016,issue #37016: Guidance on implementing a new backend,needs triage,47,needs triage,1,0,37016,,,,0,370a8f1a8585a17a0574cd71c12d3ca9cbbde9b7187849d68fca87e2b8114caf
thread:issue:37387,issue #37387: The pytorch's graph is lack of common names for nodes,needs triage,47,needs triage,1,0,37387,,,,0,c86f3f076945c9ec9e8ed12ff13a42ec951d660b7eed11faa1831dec2166f561
thread:issue:37442,issue #37442: Tensor.is_distributed not documented,needs triage,47,needs triage,1,0,37442,,,,0,5a7c9f17c14b5c2915526c9bb9459c7bca13bebc9cbac9e2c07fe810af5ea77b
thread:issue:37740,issue #37740: torch Summary writer does not display torchvision.io.read_video output,needs triage,47,needs triage,1,0,37740,,,,0,d413beacff1674202a7eee6768cfd4a28a6068a1b595d16e63bee95d3f056b33
thread:issue:37883,issue #37883: Support @property decorator in TorchScript,needs triage,47,needs triage,1,0,37883,,,,0,62e55240c103ccee39f2ba5860c0628e02797aea7614c4d0e9387a0b7b8f23a4
thread:issue:38273,issue #38273: Adding model metadata in TorchScript model file,needs triage,47,needs triage,1,0,38273,,,,0,4dcd9c7c4aa0c9a5d90ffdb5ca5a60a02b1a8c88cf552ceb9f5a4f73b7fadcbb
thread:issue:38312,issue #38312: Status of support for training on mobile,needs triage,47,needs triage,1,0,38312,,,,0,60f4b501a0c8ccf15ec5acf756191a10d3cf943b307dbc368ff4b674d912b52e
thread:issue:38744,issue #38744: [jit] `jit.annotate` and multiple compilation of functions with different types,needs triage,47,needs triage,1,0,38744,,,,0,801decf91bccc27ac9f83550acc854ea5eb3334387110286c0c877d580468d86
thread:issue:39240,issue #39240: [mobile] Running on armeabi-v7a is inconsistent with arm64-v8a,needs triage,47,needs triage,1,0,39240,,,,0,074a9adadab450f76f48a3eabaf31ff72674cb08ca4f2e3d8fd330a5bbb53ec4
thread:issue:39250,issue #39250: SummaryWriter `add_hparams` should support adding new hyperparameters,needs triage,47,needs triage,1,0,39250,,,,0,605dc4231e25ea45e481f49d2d71ed5cd05a1e61f699778d95b0d3d6ab70ea8b
thread:issue:39292,issue #39292: Tensorboard，graph in pytorch 1.4 is more complicated than pytorch 1.1？,needs triage,47,needs triage,1,0,39292,,,,0,c0b4012f2b0c506ad5fa4454ce40f940f652e3d0cae7287d86acdaaa3411e2f0
thread:issue:39309,issue #39309: [JIT] nn.ModuleList loses None objects inside it after scripting,needs triage,47,needs triage,1,0,39309,,,,0,dc90cec3c96d1de17cf661d53924a4d853f81a8fd246d189b9412a9750bf8b02
thread:issue:39539,issue #39539: [JIT] Print out mutation in IR Dumps,needs triage,47,needs triage,1,0,39539,,,,0,e6dcd30476616afdc193bee59f7399443615e07e082283879074bd8bb470c60b
thread:issue:39569,issue #39569: [JIT] OrderedDict doesn't support custom objects,needs triage,47,needs triage,1,0,39569,,,,0,391b01cac59e48e0b339b869cb4f35251aa8cd4288e2e5069e2aad03698b6e19
thread:issue:39690,issue #39690: [Quantization] Output tensor type is lost after serializing and loading back a quantized model,needs triage,47,needs triage,1,0,39690,,,,0,c0d0d68fd46ac1d43a38616e4ceedf95b837d564cc34299e569ea9f04f692102
thread:issue:39724,issue #39724: pickle_save on mobile (no longer works with 1.5.0 release),needs triage,47,needs triage,1,0,39724,,,,0,1a35aeaa3f301b11179735b1130fb5948fd49eb01aaeb99e24e8d96a5e508f61
thread:issue:40500,issue #40500: Not able to launch tensorboard using pytorch,needs triage,47,needs triage,1,0,40500,,,,0,ad00cc664ed752aad481cf27e03aff186cb73c4abcaa176c7d190d713e5c2885
thread:issue:40561,issue #40561: Cannot manually assign a tensor to .grad from TorchScript,needs triage,47,needs triage,1,0,40561,,,,0,ccf2d564b93ae88edb6914a612c1843c7e35a9bdc71a915e19ee4a241d063bf8
thread:issue:40696,"issue #40696: Trace model with floor, Inconsistent performance",needs triage,47,needs triage,1,0,40696,,,,0,8ce41c602ad4ec283f1c44d2bd754b516a4f197378960115f78cc528b87b6ff8
thread:issue:40870,issue #40870: [JIT] cannot statically infer the expected size of a list in this context for positional arguments to Tensor.view(),needs triage,47,needs triage,1,0,40870,,,,0,5c95accdec3ae2a932aebc63f0c8d98f1f357b1c4885133bab2a22391ff6e353
thread:issue:40906,issue #40906: will you support SRU on mobile?,needs triage,47,needs triage,1,0,40906,,,,0,a92adb6b797b2f65bdbc1643f4852fbf4644eb664d13be0313b28c4afdc84390
thread:issue:41046,issue #41046: Human-readable names and operations for TorchScript model graphs,needs triage,47,needs triage,1,0,41046,,,,0,5730648e2ddc22bba29c789d4f90b59e295d73fcf85e27ecfa6f8cf1a63f6b53
thread:issue:41141,issue #41141: Sum of two tuples becomes a list in torchscript,needs triage,47,needs triage,1,0,41141,,,,0,cb01e6161a70197f01dede082f8e21d292662364602a7bf85c0feb7f35d9c3cb
thread:issue:41242,"issue #41242: test_backward_deadlock fails with ""Directory not empty""",needs triage,47,needs triage,1,0,41242,,,,0,76b4459c136a0c81ca9a6a98f071d8637172a0b20727fb699a971936d0ae4fa5
thread:issue:41291,issue #41291: Return list of tuples from custom operator,needs triage,47,needs triage,1,0,41291,,,,0,262e65e96e004d99c9aa33b4e5baa341a388cfb13821d7f799e3aca7847faac3
thread:issue:41415,issue #41415: [jit] `set` builtin support,needs triage,47,needs triage,1,0,41415,,,,0,f0348b15a7767a2faf8f40718326c49232f2a0790f6b9b9c155f2063bd884976
thread:issue:41416,issue #41416: [jit] support for generators and `yield`,needs triage,47,needs triage,1,0,41416,,,,0,6727d5f51f570ccb979812cb1769e8777d09ca19b9ed161ff3b79a84ce57f27c
thread:issue:41417,issue #41417: [jit] support class polymorphism,needs triage,47,needs triage,1,0,41417,,,,0,63325d0955324ee63c14e09bf4f98dbb5606d99a6e4cb669b6a818798ac7c96e
thread:issue:41418,issue #41418: [jit] support `rpc_remote` and `rpc_sync`,needs triage,47,needs triage,1,0,41418,,,,0,f4eaeb494e6f0ce80742e37bca9e25ffcc0c5b1a9ddac63084df39fb20b0539b
thread:issue:41419,issue #41419: [jit] Support NamedTuple in tracing,needs triage,47,needs triage,1,0,41419,,,,0,00fbf185328573ea5f32ebc38949bf982cf95f48d255d03f4cf7ba6050073c50
thread:issue:41433,issue #41433: [JIT][to-backend] `selective_to_backend` infra,needs triage,47,needs triage,1,0,41433,,,,0,5d68c04056b0a2b216076b817905001a84167fc27f55ce6429b8d953bb18dbf6
thread:issue:41492,issue #41492: Make torch.iinfo/torch.finfo torchscriptable,needs triage,47,needs triage,1,0,41492,,,,0,85037809025dea98fe97a46b2cc20416084563acf1c16d54e33326a6a856f287
thread:issue:41512,issue #41512: Tensor.new_tensor is not supported,needs triage,47,needs triage,1,0,41512,,,,0,3c39b91c98934bbd25518366d02d81b0e79d101a5031ad6cb6d4613c36347b37
thread:issue:41513,issue #41513: Python object None check not supported,needs triage,47,needs triage,1,0,41513,,,,0,f2c3c3e2e2cd2d126ef1f10ed73abe78e6c49bca4a685bdb59483ecd20dc6b6c
thread:issue:41683,issue #41683: 'torch::jit::script::ErrorReport' from 'torch::jit::load('modelpath') in Visual Studio 2017,needs triage,47,needs triage,1,0,41683,,,,0,febb6282ccc6b92d71db3892f04d7ee87efcf0c0a333507358b841335892ed11
thread:issue:41869,issue #41869: TorchScript pack_padded_sequence and pad_packed_sequence run time error,needs triage,47,needs triage,1,0,41869,,,,0,2948b8d1ffe1218f55b55c66c9df391bd1fa5d7b9e30e80810899474dc263da0
thread:issue:41909,issue #41909: Dropout in TorchScript under torch.no_grad() produces incoherent results,needs triage,47,needs triage,1,0,41909,,,,0,0ac9f0d572209a058cc77d8ef38d96464b7fd9ce3dc61db4646921324abc8f71
thread:issue:42039,issue #42039: Segmentation fault (core dumped) when running optimize_for_mobile,needs triage,47,needs triage,1,0,42039,,,,0,000b22d6819e667837dc7d806ceeec1d7237e222fe650022345eebd5366626fa
thread:issue:42081,issue #42081: Document optimizations enabled by JIT scripting and tracing (TorchScript),needs triage,47,needs triage,1,0,42081,,,,0,82174f71a75b9aa2bc2851fef40cf9e37cd3d03f0baae463bde9a7ddfd7dffc6
thread:issue:42145,issue #42145: Support Python type() in TorchScript,needs triage,47,needs triage,1,0,42145,,,,0,43d71253f01b267ca213dca49263fe6e98042a6294190aac9d85f058014e51dd
thread:issue:42170,issue #42170: Tracing a[b] records converting b to fixed device type of a,needs triage,47,needs triage,1,0,42170,,,,0,6f90d2ac1b421631c8c8c7082d6bd8861218af418b6e8ec774bc984fc913c5f8
thread:issue:42207,issue #42207: torchscript does not support unicode strings,needs triage,47,needs triage,1,0,42207,,,,0,9d0657043cee82e37f69effdce2138d00a0ae934c07862ec08a7a27d1fafd6cf
thread:issue:42267,issue #42267: assertRaisesRegxWithHighlight isn't triggered correctly,needs triage,47,needs triage,1,0,42267,,,,0,7dd0167c8d30eb8c4ff87ca02e52e121052bf5a4a270488bdb5dab9d901258e3
thread:issue:42308,issue #42308: Not able to run quantized model on android,needs triage,47,needs triage,1,0,42308,,,,0,64d83982874f88a93fd8478c8c3c8407080c5ec4daa3f8516304406e634c667b
thread:issue:42379,issue #42379: TensorBoard Summaries Written with `SummaryWriter.add_scalars()` are not Purged,needs triage,47,needs triage,1,0,42379,,,,0,7d029c706e407e7b943de5e134ca5f01934cf52ba8b7e61f6dfc7e33ed687771
thread:issue:42592,issue #42592: Lower performance when forwarding JIT model with libTorch 1.6 compared with 1.5,needs triage,47,needs triage,1,0,42592,,,,0,2a16d192f75c0444d4c0ef2f1ef172f85dc474e97cc185d7ce856ac9443d05bd
thread:issue:42673,"issue #42673: Unexpected behavior of ""to"" method inside a torch.jit.script decorated function",needs triage,47,needs triage,1,0,42673,,,,0,4e8e801b3ec48ca68159cf099d0059dae5e2832e0d2d16ada9cde3b4a63808cd
thread:issue:42885,issue #42885: Unable to call `super` method with TorchScript,needs triage,47,needs triage,1,0,42885,,,,0,79964aae582732f52c008d9ef775862f2f2380c772e783fdbed354e1cc38165d
thread:issue:42947,issue #42947: [jit] std::exception_ptr can't be torchbind/convert to ivalue,needs triage,47,needs triage,1,0,42947,,,,0,a5c6f9cbe4f6b03fb6eb838cfdbfe98ef2e8975d224e4512c83a7b3cf3fc48f9
thread:issue:42948,issue #42948: [jit] TorchBind std::chrono::milliseconds,needs triage,47,needs triage,1,0,42948,,,,0,ca9011bc0cb8fbeb2dc72ef787bb7f2c00d8052e2e8363352cecab1a49f56868
thread:issue:43119,issue #43119: Tracing can leak CUDA memory when using tensor constants,needs triage,47,needs triage,1,0,43119,,,,0,348dd12be33c299203853e1aebdc1adbcc7d74de8d26abee520851b70eeac66e
thread:issue:43134,issue #43134: [jit] need a better way to handle mix CPU/GPU (Inference/Training) for tracing,needs triage,47,needs triage,1,0,43134,,,,0,bebeb423eb415b24328671776ade31aeb76e1a753bc7d9064b4d01291f898306
thread:issue:43237,issue #43237: [jit] remove shape/device specialization from tracing after ONNX done with its own shape inference,needs triage,47,needs triage,1,0,43237,,,,0,818e797cfe7f05fc6764880ba50db35d4ce8c082c1a5db952977507b70ce03e0
thread:issue:43249,issue #43249: [JIT] Cannot iterate over nn.ModuleList in JITted code when accessed through one level of indirection,needs triage,47,needs triage,1,0,43249,,,,0,f377ade3f012cfd823cded65f75ae66126368e081ba6e02bbd05e0420018bb9c
thread:issue:43261,issue #43261: OSError: could not find class definition when export torchscript module,needs triage,47,needs triage,1,0,43261,,,,0,90de6b586166ff9f5d8ef038cd72b948bcf6add671e3091d04b39b4e63987057
thread:issue:43323,issue #43323: Set SummaryWriter step globally,needs triage,47,needs triage,1,0,43323,,,,0,10ca051e06cff780accca8c1d9c72325fdba211c8f9db1601a9679fd13b0bc66
thread:issue:43328,issue #43328: torch.utils.tensorboard.SummaryWriter.add_histogram() execution time explodes with each epoch.,needs triage,47,needs triage,1,0,43328,,,,0,06425a105b569637578688ec032459c926d59b11bafab8de2098c9aff679e6f5
thread:issue:43378,issue #43378: [JIT] Using Any type variable inside can produce incorrect results due to type unification problems,needs triage,47,needs triage,1,0,43378,,,,0,215669253c761bdad5ede26c1e89a5bcea75786a2332e0c5dee888cafd401d00
thread:issue:43388,"issue #43388: Abort message: ‘terminating with uncaught exception of type c10::Error: _ivalue_ INTERNAL ASSERT FAILED at ../torch/csrc/jit/api/object.cpp:19, please report a bug to PyTorch. (_ivalue at ../torch/csrc/jit/api/object.cpp:19)",needs triage,47,needs triage,1,0,43388,,,,0,54442347319a7b7917afac36ee2145af4591a266545d1965d6b266b261720bf7
thread:issue:43409,issue #43409: Switch C10_EXPORT_CAFFE2_OP_TO_C10 to new operator registration API,needs triage,47,needs triage,1,0,43409,,,,0,9a0b5ca6406a40b7afc7edf4aba0e93c8ee489d99900dda8355eb52cb4f855dc
thread:issue:43488,"issue #43488: c10::Error: Couldn’t find an operator for aten::dropout(Tensor input, float p, bool train)",needs triage,47,needs triage,1,0,43488,,,,0,407ba71a4d40941b203f05caea38ee561286dbaa8f36089659166b8f330d3573
thread:issue:43520,issue #43520: [jit] cannot use functools.partial with TorchScript,needs triage,47,needs triage,1,0,43520,,,,0,2a36baa998e35c10ff12944cdd47bf0503e7968feba6e2ebbc86157397c2b33e
thread:issue:43765,issue #43765: JIT fails sanity checks during tracing torch.rand_like,needs triage,47,needs triage,1,0,43765,,,,0,c5794d3aff1ccf7bbf3d45c82b36064163c1f11d1874550312ae90e314f09c1c
thread:issue:43873,issue #43873: PackedSequence objects that are created within jit.script are not PackedSequence,needs triage,47,needs triage,1,0,43873,,,,0,6007f8678d2d453b3023ae1907077375b6b85ad2ef75699987831cfbe4d238d3
thread:issue:44446,issue #44446: [jit] class type optional type annotation not working properly,needs triage,47,needs triage,1,0,44446,,,,0,28edb9957a8692a333894e173b600414d79e8859a3098faaa18d10aa565ea8cc
thread:issue:44469,issue #44469: Error in the converting to pytorch mobile..,needs triage,47,needs triage,1,0,44469,,,,0,6789b3106b90e2111f7ec68b06ec24bb776305f4d4795428196832831e658a0e
thread:issue:44503,issue #44503: TestJitGeneratedModule.test_nn_CTCLoss_lengths_intlists fails if not skipped,needs triage,47,needs triage,1,0,44503,,,,0,c3654069c2a6fcf7163588346daf1b079a2bddc96c925aa3818baf3d6a9c404f
thread:issue:44608,issue #44608: building torchscript extension results in INVALID TYPE: Only int64_t and bool are supported as an integral argument type custom_class,needs triage,47,needs triage,1,0,44608,,,,0,a18938b483fc691892424bf611ad4a2011f9a1d594ab14629464ff081e753c03
thread:issue:44630,issue #44630: No pytorch_jni in StartLocal's zip,needs triage,47,needs triage,1,0,44630,,,,0,c9e6ee059d0c643073c2b224800ee892c59a88cf2bbffedc9cf5ad173ecb917f
thread:issue:44800,issue #44800: torchscript continue training,needs triage,47,needs triage,1,0,44800,,,,0,1888da789592e978b97181499750f3aaa4c977de1ca3d05a2eb5ff36e2b4ec19
thread:issue:44951,issue #44951: copy.deepcopy not working properly for jit.TopLevelTracedModule,needs triage,47,needs triage,1,0,44951,,,,0,1182d697b953e5774898331f13c439e427054c99bd7cfa8df0d63aadb56a7bd1
thread:issue:44992,issue #44992: [jit] cann't pass class objects across boundary of tracing/scripting,needs triage,47,needs triage,1,0,44992,,,,0,544facfae0f9ec1180e999d0e12b6054cfda34b763b8b81635e9021d8e050ac6
thread:issue:45012,issue #45012: torch.jit saves nonpersistent buffers,needs triage,47,needs triage,1,0,45012,,,,0,4052cf45f1652f858c1fdf6f4c71112ccf40d60db91cfd02c19a5a29bb6d339c
thread:issue:45199,issue #45199: JIT doesn't support rfloordiv with scalars correctly,needs triage,47,needs triage,1,0,45199,,,,0,6726c3d3f4f6d6a897e513a24d9477876f9a41fd5cc0e7056e0803ed44e3c118
thread:issue:45327,issue #45327: Tensorboard summary_iterator,needs triage,47,needs triage,1,0,45327,,,,0,762771e6d5abf6bf38193581f1199891239014a1fd79614dc294dfea6ec59d17
thread:issue:45398,issue #45398: build_android.sh,needs triage,47,needs triage,1,0,45398,,,,0,b91a3c11ce24fba15588c57fea7eb0a17c1d5de59745782cca34b03062063904
thread:issue:45524,issue #45524: [JIT] JIT Autograd is saving more results than are necessary for bprop like the result of RELU when it is an intermediate,needs triage,47,needs triage,1,0,45524,,,,0,7f6df5a4f2275226bfffb93c1b3523210979b0cf5359f1fbfa20ac9bee557e13
thread:issue:46368,issue #46368: [JIT] Inconsistent behavior of torchScript parser in python and c++,needs triage,47,needs triage,1,0,46368,,,,0,1852e0b05ed8300451289439150100ab6bb7b1ec0a4dc4f262eaf33b44c61780
thread:issue:46375,issue #46375: JIT: error in LSTM with `flatten_parameters`,needs triage,47,needs triage,1,0,46375,,,,0,2fe0cb80ae80253d1686e733f4b5e4a5364788e2d641a612d70fd660fb7612ca
thread:issue:46460,issue #46460: C++ API: Training and inference of torchscript modules on multiple GPU,needs triage,47,needs triage,1,0,46460,,,,0,f304dbaf35ea50a9ac9d71b4b19b9b8ce3848f887fa349dcbe49b78013cb1ddc
thread:issue:46505,issue #46505: Cannot script when a static method or cross module function calls custom op,needs triage,47,needs triage,1,0,46505,,,,0,b73a1ce2e66a7e498790597a7e73fa5b6aa199848e26af43d9cd9d47f353a852
thread:issue:46574,issue #46574: Mismatch between the documentation and implementation of caffe2 PReLU operator,needs triage,47,needs triage,1,0,46574,,,,0,4bdd49bbd940207844e7c5a462853b3b7bc9732bbd5e8392532543af471cdb38
thread:issue:46654,issue #46654: Converting object detector model to TorchScript,needs triage,47,needs triage,1,0,46654,,,,0,6e8cdcbd86f00159dcfa62364b44b5224b3ed8566d805c483c4f2576113d379a
thread:issue:46907,issue #46907: SummaryWriter deletes data automatically when there are too many,needs triage,47,needs triage,1,0,46907,,,,0,bd0b102f6d2b5096352cb149427e6211b4ad1079731b53fbc246bab2b49e2e3e
thread:issue:46979,issue #46979: [JIT] scripting torchaudio.transforms.MFCC() error,needs triage,47,needs triage,1,0,46979,,,,0,7561fec7715f3f253cc66ab679449e5b9e1018b58cfa38039817ba36b366f8f7
thread:issue:46995,issue #46995: max_pool2d always fails on native Android app,needs triage,47,needs triage,1,0,46995,,,,0,01ee4cddc4bec94d0adfb073ce757ab9f037a7aac637f3ce39a850f980e268c5
thread:issue:47010,issue #47010: [JIT] Tracing turns None Device Input into a Constant,needs triage,47,needs triage,1,0,47010,,,,0,35ea7245c0ec8eaa9c26da95fe420d3866b05bcc1bf7b6a7f7542ad0425a6925
thread:issue:47162,issue #47162: Custon C++ classes are not traceable,needs triage,47,needs triage,1,0,47162,,,,0,233897db3eb913aebefe041c4b899a60d9595868814bc64a291a2900851f29ca
thread:issue:47163,issue #47163: Grad strides do not match bucket view strides.,needs triage,47,needs triage,1,0,47163,,,,0,1319138e671376b605b604e181e94fd8d94e081e2a1feb95260ebede25b547d2
thread:issue:47177,issue #47177: LayerNorm on Android will cause very high cpu usage,needs triage,47,needs triage,1,0,47177,,,,0,13799c8666d0a4c34faff6ed4ff8ee0b736eb162d2313640c95d94fd3b0a0a14
thread:issue:47265,issue #47265: [Feature Request] tensorboard add_graph support for multiple method,needs triage,47,needs triage,1,0,47265,,,,0,3b8f660dc50444d9e7c0b14dc836c270fe544024f11e2d41a578cb56afb86a39
thread:issue:47335,issue #47335: [jit] support length inference for list comprehension,needs triage,47,needs triage,1,0,47335,,,,0,1d237ae8c5fc5b281bdb765dba9d0b1599b15094d5ee00b4d06a3a01bb95e409
thread:issue:47336,issue #47336: [jit] Properly support slicing of ModuleList,needs triage,47,needs triage,1,0,47336,,,,0,c8636f8e67cf09de7b2a5c0e3075fb432c285cf69e676e34d34b81734f0b1501
thread:issue:47342,issue #47342: My demo app uses other package name will crash but org.pytorch.demo,needs triage,47,needs triage,1,0,47342,,,,0,e0bf70ba41b1555b48e20e7eab5e822182f3ff64e9fb54c1b8ca36f26f3ad942
thread:issue:47565,issue #47565: Keyword-only function not allowed with TorchScript,needs triage,47,needs triage,1,0,47565,,,,0,61e138c4bc70830879fcd3bb7247eddb6360ddc42ef29331b09f98f0eb3381af
thread:issue:47682,issue #47682: Common dunder methods fail when called directly,needs triage,47,needs triage,1,0,47682,,,,0,5da7e0e25fe258df928a32c70501b4183fd37c77cbc7bd370601cabd92372a93
thread:issue:47692,issue #47692: pyre not able to infer the type of torch.jit.script,needs triage,47,needs triage,1,0,47692,,,,0,1eafefe22476e91bb01d81e89396bc18a8a9aac13156fc3eec14ceb75f50372d
thread:issue:47883,issue #47883: `.modules()` is not callable in TorchScript when any submodule is a module interface type (i.e. a class decorated by `@torch.jit.interface`),needs triage,47,needs triage,1,0,47883,,,,0,55f676f1e0318b8e2980c6eacd76eb491cd95683af897dc269e389dff039a1bf
thread:issue:47887,"issue #47887: ONNX export of a scripted submodule fails with ""Modules that are called during a trace must be registered as submodules of the thing being traced""",needs triage,47,needs triage,1,0,47887,,,,0,b171ca9263936d5b4fa3d627e6c592dec6c93075371f27bfe7cccf62d7dfe28a
thread:issue:48045,"issue #48045: [NNC] Improve ""UNSUPPORTED DTYPE"" error messages",needs triage,47,needs triage,1,0,48045,,,,0,619193a16d33b2759b24c07bb9d0f68d7cac29410c3a8a5588f1804ad4483833
thread:issue:48066,issue #48066: [JIT] Saving and Loading JIT Model Does Not Preserve __dir__,needs triage,47,needs triage,1,0,48066,,,,0,c4cad670edf538972351dc5c44cb40ef8321ef6334dc77a1171f9ee35b317bcc
thread:issue:48082,issue #48082: Microsoft C++ exception: torch::jit::ErrorReport at memory location 0x0000001C0C3BBB40.,needs triage,47,needs triage,1,0,48082,,,,0,c3f89edd8e2933be2844897b5cff4a75c46814656265373520f2c72a21a6e010
thread:issue:48089,issue #48089: [JIT] Optimization pass in profiling executor to fold away conditional branches,needs triage,47,needs triage,1,0,48089,,,,0,22ebae5ab3a0d9facc55e09ae5420d73993bd191ba1e4fb5b22610e0312a5c9a
thread:issue:48090,issue #48090: [JIT] Constant folding on no-op `aten::mul`,needs triage,47,needs triage,1,0,48090,,,,0,ca62f065b08c0535d223d1ffa9f568346bca74d753cfb5497636c41815f3b43c
thread:issue:48094,issue #48094: [JIT][PE] Profiling Executor should re-profile `DifferentiableGraph`,needs triage,47,needs triage,1,0,48094,,,,0,5def824caf22ec8fa7a7b39a5047cc358b8539bd91c2614c56580c174134f48b
thread:issue:48191,issue #48191: [JIT] Contained Module Attributes should be recursively compiled,needs triage,47,needs triage,1,0,48191,,,,0,d015182d97c60cb63afefca809fbc356264c77bc6a7d12ded255ee68464d6465
thread:issue:48307,issue #48307: Provide torchscript support for staticmethods for classes derived from Enum,needs triage,47,needs triage,1,0,48307,,,,0,3f77e33bbb382e41d38915479179939b99be30cd2a291e2ccf6669b11e18bd24
thread:issue:48325,issue #48325: [JIT] Constant Propagation Shouldn't Run on Values that Escape the Graph,needs triage,47,needs triage,1,0,48325,,,,0,4840495cb75999ed0028383a03ab7c795b04c2ad364c4e14a5833caf6d48198e
thread:issue:48387,"issue #48387: Error occurred when trying to call ""torch::jit::load""method for loading "".pt"" file on IOS",needs triage,47,needs triage,1,0,48387,,,,0,9cb260213893ad6896b5443bcd880acd64f2c963cebdc3757a02266f8743d1ef
thread:issue:48408,issue #48408: TorchScript: Optional sized lists cannot be called with optional scalar values,needs triage,47,needs triage,1,0,48408,,,,0,613696d319f770bc5d07f5d7c6b25b8b684d6d71cf529e35077c380fe0810a17
thread:issue:48410,issue #48410: PyTorch Mobile speed_benchmark_torch crashes on mobile optimized model,needs triage,47,needs triage,1,0,48410,,,,0,57bf00d6465cf2bce11ad9b19eb9943579e9ee66b0d2aa811a7789ad99fbca69
thread:issue:48452,issue #48452: After updating to Xcode 12 and LibTorch to 1.7.0. Facing issue when running unit test.,needs triage,47,needs triage,1,0,48452,,,,0,0ace5d8a6bcd576c3ea88a638fc2c78c8275b406e3dee7451f935702e2ed0d5f
thread:issue:48478,issue #48478: Callbacks of Futures shouldn't wait inline on another future,needs triage,47,needs triage,1,0,48478,,,,0,ca80d1fdaa30fbab07f58b08b8fb8775c612d691902083219c77ca985ae75c93
thread:issue:48703,issue #48703: Caffe2 upsample do not support float times output_size,needs triage,47,needs triage,1,0,48703,,,,0,96844e2dace2a525e6c92c318270b9b4c43a11642dc78d6ff76051d07b9b8c51
thread:issue:48710,issue #48710: RuntimeError: CUDNN_STATUS_BAD_PARAM when testing environment,needs triage,47,needs triage,1,0,48710,,,,0,ad9d922f0c6e190d6c67e1c4e44b1efaa50410e41f33ecb577bc32639c68ce7e
thread:issue:48869,issue #48869: Tracing fails sanity check,needs triage,47,needs triage,1,0,48869,,,,0,11fe2a03da2d02cf66b645c0b82ee636f13da317e517d85a9b7640f1eea65c2f
thread:issue:48928,"issue #48928: When multiple GPUs run multiple processes, it is found that any process not running in GPU 0 will have some more memory (such as 200m) in GPU 0. What is the cause of this?（多个GPU跑多进程时候，发现只要不在0号GPU跑的进程都会在0号GPU多出一些内存(如200M)，请问这是什么情况导致的？）",needs triage,47,needs triage,1,0,48928,,,,0,3d72bf3bc16cec72bb84e1cffde38044c580e026d7d4c196c083a7fb4bf9a4b4
thread:issue:48984,issue #48984: Scripting silently ignores methods of class derived from `NamedTuple`,needs triage,47,needs triage,1,0,48984,,,,0,2956143654928bf9e89c1edef62a8cbceb59875e2eecad9b794bb897b0b2bd11
thread:issue:49253,issue #49253: Differentiate between objects with a true `Tensor` type and objects with a default `Tensor` type,needs triage,47,needs triage,1,0,49253,,,,0,4a3e77872fb5bebbd5d2a3a94e42bf40de0cd139ea4725013b437e9cd3f5e400
thread:issue:49261,issue #49261: [TorchScipt] Custom Objects C++ API,needs triage,47,needs triage,1,0,49261,,,,0,7a96757e6498fabc24faa7ed33ceed391fbc9e3d57b750e6647354c46edbc455
thread:issue:49377,issue #49377: [JIT-DEBUG] Suppress GRAPH dump in profile-graph-executor,needs triage,47,needs triage,1,0,49377,,,,0,9314f05e6e16cbc5ef8eb6a365f4f5ec05f83d3bb617873097dee6d371bd0b8e
thread:issue:49460,issue #49460: Libtorch: Segmentation fault when running torch::jit::load,needs triage,47,needs triage,1,0,49460,,,,0,2448af8c3c383718ba4dd4259fea5d7cd60c16dc29c341de749c8ef7a911e2c9
thread:issue:49536,issue #49536: Pytorch mobile Vulkan API bug during call to ReLU,needs triage,47,needs triage,1,0,49536,,,,0,6fa82464e5d01d96e7d6749b3e1bb8cbe5e83d1f4f526735caadad5fc74fbe7b
thread:issue:49538,issue #49538: jit tracer doesn't work with unflatten layer,needs triage,47,needs triage,1,0,49538,,,,0,2efab8cc63b6301fdb6bd5435cf744f2afae174361316f4855819bfb061430c5
thread:issue:49569,issue #49569: Update jit/OVERVIEW.md,needs triage,47,needs triage,1,0,49569,,,,0,2671607b6dec822c28df1fc7f66e9bd750b3c299da2836044206e582a58e7597
thread:issue:49585,issue #49585: [JIT] Serialization forward compatibility tests,needs triage,47,needs triage,1,0,49585,,,,0,84b0fbf36c5e6240eaf830f818dc3a1203304c2363b5609305749f64c3519a6b
thread:issue:49614,issue #49614: Improve error on missing/problematic Module member type annotation,needs triage,47,needs triage,1,0,49614,,,,0,b07634c64063b83991b8736f495621b31b6c2505808d8a79ef44360a3c940bc8
thread:issue:49643,issue #49643: Tracing doesn't work with spectral normalization,needs triage,47,needs triage,1,0,49643,,,,0,f18207ce6e53490ebe085ab4114a879cac5aed0f2e8579259b31f4e8943035c0
thread:issue:49650,issue #49650: Annotating submodule with correct type fails scripting,needs triage,47,needs triage,1,0,49650,,,,0,1ae22555a5e4dc1ba6b8640f8a64e35d9550f75748e153d6c324005c5285e9b8
thread:issue:49782,issue #49782: TorchScript model's output different on Mobile and PC,needs triage,47,needs triage,1,0,49782,,,,0,aaf3ffacf2bfbabaceb881f53e97e32d226da6742dc72b02a1987f2ca4cc738a
thread:issue:49820,issue #49820: [JIT] Don't support math.exp and math.log in the Script,needs triage,47,needs triage,1,0,49820,,,,0,c53ca7fdf1bbdf15ac2754ca3aebdbb5c794c827fe8d0a09f0de7c8c7fb16d0e
thread:issue:49823,issue #49823: Why trace method used insted of script method in graph creation?,needs triage,47,needs triage,1,0,49823,,,,0,9c84de263798d128c2343e21ea5786e473c9b05a6a64dd9100181f74ed277b59
thread:issue:49935,issue #49935: [FX] intermediate types of empty lists/dicts not preserved during torch.fx tracing,needs triage,47,needs triage,1,0,49935,,,,0,a696acdb50d64fc2092d5ed1b831221070fd8f2ca070ff237401811c23021477
thread:issue:50127,issue #50127: JIT does not correctly compile custom classes derived from torch.Tensor,needs triage,47,needs triage,1,0,50127,,,,0,626ed8ab5bf5f4d0940fa8584e1296d8d71c721b7a9f03bf3dc87fba113c0eb7
thread:issue:50231,"issue #50231: Tracing with autocast failed with error: ""Cannot insert a Tensor that requires grad as a constant""",needs triage,47,needs triage,1,0,50231,,,,0,e10cbb23f375f09e20326d1962a7c747f868cd5db9b2459d6369544df3a9eeec
thread:issue:50531,"issue #50531: PyTorch mobile perf recipes, fails when building android",needs triage,47,needs triage,1,0,50531,,,,0,daf7040939d7af99b46040508a37878bf1f9ab329f204ae7110dbce6db7580e5
thread:issue:50714,issue #50714: [PyTorch Mobile] Android speed benchmark binary crashes when using vulkan,needs triage,47,needs triage,1,0,50714,,,,0,bd63c1b96dad67bfee50d1df8352d054b3722a44a4b7eed94be24fb56be34ed3
thread:issue:50734,issue #50734: [TorchScript] Document TorchScript IR,needs triage,47,needs triage,1,0,50734,,,,0,6fecd58367d19ef8671a68b620809b641bb11d78f99f9abbcea3239e4bd9d5b5
thread:issue:50772,issue #50772: [TorchScript Usability] Incorrect default types,needs triage,47,needs triage,1,0,50772,,,,0,6bc8658b2efb2a442379af5ccff8331f3e81f446523d1a98bc65e9405f026b03
thread:issue:50834,issue #50834: Precision of scalars is lost in script mode for binary operations with tensors,needs triage,47,needs triage,1,0,50834,,,,0,d9c82fa52f15f67e0c00a1169116e1d3d5d428cb2e9315f3c7e3f806092fed16
thread:issue:50933,issue #50933: [JIT] Type refinement failure in if-elif-else structure,needs triage,47,needs triage,1,0,50933,,,,0,835dddde626cdb3429a47a07b6e0d235ecbc946b77cbc8c4b290111cf39d208f
thread:issue:50943,issue #50943: Exporting opnames fails when using custom classes,needs triage,47,needs triage,1,0,50943,,,,0,2c1e86f24b2ada94cd72015c15a2de57d107aa365e35337c0b0313269bc2df6a
thread:issue:50971,issue #50971: Failing to trace Huggingface models using Trace API from Torchscript,needs triage,47,needs triage,1,0,50971,,,,0,bad2a63dd846fee5f838daed36ca90c31eb875085cfd01ed4b3fe9a83ba7a6c7
thread:issue:51055,issue #51055: Native code assert or hang on process exit with a torch.jit.script model on multi-GPU,needs triage,47,needs triage,1,0,51055,,,,0,3a8efd586dad3929f92f97b5793cd754e7c1373c5edfaf7246958267196dd185
thread:issue:51154,issue #51154: [JIT] Rough edges found while exploring expressions supported by TorchScript,needs triage,47,needs triage,1,0,51154,,,,0,ec8fdc60a98a95c8d1fde7209feed9741a0f60fc9eea2ce5b7f3652e1c9c2b9a
thread:issue:51328,issue #51328: Support torch::jit::script::Module converting to torch::nn::Module.,needs triage,47,needs triage,1,0,51328,,,,0,ceab9fc809dee737a282e6ed3ab8892f172a5133470a67571441674707b33e98
thread:issue:51376,issue #51376: Huggingface model not being torch scriptable,needs triage,47,needs triage,1,0,51376,,,,0,061e1c87fb9ffaa4e0a986b76efcfc243b960e02370083a1b1d007bb595e2dda
thread:issue:51379,issue #51379: TorchScript class compilation failure with pybind11-2.6.2,needs triage,47,needs triage,1,0,51379,,,,0,3a14f15d0aed236dc5b0be760db74ca9748b8c414b0b84fb1787def3af103b75
thread:issue:51409,issue #51409: C++ load model error,needs triage,47,needs triage,1,0,51409,,,,0,1b6176faf50e031833beb0952d270c15ca0259ebd8ff214f4f4d7d29c65d4ee5
thread:issue:51445,issue #51445: Saving embedding vectors as TSV for Tensorboard is not efficient,needs triage,47,needs triage,1,0,51445,,,,0,0bac7cbf911259719f948d4ec95a42a63d8722e6557779cb2de13bba0cb2edf4
thread:issue:51465,issue #51465: torch::jit::Module cannot be forward declared and used into a `unique_ptr`,needs triage,47,needs triage,1,0,51465,,,,0,ca34ebb18966f2b9df031cdf03a75da5b9efd1356f24b0b924d5147f9f640b70
thread:issue:51574,issue #51574: [Pipe] conversion to Sequential: various complex scenarios,needs triage,47,needs triage,1,0,51574,,,,0,d408b354db25eb7d5900e1c44c399f7ff3e4456960551e34955f2c8249d279fa
thread:issue:51616,issue #51616: [RPC] rref.get_type() should have default return type of future,needs triage,47,needs triage,1,0,51616,,,,0,7d752b3fd9ab5a9af8bf6077db67c7b4745866c7f3d00375c2a38710b322d1ce
thread:issue:51673,issue #51673: [JIT/Futures] Future callbacks/then() APIs should throw the correct exception,needs triage,47,needs triage,1,0,51673,,,,0,5449b046baa1e27da7782f18bf6b59e0dfaab6a88e395e16c593a26f7b50d2b9
thread:issue:51808,issue #51808: [jit] Keyword-arg expansion for constants,needs triage,47,needs triage,1,0,51808,,,,0,d6f6bd4501cef778779553e9ddf4c6cc311f73388ae3a75dd0c2fd231640aef7
thread:issue:51892,issue #51892: [cuda jit extentions] global shared destination conflicts with virtual envs,needs triage,47,needs triage,1,0,51892,,,,0,b92434ca61afebc24480d3ca7203128a219d40edb859b04bac9756d0dd1b6826
thread:issue:51931,issue #51931: [Pipe] Tied layers,needs triage,47,needs triage,1,0,51931,,,,0,156b309265980fac3df3acc7b10a9b32e9c9269f745c713ca0aa85b910539398
thread:issue:51932,issue #51932: [Pipe] Memory-Efficient Model Construction,needs triage,47,needs triage,1,0,51932,,,,0,910f9361674edcf7aa22323206aae22dbf223b434388bc8d99b3817a68fa9900
thread:issue:51941,issue #51941: RuntimeError: torch.dtype passed as int from JIT to ignored function,needs triage,47,needs triage,1,0,51941,,,,0,f4598553bab291c3b160c21e93aa14289c2274545bff044ea725ce567d7d4eb7
thread:issue:51942,issue #51942: Overwritten method has wrong self when JIT'ed,needs triage,47,needs triage,1,0,51942,,,,0,3f218d950facf9b46d66322ad1693aef887ad13d2c206a3762c79a084daba518
thread:issue:51969,issue #51969: Support backward hooks in JIT,needs triage,47,needs triage,1,0,51969,,,,0,f0ab4a2eb5d2852f7552a7a716f96de242e099740c28d8a1f2079d3f91ba0994
thread:issue:51994,issue #51994: Allow Custom Classes to register a handler for .to operations,needs triage,47,needs triage,1,0,51994,,,,0,5242a9ebabcb9c8185a55c6a80b1bd11c39ce8261d2f28f4e34d0de6ebf48532
thread:issue:52047,issue #52047: Detect OpenMP Loop and this application may hang warnings,needs triage,47,needs triage,1,0,52047,,,,0,6b97c3bfc5e10aa3d12ee3d984cdc2b46a5bf63158dbf09155d2f8c2fbfb9389
thread:issue:52144,issue #52144: Concatenation of Java tensors (Android for mobile),needs triage,47,needs triage,1,0,52144,,,,0,6a66dd72cd62974febdbe5028dd93f618d2bee84b7bffba1981f48cd370ffedd
thread:issue:52147,issue #52147: Pointer passed where number is expected for PYTORCH_CUDA_FUSER_JIT_OPT_LEVEL leading to crash,needs triage,47,needs triage,1,0,52147,,,,0,60fd7a3d3ecd2d80c09a9ab2fe2ab13a75b9c7afd2943240ca3885cf7a28b798
thread:issue:52214,issue #52214: `hasattr` check not supported on objects other than `self`,needs triage,47,needs triage,1,0,52214,,,,0,db63776ad8f497e3f851e5d7394b79a867c94eadd2b62303000eb77876905b38
thread:issue:52271,issue #52271: torchscript should work with type: ignore comments from mypy even if they are not at the end of the line,needs triage,47,needs triage,1,0,52271,,,,0,6093aca66475f6e22607a22cb1902021648a757f589b15ecb468fc81485138a8
thread:issue:52309,issue #52309: [JIT] __getitem__ on scripted ModuleDict returns unscripted module,needs triage,47,needs triage,1,0,52309,,,,0,5819f640d77b78f3245c673e4c0b35f7fdb6fa968eed4ce5b3646a89f6d4bfb2
thread:issue:52363,issue #52363: RuntimeError: Can't redefine method: forward on class: __torch__.torch.nn.quantized.modules.conv.Conv1d (of Python compilation unit at: 0x5571a223a770),needs triage,47,needs triage,1,0,52363,,,,0,2853e67e666a038695981accc82c9ed030c2a9bb7af034bb2082cc9d8a15274e
thread:issue:52471,issue #52471: all reduce hangs and does not throw exception when CUDA_VISIBLE_DEVICES is not set properly using the NCCL backend,needs triage,47,needs triage,1,0,52471,,,,0,68f35749b4b9d21d43428995bdd0d8abb051199967adb270ea84c599677f87ed
thread:issue:52517,issue #52517: [doc] JIT-compatible c++ custom ops,needs triage,47,needs triage,1,0,52517,,,,0,a4ca796183f37d2f5e79b31bd76c9b55f04be88511287404fad6f94e8cf287ac
thread:issue:52589,issue #52589: [JIT-autodiff] batch_norm isn't differentiable,needs triage,47,needs triage,1,0,52589,,,,0,637c65012e72132ab28046a5dc1f1a4dc848ff16e5bff92d5ed768e96970fb72
thread:issue:52599,issue #52599: torch::jit::Node::isNondeterministic() does not include aten::feature_dropout or inline kinds,needs triage,47,needs triage,1,0,52599,,,,0,21eb550568e9c10bf7ca0a584535411cea810e51856f97ebce9eb53a95fa30db
thread:issue:52607,issue #52607: Enable timed barrier in c10d,needs triage,47,needs triage,1,0,52607,,,,0,6cbe27a1928057621ce354a44f550e0cd28553780ff29c45fbf68e653dad22ff
thread:issue:52655,issue #52655: Optimize `einsum` in TorchScript profile guided optimization,needs triage,47,needs triage,1,0,52655,,,,0,4e796fa42e99d6ee1cdc971335bd95d4319e539d35a5e0bd09d898a7e6a3b461
thread:issue:52720,issue #52720: [JIT] Finish migrating torch._C._jit_ bindings to torch._C._jit,needs triage,47,needs triage,1,0,52720,,,,0,f9c70f66dca5c748f29eb42249a8b98eaf05fa4e86c0654670fa67b16fd17c55
thread:issue:52768,issue #52768: [nnc] Optimize tiny convolutions,needs triage,47,needs triage,1,0,52768,,,,0,036312ce47cf025f4cafd1f2c756104b17e57a724ab6e63572f5ed8942b6cd17
thread:issue:52772,issue #52772: [nnc] Optimize 1x1 convolutions,needs triage,47,needs triage,1,0,52772,,,,0,856f2d655bad0f123c64def98507c2f7703921f83bd0ea5be5bee124414f545f
thread:issue:52777,issue #52777: [nnc] Placeholder::load braced init list fails to deduce,needs triage,47,needs triage,1,0,52777,,,,0,4ebf50dba7b7387ea739e075843669380269370157e5413c00898b74749ea713
thread:issue:52795,issue #52795: Add format indexed argument support in JIT,needs triage,47,needs triage,1,0,52795,,,,0,ceb14fb8a3c99d21dab6abd32ef304e1f468aed251da540f23663f728d8e3733
thread:issue:52821,issue #52821: [nnc] Compute expressions with > 4 axes is pretty tedious,needs triage,47,needs triage,1,0,52821,,,,0,0c2ca4883b5dbec5f7c2496345a2c75b2342adec2b76634630ae40631ef16c80
thread:issue:52970,issue #52970: Wrong link in readme for the android nightly version,needs triage,47,needs triage,1,0,52970,,,,0,e696edeba4e71634f796e39561df6422676f506ed388a7b5c720a8596f39e54c
thread:issue:53140,issue #53140: [JIT] Document __jit_ignored_attributes__,needs triage,47,needs triage,1,0,53140,,,,0,cbf7275035dcf3f746116d9f981cb74d8f34646e86427af2037d8897767f83e6
thread:issue:53264,issue #53264: How to load trained . torch from conversion to .mlmodel,needs triage,47,needs triage,1,0,53264,,,,0,b351124167876b48b9ca930ad00fc75ced18023c26082dc09800087565296e03
thread:issue:53344,issue #53344: [JIT] raise unittest.SkipTest and self.skipTest cannot be used to skip tests in JitTestCase,needs triage,47,needs triage,1,0,53344,,,,0,c181db73fc0bfb0a7df81fe878b7034be1f71acfaa5b51b94b0971aee5923f45
thread:issue:53414,issue #53414: [JIT] Metaprogram non-type based expression trees,needs triage,47,needs triage,1,0,53414,,,,0,be74b735f353cdc8a7971ff1694639b2523592d1aa93001b8943ad6a2cea7126
thread:issue:53421,issue #53421: [bug][JIT] Error when tracing model with partial hooks in PyTorch v1.8.0,needs triage,47,needs triage,1,0,53421,,,,0,b3d429cef543c3bb45f2142cddabb6ba90b93cfdd927f5287dfb53f09dc59b1f
thread:issue:53451,issue #53451: Model loaded from full model state checkpoint does not run on multiple GPUs (device mismatch),needs triage,47,needs triage,1,0,53451,,,,0,b13ceecb71b74f82011af13c48248d52fe7dcb6d273a2f4bc49df5eb7c062076
thread:issue:53472,issue #53472: After upgrade to 1.8.0 memory leak in libTorch after using torch::jit::load(),needs triage,47,needs triage,1,0,53472,,,,0,5453476a48d2e627845cb8669a39c22b9f776a4e33d34e7245de18b6ed9a8e18
thread:issue:53744,issue #53744: TorchScript divison by zero fails to error out,needs triage,47,needs triage,1,0,53744,,,,0,8ba60dd336b82319fe7fb93a82e3b8a36d623b011606c7aa2648a50c1bfb5a09
thread:issue:53753,issue #53753: TorchScript cannot handle declared module attributes of Python types,needs triage,47,needs triage,1,0,53753,,,,0,38004478b14061beac39bdfceb1af01b308e49014c0a5baefbef8c72341ad725
thread:issue:53824,issue #53824: [JIT] torch.jit.optimized_execution(True) greatly slows down some operations in PyTorch 1.8.0,needs triage,47,needs triage,1,0,53824,,,,0,96abf5fefec54fa53feddb2b3c999e0eff147f6f38d74e5bfe2f79a2550e431f
thread:issue:53862,"issue #53862: After the source code is compiled, an error will be reported fvcore",needs triage,47,needs triage,1,0,53862,,,,0,e06d8ad92bf2533c1afca53bbf8c20303b343e4775b7c12c6167600fe9dc712b
thread:issue:53867,issue #53867: [JIT][Static Runtime] Memory optimization for output tensors,needs triage,47,needs triage,1,0,53867,,,,0,a1f1f69cc7cb1aaddd24da5bba3acc53506fe09f56113e03afb1d0d2d44c28d6
thread:issue:53877,issue #53877: [JIT] Bad interaction between if-else-style and assert-style type refinement,needs triage,47,needs triage,1,0,53877,,,,0,69b6bd316870ea7cab0e2a3457acee60796e4ae8d192f783b2f9c8b5f3c3723c
thread:issue:53878,issue #53878: [JIT] TorchScript errors out when assigning a value to an existing variable that's defined in outer block,needs triage,47,needs triage,1,0,53878,,,,0,d10413c2c9ce1dacca42171ff6e74f7ae347b85eae22bf91d01f43f375d0a103
thread:issue:53904,issue #53904: [JIT] Support `torch.ones([])`,needs triage,47,needs triage,1,0,53904,,,,0,823ac4a569d34d63db2a699bf24e3d6a34013a64d77a5b3bd692404d2e49fbeb
thread:issue:53921,issue #53921: [JIT] TorchScript infers incorrect type for `or` in non-boolean context,needs triage,47,needs triage,1,0,53921,,,,0,2940c4c53e2e06be28f4476ba1672c093ec7f34fad7ca0045f39427b84298851
thread:issue:54140,issue #54140: [Torchscript] Error detected in torch::jit::(anonymous namespace)::DifferentiableGraphBackward on custom RNN model,needs triage,47,needs triage,1,0,54140,,,,0,49095d67d85f7508c0d9d2fdab6a3c4d5a66481c7c4c6fd2ecb4e1b76dff07de
thread:issue:54146,issue #54146: Conversion of Pytorch model to Torchscript for model LaneATT(Lane Detection),needs triage,47,needs triage,1,0,54146,,,,0,d270b8c546246158fd570e1a6e65523b0f7d41f71badf3bcb919e86ab921a1c0
thread:issue:54348,issue #54348: [JIT] TorchScript does not correctly handle `torch.` namespace functions called with `kwargs`,needs triage,47,needs triage,1,0,54348,,,,0,39a7bad67ffcce0806802653ffbecbb834a130c2529b4ed3197bd29690d8b49e
thread:issue:54365,issue #54365: Support `Generic` Type in TorchScript,needs triage,47,needs triage,1,0,54365,,,,0,d111e16438a6618bdb5f2e3f36bd7a80c52ffc6837177c51741fc7aef939a230
thread:issue:54435,issue #54435: [JIT] prim::ModuleContainerIndex doesnt work with freezing,needs triage,47,needs triage,1,0,54435,,,,0,86319a0bc2410c9bd34d6ef0c8571c269114c837fdd8fcd735509b7a898799a7
thread:issue:54593,issue #54593: Multiple refinement with `isinstance` doesn't work,needs triage,47,needs triage,1,0,54593,,,,0,79c071acd32770e85a57782a6e5c911fcac82858f7572c892e5ab5d61c2595db
thread:issue:54598,issue #54598: Type refinement with `isinstance` only works with generics,needs triage,47,needs triage,1,0,54598,,,,0,a0867d7deccc42cedc266acbcdd528be1b6e95c135c575070837c6ba986efae0
thread:issue:54599,issue #54599: Better documentation for `torch.jit.isinstance`,needs triage,47,needs triage,1,0,54599,,,,0,e9de7670c8d6d504fbe7f4fa559058f795dac0d1925d3fe018386008cb815df7
thread:issue:54612,issue #54612: Exceptions thrown in JIT interpreter always are translated into RuntimeError,needs triage,47,needs triage,1,0,54612,,,,0,a5c200339f2125a093904292eaf159d41dd6192a5cf658932facd0db7348c438
thread:issue:54625,"issue #54625: [torchscript] nn.Embedding(_weight), tensor.masked_fill_, and torch.autograd.grad causes wrong gradient",needs triage,47,needs triage,1,0,54625,,,,0,6510a9aade67fc2c910a6fda8922df125aff8221b630d3686c1f68628094588a
thread:issue:54728,issue #54728: JIT prim ops v.s. dispatcher,needs triage,47,needs triage,1,0,54728,,,,0,25ed06a1041c4c0d289e4d48b82379dc09c6616a6a28a36ff0c61bcf49fa04cf
thread:issue:54821,issue #54821: Decouple TorchScript compiler from operator library,needs triage,47,needs triage,1,0,54821,,,,0,61b5c1fb58308ab09cae249e3da1262e8f83ba64ae160ae3d622bc2deac252c1
thread:issue:5489,"issue #5489: torch.jit.trace(network, data) fails if data is an OrderedDict",needs triage,47,needs triage,1,0,5489,,,,0,57c09d7e5db8b7408486d57e27b25f87b4e831630b2f27374f13bfdc9946454e
thread:issue:54912,issue #54912: [autocast] DataParallel in a single process - possibly outdated docs,needs triage,47,needs triage,1,0,54912,,,,0,3c1d041217ede452219b81d83129a4effbaf7e72ef088f1812cbf6a499f9ebd3
thread:issue:54937,issue #54937: [JIT] export fast-transformer to TorchScript,needs triage,47,needs triage,1,0,54937,,,,0,c47562c6a622b2e0f990ae12d73f16310ae30327dde53715b3cda6a41665d109
thread:issue:55072,issue #55072: [JIT][Testing] Replace use of `assertRaisesRegex` with `assertRaisesRegexWithHilight`,needs triage,47,needs triage,1,0,55072,,,,0,a232d1f91d115d5986150bd55614eb86a87b509ac9b0f8bae8e1a55355480a5a
thread:issue:55194,issue #55194: Tracing `len(tensor)` gets static constant value in result graph,needs triage,47,needs triage,1,0,55194,,,,0,5dfea5f05906bc4cc9046777493ccd379e82717599b6faeb159c63502ffbfe8c
thread:issue:55211,issue #55211: [TorchScript] Can we support type refinement by Python type annotation?,needs triage,47,needs triage,1,0,55211,,,,0,a1377dd1d09aae7bb10b0556643b66b654055d3d63de427f0503e4dda4fde8ea
thread:issue:55282,issue #55282: Poor error message when trying to jit a function instead of a module (RuntimeError: Cannot insert a Tensor that requires grad as a constant.),needs triage,47,needs triage,1,0,55282,,,,0,8832118a4fc3fa119c3bb85490923dc126633fee0a4219f4706cacde33e87282
thread:issue:55541,issue #55541: `torch.jit.interface` does not understand dunder methods (e.g. `__call__`),needs triage,47,needs triage,1,0,55541,,,,0,6188b38c3112918319be4a13c90dc3d27b6a748ddcc85b459dcc25b78bd72199
thread:issue:55603,issue #55603: Faster Inference time for first .pt model prediction,needs triage,47,needs triage,1,0,55603,,,,0,85ac5c40625fdaface9a87b342fee1f1e1cca623c60249b35924062892218193
thread:issue:55742,issue #55742: [JIT] TorchScript to represent typed tensor annotations in IR graph,needs triage,47,needs triage,1,0,55742,,,,0,c2222e54e9c91d5785da9582fbfadfa0ffa5fbf9bb82ee38ad66f8d7987dfdc9
thread:issue:55749,issue #55749: Fail to introduce torch::jit::Module as a parameter of a cusomized operator.,needs triage,47,needs triage,1,0,55749,,,,0,bd390735406d6ec978b825160c785566f6bf10cce88e034f59bb14454cb9259c
thread:issue:55779,issue #55779: Request setNumThreads API for Java,needs triage,47,needs triage,1,0,55779,,,,0,91f8a36f89dc4505b2f97f350a42a053847347af9f97c89e311030f4ebf9de58
thread:issue:55782,issue #55782: YoloV4 Pytorch model inference fails on NNAPI Android,needs triage,47,needs triage,1,0,55782,,,,0,f5ff61a8eb300ba8b6ff15e1320915606bfd8350d2664f3c554a618f8a6662bf
thread:issue:55999,issue #55999: [JIT][ProfilingExecutor] A reliable and consistent API/protocol to query & propagate profiling data,needs triage,47,needs triage,1,0,55999,,,,0,f35c3bac91a91924deb337372365b209e48881e148477e9a468fe8ca3fac49ae
thread:issue:56006,issue #56006: Unable to get a Vulkan tensor using `to('vulkan')`,needs triage,47,needs triage,1,0,56006,,,,0,8b8b5606e4fc4c7cfcc74a751efd33cf651f5fa3c3ec6343c8e7b64484ea4095
thread:issue:56007,issue #56007: NNAPI support for torch.cat([...]) when dim=-1,needs triage,47,needs triage,1,0,56007,,,,0,623b9b644a838294e144862858371e421d70de13c6174d1e5878070c145457be
thread:issue:56010,issue #56010: [tensorboard] Expected data-type of images in tensorboard SummaryWriter,needs triage,47,needs triage,1,0,56010,,,,0,9c3ecceb130bbd2fbf126c7c1616a89831b428434158e3d9e957cc7c59b7ef19
thread:issue:56024,issue #56024: Support for mkldnn + ddp,needs triage,47,needs triage,1,0,56024,,,,0,23fe420f59f297b74ce843b58b333c3fc0b30c53ef2652cd824b74bb0a14bb94
thread:issue:56096,issue #56096: [JIT] Merge/reconcile `__jit_unused_properties` and `__jit_ignored_attributes__`,needs triage,47,needs triage,1,0,56096,,,,0,f8127f95ecd9443c3d5733aa3624f043ad7a365d1aa68b9ef9d8ea1da1978a1c
thread:issue:56111,issue #56111: ProcessGroupGlooTest and other distributed tests don't capture stdout/stderr in test UI view,needs triage,47,needs triage,1,0,56111,,,,0,01ac6141bac7b66e28fe78ab3c2f0a2987a3fe39249f019d4a57bf0fac73bcaf
thread:issue:56159,issue #56159: Addition/subtraction/multiplication/division of bool variables are not supported in JIT,needs triage,47,needs triage,1,0,56159,,,,0,8b9fbe300764a0575cb597e1727255bf1a4e7aaac04b66fdf39d1e93ef8ddd30
thread:issue:56314,issue #56314: test_variant_consistency_jit fails for torch.tensordot with dtype float32 with error INTERNAL ASSERT FAILED,needs triage,47,needs triage,1,0,56314,,,,0,216d4e4b017d03c86fadbbcc4a0bb8151ce631448f0136b8894393968e82db3a
thread:issue:56571,issue #56571: Python static variable access for TorchScript's custom C++ classes,needs triage,47,needs triage,1,0,56571,,,,0,251a9055a36499fc17f499b0fa486bbc4704deffadd66bb7a9103edec654811f
thread:issue:56595,"issue #56595: TorchScript breaks (""internal assert failed"") on `torch.set_grad_enabled`",needs triage,47,needs triage,1,0,56595,,,,0,dae570a1557a339b5927cba7a7f980b646b0b41a1c0d1abc186cc30540d6d367
thread:issue:56660,issue #56660: consistency jit tests gives error for gradient operator,needs triage,47,needs triage,1,0,56660,,,,0,8be83479782389356dc3821ea9cc535e4af8fade00d198675e67490dcd3defbf
thread:issue:56703,issue #56703: [Pytorch Mobile] Error running build_pytorch_android.sh,needs triage,47,needs triage,1,0,56703,,,,0,a4265376552a34d2b25ea201e71d6df198fc40d273f92ac6d0ab6ab72d2476f1
thread:issue:56969,issue #56969: [JIT / NNC] Incomplete Ops Coverage,needs triage,47,needs triage,1,0,56969,,,,0,571081a1ab61a7041a24c847d9d990b4528f363024a5e148a6484e38babb0bbe
thread:issue:57035,issue #57035: `typing.List` need to be on the same line as the variable declaration?,needs triage,47,needs triage,1,0,57035,,,,0,f501394a2c5fc5ad5f784e35f70ac651e61e351d8e5bf29a44e52075ba3577c0
thread:issue:57163,issue #57163: Segmentation fault when using add_graph from tensorboard in combination with try except in forward pass,needs triage,47,needs triage,1,0,57163,,,,0,c1b7634a57c1a34c21148a0761daf113f5c82dadf75d888b942326c0aa359eb6
thread:issue:57271,issue #57271: RuntimeError after optimize_for_mobile,needs triage,47,needs triage,1,0,57271,,,,0,ae5553e628647a40b1a3e8f74d9cbe8891e9d34a8cd5287f0c6a497d2eef3ef3
thread:issue:57313,issue #57313: [jit] pickling custom classes may lead to invalid pickle files,needs triage,47,needs triage,1,0,57313,,,,0,798d748d49ed7037324209dc9e138affcc16b0efc8bc47ae067fe8bf32c41662
thread:issue:57457,issue #57457: Valgrind Shows Potential Memory Leaks,needs triage,47,needs triage,1,0,57457,,,,0,e05fa0cb69584087de48b08ba6b803559590c2ead70f86df34929a5c4869920d
thread:issue:57559,issue #57559: [JIT] Investigate cloning/copying object before tracing it,needs triage,47,needs triage,1,0,57559,,,,0,cb5225e9f50aff0695c5e1468edb2ae0c9b7826a0473c73df8c2e41decf8afe3
thread:issue:57592,issue #57592: `jit.trace` fails to capture single-element tensor passed into op requiring scalar-type argument,needs triage,47,needs triage,1,0,57592,,,,0,d8164d432eb36ae9b0201500afadb2624f6c5388fb320386a74fdf7bc6cbe7a6
thread:issue:57639,"issue #57639: After tracing, the device of the named buffer is still `cuda` although it was moved to `cpu `",needs triage,47,needs triage,1,0,57639,,,,0,e360ff4485a68e6d7b79177476230fb7eb2ed59c74b201bd5d45f5def5574416
thread:issue:57716,issue #57716: Unknown builtin op: torchvision::nms in LibTorch,needs triage,47,needs triage,1,0,57716,,,,0,6cc474786cff21718f242717bb868d164de30e49fb2466b8ba42818b6b8a711f
thread:issue:57717,issue #57717: TorchScript JIT Compiler Can't Recognize Repeated Branches (Scripting),needs triage,47,needs triage,1,0,57717,,,,0,26ce2282e7f9db6bf4620164496ac17963ad1462b3aeca4b001cbab2edc41d33
thread:issue:57739,issue #57739: object with __torch_functon__ does not work with jit function,needs triage,47,needs triage,1,0,57739,,,,0,8ba2419a7d37e55b7031d085f6c3cf646bc8e0057410e16e685da68c0944df27
thread:issue:57793,"issue #57793: LibTorch inference script model is slower than PyTorch, and the speed is not stable. Why?",needs triage,47,needs triage,1,0,57793,,,,0,3af2a2e90dc8b1b5857529e88fdcbd64fd27592e82b7d0ee0cf2dde61388fd4d
thread:issue:57887,issue #57887: [Mobile] Valid scale ratio with invalid error,needs triage,47,needs triage,1,0,57887,,,,0,c0c7ef72e8d20eba4d8161c0f0ceccc8e7e712a8a768cef425f5673789f1347f
thread:issue:57893,issue #57893: Android API to run model on Vulkan backend,needs triage,47,needs triage,1,0,57893,,,,0,412606ed8cc4779ab69b01aa9de510fee62fd83045f22720614ac9950d39565e
thread:issue:57894,"issue #57894: JIT ""optimizations"" take way too long",needs triage,47,needs triage,1,0,57894,,,,0,4c8bfb097ca37af9eb84f150f8066a8ab7028372e1177cb69bffc7edce08967c
thread:issue:57929,issue #57929: [JIT] provide a pass for alias reduction,needs triage,47,needs triage,1,0,57929,,,,0,cf4bb380941cc4b376d9aae950fcc5cff01d7bdd0210a29013745bcc6a405c74
thread:issue:58109,issue #58109: torch.jit.trace memory leak,needs triage,47,needs triage,1,0,58109,,,,0,64de6d7a9bc767dfb2f989a929df75b9d45498684fbb27496d64b632c42001ed
thread:issue:58298,"issue #58298: [JIT] Can't handle creating new objects in compiled methods: ""Class does not have an __init__ function defined""",needs triage,47,needs triage,1,0,58298,,,,0,b54a0806552b2b6b2ab392d98702ae5402e503b27e0a9134c11731f86a596f13
thread:issue:58456,issue #58456: TSAN issues in torch::jit::ProfilingRecord,needs triage,47,needs triage,1,0,58456,,,,0,12a04467863adba22bac705dad650f9a9b0a3ea90780548edab34ec6e5279fde
thread:issue:58507,issue #58507: OpInfo JIT tests do not work with Tensor kwarg arguments,needs triage,47,needs triage,1,0,58507,,,,0,9b02e207de4514ec55fea8256ff98fe1fb80a9f9bc1fa10b107438cb1da2731e
thread:issue:58516,issue #58516: Unhelpful error message when type annotation is violated.,needs triage,47,needs triage,1,0,58516,,,,0,d14b627b87154199360df2dc541fadead0d0ee11ae3931c5e20b5c7673f3517d
thread:issue:58744,issue #58744: Using pytorch 1.1 model on pytorch 1.8 environment,needs triage,47,needs triage,1,0,58744,,,,0,df50e2aff50fd5daef54f79120e72ade825e608041ea0ea9f0c9ee94dbd6131e
thread:issue:58789,issue #58789: Type annotations for torch.jit.* decorators,needs triage,47,needs triage,1,0,58789,,,,0,fd917a76fe3ff3eaa69c853a3cf99d22b3b896c96ef3e77301f9cb96f766e6bf
thread:issue:58804,issue #58804: A Null pointer dereference bug,needs triage,47,needs triage,1,0,58804,,,,0,4cfdd30b3ae8abec28c3657f036bcc3043f3a3b17d3fc192e64c114768e555ad
thread:issue:58836,issue #58836: Latest nightly produces CUDA torchscript that cannot be run,needs triage,47,needs triage,1,0,58836,,,,0,766ce9e5945f37e0e117bc36dc3dba40003af8bdf9c37fbab8f3917339ee7139
thread:issue:58845,issue #58845: Use jit in flask,needs triage,47,needs triage,1,0,58845,,,,0,23b008ac0b6ec47556b3d2ac4cfb25fb9028fd1f7836a6892b8fc489e9047a15
thread:issue:58857,issue #58857: Call non class methods from `torch::jit::Module` in C++,needs triage,47,needs triage,1,0,58857,,,,0,10efae93db32e8c752a752cb9f9d5b9019d53a938014609198e4c729ceca7396
thread:issue:58907,"issue #58907: JIT fails when calling an exported method inside another one, if kwargs with defaults are not passed",needs triage,47,needs triage,1,0,58907,,,,0,1e25768d0f84361a5f0e3ddce424deab250891786b530c2517c48753ebd1ec17
thread:issue:58982,issue #58982: Can't re-init rpc with a different rank after rpc shutdown,needs triage,47,needs triage,1,0,58982,,,,0,849851682955c2c14c50b4caf58dbf2630b72e2f1126fb5bf1911b636c786d47
thread:issue:59114,issue #59114: [JIT] OpInfo / Autodiff tests dont seem to actually invoke symbolic backwards,needs triage,47,needs triage,1,0,59114,,,,0,b2ddbfa7e72f81151a818d041dda866d22faa97de68d1707e284297e079a81d2
thread:issue:59153,issue #59153: JIT compiling a ParameterList raises an unnecessary warning,needs triage,47,needs triage,1,0,59153,,,,0,28dce136c2b4921833b535bd8266fe488b1ff1951dd8a904c8fec3f497b39ddd
thread:issue:59174,issue #59174: [jit] Error: Results of original model and exported/imported version of model differed,needs triage,47,needs triage,1,0,59174,,,,0,78bfaf1ccd16ae224cc37ce3925fd923658effb3a94b1cef87c4f0bed14b960b
thread:issue:59257,issue #59257: Optimizer step stuck during distributed training,needs triage,47,needs triage,1,0,59257,,,,0,5cbec39a5d2fe5eab9a256f79295f61d7211c5c6c6982885c5e387d814f42861
thread:issue:59441,issue #59441: distributed/test_jit_c10d.py fails with RuntimeError: Address already in use,needs triage,47,needs triage,1,0,59441,,,,0,d3ce106cce3264dff6afca3b00381d07acfaf3ea472a0db75691baa7f0d413c8
thread:issue:59528,issue #59528: AttributeError: module 'torch.jit' has no attribute '_script_if_tracing',needs triage,47,needs triage,1,0,59528,,,,0,338ff184861197315c69f457e469e5c5ccc0e7534e66d85bb1fea74aecc9cf09
thread:issue:59530,issue #59530: Batch size is hardcoded using torch.jit.trace with LSTMCell,needs triage,47,needs triage,1,0,59530,,,,0,70825abc5df38e5f53c44ec1b338fd700d3bd0e86224d15cb01046e7912e7067
thread:issue:59551,"issue #59551: error when trying to call ""torch::jit::load""method, use Metal backend in PyTorch Mobile",needs triage,47,needs triage,1,0,59551,,,,0,deae2d428bee5ba518f311782519dc53c4f8c8ca6cd58568d674abb16ce86c49
thread:issue:59612,issue #59612: Why Pytroch.distributed does not expose NCCL cuda stream?,needs triage,47,needs triage,1,0,59612,,,,0,4308b5b1df93eeb493f84aa429925e1c2f6f28cd1b248d06c7918e61030ec2cf
thread:issue:59652,issue #59652: pin_memory mutates its input but this is not reflected in JIT schema,needs triage,47,needs triage,1,0,59652,,,,0,ae341263d6af51a75a134bc154017a2408bb3c4bf7abadc234767bbecbb24291
thread:issue:59690,issue #59690: Inference ran on new thread leak memory on Android,needs triage,47,needs triage,1,0,59690,,,,0,ba3d62661cfed9fcdf20eb83c6c2a31a643971570f140f54c9126a56e60dde15
thread:issue:59806,issue #59806: Context manager for torch.jit.ignore is not covered by test coverage,needs triage,47,needs triage,1,0,59806,,,,0,6442c6cd034aa7c51af6238d8f166af062043a6ebb7b7bb85815f8590e0b549a
thread:issue:59823,issue #59823: print_regression should handle sharded test properly,needs triage,47,needs triage,1,0,59823,,,,0,0e1194ffdfa875851fce44715a7e105bf230000d71034dc64d7266934cbafe37
thread:issue:59865,issue #59865: Custom build of c++ libtorch for Android program or dynamic so lib does not reduce program or lib size,needs triage,47,needs triage,1,0,59865,,,,0,fa1f72c83b252b523099d89507bf7bb945e0786085fafb10b90668a4e5cd91b8
thread:issue:59905,issue #59905: [JIT] batch operators in training/inference,needs triage,47,needs triage,1,0,59905,,,,0,9b2901546a40fe44f4d6e63270e4359c0ef85c32da8f7618df33241d06cc1f6d
thread:issue:60062,issue #60062: [c10d] Make `broadcast_object_list` accept a device parameter,needs triage,47,needs triage,1,0,60062,,,,0,fea052870e836c267399338bd4105f1e3d9f485dd5554eaa8201b93672ace725
thread:issue:6007,issue #6007: Clean up the extra copies that occur in the execution engine and other places,needs triage,47,needs triage,1,0,6007,,,,0,06a0f2d126c3ad4dbd63040e48c23769f4ac8e4da0db8e0102fc9ac289f7b7e9
thread:issue:60089,issue #60089: torch.package: TypeError: 'NoneType' object is not iterable,needs triage,47,needs triage,1,0,60089,,,,0,ac057cd8707d366125f8a33e2d9260addd8596c1056ba487af3c19eb0795c618
thread:issue:6010,issue #6010: Develop a strategy for writing leak tests for pytorch.,needs triage,47,needs triage,1,0,6010,,,,0,e8ece1808a4bebcad22029efb53b83b00efa893f232685779ae65480e1e8ac74
thread:issue:60143,"issue #60143: torch.jit.save gives error - RuntimeError: Could not export Python function call 'NMSop'. Remove calls to Python functions before export. Did you forget to add @script or @script_method annotation? If this is a nn.ModuleList, add it to __constants__",needs triage,47,needs triage,1,0,60143,,,,0,8da08e32c139d92c1099e9156ebf47beaa413fd8b198cec5d5b76f2ce07e0a7a
thread:issue:60172,issue #60172: [Android] Upgrading to 1.9.0 causes NNAPI model loading to fail,needs triage,47,needs triage,1,0,60172,,,,0,3c01efa157383715c23d2509c69b53f896ccc05b49614a6090679430b9c42ece
thread:issue:60189,issue #60189: [package] Tutorial not accessible,needs triage,47,needs triage,1,0,60189,,,,0,0153ca557f0d96835e9db0ba087d67b1380ac1ea7ba14a6131507fcb60c62f35
thread:issue:60359,issue #60359: Expose the .pt model file path during TorchScript custom class Serialization/Deserialization,needs triage,47,needs triage,1,0,60359,,,,0,8d7df9a592b5afaedc3127f759b4452d8f293ac9c315743e4c1ec9d25c4cc86e
thread:issue:60440,issue #60440: DDP with cuda rpc failed with DistributedOptimizer Adagrad,needs triage,47,needs triage,1,0,60440,,,,0,0226095588426b65f512079780b00588537cf1d3bd427835bd672c68ae844ca2
thread:issue:60518,issue #60518: different gpus to train,needs triage,47,needs triage,1,0,60518,,,,0,2cf58829f03c6379999253b125ccaa675d4695f1df4570a84cfc6f027374f092
thread:issue:60640,issue #60640: Building fails on a node with libibverbs installed with no Infiniband hardware present on RHEL8.,needs triage,47,needs triage,1,0,60640,,,,0,87f7c656ecaf2717d37564c6502db3c8602a93dfc0cd448fc76b3ac2573a7a2d
thread:issue:60732,issue #60732: Potential memory leaks reported by static analysis tool,needs triage,47,needs triage,1,0,60732,,,,0,000e22d43645ee8a1edf3401abadfd11d637510313f4b0dff2bff7f751821cda
thread:issue:60844,issue #60844: DDP fails if you have multiple forward passes and a single backwards pass with `find_unused_parameters=True`,needs triage,47,needs triage,1,0,60844,,,,0,02c274b5dd1d3f0e3a4de0b834393fbf1cdccd45ac02a89caacbabdd43d6ffaa
thread:issue:60847,issue #60847: How to release CPU memory cache in Libtorch JIT ?,needs triage,47,needs triage,1,0,60847,,,,0,40a7029eef2de8182ac6b992280f6cba1ee8892f7456099b7d3121d0648ddb71
thread:issue:60937,issue #60937: [jit] About traced script module using in c++,needs triage,47,needs triage,1,0,60937,,,,0,c70040cbd8f1d3dbcc8cef28787fc2b5213fa58f6b336157bfbcdd545383042e
thread:issue:60971,issue #60971: jit scripting for the parametrizations,needs triage,47,needs triage,1,0,60971,,,,0,980efaa36236c994eb72cad5542c842e0cbe4686801a345393f1cf2771a0891b
thread:issue:61221,issue #61221: Unable to add a scalar to an MKLDNN tensor,needs triage,47,needs triage,1,0,61221,,,,0,054d02e95eaf321b0bd8ed9614183867ab2437579072ade22520e704bed4a3a0
thread:issue:61267,issue #61267: Incorrect trace with MKLDNN (adding scalar),needs triage,47,needs triage,1,0,61267,,,,0,ab001e6621a707706533119f9f1ad690de41b94f6c4f82802768204af3ef8e45
thread:issue:61268,issue #61268: Unknown type name '__torch__.torch.classes.metal.Conv2dOpContext',needs triage,47,needs triage,1,0,61268,,,,0,06be0532134a017d09e06e74dd3014f1db5abe26f418b2236626fa769364b909
thread:issue:61344,issue #61344: Get TensorType's device in python,needs triage,47,needs triage,1,0,61344,,,,0,1cf583ee297a80c03891da6319b35f66d22c915b452ac58de4010ac527fe5f99
thread:issue:61709,issue #61709: [pytorch android] use Vulkan backend crash,needs triage,47,needs triage,1,0,61709,,,,0,b1fd94bd1c1f1342447921e9c5b068575de7827463c321ffac7935d22795a010
thread:issue:61765,issue #61765: How to save tensors on mobile (lite interpreter)?,needs triage,47,needs triage,1,0,61765,,,,0,bd22a99164ada6d05e079f9e3f8923f112de91ed3c632e09bbf362fba917d84d
thread:issue:61951,issue #61951: Vulkan: error -5 when trying to use it with Nvidia card,needs triage,47,needs triage,1,0,61951,,,,0,f70a337437e28592c20ffde22648216e508b4b451efd0cc445513f2555211caa
thread:issue:61966,issue #61966: Bootcamp Task: Add `prim::to_mkldnn` to convert from aten's nchw to mkldnn's nChw8c directly,needs triage,47,needs triage,1,0,61966,,,,0,453e1e6d2d75f693ef105578c2c2ca72fe83bf7d94dd980199de23a6c04275ed
thread:issue:62007,issue #62007: Add Future return from default process group collective communications,needs triage,47,needs triage,1,0,62007,,,,0,e32a696cbbbd7c7068659b1a995d7e587f3dce43bd4438f7c015e554b820eab4
thread:issue:62154,issue #62154: AttributeError: module 'torch.jit' has no attribute '_script_if_tracing',needs triage,47,needs triage,1,0,62154,,,,0,d5851f6791d3b94f7b2b8ac33a754ba570550519a309e9158067ff98909b1e63
thread:issue:62165,"issue #62165: RuntimeError: xnn_status_success == run_status INTERNAL ASSERT FAILED at ""/pytorch/aten/src/ATen/native/xnnpack/Linear.cpp"":158",needs triage,47,needs triage,1,0,62165,,,,0,906ffe156c7d9ec3d00f9515fe26be8132d031ad29ac7ac89ad035209531adca
thread:issue:62221,issue #62221: Torchscript Optimize for Mobile RuntimeError: Can't deepcopy IValue with tag: Enum,needs triage,47,needs triage,1,0,62221,,,,0,a740264aa5b922ec700e1a30df6b62c79adbea24fbbf1f74ff62899970e40983
thread:issue:62283,issue #62283: c10d and discontiguous tensors with mismatch strides,needs triage,47,needs triage,1,0,62283,,,,0,6464ac603684c7bf0074ff51a3aed3ed4bf43d6e4b920adfc37a1067b28a7d40
thread:issue:62300,issue #62300: ProcessGroupGloo CUDA Comm Synchronization Looks Wrong,needs triage,47,needs triage,1,0,62300,,,,0,ff14750f8cd28459702370d6f95b283430a294b282f8099b7d51f3fa7b473c42
thread:issue:62313,"issue #62313: On Triton inference server, the TorchScript .pt file, exported by 1.9.0, serves unsuccessfully",needs triage,47,needs triage,1,0,62313,,,,0,7c8aef2846a41cfe25e26448171f469fa15db697afd4c63b6261424a012b638b
thread:issue:62314,issue #62314: dist.destroy_process_group() don't release master_port when there're at least 2 process groups,needs triage,47,needs triage,1,0,62314,,,,0,b1dd8a5695e77bd889d41a6ea76e337b0ee15a03b6baf232d1ba9ed4130fb17e
thread:issue:62325,issue #62325: [jit]][script] handling of 'void' returns,needs triage,47,needs triage,1,0,62325,,,,0,76aec41cdbe55ad72ef871deea60cd937b0360d9b31a7cba8af23a3a381db6f8
thread:issue:62381,issue #62381: torch.distributed and subprocess do not work together?,needs triage,47,needs triage,1,0,62381,,,,0,b7cf6a7f572bfa41a8cfed39c7d73cade9cd5f63698a3d7bf437bedea73e0d99
thread:issue:62394,issue #62394: crash in libtorch jit,needs triage,47,needs triage,1,0,62394,,,,0,bda652078ce39e82c21e06a68cd2618c1aa01153a720c1c83c8b223443df1710
thread:issue:62403,issue #62403: DDP,needs triage,47,needs triage,1,0,62403,,,,0,1d27917e1217ad9403d68f2acd10c0703085356c7cdf144878f0676d36dac5e0
thread:issue:62474,issue #62474: Distributed: RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation: [torch.cuda.FloatTensor [2048]] is at version 4; expected version 3 instead,needs triage,47,needs triage,1,0,62474,,,,0,e62657ef1b10c905e24a648a3912116a1ac1fdc6af262fc9b9772c165ebdcece
thread:issue:62542,issue #62542: 3 Times memory cost when loading the model to torch.nn.parallel.DistributedDataParallel,needs triage,47,needs triage,1,0,62542,,,,0,8ee3f1a070661787c060930527b6339caf85535881c2f00e63efbc5256180887
thread:issue:62543,issue #62543: FAIL: test_threshold (test_jit_fuser.TestFuser),needs triage,47,needs triage,1,0,62543,,,,0,a5337cc060cd57596a5e498fa49b9fd56b234b9d801c68cd1e6be5fe5ce7c33f
thread:issue:62556,issue #62556: [TorchScript] Inconsistency between Python and PyTorch when raising 0 to a negative power,needs triage,47,needs triage,1,0,62556,,,,0,039c43c2efa02d8bccf9505954aecabdcedb40ad193e961376a8e6f98c031748
thread:issue:62603,issue #62603: Unexpected `is_training` after using `train` in a JIT Module,needs triage,47,needs triage,1,0,62603,,,,0,1bb850d51a4903d3183cc8d444540d564bf9e9d00b36ed6c2d5d49e97e1d9dcb
thread:issue:62676,issue #62676: pytorch_mobile custom build Module.forward null reference.,needs triage,47,needs triage,1,0,62676,,,,0,ab59d9f429236dbc9d25803ee186e06c9ad9fac0901894c0ce061de6b0254772
thread:issue:62775,issue #62775: [TorchScript] Support for registering builtin operators with control flow,needs triage,47,needs triage,1,0,62775,,,,0,fc9052aa6e386093b890db3a29409ea3f29976067020d21a1119fedae061b7dd
thread:issue:62821,issue #62821: Enable fft support for mobile builds,needs triage,47,needs triage,1,0,62821,,,,0,178be5ef92818b6e2f61f86a8a6d56d9f6524d94d73ce2f27e7a01a2d84fa06e
thread:issue:62854,"issue #62854: aten::pixel_shuffle(Tensor self, int upscale_factor) -> (Tensor): Argument self not provided.",needs triage,47,needs triage,1,0,62854,,,,0,00803df2bded862ee028f57e3f05939bcf95d55f3de85e5abd6834dbf6ad353b
thread:issue:62885,issue #62885: Type of first constructor parameter of ElasticDistributedSampler and DistributedSampler,needs triage,47,needs triage,1,0,62885,,,,0,3319482131bd8a6d893c04efb6ea6b7d8bddf43099fff7958067357538bafb8d
thread:issue:62962,issue #62962: [JIT] NVRTC unknown error,needs triage,47,needs triage,1,0,62962,,,,0,bfe15ff6d1ed91a25970d450d4fd121af165f05a36afeaee11b92be49c712233
thread:issue:63124,issue #63124: torch.jit.trace quantized bigbird leads to 0INTERNAL ASSERT FAILED runtime error,needs triage,47,needs triage,1,0,63124,,,,0,f81c43b4cad077ae1f362e9dc55d50c02240ce1f97114acfcc88c5f04f970c3d
thread:issue:63171,issue #63171: [BE] Clean up DDP Single-Program Multi-Device Vestige in reducer.cpp,needs triage,47,needs triage,1,0,63171,,,,0,622407c42f66367f3b341472a35a91e9f2700dd1c104918f2e3200b690fada42
thread:issue:63247,issue #63247: Add a timeout argument to RPC shutdown(),needs triage,47,needs triage,1,0,63247,,,,0,e171af760de8e768a642068491dc2886d984d3e88b3a97dec578b4f19fef12fb
thread:issue:63395,issue #63395: How to efficiently (without looping) get data from tensor predicted by a torchscript in C++?,needs triage,47,needs triage,1,0,63395,,,,0,a8d64fbd7748605514733329ae377498187c93267ff4a1c86e4685c5dc89df1f
thread:issue:63403,issue #63403: Consolidate distributed benchmark folders,needs triage,47,needs triage,1,0,63403,,,,0,ada40578ec0125fd90c19a991fd2bc62b50675a83d5d2c641765c76e707778d8
thread:issue:63405,issue #63405: torch.arange has issue tracing with bool input,needs triage,47,needs triage,1,0,63405,,,,0,1f3fccf68845b7a6dbe0fd5c73051ae0b59bc2786fd3f30d654c7a5230419d85
thread:issue:63445,"issue #63445: getNcclVersion() in NCCLUtils.cpp handles ""2.10.3"" incorrectly",needs triage,47,needs triage,1,0,63445,,,,0,61340dc56ee457624ae46b401de4785952d9625635873ccded32a17ec7c8363c
thread:issue:63455,issue #63455: distributed_error,needs triage,47,needs triage,1,0,63455,,,,0,465c46bdd6e1d362871a09e18957780703a3a34c30d023228aa6818a783acfe0
thread:issue:63476,issue #63476: Error loading Pytorch lite model in Android,needs triage,47,needs triage,1,0,63476,,,,0,fc4b05f2d850e13c09b8da57dbbb22e85e97435ee3744f13a1240b8284cdfc49
thread:issue:63610,issue #63610: Enable some amount of CI tests on lazy_tensor_staging branch,needs triage,47,needs triage,1,0,63610,,,,0,88ac2e17b70bdc100a89f0e75e32c8de36e4430b78c50d77bfe05aa67ce4564e
thread:issue:63648,issue #63648: [JIT] Typing on math.ceil is inaccurate for Scalar input,needs triage,47,needs triage,1,0,63648,,,,0,d2b589352126da65713b127d5c5b983a26cd19967fdd1950de21e2f62fdcf597
thread:issue:63832,issue #63832: [RFC] Add `torch.distributed.run` as a console script in pytorch's setup.py,needs triage,47,needs triage,1,0,63832,,,,0,c4a3ca818c19ff9f41e2c9d766fc11c6d408f323401c58976f9105c1cfb98a01
thread:issue:64005,issue #64005: [LTC] Find a way to convert at::Generator into torch::jit::NamedValue,needs triage,47,needs triage,1,0,64005,,,,0,60a7d8073c335ea7f671d30ff5d1df316c0ca422f8c99b11949867b5b831ff6b
thread:issue:64060,issue #64060: Consolidating NCCL version parsing functions,needs triage,47,needs triage,1,0,64060,,,,0,565c3cfc9a2501d887704480beefa98ef396652d7a61e9fb1fef8093a69343f0
thread:issue:64096,issue #64096: String frontend doesn't support tuple unpacking,needs triage,47,needs triage,1,0,64096,,,,0,e0b9109c959f6e1b118a108693a0e1579bd3b96257deb01f8a37acfaf57bae5e
thread:issue:64119,issue #64119: Cleanup tests in rpc_test.py,needs triage,47,needs triage,1,0,64119,,,,0,b3fa45185960b38c2ca6f826e97140ffbe89a3bc75ffea07b9e34ca838458341
thread:issue:64130,issue #64130: Every process takes a tiny memory on each nvidia GPU card when using DDP.,needs triage,47,needs triage,1,0,64130,,,,0,1b2ec2cb8a127f8d1619dbff020104f31e64e7b0bf4a20ca42e9937019cc52ce
thread:issue:64150,issue #64150: Enhance DDP doc to have a complete example including data loading,needs triage,47,needs triage,1,0,64150,,,,0,f883606c25dea9652bf29ef7058086e98b29afc1f76bbed6f30d279b99df500e
thread:issue:64247,issue #64247: How to optimize jit-script model performance (backend device is gpu),needs triage,47,needs triage,1,0,64247,,,,0,7ef2b035682e5c4df4ede1128aca9df1effd5fa111837fceebadfbf9faeb6210
thread:issue:64258,issue #64258: Incubation of Graphical Representation in TF2 of HParams,needs triage,47,needs triage,1,0,64258,,,,0,0f9fa0ffdcf55936dec8fc2db5d73683f7cd8b03a9019b1f90d1e5ceb984cbb0
thread:issue:64260,issue #64260: Using Cloudpickle for pickling jitted functions,needs triage,47,needs triage,1,0,64260,,,,0,78469f80229efeb01cad042acd7211f5df5783a42998075d4123ce0dd6294ead
thread:issue:64333,issue #64333: When compiling TAG 1.9.0 or master for android，An error occurred and the compilation failed！ABIS_LIST=arm64-v8a,needs triage,47,needs triage,1,0,64333,,,,0,0c7fed34ec3d4b048b72e50e72478e6444199d0911e96fbe5ca17b8b5ae351c2
thread:issue:64334,issue #64334: How to add nan value judgment for variable t0_1 in fused_clamp kernel generated by torch/csrc/jit/tensorexpr/cuda_codegen.cpp.,needs triage,47,needs triage,1,0,64334,,,,0,1cde1cb27c5e1d4b28cd0bb9087b7293d6e82ee788890cf7cde048627a32b81a
thread:issue:64419,issue #64419: Windows Torch.distributed Multi-GPU training with Gloo backend not working,needs triage,47,needs triage,1,0,64419,,,,0,3ad42cf8ac7c721788b0e5afb1cfd282af86770edb4cd7a716c298bcac06cc06
thread:issue:64462,issue #64462: torch.package: re-export docs suggestions don't work + other sharp edges,needs triage,47,needs triage,1,0,64462,,,,0,1149c2ee5bee3636c262407771a1cc2591955c38f39dfd19d38f9ee816ef6a4e
thread:issue:64487,issue #64487: Error in torchScript based on Pytorch tutorial,needs triage,47,needs triage,1,0,64487,,,,0,d6e68d0a848e7afe5a6cfd05ec4b53cd2faeda49ecc1ef4b49bcd76127661786
thread:issue:64488,issue #64488: It' so strange!!!,needs triage,47,needs triage,1,0,64488,,,,0,8b2a5e1b58e61ca63883d4d1ebc3f51b00152639e133d5a5218f66642923ba63
thread:issue:64490,issue #64490: Load torchscript model with custom operators (GNN) on Windows,needs triage,47,needs triage,1,0,64490,,,,0,ed10660027ec0b3de2f4ac3868072086fe4d64c2004c971a5d27310447847069
thread:issue:64628,issue #64628: [jit] cannot compile staticmethod of classes when they have a unscriptable __init__,needs triage,47,needs triage,1,0,64628,,,,0,e3b330d9aa328ab5e544f14a2fae70d094fb3c96ad834f16ffacdca2613cc3ca
thread:issue:6465,issue #6465: [caffe2] ChannelShuffle example,needs triage,47,needs triage,1,0,6465,,,,0,8c2b0cbe7d8cae6f567a70a51868165446599387c97c724fe7e26c09b1b2e011
thread:issue:64775,issue #64775: [JIT] Canonicalize aten::rsub,needs triage,47,needs triage,1,0,64775,,,,0,685fc9f6fa3fe05c631b6506a6bea000f33ea9b9aadf7ee746e3cb7f90a389cb
thread:issue:64795,issue #64795: profiler crashes the program when trying to stop on different threads without context manager,needs triage,47,needs triage,1,0,64795,,,,0,7bf4678bacaa558e4ff3acf02dd4208210a416abfc2e2e22bd2ca9eab80e1e98
thread:issue:64799,issue #64799: Add support to profile all RPC worker threads that are run on a process,needs triage,47,needs triage,1,0,64799,,,,0,0bf51fb902cbcdc32193fd12c18140cc3915b2b547d632b8fdafb599f9518736
thread:issue:64896,issue #64896: pytorch1.9.0 can not use torch.jit.load() to load model which save in pytorch1.3,needs triage,47,needs triage,1,0,64896,,,,0,65d276b3593322c71613dd17e80d0512c7a799800b77cc9de684940c909b0b47
thread:issue:64918,issue #64918: TorchScript error loading module with float literal default argument,needs triage,47,needs triage,1,0,64918,,,,0,472c472c3689ee681b02d319af11048bb07275ca34ef2cc2f1821caa69663e27
thread:issue:64957,issue #64957: `INTERNAL ASSERT FAILED` after `torch.jit.freeze`,needs triage,47,needs triage,1,0,64957,,,,0,50aa8f2b04c9fa32b9b7b509f61a1365adea0e9d4995be562fea59dbf068ccba
thread:issue:64986,issue #64986: RandomSampler / DistributedSampler does not seem really random,needs triage,47,needs triage,1,0,64986,,,,0,b67818cc3bb732ad644c420782db75b0d47b9363fee25f7d02bc691694d8f142
thread:issue:65052,issue #65052: Use hyphens to separate long CLI parameters instead of underscores,needs triage,47,needs triage,1,0,65052,,,,0,ceb1d0519dc919b31130754c2cd47c613eec91e7f9b0de992f25caf4888a063f
thread:issue:65058,issue #65058: Add nnapi serialization for module components in Wav2Vec2Model,needs triage,47,needs triage,1,0,65058,,,,0,37505587e3c9575499af96d0dd0de4afebde0a579b2c133453f20cb5a3fb87a9
thread:issue:65098,issue #65098: Bug: torch.jit.load cannot load from IOBuffer,needs triage,47,needs triage,1,0,65098,,,,0,b6254911258b2b0088112c47ee25cf475bfc6feb1043913afa24799fd5ba5362
thread:issue:65144,issue #65144: jit.trace's returned module returns a tuple instead of a named tuple,needs triage,47,needs triage,1,0,65144,,,,0,e5056bd713db1b47970dec074f5dad8c40fe48609e37306d069097686d100f85
thread:issue:65297,issue #65297: [JIT] TorchScript not works as expected when ModuleDict has modules with different signatures,needs triage,47,needs triage,1,0,65297,,,,0,c8fab588b7a148c0474a3fe5a6e9077e188929569d897abbf9292f39e624cece
thread:issue:65305,issue #65305: Add support for HF NLP models in torch.package/deploy,needs triage,47,needs triage,1,0,65305,,,,0,b1714b4ae39d68a581b6a1a681698c163ef0e87191d28cbb8eb0de198013fcee
thread:issue:65397,issue #65397: Vague linkage when using libcxxabi built torch in torch_glow,needs triage,47,needs triage,1,0,65397,,,,0,b8609a792cf08aacfc7a7a29084cccea997acc5d3156c25741ceb93490b68131
thread:issue:65398,issue #65398: [JIT] JIT does not support non-persistent buffers,needs triage,47,needs triage,1,0,65398,,,,0,3c67c8b4f5f7e785aef55d9e87be5aee2e13a39217d27a692e8fde340498b0a1
thread:issue:65449,issue #65449: Make SummaryWriter.add_image supports PIL image directly,needs triage,47,needs triage,1,0,65449,,,,0,302288b704830d39153179abee130276026b902135f2c1a46407b21d65f43b2b
thread:issue:65490,issue #65490: optimize_for_mobile() assert fail: missing op for prepacked::conv2d_clamp_prepack,needs triage,47,needs triage,1,0,65490,,,,0,66ff0e3901b6a364ff168879e43aef327df37c3bbb058d1ad7cced90243c8d35
thread:issue:65588,issue #65588: Make torch.utils.tensorboard.add_scalar accepts float global_step,needs triage,47,needs triage,1,0,65588,,,,0,492bc3e774dfe9a773de0e03f0cb9fd68ec3e1116fc2419a99ed686ebfdccdca
thread:issue:65607,issue #65607: torch.dtype is int in JIT instead of torch.dtype,needs triage,47,needs triage,1,0,65607,,,,0,ba2b5db2b987028121fb40807180e66288392292acc2167916ee7f0e5396b8f3
thread:issue:65611,"issue #65611: `torch.jit.trace` fails claiming ""forward method already defined""",needs triage,47,needs triage,1,0,65611,,,,0,453670e927611d529189ca6664b2f8c9212aac369a187eef10f2388682dd8ef9
thread:issue:65630,issue #65630: Constants in torch.Jit.script,needs triage,47,needs triage,1,0,65630,,,,0,19be39d02e3b9cb7e64dcc62faa45f2c44cc268d581f4aa90cf685c9dafd0184
thread:issue:65643,issue #65643: [JIT] Start Using and Optimizing on Symbolic Shape Inference in Freezing,needs triage,47,needs triage,1,0,65643,,,,0,e083b0ff13c89862f05d122698d76f36d55dd73d08137c6966282403645bc755
thread:issue:65662,issue #65662: mobile deployment: optimized_for_mobile---wrap_cpp_module error,needs triage,47,needs triage,1,0,65662,,,,0,614e6d0ea19b65b3df4841a3a4091a07c933424546e087dceb36d523ad833530
thread:issue:65734,issue #65734: pytorch mobile: some quant models on android cause memory leak,needs triage,47,needs triage,1,0,65734,,,,0,741e326407966da9867b3193ec532f92ea2e31c041564cca6a0c2cc6e025d9f4
thread:issue:65776,issue #65776: `torch.jit.save` fails on Windows,needs triage,47,needs triage,1,0,65776,,,,0,75e51fd903f51c81fb727927db3deaf6a9eeabc600e39040d65d62425f0747ba
thread:issue:65813,issue #65813: [RFC] Low-level speed optimizations for PowerSGD,needs triage,47,needs triage,1,0,65813,,,,0,77ecca0cd7ed78d62756f91727349a880c8891e7cc2e62ec663ac3cba30503ec
thread:issue:65814,issue #65814: INTERNAL ASSERT FAILED when reusing JIT compiled module,needs triage,47,needs triage,1,0,65814,,,,0,67aee9d0603859c9ef1c60b49759256f9a76baffe33857c62878a0305addee0b
thread:issue:65824,issue #65824: Failing to compile /torch/csrc/jit/frontend/ir_emitter.cpp with pointer is null warning,needs triage,47,needs triage,1,0,65824,,,,0,af6fd8452cc394d62916bb8c0d0b9302e2d677aaf7aa84b14ac6465fa8392d7e
thread:issue:65907,issue #65907: Inplace error of BatchNorm layer in DistributedDataParallel module,needs triage,47,needs triage,1,0,65907,,,,0,b8986d59cbb87a23859d3ffbb79fef55caef454f73eb7af8899460ff8c76bdcf
thread:issue:65915,"issue #65915: I am getting undefined symbol: _ZN5torch3jit17parseSchemaOrNameERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE error. This I am getting when I am trying to ""import torch from nemo.collections import nlp"". I am trying to use pytorch ngc container 21.05. I tried to import torch before the nemo extension. Please suggest how I can resolve this.",needs triage,47,needs triage,1,0,65915,,,,0,60cf80ebfa81aef0ce4691b063e72f536c09503a24f59e19a424b57e4ef9a85a
thread:issue:65918,issue #65918: C++ extensions don't correctly set USE_C10D_FOO preprocessor macros,needs triage,47,needs triage,1,0,65918,,,,0,ed12d01be361bb7d3ff79d668e419436b0064bab32549657a7a82196ecf64009
thread:issue:65983,"issue #65983: traced EmbeddingBag cannot be used with different batch size, functional.embedding_bag is not de",needs triage,47,needs triage,1,0,65983,,,,0,a081e49a9fba6b8f41ee8fb5d66057e0e877f315240cb6cdd1d6a69d19f5a93b
thread:issue:65985,issue #65985: torch.cuda.power_usage (ESG 🌱),needs triage,47,needs triage,1,0,65985,,,,0,3991e786ec0c17e3a51d5a95aded1a0d763b18e3004367835c10d5830161acc0
thread:issue:66017,issue #66017: Torchscript `jit.script` breaks with OSError if dataclass used to pass results during model execution,needs triage,47,needs triage,1,0,66017,,,,0,03c8d8d1282cee6b7114763344b9679cd66ce5354555cf39ab6e007ec57e6ef3
thread:issue:66046,issue #66046: PicklingError when saving a ddp module with torch.save(),needs triage,47,needs triage,1,0,66046,,,,0,76351487fc6ee1e4df6d41713b681c938ffdfb6d7aa7a6b5b1ebdf68149b6fc8
thread:issue:66065,issue #66065: torch.jit.script fails to cast explicit Optional parameter to bool,needs triage,47,needs triage,1,0,66065,,,,0,92a59f5a4f1112b662f5c65b1855662482bd02a336defaf1274a95e13a7f5253
thread:issue:66094,issue #66094: [JIT][Sym Shape Analysis] Extend analysis to extract output shape logic,needs triage,47,needs triage,1,0,66094,,,,0,bfa50ac262c862acb9c71fff0d0ab93246f807bef50b12596003d3b9a1639b2e
thread:issue:66150,issue #66150: JIT: empty_like doesn't support requires_grad,needs triage,47,needs triage,1,0,66150,,,,0,7be1d40aba8be3e635980c3868dc773e31ccbb52f41483ddf6bbc2c84d06f079
thread:issue:66545,issue #66545: Forward method slows down at some point if it is executed repeatedly on Android with NNAPI,needs triage,47,needs triage,1,0,66545,,,,0,4082b85a11b89d67b0ca403713dd82914d9916573a4d221a43ae209b197bb42f
thread:issue:66547,"issue #66547: destroy_process_group() does not clear worker count, leading to ""Timed out initializing process group in store based barrier""",needs triage,47,needs triage,1,0,66547,,,,0,96c6144e68c515ea25b979f0ca66909a7023eafbf11ebb29d5ea03b43cbe4db3
thread:issue:66623,issue #66623: Constant folding in symbolic shape inference fails: expected scalar type Long but found Float,needs triage,47,needs triage,1,0,66623,,,,0,18f446358e1e42468a4900fb50f50bd7b75f1bbb1b1557db3e2c4f6f89abd60d
thread:issue:66624,issue #66624: Separate libtorch and non-libtorch specific files under `torch/csrc/distributed`,needs triage,47,needs triage,1,0,66624,,,,0,59e232f7ef0533d1f3c0117c2534ff2cfe57bdc039d24da47f34f76bde72c0cf
thread:issue:66651,issue #66651: [JIT] Profile optional tensor,needs triage,47,needs triage,1,0,66651,,,,0,4143e6306052e63d90b1e6eb587b8a648a18f6490ea47bb4e55773ad999c9f5c
thread:issue:66768,issue #66768: Error: `copy_to_metal_ is implemented only for float dtype`,needs triage,47,needs triage,1,0,66768,,,,0,b8b7043dbd362f2e7122c22909a4ae2f1bcb6acd83c3100e613018021e155f68
thread:issue:66791,issue #66791: jit tracer：Lost imported pre-model parameters,needs triage,47,needs triage,1,0,66791,,,,0,fc45c9aea1af6027b2fb6acfe5588599d3989f7b0f5d06339379c3f9f8ff2d8a
thread:issue:66816,issue #66816: Jit fuser half precision support,needs triage,47,needs triage,1,0,66816,,,,0,55353747a4b081fff47872ffe40a609847d615f1888f18dc779183c65c454e67
thread:issue:66841,"issue #66841: [Torchscript] Comparison between list and int returns true [1] == (1, ) == 1",needs triage,47,needs triage,1,0,66841,,,,0,77eaa65188777454e18bca99843fb0fa82170288f4223104a0a99bddf4356ea0
thread:issue:66930,issue #66930: JIT vs. eager mismatches for jit.traced `int8` to `int32` casting,needs triage,47,needs triage,1,0,66930,,,,0,0d6f7a052d1304946336388a7271fcc29fd6dd7cc651c24d795c0fb45d4e1146
thread:issue:67099,issue #67099: JIT: inconsistency in LSTM.forward,needs triage,47,needs triage,1,0,67099,,,,0,a4ba99735c96a728fe170b7b88cdd6b43f0ecf76004fd4cbe1b605df5d81afb6
thread:issue:67111,issue #67111: [doc] Can we update TorchScript union support?,needs triage,47,needs triage,1,0,67111,,,,0,bcfa351c925f33395d747803560fa322535b341b99b3af43ef46342eb0b19511
thread:issue:67142,"issue #67142: enumerate(..., start=idx) is not working correctly while scripting",needs triage,47,needs triage,1,0,67142,,,,0,4f11fbe6ae4960c7570ecf0ca4ee75e4da7c6b8cfd81353f5d9b99a06e9de28f
thread:issue:67146,issue #67146: `torch.jit.is_scripting()` not set when scripting a Module,needs triage,47,needs triage,1,0,67146,,,,0,b80727c26cbdf9d798577f43f0c6ec1080fbed4b17e438d8bfb9b8578e2bcf8d
thread:issue:67160,"issue #67160: Include function name in ""This Python function is annotated to be ignored and cannot be run"" error message",needs triage,47,needs triage,1,0,67160,,,,0,24b13033212e2bec40451770fb6b313ab6ffbefe3cd9defe44a5ac1a1d8c9b4b
thread:issue:67241,"issue #67241: init_process_group hangs for multi-node, Pytorch > v1.3.1 and file init_method",needs triage,47,needs triage,1,0,67241,,,,0,515c5caa52b1b037dab69db5f9b3d7bfb52cd75ba77def074a1aca26b6a8802d
thread:issue:67392,issue #67392: Missing doc for torch.jit functions,needs triage,47,needs triage,1,0,67392,,,,0,8e479e333abd7825cbd856294533abe3885f8ab407c6b222e2a8019a82cf3e04
thread:issue:67420,issue #67420: Avoid warnings when jitting pytorch modules,needs triage,47,needs triage,1,0,67420,,,,0,5ff864579e8d3796ac853f7cbff0e82cf295d6945200938d236e4709f55ec36e
thread:issue:67458,issue #67458: OpInfo JIT tests fail for torch.nonzero,needs triage,47,needs triage,1,0,67458,,,,0,246855cc8621247204bd72fe714ccb1cc6cdba98df74a306dc946b3e5d900585
thread:issue:67473,"issue #67473: [Feature] Separate, Queryable Alias Parser",needs triage,47,needs triage,1,0,67473,,,,0,3933778431aae795a7ec06727e76d8ec4f85aac41fb13d40df992b8e86b1fffe
thread:issue:67474,issue #67474: [JIT][Feature] Runtime Verifier of Alias Annotations,needs triage,47,needs triage,1,0,67474,,,,0,c065f11e4921d53110803c6bb031820dc914eec315a28d17e344dfab006c13b4
thread:issue:67542,issue #67542: Segfault from ONNX exporting code where torch.jit.script and torch.no_grad are used together,needs triage,47,needs triage,1,0,67542,,,,0,41a4e65fd7b8b4e1abf51529e550e11f9dec72a2bc8fadb4bc2a6b458b74cb2e
thread:issue:67599,"issue #67599: Mobile builds need to make custom build of libtorch to run, let's just include it to make it easier",needs triage,47,needs triage,1,0,67599,,,,0,f99e271fe6f5c4754a04e724482bede4a4f8526e66e3d80f85fc1b8447416a76
thread:issue:67603,issue #67603: The minimum supported version of pytorch when using toch-mobile？,needs triage,47,needs triage,1,0,67603,,,,0,ee921a58d34b09a5856d26c236090e5a4f42514f9d80d5fd3ff15ac870ea1a06
thread:issue:67659,issue #67659: `test_variant_consistency_jit` failing for `max_unpool2d`,needs triage,47,needs triage,1,0,67659,,,,0,fa2f1524c88408da24d0628ff6fe8465652658c2e594b755a258beb18a1a9a64
thread:issue:67834,issue #67834: Model tracing not working,needs triage,47,needs triage,1,0,67834,,,,0,e7e19a172016b58145a5e8d0191059f1e3262dcb8c9a69913145520cde817d5f
thread:issue:67838,issue #67838: `test_variant_consistency_jit_contiguous_cpu_float32` from `test_ops` fails,needs triage,47,needs triage,1,0,67838,,,,0,972c8162fd047631d819a68689fb5856fd055bb43a908af29caf6bd481c2239c
thread:issue:67889,issue #67889: [libtorch][torch.jit.script][input dynamic shape]Error in loading “. Pt” file on C + +: index out of range,needs triage,47,needs triage,1,0,67889,,,,0,def1d21371cacc4c92f8d6f068509ffcdba868f242acaf0616370d6bc20c9ba2
thread:issue:67972,issue #67972: Distributed.TCPStore,needs triage,47,needs triage,1,0,67972,,,,0,316951c64afd68ea62e381b0d5ae2ed82c33693d6dc34a359878d32cc78aa2e5
thread:issue:67983,issue #67983: torch.jit.export does not working,needs triage,47,needs triage,1,0,67983,,,,0,024d78301faaa863b32ffef6d7883863132f7a68d483316e6f97b229d18d3338
thread:issue:68014,issue #68014: Add TORCHELASTIC_RESTART_COUNT env variable to DDP logging,needs triage,47,needs triage,1,0,68014,,,,0,81fca67d8af2a956b8449c934831ae0f21f1a40738f69751424876ee1a1f7347
thread:issue:68049,issue #68049: Request to revise the Pytorch tutorial.,needs triage,47,needs triage,1,0,68049,,,,0,c49edbd911f421851b18bd6780d20853600c20195974a50b1a30b6459e19223f
thread:issue:68064,issue #68064: In place `or` operator doesn't work in torchscript,needs triage,47,needs triage,1,0,68064,,,,0,d201b750f86c07fa981bce55672b7d7d05ae5a253cd1b80ae1baf87bbb350e53
thread:issue:68065,issue #68065: Incorrect return type annotation for _get_mobile_model_contained_types,needs triage,47,needs triage,1,0,68065,,,,0,9677ee9756a710267a40e4d9b555de923259081764097d94124178d636cc8018
thread:issue:68088,issue #68088: [distributed] Restructure our distributed collective tests,needs triage,47,needs triage,1,0,68088,,,,0,4a15d2c476da1855782491a79ded6a926ab726c2d29e5d5b8da9f7db058c46d6
thread:issue:68256,issue #68256: RuntimeError: CUDA error: initialization error when calling torch.distributed.init_process_group using torch multiprocessing,needs triage,47,needs triage,1,0,68256,,,,0,37eabeaa99795612dfb565f6df54bd65a578a22ff94fce584b4f73144fa9ec17
thread:issue:68297,issue #68297: [DDP] Debug mode should ensure reduction is finished in backward pass,needs triage,47,needs triage,1,0,68297,,,,0,f69c8700f96871d4a522154f7b37e4b221342ef00bc5040064b526feb45d9a89
thread:issue:68330,"issue #68330: isObject()INTERNAL ASSERT FAILED at ""../aten/src/ATen/core/ivalue_inl.h"":115, please report a bug to PyTorch. Expected Object but got Tensor",needs triage,47,needs triage,1,0,68330,,,,0,4aa704aa08f3c983c4fb89f7799e3d0e73ada5c067c30da06f2e3b5963dc3ea3
thread:issue:68536,issue #68536: Exporting a model with `_save_for_lite_interpreter()` fails with Segmentation fault,needs triage,47,needs triage,1,0,68536,,,,0,b99091957004510f372dccf66ea12eb61f6ac8c215957c5cb27100335dd08c64
thread:issue:68568,"issue #68568: About ModuleDict indexing when save model based on ""torch.jit.script()"" method",needs triage,47,needs triage,1,0,68568,,,,0,7e22a95f2bd90227c581512a28d8424df5e110e7000d214bf0cf2216c83a4d9f
thread:issue:68759,issue #68759: `test_variant_consistency_jit` errors for `torch.nn.functional.prelu`,needs triage,47,needs triage,1,0,68759,,,,0,8da73f94671c5e586db9693d3aa323a1f3ce09e8249c0c08fafc846d468251ba
thread:issue:68785,issue #68785: [elastic/distributed] API to retrieve consolidated worker information,needs triage,47,needs triage,1,0,68785,,,,0,c7ed7a55c5f3bc21d0f9f4cd2ed558ba9754cd5b2e1d60390a985a350c442712
thread:issue:68800,issue #68800: [1.10 regression][jit][cuda fuse] cat codegen wrong var name,needs triage,47,needs triage,1,0,68800,,,,0,d787bb46d81941399edb766219ad0df149b506bf31321f4cf30e511bab695ddc
thread:issue:68862,issue #68862: find_unused_parameters of DDP,needs triage,47,needs triage,1,0,68862,,,,0,c5bc12be6acfb2d93983809bf6be1941fecac24d69fa2d05bf99152212750f3d
thread:issue:68893,issue #68893: NCCL Network is unreachable / Connection refused when initializing DDP,needs triage,47,needs triage,1,0,68893,,,,0,2f97fb5f833bafc041d0faab8b33c79ee0f929c51464f24e329f54a61686b63f
thread:issue:68917,"issue #68917: Bugs related to NCCL on RTX 6000, code freezes with no output when using DistributedDataParallel",needs triage,47,needs triage,1,0,68917,,,,0,f1f6ae3b43c0df0af199f399a2c45f2637752a8ec11d1e449d1c46bb53c3dfc6
thread:issue:68921,issue #68921: [Bug] forward function error occurs when the scriptmodule (made in Python) including the hook function is loaded from C++(Libtorch1.8.2),needs triage,47,needs triage,1,0,68921,,,,0,f7cba427442bfe632252294801b55bf75dbb70b88c1ca376f64a845ce8236db2
thread:issue:68927,issue #68927: Thank you for the diagnostic script.,needs triage,47,needs triage,1,0,68927,,,,0,16afb18feddb6914b7734bf60cb2972acab41c228dffac6516bf82f7a29bf161
thread:issue:68940,issue #68940: Breakpoint training in a Tensorboard log,needs triage,47,needs triage,1,0,68940,,,,0,affa7d159ac7b6b406d1fcbe2a473d39cae448d04ad41c9baf8a3750774bbbca
thread:issue:68968,issue #68968: torch.jit.trace with pack_padded_sequence cannot do dynamic batch,needs triage,47,needs triage,1,0,68968,,,,0,286858c92846380693d3da9f74a5af449dff25d7abce20cdf0be14bfbc57f4eb
thread:issue:6898,issue #6898: [Caffe2] cudnn versions compatibility issue.,needs triage,47,needs triage,1,0,6898,,,,0,403d6a4ef685bae8cb92173bf69799ff623aaecaaa5aaa04324bedb1a2c8e7ff
thread:issue:68984,issue #68984: speed_benchmark_torch crashes when trying to run a model using Vulkan backend on Android,needs triage,47,needs triage,1,0,68984,,,,0,4f77e96503e2caf7e18ab564ffbbce32e68c04c0c7e033e26a5c52fa089de57a
thread:issue:68987,issue #68987: Operator is not supported in mobile module.,needs triage,47,needs triage,1,0,68987,,,,0,7b175398584544bebdd1f1610aa80af51ac1594cbdbe1a8804c483fe13677b2a
thread:issue:69006,issue #69006: [JIT] allow `Optional[torch.Generator]` be tracable in Python,needs triage,47,needs triage,1,0,69006,,,,0,9ad09fcef92c0f065ad42110f741ac6050843ad0e05670810dc279582edc34ff
thread:issue:69084,issue #69084: Unicode Normalization with Torchscript,needs triage,47,needs triage,1,0,69084,,,,0,3bff75d012dfbf6b1204e996795fa9001a4953b4a06a28aea6391fa901ad81a5
thread:issue:69250,issue #69250: TorchScript incomptaible with non-persistent buffers,needs triage,47,needs triage,1,0,69250,,,,0,b986297b5a7ed15f0cec1bb3d1f65056bc9ec2e15e64558cf4ca737c0f4a3067
thread:issue:69308,issue #69308: Enable DDP checkpointing tests for Gloo backend,needs triage,47,needs triage,1,0,69308,,,,0,0fa694fd41d386b3eaa7becd9ef78168b8558fefc46d42966de1cf9ef6bc04da
thread:issue:69320,issue #69320: [JIT] Break up JIT Opinfo Tests,needs triage,47,needs triage,1,0,69320,,,,0,593d7c409d644885424ff6a43d02fb1a45c64b7b8b9c856411ef39377fc1c4c4
thread:issue:69347,issue #69347: Memory leak when moving tensors between Metal and CPU on iOS,needs triage,47,needs triage,1,0,69347,,,,0,e03f26447687f78b2f6c7f471fff8bad792651f95e8103882cf58034304d8b1e
thread:issue:69352,issue #69352: I want to know how to read the LMDB file once when using DDP,needs triage,47,needs triage,1,0,69352,,,,0,3e443d3e8f53535af144f4e866fc2985ba34a72c691c7844f259705ebf9bef07
thread:issue:69354,issue #69354: Missing instruction in recipe for Andoid/mobile interpreter,needs triage,47,needs triage,1,0,69354,,,,0,d292eaa97b9dcb43922db058a9b382ac4ca9fce1e7463ff1f17e20df2a75f837
thread:issue:69425,issue #69425: Integrate Libtorch into Unreal Engine 4: _ivalue_INTERNAL ASSERT FAILED,needs triage,47,needs triage,1,0,69425,,,,0,149e3cc7a9e3fcd490e8fc56b3f8a83227910531e54dfd53a4a15660c37a168b
thread:issue:69434,"issue #69434: JIT RuntimeError: 'Union[Tensor, List[float], List[int]]' object is not subscriptable",needs triage,47,needs triage,1,0,69434,,,,0,9887c08a3caeeb23a35ef53af9d7c7c8079be95d936e04959a388d33e0c18c63
thread:issue:69435,"issue #69435: torch.is_tensor(obj) doesn't work with JIT, despite the fact that isinstance(obj, Tensor) already works with JIT",needs triage,47,needs triage,1,0,69435,,,,0,0a7f6c55724a435921d6aa13777f1c796c2a952d0a8199f6ec1230ec031f6fbc
thread:issue:69506,issue #69506: [performance] a profiler util to show a rough break-down of the types of ops used by a model,needs triage,47,needs triage,1,0,69506,,,,0,e13752d339b0addc590df63076a183de2fffd2a9730c13bb1fc9b900e5df1bf8
thread:issue:69522,issue #69522: PyTorch 1.9.1 incorrect result for all-reduce,needs triage,47,needs triage,1,0,69522,,,,0,8466f9b44d0234beb5ac410a987ce2d25d37daad89ada5977886b5468af46c2b
thread:issue:69563,issue #69563: [Distributed Tests] Print stacktraces of all processes when one fails,needs triage,47,needs triage,1,0,69563,,,,0,5bfdaa8bde4da5f2553a08abac844f8c0ed6ba09857728d1b3533facbd35bded
thread:issue:69591,issue #69591: Make DDP + Zero have the same communication volume as regular DDP,needs triage,47,needs triage,1,0,69591,,,,0,cda4a07053fcccdc02913ac604d28249286043fd2bc08f2a450470f603fae9ba
thread:issue:69594,issue #69594: RuntimeError: Operation not supported,needs triage,47,needs triage,1,0,69594,,,,0,a19578a40856f1d7cc0ca621c583dd436dd6e8b5c3f88941cecfce47e05c3ad0
thread:issue:69608,issue #69608: Optimize For Inference - add warning/error upon serialization,needs triage,47,needs triage,1,0,69608,,,,0,89df93ca36d284add39f613ec21c061d0d5633ce279e5f8e07f4b6dd371efe47
thread:issue:69609,issue #69609: [mobile_optimizer] RuntimeError: We don't have an op for metal_prepack::conv2d_prepack,needs triage,47,needs triage,1,0,69609,,,,0,e45b936aa8035517cab165f96540f26b746b951aef3914d28710a6085501d7dc
thread:issue:69616,issue #69616: Multi-GPU training stuck when using grad scaler,needs triage,47,needs triage,1,0,69616,,,,0,28de14b4fcd0de2c855ea199d310ae0b55256d5544109e38cb5725a90a8ff070
thread:issue:69782,issue #69782: Torch Profiler does not count FLOPs for backward pass,needs triage,47,needs triage,1,0,69782,,,,0,87dbead3ee2a56736332b9f882dd937fcda797b1e16f651838c9e55213be4ee8
thread:issue:69814,issue #69814: JIT / TorchScript docs missing any mention of typing.cast,needs triage,47,needs triage,1,0,69814,,,,0,60da939f09450b13598dcd3054a904ff6f687fc465ce378c41a0fdb1d29779f6
thread:issue:69830,issue #69830: Conversion error from pytorch model to libtorch model,needs triage,47,needs triage,1,0,69830,,,,0,1805304c8489136f2d38cd664258d9146672bc5a5b212bab10ea0eae5e751dcf
thread:issue:69931,issue #69931: [JIT] torch.exp roughly 20 times slower in TorchScript vs. PyTorch,needs triage,47,needs triage,1,0,69931,,,,0,a24c524888d2d5a0f6cb48b7e41890932ef8bcc0b8db8d9fd9a8f84fda53fee2
thread:issue:69938,"issue #69938: JIT is overriding variables of type List[int] to type Tuple[int, int] for an unknown reason during function calls",needs triage,47,needs triage,1,0,69938,,,,0,b27ca7044f76b6dfb5312607f307b5240e8b743c6b490d0c435d40098a1bc02b
thread:issue:70037,issue #70037: Question about collect tensor in distributed dataparallel?,needs triage,47,needs triage,1,0,70037,,,,0,bc6d47cdb9a659091c29da9a7982c54d7098b11a5b53a7b623fc582048b041c8
thread:issue:70051,issue #70051: [feature request] Support python decorators in TorchScript,needs triage,47,needs triage,1,0,70051,,,,0,38dfdc67feab30020c87ab38821f6a72abb511ca37e95c11c08061f7bd496878
thread:issue:70054,issue #70054: [c10d] have a way to determine two ranks on the same hosts or not,needs triage,47,needs triage,1,0,70054,,,,0,e92735cf130c87e3cf44fae00ea76138c944a5f7b8cff864e44ac0fbc9bf7e1b
thread:issue:70058,issue #70058: Unknown builtin op: spconv::get_indice_pairs,needs triage,47,needs triage,1,0,70058,,,,0,54ee88898d35abcf14613a1ddb7d6ed2c4a5cc08416cfabc30682686422df993
thread:issue:70099,"issue #70099: Question: what is ""Parameter indices""?",needs triage,47,needs triage,1,0,70099,,,,0,04bb77d2981a61858bacea28ccb68d815ef878086c57bd647c2642ad916b4f0d
thread:issue:70102,issue #70102: torch jit script segm fault,needs triage,47,needs triage,1,0,70102,,,,0,ca8c3cf5b927c459848a52b4a0a280bedbc1fb5ad247cc5d25393fad26288860
thread:issue:70155,issue #70155: RuntimeError: Tensor must be CUDA and dense when calling all_gather_object even though there is no tensor in the object.,needs triage,47,needs triage,1,0,70155,,,,0,e711d6b64afd7fdd798381da601a132c2fa0357b155e98eb3a24db7030e269dc
thread:issue:70159,issue #70159: [Android] Unknown builtin op: aten::reflection_pad3d,needs triage,47,needs triage,1,0,70159,,,,0,67bda0a1eb1de0a8277d2f0c3785360e71115fe59cbd7598a08035e2255e03ba
thread:issue:70223,"issue #70223: jit: Confusing behavior w/ torch.autograd.grad, iterative loop, and printing?",needs triage,47,needs triage,1,0,70223,,,,0,47c1b7b73b6bdeda6e99307d1d8bbfcf1df30ae487fbbcb59632ce79d31cc5ab
thread:issue:70240,issue #70240: JIT: wrong list/tuple length if using Union,needs triage,47,needs triage,1,0,70240,,,,0,72e8a546875ee0bbb0ec38a7f149ad6c09ee6e53ab83a438a189bde65627f995
thread:issue:70267,issue #70267: Inconsistent multi-node latency with NCCL and OpenMPI,needs triage,47,needs triage,1,0,70267,,,,0,77d6c6654ba0f1f5bf8ae41d8bde7d93aa856faa4388c53443ea22a191922b1d
thread:issue:70450,issue #70450: torch.jit.optimized_execution is not mentioned anywhere in the docs,needs triage,47,needs triage,1,0,70450,,,,0,9525d82eb7dc7d6553c85720c0a1aaffca460384ee7fca1e9bc08f38cb3a6e71
thread:issue:70511,issue #70511: Strange behavior of torch.jit.trace when moving parameters across device,needs triage,47,needs triage,1,0,70511,,,,0,40a4b93922f68c947d45b1ddf4906d902baf820244f463875f35d0594418440b
thread:issue:70544,issue #70544: Tensor loses `bool` method during scripting,needs triage,47,needs triage,1,0,70544,,,,0,5ffd54aad9caa2154bb9d9637d80ac426c6d57bc22b28b43e47c642dcf45d66d
thread:issue:70573,issue #70573: JIT magic method which returns class instance fails.,needs triage,47,needs triage,1,0,70573,,,,0,e13f4ea5f4a3ea64266f66557cc49bc3684bf4aa25b82ac9e25ff8d15984286c
thread:issue:70627,issue #70627: Torchscript Compiled functions don't support keyword-only arguments with defaults,needs triage,47,needs triage,1,0,70627,,,,0,17cc4666c327e56f7409d7e35b76e1ea43c51703877b9a76cd3aec796b707ca1
thread:issue:70951,issue #70951: A diagnostics mode to report when constraints aren't being met for optimal performance,needs triage,47,needs triage,1,0,70951,,,,0,cf9876508ef32710d04087bbeba183de34ddd85296dfe650a9b095473c64e497
thread:issue:71151,issue #71151: The implement of `containsTensorType(const TypePtr& t)` in jit pass `PropagateInputShapes` ignore some situation?,needs triage,47,needs triage,1,0,71151,,,,0,0e209a771cc1ec4281bb0a6448f9ae80f3fcf219a92439887be11e13d853314d
thread:issue:71165,issue #71165: traced module fails on second execution,needs triage,47,needs triage,1,0,71165,,,,0,576dd5eb0971b8a82cca3f1754164c0192a3c13a7936bbf0dce3ebe3c2e8acc3
thread:issue:71261,issue #71261: The jit model will fail when calling the torch.autograd.functional.jacobian with multiple inputs and setting the vectorize to true.,needs triage,47,needs triage,1,0,71261,,,,0,ae9a348e1ebe937b986917939f811314e77c20c6ebfb64fb6da5bae604df4429
thread:issue:71277,issue #71277: JIT support for `torch.__version__` & str comparison operations,needs triage,47,needs triage,1,0,71277,,,,0,753a66b29655b882c8303a1a753e77d9c6e121751b45e0502b4964e2eea44db7
thread:issue:71286,"issue #71286: [JIT][tensorexpr] cat, batch_norm opinfo tests failing",needs triage,47,needs triage,1,0,71286,,,,0,ae4d102ddc666b2beeedda9a14868072e80a6d5e73776879563c925d242ad1c4
thread:issue:71288,issue #71288: `memory_format` argument isn't supported in torchscript for tensor.is_contiguous(),needs triage,47,needs triage,1,0,71288,,,,0,0f28a8d8928870b82dff1b2e3f305d1a24c5d1172cd10328f6adedb286248593
thread:issue:71784,issue #71784: [JIT] addmm differs from eager mode,needs triage,47,needs triage,1,0,71784,,,,0,3ab41f90f456ddfbc85c0281797eadd133cb90eb6d513914d3c4d5ca54c34cdc
thread:issue:71849,"issue #71849: addcmul, byte_channels_last fail test_nnc_correctness opinfo tests under UBSAN",needs triage,47,needs triage,1,0,71849,,,,0,c69839727e297e073a9224e387de96631eabec9e2deeeb1c4955b690eca615cc
thread:issue:71912,issue #71912: `torch._sources.normalize_source_lines()` and consequently `torch.jit.script` fail with lambda functions,needs triage,47,needs triage,1,0,71912,,,,0,624c265729153dad94a9e5f97e3fc5a2a8c5a9371e3612dfe08f5b537e0d881b
thread:issue:71929,issue #71929: [JIT] Results are different when saving in TorchScript Format through train/eval mode,needs triage,47,needs triage,1,0,71929,,,,0,d765d83098f29b5cfa5c11c4f927489f67684f5b49c30860ae24b8cba7fb7870
thread:issue:72029,issue #72029: JIT function crashing or failing (depending on profiling),needs triage,47,needs triage,1,0,72029,,,,0,a3f1af47b13758e79694c8c6ede2e1ec160dd9aa937a063b34a4939248676aab
thread:issue:72110,issue #72110: Conversion Error in pytorch mobile with metal,needs triage,47,needs triage,1,0,72110,,,,0,10c05593f065a04679256abcf8e81d660c65af1b7397b7767e5051c669c36ff2
thread:issue:72220,issue #72220: [JIT] Assert that the autodiff implementation of backward() returns the correct number of values,needs triage,47,needs triage,1,0,72220,,,,0,64444287e6d39c8ee9dc74178659a3c8de04ac238813c73947bbe98feee28f20
thread:issue:72252,issue #72252: Mobile: minimize inferencing latency by having better control over available CPU cores,needs triage,47,needs triage,1,0,72252,,,,0,325847aafe1988cb39f5a76089abf7b9649e4db43d8f9b841c3ba5f116103b87
thread:issue:72330,issue #72330: Add and Mul torch tensors on Metal (IOS),needs triage,47,needs triage,1,0,72330,,,,0,b1fcc152e273cb2651b739a9a3753d544806c899b414faa737c56ee0b461ca21
thread:issue:72338,issue #72338: [JIT] Channels last optimization pass,needs triage,47,needs triage,1,0,72338,,,,0,8b84fc36b7511ff2bb6d8ea020d8625554b68839b1e26515979be0cad2d24125
thread:issue:72498,issue #72498: torch.jit.script + torch.split + onnxruntime incompatibility,needs triage,47,needs triage,1,0,72498,,,,0,3cb439f91d381ee1bbe9c4d2333553588dee11ae21b99a53f4079dc98aaaeaa7
thread:issue:72516,issue #72516: test_del (jit.test_builtins.TestBuiltins) fails due to highlight assertions,needs triage,47,needs triage,1,0,72516,,,,0,b7284fc57f948535a447cac54c082244c36d0ecab92a2a21f0b73a0503b85335
thread:issue:72617,issue #72617: TorchScript assertion failure for a `List[...]` inside a `NamedTuple`,needs triage,47,needs triage,1,0,72617,,,,0,1f560138475a93f4cca44d6f57a5038326eb47bf65ee8ad2896817ff67554a46
thread:issue:72659,issue #72659: Add a unit test which uses an extension module + ordered importer,needs triage,47,needs triage,1,0,72659,,,,0,183c0c6b06c350e3e2b576b0ccef196d184a6f7aee00ba3784eccef1f066bae6
thread:issue:72746,issue #72746: CPU execution/dispatch time dominates and slows down small TorchScript GPU models,needs triage,47,needs triage,1,0,72746,,,,0,abe17eb107f9787345bcfa8f4a98dddb940477a0e013426c54c3c0d9dfd1949a
thread:issue:72906,issue #72906: Replace deprecated distutils package in torch/utils/tensorboard/__init__.py,needs triage,47,needs triage,1,0,72906,,,,0,51cbb9ff5f839f631ecf5eab13ca952a983cd81a8bbc26d04982b86a3591bd01
thread:issue:73072,issue #73072: Exporting model to TorchScript fails once I load the saved torchscript file,needs triage,47,needs triage,1,0,73072,,,,0,9105a2d365b59f373c163d7319d6a522a9b01043d494e5e002a0e0ba9642f8b1
thread:issue:73142,issue #73142: Support passing Python code as a string to torch.jit.script,needs triage,47,needs triage,1,0,73142,,,,0,8374b93bcc43fc6313ddac668ba51c06e1da2860c2583a800600293e4610ef7d
thread:issue:73419,issue #73419: test_jit_cuda_fuser fails on non-CUDA node for 1.11.0rc3,needs triage,47,needs triage,1,0,73419,,,,0,49a252de54267664ccbd43ef393f5af7f31b88a09ca2a1a41c78e7c943058833
thread:issue:73571,issue #73571: conv with padding='same' fails on lazy torchscript nvfuser,needs triage,47,needs triage,1,0,73571,,,,0,f501cb36085462048e70878d021a8370eb40e72388f3e747df21770af6632ac0
thread:issue:73608,issue #73608: `torch.jit.load` fails when function parameters use non-ASCII characters,needs triage,47,needs triage,1,0,73608,,,,0,90defe4bcef4a3f0fef684471a8eed2675a194e4961040635d3595672c760300
thread:issue:73972,issue #73972: Forward_AD and Torchscript Functions results in Nones or wrong values.,needs triage,47,needs triage,1,0,73972,,,,0,7ee25b46057d38a40155f0c8215bb151de918b5e5fbfb3cf1fb4aa948ef38174
thread:issue:74030,"issue #74030: [JIT] tuple of float is accepted in List[float] but not in Union[float, List[float]]",needs triage,47,needs triage,1,0,74030,,,,0,f1b9627e12ed35e18fc4c9963f30b5c67275e983c8c7b6b18638c4540433e4a9
thread:issue:74095,issue #74095: torchscript RNN modules cannot move between GPU,needs triage,47,needs triage,1,0,74095,,,,0,0359852d33ff351b92a17860ffe47d48de0171a43ab05e579af52a7ce16b25f4
thread:issue:74221,issue #74221: `package.PackageExporter` does not actually appear to have a `file_structure` method,needs triage,47,needs triage,1,0,74221,,,,0,0e0542850bb7dc8854634297b0eb9b35ed93ad3e41e8bac92ce4abc11f70ab57
thread:issue:74272,issue #74272: [PyTorch] Lightweight dispatch,needs triage,47,needs triage,1,0,74272,,,,0,3a1df6e2036ec01af61986ca0dadce4587ba76cb7c3e4729dda5fbae47fa017d
thread:issue:74589,issue #74589: Test test_ops_jit is taking too much time on ASAN,needs triage,47,needs triage,1,0,74589,,,,0,a7f5b8279ac0db11faf3ee1fb3598e44f1be7d005134f5cf8a3e635f73891c73
thread:issue:74605,issue #74605: torch.profiler.profile does't work well for CPU model when not using torch.profiler.schedule,needs triage,47,needs triage,1,0,74605,,,,0,89cd695ae37faea4bd05ab3916f627e84dc2d0e16bfbc1fe4e032be0dab9eb11
thread:issue:74799,issue #74799: Irrelevant warning during ONNX export of torch.jit.ScriptFunction: Model has no forward function,needs triage,47,needs triage,1,0,74799,,,,0,510f06f4da0f002438b6a8167f1ce83973e338473d15e0e7c12dc85854525e98
thread:issue:74909,issue #74909: torch.package fails to import if dataclass and __future__.annotations present,needs triage,47,needs triage,1,0,74909,,,,0,0cdf96b59c9766001a1408364009c5feb2ce26b67b5e514057693b52fcc644c3
thread:issue:74956,issue #74956: Allow `torch.package` to accept `*.pyc` files,needs triage,47,needs triage,1,0,74956,,,,0,3501a5f175afa78e513ec4de0969743cb3653e43bb08de20384c112b58f8ee42
thread:issue:74964,issue #74964: Add Unit Tests for Torch.Package Subclasses [Waiting on changes to land],needs triage,47,needs triage,1,0,74964,,,,0,ebc721c0dffe53bc91a9eaf6351458fc837553a498b059193a5bcfd73ecd7c53
thread:issue:75005,issue #75005: Cannot create TorchScript file after converting into a quantized model,needs triage,47,needs triage,1,0,75005,,,,0,2b4de0787d742fc4c67fc948a81b4d9ed6f4ec39c13e0d63b2f029c5eeac6ee8
thread:issue:75038,issue #75038: Generate source location information in TS LTC backend,needs triage,47,needs triage,1,0,75038,,,,0,ee9bf08ed01836788b05092cbfcb111b6e4f1100e70d8493fdf5af0d581912aa
thread:issue:75217,issue #75217: Storing LTC tensor shape information in jit::Value,needs triage,47,needs triage,1,0,75217,,,,0,db280bece21f672bfdb6fefe231dd3e57a7eb25bee8eb59caf117a64de4da08c
thread:issue:75280,issue #75280: [JIT] log_extract.py improvements,needs triage,47,needs triage,1,0,75280,,,,0,61dca518d0e03935dd4df1a7a8d7950a0a132398e1e644b98137a8588b122539
thread:issue:75308,issue #75308: Using Adaptive Model for Pytorch Mobile,needs triage,47,needs triage,1,0,75308,,,,0,0561ebf37d55a776597744371087d4572dcf0ceba7f9291d1ef9a25e4d82e555
thread:issue:75309,issue #75309: [jit] failing OpInfo JIT tests for conv1d and complex input,needs triage,47,needs triage,1,0,75309,,,,0,d0986fdc8df655d47e1b066cbb1d270c07b49ac7ba26baad4a5fccb69e0b5c81
thread:issue:75467,issue #75467: torch.package: log.info when loading packages w/ externed modules + documentation,needs triage,47,needs triage,1,0,75467,,,,0,44e1450387f09aadd00dd5f8c5bbb4900d2692bf9f967618b865402e5cd7c6ab
thread:issue:75502,issue #75502: TensorBoard frontend fails to display embeddings when `add_embedding()` writes large `label_img`,needs triage,47,needs triage,1,0,75502,,,,0,34a8f3e450d6328f1fe54d29f1ebe34e9b25808c65b4a36837590a8942b0a60a
thread:issue:75556,issue #75556: [JIT] `torch.jit.ignore` is not working on hooks,needs triage,47,needs triage,1,0,75556,,,,0,63bbd1f46949f5804a414146aca12879f3ac6737c1f858f262468ccad679acc6
thread:issue:75654,issue #75654: jit fails when trying to assign values to model via hook,needs triage,47,needs triage,1,0,75654,,,,0,fb79624037b2deb9b3d6ae4d44f1a44b53bf9d0e1fbbd994d639bd30a48b0d99
thread:issue:75655,issue #75655: `jit(Function)` results in double execution,needs triage,47,needs triage,1,0,75655,,,,0,1d2d3215ccc5952479c61178632c8e548035c16bcf48d592e6bb41044ce16c45
thread:issue:75673,issue #75673: Tensorboard Issue with visualizing the connections of encoder-decoder network,needs triage,47,needs triage,1,0,75673,,,,0,b41c8cbc550b8ce98792e4993e902b2de4caffe082b35f89e5c24dbc88681467
thread:issue:75788,issue #75788: `torch.jit.script` Script functions do return `requires_grad = False` if `torch.no_grad()` has been used,needs triage,47,needs triage,1,0,75788,,,,0,b8364dff58329f10af5d32547c70ab19ac1fe89df1d1ddc6a9b08f5c7415cbbc
thread:issue:75794,issue #75794: Different result with JIT,needs triage,47,needs triage,1,0,75794,,,,0,0cb8627cf22cccee0040f9ba0165c340fea28725ad5bc4083d9e574218b0ea06
thread:issue:75864,"issue #75864: INTERNAL ASSERT FAILED at ""vulkan_rewrite.cpp"":272",needs triage,47,needs triage,1,0,75864,,,,0,849fb96071e53eb5ee76f0847deed7964436a168adfabf2e3804486e17a076e9
thread:issue:75903,issue #75903: torch.jit.script'd function very slow on first invocation on latest nightly,needs triage,47,needs triage,1,0,75903,,,,0,4250cecb6f46fa0bf719321577e1d7b5f411a2f10a59db099e60b0f22c66a27c
thread:issue:75935,issue #75935: torch.jit.trace error when custom autograd function used in the model,needs triage,47,needs triage,1,0,75935,,,,0,3a16f32371463ff4ad0815cbf52b1b89f584eab67979493cca008abac02165a2
thread:issue:75956,issue #75956: [JIT] [Autocast] JIT Autocast Pass operations' list should be extendable and consistent with imperative path,needs triage,47,needs triage,1,0,75956,,,,0,2069b82860efa5cbcc8983cdfd05d996904bd22498002411a452839fb86bcefd
thread:issue:75960,"issue #75960: jit/_trace.py"", line 71, in _unique_state_dict filtered_dict[k] = v.detach() AttributeError: 'torch.dtype' object has no attribute 'detach'",needs triage,47,needs triage,1,0,75960,,,,0,533932ab19cca95eecec6cc9d0e1c3d1d5ac0595d6ce6fd291d5eb756a74432f
thread:issue:76013,issue #76013: test_jit.py TestWarn.test_warn and friends doesn't work under pytest,needs triage,47,needs triage,1,0,76013,,,,0,7114cf6e0b652501849fe34eb67c546140945a8b41eb0ace76878cb5fb3ca8b6
thread:issue:76026,issue #76026: Jit torchscript for prediction is missing 'forward' when using forward hooks,needs triage,47,needs triage,1,0,76026,,,,0,df4468f6fbc283dd1666614eedc0358b324ec15dcc63bf8caa39ef7d73fbdb22
thread:issue:76029,issue #76029: Mobile assets upload could break third party mirrors due to binary data size,needs triage,47,needs triage,1,0,76029,,,,0,72848537a7e200009640dc8bebd80850db37c6b006210e80ee759a42bb7f4b04
thread:issue:76181,issue #76181: Allow `torch.fx` tracing on TorchScript models,needs triage,47,needs triage,1,0,76181,,,,0,4164d322efb32e8af8dd5ff80b722ff663197a274469720df275351c07a4c23b
thread:issue:76295,issue #76295: TorchScript: jit.script fails when using 'tolist' with int32 tensors.,needs triage,47,needs triage,1,0,76295,,,,0,3522eb55dbe0750db2ce5f520d8c1058ddb391d0c21e49d8125e79b95d6761cd
thread:issue:76309,issue #76309: More understandable name column of the table of the profiling result.,needs triage,47,needs triage,1,0,76309,,,,0,18803eb6f933894961fe3084606d7c3a070d0cb5202378f08f3a2d236c786289
thread:issue:76486,issue #76486: [Tracer] RuntimeError: _Map_base::at when tracing fake quantization,needs triage,47,needs triage,1,0,76486,,,,0,11fc14737d1f3399df129fc975bdcb8d2dce5108478c2c5c45171e6440ff5bd0
thread:issue:76503,issue #76503: [JIT] make IRAttributeError extend jit::Error,needs triage,47,needs triage,1,0,76503,,,,0,b3f4daf92eab9bb30d3c922c4053517bb38ddef02e250fa8322a4250553f2c5c
thread:issue:76672,issue #76672: C++ CUDA assign existing memory to forward method.,needs triage,47,needs triage,1,0,76672,,,,0,fc84592e605a8a1033ddb830fd38d63b7af3bfd5300765fef4576f6f6b67bf64
thread:issue:76682,issue #76682: [torch::deploy] Remove manager_ from AquireSession,needs triage,47,needs triage,1,0,76682,,,,0,568c8f02d87b9687a9f87d34d887ef68b3eec575e2c90d289658db52f89eb52d
thread:issue:76683,issue #76683: [torch::deploy] remove reliance on manager_ for unload,needs triage,47,needs triage,1,0,76683,,,,0,41f8f9d7e58d63459e7f50b5c497216b992fdb89d07cbb35d0af6cca34822183
thread:issue:76685,issue #76685: [torch::deploy] move create_movable to interpreter_manager,needs triage,47,needs triage,1,0,76685,,,,0,8c037e83aaf12fa2eed1097672d801aec3bdded2017a9474986e710309c413b7
thread:issue:76686,issue #76686: [torch::deploy] Remove `manager_` from the constructor and deconstructor of `InterpreterSession`,needs triage,47,needs triage,1,0,76686,,,,0,d6811eb1813712b665b899791991213ea59dded8fb32545d6361c6e6046fe72d
thread:issue:76774,issue #76774: Support Positional-only Arguments in JIT,needs triage,47,needs triage,1,0,76774,,,,0,0d07ff928f46a6aa93aada05915a027511b70debb8de39d900182d34bc228867
thread:issue:76807,issue #76807: There is a bug with latest stable torch version and the following Nightly versions related to `optimize_for_mobile`,needs triage,47,needs triage,1,0,76807,,,,0,03350384b944f3b683c185e089ea19526e2f5af238fd8927daf7b3cfa1920e23
thread:issue:76881,issue #76881: RecursionError when running torch.jit.script inside JitTestCase,needs triage,47,needs triage,1,0,76881,,,,0,708c93aecf4e5dc28519fd7aa6e9931a165841962df1eb65cad3e8316f125a4b
thread:issue:77027,issue #77027: Kill use of TensorImpl::ShareExternalPointer in torch/csrc/jit/tensorexpr/external_functions.cpp,needs triage,47,needs triage,1,0,77027,,,,0,6775523c1e818e632b8506193ce7abf17da91cb822641585c4cfaefc0b06b9e5
thread:issue:77167,issue #77167: nn.functional.pad accepts bool values but raises internal assert when converted to JIT,needs triage,47,needs triage,1,0,77167,,,,0,9db3459c1a498d2699960afc14bf5a5532798bb6a0670f08dd4104e03df0ada3
thread:issue:77354,"issue #77354: outputs_[i]->uses().empty()INTERNAL ASSERT FAILED at ""/opt/conda/conda-bld/pytorch_1646755853042/work/torch/csrc/jit/ir/ir.cpp"":1314, please report a bug to PyTorch.",needs triage,47,needs triage,1,0,77354,,,,0,f99c431590ba809b83bcb6b4f12816d0af9ce2f96141076d5712ead406f81967
thread:issue:77837,issue #77837: TorchScript attempts to compile dead branch of torch.jit.is_scripting,needs triage,47,needs triage,1,0,77837,,,,0,611c2b3b72adf1b38fa282e34e92a74d0c139f86554660dd0efc214321f51826
thread:issue:77840,issue #77840: Legacy model format is not supported on mobile,needs triage,47,needs triage,1,0,77840,,,,0,1d7420d34e6463d60019fd5f3b90c0aeca2a905aae8e717fbe95e17f4b67605e
thread:issue:78133,issue #78133: Enhancements to AliasDB to handle in-place operations,needs triage,47,needs triage,1,0,78133,,,,0,b1e5c60d53907deefc07d5bbd7243347aa9409aa00e4463ac39f16baf903b857
thread:issue:78172,issue #78172: Allow specifying pickle module for torch.package,needs triage,47,needs triage,1,0,78172,,,,0,d73aedffe64ff54b3eee52b367ac3065699703ad8fb521d46e628d9e92eb08be
thread:issue:78483,issue #78483: is the issue resolved? windows not pytorch_jni in path,needs triage,47,needs triage,1,0,78483,,,,0,70aa73e95be235536997e2434ca3914d58a785d69c26e4a8705b109fcb65299f
thread:issue:78486,issue #78486: Exception in torch.jit.script doesn't indicate where in the code the problem lies.,needs triage,47,needs triage,1,0,78486,,,,0,795c344c0c41b963b024e5ce1904a7fa7bdce0279c071e33e0a0c22c492fe23b
thread:issue:78559,issue #78559: [JIT] autodiff implementation of rand_like function is outdated,needs triage,47,needs triage,1,0,78559,,,,0,5cfda685492cfabe73a222631c552a0b9130b8d9c4314a95af0fe7d577f75dd4
thread:issue:78805,"issue #78805: torch.jit.script segmentation fault (pytorch debayer module) 1.10, 1.11 and nightly",needs triage,47,needs triage,1,0,78805,,,,0,b605206418c94026515f64c0f0f059c13b129f903bb05e025cf1347fa2e97083
thread:issue:78812,issue #78812: TorchScript inference get intermediate result?,needs triage,47,needs triage,1,0,78812,,,,0,831212ae6d6bcbbde5ba62546d3ffd6ca27f3443ad161941cbe953c5d3f4b4fb
thread:issue:78920,issue #78920: Strange tracing result with torchscript,needs triage,47,needs triage,1,0,78920,,,,0,c8661752baa1d393d95425b85716048bd4a097f58f572aa46d92cb8c01c211e1
thread:issue:79003,issue #79003: Libtorch C++ mobile build linking error,needs triage,47,needs triage,1,0,79003,,,,0,e33e731ff01a1d84025a0bb391fd317719126f6aeeffc2c59a70991954dafa35
thread:issue:79008,issue #79008: torchscript jit trace support custom op without specific csrc and .so,needs triage,47,needs triage,1,0,79008,,,,0,ba99b7b03183e4cf0480563344b502b608793649933f47589b1da427ab1c7b7a
thread:issue:7904,issue #7904: TracedModules don't support parameter sharing between modules,needs triage,47,needs triage,1,0,7904,,,,0,6dc177007b1bedefec866749ff3701bf87a0b70b2d1ec019d1847cad5ced58ac
thread:issue:79275,issue #79275: testSerializationInterop in test/cpp/jit/torch_python_test.cpp has not run in over two years,needs triage,47,needs triage,1,0,79275,,,,0,b7a7905336c65ca96061d64c37c51d1c9f2891b2d56d2fa0e6275618aa5b6c6c
thread:issue:79375,issue #79375: TorchScript bidirectional lnlstm from example doesn't work,needs triage,47,needs triage,1,0,79375,,,,0,9066d363585ec20f267963d38f335ad77135557a4b16498e5de0690f50d68dad
thread:issue:79684,issue #79684: Does Torch JIT Support Trace High-level Custom Op?,needs triage,47,needs triage,1,0,79684,,,,0,eb324e5454a890d883dec364c39b8b6f139ff6c9d2bdfe1d4b5166e8abec2182
thread:issue:79751,issue #79751: [JIT] failures with nested with blocks + loop continuation,needs triage,47,needs triage,1,0,79751,,,,0,75c082e7f0d989575033c51a9894812639fdd817daa2bc36392765b48556c26f
thread:issue:79875,"issue #79875: A/libc: Fatal signal 6 (SIGABRT), code -6 in tid 11742 (objectdetection)",needs triage,47,needs triage,1,0,79875,,,,0,1117ee1eb80600e7b1228299aac8a101cb26f1d62f2a74888a1419efdd2d85a6
thread:issue:80007,issue #80007: when forward use **kwargs，how to construct the example_ Inputs parameter in jit.trace?,needs triage,47,needs triage,1,0,80007,,,,0,f73b60c3d6001ee93db21d06e65de0b61da0cea5fcd6944e582233729004a1c2
thread:issue:80142,issue #80142: CPUProfilingAllocator greedy allocation plan generation failed,needs triage,47,needs triage,1,0,80142,,,,0,df19f28168f492d4f00d9471c29d31ab03995e65625609ab832cbd1e1a82f44c
thread:issue:80321,issue #80321: java.lang.ExceptionInInitializerError at org.pytorch.NativePeer.initHybrid(Native Method),needs triage,47,needs triage,1,0,80321,,,,0,6553ffcff5e41b2f0e1a820b6c0d8133969ce006151753b5b168222c9cc0e6e4
thread:issue:80606,issue #80606: [jit.script] jit.script give uncertain results using torch.half,needs triage,47,needs triage,1,0,80606,,,,0,fd5428c46ff1a41e152778b1a99f64ebc7c437f29b9e0ff4d59ceb1143628aec
thread:issue:80829,issue #80829: Can torchscript dump backward graph?,needs triage,47,needs triage,1,0,80829,,,,0,aa60f0f590507b73a48abd8f42d16c3961b58fd11cfda8e5511f111f557c7294
thread:issue:80861,"issue #80861: jit.freeze throws RuntimeError: stack_out && stack_out->size() == 1 INTERNAL ASSERT FAILED at ""../torch/csrc/jit/passes/frozen_conv_folding.cpp"":281",needs triage,47,needs triage,1,0,80861,,,,0,28e04b91d6fea196a60a7d2845fefd89b3028ba710968b6a61f63cd2c45deacf
thread:issue:80863,issue #80863: SummaryWriter add_embedding issue with label_img,needs triage,47,needs triage,1,0,80863,,,,0,99296398d41c4bc3e055f7ae567f5fccc5bd6d2d5fc4cb1b7cff4ad9eec6d7a0
thread:issue:80939,issue #80939: [jit] script backward wrong gradient,needs triage,47,needs triage,1,0,80939,,,,0,70a5638aa5b6ad99bce33616222f1c5b9f74b0ee790a8d722bc7a07a25c901b1
thread:issue:81100,issue #81100: [jit] Failed to load a saved scripted function,needs triage,47,needs triage,1,0,81100,,,,0,5f1b853e644db685d71f4e5bf88d1f369f16430d32dabc72ef5c51ae4a25f96c
thread:issue:81245,"issue #81245: Libtorch cannot load TrochScript Module correctly, when a network contains conv2d(inchannels=64, outchannels=128, kernelsize=1) .",needs triage,47,needs triage,1,0,81245,,,,0,ca04b73f54daa34b1cc87cec24cd366e5a4ac1b1ef7c8d5129ea83f3d51aee04
thread:issue:81337,issue #81337: JIT trace takes forever on a simple method,needs triage,47,needs triage,1,0,81337,,,,0,5eab3a04395c2cedeba39110881d4f75dde19c4c3898e6736b0aabd1d9f48991
thread:issue:81460,issue #81460: [JIT] Request Constant Propagation to keep fake_quantize_per_tensor_affine and fake_quantize_per_channel_affine on the graph,needs triage,47,needs triage,1,0,81460,,,,0,71dfed26d29b18b4716747ed8a5285cb4360feb46e9619932d66581c7efbb85d
thread:issue:81552,issue #81552: Support Swift Package Manager (SPM) for iOS,needs triage,47,needs triage,1,0,81552,,,,0,f22eb575e59b1b1a3f44961330a81132e28099a51b5e9bbfff79164cceb49e5d
thread:issue:81635,issue #81635: Windows Debug binaries crash on forward: assert fail on IListRefIterator destructor,needs triage,47,needs triage,1,0,81635,,,,0,b9856bbdaadffd01e5e797024da6fda4558e4968464ec96289ee821c81fb4984
thread:issue:81651,issue #81651: optimize_for_mobile has an issue with constant operations at the end of a loop,needs triage,47,needs triage,1,0,81651,,,,0,40b1c8bdced369506cd5c4871ff5e007694e1d03125d76d0c102ec50978c3c96
thread:issue:81855,issue #81855: Enable jit error when using FSDP,needs triage,47,needs triage,1,0,81855,,,,0,b5091ae83da322b2722822e29851b11df782f9b91386add2a97138b100a65144
thread:issue:81912,issue #81912: Unknown builtin op: aten::broadcast_shapes,needs triage,47,needs triage,1,0,81912,,,,0,b5d6db464f99088ac5396481e75ebd708bba15e8227086511cec305a7cbe64ef
thread:issue:82065,issue #82065: optimize_for_mobile vulkan_prepack::conv2d_clamp_prepack,needs triage,47,needs triage,1,0,82065,,,,0,fc0ff4f26ad416117e499ac30359e54c8c93d4e7646b563ef564fdc7d69e9128
thread:issue:82084,issue #82084: Untangle TorchScript prim ops in aten namespace,needs triage,47,needs triage,1,0,82084,,,,0,b9776c4819b8c47615493569bfc202312f5b0c76a1c13f3f960e98384029539e
thread:issue:82099,issue #82099: UserWarning: operator() sees varying value in profiling,needs triage,47,needs triage,1,0,82099,,,,0,04f5047aab451bb31f0ce9289043d1ff336e471c973530b9c7b92b33a10b06f4
thread:issue:82216,issue #82216: iOS TestApp from mobile performance recipes tutorial doesn't build on macOS,needs triage,47,needs triage,1,0,82216,,,,0,a1f0b1b2a54da4bb21197a970cf2ee264cd54372c9f537f1fd6608fdc5a3a40f
thread:issue:82312,"issue #82312: A/libc: Fatal signal 6 (SIGABRT), code -1 (SI_QUEUE) in tid 9792 (Background), pid 9674 (ample.testtorch)",needs triage,47,needs triage,1,0,82312,,,,0,7d2d974e1b0cea60f669a65ffc72094602234e2b16d4b2d95b8b0f85993608db
thread:issue:82324,issue #82324: [JIT] SchemaInfo warning appears out in the wild,needs triage,47,needs triage,1,0,82324,,,,0,85120ea430f0c71fa3f6390b4b8f52492902a37c33a41c58add119c9ffe62b7f
thread:issue:82417,issue #82417: [JIT] _unsafe_view returns alias when size(input) = size argument,needs triage,47,needs triage,1,0,82417,,,,0,a99d148e1db2ce4141e51b518fc9fc59edde95f4ad6ca0aa11483a684435b510
thread:issue:82451,issue #82451: add support for bitwise operations with floating point numbers,needs triage,47,needs triage,1,0,82451,,,,0,0e7c341402989309469f3cba05e7a5d5733c670f031e38f9335a10218f1040ce
thread:issue:82550,issue #82550: `torch.cat` can break `torch.jit.ScriptModule` when in inference mode,needs triage,47,needs triage,1,0,82550,,,,0,8d7a0f51c01c4c9b2c7ff7693487619af81615a0798751eb9933da5abb6c0cc7
thread:issue:82598,issue #82598: Deep copy models with `create_feature_extractor` produces different parameters,needs triage,47,needs triage,1,0,82598,,,,0,a3c54831816080280d68a2f218538375cfb7429d7f795c60e34bf19f0cf66b3d
thread:issue:82634,issue #82634: [Torchscript] some activations backward are not fused when used with linear,needs triage,47,needs triage,1,0,82634,,,,0,d08631d9060718550daa29bdf716855a699fb8f6d20c70c67d5bf8db9dd9dd7e
thread:issue:82635,issue #82635: [Torchscript] torch.min returns wrong gradient when inputs are equal,needs triage,47,needs triage,1,0,82635,,,,0,bcdca3e32d969ec7bf7ff46865dba75f8a88c88ba6e2a4999015927da3069db0
thread:issue:82762,issue #82762: Fill in a bool Tensor not supported in jit,needs triage,47,needs triage,1,0,82762,,,,0,95998ea34343c6d08fe39ba132899568d4d8efb96f33e9e3bab298534d4478eb
thread:issue:82872,issue #82872: Unknown builtin op: torchvision::deform_conv2d,needs triage,47,needs triage,1,0,82872,,,,0,5256f23fb3be0aa5e7d32dc92e4cecf5a3e7f48eb6780db26e8b357f812141d3
thread:issue:82951,issue #82951: torch.profiler's FLOPs measure only counts operations involving '+' and '*' .,needs triage,47,needs triage,1,0,82951,,,,0,fcf66680798ff96fa773a8a1dcdd83649cdf50f2b58e538ab05f150de17c402d
thread:issue:83045,issue #83045: [JIT] Scripting modules fails for modules that contain nested NamedTuples,needs triage,47,needs triage,1,0,83045,,,,0,9bd2298fc67e0e0e374702ec2d7608649eb99b95c4dfd61f32dda036efc6cf72
thread:issue:83143,issue #83143: Bug in building pytorch deploy from source in macos USE_DEPLOY=1,needs triage,47,needs triage,1,0,83143,,,,0,ffc4fce0330a4976e127badaf5b6753fbf81fa44e41c8968adbb382c55c9d8a6
thread:issue:83161,issue #83161: Optimize for mobile metal model,needs triage,47,needs triage,1,0,83161,,,,0,520fb2f2e5bc98e8c4bd2bf725879da9feeed299ed0dc0e3f79c26a42e880946
thread:issue:83250,issue #83250: Using Pytorch and Mapbox in the same project,needs triage,47,needs triage,1,0,83250,,,,0,d58e8d533946ca4214802bea2c19cd16db98a49844dbfd307ff40b383578f494
thread:issue:83349,issue #83349: checkpoint function is not jit compatible,needs triage,47,needs triage,1,0,83349,,,,0,58a2d203ad4496acf6e028d16fe679fe486a54368061dd0c668499480ec75be1
thread:issue:83551,issue #83551: OpenJDK libtorch_cpu.so stack guard warning,needs triage,47,needs triage,1,0,83551,,,,0,f02f63fcdd128eeb1e9ba12647612a950a9d06617d75e01b50801a42c9f472e5
thread:issue:83585,issue #83585: Segfault when profiling with_stack=True on model with jit.optimize_for_inference,needs triage,47,needs triage,1,0,83585,,,,0,e455987ecbd8f9eb2222bfe7b47068c83455db9e038065eafb48a8a4dd907798
thread:issue:83589,issue #83589: Speedup for adding images to tensorboard,needs triage,47,needs triage,1,0,83589,,,,0,3f11c1da4228f26a94021f61523bfd8fbcd457ccf766a3a435bc52e39c764280
thread:issue:83769,issue #83769: [TorchTidy] Check if `set_to_none` would change optimizer semantics.,needs triage,47,needs triage,1,0,83769,,,,0,9d1013026ce561d7283006cfc2cc7c6df249269f80e5e98d05004390d80b6ba1
thread:issue:84049,issue #84049: TorchScript unsupport tuple unpacking as function inputs.,needs triage,47,needs triage,1,0,84049,,,,0,bca1fad3d5f7b9fbf4e087162822ce2daafd4d2eb62f340b934256c18110e375
thread:issue:84055,issue #84055: scripted fasterRCNN model cannot be loaded with libtorch c++ API,needs triage,47,needs triage,1,0,84055,,,,0,c85da58778a2c963e1004ac2a7357401bcbd159ce8788c4d496bbd637c5d611f
thread:issue:84135,issue #84135: [jit] ignored method calling static method results in an error,needs triage,47,needs triage,1,0,84135,,,,0,8fc9282e4da9bcf79a22d37d97fd35f0989ef276aeb79872e7eba8773a96e06f
thread:issue:84187,issue #84187: [jit] WithInsertPoint can't get back to the prev_ node if the prev_ node has been destroyed,needs triage,47,needs triage,1,0,84187,,,,0,872d2850bddbedacc26ba98f545eec7db4815987915e398dcf014db44aa2d17c
thread:issue:84189,"issue #84189: RuntimeError: outputs_[i]->uses().empty() INTERNAL ASSERT FAILED at /pytorch/torch/csrc/jit/ir.cpp:1027, please report a bug to PyTorch. (eraseOutput at /pytorch/torch/csrc/jit/ir.cpp:1027)",needs triage,47,needs triage,1,0,84189,,,,0,d6be7fca7a0841c3c3ab0f50ed9e916239c4357b22acaf7455a3356a1c8d8a27
thread:issue:84304,issue #84304: op for aten::bitwise_and during torch.jit.trace,needs triage,47,needs triage,1,0,84304,,,,0,a360b2f3831c92ecafd98fad6bca6d5ab8ef52d74ebf3b04daaaa0ccf35b66ca
thread:issue:84318,issue #84318: [Profiler] Generic Tensor summary,needs triage,47,needs triage,1,0,84318,,,,0,69abef25e1caea07d022717d93c6c45c6e3e55dbcb153c1a697cbea73767ea26
thread:issue:84340,issue #84340: Floordiv is deprecated.,needs triage,47,needs triage,1,0,84340,,,,0,667485b3ccd0fd6deca04bf32805347244cd9b9b9fffb3231cd4601ba29b2863
thread:issue:84495,issue #84495: functionalize: Does not compose cleanly with torch.jit.script/torch.jit.trace,needs triage,47,needs triage,1,0,84495,,,,0,4a0fc9117b6b2a2238f5cc1b10c1d61f1f4e4801a8552d57c749e916a0717a17
thread:issue:8453,"issue #8453: [JIT] Generalize checkTrace() to also compare gradients of parameters, not just inputs",needs triage,47,needs triage,1,0,8453,,,,0,13a8145075534f1e741f2986056fab5446a7973b10751c1f5292c4df6d184341
thread:issue:84588,issue #84588: torch.jit.script IndentationError: unexpected indent,needs triage,47,needs triage,1,0,84588,,,,0,018504cf354a6c624b29ad170e3ae5bcfb536d1f8f599d99e1aff4089205866e
thread:issue:84615,issue #84615: Serialize the warmed up torchscript module,needs triage,47,needs triage,1,0,84615,,,,0,25f7d32be80742c562b8a022b5c6d51f66be27750195a6594a8705841ea08f97
thread:issue:84628,issue #84628: INTERNAL ASSERT when the type of argument is not considered in JIT,needs triage,47,needs triage,1,0,84628,,,,0,37f0a5b2b57afa548e3f0e998ab404db55011f898b36c84b5a986a8f1ea59e02
thread:issue:84646,issue #84646: JIT script calculation/dtype inconsistent depending on operator expression,needs triage,47,needs triage,1,0,84646,,,,0,63ff9e3c74f02673ce477ac2c15ffceefa9c60ef5f464bc3b2cdb4a2693dcadb
thread:issue:84673,issue #84673: Autograd will take `init` module API into account when using `jit`,needs triage,47,needs triage,1,0,84673,,,,0,f4e2351d1b5eab9b0462a83147157c224b8e11c82be948497279bf46d083b666
thread:issue:84681,issue #84681: JIT will affect the gradient computation of forward mode,needs triage,47,needs triage,1,0,84681,,,,0,ad5e1a7dd04ca91c5a7e241af02f8d2e09f5bd811b8841145a5000dc253c4501
thread:issue:84685,issue #84685: [TensorExpr] applying `rfactor` for a Mul Reducer with init value different than 1 results in wrong results,needs triage,47,needs triage,1,0,84685,,,,0,8707f92b6156a39ef8a692d3e01a85e6323cc5e3795dc1a80eeafd078f8229ed
thread:issue:8473,issue #8473: RuntimeError: /pytorch/torch/csrc/jit/tracer.h:117: getTracingState: Assertion `var_state == state` failed.,needs triage,47,needs triage,1,0,8473,,,,0,f6fd073abfd51f7dd821cfa40c3f01839ba9b7bdaad4b498524f4388ef3b10f9
thread:issue:84831,issue #84831: INTERNAL ASSERT FAILED for _jit_pass_vulkan_optimize_for_mobile (Google Colab),needs triage,47,needs triage,1,0,84831,,,,0,89079c40fbbd01c82c324bcad7324101ea53cdd281014381639a8e1754474d27
thread:issue:85072,issue #85072: Segmentation fault in `torch.jit.wait`,needs triage,47,needs triage,1,0,85072,,,,0,29600b6cd94688b8b6d3a4345a3e4362308c7faf9465f96abed50621cf781ed2
thread:issue:85149,issue #85149: Pytorch on iOS (iPhone X & XR) throwing can't allocate memory exception. Ref Logs:,needs triage,47,needs triage,1,0,85149,,,,0,d77ad5a8151b7b960baadeacc0db70deb3378994e3c550c3c1fc6f97223dee6d
thread:issue:85342,"issue #85342: JIT fuser issues with {ceil,floor,round,trunc}(int8)",needs triage,47,needs triage,1,0,85342,,,,0,09787a82cf5321fa886e6e82db34ecbd6131ee74d0c05f00d617bae9b4e66a8e
thread:issue:85505,issue #85505: The reload model has different (and strange) forward computation from original model with `LSTMCell`,needs triage,47,needs triage,1,0,85505,,,,0,07a44e9c0f749fe0f650f292075d44d03acda07aa13ba93a50e6864ba2c8be7e
thread:issue:85513,"issue #85513: Measure impact of JIT decompositions, reconsider the design",needs triage,47,needs triage,1,0,85513,,,,0,a89d113e0a4073050858c4a5b7e3e19b7069e69eb6b480bbaaadd59b8a32e194
thread:issue:85514,issue #85514: The reload `MultiLabelMarginLoss` will have different gradients on cuda,needs triage,47,needs triage,1,0,85514,,,,0,1d95f3fb386726f2ea20518ad4acfbb4be6546569704ae151ef1c1e8eab64e97
thread:issue:8556,issue #8556: [JIT] Interleaved C++-Python execution loses inner Python stacks,needs triage,47,needs triage,1,0,8556,,,,0,3981e5dd8764768df562508ad7f67b1c938540c97e0672d691978111ffd43e3e
thread:issue:85568,issue #85568: `detach_` behaves differently when computing the gradients in forward mode w/ `jit`,needs triage,47,needs triage,1,0,85568,,,,0,0f786095d234fbc1c01ea054c578546709edba980116e46bc9b0c89f15a8dcb9
thread:issue:85570,issue #85570: `mvlgamma_` will fail when compiling with trace `jit`,needs triage,47,needs triage,1,0,85570,,,,0,b563582d4f8d2b94b0d4ae3405fd3a7a0e5fcb8c6626c36f406d175f2651091c
thread:issue:85573,issue #85573: `jit` could make some undifferentiable APIs differentiable,needs triage,47,needs triage,1,0,85573,,,,0,9df0f29966502593ea936e1c5361204b8c18c95ea32404c45fc1e400086c0edc
thread:issue:85585,issue #85585: `as_tensor` will return a different dtype with script,needs triage,47,needs triage,1,0,85585,,,,0,b5baafa52f310daf0e58b5e81063307912974d9852c89b6c9a51e98731f8d9fe
thread:issue:85588,issue #85588: `linalg.norm` cannot compute the grad in forward mode after script,needs triage,47,needs triage,1,0,85588,,,,0,7337c928d4a0e9582926981a03f415cc58a74dfdd3353957b71a5a3d7109806f
thread:issue:85706,issue #85706: RuntimeError: [enforce fail at CPUAllocator.cpp:68] . DefaultCPUAllocator: can't allocate memory: you tried to allocate 4860032 bytes. Error code 12 (Cannot allocate memory),needs triage,47,needs triage,1,0,85706,,,,0,e7c02f7ab987224f356d02fccb8732e7fd9d56e72f7d05143affbe336794dfb8
thread:issue:85804,issue #85804: JIT fails to trace `sparse.mm` with a strange error,needs triage,47,needs triage,1,0,85804,,,,0,63c4fca15f4c68dcbc12092163b952586a14efa6cc7b2871b94b9365b7d81955
thread:issue:85805,"issue #85805: `vector_norm` will trigger ""Tracing failed sanity checks"" for JIT when ord is boolean tensor",needs triage,47,needs triage,1,0,85805,,,,0,735cba9830659f1f48d2a736d90dec6bcb508c4c690d51aac55f190a45b6dccd
thread:issue:85813,issue #85813: TorchScript error for `Enum` inside a module,needs triage,47,needs triage,1,0,85813,,,,0,f4cbb0fb18d4782aabb2d862b870b8fa89fc13bb674a1466f883c1b5c23b8d66
thread:issue:85893,issue #85893: TorchScript does not recognize mix-in types with `Enum`,needs triage,47,needs triage,1,0,85893,,,,0,ab54a8a225c8605a99e90b56f10f4f4ab7108f47bedc24569193b8c46f032043
thread:issue:86110,issue #86110: JIT return a tensor with different datatype from the tensor w/o gradient and normal function,needs triage,47,needs triage,1,0,86110,,,,0,c6848ee60209c458418015a1b75ac02c39944c5e74c7102ea4c96288d2b1a180
thread:issue:86112,issue #86112: `cdist` should succeed when `p` is integer in JIT,needs triage,47,needs triage,1,0,86112,,,,0,02e91b05a69bdb15bfb1edb856fe5de399e0dcc51b2a643e144b127e2792020f
thread:issue:86116,issue #86116: JIT fails to trace binary cross entropy with a strange error msg,needs triage,47,needs triage,1,0,86116,,,,0,356bcfb2dadf10fdcaf29f4bdb54acdeb8113d02fd9d8a1ab6a410d9e2acdf99
thread:issue:86124,issue #86124: torch.jit.trace throwing Invalid name for qualified name eror,needs triage,47,needs triage,1,0,86124,,,,0,7389ae7218715ea9d7938f664927a3ce44fb74a60963c78dae8b938e98df3f7b
thread:issue:86276,issue #86276: JIT miss the argument `as_tuple` for API `nonzero`,needs triage,47,needs triage,1,0,86276,,,,0,3709b5550b2408ce26168ca4cc4812c6cb79473e68df13af0d6fa7ba98e29fe4
thread:issue:86287,issue #86287: JIT `lgamma` will return `inf` only with dual input in forward mode,needs triage,47,needs triage,1,0,86287,,,,0,f8e9b1165fb21ef64cfabdaffb05cd2f096ddd6ed0f7cae45e0dcbe08e7484bf
thread:issue:86351,issue #86351: Jetson JIT: Memory Leak on inference after optimize_for_inference,needs triage,47,needs triage,1,0,86351,,,,0,99a2354602b1f50ce85b9f79b0e208fed93d0bc0895a05e9bbeb234004f5df75
thread:issue:86455,issue #86455: (JIT) x:Optional[T] cannot not expect content type after `if x is None or x.shape[0]==1`,needs triage,47,needs triage,1,0,86455,,,,0,dc304193feb91dd4749fe8bb0667fb9927b8f402549ee83575e3370649ff9acc
thread:issue:86552,issue #86552: `max_unpool` and `max_pool` will trigger INTERNAL ASSERT FAIL in JIT,needs triage,47,needs triage,1,0,86552,,,,0,a0266ae1e6902f194aaea20d28ee422a2905fbf04cd73a02d3324101df1bdd8a
thread:issue:86553,issue #86553: `topk` will return the wrong value and could read out-of-bound value after jit,needs triage,47,needs triage,1,0,86553,,,,0,e4306828868aa4a1618c30cf12f08ae7df9976a8dcd1c59516e7c3f66e209978
thread:issue:86572,issue #86572: Instructions for Selective Build for Mobile Linux Platform,needs triage,47,needs triage,1,0,86572,,,,0,ad74a9858ec180e60f9ca00bd9f4da71c03b2b88ae4f8e2ac534464998a5ae6a
thread:issue:86597,issue #86597: JIT returns different values for `cos + frac` on cpu,needs triage,47,needs triage,1,0,86597,,,,0,963597ee4aab2fdf72eacadc9ab6320c5e0b622a9cde46c46343aa03754535d6
thread:issue:86598,issue #86598: `squeeze_` fails with JIT but succeeds without it,needs triage,47,needs triage,1,0,86598,,,,0,fe9a63159604a956224e718b467a717d452d51360c9dfa0afb3f7aa70acf17c5
thread:issue:86616,issue #86616: JIT returns different values for a model on cuda and returns a strange error message on cpu,needs triage,47,needs triage,1,0,86616,,,,0,c10d6320bfd45eaeca08f7ae00bd4e553a9719b6b5781b8ed5d516360e909671
thread:issue:86662,issue #86662: libtorch throws `required keyword attribute 'profiled_view_size' has the wrong type` on Linux,needs triage,47,needs triage,1,0,86662,,,,0,ef5e697a8b15296edc824e0e1ab35d2b3b799a0c12b8f0a9b410189d4347649b
thread:issue:86667,issue #86667: Adding a linear layer leads to failure of `optimize_for_mobile`,needs triage,47,needs triage,1,0,86667,,,,0,3fc4a56557f74159108244392478e01a6d62898072cf44c42a00f9bee8e56300
thread:issue:86816,issue #86816: JIT model returns different value on cpu with uniform-initialized input,needs triage,47,needs triage,1,0,86816,,,,0,9ad1dd6b8090169f7a25caade837a082ec06014956c48bdf4b21d28ff8045acb
thread:issue:86818,issue #86818: Forward hooks for ScriptModules,needs triage,47,needs triage,1,0,86818,,,,0,9715fddd1dbf812a77c8adfef093f8133d1bc766f4da80f03a934b583b5bb91a
thread:issue:87145,issue #87145: Unrecognized data format when using release libtorch libraries in debug build,needs triage,47,needs triage,1,0,87145,,,,0,cee454e86c5e7aac3cb5cb239286ad032429287afba95fcc36a6967b7e3ec4d8
thread:issue:87268,issue #87268: register_package has no further documentation,needs triage,47,needs triage,1,0,87268,,,,0,7691013cce86b12715c1faee95e119fa561606ffe6f6c0b64f47ef70cdb8206e
thread:issue:87289,issue #87289: 'str' object has no attribute '__module__' in jit is_final,needs triage,47,needs triage,1,0,87289,,,,0,6ee85aa37e8dae81982836c0093e07ad71a3b3efabedf16e99643acb32062dc7
thread:issue:87390,issue #87390: `chunk` a 0-dim tensor will crash in JIT script w/o profiling executor,needs triage,47,needs triage,1,0,87390,,,,0,fe138e094bfb3b875d9609f9cf58cfea6acfef32d589bd8211885741fe1b2da1
thread:issue:87487,issue #87487: [JIT] Inconsistent handling of tracing dict output leads to assertion,needs triage,47,needs triage,1,0,87487,,,,0,84f88ae085ac4dd4e828cbb66b02e260fc7e3adc5a4f7ef40f620f504ae77902
thread:issue:87781,issue #87781: `torchtyping` annotations make saving to Torchscript fail,needs triage,47,needs triage,1,0,87781,,,,0,22d3c0457d32f42c9518595a2cbc42628a7c1ceb328703cc05010551c46fceef
thread:issue:87787,issue #87787: `BatchNorm` a 0-shape tensor will crash in JIT trace w/o profiling executor on cuda,needs triage,47,needs triage,1,0,87787,,,,0,2a6f5ead918ab53e84768edf9ae5e5bc2d1ea193c76c6f31bb2ff150f1b1cd2f
thread:issue:87792,issue #87792: `AvgPool` and `MaxPool` will crash in JIT w/o profiling executor,needs triage,47,needs triage,1,0,87792,,,,0,d29dee88d94db1a5759e7880fc65462e814e769679f5d2f1a07b8d0f2ba11c55
thread:issue:8818,issue #8818: MPI causing job to hang --- unresponsive to external (termination) signals,needs triage,47,needs triage,1,0,8818,,,,0,cdb230fac50b73758601d9ecabab6ccc9aac673f1ec1947165538c0a0e17515c
thread:issue:88301,issue #88301: 1.12.1 incompatible with c++ built for 1.12.0 and vice versa,needs triage,47,needs triage,1,0,88301,,,,0,23656152caece303142e688815593b949264ebcb28125353b7248673a60216f4
thread:issue:88491,issue #88491: torch.set_grad_enabled results in RuntimeError with torch.jit.script,needs triage,47,needs triage,1,0,88491,,,,0,e5250840ffcfa9655d97497b532184e289b93554774932cd84ceed5293368ead
thread:issue:88574,issue #88574: Add support for `torch.Generator` in the FX IR,needs triage,47,needs triage,1,0,88574,,,,0,a99687284b34536cf4a8188a2d7a4453c7a8e5fefedb1be5ea9e1cf3c45790a6
thread:issue:88617,"issue #88617: The libtorch tests Simplify.{SimplifySymbolicMinMax,SimplifyNestedMax,SimplifyNestedMin} fail on Apple Silicon",needs triage,47,needs triage,1,0,88617,,,,0,bd5ad6d2b8991fcdf5c5e5924d991a739c362206b338f245ce40adfd6b5e0460
thread:issue:88735,issue #88735: RuntimeError: method '__torch__.___torch_mangle_0.MyModule.sin' already defined.,needs triage,47,needs triage,1,0,88735,,,,0,5c2dc795a4e3719ca90c4f99e9113ad734e1df00ce3d4d683b66af8a2f7a875f
thread:issue:88765,issue #88765: torch.jit.trace() - AttributeError: 'NoneType' object has no attribute '__module__,needs triage,47,needs triage,1,0,88765,,,,0,11a7f4ce6af9addb0cd7991be1dfaf648fb9045b98543c1f366408702391c95c
thread:issue:89064,issue #89064: UserWarning: The TorchScript type system doesn't support instance-level annotations on empty non-base types in `__init__`.,needs triage,47,needs triage,1,0,89064,,,,0,47c15eedde8ec59e98721829c6d5d4c4357d3a843bdf854a7ff7de2dcbbc8ecf
thread:issue:89065,issue #89065: Improve clarity of meaning of `torch.jit.trace`'s `example_inputs`,needs triage,47,needs triage,1,0,89065,,,,0,476bc23ce3048152c28bddd8c9f09b00d01bd3cbdc0ffa172a967a658cc2cac5
thread:issue:89076,issue #89076: Transformers model tracing not working,needs triage,47,needs triage,1,0,89076,,,,0,bf4b2db262a6e6df0328fe233c4c949eaa3afaaefce6e5c41ba5e26e941057dd
thread:issue:89114,issue #89114: jit.script() fails to resolve/cast Optional[Tensor] fields of sub-modules or base classes of the object being scripted,needs triage,47,needs triage,1,0,89114,,,,0,fe080aeea4331f458b20e8474b96cba38b1a226a7e1ef20648f0d310817135a8
thread:issue:89185,issue #89185: [feature request] Add ability to preserve traced shape during torch.jit.save and torch.jit.load,needs triage,47,needs triage,1,0,89185,,,,0,5e144e04e696b72b70a317050daa8d19cd223d2a7fbe4edd644297a69d1975b9
thread:issue:89386,issue #89386: How can i patch the torch.jit in the second solution? Could not figure out entrypoint ?,needs triage,47,needs triage,1,0,89386,,,,0,97e0cf74c8fc768fe4456740ce0a7e2594afb7d196208bceb4ccb0ad60536d6f
thread:issue:89483,issue #89483: Unexpected behavior from torchscript (mixing trace with script),needs triage,47,needs triage,1,0,89483,,,,0,ce72a0543fb668cf987df90655a1969170e17a193b52e745d764472a3542efcd
thread:issue:89484,"issue #89484: kind_.is_prim() INTERNAL ASSERT FAILED at ""../torch/csrc/jit/ir/ir.cpp"":1098",needs triage,47,needs triage,1,0,89484,,,,0,681c0e45ced8f333d1a96faf21732d53e8c689f8ac93b343f34db6b94468103f
thread:issue:89868,issue #89868: No pytorch_jni.dll file in libtorch 1.13.0 lib folder,needs triage,47,needs triage,1,0,89868,,,,0,c3544f84ac357d9cc7e3c1804b677bcbc2db2c431836234c4db39acc97496546
thread:issue:90181,issue #90181: Graph is renamed in torch.jit,needs triage,47,needs triage,1,0,90181,,,,0,b284570204f865b21ed2bc5ecf538f90cd50867c1d8d9b3314fedb3495370a91
thread:issue:90301,issue #90301: Unknown buildin op: aten::pad,needs triage,47,needs triage,1,0,90301,,,,0,40919a50d7f194daa146483b7b75b29aac49a5ef5ef76c660fb4be55fa913c19
thread:issue:90305,issue #90305: Inconsistent Hash of IValue between aten/src/ATen/core/ivalue.cpp and aten/src/ATen/core/Dict_inl.h,needs triage,47,needs triage,1,0,90305,,,,0,6f6aa377c1674106ee1fdd0e3d714087e783bf411efb258901343a36d17929bb
thread:issue:90369,issue #90369: [JIT] Wrong type inference leads to misleading error message,needs triage,47,needs triage,1,0,90369,,,,0,783d0327edab3d504065ce53c8d3a386ed32394a6b50db4d6a4413dfec4c97d0
thread:issue:90374,issue #90374: torch.utils.tensorboard import fails if a new protobuf > 3.20 is installed (bug in tensorboard/tensorflow but better guard against it),needs triage,47,needs triage,1,0,90374,,,,0,bec3435365cf04b359b91fbf397ea285393d71e6d1df396389cfcd93e67248e4
thread:issue:90466,issue #90466: Saving a scripted module to a buffer does not work.,needs triage,47,needs triage,1,0,90466,,,,0,22edf4b1121d281d16190701e276a5874c610371242b2f0f343be7753bc71bc8
thread:issue:90885,issue #90885: TorchScript with complex abs doesn't work in backward,needs triage,47,needs triage,1,0,90885,,,,0,d872fc785c4265146adf447db9a25812a6803e4140a4ba3058697bd5587b0949
thread:issue:90915,issue #90915: Torch SummaryWriter import fails with torch 2.0 with an error on numpy.object,needs triage,47,needs triage,1,0,90915,,,,0,34a5577abfe38e23dcc63e0d0838afd06e29aec032012e86ac83392e2e11df56
thread:issue:91000,issue #91000: node.stack_trace does not handle escaping correctly,needs triage,47,needs triage,1,0,91000,,,,0,b86133c3c2bbbbe0f55279c959e9ae5f161ae775fd1bda016885f2c345f7f127
thread:issue:91098,issue #91098: [TorchScript] Failed to Forward Correct Number of Arguments to Different Functions,needs triage,47,needs triage,1,0,91098,,,,0,dcdb6b2bf7366123f48d9a20899fe48c31055f2969954cf97e912bfa765a93dc
thread:issue:91229,issue #91229: JIT mishandles torch.round() in PyTorch 1.10.1,needs triage,47,needs triage,1,0,91229,,,,0,8850735ea4bfd1330a9554300385cf0a1340c4e8bfd2a7af3b0169a5e93a39cb
thread:issue:91311,issue #91311: Profiler is not properly recording seq number when any key above autograd is used,needs triage,47,needs triage,1,0,91311,,,,0,cbc374bb7ebf24aeaf0fca268c2a422ab1921b23dad5d0c376dd7e4828ac10c1
thread:issue:91358,issue #91358: [JIT] .backward() not supported by JIT trace,needs triage,47,needs triage,1,0,91358,,,,0,7ad05448c70b41f81b61d6564bb9ab96410b3e1180d51011a2b41f2a6c644e2f
thread:issue:91360,issue #91360: Offer `get_buffer` and `get_submodule` in `ScriptModule`?,needs triage,47,needs triage,1,0,91360,,,,0,276ac21047956592e940a8d97a701fdada8fb4f5fb9f254ac274628028c92440
thread:issue:91387,issue #91387: getting issue 'typeindex' file not found in Littorch-Lite/install/include/ATen/core/custom_class.h,needs triage,47,needs triage,1,0,91387,,,,0,924099d44e419370d4da242b32dfd1b9294f09c7a95f7b1609fba1f0bf9fe50a
thread:issue:91396,issue #91396: [JIT] Zero-channel conv2d cannot be applied with `optimize_for_inference`,needs triage,47,needs triage,1,0,91396,,,,0,921fa26680c67e444cc2275308c7ab732b480306c8cd7a5d587ff39345388e4e
thread:issue:91504,issue #91504: torch.jit.script ERR: RuntimeError: Can't redefine method: forward on class: __torch__.SoSadModule,needs triage,47,needs triage,1,0,91504,,,,0,085f348fc355dfc921bbbe813281f0550746aa0c04bfbff8fff50cb2d9a76c98
thread:issue:91562,issue #91562: `torch::jit::optimize_for_inference` doesn't preserve exported methods when calling `freeze`,needs triage,47,needs triage,1,0,91562,,,,0,2345d832500aafde7d25377fb496fd4b35f52251073d2c81e0de5cf38fd28a83
thread:issue:91686,issue #91686: Sparse tensor not supported (Minkowski Engine),needs triage,47,needs triage,1,0,91686,,,,0,8e8f33eb9521103677f5787de431edc63905e01c8be49fb9b7022f249e842183
thread:issue:91738,issue #91738: tracing torchvision detection model results in an error,needs triage,47,needs triage,1,0,91738,,,,0,369b85921c9ea037891a31272f6e32ff004dfebdb135bb0f647223f56779379d
thread:issue:91903,issue #91903: Profiling with stack enabled results in error when Python's cProfile is also running,needs triage,47,needs triage,1,0,91903,,,,0,3c49437cc7ce577b5c8965e91faadebf45f4343a421964551cacb3649711bb50
thread:issue:92041,issue #92041: jit testing fails on 3.11 debug build,needs triage,47,needs triage,1,0,92041,,,,0,f634aef28934bdfba997238d0aa076ca2fe770b9fd158481d158768e0a0a2567
thread:issue:92175,issue #92175: test_fx_passes generate bad test names,needs triage,47,needs triage,1,0,92175,,,,0,ae99922c8c8da2d19bd2369079a8efce0ec68832432b9a2b78a5dab427e0443e
thread:issue:92230,issue #92230: Add torch::jit::ScriptModule to the C++ API,needs triage,47,needs triage,1,0,92230,,,,0,b5cd4ac8ba84c5738dd3ed5e4279124ca6cbc8788fda36aa0480ebca89050396
thread:issue:92246,issue #92246: test_qnnpack_add fails,needs triage,47,needs triage,1,0,92246,,,,0,e516479ff750268acce8c807b774ac438b1c79d7fbcb45679f350a366dc2bf20
thread:issue:92294,issue #92294: jit.fork stalls multiprocessing dataloader,needs triage,47,needs triage,1,0,92294,,,,0,b67c1527a8fc4ee930844fa3e5ee06fe5a1d06726d280d6307f87e1bd552348b
thread:issue:92375,issue #92375: operations failed in TorchScript interpreter,needs triage,47,needs triage,1,0,92375,,,,0,554c3ac76c51f19ae573dbeb19229185af9e8ae00806617638bb608a9a2ebc39
thread:issue:92548,issue #92548: [JIT] Inconsistency in tensor shape between eager mode and JIT,needs triage,47,needs triage,1,0,92548,,,,0,579f49da62f5dabea15d4054d3d7e8d63badfab7053bf69e6d5a88c70a91faa2
thread:issue:92558,"issue #92558: [JIT][TracingCheckError] inplace ops incompatible with `contiguous(.., channels_last)`",needs triage,47,needs triage,1,0,92558,,,,0,0105551a3f08b4ca6b31c2ae9b46a1db934a7a85a4f63baa882c27a0eb0f7a40
thread:issue:92563,issue #92563: [JIT] INTERNAL ASSERT FAILED when `Conv2d` and `clamp` used together,needs triage,47,needs triage,1,0,92563,,,,0,5226b40404986f0b547a5f71ac710e068267255863122659837e4df384862aa1
thread:issue:92654,issue #92654: Traced model output differs on C++ and Python,needs triage,47,needs triage,1,0,92654,,,,0,dedd92ab8f58dee3b3bab544ed533694bad82a67cd828ad16c748e59e2c4556b
thread:issue:92674,issue #92674: [JIT] `Linear` + `BatchNorm2d` Trigger Inconsistency between Eager Mode and JIT,needs triage,47,needs triage,1,0,92674,,,,0,98d0d0534f84ee8111695c293d88e12def673ae50e70da94413fa0987029b9f3
thread:issue:92740,issue #92740: [JIT] Applying `conv2d` over Constants Leads to Exception,needs triage,47,needs triage,1,0,92740,,,,0,54d2c413c020dc9242ae51c1926207743c648a484f3f9375887d347fa18da523
thread:issue:92742,issue #92742: [JIT] Consecutive use of `addmm` Leads to Exception,needs triage,47,needs triage,1,0,92742,,,,0,12908e71e0f1cf5a4e750985e9b1290a660f3e3286ecf91ca37514da5f92c847
thread:issue:92812,"issue #92812: torch.fx fails to trace through ""+"" op between torch.Size and torch.fx.proxy.Proxy",needs triage,47,needs triage,1,0,92812,,,,0,f2b0dfb040b70c93c0d3e4fc4547dd079b7d5f4a3833ff5297c3418d46da02b0
thread:issue:92855,issue #92855: Profiler documentation doesn't mention some exports are mutually exclusive,needs triage,47,needs triage,1,0,92855,,,,0,7b995b408fa098c1703fe2439451425e2f9055fb8103bfa7c334fe383f924f8b
thread:issue:92866,issue #92866: JIT Function Fails when run a second time,needs triage,47,needs triage,1,0,92866,,,,0,812706cfbf4f1f1f3353570b0344938b3437cd19fc6a62b28f97e4bfa0d83375
thread:issue:92888,issue #92888: [LibTorch] pickle_save output cannot be reloaded using pickle_load in Windows,needs triage,47,needs triage,1,0,92888,,,,0,6dbd2877612a27389da20a4fa2eed98c8434499b0ddae8583b892c6631da94e0
thread:issue:93002,issue #93002: Replace pattern fails on incompatible function arguments,needs triage,47,needs triage,1,0,93002,,,,0,0928aa3dc2636a01898f94a320a48612e45e3d901cfb7608f088a3d015a3ba68
thread:issue:93134,issue #93134: Failed to Open libnvrtc-builtins.so.11.7,needs triage,47,needs triage,1,0,93134,,,,0,e9b3481f0d077182a679f880fc6fc53e2ef9a41dba8050c851a03627616a3fa5
thread:issue:93255,issue #93255: MaskRCNN model loaded fail with torch::jit::load(model_path) (C++ API),needs triage,47,needs triage,1,0,93255,,,,0,9c5fc3122478a6d7c6218a2c3a9bbd5713fe6ad4987dedbbb6361bd356b0d9bd
thread:issue:93350,issue #93350: Tensorboard SummaryWriter with cloud storage does not work on Mac,needs triage,47,needs triage,1,0,93350,,,,0,f7494423df017e28fec7073c687399998b4d4f71502b3730a7daaabedff7f2c0
thread:issue:93826,issue #93826: torch.jit.script does not work with DataParallel,needs triage,47,needs triage,1,0,93826,,,,0,377f3029e25a2938723bfc95ec119a497355887568eec73df4ee1c75e774e9ec
thread:issue:93913,issue #93913: [BUG] jit.trace not working for torchvision ViT models,needs triage,47,needs triage,1,0,93913,,,,0,dc440c834a90ab0b04e45ddc6d7eecd2c109befffef70b138c1bfa6f77ba0fd4
thread:issue:93943,"issue #93943: `torch.jit.trace` memory usage increase although forward is constant, and gets much slower than forward with model depth increase",needs triage,47,needs triage,1,0,93943,,,,0,b75afd053e243762c813db6369823ef08705a5e7301be7a8640b7d1acecff6d0
thread:issue:94908,issue #94908: ASSERT(initialized()) Debug Error after JIT fusion on Windows,needs triage,47,needs triage,1,0,94908,,,,0,62ecd3c19a56d4c3cb8909ff97a7ebfa97131c1c315f2c277b1dde28afb79fe5
thread:issue:94974,issue #94974: Slow inference of torchscript model in C++,needs triage,47,needs triage,1,0,94974,,,,0,6d6d502c92b57422e3ea73b6de5bc00d4836d0e03196d71b1dfc4a95cae5a57e
thread:issue:95194,issue #95194: High Cuda Memory Consumption for Simple ResNet50 Inference,needs triage,47,needs triage,1,0,95194,,,,0,0a8d69bed4f07d0b039ef0954833b7b0e2b137ee15a50b15b98040fa2d326f1e
thread:issue:95497,issue #95497: Import parameters from jit,needs triage,47,needs triage,1,0,95497,,,,0,4ce23f2864eda1e99154ad91e50e1be00d7cfe57c289396760874a669e8c2f76
thread:issue:95560,issue #95560: torch.jit.load documentation doesn't specify if it is safe to load untrusted models or not,needs triage,47,needs triage,1,0,95560,,,,0,c02d054d6812a583fd1e385fa1f712eabc505aba7d927a86404a82da62ac27b6
thread:issue:95789,issue #95789: drastic speed regression of torch.jit.load starting with the 20230301 nightly,needs triage,47,needs triage,1,0,95789,,,,0,6eae96adf38cd7083f58caa3360ea1b80cc923af83cdf1c4775d1256c6f95566
thread:issue:95858,issue #95858: [JIT] Support string type annotations in NamedTuples,needs triage,47,needs triage,1,0,95858,,,,0,65b869892c359352e8d286d0f9b1c19b57ba3a33f0b73510744bec262e65beac
thread:issue:9595,issue #9595: [JIT] Unify python -> SugaredValue construction + fix inlining of graphs with Tuple-typed inputs,needs triage,47,needs triage,1,0,9595,,,,0,b4da37bb732d6469175e978bd6e0c1bf31476acfbf597589f0e14de8c4f16688
thread:issue:95960,issue #95960: `@torch.jit.unused` does not properly ignore unsupported function signature,needs triage,47,needs triage,1,0,95960,,,,0,c386c706eba19eca8d50877acd0b653665a16525d48e43a52cb1ec27c4884114
thread:issue:96098,issue #96098: Error during inference on iOS: INTERNAL ASSERT FAILED at it_type_base.h:535,needs triage,47,needs triage,1,0,96098,,,,0,8b9599592a18800e736e8e8a676a96ab6b79180f6db4b429ab49ee7199e2c1f0
thread:issue:96185,issue #96185: `ld: error: unknown argument '-force_load'` when linking libtorch on Android,needs triage,47,needs triage,1,0,96185,,,,0,d4211e60c4caab053a1232ca735bc2905d1c20ad801f87f1708ceaad3a986b13
thread:issue:96420,issue #96420: Add location information when exception are thrown in `torch.jit.annotations.try_ann_to_type`,needs triage,47,needs triage,1,0,96420,,,,0,a15a8ab49e626083f19153119474f0a6e6c24bd923390652a695c2369a79026d
thread:issue:96448,issue #96448: Unable to move torch.jit.load-ed models to XLA devices,needs triage,47,needs triage,1,0,96448,,,,0,fa13af51716f7c06522278474fc9bf6477bc738e4319b30eb4a00cd49341954e
thread:issue:96639,issue #96639: Completely different output between .pt and .ptl,needs triage,47,needs triage,1,0,96639,,,,0,b6a24502d394beceb4413ce528ec8e26a9ca85d8ce7e844ce259288b1961db6a
thread:issue:96735,issue #96735: when run python run_test.py -i test_ops_jit error like this. ValueError: option names {'--junit-xml-reruns'} already added,needs triage,47,needs triage,1,0,96735,,,,0,a27257a014519a0cfd381ba72a1597831cc43afb4e13d91b18736f1c03948508
thread:issue:97047,issue #97047: Optimize for mobile produces incorrect result with INSERT_FOLD_PREPACK_OPS optimization,needs triage,47,needs triage,1,0,97047,,,,0,22b94c45eb995107b9ab449f1245ca14a76053ebb7603499c75cef6d4aa560a1
thread:issue:97343,issue #97343: Traced module shows non-deterministic behaviour on CUDA,needs triage,47,needs triage,1,0,97343,,,,0,bf75c256048f7b3814a19d2d15dcff898ac878c9dbcbee7d605fd46d1078a981
thread:issue:97913,issue #97913: BUG torch.jit.annotate on List + torch.stack give wrong DTYPE,needs triage,47,needs triage,1,0,97913,,,,0,52c98a3d9b588d98e4af24b2ab2001492dc5e2f0d73674e5192d99337f80d082
thread:issue:9797,issue #9797: Build torch as a submodule with static linking doesn't work (CAFFE2_PERF_WITH_AVX2 is not defined),needs triage,47,needs triage,1,0,9797,,,,0,2336904663db63dc7f67d27053d0625aa47fc38f04fdbd4881e248c1d99a39ba
thread:issue:98064,issue #98064: Module 'Sequential' has no attribute '_modules' :,needs triage,47,needs triage,1,0,98064,,,,0,a2e7bf9374d58211e196e0252f4e597c4ba2a0d0dd66e2dde94ee2e4cce0459a
thread:issue:98124,issue #98124: PyTorch Profiler fails recording functions,needs triage,47,needs triage,1,0,98124,,,,0,9977401b4c6f6e5a448987569b313e68f84aa9db84b1d74675708761fd95ef62
thread:issue:98210,issue #98210: torch.jit.script + legacy executor mode has diff in some pattern,needs triage,47,needs triage,1,0,98210,,,,0,5245061f3a78246e1dcbbd0fbe1caa7aa1faecef4529370d5f48ccfe4935dec8
thread:issue:98422,issue #98422: [FX] Symbolic trace over `torch.Tensor.${fn}` APIs,needs triage,47,needs triage,1,0,98422,,,,0,2d5c74632ebf2879345ec163920bc71b99450bb6cc5172328e61be4493abaf6d
thread:issue:98695,issue #98695: Torchscript: Name Mangling prevents Type Refinement,needs triage,47,needs triage,1,0,98695,,,,0,1b5406c669326c49266916cf5660cb7abc6e8abd266924b5f4c31c58577d8132
thread:issue:98888,issue #98888: Changes to TorchScript autodiff changing default behavior are no longer accepted,needs triage,47,needs triage,1,0,98888,,,,0,4703bad29d8be31b9bd11ca79a017ccc00ecea7365d7c88b332c3dfad71f5694
thread:issue:99821,issue #99821: Tracer cannot infer type of Seq2SeqLMOutput,needs triage,47,needs triage,1,0,99821,,,,0,a0f14ff8a6cccc9221f850db9c55d48ebabc5e65d0ca37739af8670e4e9bacac
thread:issue:129553:issue:141550:issue:147513:issue:171659,pr #181109: tools/stats: add script to generate test-times.json from local JUnit XML reports,ready for maintainer decision,45,Review active PRs and linked issues as one maintainer work thread.,7,0,"129553, 141550, 147513, 171659, 180975","181109, 188999",,,14,3ce451b9ff54587e2bf27d982e1fcbd4098ed287c97f4f1a91e681b3328f55d5
thread:issue:185770:issue:187293:pr:185790,pr #185790: Fix torch.nn.functional.gelu returning NaN for infinite inputs,ready for maintainer decision,45,Review active PRs and linked issues as one maintainer work thread.,3,0,"185770, 187293",185790,,,2,f78dfb6e88a5a4a17098a1405389cae1100ea2c094a42311a764635578b58e28
thread:issue:188602:issue:189078:pr:186245,pr #186245: [1/N] Optimize PyTorch CUDA build with LLVM BOLT,ready for maintainer decision,45,Review active PRs and linked issues as one maintainer work thread.,3,0,"188602, 189078",186245,,,4,86663f9f2bd877e9dc61d3a96005f8a9f802e806376892a45187dc34c6f9fd04
thread:issue:167885:pr:180020,pr #180020: Add a debug tool to warn on NULL stream use,ready for maintainer decision,45,Review active PRs and linked issues as one maintainer work thread.,2,0,167885,180020,,,1,75af680ce566a27ef5bfe85542e3914374b9b4c4209bef01da8aab4b63ffae79
thread:issue:172987:pr:176317,pr #176317: [MPS] Leveraging unified memory memcpy instead of blit encoding for tensor device transfers,ready for maintainer decision,45,Review active PRs and linked issues as one maintainer work thread.,2,0,172987,176317,,,1,6a2cb1e207b1ab361337a36f0fb8c63aa08b44995af1bd1205116c75ff4b2045
thread:issue:181093:pr:185486,pr #185486: [OpenReg] Add minimal implementation for inductor integration in openreg,ready for maintainer decision,45,Review active PRs and linked issues as one maintainer work thread.,2,0,181093,185486,,,1,f3fae74aba0f68823b60a10993c058e66ba02681ae1fa992ad3b724177f463e0
thread:issue:188319:pr:188951,pr #188951: Add ROCm Docker release images,ready for maintainer decision,45,Review active PRs and linked issues as one maintainer work thread.,2,0,188319,188951,,,2,3e52e8c8e1c7f5a483d6a1c7fe179439a3e3dfef993e6c60957474fa26c1f55e
thread:issue:188652:pr:188263,pr #188263: (fix) NNPACK PeachPy air gapped build,ready for maintainer decision,45,Review active PRs and linked issues as one maintainer work thread.,2,0,188652,188263,,,1,1d57ccd5f9d5895b5417d57e6c99dc93977f0e69f5ec9ba9ff5ab3c8b542fa89
thread:issue:188672:pr:188681,"pr #188681: [Inductor] Use CUDA toolkit's libdevice by default, fallback to Triton's bundled if not found",ready for maintainer decision,45,Review active PRs and linked issues as one maintainer work thread.,2,0,188672,188681,,,1,a3790835253ed03ff1f40866941e85d7decd1a58ea97a9cd04d09bdc5dd3eb32
thread:issue:188727:pr:188500,pr #188500: Allow pre-split pipeline microbatches,ready for maintainer decision,45,Review active PRs and linked issues as one maintainer work thread.,2,0,188727,188500,,,1,33ba5222335e669ee6678444d38672a5f7b84bf37c97b37e71d51db5fef9ba91
thread:issue:189161:pr:189162,pr #189162: [inductor] Use non-NaN-propagating fmax in softmax persistent reduction,ready for maintainer decision,45,Review active PRs and linked issues as one maintainer work thread.,2,0,189161,189162,,,1,8abacc23a1b5b8d46fad3c56c7ae90b2971d149dee8975d855c2f853d1db2e33
thread:pr:182987:pr:183028,pr #182987: [Inductor] Parallel batched precompile in combo sequential autotune,ready for maintainer decision,45,ready for maintainer decision,2,0,,"182987, 183028",,,3,63df7e39440c277900d9f06fbb8208b57c94e63ad0aa809ab3f3886d7261c8ab
thread:pr:187898:pr:188939,"pr #187898: [profiler][cupti] Perfetto-native (.pftrace) export, full chrome-args parity",ready for maintainer decision,45,ready for maintainer decision,2,0,,"187898, 188939",,,2,18c5417e9dd8bfac82395a04abd33a0c0bb2f0f7f012946b8c0ea06cc28ff4a5
thread:pr:126348,pr #126348: [AOTAutograd] tweak min-cut partitioner to avoid saving softmax output,ready for maintainer decision,45,ready for maintainer decision,1,0,,126348,,,0,8e3c97d4878196a39e54209f91aeef4f80ae296366713313d65b8465b5b54344
thread:pr:138068,pr #138068: Add TORCH_CHECK_INDEX in convert_indices_from_coo_to_csr_cpu,ready for maintainer decision,45,ready for maintainer decision,1,0,,138068,,,0,f69af2149240ca83493782a945c0de1dcfd9c50fe5d2eccd9b53e3ea584f0596
thread:pr:151218,pr #151218: Implement MKLGenerator,ready for maintainer decision,45,ready for maintainer decision,1,0,,151218,,,0,2b4fb69deea022019263d1ef7e4f230bf1bd815060e6e443844c5f234b718d40
thread:pr:152159,pr #152159: Add dynamo config to HOP-ify context managers,ready for maintainer decision,45,ready for maintainer decision,1,0,,152159,,,0,a3e4697f7a003f1f944a184b2564c144932ede31ad1d0f22c943c0354ea78869
thread:pr:162454,pr #162454: [xpu][feature] Integrate OneDNN SDPA training forward/backward into XPU OVERRIDEABLE Backend,ready for maintainer decision,45,ready for maintainer decision,1,0,,162454,,,0,90ff8a3a8372b6b07af76c91043c0bc1daa4f822938053cf3313c6497cc95a5f
thread:pr:171313,pr #171313: Support torch.accelerator.Graph on CUDA,ready for maintainer decision,45,ready for maintainer decision,1,0,,171313,,,0,2a048b6d8a381453c9b08fb07722b9e069cc9fad4d0f5d75a1023290684c9146
thread:pr:172003,pr #172003: Add __getitems__ to TensorDataset for faster DataLoader iteration,waiting on contributor,45,Wait for contributor update on PR #172003; keep related issue/PR context attached.,1,0,,172003,,,0,1bdb247fad16b1dd026f66e68b7c63ee25d4c602d435c43b2154832ff578bb88
thread:pr:172991,pr #172991: Fix torch.compile for jagged nested tensors created inside compiled functions,waiting on contributor,45,Wait for contributor update on PR #172991; keep related issue/PR context attached.,1,0,,172991,,,0,10b4ebcea6dab3726378473c9d71e0ac44c901f4a254815d5d5c682bde528258
thread:pr:173072,pr #173072: s390x: rework ZVECTOR data into 16-byte-sized vectors,waiting on contributor,45,Wait for contributor update on PR #173072; keep related issue/PR context attached.,1,0,,173072,,,0,41757550d1f24a36f6c302881376cb1809c669b3aa9fdc425f44064f31f9c360
thread:pr:173182,pr #173182: Updated foreach_norm to accept dim and keepdim arguments.,ready for maintainer decision,45,ready for maintainer decision,1,0,,173182,,,0,b6aca77d8d777985422575cafbb4afbd785d5d411d0b75549ac47a542305094f
thread:pr:174605,pr #174605: make Store timeout_ atomic for thread safety.,ready for maintainer decision,45,ready for maintainer decision,1,0,,174605,,,0,9f65760fbeb63564811cc10073a9e3074e56bde244730a3eae05bc8b326619e7
thread:pr:175083,pr #175083: [pytree] add another simplified pytree module `torch.pytree`,ready for maintainer decision,45,ready for maintainer decision,1,0,,175083,,,0,c7c248236e0b380967d40cbec9ed0f3ea5ea849c18e645bf7eb0ee0e4376e476
thread:pr:175425,pr #175425: Improve error message for CPU-CUDA copy during graph capture,ready for maintainer decision,45,ready for maintainer decision,1,0,,175425,,,0,2b5ea82c98b4d74f2071fc6013394f819293682868f550f9ca83480899ab6e20
thread:pr:175476,pr #175476: [inductor] Hoist output buffer allocations into prior CUDAGraph partition,ready for maintainer decision,45,ready for maintainer decision,1,0,,175476,,,0,943db35ff20d0f0910fac4119e85b4d8c4dc98c328c2090ab6ad8d7e24023e9b
thread:pr:175512,pr #175512: [kineto] Generate unique group_trace_id instead of reusing trace_id (#175512),ready for maintainer decision,45,ready for maintainer decision,1,0,,175512,,,0,ba25aa64d3cc18d36f1308837e567c6b899e11e2cb0c42271584da6fe6f16e34
thread:pr:175706,pr #175706: add C++20 bit operations and contains to C10,ready for maintainer decision,45,ready for maintainer decision,1,0,,175706,,,0,1462920ebf6373633e27ad846061e0b66658f2abb02f188dd164fa9105142b4b
thread:pr:176584,pr #176584: [RNG] Decouple Philox subsequence from hardware topology for cross-platform consistency,ready for maintainer decision,45,ready for maintainer decision,1,0,,176584,,,0,e66a5598678fe14b316a1147778d272a9a409bc81169b61cbba13be497f68636
thread:pr:176936,pr #176936: Add driver api guard to cuda utils,ready for maintainer decision,45,ready for maintainer decision,1,0,,176936,,,0,93bab6fee63c5e3c8de5f178a27a8d18b54c2b6d50fb5fa8b7e6389a81ad78a8
thread:pr:177147,pr #177147: [FSDP2] Remove redundant event wait for all_gather_stream,ready for maintainer decision,45,ready for maintainer decision,1,0,,177147,,,0,a10d5cbcbf56b72594c7931100afd10d583f04ea39b59e12c8fe1e374421c6ce
thread:pr:178037,pr #178037: [MPS] Raise clear error when MPS is used in forked subprocess,ready for maintainer decision,45,ready for maintainer decision,1,0,,178037,,,0,6afb4400e217865890bdfae8eb31814bfb22e699aa212a439fa9ace1d26c34ec
thread:pr:178068,pr #178068: [docs] Add comprehensive introduction to PyTorch Elastic documentation,ready for maintainer decision,45,ready for maintainer decision,1,0,,178068,,,0,1807393a9838e51fe8621e806461309b61438ae846c99c54c3cad286ce08e084
thread:pr:178069,pr #178069: [docs] Add examples to reduce_scatter documentation,ready for maintainer decision,45,ready for maintainer decision,1,0,,178069,,,0,d5d0e479994577f7fc293d72473b296e70a0de2bfca75ece1ab1bd39b4159317
thread:pr:178082,pr #178082: Fix ninja workdir collisions in BuildExtension,waiting on contributor,45,Wait for contributor update on PR #178082; keep related issue/PR context attached.,1,0,,178082,,,0,814450279f84e3b43fa250f16af2a5c873e29abdf0ef1edc782ac8514d94860c
thread:pr:178088,pr #178088: Use more structure binding,ready for maintainer decision,45,ready for maintainer decision,1,0,,178088,,,0,e7991a4cf00ea20c29ecfd65fc0c4f77f2d05b06a86de713fa2ef995c730f6e9
thread:pr:178373,pr #178373: docs(nn): clarify return type for MaxPool1d/2d/3d when return_indices=True,ready for maintainer decision,45,ready for maintainer decision,1,0,,178373,,,0,d2baab55c01ab84eeb62e0053d905d1384f309cc18b17f7a8533f58e02470a8c
thread:pr:178545,pr #178545: [MPS] Add native ops fast path for SDPA on long sequences (~2x speedup),ready for maintainer decision,45,ready for maintainer decision,1,0,,178545,,,0,c0da57d8fb3a03e0487ef0d9b2bfffc85ec2c8a29e7267d0cf6c998ad311b971
thread:pr:178568,pr #178568: Add support for activation checkpointing with nested data structures,ready for maintainer decision,45,ready for maintainer decision,1,0,,178568,,,0,3f5ea534c499a61b1e28682634f8c4580d4a0d69ad50262974bcb0a7cf44316c
thread:pr:178590,pr #178590: Correct type hints for ConvNd subclasses,ready for maintainer decision,45,ready for maintainer decision,1,0,,178590,,,0,045606afbbbceef6639a4d21c1797ecc5ca16192e1062c809f733a89a8fd5c07
thread:pr:178924,pr #178924: [inductor] Fix FP8 dtype detection in _get_estimated_runtime,ready for maintainer decision,45,ready for maintainer decision,1,0,,178924,,,0,b232945fad45b6cada3711198376ab13885b196a0f1b54f356e4f9dcbad28266
thread:pr:178960,pr #178960: profiler: fix USE_KINETO=OFF build failure (unconditional ActivityType.h include),ready for maintainer decision,45,ready for maintainer decision,1,0,,178960,,,0,152549a753e5d7606ab420e7aa4b2184a44142298fb0d9675684398af22c2f7c
thread:pr:179000,pr #179000: Add type conversion test for IValue: implements TODO(gmagogsfm),ready for maintainer decision,45,ready for maintainer decision,1,0,,179000,,,0,cb2f272baa8d152ee07d5f63ad848eb4fe6a796e6a03ec52c9961dda386a4c85
thread:pr:179020,pr #179020: Update third-party/rocm_composable_kernel to f1746955,ready for maintainer decision,45,ready for maintainer decision,1,0,,179020,,,0,a2014ebea0bd31c560b4054535c6001bd2f84da3be15b532cd17201195d0a38f
thread:pr:179212,pr #179212: feat(pipelining): add configurable auto-partitioning algorithm for pipeline parallelism,ready for maintainer decision,45,ready for maintainer decision,1,0,,179212,,,0,74439adb309d750fbf6c6ceadb3eb68e579605d8398bcef4215ed426dc427a1e
thread:pr:179399,pr #179399: Improve GELU documentation formula,ready for maintainer decision,45,ready for maintainer decision,1,0,,179399,,,0,f8c92bef5e7767fcfda74911372130b686c535812959f115b8b031c7ebaefaac
thread:pr:179485,pr #179485: Add torchbind custom class inheritance support (#179485),ready for maintainer decision,45,ready for maintainer decision,1,0,,179485,,,0,78b3ff2d0debdc6a0975470861ae1767b7c4bf535d98265df74eff33b3f68c84
thread:pr:179594,pr #179594: [CUDA] Use #pragma unroll in Normaization.cuh,ready for maintainer decision,45,ready for maintainer decision,1,0,,179594,,,0,ff26d36ca53751e385a721b55f1b30e8e23b6a3db1e37c86a2ac7e9048196788
thread:pr:179698,pr #179698: [torch][inductor] Fix numel variable scoping with enable_kernel_profile (#179698),ready for maintainer decision,45,ready for maintainer decision,1,0,,179698,,,0,d120bec7aea7910d6797293fc134f546cc81d8cf81f38b1d342f2c6a38accaad
thread:pr:179700,pr #179700: [torch][inductor] Add RAIIAtenRecordFunctionHandle profiling to remaining uncovered kernel paths (#179700),ready for maintainer decision,45,ready for maintainer decision,1,0,,179700,,,0,a6cb521aa494128b7015c0b982444bd2f7182b5c4d724a8c02a4ae55425e357f
thread:pr:179719,pr #179719: Skip div-to-mul-reciprocal when division_rounding is enabled,waiting on contributor,45,Wait for contributor update on PR #179719; keep related issue/PR context attached.,1,0,,179719,,,0,e24cf228ff4817da0d20a48516d9eb4f25a7f870ef7d6daff91b6877a55abf76
thread:pr:179747,pr #179747: Add rotary positional embeddings support to nn.MultiheadAttention,ready for maintainer decision,45,ready for maintainer decision,1,0,,179747,,,0,675d2a1125f11c5af01ebbfcb4d56544e156517b21edac61cb0a212f5cd2b41f
thread:pr:180171,pr #180171: Fix empty changed-files handling in Filepath heuristic,ready for maintainer decision,45,ready for maintainer decision,1,0,,180171,,,0,731ec22643870bcbc5d63b8b4a88f0e59604776358893ed1f265dbf1c9045ee1
thread:pr:180187,pr #180187: add OneCycleLR degenerate phase cases,ready for maintainer decision,45,ready for maintainer decision,1,0,,180187,,,0,94e30a561418a320b69ef4315e67c5ab3f431d75f461371e44577a35fd2d84a9
thread:pr:180209,pr #180209: XPU oneDNN: thread-local LRU cache for FP matmul primitives + post_ops deduplication,ready for maintainer decision,45,ready for maintainer decision,1,0,,180209,,,0,99df6d19290e5a97a4908ca7e7d5f2568bac8d005ff76d582354b0427c03ee63
thread:pr:180230,pr #180230: [TEST][CUDA][TF32] Update TF32 test decorators to use new fp32 precision path,ready for maintainer decision,45,ready for maintainer decision,1,0,,180230,,,0,9f234ad59dd7bb73f7d1088d169039db57667e82b5772bb0f3b6089b720791b2
thread:pr:180527,pr #180527: Add thread_time_ns and idle_pct to dynamo_timed chromium events (#180527),ready for maintainer decision,45,ready for maintainer decision,1,0,,180527,,,0,af037271dee175a99486613474f839ae1bec51f9fda9220aab545c98709bc39a
thread:pr:180584,pr #180584: Test,ready for maintainer decision,45,ready for maintainer decision,1,0,,180584,,,0,643aa1a691148bb223c5ffba286ec132d65cb06fe379039c635f783264adb44c
thread:pr:180686,pr #180686: test: relax error tolerances for complex64 tests with tan and tanh,ready for maintainer decision,45,ready for maintainer decision,1,0,,180686,,,0,8f2adbf73f07ad089c28175a2bca724e32d6ac78be2d34d20b6909cf3e7ee81b
thread:pr:180758,pr #180758: Remove dead TORCH_HIP_VERSION guards,ready for maintainer decision,45,ready for maintainer decision,1,0,,180758,,,0,20a2f28e098103f2ebb9fa66c6c13c1e56a3f7211135214c74afe72c4dc03094
thread:pr:180788,"pr #180788: Use C++20 concepts/requires in Tensor, TensorBase, and IValue",ready for maintainer decision,45,ready for maintainer decision,1,0,,180788,,,0,95a3e9e2df81a63b00f03dcf2c3fdbd992b42522d573c5408716ce0d291b1124
thread:pr:180954,pr #180954: Add value property to LazyVal,ready for maintainer decision,45,ready for maintainer decision,1,0,,180954,,,0,fd8f7e8bf34329a6db715ede8615ea27864010cb171a9f37b7c470acf544e42f
thread:pr:180988,pr #180988: [fbandroid][buckconfig removal] removal of last files referencing tools/build_defs/android/build_mode_defs.bzl,ready for maintainer decision,45,ready for maintainer decision,1,0,,180988,,,0,ac30985fc089333d4e782d61a982dc4eb31e24aa2825cb8362b6d3e86d289627
thread:pr:181030,pr #181030: [MPS] Migrate SDPA from MPSGraph to Metal,ready for maintainer decision,45,ready for maintainer decision,1,0,,181030,,,0,49cfef164ae8c21e98c741ff9d4bac3ac5860fd390c2bcd792047722cdc4d1ce
thread:pr:181285,"pr #181285: [TEST][ARM] Remove Pointer Authentication Code from address, fall back to `dladdr_lookup` where possible",ready for maintainer decision,45,ready for maintainer decision,1,0,,181285,,,0,c92850dd4554d6b9ee42ff481b75b25f9751a2d96db12c4ccd344f57164aa472
thread:pr:181292,pr #181292: [inductor] Autotune cache stability and perf improvements (#181292),ready for maintainer decision,45,ready for maintainer decision,1,0,,181292,,,0,e90c75fb12fb5197a748db4ac90b9e461cd1ee7650f17b81877857b53db2ee54
thread:pr:181330,pr #181330: [Profiler] Add a skip_build_tree metadata field to Kineto events to m…,ready for maintainer decision,45,ready for maintainer decision,1,0,,181330,,,0,0d44f979b039fcceda4f1289c2954ad73f402fd2efb9254b07cd481289ba18c8
thread:pr:181622,pr #181622: [Inductor][NVGEMM] Refactor rendering,ready for maintainer decision,45,ready for maintainer decision,1,0,,181622,,,0,d47cdf103a91c4fd4384e4cd5032b7cc39bf83586620eb85d189f4c9ce9bf64e
thread:pr:181833,pr #181833: Add boolean evaluation for LazyVal wrapping str,ready for maintainer decision,45,ready for maintainer decision,1,0,,181833,,,0,bc9c022cd0ed3fda12f4902a921bdef48de33225cd89a5f6ee95e2b67d5ee755
thread:pr:181849,pr #181849: Enable FP32 emulation via BF16 in cuSOLVER,ready for maintainer decision,45,ready for maintainer decision,1,0,,181849,,,0,b1c5b95f291772ea465c1411ac57e82646e8f456f69c70b5f7d64b2bb43497cd
thread:pr:181862,pr #181862: collect_env: add uv support,ready for maintainer decision,45,ready for maintainer decision,1,0,,181862,,,0,eb4eef51e9912c9fdcd995c34f238667acf3efcefcfa6c2d514063f0bfbf699a
thread:pr:181902,pr #181902: [Inductor][NVGEMM] Remove duplicate nvgemm_max_profiling config,ready for maintainer decision,45,ready for maintainer decision,1,0,,181902,,,0,32e093189bbea638ecb7a6c92e79d48ea204b27cfa8d167c9e41690a9e2fc305
thread:pr:181903,pr #181903: [Inductor][NVGEMM] Avoid double iteration over kernel cache during choice enumeration,ready for maintainer decision,45,ready for maintainer decision,1,0,,181903,,,0,4ee3761b995d6964e1a2be7c196d9aee4c9e4aeb8a498831cfeb2b20ce085193
thread:pr:181910,pr #181910: [Inductor][NVGEMM] Link tracking issue in supplement configs comment,ready for maintainer decision,45,ready for maintainer decision,1,0,,181910,,,0,9e602bbd9ce1734932241b48e3fcc3505bfa4e4d88ae37b9375ca560e8aaf6cf
thread:pr:181996,pr #181996: [torch.package] Register package modules in sys.modules for Python 3.12 dataclass compat,ready for maintainer decision,45,ready for maintainer decision,1,0,,181996,,,0,d64525bf3a1578a18883e536cdc3694da2bcc153d06d8f09ebad10df71ceb231
thread:pr:182024,pr #182024: [docs] Fix typo in data page,ready for maintainer decision,45,ready for maintainer decision,1,0,,182024,,,0,f7cad93c6017a77d86f044ba7ed4e3fdcced5e016ba639047580cdf53755727f
thread:pr:182030,pr #182030: [ROCm] Make AOTriton bundling optional via `BUILD_AOTRITON_INTO_WHEEL` flag,ready for maintainer decision,45,ready for maintainer decision,1,0,,182030,,,0,282803d72c5ed02f4fa52783396f135595a260c53a946ab7aad8c399445333b8
thread:pr:182040,pr #182040: Download magma & sccache install scripts over https instead of aws s3,ready for maintainer decision,45,ready for maintainer decision,1,0,,182040,,,0,8ee8a9b67d764ddf0cc9ec891a2168d22745e4d866c6abd9779f956fec8552fb
thread:pr:182053,pr #182053: Partition load_state_dict child keys once,ready for maintainer decision,45,ready for maintainer decision,1,0,,182053,,,0,cf506f75b20128265f6eea7d346495deebf061262af5c761dd03572147746eac
thread:pr:182061,pr #182061: Document low-memory state_dict loading with meta device,ready for maintainer decision,45,ready for maintainer decision,1,0,,182061,,,0,7782a0d5dcf396c685becada4aad37e8affdb83af018311f69574b041bd75ab0
thread:pr:182159,pr #182159: Fix List isSubtypeExt,ready for maintainer decision,45,ready for maintainer decision,1,0,,182159,,,0,392fb4a8777fc8f7cfd0d7b123475bbeaeffc62a1badd3bb7af805c228f91256
thread:pr:182289,pr #182289: [MPS] Pinned memory - Add pinned memory allocator,ready for maintainer decision,45,ready for maintainer decision,1,0,,182289,,,0,b990fd3405a8c1107c534a48474efdfe602352efc9d2ac6de79cb6de52033ad3
thread:pr:182340,pr #182340: Bump Windows CPU trunk test shards from 4 to 5,ready for maintainer decision,45,ready for maintainer decision,1,0,,182340,,,0,d9983e85f6cc041ff13619808f8e17ce67c326fa76ce0f021800c69f2c496497
thread:pr:182426,pr #182426: Enable FP32 emulation via BF16 in cuSOLVER for qr and lu,ready for maintainer decision,45,ready for maintainer decision,1,0,,182426,,,0,b611b30e6c99f315d2823143ef896e4afc828ad64b48956711e5315c4e98c855
thread:pr:182582,pr #182582: Add addmm quantization graph passes for PT2 readiness benchmarks (#182582),ready for maintainer decision,45,ready for maintainer decision,1,0,,182582,,,0,690aeee345b2dfac4656b5d48170feedfeca64be57927e243a6437ab1f8213bb
thread:pr:182687,pr #182687: Support generator arg in dirichlet and beta torch.distributions API,ready for maintainer decision,45,ready for maintainer decision,1,0,,182687,,,0,508c6258b50e13d5cfa6e0099567cc08101b83ab2196995d11baa9e89ac9e084
thread:pr:182747,pr #182747: Resolve PYTHON_SIX_SOURCE_DIR from installed `six` to skip pypi download,ready for maintainer decision,45,ready for maintainer decision,1,0,,182747,,,0,9c000936d5facc592b7ed4d2ee48f7c84a372a23369d118a1006dff91ada7a7d
thread:pr:182785,pr #182785: [Inductor] Fix FX wrapper codegen for symbolic floor/ceil offsets in ReinterpretView (#182785),ready for maintainer decision,45,ready for maintainer decision,1,0,,182785,,,0,bca3d6c82fe6ac8670a467df77427dc4b13108e81b6a35792df5594045456ccc
thread:pr:182919,pr #182919: Add forward-only cuDNN FP8 per-tensor attention,ready for maintainer decision,45,ready for maintainer decision,1,0,,182919,,,0,6d852ee6d6be7c7db02cdc1a44f7c31d9f0d4cdb3eaa25861407cb74ae34a6ea
thread:pr:183343,pr #183343: dynamo: trace torch.cuda.Event.record / Event.wait under fx proxy mode,ready for maintainer decision,45,ready for maintainer decision,1,0,,183343,,,0,28db086fea1184d9feb641968f2944eb5fd5546cd8a774a4681debb029f6dcb4
thread:pr:183501,pr #183501: [PyTorch][Pinned Memory] Fix stuck event blocking in CachingHostAllocator,ready for maintainer decision,45,ready for maintainer decision,1,0,,183501,,,0,95c5fac2fcec45da64c6740d0b74c3a05c7a07d64c40495ca5e9ee4bd224440f
thread:pr:183716,pr #183716: Validate dim in cummax/cummin meta for empty tensors,ready for maintainer decision,45,ready for maintainer decision,1,0,,183716,,,0,f5993c27a4768515cd20638990e2ba9f5fd478c98e1a8d76fbfd52db6b68e029
thread:pr:183738,pr #183738: Multi-warp TMA scatter_add via PerWarpTmaPipeline,ready for maintainer decision,45,ready for maintainer decision,1,0,,183738,,,0,e318fc195a8b963d6d18dad03dfc72e15c0c1ef187be16a6a99a7ea5d14dd0c0
thread:pr:183739,pr #183739: TMA scatter_add: 2D-tile multi-row path for small-N shapes,ready for maintainer decision,45,ready for maintainer decision,1,0,,183739,,,0,176d0ef53ed1148a2acb1ca5ff6fc7264544572c9ae93adcb1a14353bc16410b
thread:pr:183752,pr #183752: [BE]: Revert back to upstream uv action to solve rate-limiting,ready for maintainer decision,45,ready for maintainer decision,1,0,,183752,,,0,b3718b5766c183721523d6938224477fe8747c287b760ccca40397beed477618
thread:pr:183755,pr #183755: [MPS] Enable bf16/fp16 ConvTranspose3D by removing the gating,ready for maintainer decision,45,ready for maintainer decision,1,0,,183755,,,0,0c9f1ad1067d5f3c77bc35b7bb37fc857e356cc9aef4a5a64816fb5ab657ecfd
thread:pr:183811,pr #183811: [docs] Document 10 public APIs in torch.nn.parallel submodules,ready for maintainer decision,45,ready for maintainer decision,1,0,,183811,,,0,e662bccfcb3a595068415314d05538d25a8e7f9d06efc8f629c76c0e608515bf
thread:pr:183852,pr #183852: ci: declare workflow-level `contents: read` on 8 workflows,ready for maintainer decision,45,ready for maintainer decision,1,0,,183852,,,0,68a65b11a71b05017398b1953433a3a573e234d7fdacd775c6391f92ebd6d909
thread:pr:183855,pr #183855: [c10d] Surface operational metadata in WorkNCCL operator<<,ready for maintainer decision,45,ready for maintainer decision,1,0,,183855,,,0,5d99149c22af3eaecf8d2b00ca7bc5ccc9c66d6d62382f311e6cc505c65cc4e9
thread:pr:183870,pr #183870: [stats] Handle alternate test-reports path in backfill_test_jsons_while_running,ready for maintainer decision,45,ready for maintainer decision,1,0,,183870,,,0,4d0a34c095f8e33631f6e9a2a10b59361cefe6494f930e803e969ab74990748d
thread:pr:184142,pr #184142: Fix LPPool signed fractional inputs,ready for maintainer decision,45,ready for maintainer decision,1,0,,184142,,,0,f5aa82c360785818ee628d4a2d733e33dbe55abc221793903f09e2f1b713ea3e
thread:pr:184190,pr #184190: [fx]: Add SymNode helper to get raw expr,ready for maintainer decision,45,ready for maintainer decision,1,0,,184190,,,0,570fa298f5dc89ca26dcef58f70c8bc5ab67f780489503cc83fa691eab675edb
thread:pr:184398,pr #184398: Enable 3.15 in CI,ready for maintainer decision,45,ready for maintainer decision,1,0,,184398,,,0,d5efb367d613f95bd103951876d2c6ab301b1aa27aeebc2ff4addc00d9ac4fe5
thread:pr:184426,pr #184426: Support getting CUDA patch version,ready for maintainer decision,45,ready for maintainer decision,1,0,,184426,,,0,516d45f9bc5426854230a84bb7dcb1165f3804744a5502af75bed57eb89b16e5
thread:pr:184548,pr #184548: [Test]Refactor test/test_sort_and_select.py to be device generic,ready for maintainer decision,45,ready for maintainer decision,1,0,,184548,,,0,72a19673c979a650535ba53c4d0a8f474aa29f035f3c14670e074ebf224cbbc2
thread:pr:184559,pr #184559: [Test]Refactor test/test_reductions.py to be device generic,ready for maintainer decision,45,ready for maintainer decision,1,0,,184559,,,0,37b47158c6dc7a1b3ad83f962e3caa91fdd006cf76005b02a0ef3a97e167d44a
thread:pr:184585,pr #184585: [MPS] Wire SDPBackend dispatch for scaled_dot_product_attention,ready for maintainer decision,45,ready for maintainer decision,1,0,,184585,,,0,7c2ef6f0612a6ce01fe431e1f60472448c85e32fb5800cda5ec552164a2a2f9a
thread:pr:184718,pr #184718: [Test] Make test_testing.py device-agnostic for out-of-tree accelerators,ready for maintainer decision,45,ready for maintainer decision,1,0,,184718,,,1,a13c8bd3453c55751b8e581bef8770df3a4a0f9a88f3b382b4fe6123bcb0246a
thread:pr:184742,pr #184742: [Test] Make test_expanded_weights.py device-agnostic for out-of-tree backends,ready for maintainer decision,45,ready for maintainer decision,1,0,,184742,,,0,d2d86f6cc624491ca1dd8fa1654e748e4daabbe085a39f9f9da8bdc28f3b4bbe
thread:pr:184859,pr #184859: Deprecate c10::ssize as alias for std::ssize,ready for maintainer decision,45,ready for maintainer decision,1,0,,184859,,,0,4996a7f6d15198b85ca08f350c5c459828b2d2da842db5a282e78f1594380492
thread:pr:184866,pr #184866: [distributed] Handle empty inputs and kwargs in _to_kwargs,ready for maintainer decision,45,ready for maintainer decision,1,0,,184866,,,0,8d029cd710b59a6b09cdb421e051a9112400d26fca0a605583df2403f4d1432e
thread:pr:185125,pr #185125: Remove unused caffe2/utils/string_utils,ready for maintainer decision,45,ready for maintainer decision,1,0,,185125,,,0,cd8dc5b61881bddaa3f39b4c60de21ec83c66bea1992ae54dff1c5e93ce418b0
thread:pr:185137,pr #185137: [fx][_library] Honor wrapped op's schema.is_mutable in is_impure's auto_functionalized HOP branch (#185137),ready for maintainer decision,45,ready for maintainer decision,1,0,,185137,,,0,4cc433f0028432b86b197864dac43e319206b3deafcf42840a8ada36941aaf41
thread:pr:185184,pr #185184: [Test][distributed] Refactor some distributed test helpers to be device-agnostic,ready for maintainer decision,45,ready for maintainer decision,1,0,,185184,,,0,32429b86b5962fbae4b880d848ed6b697899069df57ea324f64212399e31e83b
thread:pr:185206,pr #185206: Make minifier_tests more device agnostic,ready for maintainer decision,45,ready for maintainer decision,1,0,,185206,,,0,18b4bd8c4dbb056684dc69810561d3d50d30aee190672a65c5ee762a2be7cbfb
thread:pr:185446,pr #185446: Adding a test case using fx.wrap to test whether TLX user-defined kernel is visible in the inductor output (#185301),ready for maintainer decision,45,ready for maintainer decision,1,0,,185446,,,0,e5cbd954a50c2491a3bc0a1f5f6e4004b31f22e5052121fb75762075b50cae87
thread:pr:185507,pr #185507: Pre-compute igates for XPU LSTM to reduce per-step kernel launches,ready for maintainer decision,45,ready for maintainer decision,1,0,,185507,,,0,eebb2ed7deaa3993a86c9d2c211d168ba39b86acb9da07f897884d2fb541ea30
thread:pr:185648,pr #185648: [Test] Make sigmoid_add extension test device-agnostic in test_cpp_extensions_aot.py,ready for maintainer decision,45,ready for maintainer decision,1,0,,185648,,,0,9355bc0a0938c45d700ed4fae32c79d9492f4961a972103660c8ed72d84529e1
thread:pr:185743,pr #185743: Mark CUDA kernel helpers static,ready for maintainer decision,45,ready for maintainer decision,1,0,,185743,,,0,56d4b7e525963cf580c0fe09ca61f9881c3ea1d8aa6ba329bfbeaf912f35fc17
thread:pr:185758,pr #185758: Tidy MPSGuardImpl: name capability bitmask and fix clang-tidy nits,ready for maintainer decision,45,ready for maintainer decision,1,0,,185758,,,0,4782b00bd9250bf7bbb838a27bc5522c9ab35563dcaf90151d44aabb90de3a14
thread:pr:185788,pr #185788: Support 64-bit indexing in channels_last bilinear upsampling,ready for maintainer decision,45,ready for maintainer decision,1,0,,185788,,,0,8a3e3111205ac899f3e74adf058e7c19ffb66f39a488f68c5b9d61e8b6ad9665
thread:pr:185881,pr #185881: [Test] Refactor test/test_reductions.py,ready for maintainer decision,45,ready for maintainer decision,1,0,,185881,,,0,2183fb9c4a0979944b9fc08da268410f0a74db3b6be95e8f0866e42ab54e64f6
thread:pr:185883,pr #185883: Simplify ostream formatting with fmt::join and fmt::print,ready for maintainer decision,45,ready for maintainer decision,1,0,,185883,,,0,2fd4be48aa1c56c0b50db6a6debe1d3d3abea894d52e158f592655b29545db3f
thread:pr:185901,pr #185901: [Test] Refactor test_schema_check.py to be device-agnostic,ready for maintainer decision,45,ready for maintainer decision,1,0,,185901,,,0,214a059b19e41f236f6038e4f337696ef8eddc63c7a6b8aa57d522c44410ea91
thread:pr:185906,pr #185906: add type hint to seed argument in torch.random.manual_seed,ready for maintainer decision,45,ready for maintainer decision,1,0,,185906,,,0,8df2128447bbb0b27fcb3a7d2f3fa8c35d56dcd16c9d02449a755cb62d34f11b
thread:pr:185930,pr #185930: [Elastic][Distributed] Flush dirty rendezvous state immediately,ready for maintainer decision,45,ready for maintainer decision,1,0,,185930,,,0,362de14744af8be8f1208bef1c32441ad9667d7305d524f4208f7ef99e645ec6
thread:pr:185993,pr #185993: Fix SM121 compute capability corruption in CMake GPU arch detection,ready for maintainer decision,45,ready for maintainer decision,1,0,,185993,,,0,cec3d0d0d2c7c972fd01aa935c11417682264be2c15cc43c77c6aea227f28c77
thread:pr:186045,pr #186045: Merge identical elif branches and inline single-use locals in torchgen,ready for maintainer decision,45,ready for maintainer decision,1,0,,186045,,,0,8cdf1c50dc0fd278672b7c34412f9d46ad84f1f3d6f1d7e4a895fb0938b5259a
thread:pr:186046,pr #186046: [MPS] Remove dead xfail/tolerance keys from MPS test infra,ready for maintainer decision,45,ready for maintainer decision,1,0,,186046,,,0,b3c5f5db9f95f5dc80fa7692db547ae91736b558b3f229cb4b8a205657a6e2b0
thread:pr:186110,pr #186110: Remove test_transformers from windows blocklist,ready for maintainer decision,45,ready for maintainer decision,1,0,,186110,,,0,6e1be3824fdf7b90ab1057833659fde5d79c9efda15b2f5f4a6f1fb9626e6f1a
thread:pr:186145,"pr #186145: Spelling fix: Change ""structeur"" to ""structure"" in Lumen CLI VLLM test",ready for maintainer decision,45,ready for maintainer decision,1,0,,186145,,,0,18f392576b09aa9b857524649e7de5240e49d086817a87e5af42da4bdf903623
thread:pr:186232,pr #186232: Add opt-in cross-stream segment reclaim to CUDA allocator,ready for maintainer decision,45,ready for maintainer decision,1,0,,186232,,,0,d8d53713d97d9d4b000aaf81753b7d1f5090f1b9935ec4c5bba8ff71792cb7cf
thread:pr:186239,pr #186239: [ROCm][CI] Add linux.rocm.mi350.docker-cache to docker-cache-rocm,ready for maintainer decision,45,ready for maintainer decision,1,0,,186239,,,0,dfd6d28d38c10dd88101926be82faf1f35c59b4f7f0785ff693fca189e82bc1f
thread:pr:186273,pr #186273: [TEST][CUDA][TF32] Re-enable TF32 setter check test,ready for maintainer decision,45,ready for maintainer decision,1,0,,186273,,,0,98db0f191cc5ad41d95ffb244fc79ff644dfec70ff1e8656584a6dc94b7e13d9
thread:pr:186307,pr #186307: Add contains() to ska::flat_hash_map/set,ready for maintainer decision,45,ready for maintainer decision,1,0,,186307,,,0,3dd6d8e3739f4f63b05e10f1804b392728798edcb24fba9defdb80d7fcb6e75d
thread:pr:186345,pr #186345: [Test] Make test_comparison_utils.py device-generic,ready for maintainer decision,45,ready for maintainer decision,1,0,,186345,,,0,24034618973d47b8b7ec3ee1286d6a1d3038956212073b77dcb09e3f65972464
thread:pr:186473,pr #186473: [accelerator] Support multiple private backends,ready for maintainer decision,45,ready for maintainer decision,1,0,,186473,,,0,59aefc3251cd1d25b5b3f6a1f1ee7c8215ed6a30e23b52bb46bc734664542d1d
thread:pr:186559,pr #186559: Remove deprecated pytree_to_str and str_to_pytree functions,ready for maintainer decision,45,ready for maintainer decision,1,0,,186559,,,0,af7c6a7cebff555d7d359242171553e4a3d603990a9420b40b36f427daa16531
thread:pr:186565,pr #186565: Support parametrized modules in functional_call,ready for maintainer decision,45,ready for maintainer decision,1,0,,186565,,,0,d752a9c065eae3c8ef3f4f8684ec9179bd426e22cdf8ff6b722ec38a5aedf330
thread:pr:186568,pr #186568: Add expected-failure test for ConstantLR factor=0,ready for maintainer decision,45,ready for maintainer decision,1,0,,186568,,,0,e0829a89e6336b271a2aad45c2d5890cb6a5275845f3b7cb480c0149373dc77f
thread:pr:186603,pr #186603: Exposing MiMalloc options and stats,ready for maintainer decision,45,ready for maintainer decision,1,0,,186603,,,0,9fb8edb5038b633dd684c9242b4e49b1b81072e7f623a2353787970af22448d9
thread:pr:186617,pr #186617: Added profiler record for THPVariable_dealloc,ready for maintainer decision,45,ready for maintainer decision,1,0,,186617,,,0,027233bcf476f5c773613a46e34f45c9c09678d6817d7270947f0f9d42cd1863
thread:pr:186624,pr #186624: Fix CLANGTIDY errors in c10/cuda,ready for maintainer decision,45,ready for maintainer decision,1,0,,186624,,,0,e38d6aafb2f2ff034fda61a2b3a39d0e592c0b81b08fe041a1884aa8a8bbbdd9
thread:pr:186881,pr #186881: nn/utils: skip no-op .to(first_device) in _get_total_norm,ready for maintainer decision,45,ready for maintainer decision,1,0,,186881,,,0,40f7e4e037c710ad83a4295efb5c2a06d806f0eb5a1c25b32d88ffc97a4510b8
thread:pr:186887,pr #186887: Update test_mm_sparse_first_NT/NN input generation,ready for maintainer decision,45,ready for maintainer decision,1,0,,186887,,,0,9d796a32f239bb4a5ef88e51decd211495fec7880788ec4fb8836b5caf827601
thread:pr:186924,pr #186924: Also search $MKLROOT/lib in Linux for mkl wheel and install_mkl.sh,ready for maintainer decision,45,ready for maintainer decision,1,0,,186924,,,0,03329bfae0d89a8c587b53582e2a706b846a1a86f650321c62518b1c46652d1a
thread:pr:187006,pr #187006: Avoid duplicate sm_103 family-specific CUDA codegen,ready for maintainer decision,45,ready for maintainer decision,1,0,,187006,,,0,4c6e9779fc536a1d9bcca591e5a62f0a176c32e127edf7cfd962db777d6f730a
thread:pr:187097,pr #187097: [BE] Fix B007 unused loop variables,ready for maintainer decision,45,ready for maintainer decision,1,0,,187097,,,0,6398302925c5e95364ec6b4cb7d83ed7b298413e805810b42078060ee81bc4ff
thread:pr:187126,pr #187126: [c10] Remove deprecated CUDAAllocatorConfig forwarders,ready for maintainer decision,45,ready for maintainer decision,1,0,,187126,,,0,49ba310f2fbfff1567f4673ace2375f184a13b6d9f0d9910f169ddea0e10a3fd
thread:pr:187135,pr #187135: [xpu] Remove deprecated at::xpu CachingHostAllocator forwarders,ready for maintainer decision,45,ready for maintainer decision,1,0,,187135,,,0,162f5f035c388ffb846a943a78e1345e057abd478c6ba945c02da356be065f3f
thread:pr:187146,pr #187146: test all named dims test with dynamic sepcs using automatic conversion,ready for maintainer decision,45,ready for maintainer decision,1,0,,187146,,,0,92e6c0ab4eac3426d247cfe398767c521b50f5c55b72a5f347d1bffddf10b7c2
thread:pr:187192,pr #187192: Protect GraphPassRegistry map with shared_mutex against concurrent access (#187192),ready for maintainer decision,45,ready for maintainer decision,1,0,,187192,,,0,186e705294aa264fc70ca6ce528617c06e6c4a4b8c115c5f819cd07609c0082f
thread:pr:187245,pr #187245: update backend parametrization for varlen attn,ready for maintainer decision,45,ready for maintainer decision,1,0,,187245,,,0,7fda4d145ba61b41dac5de130051a44e7806a10a46f76fb4bf734067b6aaf363
thread:pr:187317,pr #187317: [pr-review skill] Add device portability checks for shared accelerator code,ready for maintainer decision,45,ready for maintainer decision,1,0,,187317,,,0,bc11a1ae2c5bb3f0bbe7180b2d54151f2982aeac889ac5ebffb09bd16d546133
thread:pr:187333,pr #187333: [CUDA] linalg: overlap looped cuSOLVER LU (getrf) across auxiliary streams,ready for maintainer decision,45,ready for maintainer decision,1,0,,187333,,,0,bd9914a5a460eec4c3ffc54967df8cd2a5153a7ed05785a6d7d0425383c01da9
thread:pr:187404,pr #187404: [inductor][cutlass] Normalize reshaped external reads in EVT epilogue fusion,ready for maintainer decision,45,ready for maintainer decision,1,0,,187404,,,0,9d8be3a39c4564e0f6dc809788c764cc56d99b054741109f6528ebb209fca586
thread:pr:187410,pr #187410: [distributed] Use node.target instead of node.name in apply_manual_re…,ready for maintainer decision,45,ready for maintainer decision,1,0,,187410,,,0,d150d075f586ab4312652fa00a759c7225320383bf63da5ec8a06116bc9b656a
thread:pr:187566,pr #187566: [docs] Improve autograd anomaly mode documentation,ready for maintainer decision,45,ready for maintainer decision,1,0,,187566,,,0,cf4b3f00d8244e21b596555827dc4c716339c684fb70695fa22ab653d45d8355
thread:pr:187687,pr #187687: fix: aten embed bwd thread sync before proceeding to next batch,ready for maintainer decision,45,ready for maintainer decision,1,0,,187687,,,0,3ee5ec9a1d338f851df0e7f421d56c2cc60e70e999434ae6ae52fef3fb031728
thread:pr:187694,pr #187694: Unpack Python autograd saved tensors in reverse order,ready for maintainer decision,45,ready for maintainer decision,1,0,,187694,,,0,681fa3fe24062ef93d3e7c6b88cdb994b1344f826a2ae49fafcce97df677daa0
thread:pr:187713,pr #187713: Fix B008 flake8-bugbear violations (function calls in default arguments),ready for maintainer decision,45,ready for maintainer decision,1,0,,187713,,,0,f36ef3cd3b8345b1d4e799c23a310e3a7c2fd06ab260aecd884ce95334b9b056
thread:pr:187717,pr #187717: [CUDA] solve_triangular: fix silently wrong batched results and slowness in large-RHS workaround,ready for maintainer decision,45,ready for maintainer decision,1,0,,187717,,,0,c51dbc8b2521d51689015bcfae3a14f44cdc60bc107eb7822bcd81ada5530380
thread:pr:187742,pr #187742: Fix calculate_out_dimension using stride[0] for the width dimension,ready for maintainer decision,45,ready for maintainer decision,1,0,,187742,,,0,ca09462278d21eb581da42a56515c987578b7bb75c2a95ce4fb0e0bb53944b7e
thread:pr:187769,pr #187769: [docs] Clarify Softplus overflow behavior for reduced-precision dtypes,ready for maintainer decision,45,ready for maintainer decision,1,0,,187769,,,0,e161a56c0ddec73c13acfea72d9300661f898e564e6a5f5481a22c5b70912baa
thread:pr:187796,pr #187796: Add native MPS SVD path,ready for maintainer decision,45,ready for maintainer decision,1,0,,187796,,,0,b2d4d05aa48a94b77ac933f531dfa6b52e18a076c9b12a9afb1d5c4382ed5a98
thread:pr:187799,pr #187799: Fix ROCm libdrm readlink buffer bounds,ready for maintainer decision,45,ready for maintainer decision,1,0,,187799,,,0,37e0512856bbcd5174a4ed7570a605d30ffb4d56f9796fd737c7b1e673118dbb
thread:pr:187800,pr #187800: Verify Magma archive before extraction,ready for maintainer decision,45,ready for maintainer decision,1,0,,187800,,,0,5ff3865830c76bc61a32e7f904ee8181ce72840047e124aa19efebe2e2c13540
thread:pr:187813,pr #187813: Generalize memory viz and cycle detection,ready for maintainer decision,45,ready for maintainer decision,1,0,,187813,,,0,a79381674fd9ee66a01477db6bdb901d4eda789080cb32c53423f9f53cc88a83
thread:pr:187816,pr #187816: Clean up mentions of deprecated nvidia-docker in README files,ready for maintainer decision,45,ready for maintainer decision,1,0,,187816,,,0,ae75e31c28f83a3e7704e5ccb8f867b2e49d9a1cef8822a1919a4714df657081
thread:pr:187847,pr #187847: fix(nn.utils.prune): don't crash adding a method to an empty PruningContainer,ready for maintainer decision,45,ready for maintainer decision,1,0,,187847,,,0,e63fee4228c32ed0780f279b42de53dfb882f4500cc142b7874327ea48bca86b
thread:pr:187881,pr #187881: Fix margin docstring rendering in HingeEmbeddingLoss,ready for maintainer decision,45,ready for maintainer decision,1,0,,187881,,,0,6d2bcc1de44af911314f980c9f8de5dddd95b4297b2be42a860ca001d80962e9
thread:pr:187901,pr #187901: [TEST] Prevent PLATFORM_SUPPORTS_* from skipping tests on privateuse1 backends,ready for maintainer decision,45,ready for maintainer decision,1,0,,187901,,,0,860291f6c737a3837434d12ca9749ac3b93a7cc95601bcf175ecd3fafbc0eb87
thread:pr:187921,pr #187921: Support freeing blocks with stream uses during graph capture,ready for maintainer decision,45,ready for maintainer decision,1,0,,187921,,,0,9f33043521ffce902841d411d71c47eb471bd8b0b3016adce0a7e028755f6f87
thread:pr:187999,pr #187999: Fix uninitialized memory warnings in CPU linear algebra kernels,ready for maintainer decision,45,ready for maintainer decision,1,0,,187999,,,0,ca3e7b6a57e51cd997b9af828ff6c1f74b8aa3973103b10301107ae8fa30440c
thread:pr:188031,pr #188031: Fix 64-bit indexing in spatial softmax backward,ready for maintainer decision,45,ready for maintainer decision,1,0,,188031,,,0,6e4f6867948f9b4185fc9430da64f007d70d2ca16274d5cb5f172fc64092df47
thread:pr:188045,pr #188045: [MPS] better memory allocator,ready for maintainer decision,45,ready for maintainer decision,1,0,,188045,,,0,57bfdeefc34ac8c87b717bf0acde414a60f6ea53b958901b570631857199b94a
thread:pr:188089,pr #188089: [forward AD] Remove obsolete __debug__ guard when loading JVP decompositions,ready for maintainer decision,45,ready for maintainer decision,1,0,,188089,,,0,48e1f5addd0cbc8ea82233c17a808ee95f73e122a268f4b03f9f8bb09c213906
thread:pr:188125,pr #188125: [AI Codemod][PerfAICT-General] perf: Special case non-symbolic set_sizes_contiguous,ready for maintainer decision,45,ready for maintainer decision,1,0,,188125,,,0,1990c7aa6563b72566e90693de76757f2339241a2c02721b8bad479624d21102
thread:pr:188126,pr #188126: [inductor] Enable cudagraph partition inside invoke_subgraph subgraphs,ready for maintainer decision,45,ready for maintainer decision,1,0,,188126,,,0,ed45e7fde11e8c51c5d6fd91957a5c52b6694790978842c63dc4b89c5fa77f7f
thread:pr:188164,pr #188164: Fix THPDTypeInfo_compare reading off the end of non-finfo/iinfo objects,ready for maintainer decision,45,ready for maintainer decision,1,0,,188164,,,0,5ee30e1101a2b35453d9cc7f8f0fdf6843ab29759cdddc2e45bf688441a4925e
thread:pr:188191,pr #188191: Add sm_121a (NVIDIA GB10/DGX Spark) to default CUDA arch list,ready for maintainer decision,45,ready for maintainer decision,1,0,,188191,,,0,cc6055cf44522fa1e593466927cae2ccc5488d100e29daff235d58cc18b35c75
thread:pr:188250,pr #188250: Remove dead USE_ROCM block in CUDADeviceAssertionHost,ready for maintainer decision,45,ready for maintainer decision,1,0,,188250,,,0,86b05d9361fe172ec906072b79b79e7c94894201a982b7232eba868f6d443ce4
thread:pr:188252,pr #188252: Fix ROCm round-robin counter reset in initSingleStream,ready for maintainer decision,45,ready for maintainer decision,1,0,,188252,,,0,6a34ce6e6362c62894852332513a8cb5f060f1288a599d65046cd842b7d7f7da
thread:pr:188293,pr #188293: Allow tuple subclasses for run_backward inputs,ready for maintainer decision,45,ready for maintainer decision,1,0,,188293,,,0,4fd5235ca80a38bd25df19fcf367b6a489e50abc9ea81f6ab9f011c74862965f
thread:pr:188315,pr #188315: [HOP] Autograd implementation for switch,ready for maintainer decision,45,ready for maintainer decision,1,0,,188315,,,0,d8c5906033c317d8a7d6fa443aaa6066fbdb7d0449ce21ea51d354a4712d2261
thread:pr:188329,pr #188329: Optimize all-gather copy-in with a flat parameter buffer,ready for maintainer decision,45,ready for maintainer decision,1,0,,188329,,,0,483a5b394c39eee557b7f8730cdeed1e74f01898502e8e34d60d2b2de82f9be1
thread:pr:188348,pr #188348: [HOP] Fixes for zero-length scan dim,ready for maintainer decision,45,ready for maintainer decision,1,0,,188348,,,0,79c3e29de5e37532f8159c07b9ad5ea3161591393381580daf411f2c0a698b59
thread:pr:188364,pr #188364: Allow from_blob / TensorMaker to attach BackendMeta at construction,ready for maintainer decision,45,ready for maintainer decision,1,0,,188364,,,0,392050a1eb7924500d251beb570591f7e5150cf4a14010a3d7ab437ce91b62d1
thread:pr:188379,pr #188379: [HOP] None input handling for scan,ready for maintainer decision,45,ready for maintainer decision,1,0,,188379,,,0,915cb610d5dc6af32a5a98b492b6edf988e871a937ed3b705f6dc2a811fe0f81
thread:pr:188402,pr #188402: Fix matrix_exp crash on non-contiguous (channels-last) inputs,ready for maintainer decision,45,ready for maintainer decision,1,0,,188402,,,0,fc48dc52f585ed123876031e798b5891164314ca1309aebcbf438a7a2b4dda9d
thread:pr:188407,pr #188407: [Windows] Add triton_windows commit pin for Windows nightly builds,ready for maintainer decision,45,ready for maintainer decision,1,0,,188407,,,0,9411dffb28242b81c1561e8c8bb117287427470a4988611438f536d64488298a
thread:pr:188433,pr #188433: linear_cross_entropy: test and document bias with probability targets,ready for maintainer decision,45,ready for maintainer decision,1,0,,188433,,,0,7f3ccaa61ccc614c65d299374b9fb13e3fd4243ff3754ab14e2bef4da4a17660
thread:pr:188434,pr #188434: [dcp] Free staged dict on caller thread via weakref.finalize in async_save (#188434),ready for maintainer decision,45,ready for maintainer decision,1,0,,188434,,,0,bb188a0559b033e7f0b6b9726df3fcdce6083d45b966562f06f51854729b1b5d
thread:pr:188530,pr #188530: optim: remove redundant Adafactor assert,ready for maintainer decision,45,ready for maintainer decision,1,0,,188530,,,0,d0fd4c9a23699393da51fcd0c2e0a57c88a63ac949e6da2c00c90b7be9ad7acb
thread:pr:188555,pr #188555: Fix rocm lazy context initialization in test_set_device_0,ready for maintainer decision,45,ready for maintainer decision,1,0,,188555,,,0,ef16ebdc041efda73e305a4d03e6ce52f6ea6175468303c1ac49113bd92c0d10
thread:pr:188606,pr #188606: Relax complex tensor consistency tolerances for CUDA binary nonlinear ops,ready for maintainer decision,45,ready for maintainer decision,1,0,,188606,,,0,b8d876d6fbfd78001c393a04c2ef6e5c6774b58f2f3267d96e14a1986e7ee1f4
thread:pr:188613,pr #188613: [MPS] copy for contiguous same-dtype using a compute kernel,ready for maintainer decision,45,ready for maintainer decision,1,0,,188613,,,0,cbe2e3cdeca1212f5bc38ad2dda3e4f7622e841b7dad7d4c38ba7d7e9b903537
thread:pr:188625,pr #188625: binary-docker-build: add opt-in remote BuildKit (OSDC) mode,ready for maintainer decision,45,ready for maintainer decision,1,0,,188625,,,0,4e03cda2007a65bff9606bc8bd5626ecd36d1986d168cf42fba9f6d00387645a
thread:pr:188626,pr #188626: build-almalinux-images: migrate to OSDC via remote BuildKit,ready for maintainer decision,45,ready for maintainer decision,1,0,,188626,,,0,63f7ca1f1ee644c06235d6563732a63129d6775406484fc97dda55efdaf29643
thread:pr:188660,pr #188660: [Inductor] Enable mix-order reduction for wide reductions + tests,ready for maintainer decision,45,ready for maintainer decision,1,0,,188660,,,0,f74aac01d01f6abfac94f095cedfd2d842e34acc67a135d190e476ff3048eb84
thread:pr:188760,pr #188760: Guard user stream compile tests with xfailIfNoAcceleratorTriton,ready for maintainer decision,45,ready for maintainer decision,1,0,,188760,,,0,5493e01ba8f560d0d4a062f615931669d203efdefd110808e488eca79149e385
thread:pr:188761,pr #188761: MSVC compiler flag change for JIT cpp extensions on Windows,ready for maintainer decision,45,ready for maintainer decision,1,0,,188761,,,0,d045c1a52e9315658974bc562c81483475986a25e2e88c89e7f67cf8739fe79d
thread:pr:188802,pr #188802: [MPS] Migrate conv3d to metal kernels,ready for maintainer decision,45,ready for maintainer decision,1,0,,188802,,,0,3da2b96d8954a4b30e3f6f60d3f8b391a936ad2f46013116eac8e1b6a2628b9d
thread:pr:188804,pr #188804: [MPS] Add solo large-tensor allocator + torch.mps.set_allocator_settings(),ready for maintainer decision,45,ready for maintainer decision,1,0,,188804,,,0,aa7f244c124efcfdc714affee035622b5f940ecff80835f3a446f5e893f367db
thread:pr:188814,"pr #188814: Fix typos in comments and docstrings across ao, distributed, masked, nn, and sparse modules",ready for maintainer decision,45,ready for maintainer decision,1,0,,188814,,,0,cdbaa285cd6c573c915d61c4dbdb00acc0f065d234b17d246f86a026b66923b0
thread:pr:188823,pr #188823: [Inductor][Host-side TMA] Cache host-side TMA descriptors across launches,ready for maintainer decision,45,ready for maintainer decision,1,0,,188823,,,0,5c33ed93749669fd8bb0441d2ce0a3f9081d4641fea3aa7d03d94aeb22467545
thread:pr:188846,"pr #188846: Fix torch.pow docstring: scalar params are Number, not float",ready for maintainer decision,45,ready for maintainer decision,1,0,,188846,,,0,5111d2627d705387a8e08393aee0e14f530aef1d9adaa5073f9f46180ad49a41
thread:pr:188870,"pr #188870: Fix typos in comments, docstrings, and error messages",ready for maintainer decision,45,ready for maintainer decision,1,0,,188870,,,0,2e556e8b3f476e3c57c36f97b6ad9f43f10012d0555dc65805576e287003bd93
thread:pr:188874,pr #188874: Enable XPUGraph native recording mode,ready for maintainer decision,45,ready for maintainer decision,1,0,,188874,,,0,9a8e0213f8b9ca8f2eec8892fead647a13a79817981a9f89f498acd77e20b1a1
thread:pr:188885,pr #188885: [DataLoader] Clean up stale Py2 comments and pickle-error warning,ready for maintainer decision,45,ready for maintainer decision,1,0,,188885,,,0,b9e2062cccb27ae7f2d3d93f790e60a57c700356b2a141f4f16551b085a598bf
thread:pr:188903,pr #188903: [HOP] Adding input mutation support for switch (2/2),ready for maintainer decision,45,ready for maintainer decision,1,0,,188903,,,0,9be1a7d8ce18cd56540601fa5f6e3b9432f4bf6bdbcf2ea25caa6d5664d28c05
thread:pr:188944,pr #188944: [NCCL] try to remove the redundant copy in functional collectives,ready for maintainer decision,45,ready for maintainer decision,1,0,,188944,,,0,1702c686b8014846cf6bfe28a6e42fe3e1d579e8a8a23d61c13cd3951efc213e
thread:pr:188946,pr #188946: [Bugfix][FSDP] Capture out-of-place redistribute() return value in optimizer state reconstruction,ready for maintainer decision,45,ready for maintainer decision,1,0,,188946,,,0,eb8fc65da09729f3fc233bc17c310dd06c33989d2f4bb75c03dc8f5a988a0187
thread:pr:188947,pr #188947: [Bugfix][DTensor] Fix wrong shard offsets for uneven DTensor shards in TP-FSDP metadata,ready for maintainer decision,45,ready for maintainer decision,1,0,,188947,,,0,853f195258461f3733cf3974b7c46d7c0e1d1f13275ace968342fe0bcb9e348f
thread:pr:188973,"pr #188973: Reapply ""Avoid redundant scalar-to-tensor conversion in `real`/`imag`setters on complex tensors (#188603)",ready for maintainer decision,45,ready for maintainer decision,1,0,,188973,,,0,129dceb9e556e1f250cbc8d285e4c016c8ae4728c4a9ffaa066f15a1b3ca9935
thread:pr:188976,pr #188976: [HOP] Inductor lowering for switch (1/2),ready for maintainer decision,45,ready for maintainer decision,1,0,,188976,,,0,202dc8d0b9d355456d1d9b30e451eb167a4191b62ecc5838d6dc12017e07f184
thread:pr:188977,pr #188977: Fix typos in docstrings and comments across torch modules,ready for maintainer decision,45,ready for maintainer decision,1,0,,188977,,,0,d9ff855fcb741745716ccb9018d1efd76a93686a394e5fd332f57fb45112eec6
thread:pr:188991,pr #188991: [NVGEMM] Support `cutlass.operators`,ready for maintainer decision,45,ready for maintainer decision,1,0,,188991,,,0,cff8b0ba96590ff0b9008b47f0b913f3d83557aed46b1940e427da22277e3106
thread:pr:189011,pr #189011: include activation epilogue in GemmAndBiasParams::Signature(),ready for maintainer decision,45,ready for maintainer decision,1,0,,189011,,,0,166c3207e48485ee828ecdadb8cdb3495e17a39c9fd95695bbeaf561b832ce47
thread:pr:189028,pr #189028: [HOP] Moved branch clamping to user surface for switch,ready for maintainer decision,45,ready for maintainer decision,1,0,,189028,,,0,6978e0bacd5c3c164d26307b5e38aabd20fd4a3e2e9f40978c39ea33c6871f0f
thread:pr:189037,pr #189037: Preserve strides for CuTeDSL grouped MM fake tensors,ready for maintainer decision,45,ready for maintainer decision,1,0,,189037,,,0,b09a22d01a98075ab2512c2a4db6128430a4cf6eaeca8c45450adf690ad7cc40
thread:pr:189089,pr #189089: [fmha_fwd][rocm] Extend CK ck_tile FMHA seqlen-pad guard to the GQA swap path,ready for maintainer decision,45,ready for maintainer decision,1,0,,189089,,,0,3d0f8211bebe29cad736f9e6bb319aaabeb98d00cd8aa0fc405fe74878519aea
thread:pr:189100,pr #189100: [ROCm] Fix test_is_pinned_no_context for Python 3.14 forkserver default,ready for maintainer decision,45,ready for maintainer decision,1,0,,189100,,,0,a1f2b2ab28623144c07e9747adc9fe10ec93d115f9f3ae3cb910f025cbfc106e
thread:pr:189101,pr #189101: [inductor] Fix data-dependent GuardOnDataDependentSymNode in get_fill_order,ready for maintainer decision,45,ready for maintainer decision,1,0,,189101,,,0,7b2a302de103dd657d682ede8fb22a3e9b1610c4686a95a0a646dd4dfad9a157
thread:pr:189102,pr #189102: [TEST][Inductor][CUDA] Bump tolerances for `test_comprehensive_combinations_cuda_float16`,ready for maintainer decision,45,ready for maintainer decision,1,0,,189102,,,0,dd248537fc73561e42ddf31cacafc5200cab487ac4083fda340c59f1c49d4d9b
thread:pr:189108,pr #189108: [Inductor] Bound exact index expressions for int32 indexing eligibility,ready for maintainer decision,45,ready for maintainer decision,1,0,,189108,,,0,7cecbbbba0225acb658dbd33f075b5d2d31528d7deb95d4feade42bb7be2ab1a
thread:pr:189140,pr #189140: [Test] Update device type tests in TestVarlenAttention for device-generic,ready for maintainer decision,45,ready for maintainer decision,1,0,,189140,,,0,a1d940ebb54fec51986bdd1d28e06dffd03855c9bf4cdd9a5c6b51926a082656
thread:pr:189143,pr #189143: [Test] Make test_native_mha.py testcases device-agnostic,ready for maintainer decision,45,ready for maintainer decision,1,0,,189143,,,0,f1457b7610ca83a8e50eec9e2a77697a23697deea12ba8febd4843f3407bb601
thread:pr:189148,pr #189148: Add maybe_unused label to strides when strides is unused,ready for maintainer decision,45,ready for maintainer decision,1,0,,189148,,,0,770518f69f245bd6980c2a9edf1eae52ddfe76606dd503e17b607d237a948913
thread:pr:189155,pr #189155: Add CUDA Support for Fused `_add_relu` With Half and BFloat16,ready for maintainer decision,45,ready for maintainer decision,1,0,,189155,,,0,bf05f0cd5609f498b951fe5a2499f15233a3bc9b75530ac2f005c823ceb6d653
thread:pr:189160,pr #189160: Use SafePyObject in PyAnomalyMetadata and PySavedVariableHooks,ready for maintainer decision,45,ready for maintainer decision,1,0,,189160,,,1,9d9e1ba60d0226d08e8a1c4b04464bcac3128580d49cb400e956ee194618fccf
thread:pr:189163,pr #189163: Enable expandable_segments on Windows (port NVML driver API as prerequisite),ready for maintainer decision,45,ready for maintainer decision,1,0,,189163,,,0,b188dd3af6b47db1dbb24a638eed6a13a3f64970837e85abf58a39dce91db089
thread:pr:189177,pr #189177: [inductor] Drop unused cache_info from CacheCompiledArtifact._validate_and_unpack,ready for maintainer decision,45,ready for maintainer decision,1,0,,189177,,,0,17cb2f6f1daed5c2d858c256601633b0bb5a7d7ddfdba23b023a52653f3b1985
thread:pr:189182,pr #189182: make the return more specified,ready for maintainer decision,45,ready for maintainer decision,1,0,,189182,,,0,0fbc1fdc5a03810f0ffac72c11f652693d2361e7c0220efb9655dc4360d00c8f
thread:pr:189184,pr #189184: [inductor][fb] Whole-graph batch_addmm->bmm fusion for independent same-shape addmm (#189184),ready for maintainer decision,45,ready for maintainer decision,1,0,,189184,,,0,2655c26aaa32935d023fba7baa43efb8dfef336f58f855702407295ca66d520d
thread:pr:189199,pr #189199: [DTensor] Fix spurious cross-device FakeTensor mismatch in multi-threaded tests,ready for maintainer decision,45,ready for maintainer decision,1,0,,189199,,,0,85cb5c565a545661464184a51a8fb8cb38c1aad32afcb40e36e4f1b86fa69519
thread:pr:189251,pr #189251: [Dynamo] Store backend tags in the registry,ready for maintainer decision,45,ready for maintainer decision,1,0,,189251,,,0,52a010df7e86b60b1feb2d03d4a6f10ab9382c27e1745ab017fbc4970d971b65
thread:pr:189260,pr #189260: PERF: improve cuda/xpu Stream.__hash__,ready for maintainer decision,45,ready for maintainer decision,1,0,,189260,,,0,f1bb78a37a237022d0c34c22a0d2dcc50a3176375cdf5569480eb8b0bb2882d0
thread:pr:189274,pr #189274: [dynamo] Route set/dict-view/list dunder methods through tp_slots,ready for maintainer decision,45,ready for maintainer decision,1,0,,189274,,,0,fbb3daabc8ea243ee97d5b41de8f5165a976faf23958542f020d6a735d3538db
thread:pr:189303,"pr #189303: Reapply ""Dispatch index_put_ to fill_ for scalar values (#188267 / #189105)""",ready for maintainer decision,45,ready for maintainer decision,1,0,,189303,,,0,b1e1d4a5f765d8007489561283b4036cc5c84a54179acce9a524578c0e94c2b0
thread:issue:158698:issue:158710:issue:179128:issue:69431,issue #158698: [fsdp2] Possible to fuse reduce-scatter (in fp32) + div + dtype conversion (to bf16) in post-backward?,needs maintainer decision,44,needs maintainer decision,5,0,"69431, 92801, 158698, 158710, 179128",,,,4,42c00e960c2c43ac6a18cfd91d922b25953670f0d18ca52c7dc56c12a8e63766
thread:issue:145374:issue:154329:issue:155060:issue:164299,issue #145374: Memory Leak in MPS Backend During LSTM Iterations (Out of Memory Error),needs maintainer decision,44,needs maintainer decision,4,0,"145374, 154329, 155060, 164299",,,,4,5234aa30796d3ef9cd559f092aa2a1c7bacd663761fce0dd2029ddc4e8069793
thread:issue:184036:issue:187178:issue:187179:pr:187243,pr #187243: Reject p < 1 in triplet_margin_loss to prevent NaN from p-norm overflow,has linked issue,44,Review active PRs and linked issues as one maintainer work thread.,4,0,"184036, 187178, 187179",187243,,,4,e709325134bc775b7093bd353e779487ff83b14c9f772bade89c49719fe44d3a
thread:issue:109706:issue:146938:issue:80821,issue #109706: Make standard container classes satisfy container Protocols.,needs design decision,44,needs design decision,3,0,"80821, 109706, 146938",,,,2,d8db48914b1e939f1565847043b90a8d657296d73d41279193218276d4354cbf
thread:issue:188325:issue:188405:pr:189145,pr #189145: [Dynamo] Preserve Stream/Event subclass types in graph reconstruction,has linked issue,44,Review active PRs and linked issues as one maintainer work thread.,3,0,"188325, 188405",189145,,,4,4ae237a1e9f14efde53093497911202d41282f253888c2c4f4325e6de14bfd19
thread:issue:100347:issue:181725,issue #100347: nn.MultiheadAttention doesn't use efficient scaled_dot_product_attention,stale/low urgency,44,stale/low urgency,2,0,"100347, 181725",,,,1,467dede8dc872d25757f84ad8de42b05d1a031801374d137ca876a5b89e57505
thread:issue:102207:issue:108798,issue #102207: skipIfTorchInductor Tracking Issue,good volunteer slice,44,good volunteer slice,2,0,"102207, 108798",,,,1,14b42c97b2dfb618018a9fa7bf866270e4a3453188ca10773b762ee566b4a6fc
thread:issue:10454:issue:117122,issue #10454: [feature request] Rank-Revealing QR - Adding dgeqp3 support to torch.qr,stale/low urgency,44,stale/low urgency,2,0,"10454, 117122",,,,1,fd8e56d2c4935d17587d2e22f4a11e1d75b623f86a4d366f9258aba9524f7752
thread:issue:127078:issue:173881,issue #127078: Wrong Flag defined for ppc64le arch,needs maintainer decision,44,needs maintainer decision,2,0,"127078, 173881",,,,1,59429e8ba6a277e08dff638f5d54d48c0c5c0e3a5c011d57a3c812c2a5ea1eda
thread:issue:137635:issue:170168,issue #137635: [MPS] `torch.linspace` produces different results that CPU implementation for integral types,has linked PR,44,has linked PR,2,0,"137635, 170168",,,,1,68d5ccae6e97c700a82f63923148c742b1bd91d0063556d1b253b813e4148702
thread:issue:139019:issue:173912,issue #139019: Stabilizing the Kumaraswamy Distribution,stale/low urgency,44,stale/low urgency,2,0,"139019, 173912",,,,1,000fab4d25885c3517ee98a5868f75a4812bcb978b244746bc2cafbf854e813e
thread:issue:142228:issue:162588,issue #142228: Scaled Dot-Product Attention Invalid Configuration Error on Large batch size,has linked PR,44,has linked PR,2,0,"142228, 162588",,,,1,3c6864e68d54a032197eee8f65ed83415fb82eb87398395c4396f49a48a8eb1f
thread:issue:150199:pr:188795,pr #188795: Fix `torch.asarray` to preserve an input's device when a default device is set,has linked issue,44,Review active PRs and linked issues as one maintainer work thread.,2,0,150199,188795,,,2,fde3d6306dbb54c610a9d420703455652dbd0b84818231a6a3fa5667ef67a8cb
thread:issue:154356:issue:180377,"issue #154356: Q.size(-1) == m INTERNAL ASSERT FAILED at ""/pytorch/aten/src/ATen/native/BatchLinearAlgebra.cpp",needs maintainer decision,44,needs maintainer decision,2,0,"154356, 180377",,,,1,8010dad5497c3ddf00a5ee577b6d506e0dd12a279638ef8194557337bc184f4e
thread:issue:157431:issue:64947,issue #157431: ``torch.quantile`` edge case,stale/low urgency,44,stale/low urgency,2,0,"64947, 157431",,,,1,6c69e3dda66872146f8b864acaa681882042b7f879236296ae59feb7dd47e4ac
thread:issue:173382:issue:184084,issue #173382: Make torch.cuda.empty_cache() actually empty the GPU cache,needs design decision,44,needs design decision,2,0,"173382, 184084",,,,1,f842a6fe81f1949ec07674021fe8b0752be1ca3d1eca0c1fee63272a6cf4435e
thread:issue:175211:issue:177969,issue #175211: CUDA/ROCm/Accelerator testing should replace get_device_capability() with feature queries,has linked PR,44,has linked PR,2,0,"175211, 177969",,,,1,bb086dc9b940626c5e046004e1080cdc5b855c7eb5a6d24d6c689c6297161a29
thread:issue:175267:pr:184830,pr #184830: [Dynamo] Guard source-backed identity comparisons,has linked issue,44,Review active PRs and linked issues as one maintainer work thread.,2,0,175267,184830,,,1,593933604ead60c283ce16748cf0295e2ffdf25e82ab83a1b0703f4442d7417e
thread:issue:175408:pr:184809,pr #184809: Improve custom op Tensor subclass fake error,has linked issue,44,Review active PRs and linked issues as one maintainer work thread.,2,0,175408,184809,,,1,5cfaa1f517b5030d4c108ebc06bf61cc134be9f376a044d0937236aa97f6db4b
thread:issue:176296:issue:176298,issue #176296: [MPS] Binary ops on uint16/uint32/uint64 produce garbage values,has linked PR,44,has linked PR,2,0,"176296, 176298",,,,1,54cc3c52a064800aa2b272ff0b3a11b3b002082afaf8c8c87d5ae541df47648e
thread:issue:177827:issue:182277,issue #177827: Internal Assert Failure in `torch.cuda.memory.caching_allocator_alloc` with negative allocation size,has linked PR,44,has linked PR,2,0,"177827, 182277",,,,1,c87f9cb7452acb2d8940cbad019b142a3195f7aaa11fcfe78a527729e53cea44
thread:issue:178593:issue:187338,issue #178593: torch.gcd returns incorrect results due to automatic dtype casting between signed and unsigned integers,needs maintainer decision,44,needs maintainer decision,2,0,"178593, 187338",,,,1,9c2c5acfb5e99c005cdc69c692fb699be9f94f8a1eaad15a82daed82f26f8ee6
thread:issue:178977:pr:182273,pr #182273: [c10d] Fix device_id not propagated through ProcessGroupWrapper,has linked issue,44,Review active PRs and linked issues as one maintainer work thread.,2,0,178977,182273,,,2,2d790e8e6a0d6d47b2ea2bb2af436e986c977133428c658a71f125564d315469
thread:issue:180762:issue:44279,issue #180762: torch.utils.cpp_extension.load should support CUDA device linking (-dlink) similar to CUDAExtension,has linked PR,44,has linked PR,2,0,"44279, 180762",,,,1,4fbb0e2e7b7a69acebb47d0115ce98ddb5ecec8d7c2f0ec350aca68bd3ebbd61
thread:issue:182817:issue:183621,issue #182817: [MPS] Switch test_output_match / test_output_grad_match to magnitude-floored relative tolerance,needs maintainer decision,44,needs maintainer decision,2,0,"182817, 183621",,,,1,e3cb1871406f381fd79d71d0f963db5113d688efd6ed9a401c47cff5cec7f8e4
thread:issue:184361:issue:185780,issue #184361: [pipelining] PP fails when trying to pipeline any HF transformers model that uses RoPE,needs maintainer decision,44,needs maintainer decision,2,0,"184361, 185780",,,,1,9fb3d7a1b35851a3d5569a57ffe5f5db37da41f6f9fdc490765fe7ce501f4f06
thread:issue:184575:pr:188066,pr #188066: test: add F.normalize zero-input correctness tests (fixes #184575),has linked issue,44,Review active PRs and linked issues as one maintainer work thread.,2,0,184575,188066,,,1,95943abd0c065221e18c6fc5d97df9d619f31e71c020629e6552ccc7f939ddf4
thread:issue:184842:pr:185054,pr #185054: [autograd] Validate forward-mode jacobian/hessian require vectorize=T…,has linked issue,44,Review active PRs and linked issues as one maintainer work thread.,2,0,184842,185054,,,2,8a8c19511c0b780ddf24af416177ab9f441b54902d428023b97c2aa4a110ce2b
thread:issue:184845:pr:185751,"pr #185751: [nn] Raise ValueError early for invalid (ndim, pad_size) in non-constant F.pad modes",has linked issue,44,Review active PRs and linked issues as one maintainer work thread.,2,0,184845,185751,,,2,841ca637cc18e051d8ebf726b4588e5d8b02f57d3b829535c64b8ba6fad8df21
thread:issue:186824:issue:186826,issue #186824: torch.distributions.Cauchy.icdf(0) returns large finite value with wrong sign in float32; Gumbel.icdf(0/1) also finite instead of ±inf,has linked PR,44,has linked PR,2,0,"186824, 186826",,,,1,c9c5c4b28a4cd37796d289dbd59e590bc78765bdb175fcc9a792a3f19e4c1210
thread:issue:187093:pr:189127,pr #189127: [inductor] Decompose broadcast-bias baddbmm before pointwise,has linked issue,44,Review active PRs and linked issues as one maintainer work thread.,2,0,187093,189127,,,4,50f071f9c4178d6f9eaef45d3e3e546e89a8aa4f6c47dd668375b07c3bb83c38
thread:issue:187177:pr:187242,pr #187242: Validate LocalResponseNorm parameters to prevent silent inf output,has linked issue,44,Review active PRs and linked issues as one maintainer work thread.,2,0,187177,187242,,,2,fe49ae162db2b9ae26a5438262b37436f8aa8f239784d65db51d62e75fa94a34
thread:issue:187759:pr:188053,pr #188053: test: add svdvals NaN propagation test (fixes #187759),has linked issue,44,Review active PRs and linked issues as one maintainer work thread.,2,0,187759,188053,,,2,ed0a540a250a1a77c14d1a5224df0a9171ea197ff1290afdeca7c6eb26030bc8
thread:issue:188718:pr:188719,pr #188719: Fix Inductor KeyError on view_as_complex of non-contiguous CUDA tensors,has linked issue,44,Review active PRs and linked issues as one maintainer work thread.,2,0,188718,188719,,,1,658a70a902db054103dec6cd20d24c7c37788018c1bcd02cf4c49760fbceb417
thread:issue:189118:pr:189119,pr #189119: Fix test-jsons artifact upload path,has linked issue,44,Review active PRs and linked issues as one maintainer work thread.,2,0,189118,189119,,,2,cf2e9916c13189682570d91cdc5a03ec383fbcda66cf9dc17ee1378af600f617
thread:issue:189157:pr:189158,pr #189158: Skip GHA stats artifacts from other workflow attempts,has linked issue,44,Review active PRs and linked issues as one maintainer work thread.,2,0,189157,189158,,,2,28e60907dc7136081eddf65eb605ac49497d73c9f32cead0ab8eabd9bb56053c
thread:issue:189271:pr:189288,pr #189288: [inductor] Fix compute_ancestors KeyError from self-referential mutation copy,has linked issue,44,Review active PRs and linked issues as one maintainer work thread.,2,0,189271,189288,,,1,99e28ac7bb0fa57ba48e10e089e042b21e3edf622a4bad4dd48bf076050e7dbf
thread:issue:3867:issue:80301,"issue #3867: [Feature Request] Implement ""same"" padding for convolution operations?",has linked PR,44,has linked PR,2,0,"3867, 80301",,,,1,47d95143734998d2a55cdba4f8fa09730622d0875d0955bed43d018021da0140
thread:issue:85852:pr:184094,pr #184094: Clone aliased mm/addmm inputs on CPU to avoid in-place gemm,has linked issue,44,Review active PRs and linked issues as one maintainer work thread.,2,0,85852,184094,,,2,1731f892cc63793e365d99e8d6e95d0eb6bc2e67ac9c886bdae3c1ddbd4e8160
thread:issue:112044,issue #112044: DDE in pack_padded_sequence/pad_packed_sequence decomposition.,has linked PR,44,has linked PR,1,0,112044,,,,0,1129c95ddf2dd7c5138016e63429af98b4e5e2081a3b30b664c589b526c9ac9c
thread:issue:116203,issue #116203: Refactor precision handling in test/test_optim.py,has linked PR,44,has linked PR,1,0,116203,,,,0,19496ece329bc28db6a5abaebf10c77fc42cd40f445a1588d49aee681baf4d61
thread:issue:116465,issue #116465: torch.cartesian_prod inconsitent return shape for one input tensor,has linked PR,44,has linked PR,1,0,116465,,,,0,d1f112f40ce1990adf815b43713eaf7bad2cfe76c8451ee9cfb65335c35532b7
thread:issue:123339,"issue #123339: Inconsistency: BCELoss asserts target in [0, 1], BCEWithLogitsLoss does not",has linked PR,44,has linked PR,1,0,123339,,,,0,db0ebe14aebfc0da72d7029f76cbe29576d259dbddecd9bd818622f60a52e4a4
thread:issue:128712,issue #128712: Add Swiglu activation function,has linked PR,44,has linked PR,1,0,128712,,,,0,09d7fe58d85ce8c1fbdf1cfd621211cb432299dfffbe988203b923d778ad040b
thread:issue:129020,issue #129020: Expand Tag Set: views & reductions,has linked PR,44,has linked PR,1,0,129020,,,,0,6981f6e9ca0e13ecef7aad7e626e73bede3cff328e2ec97795a6466407a2d161
thread:issue:131765,issue #131765: [typing] Add missing `__all__` to modules.,has linked PR,44,has linked PR,1,0,131765,,,,0,86149116f48505f2f2d8d5755aef36e66c09728702f1c790a8ece2fd360d917e
thread:issue:132896,issue #132896: AOTAutograd.process_inputs doesn't work on nested subclasses,has linked PR,44,has linked PR,1,0,132896,,,,0,e450056039980f6755af87edb53129fa6a72e3aba51c29737e5072b34a1b0b13
thread:issue:133254,issue #133254: Shared memory out of resource when using flex attention,has linked PR,44,has linked PR,1,0,133254,,,,0,fcff09adb8371043a3b1bdfc9b801cd5424628b893663db6ce1d576ca5257788
thread:issue:133512,issue #133512: functionalization doesn't faithfully reproduce strides of intermediate values,has linked PR,44,has linked PR,1,0,133512,,,,0,318766c7061496c4a94af54842441ec8bca38cee3c571dddd4ea638885475b0a
thread:issue:135011,"issue #135011: dynamo (re)compilation issues: shape (1,1), nn.Parameter, mark_dynamic",has linked PR,44,has linked PR,1,0,135011,,,,0,1ef20aa304b307b9414721973ca9c6ee2e814ba769365198877e4000e53c93fc
thread:issue:138795,issue #138795: compile + allgather with group will fail for stack-style allgather,has linked PR,44,has linked PR,1,0,138795,,,,0,5fa7b5c9250c6d4fd03369aa9eb0f06d9fee36e9ce92f3f81bbfc7df2e74a6c4
thread:issue:140000,issue #140000: torch.argmax slow on CPU,has linked PR,44,has linked PR,1,0,140000,,,,0,da22c4fc97b7f6d1535c2d401d559a5f382ba7cc5033ee82c03e115db6c39f3e
thread:issue:140497,issue #140497: C++ standard library functions not found when compiling c10d debug handlers,has linked PR,44,has linked PR,1,0,140497,,,,0,4bad268f48ab1d68e107c6720ff7d870020f54958108cf8d505ca1dca58c1d9c
thread:issue:141595,issue #141595: FX graph cache does not do provenance tracking for inductor guards,has linked PR,44,has linked PR,1,0,141595,,,,0,6a5fcee18eae90e055b48d5a1866286177680fa217075429d355c8429c2e67eb
thread:issue:144580,issue #144580: `torch._foreach_mul` does not support autograd,has linked PR,44,has linked PR,1,0,144580,,,,0,9e7dae29ce3a9685602376587468c466599ea85619b1eee6bad80169b80a9234
thread:issue:145208,issue #145208: Some FlexAttention learned bias bugs/limitations,has linked PR,44,has linked PR,1,0,145208,,,,0,a33c12fe1e2403cdf417e4870577f1c15c35653b034c79698c2232dea92dc63d
thread:issue:146249,issue #146249: Replace `*args : Any` with `typing_extensions.TypeVarTuple`,has linked PR,44,has linked PR,1,0,146249,,,,0,4a5dcd04c98fd245c330423574b3e1e288d0408bec1bf267315f2e112b9e7ea4
thread:issue:146306,issue #146306: torch.compile failed on single-node multigpu setting.,has linked PR,44,has linked PR,1,0,146306,,,,0,7a211907ef6237f847019cffacf8ed71db0f96ae47537a667ff3fb256571f6ee
thread:issue:146508,issue #146508: Something very wrong with float16 CPU implementation,has linked PR,44,has linked PR,1,0,146508,,,,0,b1c4295cafeaf7dd55716dfd60b8ee0f90efbd7f6755da21545d8b869cfec95c
thread:issue:147075,issue #147075: [inductor] Performance Degradation and Hang in `torch.diff`,has linked PR,44,has linked PR,1,0,147075,,,,0,4794eb25ea7f7b7d3fb49919a6f23ee9ff9a718e0ca06191cab523424058de11
thread:issue:151027,issue #151027: Add size/strides/alignment assertions for cpp_wrapper,has linked PR,44,has linked PR,1,0,151027,,,,0,c2e2a3ae95626cb4e9206bfbafa8e861b62e9d23f95ec45a502a59f73dad816b
thread:issue:153485,issue #153485: Double backward error in Pytorch PP,has linked PR,44,has linked PR,1,0,153485,,,,0,19fe088aa2f9c9a8b002c4abb7768e581e8203bc5a14469bade9d81af2596d90
thread:issue:153825,issue #153825: 'torch.sparse.to_sparse_semi_structured' significantly worsens performance on H100 GPUs,has linked PR,44,has linked PR,1,0,153825,,,,0,32b6b6646079335bbb3745d3b920ece5670b7deaeb3195e355836255da917dc8
thread:issue:161765,issue #161765: Add torch compile Dynamo support for protobuf,has linked PR,44,has linked PR,1,0,161765,,,,0,82357846416457124923ab3b37b7fb46463147b48a43df9ad6625d9892c78756
thread:issue:161863,issue #161863: torch.testing.assert_close doesn't accept tensor-likes,has linked PR,44,has linked PR,1,0,161863,,,,0,9e41f7b10ad2e7755c0b79e5c7c4ea778c9c3ebc88c29c79154a3cac02d641a2
thread:issue:165650,"issue #165650: `torch.remainder` with integer divisor 0 should raise, but returns large values in CUDA",has linked PR,44,has linked PR,1,0,165650,,,,0,81b26abd4e9045e505f7f0cba0f5174137f61b5df5e862e744d2c010237ffbfb
thread:issue:167223,issue #167223: [feature request] Re-enable `__torch_function__` dispatch when invoking autograd hooks for Tensor subclasses,has linked PR,44,has linked PR,1,0,167223,,,,0,d6cde7d989115a0d2de559a644c1d1ccae3d20f239eb97dc98ce603d9aa6b577
thread:issue:167818,issue #167818: undefined symbol for `at::meta::_index_put_impl_` when running or compiling executable on my own torch-related project.,has linked PR,44,has linked PR,1,0,167818,,,,0,38527c7f2b4a3989735cc0fcd5837f132f16a7f9bd50fa8ba15a71e0e3fd576e
thread:issue:168972,issue #168972: Aarch64 unit test failures - test/test_cpu_repro.py CPUReproTests.test_float32_to_uint8,has linked PR,44,has linked PR,1,0,168972,,,,0,e6f82fdb9b82cdfb80f50f7dc882fa7ad2e69d388310ac33f04ba06d15cfb30d
thread:issue:170367,issue #170367: DTensor redistributes on foreach_X operators are inefficient,has linked PR,44,has linked PR,1,0,170367,,,,0,1a1b0a75f24d5729b29187537891da5f70e283b9ae7b76abbe813c7b794ac684
thread:issue:170737,issue #170737: Expect to open the is_privateuseone branch in the LSTM interface of PyTorch's RNN.cpp to develop corresponding LSTM functionality on NPU.,has linked PR,44,has linked PR,1,0,170737,,,,0,7bfd240d9959d5cc20fa841222aef6d5573c775fd69c7fa9203ac61ee336e171
thread:issue:170798,issue #170798: nn.functional.pad silently changes the padding value if input dtype is int64 and padding value is larger than 2^53,has linked PR,44,has linked PR,1,0,170798,,,,0,d68e30d6bf1f5a151dea4b52a54f4d1171976ecae3c0b528e79f9ceb96b8259d
thread:issue:170869,issue #170869: Feedback about torch.addcdiv,has linked PR,44,has linked PR,1,0,170869,,,,0,26fd4dd05285d66e346b32829bc8e7897184ac9d8e66123c28a4eea1693ff76a
thread:issue:171249,issue #171249: [Bug] softplus returns inf for large beta while other frameworks return finite value,has linked PR,44,has linked PR,1,0,171249,,,,0,c744c087169a67c79d17d0241d2f97e5897df508265a5d6416112e48ff24962c
thread:issue:171355,issue #171355: torch.utils.checkpoint.CheckpointError with qwen3 MoE,has linked PR,44,has linked PR,1,0,171355,,,,0,96c4f39409550c8076c566b174cfceaca765344e9cd940b3315cf276a67309b6
thread:issue:172321,issue #172321: [OpenReg] Extend test_autograd.py coverage,has linked PR,44,has linked PR,1,0,172321,,,,0,7917ece9226390b651c6c6736d1f6a96e74132f84922361b88862fb56cc87c87
thread:issue:172324,issue #172324: [OpenReg] Add documentation for standardized testing patterns.,has linked PR,44,has linked PR,1,0,172324,,,,0,0fabfcdb4a8398c9d3c40d19cc7c277c6e28d549e9ccdfb694e7aef0f41e7d0b
thread:issue:172538,issue #172538: Support native structured kernel and out_as_primary: true for PrivateUse1,has linked PR,44,has linked PR,1,0,172538,,,,0,ace8981fd9af070b6af739786cd52534a9070cbea82012cadaca5d194b98c7f0
thread:issue:173049,issue #173049: OOM error suggests expandable_segments even when enabled,has linked PR,44,has linked PR,1,0,173049,,,,0,4396e976d3c04f465827e74220f88dbaa2c754d247cd3707c631bbf22234e0d3
thread:issue:173232,issue #173232: error C2872: 'std': ambiguous symbol when compiling `csrc/dynamo/compiled_autograd.h`,has linked PR,44,has linked PR,1,0,173232,,,,0,e452d8243674928745b4c19ab63e84a78980f61235036eafc832c8459b0f160e
thread:issue:174017,issue #174017: Custom autograd Functions are not aware of partial backward pass execution,has linked PR,44,has linked PR,1,0,174017,,,,0,2e78f3dcd6942ac5c3e56963cfea190eee494ef88bd213a862a37b872859e3bb
thread:issue:174177,issue #174177: [test] Add error_inputs for nn.Linear module,has linked PR,44,has linked PR,1,0,174177,,,,0,1182b16e7baa19437c3a252ca01198e49ec31ea7f816f5d73265fbaab73d7e42
thread:issue:174625,issue #174625: document `torch.cuda.memory.mem_get_info` in,has linked PR,44,has linked PR,1,0,174625,,,,0,14b427e4fa5f5f4a0e4ef8440531f099ce7c6af0cb171ea1b26b9cbfd704af6e
thread:issue:174668,issue #174668: torch.histc is not consistent,has linked PR,44,has linked PR,1,0,174668,,,,0,07b1cd37e99be7f8151aa386a62274c1e29625cb88f89817366a5a09027027f2
thread:issue:174682,issue #174682: spin lint doesn't regenerate cuda headers,has linked PR,44,has linked PR,1,0,174682,,,,0,280aae003a4723e5c0dadbcffc93b6a66ced42300980068c34aebd0574813033
thread:issue:174984,issue #174984: `torch.fft.rfft` errors out with MKL FFT error when given an empty tensor,has linked PR,44,has linked PR,1,0,174984,,,,0,ee80d5fdd4bdadbf7169bb4ebe2ea0b4e53fc515a74cf284bf331a59c74b5a0f
thread:issue:175001,issue #175001: torch.cond() and cuda graph mega issue,has linked PR,44,has linked PR,1,0,175001,,,,0,7da81e0bab9d3e71e7977f7c7fc593d344e91489dfdea15e3d06c9409fe431f4
thread:issue:175324,issue #175324: Should `overrides.is_tensor_like` return a `TypeGuard[Tensor]`?,has linked PR,44,has linked PR,1,0,175324,,,,0,af1df7aeeac1c56964fd8bd3bdb234560765aa220694519bd4815f74728c884b
thread:issue:175368,issue #175368: `F.embedding_bag` segfaults when intermediate offsets exceed indices length,has linked PR,44,has linked PR,1,0,175368,,,,0,16f8e0a6b91f4df89dd4cfb59aeacaa56c2c99f3af8c713d6d4f907c3a2f478d
thread:issue:175489,"issue #175489: CPU `F.interpolate(..., antialias=True)` raises NotImplementedError for bfloat16/float16 (bilinear/bicubic)",has linked PR,44,has linked PR,1,0,175489,,,,0,8608856820be5b0b945a8567f93d29757d9ce3cb27f98004237e3b63e7d6e726
thread:issue:175521,issue #175521: torch.vmap working with non-tensor leaves in output pytrees,has linked PR,44,has linked PR,1,0,175521,,,,0,87ad4574d36c02b8cd80e56cbf85230edced288616431f499b5da52e7d558797
thread:issue:175522,issue #175522: while_loop vmap compatibility,has linked PR,44,has linked PR,1,0,175522,,,,0,cf3aae44c0077c4556747bf44fdd56debd6544b73d58ffc418bfed530a9e8f05
thread:issue:175994,issue #175994: CI: Dead CentOS code in docker setup scripts,has linked PR,44,has linked PR,1,0,175994,,,,0,5ab4f121dc137892a13489221038a4d6c69e96a880a62b411b2ff8c7f024caa0
thread:issue:176244,issue #176244: Bug: gaussian_nll_loss — possible issue with broadcastable variance shapes having multiple unit dimensions,has linked PR,44,has linked PR,1,0,176244,,,,0,f1284e4940c272609c93cc2d53696a2648f236649effb71b43a10b34476c5192
thread:issue:176381,issue #176381: Output difference anomaly under equivalent migration in RReLU operator,has linked PR,44,has linked PR,1,0,176381,,,,0,6d95db4c7803e744574124f41749c29d9076fc89e009404a1d05073aa72071e9
thread:issue:177109,issue #177109: [RFC] Make torch.library.custom_op faster in the common case,has linked PR,44,has linked PR,1,0,177109,,,,0,417b8bb73369b8e15975a895401707d64c8b253133e47e27aef618dcc949c87b
thread:issue:177127,issue #177127: torch.multinomial: add flag to skip input validation (66% of GPU time),has linked PR,44,has linked PR,1,0,177127,,,,0,76b96dbe4ddeaee4573486fa4bb86d87e3b20795ed1f088b54dcab7ef20f7037
thread:issue:177237,issue #177237: torch.conv1d incorrectly accepts invalid weight shape in meta backend while raising correct error in eager execution,has linked PR,44,has linked PR,1,0,177237,,,,0,9686695a04dcf3a947377dc3a7fac20cc146da099ec0ee5d2772be7a783961cf
thread:issue:177250,issue #177250: AArch64 Unit Test Failure - TestNN test_upsampling_bfloat16,has linked PR,44,has linked PR,1,0,177250,,,,0,8025dfd2207be14bd221901da61f3b2cedb3dec4725349295c128a2e6b73ac76
thread:issue:177265,issue #177265: Compiling CppExtensions in parallel causes race condition,has linked PR,44,has linked PR,1,0,177265,,,,0,0c378c8d338ac9c756581280ea632d67c3b27c8dff05532c019b3633fcde22de
thread:issue:177968,issue #177968: Add custom precision or tolerance support for out of tree backends,has linked PR,44,has linked PR,1,0,177968,,,,0,0ca2cc90f00100d17487db9d077aa2e8227bc847df5d96b029bce73281c74f1d
thread:issue:178125,issue #178125: conv_transpose1d meta implementation allows invalid output_padding that real CUDA execution rejects,has linked PR,44,has linked PR,1,0,178125,,,,0,05ce0216066bfa3192ddad827c7a73967f31cf2a946b0601377b4abbe4457f79
thread:issue:178128,issue #178128: conv_transpose3d meta execution produces output for invalid bias shape instead of matching eager validation,has linked PR,44,has linked PR,1,0,178128,,,,0,d94d8f49bab64fbcac6447cac8639055016d2ee5904a541dba2fd5e7cb499117
thread:issue:179259,issue #179259: test_torch.py does not run any tests on MPS despite number of `skipIfMPS` decorators,has linked PR,44,has linked PR,1,0,179259,,,,0,d0692cfabdde435a32ec6552f29eede02722e62d1da9514ac4479e45b948fd9c
thread:issue:179272,issue #179272: `torch.compile` returns different output stride/contiguity from eager for upsample_nearest3d on non-contiguous input,has linked PR,44,has linked PR,1,0,179272,,,,0,e7bef232139e74e8bf5916b1af439f860e101413f591be3edb977cfbc484f3b7
thread:issue:179608,issue #179608: [MPS] avg_pool1d produces negative values from non-negative input when zeros follow large floats,has linked PR,44,has linked PR,1,0,179608,,,,0,a90065a415c59ac5b4b9a1dd6f47c11bedcb0fdeb3a082e688f5e8f827e6e2cb
thread:issue:180119,issue #180119: Feedback about torch.Tensor.index_add_,has linked PR,44,has linked PR,1,0,180119,,,,0,2ae34c9af8c8a8bf3e30a19ca8bedce6e2eb6a0fb0fff77d5fa623e78a63f190
thread:issue:180346,issue #180346: [DTensor] clip_grad_norm_ fails when per-parameter norms come from different DeviceMesh objects,has linked PR,44,has linked PR,1,0,180346,,,,0,be600013cb1c659e39446f231d158966e87100ecb0604b368dbffe771e53f1c8
thread:issue:180467,issue #180467: remove TORCHDYNAMO_DYNAMIC_SHAPES=1 from docs,has linked PR,44,has linked PR,1,0,180467,,,,0,e0ed87f01e27eba5e8436df99a002a0a5d123155393be8f1cd8ab3599241255e
thread:issue:180649,issue #180649: torch.linalg._powsum docstring advertises out= but it is not implemented,has linked PR,44,has linked PR,1,0,180649,,,,0,c9c062d6aa04ccf07a7dad27fef831812725616cdbbc5ed4869755b731f1ffd7
thread:issue:180823,"issue #180823: [Inductor] RuntimeError: weakref to UntypedStorage when compiling transpose(-2, -1) under inference_mode",has linked PR,44,has linked PR,1,0,180823,,,,0,ebc177281932ac8a6cccdf9d0b17aaa7541809df5f7188f6b6e9d9b75639fba7
thread:issue:180855,"issue #180855: `""index_copy_cpu""` not implemented for `'Float8_e5m2'`",has linked PR,44,has linked PR,1,0,180855,,,,0,72a443094448142987534308191dd22550ebb534c4b9af096696624f131ca6a3
thread:issue:180957,issue #180957: Missing CSE in TorchInductor,has linked PR,44,has linked PR,1,0,180957,,,,0,35557b48730637af29a339e81bffd30d920b912e67d631edaf023b128704605d
thread:issue:181565,issue #181565: [C++ API][Packaging] nlohmann/json.hpp is exposed again by public headers in PyTorch 2.10 wheels,has linked PR,44,has linked PR,1,0,181565,,,,0,a69a9686590ea9456c9bfc06353db96b1bf7aad9bd4d9cd17fb3993b809785d9
thread:issue:181576,issue #181576: torch.meshgrid raises RuntimeError if no arrays passed,has linked PR,44,has linked PR,1,0,181576,,,,0,6379a070a95b7c0dda7eb2f543e4f99046d457ed6512a6e406931d261c6018a8
thread:issue:181805,issue #181805: torch.fmax with uint8 out produces inconsistent overflow casting on CPU and CUDA,has linked PR,44,has linked PR,1,0,181805,,,,0,a9f93553f0c7498c87f2d231f4c2988f4832ff0b7f7b833f7ccd5df58e78f98f
thread:issue:181806,issue #181806: CUDA torch.signbit returns false for negative float16 NaN,has linked PR,44,has linked PR,1,0,181806,,,,0,3fef296f19af7f6c1e16a5854e6b58eb219602e221f777313e6a1c4f4519dadd
thread:issue:182318,issue #182318: Dynamo codegen leaks torch function enabled state via _EnableTorchFunction destructor,has linked PR,44,has linked PR,1,0,182318,,,,0,fa0823213c4794de2f8eeac4d0bd7bca15eeb783cb77e334de11e078e5d4b7aa
thread:issue:182430,issue #182430: Inductor generates redundant upcast_to_fp32,has linked PR,44,has linked PR,1,0,182430,,,,0,f45dfc760b8b98333ddc66e931e71404e47614efa215829cdb4618f2fbc5acf4
thread:issue:182706,issue #182706: Dynamo is leaking internal state during optimizer tracing,has linked PR,44,has linked PR,1,0,182706,,,,0,b9ced6a94c98ab692b2630d179c4dc6c4c21713ddfc164b0e09bbd38fbf9ebde
thread:issue:182928,issue #182928: [Inductor] Filter extremely bad triton configs by device information for triton heuristics,has linked PR,44,has linked PR,1,0,182928,,,,0,3ea038b5454e0acf5b298ac20dc1be26c24c089cf5546d17b9da3240b736daee
thread:issue:183899,issue #183899: torch.compile fails on addmm with beta=0 and non-broadcastable bias shape,has linked PR,44,has linked PR,1,0,183899,,,,0,acef310afda162d2ec375108bdb059ae1db15a10354ed0e9b5eaaeb95f272a2e
thread:issue:184031,issue #184031: torch.nn.functional.gumbel_softmax returns NaN when tau=0 (division by zero),has linked PR,44,has linked PR,1,0,184031,,,,0,b06461dc1f36cad8f1b7727b0c32b9b94d0ef87872fffaab119e0c004927bd1d
thread:issue:184037,issue #184037: torch.nn.LPPool2d returns NaN for non-integer norm_type on signed input (x**p instead of |x|**p),has linked PR,44,has linked PR,1,0,184037,,,,0,dfe3185e9dea79683dbe497793ce6f267aac6856be0245033f978a71d361a85e
thread:issue:184219,issue #184219: serialization_id is not deterministic across machines,has linked PR,44,has linked PR,1,0,184219,,,,0,4d44b5f0ad0c6d4536ebb76995e79eebfbcce0f233a7408b336fffd5e6adcbeb
thread:issue:184282,issue #184282: LazyBatchNorm2d complex64 input has inconsistent CPU and CUDA errors,has linked PR,44,has linked PR,1,0,184282,,,,0,854a5a2d2818402e47659097c346f315925481d19b5162072f23dafac158d65d
thread:issue:184572,issue #184572: `torch.maximum` / `torch.clamp_min` gradients at tie point violate PyTorch's documented subgradient convention and are inconsistent with `F.relu`,has linked PR,44,has linked PR,1,0,184572,,,,0,8f8c1236417bcab96e9c410435b2ddbb1e35e9d613d7370624da1f71b80eaa01
thread:issue:184642,issue #184642: `Module.to()` has no overload for `.to(memory_format=torch.channels_last)`,has linked PR,44,has linked PR,1,0,184642,,,,0,728a517370cb3c26ed3c117afe1693a802d32b2949f1cee9d315b9514372ddae
thread:issue:184783,issue #184783: `set_multithreading_enabled` still spawns autograd threads,has linked PR,44,has linked PR,1,0,184783,,,,0,5b8bc492cd5fb524c0bbe235d5903c3605442f6155f03af5f7074a56164f0516
thread:issue:184890,issue #184890: vmap of autograd.grad fails on the output of model run on AMP,has linked PR,44,has linked PR,1,0,184890,,,,0,43b130513a92d1934fc0a9894a819c4341d888cb5fb628c925b760d001dc1d3f
thread:issue:185006,issue #185006: ConstantLR accepts factor=0 and crashes with ZeroDivisionError after total_iters steps,has linked PR,44,has linked PR,1,0,185006,,,,0,bec2110ad485eeb69ff5b3303f411a23d12fe65f11d7eb20d603610d571e0ee8
thread:issue:185022,issue #185022: ModuleList not recognized as Iterable by type checkers,has linked PR,44,has linked PR,1,0,185022,,,,0,5fb177fb827abbd7e7de3eb77e97323b6afa8362bc79f2187be2e91ef964bf80
thread:issue:185023,issue #185023: register_buffer-created attributes and Parameter.data incorrectly typed as Tensor | Module,has linked PR,44,has linked PR,1,0,185023,,,,0,f59cef1ba32daff49f4e1520a9a81566c55acbeaa2701f592160b57b183c7949
thread:issue:185046,issue #185046: [Bug] `torch.library._register_fake` crashes when any unrelated `LazyLoader` module in `sys.modules` fails to import,has linked PR,44,has linked PR,1,0,185046,,,,0,cb21a1f99e8dcdaf619f7b3264bf11f1ae86614d2edc47ad5adf89a6ff5a8eea
thread:issue:185216,issue #185216: `torch.ops.symm_mem.one_shot_all_reduce` causes `CUDA_ERROR_ILLEGAL_ADDRESS`,has linked PR,44,has linked PR,1,0,185216,,,,0,3db99ba9d75f183e19b906280c165f3b7d52ea55fcadcb3e5f4a805706819aa7
thread:issue:185246,issue #185246: torch.compile silently accepts invalid Bernoulli probabilities that eager mode rejects,has linked PR,44,has linked PR,1,0,185246,,,,0,a8ef8f21f615dba8aaf04ab2c3f47c29aaeb692e1a24944a4b0d799aa9bba69b
thread:issue:185248,issue #185248: `torch.compile` silently accepts negative standard deviations in `torch.normal` that eager mode rejects,has linked PR,44,has linked PR,1,0,185248,,,,0,efc30a0ee575713c7809b1f31c09f77a166c15bbccb785a1b0a244dccdac7d9c
thread:issue:185296,issue #185296: [AOTAutograd] [DTensor] subclass output SymInt arity mismatch causes partitioner assertion,has linked PR,44,has linked PR,1,0,185296,,,,0,2b7184bceb3dabf9c685174e8b9817eb65a1cb910ba5960d969bdc332ae29fe7
thread:issue:185331,"issue #185331: [pipelining] Pipelines reuse recv buffer tensors and directly pass them to the stage, causing backward hooks on stage boundary activations to persist between steps",has linked PR,44,has linked PR,1,0,185331,,,,0,9d898f442fe6dcc9373771307e55793d2baeb0573eb87293097f7eda5f0683e2
thread:issue:185464,issue #185464: torch.compile produces wrong results for F.multilabel_margin_loss when target contains -1,has linked PR,44,has linked PR,1,0,185464,,,,0,181dcbd9efb857dd899cc3a496848e31b48fcf80f4c26c3d28f24c563ef301d9
thread:issue:185481,issue #185481: [inductor] argmax over comparison mask is accepted and used for indexing while eager raises,has linked PR,44,has linked PR,1,0,185481,,,,0,6877bf26fc1f67247131374f3be35bbfde08181e58c2bea17d102b15453ac227
thread:issue:185533,"issue #185533: torch.compile(mode=""max-autotune"") fails with NotImplementedError: SliceView in get_stride()",has linked PR,44,has linked PR,1,0,185533,,,,0,eb706c408665c7a0286f249000c0326c26afd99f684a783ae15d49ac9314794b
thread:issue:185543,issue #185543: [inductor] Gradient mismatch between eager and inductor for torch.quantile on tied values,has linked PR,44,has linked PR,1,0,185543,,,,0,ca0ea8a0a231aa84a10ac832b29ee76a198971f040fcf2c0ab54fa5d95803ec0
thread:issue:185581,issue #185581: `torch.compile` fails at FakeTensor validation for ConvTranspose2d,has linked PR,44,has linked PR,1,0,185581,,,,0,fe459a2c8ca48871661a8d40fd4b38fd884580098bccbbd2ba74132d784cae3c
thread:issue:185640,"issue #185640: `make_fx` generates broken input codegen for a function with a `(tuple, dict)` positional signature",has linked PR,44,has linked PR,1,0,185640,,,,0,ffe93e55a04f7c7ec3c2a76f974d4b4be29f0ac5295aed7d8d7c190a2533bfbf
thread:issue:185647,issue #185647: Unexpected storage from MSE loss output,has linked PR,44,has linked PR,1,0,185647,,,,0,05c89978b09fbca3b83af56255ed7176b3ead4a7181d4d203f9c53ab66c2f35a
thread:issue:185696,issue #185696: [Inductor] torch.atan2 yields incorrect result for x=0.0 with float16 on CUDA (loses signed zero),has linked PR,44,has linked PR,1,0,185696,,,,0,019e000c9e752490b697b83821298aa614df491ee0df5cd43243be8faf2c127c
thread:issue:185885,issue #185885: [Inductor] torch.index_add accepts negative index while eager raises out-of-bounds error,has linked PR,44,has linked PR,1,0,185885,,,,0,1d86048d3da4e483c2a7c459c493732b7905e006ace3408b90b4274378380da9
thread:issue:185975,issue #185975: torch.accelerator.default_stream API is missing,has linked PR,44,has linked PR,1,0,185975,,,,0,7fe715709096248ee456d3e417c65ddc4514386545797a94fec06fd6980ce6da
thread:issue:186140,issue #186140: fix qualname for OpOverload/OpOverloadPacket,has linked PR,44,has linked PR,1,0,186140,,,,0,5bb7210a05c0554617a73e6952300d6c2be8009d65796f79fcbe641e53e82163
thread:issue:186148,"issue #186148: `torch.matmul` misses `mm` fold for `[S, 1, H] @ [H, K]` when the size-1 batch dim has a non-contiguous stride",has linked PR,44,has linked PR,1,0,186148,,,,0,4f9fe9f9f2239790bed0fee69ae00837deb2e9ee87ba542bd4857b7edba8da50
thread:issue:186176,"issue #186176: AOTInductor serializes entire underlying storage for each constant tensor, causing severe bloat for shared-storage weights",has linked PR,44,has linked PR,1,0,186176,,,,0,1630f251af8a2cc02a7d161e701bee1c6cfba6da446d57d15ac8ac86399a4d4a
thread:issue:186544,issue #186544: [Windows] Debug build fails: LNK1104 cannot open file 'python310_d.lib',has linked PR,44,has linked PR,1,0,186544,,,,0,82d0c5554b6468b328a458d4feb569c99641046b398e8a605cad76f41e718fbe
thread:issue:186556,issue #186556: torch.ldexp returns incorrect gradient w.r.t. mantissa for negative exponents,has linked PR,44,has linked PR,1,0,186556,,,,0,643a66a1bdd1682df2b13c4170630885efed7cfe1c4706d2eec188f6624159c9
thread:issue:186572,issue #186572: [Inductor] Checkpointed reductions can choose different configs between forward and recompute,has linked PR,44,has linked PR,1,0,186572,,,,0,5109100ace1bfea59b4b831de6d1ad3df8e8dc9e25a1e04fca7a756a7e344721
thread:issue:186901,issue #186901: Inconsistent behavior of torch.sign on NaN input values,has linked PR,44,has linked PR,1,0,186901,,,,0,11a360e6e482c3f82b8d944b1fb14781b228821e5c42060e8629095ff76a8686
thread:issue:187089,issue #187089: Support BFloat16 in ATen CPU tensor creation (tensor_cpu),has linked PR,44,has linked PR,1,0,187089,,,,0,25288d57dde802f3775ce8b85d46a83e1b0a3a640847b337aa3e33bd9a0b7ee3
thread:issue:187107,"issue #187107: [MPS] linalg.lu_factor not implemented, causing RunTimeErrors on torch.linalg.det and torch.linalg.solve",has linked PR,44,has linked PR,1,0,187107,,,,0,f6722a966145feb5646f8376e0886da9de4f7097d64fc51161ceddc2b46d1fc3
thread:issue:187233,issue #187233: torch.nn.functional.max_unpool3d — internal MPS RuntimeError for complex64 input,has linked PR,44,has linked PR,1,0,187233,,,,0,1a5e4ebfc6414b021fef83c51a3d7ee1ac75412e29bfd60403159d6b3b94cacb
thread:issue:187294,issue #187294: torch.special.zeta returns NaN for q=+Inf instead of 0,has linked PR,44,has linked PR,1,0,187294,,,,0,2a59ae470d0d929dcd0e18b0fe4fb6b7297fbe2a4278086fb2f2332351139181
thread:issue:187295,issue #187295: torch.sign returns 0 for NaN inputs instead of propagating NaN,has linked PR,44,has linked PR,1,0,187295,,,,0,9b86b1ba2a27c394641ebbc3b7f716e03675e1f4cafe0aaa04f32a8f9c7add64
thread:issue:187296,issue #187296: torch.heaviside returns 0 for NaN input instead of propagating NaN,has linked PR,44,has linked PR,1,0,187296,,,,0,35bab662c0fcc15ba40418bf48b6a5f69b833b160b712f9fb03eda346deb180d
thread:issue:187298,issue #187298: convert_sync_batchnorm silently drops the activation / extra behavior of `_BatchNorm` subclasses (e.g. timm `BatchNormAct2d`) - please emit a warning,has linked PR,44,has linked PR,1,0,187298,,,,0,c72f10fbc9e632a1e9aeaadc47bce3f4e3c4fce80866007367cbdba34c57515d
thread:issue:187299,issue #187299: torch.nn.functional.unfold — RuntimeError instead of ValueError for zero stride,has linked PR,44,has linked PR,1,0,187299,,,,0,c6c91816cd86ba3370182cb76f9557e1525d02d1ef01f9634cecda8f161b99c7
thread:issue:187547,issue #187547: Dynamo graph-breaks on raw unbacked SymInt BlockMask lengths in FlexAttention,has linked PR,44,has linked PR,1,0,187547,,,,0,a0824fed18426039fdf5089734ee7546213879e0df16e7d5c8d2efcfce2c5301
thread:issue:187593,issue #187593: [Dynamo] Avoid guarding on Unused variables,has linked PR,44,has linked PR,1,0,187593,,,,0,6083fe5808afaee29f19ceb1ad33288370e42bee832b8e17ab97f9663081f857
thread:issue:188052,issue #188052: Allow PrivateUse1 backends to opt out of CompositeExplicitAutograd decomposition,has linked PR,44,has linked PR,1,0,188052,,,,0,dae47a1c33d15c0fe378c0b8c431cff731e352f40e29683f8d39859c01b15d2e
thread:issue:188088,issue #188088: torch.clamp rejects mixed Tensor and Number min/max bounds although Tensor+Tensor and Number+Number both work,has linked PR,44,has linked PR,1,0,188088,,,,0,65c89a67b1d75b6cf1f663abcbb67973872d5c0b9c1509c6a31a38248db591a1
thread:issue:188166,issue #188166: load_state_dict(assign=True) mutates the caller's state_dict._metadata,has linked PR,44,has linked PR,1,0,188166,,,,0,2359b6b12869611c139537892018728955ae8350543a836d67b7a11175dc81d6
thread:issue:188182,issue #188182: torch.fft has no sharding strategy,has linked PR,44,has linked PR,1,0,188182,,,,0,88598ff03d7950db9d940463bfacafdbdade55df3a7c4e91615512142bbe6fdb
thread:issue:188213,issue #188213: [inductor] `LoweringException: AssertionError` in `scatter_reduce_` lowering,has linked PR,44,has linked PR,1,0,188213,,,,0,d74eacb96214ef98025860b8ec094290014e9049033b02f90699bc01efe38ffe
thread:issue:188217,issue #188217: [inductor] `TypeError: reduce() of empty iterable with no initial value` during lowering,has linked PR,44,has linked PR,1,0,188217,,,,0,f736d61f4f1331a1ff3b9978e35ba861d29845346849d782ac45e6a239d9ca44
thread:issue:188225,issue #188225: [inductor] `index_propagation` `ValueError: The argument 'nan' is not comparable`,has linked PR,44,has linked PR,1,0,188225,,,,0,515cd6d10f86f8e600e02366001dc21d0e3e258716ebec55db3f5ef5943fbafc
thread:issue:188226,issue #188226: [inductor] `ValueRangeError: Invalid ranges [0:-61]` during index propagation,has linked PR,44,has linked PR,1,0,188226,,,,0,fb3f5c776e74519038aee6e1738605f2c5d84e9b7ec4c827e4e35b0b008af11b
thread:issue:188245,issue #188245: Broken tutorial URL,has linked PR,44,has linked PR,1,0,188245,,,,0,17f53c3d67520cd0a0f132c4a3c9945f9868d0b9c1953dd72bc04fa0f1d8e20b
thread:issue:188338,issue #188338: [ROCm]: CUDAGraph safe Convolutional Kernel e.g. `F.conv_transpose1d`,has linked PR,44,has linked PR,1,0,188338,,,,0,213e7ee19cd6401d64887ffe7e0d358da2676c87b371099cbb93234a541e7017
thread:issue:188370,issue #188370: [Meta][PrivateUse1]: Data-dependent ATen ops need eager shape-oracle support for PU1 backends.,has linked PR,44,has linked PR,1,0,188370,,,,0,50d69c7e33b5c6d2126c1dd5155ead62825ef8555b070fea66007beffc782093
thread:issue:188390,"issue #188390: [DTensor][Compile][AOTAutograd cache] DTensor stable hashing generates identical AOTAutograd hash keys across ranks, causing false cache hits.",has linked PR,44,has linked PR,1,0,188390,,,,0,f94ce0388e7d643c68bb17b232300b74acad9957f9300eeed2ebe022b1da182d
thread:issue:188398,issue #188398: [Docs] torch.utils.mobile_optimizer link broken in 2.7 API documentation,has linked PR,44,has linked PR,1,0,188398,,,,0,e436b589394adedcde3a9ebea084fbf9694a5d3c0069b1e74f3e96cdce7c169e
thread:issue:188536,issue #188536: Graph-break reconstruction loses identity of graph-created torch.Event subclasses; record() and wait() operate on different event objects,has linked PR,44,has linked PR,1,0,188536,,,,0,3b508e4fd579b956993da1628307a20b9bfcb4edaa9356765a8ae559ff0246dd
thread:issue:188538,issue #188538: torch.fx _format_target emits invalid or semantically incorrect Python for valid module names,has linked PR,44,has linked PR,1,0,188538,,,,0,a66e18cd8624574b099e50619a12f578002d335a9e4e4bfb4648c649774a4c73
thread:issue:188635,issue #188635: torch.compile misses fusion for producer -> view/transpose/contiguous layout clone in attention-style paths,has linked PR,44,has linked PR,1,0,188635,,,,0,7a89cb102b3f61d91b6bdccee0e4a94762125b4d084a3ccd4007a18cd3f8af79
thread:issue:188728,issue #188728: [MPS] F.linear does not respect the PYTORCH_MPS_PREFER_METAL flag,has linked PR,44,has linked PR,1,0,188728,,,,0,36a037d9ea98d9b8ae7c33fe63a8d10314edd158f41cddf51893c91e3850b86b
thread:issue:188796,issue #188796: [opaque object] Callable value Opaque Object on a tensor subclass is not accessible for method calls in torch.compile and leads to graph-break,has linked PR,44,has linked PR,1,0,188796,,,,0,2ea957a682d2ec57e2ef64a2596e120ba44c862c23f6c6b13cce901097a33563
thread:issue:188882,issue #188882: Empty autograd section in backend YAML is rejected,has linked PR,44,has linked PR,1,0,188882,,,,0,2957566356d4a9ea53885bd03516c7a48a8b28a682e441e7145fa9b26cfa9192
thread:issue:73009,issue #73009: Improve loading for nn.modules.lazy.LazyModuleMixin,has linked PR,44,has linked PR,1,0,73009,,,,0,c5d7dcf276799787cafbce53bae82fee759be02c0990f93224bf583ac1b5f670
thread:issue:77176,issue #77176: Improve the overall design of MPSGraphCache,has linked PR,44,has linked PR,1,0,77176,,,,0,5a0e79d4c30d4f3f756bb72487084c79272ac2ab0b4ad36fa0642aa2b46954a8
thread:issue:80025,issue #80025: `torch.special.gammainc` backward pass with respect to the first argument,has linked PR,44,has linked PR,1,0,80025,,,,0,723edeb3a246e38feb37578c5796e47075a63336cdb002bcb9af978b8e3c6a92
thread:issue:89459,issue #89459: Higher order derivatives of sinc explode,has linked PR,44,has linked PR,1,0,89459,,,,0,b3de9a44854acc032ecb5b4205f5a3ee7aaf6fbbee7ac77534a29add52466aa0
thread:issue:89489,issue #89489: build: cmake: need to uniformize installation of libraries in CMAKE_INSTALL_LIBDIR (not lib),has linked PR,44,has linked PR,1,0,89489,,,,0,9daf5545adf104cdb3b7fc48ea47f9438f6ed1066cc360e54a7c899fe6af59c9
thread:issue:91898,issue #91898: Lazily start worker threads in the autograd engine,has linked PR,44,has linked PR,1,0,91898,,,,0,a8023bfae0149a240c283a520834a417e12a204daa467ed71760c8c5e2ac9902
thread:issue:96409,"issue #96409: Initialization on `meta` device failing for models containing `nn.utils.weight_norm`, with `NotImplementedError: Could not run 'aten::_weight_norm_interface' with arguments from the 'Meta' backend.`",has linked PR,44,has linked PR,1,0,96409,,,,0,29c7a332e23cbdf29bb7bd60f1fcbbf26c6e9cb59cc6bb4201a01d52b5da1a27
thread:issue:181946:issue:183758:pr:182712,pr #182712: [MPS] Enhance Linear tensor ops to avoid unnecessary .contiguous() calls on weight tensor,waiting on contributor,43,Wait for contributor update on PR #182712; keep related issue/PR context attached.,3,0,"181946, 183758",182712,,,2,d237bf8dea545d5318522d720cd33353d44b18039f4a838b1ed1ca30028051ed
thread:issue:170753:pr:176678,pr #176678: [CI][Windows] Add NVIDIA RTX workflow,waiting on contributor,43,Wait for contributor update on PR #176678; keep related issue/PR context attached.,2,0,170753,176678,,,3,565aa75d06903e926ddef4c2d41375e3b18b232aeac1cbf92991f4376ba40099
thread:issue:180362:pr:173519,pr #173519: Add on-demand ppc64le wheel build support,waiting on contributor,43,Wait for contributor update on PR #173519; keep related issue/PR context attached.,2,0,180362,173519,,,1,654ec4cefdc221b047664cf4452c200e06d0d1b9f12bc4fb393fc003b2329c31
thread:issue:183434:pr:185813,pr #185813: [ROCm][Windows] Audit license-files: explicit enumeration + audit test (#183434),waiting on contributor,43,Wait for contributor update on PR #185813; keep related issue/PR context attached.,2,0,183434,185813,,,1,7f97451252b38115274a3f719b3e8e4fcffc5194f204def520d58459980999e5
thread:issue:186449:pr:186965,pr #186965: [RFC][dynamo] Add ContextVar.get() tracing support (Phase 1),waiting on contributor,43,Wait for contributor update on PR #186965; keep related issue/PR context attached.,2,0,186449,186965,,,2,dc0f2a032c66a73fac44244d751a0a78c1bf6bb6247b88dabcdbe9aeb70b8fe0
thread:issue:188020:pr:187696,pr #187696: Add DataLoader worker signal message affixes,waiting on contributor,43,Wait for contributor update on PR #187696; keep related issue/PR context attached.,2,0,188020,187696,,,1,d3f622fb6c4fdc018836ae99d4c35ccf7b5e8da4aee04343f3a591d5c87928b8
thread:pr:130856,pr #130856: [pipelining] lazy shape inference for stage,draft/noise,43,draft/noise,1,0,,130856,,,0,00c3574e4a21e8f5c97f7f50aec596a98e2277bdb21bbf5f14b9453d785b608e
thread:pr:166063,pr #166063: Add Gluon based grouped MM kernel for Blackwell into Inductor,draft/noise,43,draft/noise,1,0,,166063,,,0,dea4537ac668b392a1d9c5697c71ac011a1b606a8c1905019952997c0cfc718f
thread:pr:166396,pr #166396: [xpu][test][1/N] Enable tests of test_nn.py on Intel GPU - instantiate TestNN with instantiate_device_type_tests,waiting on contributor,43,Wait for contributor update on PR #166396; keep related issue/PR context attached.,1,0,,166396,,,0,3deae785fa4f43b2e06fdfe7756469b3677094df13d25d9be94922f70b38b90a
thread:pr:169916,pr #169916: [CPU] Support multiple ncores-per-instance settings in xeon/run_cpu.py,waiting on contributor,43,Wait for contributor update on PR #169916; keep related issue/PR context attached.,1,0,,169916,,,0,41ce5e90f3cc0a8a711a69e980366f9947bf38203722efa66b23750a62117172
thread:pr:170752,pr #170752: Support torch.accelerator.get_device_capability on CUDA,waiting on contributor,43,Wait for contributor update on PR #170752; keep related issue/PR context attached.,1,0,,170752,,,0,46e16a393b2b329769aee4c44e7c2da4a050bcf6b292d2c26be72dcd2fa96da6
thread:pr:171035,pr #171035: Add 2D Right-hand-side support for cuDSS backend `linalg.solve`,waiting on contributor,43,Wait for contributor update on PR #171035; keep related issue/PR context attached.,1,0,,171035,,,0,b81b772ff1bdc9ff0e730f6fd7df7e4514b36b7e259cff9dd508b902dd2b49a3
thread:pr:172109,pr #172109: stage executor,draft/noise,43,draft/noise,1,0,,172109,,,0,a2431b5b3f1b9b8b30564fb49e639a8e149f6537194c2605fbd69c76e83beff2
thread:pr:172603,"pr #172603: Rewrite _get_total_norm with _powsum, add skip_root flag",waiting on contributor,43,Wait for contributor update on PR #172603; keep related issue/PR context attached.,1,0,,172603,,,0,07b322250a06a62c0064bbfe6f5b6904532069d4955ce2c0498b10bba7e5804a
thread:pr:172945,pr #172945: Bf16 optimized heuristics,waiting on contributor,43,Wait for contributor update on PR #172945; keep related issue/PR context attached.,1,0,,172945,,,0,bc4a31f9903513b4b84314524e9ed449881b4cfbeef0bbe1d2b7dabd740e6d0b
thread:pr:173258,pr #173258: [CUDA] Math.cuh add missing return statement,waiting on contributor,43,Wait for contributor update on PR #173258; keep related issue/PR context attached.,1,0,,173258,,,0,ec36b4772718876c79b54bd742ec0e6bcb5eadc62adb88ddb603103470848b33
thread:pr:174491,pr #174491: fix: prevent issue with destruction of std::shared_timed_mutex,waiting on contributor,43,Wait for contributor update on PR #174491; keep related issue/PR context attached.,1,0,,174491,,,0,20a92635388e21165dc80d281f635471f479e02cb767fe944138d9e515ef8bd3
thread:pr:176420,pr #176420: [XPU] Ensure dense (canonical) strides for convolution inputs,waiting on contributor,43,Wait for contributor update on PR #176420; keep related issue/PR context attached.,1,0,,176420,,,0,62d07af13fa91d1c844a84acf7aa4216bcf961420c078005d0bf02124cb5ff9d
thread:pr:176619,pr #176619: Fix: raise RuntimeError when cudagraph partitions are empty,waiting on contributor,43,Wait for contributor update on PR #176619; keep related issue/PR context attached.,1,0,,176619,,,0,e7677c2ea80e0e7e54eff0ea9cc6815b60cb80a514024ff5a89558ab8da92013
thread:pr:176747,pr #176747: Fix root cause of extraneous termination handler error logging (eval) (#176747),waiting on contributor,43,Wait for contributor update on PR #176747; keep related issue/PR context attached.,1,0,,176747,,,0,bc005ffb0a5f58d42ed2709776bfc8468f265568de459d1258759b15d582d5f4
thread:pr:177540,pr #177540: [inductor] Handle Triton cluster_dims metadata from both binary and metadata layouts,waiting on contributor,43,Wait for contributor update on PR #177540; keep related issue/PR context attached.,1,0,,177540,,,1,02944fcf5aab88580c3be7ee8a5e108c19e43d6f4a44772f89000dc604215255
thread:pr:177605,pr #177605: Default to fused implementation when available,draft/noise,43,draft/noise,1,0,,177605,,,0,3744ad9eca825ab1709fc93c0df6a2e5e95ec605df97a31f975f8f6371db84d7
thread:pr:177690,"pr #177690: DO NOT MERGE Default pre_grad_pass_timing to ""late"" in fbcode",draft/noise,43,draft/noise,1,0,,177690,,,0,e124635df7674edc3a14617f6b715fe2808ed74fbd42ff46847b044859765755
thread:pr:177705,pr #177705: [Do Not Land] Lookup Table Recorder,draft/noise,43,draft/noise,1,0,,177705,,,0,85efb06fe9617c284dbd15812ea64e07130bb0ca1b7951c1f0b43fa0261efe20
thread:pr:178048,"pr #178048: [inductor] Add custom_pass_context for scoped, composable custom pass registration",waiting on contributor,43,Wait for contributor update on PR #178048; keep related issue/PR context attached.,1,0,,178048,,,0,df4d3e82cf40215ccb342e31f7e37c82d39e3730d8249ff2c0dbeee740b95d9e
thread:pr:178204,pr #178204: [PyTorch][OSS][AOTAutograd] Allow None values in saved tensors for vc_check split (#178204),waiting on contributor,43,Wait for contributor update on PR #178204; keep related issue/PR context attached.,1,0,,178204,,,0,c351cb79097ed0e3c4d834368361dfa74da229d464758e6f4cb26aadcc050729
thread:pr:178354,pr #178354: [XPU] Add XPU dispatch for _scaled_grouped_mm,draft/noise,43,draft/noise,1,0,,178354,,,0,62360e7c2806b8788c5418789c39117da56d51d38c6af1d3b029e4b7c0338959
thread:pr:178412,pr #178412: Move steps to CPU for fused optims,draft/noise,43,draft/noise,1,0,,178412,,,0,22dd65c7096651dc5753aa53dc1c80349f67665cc354ecfb7932e3b473be6018
thread:pr:178463,pr #178463: [Distributed] Make Ckpt Tests Backend Agnostic,waiting on contributor,43,Wait for contributor update on PR #178463; keep related issue/PR context attached.,1,0,,178463,,,0,d9d69b72afa04963a3cfa004d2b2e57975f7866566da40736f72aab062b8d8c4
thread:pr:178550,pr #178550: [DTensor] Support sharding propagation with meta device mesh,waiting on contributor,43,Wait for contributor update on PR #178550; keep related issue/PR context attached.,1,0,,178550,,,0,8e319b4f2b0140688a73fa32e87ea1854191f1f5cde61c79f82aa4832db4f179
thread:pr:178628,pr #178628: [DRAFT]Add TorchInductor documentation to User Guide,draft/noise,43,draft/noise,1,0,,178628,,,0,fdeb5af6e112d918a254c845d0d7912889d6a9e23ef814d07f495c789f8afcc8
thread:pr:178737,"pr #178737: Enable hipsparselt version, initCusparseltBindings and related UTs",waiting on contributor,43,Wait for contributor update on PR #178737; keep related issue/PR context attached.,1,0,,178737,,,2,21d432f0f03e8da37e2360f904771cb27134dcd42ead69290b853212af06eb40
thread:pr:178840,pr #178840: Add parallel subprocess compilation and disk caching for NVGEMM backend,draft/noise,43,draft/noise,1,0,,178840,,,0,e0b65067f44d057077dfffd72e176ea0644dcec4c57c626d18802fd02de0e1c3
thread:pr:179011,pr #179011: [Inductor][NVGEMM] Add fast path for GEMM dispatch,draft/noise,43,draft/noise,1,0,,179011,,,0,574a9d7b84f56d9a36ca069d1bf52162b2005522f08cde07f26cb07a270e79a9
thread:pr:179012,pr #179012: [Inductor][NVGEMM] Add supplement heuristic configs behind flag,draft/noise,43,draft/noise,1,0,,179012,,,0,8599a45759e043ab406695116e7e10db75a20f30cff77031a14981b386a02d48
thread:pr:179013,pr #179013: [Inductor][NVGEMM] Fix FP4 autotuning tensor creation,draft/noise,43,draft/noise,1,0,,179013,,,0,e639e14e287bda4e3d1578fa4ae5ca3c6e7498ae348f393709b46cce276ed738
thread:pr:179014,pr #179014: [Inductor][NVGEMM] Wrap GroupedGemmArguments in try/except,draft/noise,43,draft/noise,1,0,,179014,,,0,943e0d08b726915675075a58ae2652720d228ddb250fc03eaaa380768f66a1d3
thread:pr:179015,pr #179015: [Inductor][NVGEMM] Add fast path for GEMM dispatch,draft/noise,43,draft/noise,1,0,,179015,,,0,0f3b0c11fc295fb77418a2e055ed6829c492ba0d71fef43fff6abe1804bea25e
thread:pr:179016,pr #179016: [Inductor][NVGEMM] Add supplement heuristic configs behind flag,draft/noise,43,draft/noise,1,0,,179016,,,0,1c357948e2e5a6feffe06297465715a1c5cec04e3f2ca39b49735b0c0f40b0d9
thread:pr:179049,pr #179049: [Inductor][NVGEMM] Add supplement heuristic configs behind flag,draft/noise,43,draft/noise,1,0,,179049,,,0,2c90a72e13891aa4e547d47d7dfe725979f3087a852baaddb72ad772e08190ec
thread:pr:179167,pr #179167: [Inductor][NVGEMM] Fix FP4 autotuning tensor creation,draft/noise,43,draft/noise,1,0,,179167,,,0,d82deeb254db374aeb5096100877ae613cff02227d8f66264955e7991c8de9fa
thread:pr:179175,pr #179175: [Inductor][NVGEMM] Fix FP4 autotuning tensor creation,draft/noise,43,draft/noise,1,0,,179175,,,0,8a481920f266c988dc45f4594f3b91d1336b8b5869a42ef0376317ed49e7f779
thread:pr:179284,pr #179284: Fix AOT autograd leaf backward mutation for auto_functionalized V2 (#179284),waiting on contributor,43,Wait for contributor update on PR #179284; keep related issue/PR context attached.,1,0,,179284,,,0,4384b83bd1d1ebfa75c5c0669961d37210ee97683ac03b61787689582deb6aad
thread:pr:179422,pr #179422: [inductor] Force contiguous inputs for custom ops in FallbackKernel,waiting on contributor,43,Wait for contributor update on PR #179422; keep related issue/PR context attached.,1,0,,179422,,,0,e82336bfb0302ad91efa5bd8ea356d5620ed401ee5d040b0de9dc245811a526e
thread:pr:179425,pr #179425: [inductor][autotuning] add logging and config for incremental autotuning (#179425),waiting on contributor,43,Wait for contributor update on PR #179425; keep related issue/PR context attached.,1,0,,179425,,,1,a92cb77a96e26003830f5ddd0707b8b831c75a04ac19eb80eb6a7d1aad0fc5a1
thread:pr:179546,pr #179546: Add structured call_hierarchy metadata to FX nodes,waiting on contributor,43,Wait for contributor update on PR #179546; keep related issue/PR context attached.,1,0,,179546,,,0,444ed11af3f444b457898f23b6d50a4e109b2ace04682215bf444614dec216c0
thread:pr:180080,pr #180080: Bitwise operator tensorcompare.cpp,waiting on contributor,43,Wait for contributor update on PR #180080; keep related issue/PR context attached.,1,0,,180080,,,0,f425b260a0d568bac6b5f9de0be319790c175056a147e95425cb567c220020bf
thread:pr:180129,pr #180129: [scaled_grouped_mm] add support for out argument,waiting on contributor,43,Wait for contributor update on PR #180129; keep related issue/PR context attached.,1,0,,180129,,,0,c3fe4e36959afa526af99f58050cf7c26abb6b4e14021b7215c65b3593880344
thread:pr:180826,"pr #180826: Remove deprecated `set_group_info`, `get_group_info` and `enable_symm_mem_for_group`",waiting on contributor,43,Wait for contributor update on PR #180826; keep related issue/PR context attached.,1,0,,180826,,,0,3d283acf7ff4c5a8851d951560f6bc9b6d599115fdecfb5785ab636c816bdb6f
thread:pr:180874,pr #180874: Add CPU offload support for selective activation checkpointing,draft/noise,43,draft/noise,1,0,,180874,,,0,7f19e8a46f03ef17019e56c7243f11283cb5c276777f2f059e4e14c50fbe2362
thread:pr:180891,pr #180891: [ROCm] Migrate from rocm_smi to amd_smi,draft/noise,43,draft/noise,1,0,,180891,,,0,f4343fbe23f3e3cde0b362c095af8961931052ca97ad2456e392d51f7a298db8
thread:pr:180904,pr #180904: Enable MyST Markdown docstrings in Sphinx docs,draft/noise,43,draft/noise,1,0,,180904,,,0,1e40f94e51a6709c6d26285bc4c3f3496905101693cb2ba6b97f8cafc9bf973f
thread:pr:181089,pr #181089: [Inductor] Disable split-reduction lowering under batch_invariant,draft/noise,43,draft/noise,1,0,,181089,,,0,2781aa9bd17bdf846d9d186d46d71c2dfca45755081edf179756569e45d79b9f
thread:pr:181138,pr #181138: [Inductor][NFC] Refactor fusion legality checks for `UserDefinedTritonKernel`,waiting on contributor,43,Wait for contributor update on PR #181138; keep related issue/PR context attached.,1,0,,181138,,,0,658138894f8ba904b781578dac967b45e60705c9611142fa5fc6ee54d35250c2
thread:pr:181306,pr #181306: Integrate profiler documentation into torch.profiler docs,draft/noise,43,draft/noise,1,0,,181306,,,0,3d375514e1d0a3787fd0a155472c04455ed5368d57c9e4f7841b4ee9ec82960c
thread:pr:181730,"pr #181730: Prevent DCE eliminate ""triton_kernel_wrapper_mutation""",waiting on contributor,43,Wait for contributor update on PR #181730; keep related issue/PR context attached.,1,0,,181730,,,0,f855ad6d1c9d047a63135a8696e7803986ca77e4b125d59cdab5cc0d4e136f2f
thread:pr:182025,pr #182025: [MPS] Use MetalPerformancePrimitives for F.linear on macOS 26+,waiting on contributor,43,Wait for contributor update on PR #182025; keep related issue/PR context attached.,1,0,,182025,,,0,7b18de7c076877d8532a2b88a27d939fd40445fb23e2fc7f96aac535e1c1c037
thread:pr:182208,pr #182208: Add missing stacklevel to 57 warnings.warn() calls in torch/,waiting on contributor,43,Wait for contributor update on PR #182208; keep related issue/PR context attached.,1,0,,182208,,,0,0ae7976503b6cecb25ced86c4835734c91af25701f17df46e5c5f0a6df30f091
thread:pr:182233,pr #182233: Fix torch.combinations dynamic shape support by replacing numel with sym_numel and propagating SymInt,waiting on contributor,43,Wait for contributor update on PR #182233; keep related issue/PR context attached.,1,0,,182233,,,0,e96ccfb4c4cdd13f76c400076f43e683d412b7ae232690ba1d77091e39ead2d5
thread:pr:182278,pr #182278: [CI] Add native build image for linux-riscv64,waiting on contributor,43,Wait for contributor update on PR #182278; keep related issue/PR context attached.,1,0,,182278,,,1,1f6d703cc6e070e400f603f695f0d4432d2ff4688e6892215f805992e09e2bc2
thread:pr:182576,pr #182576: Improved static Type Hints for dataloader,waiting on contributor,43,Wait for contributor update on PR #182576; keep related issue/PR context attached.,1,0,,182576,,,0,7533e3d77f76eca8357465ed9436f9d9b0cbff5b1cb5b63ae96fdd54d41f9933
thread:pr:182725,pr #182725: [c10d] Throttle TCPStore connect-retry warnings,waiting on contributor,43,Wait for contributor update on PR #182725; keep related issue/PR context attached.,1,0,,182725,,,1,6efbe7ee9a08120a3658d5e16eef6b037d854db09434aed975523f107ae0a774
thread:pr:182779,pr #182779: [WIP] Test playground,draft/noise,43,draft/noise,1,0,,182779,,,0,d4b3eacb24077925bada9e1fdf6fd68645b108356624cf24cb18d04a62bde2bd
thread:pr:182809,pr #182809: [2/N] Replace data_ptr with mutable_data_ptr,waiting on contributor,43,Wait for contributor update on PR #182809; keep related issue/PR context attached.,1,0,,182809,,,0,50ae3d5069ee0baf9b78972526f322f79e8f8841b49bc032007b7ec322e0bcfc
thread:pr:182921,pr #182921: [DO NOT MERGE][TESTING ONLY] Set MI300 ROCm build arch to SPIR-V,draft/noise,43,draft/noise,1,0,,182921,,,0,057480b59b090504e16193966045d918b9134b8989002b17e5dea7dbb277f5dd
thread:pr:182984,pr #182984: [Inductor] Expose config for sm carveout of templates (#182984),waiting on contributor,43,Wait for contributor update on PR #182984; keep related issue/PR context attached.,1,0,,182984,,,0,4a5d0dbf134662a49fca1724df6e8ec1086777750cd8ce377987b0b0629cb8a2
thread:pr:183054,pr #183054: coverage-td plan,draft/noise,43,draft/noise,1,0,,183054,,,0,5599b98c35c379fbf30d530c8bfe3d64db5cd2b22e504b92fd3b617d1292ecc3
thread:pr:183264,pr #183264: [WIP][Dynamo] Add mark_batch_invariant API,draft/noise,43,draft/noise,1,0,,183264,,,0,581ffdad499b31f78c89e0fadd53191e58fb08888ba49b16f9311c8c169cc3be
thread:pr:183361,pr #183361: Prototype lazy module backward hook compile,draft/noise,43,draft/noise,1,0,,183361,,,0,e0a1c973789322b8448dd5080f019cff03045c7a8b573bb20894b55780fe69ae
thread:pr:183374,"pr #183374: [ROCm] [ TEST , PLEASE DON""T MERGE ] TheRock 7.13 nightly deb install path",draft/noise,43,draft/noise,1,0,,183374,,,0,4da519f74cd60b5a74f5c90dcd87032a085bc1b283719058ec5302318d55ab3d
thread:pr:183452,pr #183452: [BE]: Update cutlass submodule to 4.5.0rc,draft/noise,43,draft/noise,1,0,,183452,,,0,79ce48f134da1cb02b62a007fccdf4c7dee219d15887d738f9dfdc2d3730bfa2
thread:pr:183456,pr #183456: [ROCm] [opbench] Temporary profiler workaround for rocprofiler crash (#182719),draft/noise,43,draft/noise,1,0,,183456,,,0,a9256d6960bc053a558f524b1a7d4e58c3f567a59108a0f62184bb22587a26a8
thread:pr:183479,pr #183479: TMA scatter_add: add multi-row-per-warp path for small rows,draft/noise,43,draft/noise,1,0,,183479,,,0,9d2df53bea252f89116bea278c79a066321b337e23c73710cca86325e6fd4683
thread:pr:183556,pr #183556: [inductor] Fix MemoryTracker assertion with shared storages in reordered schedules,waiting on contributor,43,Wait for contributor update on PR #183556; keep related issue/PR context attached.,1,0,,183556,,,0,8e180c95288d91e52b33d75b22a1e91de7b944f2f75b0369916a01af01f56dbc
thread:pr:183612,"pr #183612: Update NVPL versions: blas 0.3.0 → 0.5.0.1, lapack 0.2.3.1 → 0.3.2",draft/noise,43,draft/noise,1,0,,183612,,,0,028ccde5c354c82b428e2d356f69e79965c188e51fb3f9713e0171dfd29e12d1
thread:pr:183663,pr #183663: [inductor] _make_launchers: skip rebuilding existing launchers,waiting on contributor,43,Wait for contributor update on PR #183663; keep related issue/PR context attached.,1,0,,183663,,,1,56359e74f48b1f84feab881578b97807c28f7617977bf4541cf34a740101e084
thread:pr:183669,pr #183669: [inductor] Make experimental benchmarker parameters configurable via env vars,waiting on contributor,43,Wait for contributor update on PR #183669; keep related issue/PR context attached.,1,0,,183669,,,0,e6810d076455be239b44e025824e644e3d43b1141515cf012539e7883cc46195
thread:pr:183673,pr #183673: [ROCm][Windows] torchgen: fix Windows DLL linkage for native and namespaced forward declarations,waiting on contributor,43,Wait for contributor update on PR #183673; keep related issue/PR context attached.,1,0,,183673,,,0,c9e714566274813f2669339ca8efdf166977a6344c27427bdcc34717f45f6dd6
thread:pr:183797,pr #183797: fix internal tests for [BE][Ez]: Make more dataclass slots=True (#183449),draft/noise,43,draft/noise,1,0,,183797,,,0,482589a497a5019e7b4c4bb2fc661c9d0c92d86795594e7d343a6a767aaa4c10
thread:pr:183798,pr #183798: [ROCm] Fix HIP stream pool aliasing when GPU tracing is active,draft/noise,43,draft/noise,1,0,,183798,,,0,3f9642b283f17fb22bde943ac81beef2b2ba17800bf093dc7c00627e1a704046
thread:pr:183925,pr #183925: Covert notes/cuda from .rst to .md,draft/noise,43,draft/noise,1,0,,183925,,,0,6fa5c4e713f37511aed60c879a46c843c9bcae80e201af45911d1676e1136067
thread:pr:183966,pr #183966: [CUDA] Change argmax to int64_t for better numerics,draft/noise,43,draft/noise,1,0,,183966,,,0,7157a69ca1eac156c51003aca3be81cbf120c702e2eb2872fce6fd6bd22117f1
thread:pr:184141,pr #184141: Remove pointer_to_optional_list,waiting on contributor,43,Wait for contributor update on PR #184141; keep related issue/PR context attached.,1,0,,184141,,,0,f329e0c853a04f1500bf77811f4013f6f0dc4a3382310d8cad4f2c1ad8a706e4
thread:pr:184335,pr #184335: [spmd_types] pipeline parallel support for spmd_types-annotated tensors,waiting on contributor,43,Wait for contributor update on PR #184335; keep related issue/PR context attached.,1,0,,184335,,,0,346dbc37c4e92439e94240f8b46ea38fdc1f85f6dfa82347adb5c315c1d269cb
thread:pr:184440,pr #184440: [dynamic_spec]Add optimizer to ease user's dynamic_spec adoption,draft/noise,43,draft/noise,1,0,,184440,,,0,0418a8e86aed659e9ad71b5bea3729bb0bf2c36935ead0273d72d2ff39eb6cb8
thread:pr:184464,pr #184464: Cache PyCompilerInterface method lookups during graph capture,draft/noise,43,draft/noise,1,0,,184464,,,0,08319924b3df6220f4f102b37d893500a453fda3843621bef25bdeacf5519ec4
thread:pr:184533,pr #184533: Defer AOTAutograd version-check error for forward-only graph,waiting on contributor,43,Wait for contributor update on PR #184533; keep related issue/PR context attached.,1,0,,184533,,,0,b8259359772692f2378d07a1b8593dd40e9b2a66f6f6786f6b4ee7a78d229eac
thread:pr:184654,pr #184654: Short-circuit pytree.tree_map_only when no leaves match,waiting on contributor,43,Wait for contributor update on PR #184654; keep related issue/PR context attached.,1,0,,184654,,,0,1467e8896feb73f555608da36189a74c936b3615bb5790e058e4f17f67215844
thread:pr:184739,pr #184739: Refactor test_sdpa for out-of-tree backends,waiting on contributor,43,Wait for contributor update on PR #184739; keep related issue/PR context attached.,1,0,,184739,,,0,a22a64b4036c4d9c112ffef46beff4344f10bd0a1f097329a5f902b0685144ca
thread:pr:184873,pr #184873: Scope third-party install rules out of the wheel via CMake components,draft/noise,43,draft/noise,1,0,,184873,,,0,6e563bf20302beb7321bfa2842ff25f6b98c5a08a6d6f5b411d045e8fc4477fc
thread:pr:184895,pr #184895: [Inductor] Index expression codegen for user Triton kernel fusion,waiting on contributor,43,Wait for contributor update on PR #184895; keep related issue/PR context attached.,1,0,,184895,,,0,ad43451b4c94cc74f56dc9e722e35f2d18624af3ff88e066fbbcff5dab871a16
thread:pr:184930,pr #184930: [inductor] Score broadcast reuse for expensive pointwise tiling,draft/noise,43,draft/noise,1,0,,184930,,,0,5596db0e170bea8b7354c6fab1ab24a5d06084a72fd2e01c8e3ae13729325b94
thread:pr:184933,pr #184933: [inductor] Fold sparse one-hot sum reductions,draft/noise,43,draft/noise,1,0,,184933,,,0,eb0f92b55f6f709b166afec2ca14fc976f7c0d970d820c9655bce1d2a446567a
thread:pr:184937,pr #184937: [inductor] Prototype channel-resident reduction suffix codegen,draft/noise,43,draft/noise,1,0,,184937,,,0,6a42d02b4ba2c74b6b1baba77d0bf8763967b38a22cb84eccf7cb06c37227e74
thread:pr:184949,pr #184949: Remove allow-untyped-defs from pyi files,draft/noise,43,draft/noise,1,0,,184949,,,0,16fe8f33174d1aede41291df97c372311c17315d112a69dd7d72000c8540a05f
thread:pr:184962,pr #184962: [ROCm] Drop version cap on bf16/fp16 mm/bmm accuracy skip,draft/noise,43,draft/noise,1,0,,184962,,,0,84b24265260b8b2633571caa642aad9af9e1615c9295b318ad82bc2ccc1abd38
thread:pr:184999,pr #184999: [inductor] Avoid pointwise cat for complex output branches,draft/noise,43,draft/noise,1,0,,184999,,,0,a39f29dc7fe13ea46bb4c63995ef3725d5830461b88c6e85d8734fe43e12f186
thread:pr:185011,pr #185011: [inductor] Pair maxpool value and offset lowering,draft/noise,43,draft/noise,1,0,,185011,,,0,0ce9e6d4f61a8502f8e728d58d52808ba0bc1b3ef8b021f059a5f305f8a3b0d3
thread:pr:185012,pr #185012: [inductor] Add FP64-heavy pointwise autotune candidate,draft/noise,43,draft/noise,1,0,,185012,,,0,7a65c2b03ba5e8653b9410ce40789411d5b364124f6f721f30d73985c692e684
thread:pr:185015,pr #185015: Add SAC-driven fully overlapped activation CPU offload (primarily aimed at GH200/GB200/GB300 superchips),waiting on contributor,43,Wait for contributor update on PR #185015; keep related issue/PR context attached.,1,0,,185015,,,0,0af55e8a7b2df51b08262aed67ef2e6a14d799694c74ab98edc5950972a014b2
thread:pr:185407,pr #185407: [RFC][dynamo] Narrow GB6297 to only graph-break on differentiable duplicate tensor inputs,waiting on contributor,43,Wait for contributor update on PR #185407; keep related issue/PR context attached.,1,0,,185407,,,0,ba14f853ff0278d7c41ee90dd366aa44d5764089415740ddf4fbf10e99af1337
thread:pr:185418,pr #185418: [coor] Represent get_rank as a SymInt,waiting on contributor,43,Wait for contributor update on PR #185418; keep related issue/PR context attached.,1,0,,185418,,,0,e707cf769fb0111818cbdfaa77ed2c733786c7f16b4af1eda2d75a672317ae97
thread:pr:185475,pr #185475: [pytorch] Nested custom Python process groups (experimental) (#185475),waiting on contributor,43,Wait for contributor update on PR #185475; keep related issue/PR context attached.,1,0,,185475,,,0,a96a5008c93cd29a6404a3f914ffb5453bf739a48eba2c5eec0d2c563f7e243f
thread:pr:185505,pr #185505: [inductor] Enable ROCm CK GEMM templates under JIT cpp_wrapper,waiting on contributor,43,Wait for contributor update on PR #185505; keep related issue/PR context attached.,1,0,,185505,,,0,a250f4f76abedbac0fce8242dda11ac32b0b171c17ce1282373b5ce8f213590d
thread:pr:185617,pr #185617: [Profiler] Refactor test_trace_validator.py,waiting on contributor,43,Wait for contributor update on PR #185617; keep related issue/PR context attached.,1,0,,185617,,,1,c8f060c7b0cb56ee3cc4e6813167f924a46f67eb5e0c7eb6600f691c8c19d163
thread:pr:185701,pr #185701: [Test] Make test_packed_sequence.py device agnostic for out-of-tree backends,waiting on contributor,43,Wait for contributor update on PR #185701; keep related issue/PR context attached.,1,0,,185701,,,1,dbcee7d707f0175f59f6385aaeff26c6fb78bda512178decf32a370f7f9a34e1
thread:pr:185741,pr #185741: [PyTorch][Profiler] Fix use-after-free in global RecordFunction callback teardown (#185741),waiting on contributor,43,Wait for contributor update on PR #185741; keep related issue/PR context attached.,1,0,,185741,,,0,8cf1a705cae6f7087cae290c2afb2cbbd0355f01834d10f707d327ba075da43a
thread:pr:185910,pr #185910: [AARCH64] Optimize AArch64 qlinear performance by triggering blocked layout in prepacked weight,waiting on contributor,43,Wait for contributor update on PR #185910; keep related issue/PR context attached.,1,0,,185910,,,0,b6a24e7793f8d3249a008d27aad77dc3344d38b595c297e4b0581b204e72e4a7
thread:pr:186047,pr #186047: [Test][PrivateUse1] Enhance `get_all_devices` to include all available devices,waiting on contributor,43,Wait for contributor update on PR #186047; keep related issue/PR context attached.,1,0,,186047,,,0,5cb9bc45aeb9af1fd156e0bbb583234da807f5a08849ed429875b47d1529b60c
thread:pr:186104,pr #186104: [Inductor] Allow output-input aliasing in torch.cond branches (#186104),waiting on contributor,43,Wait for contributor update on PR #186104; keep related issue/PR context attached.,1,0,,186104,,,0,40e9abcf147bfbf71cde8fc62c104d004bfd524863223ea6914df3176f3dfc3b
thread:pr:186138,pr #186138: Support data-dependent 0-dim tensor indexing under tracing,waiting on contributor,43,Wait for contributor update on PR #186138; keep related issue/PR context attached.,1,0,,186138,,,0,7a55cfb16cd9a4f092222a2d0d88b370534c9223ac34bf590bb941dcf66a7bdf
thread:pr:186251,pr #186251: Wait for Docker images before starting builds,waiting on contributor,43,Wait for contributor update on PR #186251; keep related issue/PR context attached.,1,0,,186251,,,0,8566677dc5cf6de5a63b1cb498e99653dde9ecc87a808243f1b386f2dd67076b
thread:pr:186324,pr #186324: [FakeTensor] Propagate backend metadata through fake tensor conversion,waiting on contributor,43,Wait for contributor update on PR #186324; keep related issue/PR context attached.,1,0,,186324,,,0,89967e3655adb4844c22cb85b1a93c69eb53d6c4b2a18d2a0d858a0c4102ae3d
thread:pr:186642,pr #186642: [inductor][rocm] make AMD MM matrix_instr_nonkdim configurable,waiting on contributor,43,Wait for contributor update on PR #186642; keep related issue/PR context attached.,1,0,,186642,,,0,2749ae630a167436b008e23cf0a53dab95a14b5cfbd0d433e777893212b79301
thread:pr:186652,"pr #186652: [ROCm] Fix ""Ignoring invalid value for boolean flag AMD_SERIALIZE_KERNEL: 3""",waiting on contributor,43,Wait for contributor update on PR #186652; keep related issue/PR context attached.,1,0,,186652,,,0,843395eb308e6339ba26729b4ab61e1c1b022c730c4532315c0f281b5d7ba9ae
thread:pr:186808,pr #186808: [inductor] Re-key range_tree_nodes on IterationRangesEntry rename,waiting on contributor,43,Wait for contributor update on PR #186808; keep related issue/PR context attached.,1,0,,186808,,,0,7b6475c4a3a64d8f019cdfa4a5c5e2f24ba62991e88ceba3622f977b19954ee4
thread:pr:186950,pr #186950: Fixed issue with pickling of custom decomp dict,waiting on contributor,43,Wait for contributor update on PR #186950; keep related issue/PR context attached.,1,0,,186950,,,0,ebcb6b81af952c50aa6d3ef90822d042549f66909313eea765a71bc1e63bbacc
thread:pr:187223,pr #187223: Add targeted SDPA coverage to H100 CI,waiting on contributor,43,Wait for contributor update on PR #187223; keep related issue/PR context attached.,1,0,,187223,,,0,c9005ab80ee071c424ee24a6031584da7bc43cd8939a22264227c3d6e622b63f
thread:pr:187414,pr #187414: [Inductor][cond] Allow output-input aliasing in torch.cond branches (re-land D106106380 + dev-nosan test fix),waiting on contributor,43,Wait for contributor update on PR #187414; keep related issue/PR context attached.,1,0,,187414,,,0,09377a7782b54698d6d9c21676797a0c6e1e9b43db60f27eb2ad8d67fd341375
thread:pr:187434,pr #187434: Support autograd for Tensor?[] (optional tensor list) outputs,waiting on contributor,43,Wait for contributor update on PR #187434; keep related issue/PR context attached.,1,0,,187434,,,0,73b54153b7e2ba992308693502297b709b35bb384f177cc5f154774b907842eb
thread:pr:187450,pr #187450: [ROCm] periodic-rocm-mi350: run only the 4-GPU distributed tests,waiting on contributor,43,Wait for contributor update on PR #187450; keep related issue/PR context attached.,1,0,,187450,,,0,aa00f20db6df139a60563f9d345a5310748d2cf9919c01ae57bffb84564c7062
thread:pr:187492,pr #187492: Avoid std::views::reverse in CUDACachingAllocator,waiting on contributor,43,Wait for contributor update on PR #187492; keep related issue/PR context attached.,1,0,,187492,,,0,ddb9cc0f478aa61863ee3d19b0cd46d3818cb498e9bcd53e477ab81d52f9ccac
thread:pr:187573,pr #187573: [TMA] Fix AOTI eager scratch buffer missing CACHE_TORCH_DEVICE declaration,waiting on contributor,43,Wait for contributor update on PR #187573; keep related issue/PR context attached.,1,0,,187573,,,0,90ae2b717e365ff65703336590ce1a0038b2a336a78b066cb20d5a0edb935835
thread:pr:187600,pr #187600: [ROCm][Inductor] Enable NHWC convs by default on CDNA with an Inductor layout-opt gate,waiting on contributor,43,Wait for contributor update on PR #187600; keep related issue/PR context attached.,1,0,,187600,,,0,ea588f9cb71371ccbcefe03c30a19d7a1cc587527889cbf388a0a443fd51d8f7
thread:pr:187603,pr #187603: [executorch][dynamo] Import torch._utils before Dynamo variable registration,waiting on contributor,43,Wait for contributor update on PR #187603; keep related issue/PR context attached.,1,0,,187603,,,0,91c61c691e0cd2b577ee3d7b01ed5ece6d057cbea63ce094e0415c5413d84dcf
thread:pr:187654,pr #187654: [dynamo] Remove hardcoded CUDA references in test_logging.py,waiting on contributor,43,Wait for contributor update on PR #187654; keep related issue/PR context attached.,1,0,,187654,,,0,8bece9402f724d3cd9df59e25d9b585b348c646c8dca8ebd4757e0115de32c89
thread:pr:187747,pr #187747: Replace CUDA APIs with generic torch.accelerator equivalents,waiting on contributor,43,Wait for contributor update on PR #187747; keep related issue/PR context attached.,1,0,,187747,,,0,b6586219b6b14295371d8a9702b966f49478fd5d8ba2da9b974e94c999363d99
thread:pr:187754,pr #187754: [testintro] Add a test-introspection tool,waiting on contributor,43,Wait for contributor update on PR #187754; keep related issue/PR context attached.,1,0,,187754,,,0,6f4b7b839cf1f7ea8fd2ec57227b672ad6eab505e61bf5896b8bb9ce4bc64874
thread:pr:187755,pr #187755: [testintro] CI: post tests added/removed as a PR comment,waiting on contributor,43,Wait for contributor update on PR #187755; keep related issue/PR context attached.,1,0,,187755,,,0,b47ec04bd57bfbd2b3c11e0f5523bdff0b3548dd2d44adc7e1dcb8b62ae98f8b
thread:pr:187772,pr #187772: perf(dynamo): add opt-in guard lookup token plan,waiting on contributor,43,Wait for contributor update on PR #187772; keep related issue/PR context attached.,1,0,,187772,,,0,4e66ee2bc1742048cfa40980e46179bad7d6f30d03b387f37190f09f4d76eb70
thread:pr:187828,pr #187828: Add privateuse1 backend support to has_triton(),waiting on contributor,43,Wait for contributor update on PR #187828; keep related issue/PR context attached.,1,0,,187828,,,0,b1e2fbb9f43fb987a8ac327cf17bb90acadf864dc16b64c62de785e5eddd67e7
thread:pr:187940,pr #187940: [xpu][inductor] fix the fallback behavior of maxpool backward for performance,waiting on contributor,43,Wait for contributor update on PR #187940; keep related issue/PR context attached.,1,0,,187940,,,0,3afcd4841e9b1762c3e4575e96fe3f379cad5e3e5ca6823e6611362b70af0b03
thread:pr:188015,pr #188015: [SDPA] Fix 32-bit dropout RNG offset overflow in mem-efficient attention,waiting on contributor,43,Wait for contributor update on PR #188015; keep related issue/PR context attached.,1,0,,188015,,,0,b25d75fb7df57627b4e94bc551251fadd92fd3acdf25f2e9c4ea0a0a10802d9d
thread:pr:188086,pr #188086: Migrate torch_export_python to ShapesSpec API,waiting on contributor,43,Wait for contributor update on PR #188086; keep related issue/PR context attached.,1,0,,188086,,,0,1e85bb5f257397e8c14918cfb00c6d749a4ead14d00ccc537e4903eaa4f91022
thread:pr:188095,pr #188095: [ROCm][FlexAttention] Default PRESCALE_QK and auto-detect contiguous blocks,waiting on contributor,43,Wait for contributor update on PR #188095; keep related issue/PR context attached.,1,0,,188095,,,2,111a212022d5a7c61e550dd01ddf7b30a6b4f7691fd14d5a99ac5f1bb2f993ba
thread:pr:188331,pr #188331: [TEST] Refactor test_profiler_tree.py with hw_classification,waiting on contributor,43,Wait for contributor update on PR #188331; keep related issue/PR context attached.,1,0,,188331,,,1,a33f310c5576b94f35b8e3862611ab8d4a75267632082b7f22bc2eea8a08f8cb
thread:pr:188365,pr #188365: [MPS] Remove support for macOS 14,waiting on contributor,43,Wait for contributor update on PR #188365; keep related issue/PR context attached.,1,0,,188365,,,0,88a34ad83e28c4d489a0e9f0e0c992358134a223797b40b60e704e99e2811cd0
thread:pr:188374,pr #188374: [HOP] Deduplicate lifted args for switch,waiting on contributor,43,Wait for contributor update on PR #188374; keep related issue/PR context attached.,1,0,,188374,,,0,f91cdaba07b533f1fd66a54cb0936e76675e1f0ed3a2c42ddfbe724c77ffa4c8
thread:pr:188395,pr #188395: Refactor Dynamo side effect replay into registry dispatcher,waiting on contributor,43,Wait for contributor update on PR #188395; keep related issue/PR context attached.,1,0,,188395,,,0,c99f22e9248d209273f9fd41ce9a02f7200940289c68e6469b5eb60d0ac3bfdf
thread:pr:188600,pr #188600: [ROCm][inductor] Support grouped GEMM Triton lowering,waiting on contributor,43,Wait for contributor update on PR #188600; keep related issue/PR context attached.,1,0,,188600,,,1,19b2fb7326bf17a52266266d6dec31f2c16f4f0598af70d8244338a0fbd05683
thread:pr:188608,pr #188608: [ci] Re-enable cuDSS installation in CI Docker images so CUDA CI can actually build and run cuDSS tests,waiting on contributor,43,Wait for contributor update on PR #188608; keep related issue/PR context attached.,1,0,,188608,,,0,9efeda1a140bfb5224291c6b34007a38907bc31e0534f6467cae05d08d23ab57
thread:pr:188742,pr #188742: [ROCm][inductor] Enable scaled grouped GEMM Triton lowering,waiting on contributor,43,Wait for contributor update on PR #188742; keep related issue/PR context attached.,1,0,,188742,,,1,3f85a4ad118da6f0f472247d040882a4be9ad6abfa6bf83bf907942e37cbb98f
thread:pr:188954,pr #188954: [MPS] add matrix exp on MPS,waiting on contributor,43,Wait for contributor update on PR #188954; keep related issue/PR context attached.,1,0,,188954,,,1,5e505425696354bc2ff81ec5af84e81ebc904b165e9dc9e2a61344a94b41cbdb
thread:pr:189183,pr #189183: [inductor][mtia] Fix fp16 reduction accumulator dtype mismatch,waiting on contributor,43,Wait for contributor update on PR #189183; keep related issue/PR context attached.,1,0,,189183,,,1,4bf1e5a13b6f25457dd90249c42320e759074e2f864fbf45dd100d0db80104f8
thread:pr:189207,pr #189207: Scope cuDNN Engine 5 errata to SM 8.6 GPUs only,waiting on contributor,43,Wait for contributor update on PR #189207; keep related issue/PR context attached.,1,0,,189207,,,0,f08ee38f77805050e9a127e5c5f9a6729fd41c1c0fcc333b3b4ab1a8f93487be
thread:pr:189284,pr #189284: Add torch.autograd.graph.node_creation_hook,waiting on contributor,43,Wait for contributor update on PR #189284; keep related issue/PR context attached.,1,0,,189284,,,2,209750388ef526b24fc12fcaa6a80db3b129613da7c99f6c50fadff177359810
thread:pr:133296,pr #133296: Update file name with onednn,draft/noise,41,draft/noise,1,0,,133296,,,0,9b4786559ab52abae5435ed074ea197fc20907f2b664c3bdd62aff91c6efc5ea
thread:pr:133315,pr #133315: Update non user APIs,draft/noise,41,draft/noise,1,0,,133315,,,0,aed3f01e5cd69973279e5debbc370313e5651500b5afe5dbda49b78405523fb0
thread:pr:133392,pr #133392: Update DispatchKey and namespace,draft/noise,41,draft/noise,1,0,,133392,,,0,f86144ee81b125cf35173de319afd40d9eb2e0e059661e6d40fdb08e63a1dff6
thread:pr:133419,"pr #133419: Keep is_mkldnn, add is_onednn",draft/noise,41,draft/noise,1,0,,133419,,,0,90dd35ec8278d84fd0a29eab3ba57f47254341fee293b2eed1fa99b5f370b3a9
thread:pr:174864,pr #174864: [XPU] Fix accuracy issue in addmm for bf16/f16 dtypes,draft/noise,41,draft/noise,1,0,,174864,,,0,ec02bd9885f81bb5c981b27300f01aac39edb6d45f250f1ec03154624fdbfde5
thread:issue:108158:issue:133877:issue:136312:issue:137014,issue #108158: TORCHELASTIC_RESTART_COUNT doesn't seem to be broadcasted to all worker,stale/low urgency,38,stale/low urgency,5,0,"108158, 133877, 136312, 137014, 148048",,,,4,30f329fb8aeac000958489511cd7e1745d1165fe67063db8c72af8e4df92cde4
thread:issue:146414:issue:146647:issue:146980,issue #146414: MX basic dtypes in pytorch/pytorch,needs design decision,38,needs design decision,3,0,"146414, 146647, 146980",,,,3,ab25aeb604ef814dba9c6ae0d4e210b30b909391d849decf1d822422bd05c608
thread:issue:158827:issue:164302:issue:177010,issue #158827: [RFC] Graph generalization,needs design decision,38,needs design decision,3,0,"158827, 164302, 177010",,,,2,3816063e0b4ee4b26771906d2cf002c569d9827c8aa6e58eeebeed766b3284f1
thread:issue:47117:issue:58212:issue:63485,issue #47117: ctx.save_for_backward doesn't save torch.Tensor subclasses fully,needs design decision,38,needs design decision,3,0,"47117, 58212, 63485",,,,3,bb1bcddae727d3e69440fa378975f32ac13cbce42abc2d61812bb7943bcc77c7
thread:issue:108046:issue:110080,issue #108046: Enhanced Available Backend Discovery and Selection in PyTorch 2,stale/low urgency,38,stale/low urgency,2,0,"108046, 110080",,,,2,e256a4e430339f498036a897f29470f7688d0da361c1eb112f6d0f5ceddee8ee
thread:issue:117844:issue:95100,issue #117844: torch.fft.rfft: Unsupported dtype BFloat16,needs design decision,38,needs design decision,2,0,"95100, 117844",,,,1,906b21244ffd743562d5d9975a3b03744cf23ee8ac3b773b3a0d862803013cef
thread:issue:137574:issue:170839,issue #137574: [feature request] Provide FlexAttention as a new available/selectable backend for SDPA,needs maintainer decision,38,needs maintainer decision,2,0,"137574, 170839",,,,1,8d518f1a12bc204b75845527cc6472effa54b41230a3ad67c59caef75e5c4d16
thread:issue:160230:issue:187698,issue #160230: Offer official Pytorch Vulkan backend on pytorch.org,needs design decision,38,needs design decision,2,0,"160230, 187698",,,,1,93b071a11d6a24de6c31160910d6d0c10b46b53f8d0296c542007af6a6586a62
thread:issue:105053,issue #105053: Multiple dimensions support for `torch.max`,needs design decision,38,needs design decision,1,0,105053,,,,0,ca173ecd458c07ae963db657fa1d7fba906a6ec618828fdab04652f7ff949793
thread:issue:110485,issue #110485: [export] `torch.tensor(0)` should not get burned in as a constant,needs design decision,38,needs design decision,1,0,110485,,,,0,504b6b491cc3f2f3621482747f83f4c7e2314e57edc882de4dd589ddf96a0454
thread:issue:113022,issue #113022: Operator with only `Tensor[][]` args unsupported by dispatcher,needs design decision,38,needs design decision,1,0,113022,,,,0,cfc0cb32769e589048765e91da9585c254ca8e771491386b0e934876328fb017
thread:issue:113737,issue #113737: torch.compile + SAC: mutations in backward are not preserved,needs design decision,38,needs design decision,1,0,113737,,,,0,4c8d5d9eb8d8aa3684520c8423c5a3f8d117c24f5f8cfaa50fbef4e9050e3a03
thread:issue:113800,issue #113800: InternalTorchDynamoError rewrapping loses exception chaining,has linked PR,38,has linked PR,1,0,113800,,,,0,bd77edce95ee116022f7ce7e0214ae41f6b85d2ac24c81ec34f1bed39b147263
thread:issue:115516,issue #115516: [FakeTensor] FakeTensorMode dispatch won't change generator state,needs design decision,38,needs design decision,1,0,115516,,,,0,219d424fef6799eb07cb37849c2a40cd6dd207b746481be62b36abbab2497c4f
thread:issue:120376,issue #120376: Stochastic rounding in bfloat16,needs design decision,38,needs design decision,1,0,120376,,,,0,59cdd0556a07af5373706d315b3e61590f398df3594e068f511c350bb3d9fca0
thread:issue:122192,issue #122192: extra checks for reduce-overhead inconsistent when warmup/recording,needs design decision,38,needs design decision,1,0,122192,,,,0,d73e19607d02c713154b2928956b189ca753959928664534aef05f2d7efbafc5
thread:issue:124296,issue #124296: Request for modifying native allocator to use nvidia unified memory `cudaMallocManage`,needs design decision,38,needs design decision,1,0,124296,,,,0,f4d2d49d804de9e5f3e7d45acc300af0b21467e273922b10306c86a2422e7b75
thread:issue:132651,issue #132651: fix unnecessary re-trace in AOTDispatcher in subclass code path,needs design decision,38,needs design decision,1,0,132651,,,,0,65f34ba87f75a3ec201e271855c0461ac246649fb32a9c8908782d99f003c90e
thread:issue:139444,issue #139444: registering op through function does not work,needs design decision,38,needs design decision,1,0,139444,,,,0,c40a922dc28fcd0eaaa1e7df501958de16cf83ab18f09909fa536644554b43c9
thread:issue:140835,issue #140835: auto-grad graph replicate split_with_sizes(lengths) X times where X = len(lengths) effecting compile time,needs design decision,38,needs design decision,1,0,140835,,,,0,c562b62272e393d3559667a2b76d9c675ed3d92004b55940053af9c91ae7752e
thread:issue:141452,issue #141452: [Dynamo] Keep weak reference of parameters instead of accessing them by attributes,needs design decision,38,needs design decision,1,0,141452,,,,0,c1bf79bfd87dba1925b68eb7da4e9f0cf53e85865aa9e0753cf212b9e719eb5f
thread:issue:146804,"issue #146804: `torch.library.register_fake` respects only positional order, but not kwargs order",needs design decision,38,needs design decision,1,0,146804,,,,0,196a05a148f27e57f2ce8f9d0a69e5e71065c588c5a0f897d535e99c4f4ea92f
thread:issue:149258,issue #149258: Auto-selective activation checkpointing is not optimal for speed (issue with min_cut_rematerialization_partition),needs design decision,38,needs design decision,1,0,149258,,,,0,eeaf46b5bc28dd8fc9749f5764b619dee2a414efafe61821a53ef8378f8cec37
thread:issue:149559,issue #149559: Improve handling for custom ops with no returns,needs design decision,38,needs design decision,1,0,149559,,,,0,b61619d6a6e85a053e435776bdc7e777852d32a4798dfd256f0755892190e98c
thread:issue:151142,issue #151142: AOT Dispatcher converts a single `detach` call to multiple `aten.detach`,needs design decision,38,needs design decision,1,0,151142,,,,0,14e74b9b73350654b7c1c7fcbe2df0abaa7340335e51330941595a67abef3141
thread:issue:152323,issue #152323: compile generates inefficient code when mutating small slice of a graph input,needs design decision,38,needs design decision,1,0,152323,,,,0,79664fb4058911e36efd9bef22b5283cce5cf93acc02faf780a3ecac9f32c4ab
thread:issue:158346,issue #158346: Compatibility with C++20 `std::format`,needs design decision,38,needs design decision,1,0,158346,,,,0,93c3672273fb6cd262f4acf50757de086cc94d96b1e084466f797b7a6fa2de3a
thread:issue:169803,issue #169803: [RFC] Add RGE256ctr: A Fast Counter-Mode ARX RNG for PyTorch GeneratorImpl,needs design decision,38,needs design decision,1,0,169803,,,,0,87b1567a348c23172586a3eb0ed5e20b4d2c8dee90bfa5cb5fdc052fa3d9c652
thread:issue:170224,issue #170224: Way to ask Dynamo to hoist some warmup logic to the beginning of the compile region,needs design decision,38,needs design decision,1,0,170224,,,,0,d69e4e1909d059dd76b72294fcdfdef6b4042b84d450e5b637b995ab228b68cd
thread:issue:171676,"issue #171676: PyTorch allows `tensor.storage().resize_(0)`, leading to segmentation fault on subsequent tensor operations",needs design decision,38,needs design decision,1,0,171676,,,,0,e48ea53b00656890842e8bfbcc4a559ea63ffd5b29eca8bf4515451aeeee3e52
thread:issue:175862,issue #175862: Adding Laplacian smoothing to MultiHeadAttention,needs design decision,38,needs design decision,1,0,175862,,,,0,1e40fc1dd65732ccc30905e1d5e11acb331ec80da99fb0bb0cc00a01045e3eeb
thread:issue:176837,issue #176837: FlexAttention: extend AuxRequest to return min_scores and key indices for row-wise max/min,needs design decision,38,needs design decision,1,0,176837,,,,0,08e5f5c545574576d24f5c5013e906a1fe2d0a5f59bfd5b857ab9afcfd49741c
thread:issue:177537,issue #177537: [RFC] nn.BlockAttentionResidual: depth-wise softmax attention over layer outputs as a drop-in residual replacement,needs design decision,38,needs design decision,1,0,177537,,,,0,a89e06d3ebba9f6cae37aff2ce2f33ade5f60101e587d84ef0694f0dac642572
thread:issue:177623,issue #177623: add support for Predictive Coding as alternative to Backpropagation,needs design decision,38,needs design decision,1,0,177623,,,,0,a0423d31232fab8fa4bef46b7f4cb752b959a52d99cd8e98214cdac89f331aa1
thread:issue:178586,issue #178586: `make_graphed_callables` silently drops in-place mutations on user-provided input tensors,needs design decision,38,needs design decision,1,0,178586,,,,0,e2922004eaf86ca5e1fae1484a04925dede51b625fff576bcd8f38ccd61dfef8
thread:issue:180107,issue #180107: Allow getting `AtenTensorHandle` from the tensor object via AOTI stable C API,needs design decision,38,needs design decision,1,0,180107,,,,0,641aac6a3956ca01bf3c4bc21d67a77133ed79bb33302879c5b2f5c36be1f57f
thread:issue:180215,issue #180215: [RFC] Best practices for backend installations requiring +cpu / non-PyPI wheels,needs design decision,38,needs design decision,1,0,180215,,,,0,93ec5ab9031719efbcda524aea86126564c22f250a625df3f8b627b0d34aef5f
thread:issue:180845,issue #180845: [Dynamo] Unsupported Tensor.item() call in list comprehension when capture_scalar_outputs=False,needs design decision,38,needs design decision,1,0,180845,,,,0,58b1314d9615d99c6876aafb3824694c2c7fd10b7b35caa5b9ac18dd5926012b
thread:issue:184517,issue #184517: LRScheduler doesn't use _get_closed_form_lr when initalizing scheduler,needs design decision,38,needs design decision,1,0,184517,,,,0,873412555918d31554e09597474aec8487229fb4dad982d2077fb993b543adaa
thread:issue:188216,issue #188216: Feature request: built-in loss extrapolator,needs design decision,38,needs design decision,1,0,188216,,,,0,b8d9d98675468406cc7dd3d8786f163df7586b48dcf37e16bec62ca6e679d366
thread:issue:189014,issue #189014: [RFC]: Generalize memory snapshot APIs under `torch.accelerator.memory`,needs design decision,38,needs design decision,1,0,189014,,,,0,404d018065f0eb762add79917e5e5d1622c95d916c3e2e6c2b22d22e5b6513f1
thread:issue:77171,issue #77171: Allow users to express fused matmul/bias/relu,needs design decision,38,needs design decision,1,0,77171,,,,0,3dfdd508fd8fc5cd7e434c6877bcff893d4b6dc2359e85979cdbecc8f384af82
thread:issue:9886,issue #9886: [doc] functionalities not documented,good volunteer slice,38,good volunteer slice,1,0,9886,,,,0,6583a693b3fc2758a82a67c8c9c0789f5891902122cc7319fe1d3633b26245a2
thread:issue:99149,"issue #99149: Spectral Normalization can not be applied to Conv{1,2,3}d",needs design decision,38,needs design decision,1,0,99149,,,,0,ac66ab957ab02e7f29b1e59573a1d4a8aeed65a80a59f7de3615dad2c0ac18f2
thread:issue:113490:pr:148674,pr #148674: Set /NODEFAULTLIB:vcomp for MSVC when linking caffe2::mkl with libiomp5md.lib,has linked issue,35,Review active PRs and linked issues as one maintainer work thread.,2,0,113490,148674,,,1,564fcb7b3218154a3c9c6f91bd133253a06cf9364717088e7e43194e7e486cc6
thread:issue:122191:pr:181308,pr #181308: [CPU] Fix std_mean/var_mean all-reduce ~4x slower than std + mean (#122191),has linked issue,35,Review active PRs and linked issues as one maintainer work thread.,2,0,122191,181308,,,3,3ef6e9df6948c035066d10fa9276302b9317d2172e2669ac2c16c64dd3a0c994
thread:issue:143308:pr:186039,pr #186039: Fix Dynamo verbose guard checks for frame locals,has linked issue,35,Review active PRs and linked issues as one maintainer work thread.,2,0,143308,186039,,,1,8fdf48f51e618bdd09cedc4da5f62fde0ba539f5062ce8d5f887992ed0310119
thread:issue:162980:pr:188906,pr #188906: Add torch.utils.checkpoint.register_determinism_check,has linked issue,35,Review active PRs and linked issues as one maintainer work thread.,2,0,162980,188906,,,2,b9629fd4e8f71b4490af28284dd7701c35a105a8fc735760270e6da993bca004
thread:issue:165097:pr:189174,pr #189174: Fix dcp privateuse1 stream support,has linked issue,35,Review active PRs and linked issues as one maintainer work thread.,2,0,165097,189174,,,2,568f9187960205fc1fe3472c5501cd65d73b1621622dfc3e3c5732b9ba23b68b
thread:issue:187266:pr:176265,pr #176265: Add backward batching rules for SDPA and fix attn_bias under vmap,has linked issue,35,Review active PRs and linked issues as one maintainer work thread.,2,0,187266,176265,,,1,73227a7d6a121c2f19eb6fc5870a0f4ea4ccc4e996558fbcb960ccedc6436e33
thread:pr:134942,pr #134942: Enabling ATen Distribution kernels for AARCH64 using OpenRNG,waiting on contributor,35,Wait for contributor update on PR #134942; keep related issue/PR context attached.,1,0,,134942,,,0,616221773f111e436b0cf547da1c83fe16710274d06586b291c7efcfbc2371e4
thread:pr:138388,pr #138388: Extending SVE VEC Backend Support in PyTorch to SVE128 and SVE512.,waiting on contributor,35,Wait for contributor update on PR #138388; keep related issue/PR context attached.,1,0,,138388,,,0,9a6d09a05c7cac1cbd24f88e950bd0be65567ade302847fd84a374c808507619
thread:pr:148023,pr #148023: torch.utils.checkpoint preserves torch function mode stack during recompute,has linked issue,35,has linked issue,1,0,,148023,,,0,defe1f058ac6bba53bb62c185ece58b9455190cb46c33f66008bbd951dbcd308
thread:pr:166388,pr #166388: Fixing assertion fail by checking if no_sync is true before DDPsink,has linked issue,35,has linked issue,1,0,,166388,,,0,86d055c14f3fa18efe88a1da5e79d1ccb99dec7d4b5f678a6299f65526fc9277
thread:pr:172219,pr #172219: [XPU] Adapt test_basic_chrome_trace to XPU,has linked issue,35,has linked issue,1,0,,172219,,,0,1cd3b8278901a2df2a1dc713ef7e6faf41f7d7b5665e260bc2378c2832756345
thread:pr:172539,pr #172539: [torchgen] Enable Structured Kernels and Out-as-Primary Wrappers for PrivateUse1,has linked issue,35,has linked issue,1,0,,172539,,,0,48a017b87b9b4e70aa6e212d4977149dd1a48d3efa21657b81ac5bb78ab5e18a
thread:pr:172829,pr #172829: Fixes registration of custom sharding strategies when using register_sharding with TensorList,has linked issue,35,has linked issue,1,0,,172829,,,0,eab439a9d645f94c776fec4bafb2f0fe2383c88d8e5755ecee4a6df9ca395743
thread:pr:173239,pr #173239: Test openreg inference hooks,has linked issue,35,has linked issue,1,0,,173239,,,0,7ee87658807a7a3ac8f962777e7109d331e592dfd7d731cf87b14b3125a683c1
thread:pr:173894,pr #173894: [nn] Fix softplus numerical instability with large beta values,has linked issue,35,has linked issue,1,0,,173894,,,0,f4ae7a2a2acfbb96dd67abb4e4a78118a65065142abe11ed9c70a77aa486a65b
thread:pr:174196,pr #174196: Add validation for lstm.data hidden state shapes to prevent segfaults,has linked issue,35,has linked issue,1,0,,174196,,,0,5b9e15ce7228123fe3959e09a32c870a2714668076fc056ee75a77b192d20063
thread:pr:174559,pr #174559: [lint] Enable Ruff PYI056 by replacing __all__.append/extend usage,has linked issue,35,has linked issue,1,0,,174559,,,0,eb08710a386f0619c96a4fae53915c6ca052ba0ae6bdda54449b67a74d81a1d4
thread:pr:175509,pr #175509: Add fast path for min/max dim reductions with size 2 on CPU,has linked issue,35,has linked issue,1,0,,175509,,,0,e6e3d46edd9dbb99e012c06baac16f7165ec98a99075d60f2a7085be4f02b53f
thread:pr:175525,pr #175525: vmap compatibility with non-tensor leaves,has linked issue,35,has linked issue,1,0,,175525,,,0,e58096939b6944377be072573d15bcc5e4659f8cd490cb4c51c1e41842e57252
thread:pr:175542,pr #175542: [DTensor] Register DTensorSpec as pytree constant at module level,has linked issue,35,has linked issue,1,0,,175542,,,0,20a1dfb38e83e956e470c537ddd56d0283686d96eac9e10c336249a658181991
thread:pr:176434,pr #176434: [export] Move schema deserialization to C++ for faster export.load(),has linked issue,35,has linked issue,1,0,,176434,,,0,5f1bf329569bfee2dc36d7fb55ea8a643bd70de3a9e1bae9a0cf2217774a4f0f
thread:pr:176746,pr #176746: Added 'gelsd' to CUDA torch.linalg.lstsq driver so that underdetermined least square systems can be solved on CUDA,has linked issue,35,has linked issue,1,0,,176746,,,0,b4da2ff0267ff5facc0ec16d4d978301437d3ce2cdb941841c2f352f553c1d42
thread:pr:177455,pr #177455: Add nn.SwiGLU activation function,has linked issue,35,has linked issue,1,0,,177455,,,0,fdc29c081b48a9706f7d76a0568f6df513bf758953efcff388ee957c67d3d9ac
thread:pr:177866,pr #177866: Sparse semi structured setup/teardown latency,has linked issue,35,has linked issue,1,0,,177866,,,0,e3c75037a29f86534d14563eb87707f4216c2538f6d2923a13d10515b136f85a
thread:pr:177961,pr #177961: [ROCm] Enable native AsyncTP,waiting on contributor,35,Wait for contributor update on PR #177961; keep related issue/PR context attached.,1,0,,177961,,,0,88859c3972b87c4c7b39a7c1cc1b50f9b57fad0d54e50d06e306e9bafddea716
thread:pr:177985,pr #177985: Fix SDPA mem-efficient attention crash when num_heads >= 65536,has linked issue,35,has linked issue,1,0,,177985,,,0,eb474269ee6f2587bd10227f7c4afce518abc1aad22bf8db257c871753260862
thread:pr:178031,pr #178031: [MPS] Add adaptive_avg_pool3d support for MPS,has linked issue,35,has linked issue,1,0,,178031,,,0,cc6c639d68da83d8d970fc36fa45fb1d6c7e666f7134c9f1a3506f4fbaa584fa
thread:pr:178099,pr #178099: Add TypeGuard[Tensor] return type to is_tensor_like,has linked issue,35,has linked issue,1,0,,178099,,,0,5835599cfd902a2a472deb8e253ab3830dd697c7cecbe4c21d60a9fe5f1db520
thread:pr:178371,pr #178371: [Profiler] Fix profile_memory not working with profile_all_threads,has linked issue,35,has linked issue,1,0,,178371,,,0,2041e45cd7d38446f2b1ba39307f1d1122e54adc8cce205a219f0167c78d4803
thread:pr:178390,pr #178390: [inductor] fold half addmm+relu into _addmm_activation,waiting on contributor,35,Wait for contributor update on PR #178390; keep related issue/PR context attached.,1,0,,178390,,,0,2bb90fb23cfdb48d28d19801088e87e8b2792fbde0a5207f5e1f9e6c0fea5b38
thread:pr:178797,pr #178797: Fix meta conv shape checks for invalid groups vs out_channels,has linked issue,35,has linked issue,1,0,,178797,,,0,4292cbe6f8f55def4d1401eeabd2bd76ea609a3d74ea8048aa1142f580e762fe
thread:pr:179160,pr #179160: Fix uint32 overflow in CUDA scan _with_indices kernels,has linked issue,35,has linked issue,1,0,,179160,,,0,784b3f8e7425e269f98fcafa4c8075eea9e025b757e3271f64358b7e6e0d2646
thread:pr:179377,pr #179377: Fix gaussian_nll_loss rejecting valid multi-dim broadcastable var shapes,has linked issue,35,has linked issue,1,0,,179377,,,0,44949e8d9a7d3d0e14a33e4bf2371b25312dc16dd21455b244a4f2bd30ff4372
thread:pr:179379,pr #179379: [DTensor] Convert type aliases to TypeAliasType in experimental/,has linked issue,35,has linked issue,1,0,,179379,,,0,f0924ba1554cf94f1c6387ad37d4c975804555cd4b0f5389999e19fa1a080fb0
thread:pr:179827,pr #179827: Replace BF16 matmul OpInfo checks with a feature query,has linked issue,35,has linked issue,1,0,,179827,,,0,cd8470229e54a8013a05a9ce1200c9d3459bf124a5d971194fdd5450478c9a1d
thread:pr:179941,pr #179941: [inductor] Add reduction loop peeling for Triton codegen,has linked issue,35,has linked issue,1,0,,179941,,,0,ba1b175e6cc13f8d43343459c3a8390b606dd65f57a460b79538866258fe2716
thread:pr:180184,pr #180184: Export auto-generated unstructured inplace meta dispatch symbols,has linked issue,35,has linked issue,1,0,,180184,,,0,05258c65a24b69b4bf257162da3f59cb1439597db6dcfb498def8b828bcee928
thread:pr:180188,pr #180188: Validate opt_dtype in nanmean to reject non-floating-point types,has linked issue,35,has linked issue,1,0,,180188,,,0,724609504ebbf4e8ca67a2da06e272fe00bc147e3350eca9b87d9213957b5027
thread:pr:180408,pr #180408: Fix torch.eye CPU performance cliff at d=182 (#48251),has linked issue,35,has linked issue,1,0,,180408,,,0,d70ea1737eb612ef1b6380f18da4bf384e00006e697f63c634b704a7da434a50
thread:pr:180444,pr #180444: Add `validate` parameter to `torch.multinomial` to skip input validation,has linked issue,35,has linked issue,1,0,,180444,,,0,8c6136d0753ea8dca1fc52faafb8083f3b0205ff6afd1506d1b1f0cf44248304
thread:pr:180496,pr #180496: feat: Add OpenTelemetry trace export support for torch.profiler,has linked issue,35,has linked issue,1,0,,180496,,,0,c1f0df0205973f2d5a9691b3c4ab5bde1fcc3b5ad5efdd9b8eaa7353f04f437c
thread:pr:180601,"pr #180601: Fix pydocstyle errors in _guards.py, _ops.py, _jit_internal.py, functional.py, _tensor_str.py, library.py",has linked issue,35,has linked issue,1,0,,180601,,,0,0f6808fda472b717221ee76432688aa907c4e3873f7dcadc7e503242cafe5e2e
thread:pr:180764,pr #180764: cpp_extension: add CUDA dlink flags to JIT load APIs,has linked issue,35,has linked issue,1,0,,180764,,,0,e32f857e38ced6ce384aa0de34c5ca6fe43d366d1ff2c8c9db93da7327cc4189
thread:pr:180781,pr #180781: Fix integer divide-by-zero UB in CUDA binary kernels,has linked issue,35,has linked issue,1,0,,180781,,,0,cbfa7eae383f752acd506f3a7e5dcb96ac42e1cecb8fd6378120cc031c3ca609
thread:pr:181027,pr #181027: Factor nanmean dtype checks into a helper + add kwarg coverage,has linked issue,35,has linked issue,1,0,,181027,,,0,83da43fcc8c69481bc338145b07f761e68d647f4d64bc15904fd6ec124c6941c
thread:pr:181051,pr #181051: Fix heap-buffer-overflow in quantized_max_pool2d for 1d args,has linked issue,35,has linked issue,1,0,,181051,,,0,f8e47e8f88b289d4388e4878f482daccb6c4091f74eef562d21943f41e58f39a
thread:pr:181110,pr #181110: Workaround for nvrtcCompileProgram changing locale,waiting on contributor,35,Wait for contributor update on PR #181110; keep related issue/PR context attached.,1,0,,181110,,,0,5f30f9c32e21788c0432474864101460a31e4df179bd815a42f9edcefc2251be
thread:pr:181352,pr #181352: Re-enable nightly tests on s390x,waiting on contributor,35,Wait for contributor update on PR #181352; keep related issue/PR context attached.,1,0,,181352,,,0,008699d383d60ddc24a5059c9fc4a7ffc77fb2b7c2e50345dbe100d3484712ae
thread:pr:181501,pr #181501: docs: add DataLoader performance guidance for in-memory datasets,waiting on contributor,35,Wait for contributor update on PR #181501; keep related issue/PR context attached.,1,0,,181501,,,0,bff0e0c72b0666da7b8ba8b33e1d8c694db5e27ae4328deb0975245e97b68b53
thread:pr:181520,pr #181520: dataloader: add performance warning for in-memory datasets with num_workers>0,waiting on contributor,35,Wait for contributor update on PR #181520; keep related issue/PR context attached.,1,0,,181520,,,0,090a048c2ea07b69fc0b2b2ca3dc12d277f51f1c81b82a8a3da5093ecc7e8266
thread:pr:181681,pr #181681: Fix UBSan null-pointer UB in vec256 partial-load Vec::loadu (#181510),has linked issue,35,has linked issue,1,0,,181681,,,0,1a152d32495703616b1ddfbe1ff3a5213d75fea6b2bb13e077bf1685667fcd63
thread:pr:181756,pr #181756: [Docs] Replace outdated flake8 and git-pre-commit references with lintrunner,has linked issue,35,has linked issue,1,0,,181756,,,0,b48dde8030e789f274f649c13c8194c5037a93bf0763354b8f5bed4bdd563146
thread:pr:181993,pr #181993: [MPS] fix device binary shaders,has linked issue,35,has linked issue,1,0,,181993,,,0,69032163e962d4e7ba6e303e5fbf100e50d94a216ac72f22cb22f328739bbe99
thread:pr:182076,pr #182076: [Fix] Add overflow and validation checks for pixel_shuffle in decomposition and meta,has linked issue,35,has linked issue,1,0,,182076,,,0,e6c77f94fcf66041d7a4ed72a0f48eb16ce769918217c9fb27149ba4f00a29f9
thread:pr:182133,pr #182133: Allow torch.arange() to return empty tensor when bounds are inconsistent with step sign,has linked issue,35,has linked issue,1,0,,182133,,,0,0fe137637e57d8f92c8064e69c018452dacde81a3bf769a30c4c03c155fa625e
thread:pr:182204,pr #182204: torch.cat/concat/concatenate: support dim=None (flatten before concat),has linked issue,35,has linked issue,1,0,,182204,,,0,7cbd5a6db8abcbb3437f1e0084b21489f3f8b556510faf61964a9f51fb07fee5
thread:pr:182262,pr #182262: Use TypeAliasType for type aliases and @exposed_in for function re-exports,has linked issue,35,has linked issue,1,0,,182262,,,0,e3fe72f4e0e261a1bc62cb09498d6e4c75e80485af8e99ca5bda8f78a1a1589b
thread:pr:182268,pr #182268: autograd: add `list_saved_tensors` utility to traverse saved tensors in autograd graph,has linked issue,35,has linked issue,1,0,,182268,,,0,10c8e2bacae0f06df29ae9ba7071e8dfa210ce9c97af016787ed72ba2782b2f7
thread:pr:182291,pr #182291: fix(nn): InstanceNorm always raises ValueError on channel mismatch,has linked issue,35,has linked issue,1,0,,182291,,,0,7e4f0d7e478a970d09647a447c675c14526b0029d5fdc6914c18b29bd60a8732
thread:pr:182296,pr #182296: fix(nn): add input dim validation to adaptive_avg_pool2d/3d,has linked issue,35,has linked issue,1,0,,182296,,,0,a1f0b95546cbfc409aea70ae344c18116eabeeeb32e68d1cb6ca37788e6c06f5
thread:pr:182384,pr #182384: [autograd] Refresh ctx.needs_input_grad for partial backward,has linked issue,35,has linked issue,1,0,,182384,,,0,c0f7f652c830dadcb9b6c730db3c88c5e8322e3f5267e4b8690bfa417f954d1c
thread:pr:182541,pr #182541: add regression test for weight_norm meta initialization (#96409),has linked issue,35,has linked issue,1,0,,182541,,,0,940a69ceced30c4451a604cc111ae5254ff01a94b092030cd0915974f89bea61
thread:pr:182601,pr #182601: docs: clarify tile dims argument,has linked issue,35,has linked issue,1,0,,182601,,,0,629f010c1b9696e23c4b8ff8373c62e275c12c9dea333ca0ccbc3230e1d9664c
thread:pr:182602,pr #182602: docs: cross-reference sigmoid APIs,has linked issue,35,has linked issue,1,0,,182602,,,0,2a974695944925aa36b174b0b2331308ed761649cd134cb4733f443bd1c87837
thread:pr:182603,pr #182603: docs: clarify RNN hidden state dimensions,has linked issue,35,has linked issue,1,0,,182603,,,0,2d06d02e328674aa212ddd7d0a43148fbff0a3f57ec68df3cb166b9e61d9b8b9
thread:pr:182604,pr #182604: docs: clarify multithreaded autograd guidance,has linked issue,35,has linked issue,1,0,,182604,,,0,33ed7549b20231279d108e80e72ba1b81f51748e62420da57a1a766c1bc9ae69
thread:pr:182605,pr #182605: test: strengthen optimizer step hook ordering checks,has linked issue,35,has linked issue,1,0,,182605,,,0,f478442fc3428e5dbd0025280a91f96538ae386e4ab828926ee754dc8ab26cf3
thread:pr:182612,pr #182612: Ignore decoding errors when reading CMake cache files (errors='ignore'),waiting on contributor,35,Wait for contributor update on PR #182612; keep related issue/PR context attached.,1,0,,182612,,,0,14912cd8535f545bfcf9d804d2cc0cb70c8068cb9bfcb04ce17b26e8046a6383
thread:pr:182627,pr #182627: Handle bool inputs in Scalar::log,has linked issue,35,has linked issue,1,0,,182627,,,0,6d80fe6212fc5c6beb6776793cd440f5b003217e54b2d5d4d03a7b261b5f549f
thread:pr:182945,pr #182945: Add pattern matcher to deduplicate var and std calculation,has linked issue,35,has linked issue,1,0,,182945,,,0,83f0a2835a2b6e98f8d08b47c470fd61fd0d7bd3628eba67c11f38e651362740
thread:pr:183269,pr #183269: [DTensor] Switch test_embedding_ops to DTensorOpTestBase (MultiThreadedTestCase),has linked issue,35,has linked issue,1,0,,183269,,,0,44d1762cc7153c4b5ad6d86d628e742d1212e0fd1331663421b7da7cb59bf2db
thread:pr:183316,pr #183316: Fix/signbit cuda half nan,has linked issue,35,has linked issue,1,0,,183316,,,0,c886373553a5ed2276cc7c6ff6cc3802616098ee3cdddde45a97c451309e8ec2
thread:pr:183433,pr #183433: [inductor] Fix backslash escaping order in user-defined Triton kernel…,waiting on contributor,35,Wait for contributor update on PR #183433; keep related issue/PR context attached.,1,0,,183433,,,0,f07ad4413e2f5afdd37926f1613c0e2ffb4e102b2503f56d703652ceed464fb7
thread:pr:183743,pr #183743: Add drift-based tolerance fallback to op tests,has linked issue,35,has linked issue,1,0,,183743,,,0,68aa0853c1a636751d9b88e0efba3f1f53dfc0d411dcd4bf47ba509bc8c29235
thread:pr:183749,pr #183749: Support auto_functionalized_v2 in can_fuse_into_auto_functionalized,waiting on contributor,35,Wait for contributor update on PR #183749; keep related issue/PR context attached.,1,0,,183749,,,0,1a041c839ad1455f51394648d8701c650990925cb63c49dd38fc224770244ded
thread:pr:184064,pr #184064: Fix silent int64 precision loss in F.pad constant mode,has linked issue,35,has linked issue,1,0,,184064,,,0,28c789e06adf7f7c97ae05dae2da4205fced5ac066c61f60d76cdda528e592eb
thread:pr:184577,pr #184577: nn: validate num_groups > 0 in GroupNorm.__init__,has linked issue,35,has linked issue,1,0,,184577,,,0,e981e2f76be9c40c6942aeddfc5287c817cb3539be5259833a00ded5fc31b46e
thread:pr:184641,pr #184641: Fix GroupNorm validation for non-positive num-groups,has linked issue,35,has linked issue,1,0,,184641,,,0,309d2594512a6ed958d03e0c9e002d8de96a0bfa3f390ec3f6001b0d14d8cae7
thread:pr:184698,pr #184698: Materialize lazy modules fully on load_state_dict instead of leaving them in a partially initialized state.,has linked issue,35,has linked issue,1,0,,184698,,,0,2857d3faf1fe513a21d22651f17a9a6af1f3566fade8d17d546ac0d938773896
thread:pr:184840,pr #184840: Add Module.to overload for memory_format,has linked issue,35,has linked issue,1,0,,184840,,,0,6c508ce6f6c8edfaacb275341a81b4b6a060d5a2e707f8201c6f072af27286f0
thread:pr:184857,"pr #184857: Raise ValueError early when strategy=""forward-mode"" used without vectorize=True",has linked issue,35,has linked issue,1,0,,184857,,,0,406f3cd72f7b16e92ef064e9cd79bada7ed536d631128a855aa96cb9a8eba690
thread:pr:184885,pr #184885: docs: correct torch.addcdiv prose to match implementation order,has linked issue,35,has linked issue,1,0,,184885,,,0,f17cd643be007f6594d18754e98d65e9d51c4ee08758ac2dfb9724a2339e1869
thread:pr:184898,pr #184898: autograd: raise ValueError for forward-mode without vectorize in jacobian/hessian,has linked issue,35,has linked issue,1,0,,184898,,,0,bdb82aad1d51f7106db499f6560133eedd6edb26eeb2105c3df3d3b2e0cf99ab
thread:pr:185026,pr #185026: [typing] Add Iterable base class to ModuleList,has linked issue,35,has linked issue,1,0,,185026,,,0,bf92cc8ee83d7a44cca907e6dcf5dad8b9cff01bf77175c126d03630cbac70d3
thread:pr:185056,pr #185056: Raise error when class attributes shadow registered parameters or modules,has linked issue,35,has linked issue,1,0,,185056,,,0,a4619307560376cebbbec9537f93d8f41f7a4b28d91657fffc55aeeec5e7df2e
thread:pr:185117,pr #185117: nn: add missing memory_format overload to Module.to(),has linked issue,35,has linked issue,1,0,,185117,,,0,e53a2256576caf29783c9e4249a93e96c1d13143a1092b3336eff9540a144140
thread:pr:185187,pr #185187: Validate forward-mode autograd.functional vectorize usage,has linked issue,35,has linked issue,1,0,,185187,,,0,fa11cfb694b1dfcdb762b98fe4622aaa5599fab93b15a735cd3a385f8b7081b7
thread:pr:185193,pr #185193: Validate forward-mode functional autograd arguments,has linked issue,35,has linked issue,1,0,,185193,,,0,e5659afc33a1dec93b2de9fe8d46460b1c6a36f3d64f2a3a6c3466294cd911a2
thread:pr:185319,pr #185319: Sort record names before hashing in writeSerializationId,has linked issue,35,has linked issue,1,0,,185319,,,0,5c3e6b2d5169f804871cd17ea54470f7a85e43efcf121322748a9d44ea2e36ef
thread:pr:185398,pr #185398: docs: clean up multilabel soft margin loss args,has linked issue,35,has linked issue,1,0,,185398,,,0,8d95c1a3ad445afe9931d56b62b79d034a76c3306fab1cb1806c298305473737
thread:pr:185454,"pr #185454: fix clamp and clamp_{min,max} gradient at scalar bound to be 0",has linked issue,35,has linked issue,1,0,,185454,,,0,563a09100ad95522fc8cc38844094758847562da618c9a67aff8cb6d9f644652
thread:pr:185744,pr #185744: docs: clarify LSTM equation notation,has linked issue,35,has linked issue,1,0,,185744,,,0,39bae70c09e170f38d05916c4527ae63b493b5cc7262f3978930f552e2e22cd0
thread:pr:185745,pr #185745: docs: clarify scalar fill values,has linked issue,35,has linked issue,1,0,,185745,,,0,04d8b642a32a07212b20a4be505e959f2e6ecfe4b44de7a98f1e5aa76b323aaa
thread:pr:185747,pr #185747: Use at::mps::OptionalMPSGuard for MPS dispatch keys in codegen,has linked issue,35,has linked issue,1,0,,185747,,,0,a1e0184a3a8deb7d3f9c83ed9aa82b61613478f77e534be65c1162e9f68b0e91
thread:pr:185792,pr #185792: test/distributed/pipelining: add end-to-end tests with HuggingFace GPT-2 and BERT,has linked issue,35,has linked issue,1,0,,185792,,,0,0b8ef060101774336e66017fbf8a9de478d476837bc3ac5383b98cd5e72ea9cf
thread:pr:185794,pr #185794: Fix torch.kron RuntimeError on non-contiguous tensors - noah,has linked issue,35,has linked issue,1,0,,185794,,,1,65f7b0867c3e85597afe252df5cb85316501dc6ee9628fd37cab9eabf42fd237
thread:pr:185807,pr #185807: Stable sorting for consistent gradient for torch.quantile on tied values,has linked issue,35,has linked issue,1,0,,185807,,,0,fdef6d4c750c20a080796818e54273088a543d2a8c36718eb01fce08edd331ba
thread:pr:185864,pr #185864: Fix segfault in native_batch_norm with empty running_mean/running_var,has linked issue,35,has linked issue,1,0,,185864,,,0,674874f7d1f97a51122b1fb1148064c75940142436e76c571b25613c3b15d4cd
thread:pr:185868,"pr #185868: Update clamp, clamp_min, clamp_max to use the minimum norm subgradient",has linked issue,35,has linked issue,1,0,,185868,,,0,3c0f3c429e60de29c7e90496ac81292d2e2a692572b3a8c4f4bda89931f077a3
thread:pr:185959,pr #185959: Fix SIGFPE in torch.lcm with the integer type minimum,has linked issue,35,has linked issue,1,0,,185959,,,0,b169b5b83f48d6ffa8db6c34d4da9147e46e5117d1093a00ef395879b25869f7
thread:pr:186050,pr #186050: Output Shape Check added to transposed Conv,has linked issue,35,has linked issue,1,0,,186050,,,0,3faa6fbebdfcb3dfec15ed85eb2588096e2b8b1f7a41e2d52cbf26bef76be74d
thread:pr:186073,pr #186073: Support weights_only=True in torch.export.load,has linked issue,35,has linked issue,1,0,,186073,,,0,cfc6687b34b7d1e7d8a30bf400d4e0b561ee126f1d2507f9236c7fef40af3908
thread:pr:186128,pr #186128: Reject dtype conversion of quantized TypedStorage with a clear error,has linked issue,35,has linked issue,1,0,,186128,,,0,467819cbd7cfdca73f631a0c87015befec01658145b98d60ceff04708516801d
thread:pr:186159,pr #186159: Restore Any return type for Module.__getattr__ to fix typing false positives,has linked issue,35,has linked issue,1,0,,186159,,,0,5c273564fdccdd29e97fe88d624396ff06969b0fe4af647ec71da54c2de1e197
thread:pr:186177,pr #186177: [autograd] Defer device thread creation when multithreading is disabled,has linked issue,35,has linked issue,1,0,,186177,,,0,d8c8cfa0b60e3d5dd25404216bdc61c75973cc76144b50a19e60d87fa73f0689
thread:pr:186207,pr #186207: Fix segfault and invalid cast in UntypedStorage,has linked issue,35,has linked issue,1,0,,186207,,,0,a568659adc75c56991bf575df95b8081af5372656c680f7d23bc8674cdc42a23
thread:pr:186524,pr #186524: Validate conv transpose output padding,has linked issue,35,has linked issue,1,0,,186524,,,0,d08989b3007f0fa68d79b7867cb1200f3bd957348e45b8991e2e273fb87f94a3
thread:pr:186571,pr #186571: Fix #176298: support unsigned add and sub on CPU,has linked issue,35,has linked issue,1,0,,186571,,,0,3394f3adfbcf4ce4720b44833e2d0c3ecbffe4792aff25a718ca50ccb9b261a7
thread:pr:186589,pr #186589: Fix serialization errors in SubTest calls,has linked issue,35,has linked issue,1,0,,186589,,,0,51aad1c665090fbffe74c62e62e887a99ece1632b39a78fb2e3cf1905723a025
thread:pr:186607,pr #186607: Raise a clear error for malformed allocator config instead of INTERNAL_ASSERT,has linked issue,35,has linked issue,1,0,,186607,,,0,421779236e003115ea4addca8501da86dc62769a06434ff534873064d315d2ad
thread:pr:186863,pr #186863: [MPS] Register unsigned binary Metal kernels,has linked issue,35,has linked issue,1,0,,186863,,,0,bc7d724937c1c0afdbddff3bddb662b527222bd3ab6c7494fa8aa3832fe43f85
thread:pr:186929,pr #186929: [fx] fix _patch_function using wrong function when forward is wrapped,has linked issue,35,has linked issue,1,0,,186929,,,0,7c73a5a7e61a5b8831dca30c52d4efe1912bbec3f6b5db94bb3210a4fa0a1300
thread:pr:186930,pr #186930: Fix NaN propagation behavior in torch.sign for CPU and CUDA backends,has linked issue,35,has linked issue,1,0,,186930,,,0,978c8b32fd19ab111b360f81b2260e4b8a612df5b0ab44d79b5f55fc21b308cf
thread:pr:186940,pr #186940: [ShardedTensor] use device-agnostic fallback in tensor_device (fixes #186938),has linked issue,35,has linked issue,1,0,,186940,,,0,0afb20eec3ecfb8f5961239d8bea12e133bfb2968ddd43f054e5ca2a5860e0fa
thread:pr:187101,pr #187101: Add missing dispatch for bf16 in tensor_cpu,has linked issue,35,has linked issue,1,0,,187101,,,0,8edc9f5b9370b18e1b975881fa0442a7e32aa160e2ed435299cacf3e604b2ea4
thread:pr:187153,pr #187153: jit: add bounds checking for textIndex in SourceRangeDeserializer,has linked issue,35,has linked issue,1,0,,187153,,,0,5672a190f3ba7d303e5e26b9cc0c8d88ab6d50806681d197394e73422c7b7291
thread:pr:187187,pr #187187: [typing] Replace TypeAlias with TypeAliasType in torch/types.py,has linked issue,35,has linked issue,1,0,,187187,,,0,24afd1f34c0e0181d0f39aa0115368a7258bed308a91e51c9039583e436212af
thread:pr:187198,pr #187198: Add target range validation in binary_cross_entropy_with_logits,has linked issue,35,has linked issue,1,0,,187198,,,0,100e74082891e8f6ed61498cb18c3e0c6d8a97d090a7d012c7b948010917ea51
thread:pr:187238,pr #187238: Add docs pages for torch.e and torch.pi,has linked issue,35,has linked issue,1,0,,187238,,,0,e4f00b479d943d89b6d4bf3571790e122dfc699bcb92893c6de569a9d0776c12
thread:pr:187246,pr #187246: serialization: fix nondeterministic pickle bytes in _legacy_save (gh#39383),has linked issue,35,has linked issue,1,0,,187246,,,0,1f8b4c01a37b2dfdf6bb328d6af1f58b4270e351b2732b36b28d55767d4a6914
thread:pr:187253,pr #187253: Align dynamic qparams test helper with C++ overflow check,has linked issue,35,has linked issue,1,0,,187253,,,0,32c2575a2e1f4bfc5c59930215dee0859606ec59641409dda12a762db5c07644
thread:pr:187257,pr #187257: [DCP] Raise RuntimeError on load dtype mismatch to prevent precision …,has linked issue,35,has linked issue,1,0,,187257,,,0,f8737dc5ff5075849ca27a316c8b163a67d80d86ed2a73f9651976b60b6fafee
thread:pr:187286,pr #187286: nn: add torch.nn.functional.rotary_embedding and torch.nn.RotaryEmbedding,has linked issue,35,has linked issue,1,0,,187286,,,0,4aa983bb32dacab2710121f17091e5bb00be8f06366ec867b7d953eb7ee37eba
thread:pr:187309,pr #187309: Add maximize parameter to LBFGS optimizer,has linked issue,35,has linked issue,1,0,,187309,,,0,4c6bbaa7538ab7fdb33a310474a8e4bf3e87d8f8723d2a3f28eabe07c0910b40
thread:pr:187325,pr #187325: [Metal][memory] add configurable memory budget API for MPS allocator (#105839),has linked issue,35,has linked issue,1,0,,187325,,,0,50b48480704dff531e79c3382c42b741f8a04c2c707488ecb5a4571c7b66db49
thread:pr:187327,pr #187327: [Metal][cache] dual-hash verification for MPSGraphCache and MPSKernelCache (#77176),has linked issue,35,has linked issue,1,0,,187327,,,0,923e2d63c82994f928adca6760d999d9247b4ddb3890c34b51ac28f3393469a6
thread:pr:187329,pr #187329: Raise ValueError for non-positive kernel_size/dilation/stride in unfold/fold,has linked issue,35,has linked issue,1,0,,187329,,,0,68cf8d1be3e0592c6b5108382a919107f9b07261d2fa0f2a5828e7af1adfd402
thread:pr:187330,pr #187330: [Metal][memory] clear graph/kernel compilation caches in empty_cache() (#154329),has linked issue,35,has linked issue,1,0,,187330,,,0,46e10c68d399d3b8c4e6fdf7ddbb206ca7ec1956063016b538a2dab7be15f812
thread:pr:187453,pr #187453: [MPS] Add error handling for complex dtypes in max unpooling operations,has linked issue,35,has linked issue,1,0,,187453,,,0,f56a1457bb6f61ef4d321cc8954a07a190db89e69385996eaba77049335d1ae4
thread:pr:187544,pr #187544: Fix torch.gcd returning negative results for INT_MIN inputs,has linked issue,35,has linked issue,1,0,,187544,,,0,971b0a4903b29a95ae3416071c663397d84bb8d8802c662920c87abb2ef7bcfd
thread:pr:187570,pr #187570: Fixed heaviside nan propagation issue,has linked issue,35,has linked issue,1,0,,187570,,,0,eeaf94e52907be3eecebec52f480211f6559df2e9ff8fa95e4dfb58adf14ef52
thread:pr:187595,pr #187595: Fix torch.gcd/lcm integer overflow on INT_MIN inputs,has linked issue,35,has linked issue,1,0,,187595,,,0,cd66cc8d53581d9789f95b15ebc0eb37c46574deaa6d8b6ebabbfe023427d14d
thread:pr:187597,pr #187597: Add return type annotation to nn.Module.named_modules,has linked issue,35,has linked issue,1,0,,187597,,,0,29cf5aa973025fd36483c237a6440900011091083a72cded974f856f61fa7db8
thread:pr:187630,pr #187630: fx/passes: fix non-deterministic ordering in get_source_partitions(),has linked issue,35,has linked issue,1,0,,187630,,,0,16715c502622ecb1186a7268a26bc2bf41537d7ec80a74890a1d5c3642dfc500
thread:pr:187656,pr #187656: Reject complex scale/zero_point tensors in quantize_per_tensor with an explicit dtype error instead of value cannot be converted to type double without overflow.,has linked issue,35,has linked issue,1,0,,187656,,,0,34e1fce746a792488a3f6d1637fe03a22c14592c9dcc6ff59b869cebabecfb98
thread:pr:187662,pr #187662: Reject aliased Q/R out tensors in torch.linalg.qr,has linked issue,35,has linked issue,1,0,,187662,,,0,05b2b8321ba44e8670f3992da225556b74ecebe5cc1174d259e04d128e6264f8
thread:pr:187748,pr #187748: [MPS] Build only the reduced Q in linalg.qr (fix tall-skinny perf cliff),has linked issue,35,has linked issue,1,0,,187748,,,0,66ac6bebd477e75df79ee43a7035221f81f5e699fcded7819b5d24f9c7611b02
thread:pr:187757,pr #187757: Narrow ModuleDict.__getattr__ return type to Module for type checkers,has linked issue,35,has linked issue,1,0,,187757,,,0,f377956a90af9f096674ef1aa4e8c88aac17d769a2ef36d97a34c28b360b1d33
thread:pr:187764,pr #187764: [MPS] Add matrix_exp support,has linked issue,35,has linked issue,1,0,,187764,,,0,51ef1d0f1a0384bb2b26290ee1eddbf670a08e12a84da6ebc22010e294752ccb
thread:pr:187771,pr #187771: Warn when convert_sync_batchnorm drops _BatchNorm subclass behavior,has linked issue,35,has linked issue,1,0,,187771,,,0,c9c7d7db4509b227105704cde0e1b7c00d02da38eebf7e42cd119369e72c962d
thread:pr:187783,pr #187783: fix(nn): raise ValueError instead of ZeroDivisionError for norm_type=0 in lp_pool1d/2d/3d and LPPool modules,has linked issue,35,has linked issue,1,0,,187783,,,0,1fde6646be573e4c4b42f896d2489b27e65c013f1d915d9275a3a01446160ec1
thread:pr:187790,pr #187790: [ao][pruning] Add experimental topology-aware modules,has linked issue,35,has linked issue,1,0,,187790,,,0,f39a46c6dc46d69eb0782ac818a9ddaadb3e6bac8d6395b6aeb94c909f4d7b15
thread:pr:187795,pr #187795: Guard release_available_cached_blocks during CUDA graph capture,has linked issue,35,has linked issue,1,0,,187795,,,0,247d982a278abfb3efac0ca12e8b6966774272e53ec6cc2de5283ba19f1d38a9
thread:pr:187803,pr #187803: fix(autograd): sinc_backward NaN for second derivative at x=0,has linked issue,35,has linked issue,1,0,,187803,,,0,e611c090030e69465170a9548f663b671bd073e1e23f3fa44526d7850fbd0e7f
thread:pr:187814,pr #187814: Reject factor=0 in ConstantLR to avoid later ZeroDivisionError,has linked issue,35,has linked issue,1,0,,187814,,,0,9536f721c388ff3e1a5da5b5b0dafaff36cc3fe49b9d59634df19d8b3fce320b
thread:pr:188099,pr #188099: [JIT] Use native fill/zero ops for outplace tracing of fill_/zero_,has linked issue,35,has linked issue,1,0,,188099,,,0,e44aab5c6c5b83e049b6c768d7751984cb6ad853179a6f4070b8b7bade967452
thread:pr:188138,pr #188138: fix: Docs are little bit outdated for torch logs,has linked issue,35,has linked issue,1,0,,188138,,,0,d6a2dca337fa159ce397ad3b37340d9372b388219bc59503c41806278fbf9e9a
thread:pr:188149,pr #188149: Handle mixed Tensor/Scalar min/max bounds in torch.clamp,has linked issue,35,has linked issue,1,0,,188149,,,0,7b86715353f5333f84753087c0f4b3c9841daf4c184111103e519c8708d7fe06
thread:pr:188183,pr #188183: [nn] Avoid mutating state_dict metadata in load_state_dict(assign=True),has linked issue,35,has linked issue,1,0,,188183,,,0,fc7438e47036b94c1c88070fe6a89d8609b82fd03bed1d7592794dee0f52716c
thread:pr:188200,pr #188200: [MPS] Vectorized contiguous cat fast path for any dimension,has linked issue,35,has linked issue,1,0,,188200,,,0,463ec4828259f2ab6f0acf5ad5ea653997b052a1487f5254a6034caca2e8a7d4
thread:pr:188258,"pr #188258: Fix _foreach_{add,mul,div}.Tensor autograd",has linked issue,35,has linked issue,1,0,,188258,,,0,8fbf8c855ef5fae84df2b66a3f3d099a081701d4d6a250c6dc1f30faa75c3927
thread:pr:188260,pr #188260: [nn] Fix RReLU noise tensor not being set in eval mode,has linked issue,35,has linked issue,1,0,,188260,,,0,e457ff809e295ffd599b23e979492daefb53bdcdcedf7ef6d45471ac80df8c89
thread:pr:188270,pr #188270: Fix inplace modification error with softmax when combining forward and backward AD,has linked issue,35,has linked issue,1,0,,188270,,,0,dff891c5917db40eba8f6464a97e6243776dea60c55a429595e187789f80292e
thread:pr:188271,pr #188271: Allow PrivateUse1 backends to opt out of CEA decomposition,has linked issue,35,has linked issue,1,0,,188271,,,0,ab6da7128e99277d3cad214017e362e363b619bfc518e787399e2c117aeee06f
thread:pr:188334,pr #188334: Handle negative integer exponents in SymPy utilities,has linked issue,35,has linked issue,1,0,,188334,,,0,03d91381986ffdf543ac41fb0c715bd63796b087d102b3e1f903a62580fa9bef
thread:pr:188347,pr #188347: Preserve tied parameter and buffer aliasing in load_state_dict(assign=True),has linked issue,35,has linked issue,1,0,,188347,,,0,1025375355ad1d2642511cf5ba1dfe34bbfd6b2f2f179f6cb78fc4d41e9e17b7
thread:pr:188349,pr #188349: [HOP] Adding length parameter to scan,has linked issue,35,has linked issue,1,0,,188349,,,0,d88781048212992e33ee337e58241500047c769f4c70a27b54e561b8d681246c
thread:pr:188363,pr #188363: Add kernel_size > 0 validation to FractionalMaxPool2d matching Fracti…,has linked issue,35,has linked issue,1,0,,188363,,,0,888388d73710af47af79b5a072c41934f3e9a6b7a72ffb4c838dca758d7af2f3
thread:pr:188369,pr #188369: Add DTensor sharding strategy for FFT c2c,has linked issue,35,has linked issue,1,0,,188369,,,0,5dceb9f5ca60d9e86a35fab1891c31c7736719ca188ba75cb58b0bfc20fe6129
thread:pr:188373,pr #188373: Enforce 1D tensor validation for batch_norm parameters,has linked issue,35,has linked issue,1,0,,188373,,,0,ae8d87d7df1ed1465c41755d377dce1baceed9e68e1936535e627293462c8c21
thread:pr:188381,pr #188381: aten: add output_size kwarg to masked_select for meta/eager shape oracle support,has linked issue,35,has linked issue,1,0,,188381,,,0,4ccf196e15c1b96310a5f1a7477e5e82b6c2501c78f0db1e2b07b487251949b8
thread:pr:188399,pr #188399: Fix #171905: Use typing_extensions.TypeAliasType for better reexport of `__module__,has linked issue,35,has linked issue,1,0,,188399,,,0,1d2b31f8573df29dd25387093ebe90372d2db78485409c7d52db9eeb0ccba2c1
thread:pr:188524,pr #188524: test: centralize optimizer comparison tolerances,has linked issue,35,has linked issue,1,0,,188524,,,0,4f714f967c2b73a25f84f435084e2f3868eaba00f90bc92a665d816046ffe0be
thread:pr:188525,pr #188525: test: add Linear module error inputs,has linked issue,35,has linked issue,1,0,,188525,,,0,a3244b90352b45470b0b6f65e9fafd9ff845ab54ab4860499a3c39797df10bd5
thread:pr:188526,pr #188526: error checking: validate _pdist_forward dimensions,has linked issue,35,has linked issue,1,0,,188526,,,0,7e57144b012badb395409a7d08512da59502143df301586022e2b32254af5c6f
thread:pr:188528,pr #188528: cuda: reject negative caching allocator sizes,has linked issue,35,has linked issue,1,0,,188528,,,0,4e9620ae5362f4c617d847594a3f79dd87b64772ecf680e2c9f3858264a1a889
thread:pr:188531,pr #188531: pooling: guard max_pool1d integer parameters,has linked issue,35,has linked issue,1,0,,188531,,,0,77b176169e73a917e7e268c0c90bdfac3909601a859da72d6b3c7d7c51d3dce5
thread:pr:188543,pr #188543: Remove deprecated torch.utils.mobile_optimizer from API index,has linked issue,35,has linked issue,1,0,,188543,,,0,ae6fdc48f7f3f7a029f548ab2013bb2888868b0ca464b7033232beae52013b91
thread:pr:188560,pr #188560: [MPS] Fix constant address space for scalar params in binary_dense_scalar kernels,has linked issue,35,has linked issue,1,0,,188560,,,0,79dfb5fb3bcc5ec1b2d2a1169d311c8460bfdb71750a157237bc019a018f4d5e
thread:pr:188579,pr #188579: optimize split reduction heuristics on GB200,has linked issue,35,has linked issue,1,0,,188579,,,0,876d9a9a467766820b37b763e29d16ccd734424e272ada9baf95991907c1f41e
thread:pr:188633,pr #188633: Draft: Update CUTLASS fence proxy API usage,has linked issue,35,has linked issue,1,0,,188633,,,0,4b889cff153d0fd4837667533a597e4779a059195d25bb458a6d064ec99f34aa
thread:pr:188637,pr #188637: [inductor] Reindex transpose-contiguous clone to enable vertical fusion,has linked issue,35,has linked issue,1,0,,188637,,,0,988860186baa9a0170efbb51fab05e29bb8f2d149f7cbab0b2d01f03839f71ef
thread:pr:188642,pr #188642: Fix __qualname__ for OpOverloadPacket,has linked issue,35,has linked issue,1,0,,188642,,,0,0dc575f2499056742f610e3ca64b10e40f82a44de6aae70806ae01254d49ca82
thread:pr:188647,pr #188647: torch/fx: fix _format_target to handle Python keywords and embedded quotes,has linked issue,35,has linked issue,1,0,,188647,,,0,abf24aca4352ebabac9cb3410e487a23ad1e32e2b8270224910edc77de0ee22c
thread:pr:188732,pr #188732: [MPS] Enable ConvTranspose3d for fp16/bf16 via internal fp32 upcast,has linked issue,35,has linked issue,1,0,,188732,,,0,c66da6db0b956ffb3c9da98567b36140d8b30df71c3c2f79be86cdd6ce94f0aa
thread:pr:188762,pr #188762: Fix multinomial without replacement silently sampling zero-probability categories,has linked issue,35,has linked issue,1,0,,188762,,,0,1c0da8bda00fdc7828721937ed0a945209c6c7364cfb823da26b0cda6c0a07df
thread:pr:188831,pr #188831: Fix module re-export in tensor/experimental API,has linked issue,35,has linked issue,1,0,,188831,,,0,65b0460dd3ac6f60f5bb54a3e963fb003784e4b2943269fab7eee42f4e502619
thread:pr:188843,pr #188843: Fix scatter/scatter_add/scatter_reduce src gradient when index is sma…,has linked issue,35,has linked issue,1,0,,188843,,,0,5d51f210b25aeda11bb73eaf7bed6b959e31b34ca94ab5efe71ce4a282e1db8c
thread:pr:188857,pr #188857: Fix type confusion in Stream comparison against non-Stream objects,has linked issue,35,has linked issue,1,0,,188857,,,0,fba615e9e0e4c9253709f70655076165591bc8a9f4d215fd0abe0430a3eb65f7
thread:pr:188883,pr #188883: Handle empty autograd sections in backend YAML,has linked issue,35,has linked issue,1,0,,188883,,,0,d722054f2f7399f5d695f7ac60cc3f7dc14ecf9b173f74692517a7a2e4f89e13
thread:pr:188893,pr #188893: [ONNX] Clarify that output_names labels outputs but does not reorder them,has linked issue,35,has linked issue,1,0,,188893,,,0,3322120a44ca2cbec64a44eb5e84dd2dce2958499f7cca6c8ba7c54b53be2d2a
thread:pr:188923,pr #188923: test: add gradient health validation tests (fixes #177116),has linked issue,35,has linked issue,1,0,,188923,,,0,a25a5b32b8ef9da722ae3303f85fe5c8bc40b65c61677e9e9e93105e000a4853
thread:pr:188945,pr #188945: Fix assert_close to support comparisons involving Python and NumPy scalars,has linked issue,35,has linked issue,1,0,,188945,,,0,007c82e6031f51b8d73db322966f247d084fd8817b97aa350fee71e4b0670834
thread:pr:188968,pr #188968: Fix record_stack_traces returning empty trace when no forward frame found,has linked issue,35,has linked issue,1,0,,188968,,,0,1a2b9a8d460a54f4e0e98bcad6323ce77d8e96a693bac97b206ba0656f9179b6
thread:pr:188969,pr #188969: [docs] document __torch_function__ behavior during autograd hooks,has linked issue,35,has linked issue,1,0,,188969,,,0,eaed5985abd166bd8d78b0b9f4a694e6734590caed081bcad2e4f95adc190253
thread:pr:189154,pr #189154: Bundle unversioned libnuma for ROCm wheels,has linked issue,35,has linked issue,1,0,,189154,,,0,ca9d4716691f7c5fe24de0b25359b0b4babbbca397a5ab7b7c3670ae3ef1bae7
thread:pr:189285,pr #189285: Fix MIOpen RNN dropout: gradient stall and missing per-step mask variation,has linked issue,35,has linked issue,1,0,,189285,,,0,ca5641d567a16c1bac3589d7fcd2c92c5426410fa5cdfc58be6f0e602fcc12cf
thread:pr:189310,pr #189310: Generalize GeneratorState input device type,has linked issue,35,has linked issue,1,0,,189310,,,0,1a83981773baa7edd88d4222a58f35c221a1a07df838520f02f0acd36677d513
thread:issue:126654:issue:138649:issue:178251,issue #126654: torch.nn.functional.scaled_dot_product_attention returns NaN values after backward pass.,needs reproduction,32,needs reproduction,3,0,"126654, 138649, 178251",,,,2,b35dfb7e4888a539399732b8d8e65e05752bae5a5a474e6aa4980909e6409773
thread:issue:118092:issue:97772,issue #118092: torch.load with weights_only=True show error for pickle protocol 4,needs reproduction,32,needs reproduction,2,0,"97772, 118092",,,,1,dcc499d6e47e9b6e33ee91fcfe9b4e8401e2f4eccacd8db99936b33e4e75fad9
thread:issue:138422:issue:171604,issue #138422: vmap with torch.autograd.grad does not work on output of compiled function,needs reproduction,32,needs reproduction,2,0,"138422, 171604",,,,1,72620abce5d3ee7269d23cc1318dec451bb0325b645a1c18e5364537e2644f7c
thread:issue:146941:issue:147115,issue #146941: Make HOPs more debuggable,needs maintainer decision,32,needs maintainer decision,2,0,"146941, 147115",,,,1,5a645f607d7ea7e86ce6d62e9d12776d37d065940088d5684908debd49cd12ee
thread:issue:108041,"issue #108041: Undefined Symobl: pybind11::detail::type_caster<at::Tensor, void>::load(pybind11::handle, bool)",needs reproduction,32,needs reproduction,1,0,108041,,,,0,f4b8dbc9263d217323f271b24ea1310b884161e1efb7350fef4eb600c66ae45e
thread:issue:111182,issue #111182: Training iresnet with torch.compile is slower than eager mode for torch 2.1.0,needs reproduction,32,needs reproduction,1,0,111182,,,,0,d87086eafdf685b5958cee275fc2c9dc89a28e16d2e389a49856b808c841a242
thread:issue:114861,issue #114861: Torchrun threads with different computing speed.,needs reproduction,32,needs reproduction,1,0,114861,,,,0,5c6222a093daffa4c3c0694640f6b200bd05f7b1812f605d515fe801dd7056fd
thread:issue:120383,issue #120383: Numeric stability issue with full parameter fine tuning of Llama 2 7B model with FSDP,needs reproduction,32,needs reproduction,1,0,120383,,,,0,a8234d2d8ede6f353e969c96071144f3e4b9993326f28eb08ae4a37c311378b8
thread:issue:121606,"issue #121606: RuntimeError: Rank 2 successfully reached monitoredBarrier, but received errors while waiting for send/recv from rank 0. Please check rank 0 logs for faulty rank.",needs reproduction,32,needs reproduction,1,0,121606,,,,0,9d0ec9ac52d7340e422aee3fe993c6385fc0f5b147de73e475a58819300e46f8
thread:issue:121944,issue #121944: raise RendezvousConnectionError( torch.distributed.elastic.rendezvous.api.RendezvousConnectionError: The connection to the C10d store has failed. See inner exception for details.,needs reproduction,32,needs reproduction,1,0,121944,,,,0,b625507d30b082c981af82a20a0daa6c623da475d3d1e42ff4a540421aa1faa4
thread:issue:125954,issue #125954: Error: command buffer exited with error status.,needs reproduction,32,needs reproduction,1,0,125954,,,,0,fc4efacc5731695146c3e44d4e5ceade92628b8f94e163ac2dd7c497a4cd868a
thread:issue:130253,"issue #130253: RuntimeError: CUDA error: CUBLAS_STATUS_EXECUTION_FAILED when calling `cublasGemmEx( handle, opa, opb, m, n, k, &falpha, a, CUDA_R_16F, lda, b, CUDA_R_16F, ldb, &fbeta, c, CUDA_R_16F, ldc, CUDA_R_32F, CUBLAS_GEMM_DEFAULT_TENSOR_OP)`",needs reproduction,32,needs reproduction,1,0,130253,,,,0,19fbf41d34eddbb65f90eaa486c871715bf48784f046b2883f36d1b8c97a3e07
thread:issue:133548,issue #133548: Floating point exception on H20,needs reproduction,32,needs reproduction,1,0,133548,,,,0,604b09e3f82cf01a1d156ee835dd00884a3a0ce882aa4aeb40712523da1aa9ba
thread:issue:135196,issue #135196: ModuleNotFoundError: No module named 'torch.fx.experimental.shape_inference',needs reproduction,32,needs reproduction,1,0,135196,,,,0,280393dfc8b8a2d415b1a6c4971a862dc75ef36a5bbf55eb28f69f73335aafeb
thread:issue:136344,issue #136344: Composite Compliance Tensor ut check fails in privateuse1 device,needs reproduction,32,needs reproduction,1,0,136344,,,,0,48c09902a947d7b54a932688e939bdc76a11033e90834ae87bb4e7475fc7644f
thread:issue:137001,issue #137001: BCE loss mps device,needs reproduction,32,needs reproduction,1,0,137001,,,,0,f033501eea2236cd309b91c40d226c4eb83805cef3ffe8b95a496a17bbaad069
thread:issue:139579,issue #139579: There is a bug when training SwinIR model in pychram and running it,needs reproduction,32,needs reproduction,1,0,139579,,,,0,c86a982db7b6c67ed600a249494ac647964f32890781f3ab20646eb1e0914a2d
thread:issue:141545,issue #141545: Missing Fake/Meta/Abstract Kernels for Operators,needs reproduction,32,needs reproduction,1,0,141545,,,,0,858ef683dcb612613d08ed3d6a93af5cc38dde96ec709bd50dcb8d5296e73418
thread:issue:143579,issue #143579: `assert_size_stride` failing in `_inductor/utils.py` `return model(new_inputs)`,needs reproduction,32,needs reproduction,1,0,143579,,,,0,74b97989aaa4d734aded742f65feb6698bddaaf936bd135c6dd025349ebf4b8e
thread:issue:144803,issue #144803: Something is fishy with discard_graph_changes,needs reproduction,32,needs reproduction,1,0,144803,,,,0,36e070af3c78fea4ad992a5292d7d7fc79e517a1566f3915b3c29b18859fedd1
thread:issue:145451,issue #145451: [Dynamo]while_loop raise an exception,needs reproduction,32,needs reproduction,1,0,145451,,,,0,a3419c56a09e7220c12509c59f77e02c976b6ba474930ae13b0937bfeb937e14
thread:issue:145782,issue #145782: Behavior Difference with Flex Attention + Sequence Packing,needs reproduction,32,needs reproduction,1,0,145782,,,,0,77b89d6deeb086fcd49499d4e9725d6c6e94ffb44a78bdd6025e7f7a9dd84e73
thread:issue:146994,issue #146994: subclasses + HOPs fail with `Attempting to use FunctionalTensor on its own`,needs reproduction,32,needs reproduction,1,0,146994,,,,0,bd2fc1384b8801e5a4cb99ad63f50daaa9b821b77832fa5ec93674bc1c685f26
thread:issue:148811,issue #148811: Custom Triton operator not registered when loading AOT-compiled .so in C++ environment,needs reproduction,32,needs reproduction,1,0,148811,,,,0,f7fd54b07c94c1111e368b4020cc4ed67e140e5bc71d95d93add2f9cec6c4bf1
thread:issue:153172,issue #153172: triggered internal assert in matmul,needs reproduction,32,needs reproduction,1,0,153172,,,,0,b8b41ae56e3d53262c82b5e2c964ec37811a3fab083a22f85de65aab22da2e36
thread:issue:158724,issue #158724: Problem with RPC on dual stack networks,needs reproduction,32,needs reproduction,1,0,158724,,,,0,4d1696d322c2b645bb7c2c811b8f4193df47f79bb287d0ba09c0b9c9edbf56da
thread:issue:159207,issue #159207: Add official support for CUDA sm_120 (RTX 5090 / Blackwell architecture),needs reproduction,32,needs reproduction,1,0,159207,,,,0,b79a26e2f905f071ca475b88c494ae56e06df0c7a039128d9b85318697848e83
thread:issue:162957,issue #162957: torch.linalg.eigh uses a large amount of memory in pytorch 2.8.0,needs reproduction,32,needs reproduction,1,0,162957,,,,0,f9275b1ede4f38a2857fe085c59cd0bb2ccd710862a08d649538e97918b47d4f
thread:issue:164342,issue #164342: Official support for sm_120 (RTX 50-series / Blackwell) in stable PyTorch builds,needs reproduction,32,needs reproduction,1,0,164342,,,,0,ba92f4b680c20ef76064e7d569f272df4333ff223100be5372c151769a2f424e
thread:issue:165532,issue #165532: PyTorch CPU / DataLoader: “Too many open files” (`EMFILE`),needs reproduction,32,needs reproduction,1,0,165532,,,,0,decf76e25517bf7086e165148ad8f4fe5fe1eaee9d0326aba6ecd974ac728e19
thread:issue:165959,"issue #165959: torch.distributed.DistStoreError: use_libuv was requested but PyTorch was built without libuv support, run with USE_LIBUV=0 to disable it.",needs reproduction,32,needs reproduction,1,0,165959,,,,0,fd4535aac1fe96d24841007386e1718d8fd240f2cd35c1a8f67b0ddd0d939aab
thread:issue:166633,"issue #166633: Command '['ninja', '-v']' returned non-zero exit status 255.",needs reproduction,32,needs reproduction,1,0,166633,,,,0,0b5c871227363985f2d730b03602b5bf1a170124e2a3772e473617ebb6f545db
thread:issue:167414,issue #167414: C10_CUDA_KERNEL_LAUNCH_CHECK uses getLastError without clearing the error.,needs reproduction,32,needs reproduction,1,0,167414,,,,0,12bf3f91e2830431fb6bd5ac6d1d7c6850f3cd1a502ec355ce3c8db00f0c7ce6
thread:issue:168897,issue #168897: Can't install on python 3.13,needs reproduction,32,needs reproduction,1,0,168897,,,,0,84edfb88d6a031d3a6dd83a0e2956df0ac122c7329c962ee065db846cf038deb
thread:issue:170138,issue #170138: Performance regression of depthwise Conv1d with fp16/fp32 AMP from 2.7.1 -> 2.8.0 and the issue remains in 2.9.1,needs reproduction,32,needs reproduction,1,0,170138,,,,0,cc95bd9dd0b5dcb64e34f4a8c2e720578013632ca54b12868d09e64a74cd93c5
thread:issue:170676,"issue #170676: CUBLAS_STATUS_INVALID_VALUE at F.linear((14,4096), (4096,4096)) device='cuda:0', dtype=torch.float16 on 4090 and B300",needs reproduction,32,needs reproduction,1,0,170676,,,,0,9930d905e2b31c47f1135d572b86535ebadff1730675f14c22978b19753bd558
thread:issue:171499,issue #171499: bug in torch.stack a long 5d video and yields a misordered video output,needs reproduction,32,needs reproduction,1,0,171499,,,,0,ac6249651d5df2c31ac44774d18afd56aebf5ecd51ba9cbc684784f5cbb36c00
thread:issue:171672,issue #171672: torch.compile max-autotune and reduce-overhead much slower than default due to repeated CUDA graph instantiation,needs reproduction,32,needs reproduction,1,0,171672,,,,0,dda106631b570a5c689db8e2a8c7176c52c6f3f744b6c7d0226179e00b29d031
thread:issue:172217,issue #172217: test_numba_integration consistently fails with SIGSEGV on CUDA-enabled environments (Ubuntu 22.04 & RHEL 9),needs reproduction,32,needs reproduction,1,0,172217,,,,0,139ed0fdd9ce323c0a85e16816638c984bb3a1de640d3f40e91f90cfa9be8eb3
thread:issue:172547,issue #172547: Cold start models with compile-cache have significant overhead,needs reproduction,32,needs reproduction,1,0,172547,,,,0,a45a0825510819796d34e4f38fc89c7b903451d7172ed1bbd94d4dfff97a852c
thread:issue:172628,issue #172628: pip download torch report nvidia_nccl_cu11 checksum mismatch,needs reproduction,32,needs reproduction,1,0,172628,,,,0,24544594396ab1ecc661a9c80f23452bb43a6cacd8a3e0d905f448947b07b0e3
thread:issue:172944,issue #172944: Tried to access the schema for which doesn't have a schema registered yet,needs reproduction,32,needs reproduction,1,0,172944,,,,0,abf0153dd812d85e631813685bc75407854a4c6ea5fccfcafd94207a8cd91de7
thread:issue:173066,issue #173066: Training accuracy may suddenly fall to near 0 in the early training stage using pytorch 2.9 and AMP (bfloat16),needs reproduction,32,needs reproduction,1,0,173066,,,,0,7b45e712344517cdfe2bc2f38f84edca81b163ac654529d5c71a3dc2bfcdf2a7
thread:issue:173315,issue #173315: [cuda] Numerical Inconsistency: NaN in Eager CUDA vs Valid Output in Compiled CPU for nn.LSTM,needs reproduction,32,needs reproduction,1,0,173315,,,,0,194d3ea07f52787a2248718a0063b19b00a465c31aac087a7b8db45a40c95ca6
thread:issue:173389,issue #173389: Inference latency increase on ARM after upgrading from PyTorch 1.13.1 to 2.x,needs reproduction,32,needs reproduction,1,0,173389,,,,0,fb0cb2309fdfddc5b528c976851b8cf2578f078bb60d07ccf9c8b08eebf09508
thread:issue:173476,issue #173476: [Hang] Unrecoverable CUDA deadlock in torch.ops.aten.lstm - process becomes unresponsive,needs reproduction,32,needs reproduction,1,0,173476,,,,0,ebbd4a8b59506aab9e132792d3251403342ed9eeea6cebc18ba05a1ec3743dfb
thread:issue:173922,issue #173922: [Numerical Accuracy] 100% relative error in LSTM output between CUDA and CPU backends in eval mode,needs reproduction,32,needs reproduction,1,0,173922,,,,0,5147f8157b4401e66b32fa335b24cbcf00f8919efc5c29624ba2ac2275dbb5ac
thread:issue:174010,issue #174010: Numerical divergence in GPT2 wpe layer: 2.19e-4 relative error between float32 and float16 on CUDA,needs reproduction,32,needs reproduction,1,0,174010,,,,0,cb8a63644122ae272de205dfedee14d875fc19ca692217614d57532c28da0110
thread:issue:175057,issue #175057: torch.compile / Inductor generates invalid C++ (undeclared zuf0/zuf1) during compile (CppCompileError) on RTX 3090,needs reproduction,32,needs reproduction,1,0,175057,,,,0,c82c4d8d1bfcf69b925957ad35ba2eae00d11d13488f8abeb85f22878d37b8b0
thread:issue:175637,issue #175637: Conv2d output inconsistency across different execution modes (device/dtype/optimization),needs reproduction,32,needs reproduction,1,0,175637,,,,0,820c6869ac08c0dfc2b82534ba16a04e873729fa19bd5fc216c2911449b9fb15
thread:issue:177020,issue #177020: tensor cuda memory is not released when using nested_tensor_from_jagged,needs reproduction,32,needs reproduction,1,0,177020,,,,0,f5ed7d1e387bf189e64b201ed3ece637509fd16a018dc8d01e27230ae168e663
thread:issue:177028,issue #177028: tensor cuda memory cannot be release by gc when a tensor hold ref of another,needs reproduction,32,needs reproduction,1,0,177028,,,,0,120ab0c86f80d4ecb5ee00ef7e2e9e92c736a86c38518208ebf4c5037892bc01
thread:issue:177738,issue #177738: C++ Can't capture Transformer CUDA Graph,needs reproduction,32,needs reproduction,1,0,177738,,,,0,1418821b8374e71f20a3d37e72f413ab5319fe0d66a49c1dbc226f78effac32d
thread:issue:177819,"issue #177819: MPS unavailable on macOS 26.3.1 arm64, and `device=""mps""` reports incorrect macOS requirement",needs reproduction,32,needs reproduction,1,0,177819,,,,0,65c9862f7213d2fcb0299d0cbb54df3993d0d1c8c5904f8a8bf0b9e24a202892
thread:issue:177952,issue #177952: Fix onnx tests running on GPU/CUDA,needs reproduction,32,needs reproduction,1,0,177952,,,,0,97c83c3b6efef6ab4c46a6651219297694271d49e091b4b9d107fa3f31dddda8
thread:issue:178049,issue #178049: print is repeated sometimes when the calc is done on gpu,needs reproduction,32,needs reproduction,1,0,178049,,,,0,e7a4fc4835d1f202e7375492c53860691ab59239198799e30d866222e9cbb09b
thread:issue:178145,"issue #178145: Large fp32 discrepancy in LeNet5Fashion conv2 output on CUDA eager baseline, with similar error against both CUDA compile and CPU eager",needs reproduction,32,needs reproduction,1,0,178145,,,,0,0304b4a9843424ab924a0897d08a734bebed14ff764789f92b7e13b984695f1b
thread:issue:179904,issue #179904: PyTorch 2.6 weights_only=True breaks loading of existing checkpoints,needs reproduction,32,needs reproduction,1,0,179904,,,,0,cf6aaea066b2a2040e2fbc1fb97219cd702dde2e3b7d9b0a5489caa9107e915a
thread:issue:181901,issue #181901: [Inductor][NVGEMM] tile_M<128 EFC kernels broadcast incorrectly,needs reproduction,32,needs reproduction,1,0,181901,,,,0,ac6a29073817f10b116c81d9467c188ab1f8106c5cb3f1f0f71b1b0cc109e5c5
thread:issue:182760,"issue #182760: [Build] UnicodeDecodeError in cmake_utils.py due to non-standard library version suffixes (e.g., OpenSSL 1.1.1wa)",needs reproduction,32,needs reproduction,1,0,182760,,,,0,ccee790ce5326fceba112481b27bfd74a8e46195b8ce316927fe8fde3ec15a84
thread:issue:184030,issue #184030: torch.nn.PoissonNLLLoss returns NaN for negative input when log_input=False,needs reproduction,32,needs reproduction,1,0,184030,,,,0,dbbe7825d3f9e4c63029099a82ab72e66b1f7e931a8e57df81653f7a3b3f7ad6
thread:issue:187824,issue #187824: libcudnn_engines_tensor_ir.so.9 not included in the PyTorch based on CUDA 13.0,needs reproduction,32,needs reproduction,1,0,187824,,,,0,3a4effa9c705ed863351c8ea535455057bcb6aa6a7c0b1b53fb0b8fc3bbe273f
thread:issue:188411,issue #188411: RTX 5090 cudaErrorLaunchTimeout,needs reproduction,32,needs reproduction,1,0,188411,,,,0,d5b682c197caa29c2998d15c2afe1dde403e2cf739cbbac13aeaf94cb2871153
thread:issue:88251,issue #88251: Weird random SIGTERM occurance,needs reproduction,32,needs reproduction,1,0,88251,,,,0,721782e50fb56cb2323da4d953f1ebf980cbecde7149db5dec0c8d48cea1ba67
thread:issue:97635,issue #97635: JAX + PyTorch produces `OMP: Error #13: Assertion failure at kmp_affinity.cpp(532)`,needs reproduction,32,needs reproduction,1,0,97635,,,,0,8e85fbfe97e8d64751d16cdd77fae2de392f47625eff2274361c49ea02eb9f02
thread:issue:98542,issue #98542: Training runs 50% slower when using 2 GPUs comparing to 1,needs reproduction,32,needs reproduction,1,0,98542,,,,0,b2d9f5ab0b905ed40ed4392a0694349f6396242dfeeb1ce2d611f7c73dfd0f48
thread:pr:133423,"pr #133423: Keep torch.backends.mkldnn, add torch.backends.onednn",draft/noise,31,draft/noise,1,0,,133423,,,0,4400e4bcc9013884fd7579ecae332aeddc8e841fdb3b648228aa0e71128c2c62
thread:pr:133667,pr #133667: Update non user APIs - 2,draft/noise,31,draft/noise,1,0,,133667,,,0,f94cbcf0c54baf4f38ac20c1a0e9d67dfbaa7079774026526d6036c84de370a1
thread:pr:133753,"pr #133753: Keep mkldnn_rnn_layer, add onednn_rnn_layer",draft/noise,31,draft/noise,1,0,,133753,,,0,3f34382dc98ff7a712eb5b06325351ae1f59d36d49c36df504cf0a077894c78e
thread:pr:176726,pr #176726: Fix Muon momentum/update numerics for XPU state_dict parity,draft/noise,31,draft/noise,1,0,,176726,,,0,eb0ce72051f77a59d71fb959cda4c5c7a26793389ad5d6dc8dbdeddddaf7b28e
thread:pr:181926,pr #181926: Enable XPU for sparse_sampled_addmm,draft/noise,31,draft/noise,1,0,,181926,,,0,013c48daa9bd2c6909fc85379b05795d1dac4306c5a18980941549fbf809a016
thread:pr:184372,pr #184372: [Draft] Remove ACL,draft/noise,31,draft/noise,1,0,,184372,,,0,3364e7c06749b58e6b93055167f3ee84560b2ec4a4e79351b58aeb90826935ae
thread:pr:187165,pr #187165: [ROCm] Enable FMA inductor lowering on ROCm,draft/noise,31,draft/noise,1,0,,187165,,,1,0287c0c53088e0e9beec3210fdc3d4a6dfbfd5914c2c7c3b28371a1ef33ff6b1
thread:issue:10440:issue:10464:issue:10818:issue:11869,"issue #10440: [caffe2] enforce fail at context_gpu.cu:285, cannot get GPU memory usage statistics",needs maintainer decision,28,needs maintainer decision,22,0,"6549, 6785, 6979, 7060, 7362, 7374, 7490, 7491, 7569, 7614, 7835, 8442, 8533, 8595, 9349, 9707, 9867, 10440, 10464, 10818, 11869, 128959",,,,21,eafb3e4a47c58df10739f546043308cd264921bd63c319c9a7d16eee8ce16324
thread:issue:108567:issue:112398:issue:112509:issue:114090,pr #119977: [NJT] Factory function support,stale/low urgency,28,Review active PRs and linked issues as one maintainer work thread.,13,0,"88137, 89124, 89125, 99142, 108567, 112398, 112509, 114090, 115978, 118107, 130538, 132212",119977,,,12,205f77f377f7f0ef0e408d07c234afcd6d0589c8124fefe4f64c09b3a2fdf23d
thread:issue:129627:issue:140069:issue:55207:issue:72138,issue #129627: [RFC][DTensor] DTensor Strided Sharding: A More Flexible Way To Shard Tensors,stale/low urgency,28,stale/low urgency,7,0,"55207, 72138, 88838, 89160, 89884, 129627, 140069",,,,6,c590f1ab14c18c58f0ab5ab784af319b1736cd0915292898795e835574f739e0
thread:issue:168756:issue:168766:issue:168767:issue:168768,issue #168756: TestModule: test/test_linalg.py::TestLinalg,stale/low urgency,28,stale/low urgency,7,0,"168756, 168766, 168767, 168768, 168771, 168772, 174913",,,,6,ca0f4734d2697f01e3fc65e5b3485d31d5428c2848c25472357d5678f1bdcace
thread:issue:105641:issue:108968:issue:128396:issue:58181,issue #105641: Turn indexing with a scalar tensor into an copy into a view and avoid a D2H synchronization.,stale/low urgency,28,stale/low urgency,5,0,"58181, 73175, 105641, 108968, 128396",,,,4,bb8554e54738fde60040fbe47244662200109d7fc5ffef0a9b20b925d95ab125
thread:issue:158643:issue:158970:issue:25104:issue:76580,issue #158643: torch.compile() produces inconsistent outputs with models containing dropout and in-place ReLU operations,stale/low urgency,28,stale/low urgency,5,0,"25104, 76580, 92927, 158643, 158970",,,,4,770d9ee605d8ea5377014779c2ec93ec3b114ba723108f42869096d1300f258e
thread:issue:168871:issue:168872:issue:168875:issue:168876,issue #168871: TestModule: test/test_transformers.py::TestSDPACudaOnly,needs maintainer decision,28,needs maintainer decision,5,0,"168871, 168872, 168875, 168876, 168877",,,,4,aae0f60f51199c8bcbddfc2fdb50f36fcb381673174015c452743f5bc74c8684
thread:issue:103161:issue:105982:issue:112459:issue:134606,issue #103161: Calling jacrev with LSTM and functional_call gives error,stale/low urgency,28,stale/low urgency,4,0,"103161, 105982, 112459, 134606",,,,4,7887b4c09d0e1f856b5f7b88d05a98ecc88c77385a544033ec3350b67c378723
thread:issue:105499:issue:111317:issue:111552:issue:112164,issue #105499: [FSDP] Revisit mixed-precision casting logic,stale/low urgency,28,stale/low urgency,4,0,"105499, 111317, 111552, 112164",,,,4,e7108c0e9f47659cef53540908676b30b34c3ba8b6608bb976d6a92d0af0e3b4
thread:issue:155720:issue:65462:issue:65464:issue:65465,issue #155720: [rocm] HIP Graph capture raises segmentation fault on AMD GPU but CUDA Graph capture succeeds on Nvidia GPU,needs maintainer decision,28,needs maintainer decision,4,0,"65462, 65464, 65465, 155720",,,,3,1c97f71032162db41fa6d3faf0e4d596cc81b387bb15534d8a3e974d7446624b
thread:issue:189135:issue:189136:issue:189137:issue:189138,issue #189135: [Sub-RFC] PrivateUse1 device identity/capability in torch.compile (RFC 1/3),needs maintainer decision,28,needs maintainer decision,4,0,"189135, 189136, 189137, 189138",,,,3,0c661d5f8b0830fbba8e33842de090e6fa3f32d1756f6e93ba8f327cc906cf63
thread:issue:143112:issue:165038:issue:169277,issue #143112: FFT half precision only let CUDA pass the check,stale/low urgency,28,stale/low urgency,3,0,"143112, 165038, 169277",,,,2,bb535578879a95af5993fcd9de75de879c2da624de5d3bbe16bbc3a944a24bae
thread:issue:144824:issue:188756:issue:84520,issue #144824: [MPS] Indexing Returns 0 if OOB,stale/low urgency,28,stale/low urgency,3,0,"84520, 144824, 188756",,,,2,4497fe7065be71648571e21e31aba927c263c0d8979543165453409e1b874376
thread:issue:166563:issue:167283:issue:167881,issue #166563: [RFC] Modifying Getting started page for Experimental Wheel Variant Support,needs maintainer decision,28,needs maintainer decision,3,0,"166563, 167283, 167881",,,,3,0e3c234035e36d66085601e352c69538c898cbf26705d9e2eaa07b6b0ae79c2d
thread:issue:168819:issue:168820:issue:168823,issue #168819: TestModule: test/test_scaled_matmul_cuda.py::TestFP8Matmul,stale/low urgency,28,stale/low urgency,3,0,"168819, 168820, 168823",,,,2,684fcba5407b8b1760c6193443172e0a96115fe86af39364826dd16be51bca99
thread:issue:175873:issue:176730:issue:179294,issue #175873: [MPS][Performance] SDPA op implementation is wasting resources by returning attention weights,needs maintainer decision,28,needs maintainer decision,3,0,"175873, 176730, 179294",,,,3,5cd425064b0aaced8640bbb622e87d88198257735e179ebf53ac4330a738a49f
thread:issue:180343:issue:180345:issue:180712,issue #180343: Accelerator-independent API to determine if a specific backend is available?,needs maintainer decision,28,needs maintainer decision,3,0,"180343, 180345, 180712",,,,2,1577a4a46e070171d7501c9b0bd849ba9c634b398e8f04a14816dc3479d8892d
thread:issue:101850:issue:102269,issue #101850: Unexpected modification to CPU affinity of Dataloader workers,needs maintainer decision,28,needs maintainer decision,2,0,"101850, 102269",,,,1,21cca9bc82cab3998a0514afb899e4de5c8fb2faab8b4c7dc6a702a7af67c9a0
thread:issue:106704:issue:154318,issue #106704: Dataloader extremely slow on in-memory datasets,has linked PR,28,has linked PR,2,0,"106704, 154318",,,,1,ff361c54deabb8afd8088e35884d58793b83bb263e13beee402a0ba1bda021f2
thread:issue:112883:issue:113063,issue #112883: Torchbench inference failures,stale/low urgency,28,stale/low urgency,2,0,"112883, 113063",,,,1,88c998c8f4b6b001bea9b9fb9ed13d977fe8e87f0b1df8caa55325da8fddf8df
thread:issue:114392:issue:183060,issue #114392: [BUG][pytree] equal `dict`s do not imply equal leaves and equal treespecs,stale/low urgency,28,stale/low urgency,2,0,"114392, 183060",,,,1,a4d0e43097a4234d34994d387de46560f8c47e4169d78fbd651bb0f8c6403e43
thread:issue:115305:issue:152822,issue #115305: torchrun to support NUMA affiinity,stale/low urgency,28,stale/low urgency,2,0,"115305, 152822",,,,1,3a5e5540d359e312e54a9a86ec119457709e3118e1f646cf1347b7c74f83805e
thread:issue:118492:issue:118798,issue #118492: `torch.compile(dynamic=True)` compiles forever,needs maintainer decision,28,needs maintainer decision,2,0,"118492, 118798",,,,1,ec0839ddcde7846dff3a4df81d0a83bbc7e3fb020e3cbbc263fea9074ec13f02
thread:issue:119574:issue:119613,issue #119574: Cannot configure dist_timeout when using device_mesh,stale/low urgency,28,stale/low urgency,2,0,"119574, 119613",,,,1,2b98fb43fdd95f87f4849257f2d66b2b82cbc73bef1dfb4eabd2bbbb7a2ca4cc
thread:issue:122540:issue:125958,issue #122540: [RFC] Solving Warm Start in torch.compile,needs maintainer decision,28,needs maintainer decision,2,0,"122540, 125958",,,,1,977a86118ff042c610afd55ad6c1eadae8e044496401660222554111277365d9
thread:issue:122840:issue:148402,issue #122840: a log_softmax kernel get much worse perf with padding,has linked PR,28,has linked PR,2,0,"122840, 148402",,,,1,6a03c280987424975895ea5035af5b9ca4b6f9a9bfb68e34a33572fa9f4bd245
thread:issue:123447:issue:95895,issue #123447: dist.barrier() hangs after calling async_save,stale/low urgency,28,stale/low urgency,2,0,"95895, 123447",,,,1,a5e5a3b7f1a1af4b7bc205b998e570717654dfb17ca3a385cc7a66b19dc79328
thread:issue:136003:issue:140706,issue #136003: [MPS] Inconsistent performance issues,needs maintainer decision,28,needs maintainer decision,2,0,"136003, 140706",,,,1,e197a3a444b11cdbc3e112365d4c72c3a88401d3a03097f3c0f33ce6f6dd970f
thread:issue:137564:issue:40770,issue #137564: Implement Incremental PCA with GPU support,stale/low urgency,28,stale/low urgency,2,0,"40770, 137564",,,,1,c51f3adb398b7d9d9653670defb05e8d37bdaa80241804ea65cfbff742307f78
thread:issue:140556:issue:189075,issue #140556: DISABLED test_backward_prod_cuda_float32 (__main__.TestNestedTensorOpInfoCUDA),needs maintainer decision,28,needs maintainer decision,2,0,"140556, 189075",,,,1,a687035866a0ec0d69964b262a48784297dc6e9219a717d584c261e89436c3d8
thread:issue:142315:issue:151705,issue #142315: Better fusion on triton templates,stale/low urgency,28,stale/low urgency,2,0,"142315, 151705",,,,1,9eb952052192498f700bf8904352e3a8cb472174f87649f4edf4e74be20269ab
thread:issue:144040:issue:164666,issue #144040: Torch.sparse.mm failing gradient computation at half precision.,stale/low urgency,28,stale/low urgency,2,0,"144040, 164666",,,,1,62c548b4f7da6b84196463f13cad72c3028a66eff6bd736df7e395f4a6297897
thread:issue:146108:issue:146111,issue #146108: [associative_scan] Support lifted arguments for inductor,stale/low urgency,28,stale/low urgency,2,0,"146108, 146111",,,,1,f747ba143250aaf96531db580b597695c2ff4392a9360150dc5bdd3c1d7fda87
thread:issue:149953:issue:153996,issue #149953: `RuntimeError: UR error` with XPU,needs maintainer decision,28,needs maintainer decision,2,0,"149953, 153996",,,,1,0bd1cefbfbd342d11facfb2ddb0df4a0768a1bd022b1952554c49fc29392eadb
thread:issue:151549:issue:170636,issue #151549: multiprocessing.pool.MaybeEncodingError Reason: 'RuntimeError('_share_fd_: only available on CPU')',stale/low urgency,28,stale/low urgency,2,0,"151549, 170636",,,,1,9ff760ab5c9c3ff90aa432170634fa83c2b676ae92197438b9545899b3765df4
thread:issue:152406:issue:166638,issue #152406: [DTensor] Calling .item() on DTensor with Partial placement results in local value,stale/low urgency,28,stale/low urgency,2,0,"152406, 166638",,,,1,9237d3f4ceda4b071cdb5335a7277a74e2b4f6740fd25658b6c71584ab16daf9
thread:issue:153484:pr:153557,"pr #153557: [PP] wip, allow grad to be None",draft/noise,28,Review active PRs and linked issues as one maintainer work thread.,2,0,153484,153557,,,1,48459e6d87abd7240f6e9e1e7ac899de23abd28f7bf43dbf53aa60de2ca4a977
thread:issue:154824:issue:175557,issue #154824: torch inductor cudagraph tree could incorrectly release some input nodes during replay,stale/low urgency,28,stale/low urgency,2,0,"154824, 175557",,,,1,2157078d396ad89a8b6f37250479be19da893695cff6feac4bddd0c08ea1e69f
thread:issue:156217:issue:157495,issue #156217: [Tracker] AutoParallel's feature request to DTensor,needs maintainer decision,28,needs maintainer decision,2,0,"156217, 157495",,,,1,d97c9d54e895c14eefaf5a917686e91ad74b05aab2df8144d468182b972d978f
thread:issue:161381:issue:164966,issue #161381: [XPU] ​​Incorrect free_memory reported by torch.xpu.mem_get_info()​,stale/low urgency,28,stale/low urgency,2,0,"161381, 164966",,,,1,e35cd12f813cef6109efd3a37442a876db0fa27ba961c97f2bd9b4d879e43352
thread:issue:164057:issue:164429,issue #164057: The device (Intel(R) Arc(TM) A380 Graphics) doesn't support querying the available free memory,needs maintainer decision,28,needs maintainer decision,2,0,"164057, 164429",,,,1,5c870693639bffef3cc2b9b95692424ef85c2c78529df18bbe768177b2d35e1a
thread:issue:164281:issue:173514,issue #164281: Document Symmetric Memory,needs maintainer decision,28,needs maintainer decision,2,0,"164281, 173514",,,,2,9fd8c6946cbff15e9c97d0445e88368fa8278e6c851b356054f02b542521dead
thread:issue:166628:issue:169429,issue #166628: [WinError 1114] A dynamic link library (DLL) initialization routine failed,needs maintainer decision,28,needs maintainer decision,2,0,"166628, 169429",,,,1,2d0778cadbba4a1199c02ba486ba9470c21832dcc15533bfa05b27786580f0c9
thread:issue:168651:issue:168652,issue #168651: TestModule: test/inductor/test_select_algorithm.py::TestSelectAlgorithm,stale/low urgency,28,stale/low urgency,2,0,"168651, 168652",,,,1,302187d774c31cdd5d75ec18db8405532838db9f986ef0ae3004faaa913d1e6e
thread:issue:170426:issue:72388,issue #170426: argmax over multiple axis,stale/low urgency,28,stale/low urgency,2,0,"72388, 170426",,,,1,a312ce43984633c815045bef474d566c396734a89e232691364924346c0b4669
thread:issue:171938:issue:186548,issue #171938: Device id in distributed process group with multiple backends,needs maintainer decision,28,needs maintainer decision,2,0,"171938, 186548",,,,2,c362bf39c1d8b2044435a198991d126fa887bc3f05d312d7f0cda25fa9590608
thread:issue:173164:issue:173362,issue #173164: allgather throw exception with discontigous input tensor,needs maintainer decision,28,needs maintainer decision,2,0,"173164, 173362",,,,1,dc9d74f832f05891fa4772a5d56d69c7d42383eacdc59c19dab39ef64e4685e8
thread:issue:174496:issue:175695,issue #174496: Improve performance of torch._foreach_copy_ on CUDA tensors,needs maintainer decision,28,needs maintainer decision,2,0,"174496, 175695",,,,1,efb7df8580e7b2e8274b098de1a3b1bfb5886fac5ddc74063fb053036fa6b6a8
thread:issue:179008:issue:179010,issue #179008: [privateuse1] Update the privateuse1 tutorial to mention registering HooksInterface,needs maintainer decision,28,needs maintainer decision,2,0,"179008, 179010",,,,2,e5cc39b8ebb0305b6607bb6e0b013b8bd2c6e642063f6e08bded540f6681bc43
thread:issue:180595:issue:180596,issue #180595: capture_scalar_outputs=True crashes 7/50 HuggingFace models with PendingUnbackedSymbolNotFound,needs maintainer decision,28,needs maintainer decision,2,0,"180595, 180596",,,,1,9517eaf2978a32037ac861ce290aee694f20f887f9f493c59bea4f118b0e072e
thread:issue:184352:issue:186099,issue #184352: Python 3.15 support for PyTorch,needs maintainer decision,28,needs maintainer decision,2,0,"184352, 186099",,,,1,b0eef9994244c3bf48cf372ef4f9be41e0591daa05d8b31f43ed1c43ae720e7f
thread:issue:186825:issue:72525,issue #186825: Bernoulli(probs=0).log_prob(1) returns dtype-dependent finite values instead of -inf (Geometric special-cases this and tests it),needs maintainer decision,28,needs maintainer decision,2,0,"72525, 186825",,,,1,544bc3ef5da3d972db589a90249b69832dae671898aa496ed1c0ab8dabb1888f
thread:issue:23430:issue:77154,issue #23430: [Feature request] Let DistributedSampler take a Sampler as input,needs maintainer decision,28,needs maintainer decision,2,0,"23430, 77154",,,,2,3e07e1ac7326211a06142f648844f71341e57474409b15cccf694b051409b063
thread:issue:28224:issue:60541,issue #28224: Allow to disable polling for CUDA synchronization,needs maintainer decision,28,needs maintainer decision,2,0,"28224, 60541",,,,1,878c89f3cc3bb6a8e57db4a4dc4b96946e6eaf90618ee977415290d2e7e4d52b
thread:issue:30458:issue:44027,issue #30458: add method to make tensor constant for debug purposes,stale/low urgency,28,stale/low urgency,2,0,"30458, 44027",,,,1,5d813ecf563470c3912233c5b13a11a434a4136b48575006e4aeaaa6143f8c31
thread:issue:39242:issue:72759,issue #39242: Add numerically stable log1mexp = log(1 - exp(-|x|)) function,needs maintainer decision,28,needs maintainer decision,2,0,"39242, 72759",,,,1,47f5006af72a3944c3190940c419018675b1269c49c604700ca8a4b5b48e8cbe
thread:issue:70160:issue:76103,issue #70160: NewOperatorRegistrationTest.testImplNoDefGetsCaught failed.,stale/low urgency,28,stale/low urgency,2,0,"70160, 76103",,,,1,6b72665a7d58f97af52060bac943e413a75b2cd5da1c200d252a48125e84d51b
thread:issue:74528:issue:74734,issue #74528: Supports for dist.send/dist.recv sending and recving torch.shorttensor,needs maintainer decision,28,needs maintainer decision,2,0,"74528, 74734",,,,2,09b7d42359bf28c5d5f14add7c1f2fbc95d9727fe27ce241d6d7c1790accdeb2
thread:issue:90544:issue:90848,issue #90544: [Dispatchable Collectives] Follow up tasks,stale/low urgency,28,stale/low urgency,2,0,"90544, 90848",,,,1,00345796ccef4bd4ad4cf049b785fbf87c2250280eecb3ea14d713bd2e7a6bec
thread:issue:100074,issue #100074: DISABLED test_checkpointing_resets_persistent_refs (__main__.CudaGraphTreeTests),needs maintainer decision,28,needs maintainer decision,1,0,100074,,,,0,1f4572af3928affedab7b0c908173b7740443794b51526da239ddd2dfcbe8ed5
thread:issue:100582,"issue #100582: Synchronization issue when combining DPP and RPC - ""Parameter marked twice""",needs maintainer decision,28,needs maintainer decision,1,0,100582,,,,0,fda6b01136307d89a2d7546628a2167b0fb0cc843a10a95154fa842e5dbe4da6
thread:issue:100985,"issue #100985: native_batch_norm has different size results on ""CPU"" vs ""META"" device",needs maintainer decision,28,needs maintainer decision,1,0,100985,,,,0,61abbb5f0c4db4a7fedb947e60996c309b15fa94e55e6416cda0cb5eabd40055
thread:issue:101402,issue #101402: DataParallel for nested modules,needs maintainer decision,28,needs maintainer decision,1,0,101402,,,,0,3173549bd0b1dd4d8dabad474b49497291a3a2db74c554f144f4b8c0ac32226a
thread:issue:102187,issue #102187: Extend fake fast path to more situations,needs maintainer decision,28,needs maintainer decision,1,0,102187,,,,0,b88259a5375f77ae9cbde83306d6cb02772f098d56ccd39ea1f9b10e30d0c16b
thread:issue:102197,issue #102197: torch.func.jvp fails when acting on a DistributedDataParallel model,needs maintainer decision,28,needs maintainer decision,1,0,102197,,,,0,988e3db2e696e7f0bf387b5c150e02f69470c451dae7ab992ef314f30f08e7bd
thread:issue:102402,issue #102402: [Composable] Unified way to check if modules are managed by composable API,needs maintainer decision,28,needs maintainer decision,1,0,102402,,,,0,8d59e8d5e773d64af44baf308cc8bf05c48826eb944f3cc0b9e8b6e1c4e6d762
thread:issue:102438,issue #102438: DDP multi node multi gpu inconsistent params,needs maintainer decision,28,needs maintainer decision,1,0,102438,,,,0,de09458ed1a2309c7741bdbda01fb1c22abad8e6fce879a0c913553814f49f3f
thread:issue:103222,issue #103222: compilation fails `error: invalid argument '-std=c++17' not allowed with 'C'`,needs maintainer decision,28,needs maintainer decision,1,0,103222,,,,0,78f15be091ea9cdba2db0576ab6f4fb03b0a96de71c7b6868985f88b987c67d8
thread:issue:103359,issue #103359: Libtorch compile error when defining D_GLIBCXX_DEBUG,needs maintainer decision,28,needs maintainer decision,1,0,103359,,,,0,beef015253932cf20aeb312e0c83f9ad55f0d3786fdb926977a0cc89ec4fdce8
thread:issue:103622,issue #103622: ARM based GPU support for Distributed Data Parallelism Module,needs maintainer decision,28,needs maintainer decision,1,0,103622,,,,0,837fd8fd6008b87642d7a6048ff15d13ea87b02ca0a3c6bbfa2df168a9e400d2
thread:issue:103626,issue #103626: 2D model checkpointing hangs on a ViT model,needs maintainer decision,28,needs maintainer decision,1,0,103626,,,,0,91970098be62803bb3bc0ccc49ea09eac74cc79f1505898ec3a3a2fd448c7e15
thread:issue:103650,issue #103650: AOT autograd: Avoid dependency on strides for manual regeneration of outputs that are aliased to inputs,needs maintainer decision,28,needs maintainer decision,1,0,103650,,,,0,335a20314dfe8183ca99bc0ef6c03635eb87d82c11ed949c465a731378ab7422
thread:issue:103841,issue #103841: [question] [docs] Short/mid/long-term status of TorchScript / JIT / torch.jit.trace / FX / symbolic tracing and its replacement by Dynamo,needs maintainer decision,28,needs maintainer decision,1,0,103841,,,,0,d15bf9e52788e634dd6f3bc2fd4363070999c02df748e57de26c760018dae6e9
thread:issue:104053,issue #104053: Tracking issue for optimizer graph not being an inference graph,needs maintainer decision,28,needs maintainer decision,1,0,104053,,,,0,61b89841fd1c132c846a72163c5eed5ec23f35da6ca80cea998d653e6ab9ac93
thread:issue:104146,issue #104146: make_fx: torch.where scalar promotion burns in device,needs maintainer decision,28,needs maintainer decision,1,0,104146,,,,0,42b472589ab17df1552765a80796b186c41c9d5bf91b2897378c52c304ca2bec
thread:issue:104164,issue #104164: DDP enhancement,needs maintainer decision,28,needs maintainer decision,1,0,104164,,,,0,f2c882c2d98160018538fae1c1a72e357f90f8daf4f112c8eb0cc649d6295edc
thread:issue:104194,issue #104194: View ops on fake tensors can dispatch `detach`es to backend kernels,needs maintainer decision,28,needs maintainer decision,1,0,104194,,,,0,6b77b8063303d589a40dc3347160474c77ddbe3cf01b34f6e4a6067b3a2588ae
thread:issue:104695,issue #104695: DISABLED test_cuda_memory_leak_detection (__main__.TestCudaMultiGPU),needs maintainer decision,28,needs maintainer decision,1,0,104695,,,,0,fe2a23c852650b88b2030319304ee68e5836113f01038bb1a8af796df1eaea7f
thread:issue:104701,issue #104701: System memory leak when using different input size of torch.nn.Conv3d,needs maintainer decision,28,needs maintainer decision,1,0,104701,,,,0,6e148ad7afb7997e1993675e603561f1bcd1815a50f2f3c8700213e192d08f6c
thread:issue:105632,issue #105632: torch.nn.TransformerDecoderLayer lacks parameter validation check,needs maintainer decision,28,needs maintainer decision,1,0,105632,,,,0,66996535ee7d73ae8603e8835e789734f57f01f94d9c1f26c45942891408114b
thread:issue:105723,issue #105723: Parameter ... has been marked as ready twice,needs maintainer decision,28,needs maintainer decision,1,0,105723,,,,0,36a94f754ddf57db8d53c906d52aadc6382c02334bb65f75b558a256a4d9f4e0
thread:issue:106006,issue #106006: [torch.compile] assertion sometimes ignored with inductor backend,needs maintainer decision,28,needs maintainer decision,1,0,106006,,,,0,a000dcae60fb6841654d27607bbdc42e282c6cc3f38dc2f17c1de7398566005c
thread:issue:106665,issue #106665: Inconsistency between CPU and GPU for `Linear()` layer with input size 0,needs maintainer decision,28,needs maintainer decision,1,0,106665,,,,0,cfe440185837da88e79fd02d3d2be2adc628f90eb4b90c42b7ef1dbe396d2c35
thread:issue:106692,issue #106692: torch.polygamma inconsistent with scipy.special.polygamma for n >= 1,needs maintainer decision,28,needs maintainer decision,1,0,106692,,,,0,bb5b6f01d7c38e5b768a526234801ff215345628e0d213a5601a2ae21a40d9e3
thread:issue:106732,issue #106732: Hugging Face safetensor does not work with FakeTensorMode,needs maintainer decision,28,needs maintainer decision,1,0,106732,,,,0,10f17c215e1856965d5b891e9ea4dddb151fca23bd09c5f5eb9a46efe5c4e26c
thread:issue:106851,issue #106851: RPC all_gather doesn't work with dynamic world size (world_size=None),needs maintainer decision,28,needs maintainer decision,1,0,106851,,,,0,a5f51ef2a1353e02024092a6f592ad0e870c7ca78c40557ae474397514aaf437
thread:issue:107174,issue #107174: Surface NCCL and CUDA version incompatibility,needs maintainer decision,28,needs maintainer decision,1,0,107174,,,,0,6df5453eb68845da6cce8ae207bed8db87a7baeea8a1dd4aa2ad0dcde2fe8247
thread:issue:107177,issue #107177: Timeout during NCCL initialization due to store,needs maintainer decision,28,needs maintainer decision,1,0,107177,,,,0,e0bb782fac010d88f22c42dec08ca035c6e4d58cadc59dc329aaf3973f8cc6bb
thread:issue:107582,issue #107582: [FakeTensor] `to` doesn't error with `allow_non_fake_inputs=False`,needs maintainer decision,28,needs maintainer decision,1,0,107582,,,,0,0fd9755f4ec5db127983475c6636fd966fb4a00b5b2edb062bf751d94904c24d
thread:issue:107631,issue #107631: torch.fx.Interpreter modules don't get compiled,needs maintainer decision,28,needs maintainer decision,1,0,107631,,,,0,2214f5c6cae8a832b59300d61f337b657afe4f4b672144d035d01e1bd1b14408
thread:issue:108212,issue #108212: Is there a standard procedure to check the consistency of environment across all nodes in PyTorch DDP training?,needs maintainer decision,28,needs maintainer decision,1,0,108212,,,,0,485523b4d1ffc8cdfdd8fea9ef759a77395b1fc855176b8274d76a6d64ab9e6a
thread:issue:108381,"issue #108381: FSDP always puts parameters to fp32 when loading state_dict, even if state_dict has bf16 params",needs maintainer decision,28,needs maintainer decision,1,0,108381,,,,0,463fd777c5392215acc742212199cdaf1a78de98e3602eda44c8eba2a80a4e7d
thread:issue:108521,"issue #108521: resutl of (torch.mm(a,b) does not match result of (a[:part,:], b)",needs maintainer decision,28,needs maintainer decision,1,0,108521,,,,0,5355fb84866fe398220da3eb991ae7d53d5a0998a4790bb6302162808f5a0140
thread:issue:108794,issue #108794: [C++ Frontend] Simple Changes for Cleaner Options,needs maintainer decision,28,needs maintainer decision,1,0,108794,,,,0,c6bbde0b00ea68ffc6949db4ca56ad035b4d8f439d45d09a122cae382a30fb9b
thread:issue:108971,issue #108971: Really slow compilation times for torch.compile causing distributed training errors,needs maintainer decision,28,needs maintainer decision,1,0,108971,,,,0,0baae7f6b63118e8386e770e5cc8ee40e8b7601aefd617e4f0b26fd0a50a927a
thread:issue:109103,"issue #109103: DDP - ""No backend type associated with device type cpu"" with new Model Phi 1.5 despite everything loaded on GPUs",needs maintainer decision,28,needs maintainer decision,1,0,109103,,,,0,4b84b6907e6ecc36447ff760d12c222d1629544a029f88720750b15cd3584a05
thread:issue:109204,issue #109204: Pytorch ROCM windows builds,needs maintainer decision,28,needs maintainer decision,1,0,109204,,,,0,2735e78e6c3777a47efa903df22f37d340616a3c053b50b8aabd0ffa1134198f
thread:issue:109237,"issue #109237: AOTAutograd should track view chains so it can replay them, instead of using as_strided.",needs maintainer decision,28,needs maintainer decision,1,0,109237,,,,0,9247c98fce0e0d280d756a1bf698297f99c6b1e05c6e189ebfa484c1deb40965
thread:issue:109477,issue #109477: [DTensor] optimizer step performance is still too bad,needs maintainer decision,28,needs maintainer decision,1,0,109477,,,,0,041fc48f0ea3cb30fc1ef5e35a5e32a52af2d16f83c5f73e39e7aaec5e5bc3e6
thread:issue:109592,issue #109592: test_memory_timeline fails on PPC due to extra temopraries,needs maintainer decision,28,needs maintainer decision,1,0,109592,,,,0,3ebc42eef1c309571149a590daad4dc6e2d51bfc63cf02e5723bfcf9649fa466
thread:issue:110281,issue #110281: feat(Pipeline Parallelism): use mincut optimization for local communication optimization,needs maintainer decision,28,needs maintainer decision,1,0,110281,,,,0,e57ce4e5bd0f26bbc5315634023d81719fd9d2dea1ee0147f0c4e6883f12e37e
thread:issue:110700,issue #110700: More informative variable names in AOTAutograd,needs maintainer decision,28,needs maintainer decision,1,0,110700,,,,0,ef131af2975cf2c1b2c34754707dc480ba746feed0cb472574fdf91e34dbe121
thread:issue:110775,issue #110775: torch.distributed.pipeline skip module throws assert error that portal.grad is not None,needs maintainer decision,28,needs maintainer decision,1,0,110775,,,,0,27909906dd49c3cb9f566f02eca33207226818366d4b4bb19977ca5baf09978c
thread:issue:111131,"issue #111131: ""Invalid Scalar type"" when using bf16 allreduce with Gloo backend",needs maintainer decision,28,needs maintainer decision,1,0,111131,,,,0,cc2bc294e00ff88b4c350601ab6356c28fbef78a1ed8ec45b4345a111c617725
thread:issue:111646,issue #111646: torchrun: elastic training not restarted on missing keep-alive heartbeat/scale-down event,needs maintainer decision,28,needs maintainer decision,1,0,111646,,,,0,a1a5084d47664277db015176cc23127b36db1825e0b4ce27592164db8ff85940
thread:issue:111824,issue #111824: GroupNorm & InstanceNorm does not handle channels_last correctly,needs maintainer decision,28,needs maintainer decision,1,0,111824,,,,0,da584b2cf49f18f0e314805c547a32d919873314d2755737226a42052ab6ed0a
thread:issue:111834,"issue #111834: Numerical inaccuracies in ""ddp_apply_optim_in_backward"" unit tests for gloo backend",needs maintainer decision,28,needs maintainer decision,1,0,111834,,,,0,dc87dd7f7813c4f215c98c4e4a4d5aa0789f70340ece2db7dacfc42dbd1b33fc
thread:issue:111943,issue #111943: Implement scatter and broadcast in functional collectives,needs maintainer decision,28,needs maintainer decision,1,0,111943,,,,0,8b8d2bb1b83d2d02f352ca92bece052b275d82d2e2a877712e715cfdeaf647cd
thread:issue:112029,issue #112029: DDPWrapper,needs maintainer decision,28,needs maintainer decision,1,0,112029,,,,0,1488112aa26191190928c7675a446fd3e689705e8eb78d08c4ef78cbbd294fee
thread:issue:112277,issue #112277: test_DistributedDataParallel fails with static_graph=True,needs maintainer decision,28,needs maintainer decision,1,0,112277,,,,0,19b54665d4fbe050c6d62a7336061b9d72b106728abb85d9a729bf536105f2c3
thread:issue:112344,issue #112344: [test/dynamo] BE: cleanup `test_misc.py`,needs maintainer decision,28,needs maintainer decision,1,0,112344,,,,0,12fc71432b0d49e3d6a68c46db4f54edc6d37a135bfb80aae3cbead2a00614e9
thread:issue:112827,issue #112827: Multi Scale Deformable Attention Support,needs maintainer decision,28,needs maintainer decision,1,0,112827,,,,0,fd10682c5aa95c07a63cefea58ddd04318746c8d02ccde7832918d86aa0b7795
thread:issue:112903,issue #112903: Support for Bazel workspace function or Bazel module,needs maintainer decision,28,needs maintainer decision,1,0,112903,,,,0,9d124a496354a75e460be078803c36466e004cba2761c035d2677143a24c764c
thread:issue:113089,issue #113089: Add docs for __tensor_flatten__ / __tensor_unflatten__,needs maintainer decision,28,needs maintainer decision,1,0,113089,,,,0,5c903b0641be01aa1b11ddc79dc7736a46639a7fa627a10d04085ea7d646d92d
thread:issue:113114,issue #113114: Allow to construct distributed Work object from Future,needs maintainer decision,28,needs maintainer decision,1,0,113114,,,,0,12da539835e1402e7d03d0070b1176026da9e1c5d084283972e570bfb6ffe079
thread:issue:113149,issue #113149: Compilation error on loongarch64,needs maintainer decision,28,needs maintainer decision,1,0,113149,,,,0,24241350c94b452e83f60849ef676ad6adcc5e2c1d8be0ed0fe289c7e1df3dfb
thread:issue:113228,issue #113228: dataloader use lmdb stuck,needs maintainer decision,28,needs maintainer decision,1,0,113228,,,,0,0cf97b59038fed0589e76aa700c336eae0cf88e1a66a6d0547527ae86cd019a4
thread:issue:113281,issue #113281: NCCL p2p ops hung after communicator aborts,needs maintainer decision,28,needs maintainer decision,1,0,113281,,,,0,86d36dee28e6122cc64c52145f8c4bce7782262a7c8783896a2f7ecb00d753b9
thread:issue:113376,issue #113376: Add validation for send/recv sizes,needs maintainer decision,28,needs maintainer decision,1,0,113376,,,,0,7e8f915e7f20d19c45e1cb1f67fa03321afab06337b09f9bac882d339b34735a
thread:issue:113589,issue #113589: Fake tensor produces incorrect values w/ is_coalesced and sparse_coo,needs maintainer decision,28,needs maintainer decision,1,0,113589,,,,0,f1878b463c37b48cf9da3e6a5b082186796a3107a56aaa0e97b1dd016a6ac9a3
thread:issue:113643,issue #113643: Distinguish immutable/mutable fake tensor mode,needs maintainer decision,28,needs maintainer decision,1,0,113643,,,,0,ee70df47218ed23e7c5a4253647cd1f418c90b74708b91108036f4f8812a667b
thread:issue:113669,issue #113669: document functional_collectives,needs maintainer decision,28,needs maintainer decision,1,0,113669,,,,0,20741c82dd51f3e238af5699a9d3c516cd213ffa706fdb7b0b2d7b6bb8505082
thread:issue:113708,issue #113708: `log_softmax` could be `2**124` to `2**1021` times more accurate on small outputs,needs maintainer decision,28,needs maintainer decision,1,0,113708,,,,0,88d7d53b5d3da88c4d497a942aefb3b26866e7f4d50df00e4b47f5b902332c20
thread:issue:114437,issue #114437: [Help Needed] [DataParallel/DataLoader/AMP] Severe slowdown on large dataset with fp16 precision and DataParallel wrapper,needs maintainer decision,28,needs maintainer decision,1,0,114437,,,,0,8aa050730ca411b77f917f840da6f255a43e24ce81998e3cafb2c45c96eb6c85
thread:issue:114765,issue #114765: DistributedDataParallel _verify_param_shape_across_processes causes reserved memory on GPU,needs maintainer decision,28,needs maintainer decision,1,0,114765,,,,0,adc9fc83b3ca27c95de940088ae12405b1ef031cd4db3a1bd939906078301950
thread:issue:114857,issue #114857: recursive search for DDP ignored parameters,needs maintainer decision,28,needs maintainer decision,1,0,114857,,,,0,727b28693d6a2065f11675e934d37d7926d7c93ac4077223ffdc90cf6a452a3d
thread:issue:114884,issue #114884: Undefined reference to ncclCommSplit,needs maintainer decision,28,needs maintainer decision,1,0,114884,,,,0,3730391884882d2d2b6bfec626a6ee16837993424031c4febaeebb117fff554b
thread:issue:115741,issue #115741: torch_function mode explodes with torch.compile,needs maintainer decision,28,needs maintainer decision,1,0,115741,,,,0,88665ca9369d08fed9ad3c8869eef4eca6bbf94b620be02aab1ff990df88cf4f
thread:issue:115861,issue #115861: gather_object does not work with ThreadPoolExecutor,needs maintainer decision,28,needs maintainer decision,1,0,115861,,,,0,e5be1551795e56d40324d296dd93c2ae05196ee4bc0f0493a4d311788b32da09
thread:issue:116183,issue #116183: Pipeline parallel doesn't seem to work with mixed precision,needs maintainer decision,28,needs maintainer decision,1,0,116183,,,,0,dfcab8edf34a619aa7637c20a19722b9112e9dc417288a8b606c35bce141b508
thread:issue:116683,issue #116683: DTensor tensor subwrapper not be handle with apex,needs maintainer decision,28,needs maintainer decision,1,0,116683,,,,0,8ee5766ad5ad54c523529d8688ccada15f1a4956f56334f003e8541f181c46e8
thread:issue:116697,issue #116697: Tracing GPU usage of `torch.vmap` vectorized operation using `torch.profiler` cause CUDA OOM,needs maintainer decision,28,needs maintainer decision,1,0,116697,,,,0,0b8d381345be67b598beb438dfbf1a14e94542ffb3acc29049d52ed12905e71c
thread:issue:117115,issue #117115: request: torch.compile support for tensor subclass subgraph boundaries,needs maintainer decision,28,needs maintainer decision,1,0,117115,,,,0,7428ce9099c8fad8988e1343e45c9ed430346a8ce64ea3b7a3d895d089f35855
thread:issue:117159,"issue #117159: If error is raised in distributed setting, top level exception formatting should include rank",needs maintainer decision,28,needs maintainer decision,1,0,117159,,,,0,3466d73a18ae92f0d8b1315449b3b145cc1f2a32ea109b05ab696c89f159f1a8
thread:issue:117662,issue #117662: dynamo <> autograd.Function assumes that grad_outputs have the same subclass-ness as forward outputs,needs maintainer decision,28,needs maintainer decision,1,0,117662,,,,0,11692743b12365de3f261b505b29c6ae546c49d184028939f3fb2fa09fa3db10
thread:issue:117755,issue #117755: [c10d] Enqueue work pointers instead of work copies,needs maintainer decision,28,needs maintainer decision,1,0,117755,,,,0,1d560b14e3cb7141c85c7bebfc2801767d45d0a0671e7423214c08f202ff8495
thread:issue:117974,issue #117974: RuntimeError: derivative for aten::_scaled_dot_product_efficient_attention_backward is not implemented,needs maintainer decision,28,needs maintainer decision,1,0,117974,,,,0,fd3cee7e16d4cd1c4a42ff41946ab79242e27ff4e7161e5a24147b1bc5abaa96
thread:issue:118094,issue #118094: DTensorConverter::successful returns False on no Tensor args causing test failures,needs maintainer decision,28,needs maintainer decision,1,0,118094,,,,0,35d8e7db5a40f609285fddbacaa7f1b2c96671a8987e7a9f6d827713b3f01531
thread:issue:118211,"issue #118211: Dynamo codegen for mapping outputs back to user variables should contain safety checks (no fake tensors, no variable trackers)",needs maintainer decision,28,needs maintainer decision,1,0,118211,,,,0,d65a9f17366eacb0cd01795efb388f0e706cb68c8784a43d559076f27e0007c1
thread:issue:118521,issue #118521: Improve constant propagation in the dynamic shape case,needs maintainer decision,28,needs maintainer decision,1,0,118521,,,,0,4579ed1bcc14feb7c13a2db709024ea516b987f42728d588c34f015e028313e7
thread:issue:119196,issue #119196: Possible NCCL race condition `destroy_process_group` followed by `init_process_group`.,needs maintainer decision,28,needs maintainer decision,1,0,119196,,,,0,94fae6522effe52f6d37c0908d7c199003ff8317753512d74a79368f8038ac42
thread:issue:119328,issue #119328: [pytree] Handling of `None` in torch.utils._pytree is inconsistent with JAX.,needs maintainer decision,28,needs maintainer decision,1,0,119328,,,,0,780f10db7f9a650cbb4e3690741a38bb92a10681477209dcb4c701da2d0e275e
thread:issue:119407,"issue #119407: convolution_backward has different size results on ""CPU"", FakeTensor",needs maintainer decision,28,needs maintainer decision,1,0,119407,,,,0,8483e639357b3b1a5aa9ade6a11616e7f6453093eb904c57b4be812f42e670b8
thread:issue:119455,issue #119455: Users confusing torch.distributed.nn.functional with functional collectives,needs maintainer decision,28,needs maintainer decision,1,0,119455,,,,0,5d5b9b5cc32f4847ceda8a06b874d6844b4694b1503d3f09aceab4d719c59c4f
thread:issue:119461,issue #119461: Promote pytorch nightly docker image,needs maintainer decision,28,needs maintainer decision,1,0,119461,,,,0,22fdaf4045b3a0278738ee7cb58bedc12cee69e0bb605010976bc1ff4463c2d7
thread:issue:119589,issue #119589: Free-floating fake tensor without the mode (FakeTensorMode),needs maintainer decision,28,needs maintainer decision,1,0,119589,,,,0,9b105bd50c5a4c5d929eaf111b747b8c698d981c6f9fea17a5ea911113fb877a
thread:issue:119758,issue #119758: Flight Recorder dump_entries() segfaults when used with coalesced operations,needs maintainer decision,28,needs maintainer decision,1,0,119758,,,,0,c47d6c92dc6351115885b26e598e3b0091566c3fac81399b4440608610cc3eec
thread:issue:119856,issue #119856: Timeout for process termination upon receiving sigterm should be a user configurable parameter,needs maintainer decision,28,needs maintainer decision,1,0,119856,,,,0,1d3f6a9b0014111514dc3c29aa9e9eb61da9ff00b9d958899da37c2d27ed2c84
thread:issue:119890,issue #119890: dynamo remap collective improvement,needs maintainer decision,28,needs maintainer decision,1,0,119890,,,,0,4c961470ad9bcbdf0a79e0f07298a3ce422b44f3b84255537e686adae041f522
thread:issue:119930,issue #119930: [dtensor] ops coverage tracker,needs maintainer decision,28,needs maintainer decision,1,0,119930,,,,0,b76473e80b4fae88bc282bfdc7aefc415eccd71de57def87a1672c2f7a3a08ca
thread:issue:120105,issue #120105: Weird Artifact name: 'not_implemented' not registered being raised,needs maintainer decision,28,needs maintainer decision,1,0,120105,,,,0,83af15aaeed228a3f1f246b1a2a91abb000719b729b78ccb84d160708ef3d030
thread:issue:120158,"issue #120158: The forward accuracy cannot be aligned when using DTensor and ColwiseParallel, RowwiseParallel for TP training on the Llama2",needs maintainer decision,28,needs maintainer decision,1,0,120158,,,,0,4f9c1813b18fa48fac469469f21d894d7485d0ce06f6d07efdf574d9b910d542
thread:issue:120308,issue #120308: ROCm: foreach is substantially slower than for loop MI250x,needs maintainer decision,28,needs maintainer decision,1,0,120308,,,,0,614299f799c22bcfca65853fd0ad4df6ed5b25ad037e047d0e7baae01c64f63e
thread:issue:120516,issue #120516: [C10D] coalesced collectives are not well tested,needs maintainer decision,28,needs maintainer decision,1,0,120516,,,,0,6a41916f7158c3826a994a12b83cfbb4d0543bed73e80af48b8aa20ff8e5e76d
thread:issue:120741,issue #120741: [C10D] coalesced op hang does not respect timeout,needs maintainer decision,28,needs maintainer decision,1,0,120741,,,,0,0d2b3fb283c964871f545504935d768a5fe9181e7774ba8295bf47c825b391b9
thread:issue:120787,issue #120787: Tensor Parallel Inference Performance Issue with Llama-7b-chat-hf Model caused by straggler problem,needs maintainer decision,28,needs maintainer decision,1,0,120787,,,,0,0af26cceb13d175262511357cedc31006c5f90362afac628e986605d697b3011
thread:issue:120902,issue #120902: RuntimeError: cuFFT error: CUFFT_INTERNAL_ERROR,needs maintainer decision,28,needs maintainer decision,1,0,120902,,,,0,06cf0aadde1a70e069074c48ec9edeece95fa69c4acd435c1fba2c63d0c3279d
thread:issue:120945,issue #120945: DISABLED distributed/_composable/fsdp/test_fully_shard_training.py::TestFullyShard1DTrainingCompose::test_train_parity_with_activation_checkpointing,needs maintainer decision,28,needs maintainer decision,1,0,120945,,,,0,afed714fef7f0aae6b959cd445305c7db366958373140ebf7c2e491757c30a3a
thread:issue:121122,issue #121122: Evaluate defaulting `TORCHINDUCTOR_CACHE_DIR` to `~/.cache` on Linux systems,needs maintainer decision,28,needs maintainer decision,1,0,121122,,,,0,fe951d101426e8f0c048e131df990f3185a0d6740162878279eb27d86f6a6e7e
thread:issue:121135,issue #121135: [torch.compile] `randperm_index_add_pattern` doesn't check the shape mismatch between self and source tensor,needs maintainer decision,28,needs maintainer decision,1,0,121135,,,,0,088a5cc531d7fb58bacee108db25a2278d22343e710028157cd2773c908dac2d
thread:issue:121181,issue #121181: Profiling tool for DDP and FSDP,needs maintainer decision,28,needs maintainer decision,1,0,121181,,,,0,76fc1d04a23d212b975297bba09901ac56101d1456779cccdc652b4b7279f15e
thread:issue:121486,issue #121486: Kubernetes API server as Rendezvous backend,needs maintainer decision,28,needs maintainer decision,1,0,121486,,,,0,aba7ec19252df46a242a6f4c8d73e0cf922b6df4820ea9b9023a45359b2bba4d
thread:issue:121543,issue #121543: [DCP] test_state_dict.test_fsdp is failing on unrelated tests in CI.,needs maintainer decision,28,needs maintainer decision,1,0,121543,,,,0,3d432711759e640a24a4d4697d4a249c990b3f8cda782ac2fd8c3cc6fd8c2b1f
thread:issue:121587,issue #121587: Add clearer docs for `torch.distributed.<OP>` vs. `torch.distributed.nn.<OP>,needs maintainer decision,28,needs maintainer decision,1,0,121587,,,,0,5e3fcd8aef99e12e006a8d9209816d387770c360412fa563203933fa15b9bc83
thread:issue:121594,issue #121594: [DDP] Gradient Synchronization Failure Induced by model.gradient_checkpointing_enable() in multi-task setting,needs maintainer decision,28,needs maintainer decision,1,0,121594,,,,0,45c5e9648ec82e5a232f7aaa5938c4dc09abc98eeca761ad2e357a7973dabfa6
thread:issue:121605,issue #121605: Make the data type of local_used_map_ in the Reducer class configurable,needs maintainer decision,28,needs maintainer decision,1,0,121605,,,,0,b7c54cb806116d0c768fae83beedd04731dbb5a0a87fb3b6ca21c0f6c8451923
thread:issue:121818,issue #121818: torch._check but trust me,needs maintainer decision,28,needs maintainer decision,1,0,121818,,,,0,bb6809488b05a9262e13a1046d2b5a83dd2bf79f8a5c6d14642954c2f7cc7a02
thread:issue:121837,issue #121837: Revive libtorch-shared-without-deps flavor,needs maintainer decision,28,needs maintainer decision,1,0,121837,,,,0,760aacf4fba9221efa0ef08aedf77715bc04d538f34d4fba6e6a713db00e0ced
thread:issue:122588,issue #122588: [ONNX] Applying consecutive conv2d layers in a for loop,needs maintainer decision,28,needs maintainer decision,1,0,122588,,,,0,046a574aa14799197b1a931b3a6a2725449b6421cb2635da567c346d48be010f
thread:issue:122806,issue #122806: Deprecate accessing FakeTensor.data_ptr outside of torch.compile,needs maintainer decision,28,needs maintainer decision,1,0,122806,,,,0,bdd4cac2455b07667baab252b2f5d71cf90f9f67fe46a37e7777b0684378f8e0
thread:issue:123148,issue #123148: GRU is super slow on MPS backend,needs maintainer decision,28,needs maintainer decision,1,0,123148,,,,0,b6483ee4319974bf40379d6781c91e2dc1420df2ed7d97f5bb11ca009a06cfc3
thread:issue:123252,issue #123252: FakeTensor support of pin_memory,needs maintainer decision,28,needs maintainer decision,1,0,123252,,,,0,e6b9fd24e5f1abac681b7d267cfff4430f2e6fcd0306ee48cde0272a6729816c
thread:issue:123815,issue #123815: my_role_info = role_infos group_rank index error,needs maintainer decision,28,needs maintainer decision,1,0,123815,,,,0,c75e6a403a2f048107dfeff2776148f85e66468c327b2db47614b05a061fb528
thread:issue:124019,issue #124019: [FSDP+TP] RuntimeError: 'weight' must be 2-D,needs maintainer decision,28,needs maintainer decision,1,0,124019,,,,0,a278169dac253914fa1c52858f022a55eb233ac4911cc3014d6975b0d018d7b5
thread:issue:124058,issue #124058: [RFC] Support Torchrun for Out-of-the-tree Devices,needs maintainer decision,28,needs maintainer decision,1,0,124058,,,,0,d2d3f7f862c9cb234f2c064aa104cca30a86d54fdb4583b6e22f352866621fa5
thread:issue:124101,issue #124101: [C10D] Standardize torch distributed launcher interface,needs maintainer decision,28,needs maintainer decision,1,0,124101,,,,0,4923b01837b2289506ce0f5dad7c36d3243891effabea928b929f6cfb9bc66be
thread:issue:124152,issue #124152: DISABLED test_torchvision_models_efficientnet_v2_m (__main__.TestVisionTracing),needs maintainer decision,28,needs maintainer decision,1,0,124152,,,,0,a4ce5a71059729253fe686c4f8b7e712b56ef9fbf93b76d200b4b23d604b0f1d
thread:issue:124204,issue #124204: RPC + DDP VRAM overflow to rank 0 GPU.,needs maintainer decision,28,needs maintainer decision,1,0,124204,,,,0,8d7a9ef41558f5e1c31ea4cec7aad1681cd235ce3eeb969937388923267c2d05
thread:issue:124260,issue #124260: [FSDP] Unreleased reserved memory after FSDP wrap,needs maintainer decision,28,needs maintainer decision,1,0,124260,,,,0,92d475a42c472ad9da2c3cf75a038c234cea286b1e28861513980b8f8d6c8f1f
thread:issue:124317,issue #124317: Many tests under test/distributed/elastic not running in OSS CI,needs maintainer decision,28,needs maintainer decision,1,0,124317,,,,0,32d9f8d6e18fc6f08fe314eb648b688a2cf22b01d5fa7c32b7093696b0b2ae0b
thread:issue:124372,issue #124372: Pipeline Parallelism support for ZeroBubble,needs maintainer decision,28,needs maintainer decision,1,0,124372,,,,0,59d0ac2cc4e9a90d096d0d9cc9218d411d446b9bf56c0e61c8cd535d82618324
thread:issue:124380,issue #124380: Error when running DDP examples to train between nodes,needs maintainer decision,28,needs maintainer decision,1,0,124380,,,,0,d9727fbc7f7cb3127c861c22aa27f9b05491395720a3df4dd74cdb3981a31968
thread:issue:124468,issue #124468: Race in ProcessGroupNCCL shutdown,needs maintainer decision,28,needs maintainer decision,1,0,124468,,,,0,8024469c8135db3861867fd9c7cd976b9072dab5ccd89ea9a49beafdc8baefe9
thread:issue:124680,issue #124680: Inconsistent results when training a model containing SyncBatchNorm with multiple GPUs,needs maintainer decision,28,needs maintainer decision,1,0,124680,,,,0,41a93f17a3be0fea8d14d97597058c4278bcae1906e8ace849d91d95cd0a2fdf
thread:issue:125030,"issue #125030: Allow operators with no returns to be ""out="" operations",needs maintainer decision,28,needs maintainer decision,1,0,125030,,,,0,9c3b3b7eeb62731148215f2c34e56785c2eb79612a339870ac685313ad4129eb
thread:issue:125254,issue #125254: AssertionError: Device mps not supported,needs maintainer decision,28,needs maintainer decision,1,0,125254,,,,0,081d4b7b23f281ff0e06eb0794353968d387ed49bcadd2a7d0212e16da1bf843
thread:issue:125557,issue #125557: result of 2 ** s differs between eager mode and inductor + triton + cuda when in float32 denormal range,needs maintainer decision,28,needs maintainer decision,1,0,125557,,,,0,ff4b92a3032fd8e3dd491fb790eb6e9f24c1290fa52b15f4b31e5ab88c82bf20
thread:issue:125604,issue #125604: Compiling GPT2 with tensorrt backend and dynamic shapes results in guard failures.,needs maintainer decision,28,needs maintainer decision,1,0,125604,,,,0,587e461904f7d28ae4f71ddd9328e09bdec101579c0167230ea143b8864128ea
thread:issue:125977,issue #125977: [compile time] AOT Autograd is taking long time in tracing,needs maintainer decision,28,needs maintainer decision,1,0,125977,,,,0,76e240a0690183e5c70916ddc12d820f4ec32e2fa5abe6400728e4aafb02e70d
thread:issue:126158,issue #126158: [torch.arange] Output mismatch with Numpy / TensorFlow,needs maintainer decision,28,needs maintainer decision,1,0,126158,,,,0,4bbae79cdb67692341a0e4e4bcce2391668b72e2265c37087ee9351e40f76244
thread:issue:126302,issue #126302: Make it easier to run distributed tests without run_test.py,needs maintainer decision,28,needs maintainer decision,1,0,126302,,,,0,3d368955a611fe315b1b8cd6b9ff66b515db8ac2670f5767ec571acb7222624b
thread:issue:126804,"issue #126804: Torch fails to build with cub from CCCL v2.3.2 or newer, due to -D__CUDA_NO_HALF_OPERATORS__",needs maintainer decision,28,needs maintainer decision,1,0,126804,,,,0,7ff88ca7a387d9c0a8ce837c7324fed2f04efdbe5eb5565b7b85938822621faf
thread:issue:127176,"issue #127176: RuntimeError: ""_amp_foreach_non_finite_check_and_unscale_cuda"" not implemented for 'BFloat16'",needs maintainer decision,28,needs maintainer decision,1,0,127176,,,,0,b09792a10bfb943670a58ffa828410ce7e7b2a4dfa80d4944d074243b26a52a2
thread:issue:127445,issue #127445: PyTorch version agnostic C++ extensions,needs maintainer decision,28,needs maintainer decision,1,0,127445,,,,0,043a97d65567ba47ef9d90ce1f88bdf42b14b8db5b73f26867515510bb19f0a3
thread:issue:127605,issue #127605: RuntimeError: CUDA error: an illegal memory access was encountered with specific input shape and Conv1d,needs maintainer decision,28,needs maintainer decision,1,0,127605,,,,0,f0d77ef425ba25c2bf24d1a2a601d236da007fdbafd924fef5f8da3564b2264f
thread:issue:127621,issue #127621: custom_op API: better type anntation for Tuple,needs maintainer decision,28,needs maintainer decision,1,0,127621,,,,0,f245739866121434b16184ab285ba671318b9cc67b9febc2d2ca5ea31e233e9b
thread:issue:127772,issue #127772: Run existing eager DTensor tests under torch.compile,needs maintainer decision,28,needs maintainer decision,1,0,127772,,,,0,8dc26c6f750eb9f0deca592def868446d8c611c4a0cbe4a8117a7fcceb389f13
thread:issue:127882,issue #127882: tensor.size does not accept `dim=None` argument despite documentation because of some confusion with named dimensions,needs maintainer decision,28,needs maintainer decision,1,0,127882,,,,0,2697a3e5f34668afa5d10e5f87ef9da7644f9b65c416f72525c695093d8b513f
thread:issue:128121,issue #128121: 2nd compile of deepcopy(model) fails on multiple ubuntu-pc (fatal error: Python.h: file not found),needs maintainer decision,28,needs maintainer decision,1,0,128121,,,,0,e761be10e006ba286f94fcf2b04046ba4a21f491dcba880283e8f0b59bc8b3fe
thread:issue:129279,issue #129279: [ONNX] Create unit tests for the new export path by adapting all existing tests,needs maintainer decision,28,needs maintainer decision,1,0,129279,,,,0,e45255435f368bbf83aa2e59f810cb0e8874369da8e299b753238ecb2f7ee98e
thread:issue:129395,issue #129395: [custom_op] Figure out what to do with random custom ops,needs maintainer decision,28,needs maintainer decision,1,0,129395,,,,0,ee3b91331b8b3d01e7dc85da499833df5cf4170206010f61cf2a0560d200cf48
thread:issue:129430,issue #129430: OperatorBase.py_impl should support with_keyset=True extra argument.,needs maintainer decision,28,needs maintainer decision,1,0,129430,,,,0,0160706760d3f6b43f88a867860ef3626693f96297acceb6806a223546d48586
thread:issue:129862,issue #129862: The call to ncclCommSplit does not have the correct config parameters set,needs maintainer decision,28,needs maintainer decision,1,0,129862,,,,0,dc619d73e9d5ca6b93ef0cf16fd1d31e1f7ec9b8fc5246610beec7dbc2f1061e
thread:issue:129964,issue #129964: run dynamo CI shard with aot_eager_decomp_partition backend to see fallout,needs maintainer decision,28,needs maintainer decision,1,0,129964,,,,0,c631811be6abc4f8c76015f293f75b29d26f946fa4061f346dc7723c37dc9b8a
thread:issue:130015,issue #130015: Improve the Inductor generated kernel for the pattern of `output1 = pointwise(intput); output2 = transpose(output1)`,needs maintainer decision,28,needs maintainer decision,1,0,130015,,,,0,659519f778ddf64d6d9ba1c98aae443484dff1b230571916e4f9c9873613c231
thread:issue:130028,issue #130028: Support python attribute mutations of tensor subclass graph inputs from within __torch_dispatch__,needs maintainer decision,28,needs maintainer decision,1,0,130028,,,,0,fe53f97f95f5b50e2f13016999294d88dd5a63c74176c6b92fecc5d85aad1553
thread:issue:130039,issue #130039: `min_cut_rematerialization_partition` is not optimal memory usage.,needs maintainer decision,28,needs maintainer decision,1,0,130039,,,,0,59873ab06327116bec8a8d8afb65e9595ef06eae2a46b9f1d708936626a23c26
thread:issue:130330,issue #130330: why not enable expandable_segments by default？,needs maintainer decision,28,needs maintainer decision,1,0,130330,,,,0,bd5717e06fff4f88a411ef26188d7483540bf724e8192bdecf429ab3b1ed1795
thread:issue:130614,"issue #130614: Significant Memory Increase with Conv2d using padding_mode 'replicate', 'reflect', or 'circular'",needs maintainer decision,28,needs maintainer decision,1,0,130614,,,,0,c81ccac28b6c55df7163a15b2b0214fb910aaf48e0b71761b2f74b972c7e3ab8
thread:issue:131009,issue #131009: [cudagraph] an issue related to dynamic shapes for cudagraph backend with sd2.1,needs maintainer decision,28,needs maintainer decision,1,0,131009,,,,0,ed7572f8a9bf3b34b385a3207fba44b653ee7a94feae54351e009dbb8da9883b
thread:issue:131737,issue #131737: Intra-Node one-shot/two-shot low latency kernels for fp32,needs maintainer decision,28,needs maintainer decision,1,0,131737,,,,0,37d1f22334f265d73ad57a73daf9043f076716d2aff71935bc3a082984541854
thread:issue:131739,issue #131739: One Shot Low-Latency kernel is 2x higher than expected,needs maintainer decision,28,needs maintainer decision,1,0,131739,,,,0,ed88e154948231c8a9800ae810a6c8a50cb8cf649c5c022996353f52d323eafc
thread:issue:131754,issue #131754: Coredump for pytorch ddp when job finished,needs maintainer decision,28,needs maintainer decision,1,0,131754,,,,0,b74ea65037998c2cc50b09d25b182de6c109d47f775ddbdc4bc3ca6246fb10e9
thread:issue:131972,issue #131972: backward of adaptive max pool (adaptive_max_pool2d_backward_cuda) doesn't have a deterministic implementation,needs maintainer decision,28,needs maintainer decision,1,0,131972,,,,0,1f8eda630909820ace5ff60a5154fe1ae10ed83fb16544a2a1bc7c60ff5cbb10
thread:issue:131974,issue #131974: Missing grad_fn information while torch.compile with customized gradient function,needs maintainer decision,28,needs maintainer decision,1,0,131974,,,,0,50799e2a30ed7bce34083c4a786f5eaab4399ad7f84ba0d608a871ce04a94a6d
thread:issue:132865,issue #132865: aotautograd CSE interferes with Inductor reinplacing,needs maintainer decision,28,needs maintainer decision,1,0,132865,,,,0,eed5d5b34f815aa784e819603131938f69ef7cad6aa101e45cb84093a65efa20
thread:issue:132989,issue #132989: Different random seeds can cause differences in arithmetic precision,needs maintainer decision,28,needs maintainer decision,1,0,132989,,,,0,386b2f370c15b581ad8ef52e4433d4377e44bf3c378f6ac7a9718fa06e751596
thread:issue:133168,issue #133168: The more detailed GELU formula should be added to the doc of `GELU()`:,needs maintainer decision,28,needs maintainer decision,1,0,133168,,,,0,9c2d67f5d991720864aa49f631039a21a5d1a0b63a48d26acf01149c4d91ddfb
thread:issue:133367,issue #133367: Adafactor foreach impl performance tracker,needs maintainer decision,28,needs maintainer decision,1,0,133367,,,,0,8c343353fe7d39e2886ffc27f7ecf84baf94cddba46febb02b1a0f8cd599d14b
thread:issue:133511,issue #133511: Do dynamo/inductor support custom classes like TorchScript.,needs maintainer decision,28,needs maintainer decision,1,0,133511,,,,0,f6b6a8a5de7a9b3d6f7601fed6d913a4766aff7f69468ea6c15d79dac65babb2
thread:issue:133559,issue #133559: Support for variable number of dimensions in functorch.dim dims() and autonaming induced inconsistencies,needs maintainer decision,28,needs maintainer decision,1,0,133559,,,,0,8df75d93d89b407bbd062d8450f4f267feb331a5550874d80c7bd983b15b550c
thread:issue:133702,"issue #133702: Add `beta`, `betainc`, etc. to `torch.special`",needs maintainer decision,28,needs maintainer decision,1,0,133702,,,,0,1590319a13d635b7b1eb1e493fe7a713c3f26ece79b5e83be7d2b724f4c1b5e8
thread:issue:133721,issue #133721: [HOP] Add BC tests for hop in export,needs maintainer decision,28,needs maintainer decision,1,0,133721,,,,0,ae45d964fa9181d2eab7257707b693114623d585b6c731ebda79efd92d682462
thread:issue:134767,"issue #134767: Setting a negative, `int` and `bool` value to `label_smoothing` argument of `nn.CrossEntropyLoss()` works against the error messages",needs maintainer decision,28,needs maintainer decision,1,0,134767,,,,0,2829574fdc0d5f56caf4954150aa69931d1a1ebf5312979be1f85131b7856d6c
thread:issue:134956,issue #134956: Pytorch XPU Windows build failed in cmake rerun loop due to the source code deep path,needs maintainer decision,28,needs maintainer decision,1,0,134956,,,,0,44b9f3bc3f5adcd45381d7187345e0680b033d5217efa826ecea3ef8572898d0
thread:issue:135178,issue #135178: [DTensor] loss_parallel not worked with shifted logits and labels,needs maintainer decision,28,needs maintainer decision,1,0,135178,,,,0,979c827ffb6f1e3ef273155e14fe35f2315827169039c629582ec24418ccf4cb
thread:issue:135358,"issue #135358: UserWarning: Grad strides do not match bucket view strides. This may indicate grad was not created according to the gradient layout contract, or that the param's strides changed since DDP was constructed. This is not an error, but may impair performance. grad.sizes() = [1, 64, 64, 64], strides() = [1, 4096, 64, 1] bucket_view.sizes() = [1, 64, 64, 64], strides() = [262144, 4096, 64, 1]",needs maintainer decision,28,needs maintainer decision,1,0,135358,,,,0,e84b86a510c305730890665b3ddc64fd055480baf67d68f9abd995c94ed7e590
thread:issue:135508,"issue #135508: BUG: `torch.cuda.is_available()` returns `False` in certain torch, CUDA and driver version",needs maintainer decision,28,needs maintainer decision,1,0,135508,,,,0,4c317b16b2d47b72ce9aa22dea193ba050bfc3079a66aac40c07bc22c3100ef1
thread:issue:135511,issue #135511: RuntimeError: Modules with uninitialized parameters can't be used with `DistributedDataParallel` when using subclass of tensor.,needs maintainer decision,28,needs maintainer decision,1,0,135511,,,,0,0ac24368a977fc18ae5c24a683e5d606f54180228550014f34566492006f97c1
thread:issue:136092,issue #136092: User API For Donated Buffers,needs maintainer decision,28,needs maintainer decision,1,0,136092,,,,0,733fa8b3351c1dcea8e33dff7b42c5f9a68f27afb4476e0fc805d7140a313cc1
thread:issue:136122,issue #136122: [Distributed] Decouple ProcessGroupNCCL's regular exit from abort,needs maintainer decision,28,needs maintainer decision,1,0,136122,,,,0,4f07201ae106bf5e2bfc70c32238bddce5c480390da0d15e08032cad6926ff13
thread:issue:136149,issue #136149: Dynamo: warning when we incorrectly inline autograd.Function.backward due to subclass gradients,needs maintainer decision,28,needs maintainer decision,1,0,136149,,,,0,7ce57c946874738e5aff2a196460faf0d640174eceaf0813ed77b16fe4796e1a
thread:issue:136187,issue #136187: Unable to Specify CUDA Stream for Collective Operations Using with torch.cuda.stream() context,needs maintainer decision,28,needs maintainer decision,1,0,136187,,,,0,57b76fcf900f75b76fa303b44097779b32c93ea2cfd455039e65437a775b461a
thread:issue:136226,issue #136226: [pipelining] try not to dry run module when creating PipelineStage?,needs maintainer decision,28,needs maintainer decision,1,0,136226,,,,0,a6307187b66ac1830a596efed1a5056365032d962f7bafcef1fddb9a2172eab1
thread:issue:136248,issue #136248: Async NCCL communication blocks CUDA kernel in the first run,needs maintainer decision,28,needs maintainer decision,1,0,136248,,,,0,7c1ef96d3eafac2d766e648ba2bee9c0ac85b6a85327ebc43a7aac2d9a99e191
thread:issue:136272,issue #136272: AOTDispatcher debug mode,needs maintainer decision,28,needs maintainer decision,1,0,136272,,,,0,9c9b40c86d792149d954f67b9ed0d48ba4cf35d5faab81fdbbae31fe136fa1b7
thread:issue:136275,issue #136275: Please offer `<version>+cpu`-versioned packages for MacOS and linux ARM CPU-only builds,needs maintainer decision,28,needs maintainer decision,1,0,136275,,,,0,a3b4bb46cc60de9dc5505ea840359bd9290a68d736fdd6920be5b78c6daeefdc
thread:issue:136363,issue #136363: [pipelining] output shape validation / lazy inference,needs maintainer decision,28,needs maintainer decision,1,0,136363,,,,0,446127588672fa3370fb2f8059e126f36c77d16cb9dae9cb37612feee265402b
thread:issue:136380,issue #136380: [pipelining] Improve recv buffer management,needs maintainer decision,28,needs maintainer decision,1,0,136380,,,,0,a095edf9700e7c3e46828a8832b82d27caeb4d509420941fc4cb196eedaf8a93
thread:issue:136390,issue #136390: [c10d][nccl][cuda] Regression (unspecific cuda launch error) with test_c10d_nncl,needs maintainer decision,28,needs maintainer decision,1,0,136390,,,,0,f98a5132e25e72b47befa74b91af0b5b537c9a88b677389369455bb5adeac001
thread:issue:136408,issue #136408: [RFC] Composable parallel documentation,needs maintainer decision,28,needs maintainer decision,1,0,136408,,,,0,ce48110b1f42502fb882c23fa1b640c1cc2f9620a5b7dc41ed04885de3b3d038
thread:issue:136425,issue #136425: Any updates on AsyncCollectiveTensor support for all-gather along non-zero dims?,needs maintainer decision,28,needs maintainer decision,1,0,136425,,,,0,bb9394e779f5082532d9c9972d00da11a2c0ea9cd42aae66e78311f5c6d106d1
thread:issue:136495,issue #136495: [dynamo] enable TorchDispatchMode for eager part when graph breaks,needs maintainer decision,28,needs maintainer decision,1,0,136495,,,,0,98dada74c9875f9ea8e2daa51fec4dae68c1071a6adec99430ce54080edd3139
thread:issue:136532,issue #136532: Wrong results when sampling from the beta distribution for small alpha=beta,needs maintainer decision,28,needs maintainer decision,1,0,136532,,,,0,ee4ccf62d493fc10578380d9b85f2bc369fcfbd55760fc0040c9c9e4a7e5f702
thread:issue:136567,issue #136567: [RFC] Offload collectives to NVSwitch when possible,needs maintainer decision,28,needs maintainer decision,1,0,136567,,,,0,0c9167fac5397631d7504a9110947c246894400a3a11e692489b37b0b39fc154
thread:issue:136748,issue #136748: Inconsistent behavior of cdist with half-precision inputs,needs maintainer decision,28,needs maintainer decision,1,0,136748,,,,0,f8a89798252d487da2cbbefa4fc2fefb2b7550ee737a0071822c19626a9f6382
thread:issue:136866,issue #136866: fused_scaled_matmul_reduce_scatter report error with channel-wise scaling,needs maintainer decision,28,needs maintainer decision,1,0,136866,,,,0,51bd9af83224a3aeb9ebd2a8e3b8efd23e33c6cb62663e87e845be821286cad7
thread:issue:136950,issue #136950: Incomplete Parameter Gathering on Rank 0 with FSDP Model Saving,needs maintainer decision,28,needs maintainer decision,1,0,136950,,,,0,a9691f2bbe3f5233e02d4012f14b84edcaeb6bb70874b032a95da5cb1c781486
thread:issue:137010,issue #137010: [RFC] Investigate multi-thread support of c10d,needs maintainer decision,28,needs maintainer decision,1,0,137010,,,,0,b1e4b1f9ad459986d5a966f9d2639df6826fae1f37a79993132b389e392915e4
thread:issue:137093,issue #137093: Use SymInt instead of sympy.Expr uniformly in Inductor,needs maintainer decision,28,needs maintainer decision,1,0,137093,,,,0,b88ac8ae2df9aa6218ced0864b3edd82c36ae97b7d05d921a1c27bcda6756f20
thread:issue:137505,issue #137505: RuntimeError: Backend nccl does not support allgather_into_tensor_coalesced,needs maintainer decision,28,needs maintainer decision,1,0,137505,,,,0,7304f176336a0b6a1813a03649553fb56906a40d73b49dc98facfa062dff5d98
thread:issue:137560,issue #137560: vmap specializing on dynamic shapes with index_put_,needs maintainer decision,28,needs maintainer decision,1,0,137560,,,,0,148804f86ad8299e7f6d72274bd87a19d3f3b5ef52c00498cf81c4987fa91085
thread:issue:137749,issue #137749: Graph break on wait() of Torchrec's Awaitable instance,needs maintainer decision,28,needs maintainer decision,1,0,137749,,,,0,d003df66822264513bb7250cdd511d5aed6864f31539850d4a86161477186e8e
thread:issue:138355,issue #138355: tlparse's fx_graph_runnable should include custom op info,needs maintainer decision,28,needs maintainer decision,1,0,138355,,,,0,49e7e87eccc785f47d4cc2be2bfd25f413f5a515296e7abfddbe5d9e4462c297
thread:issue:138464,issue #138464: `torch.special.zeta` ignores `nan` input when `other=-inf`.,needs maintainer decision,28,needs maintainer decision,1,0,138464,,,,0,88a191bd128fd632fea76eac8e7738d4d20b43c76a3089c9763c926a637f18e1
thread:issue:138980,issue #138980: compile_fx inplace modifly the input graph,needs maintainer decision,28,needs maintainer decision,1,0,138980,,,,0,6da098277ad3fff1d933ddcc411f6dbb552ad01700523c13b3ab5956c6ba6aa3
thread:issue:139129,issue #139129: [pgnccl] unstable restarting of PGs,needs maintainer decision,28,needs maintainer decision,1,0,139129,,,,0,b1b3a0c49f77e7314aee4e30dfbd83151cf601baf856809043a83e8ac605889c
thread:issue:139152,issue #139152: Dynamo capture of tensor.data assignment doesn't identical to eager call of tensor.data assignment,needs maintainer decision,28,needs maintainer decision,1,0,139152,,,,0,c3a94d9e577a1cf7b4c10ea491cfb7b189f434b65b740b29b1378ecb657399ae
thread:issue:139257,issue #139257: Error compiling the torch.library.custom_op with input mutations with set_,needs maintainer decision,28,needs maintainer decision,1,0,139257,,,,0,0c3a13b48e7ca7a22d39f9cc306f650f62ab2c61a857c76084fcb419902af608
thread:issue:139280,issue #139280: Possible bug of tools::flight_recorder,needs maintainer decision,28,needs maintainer decision,1,0,139280,,,,0,d3f650bd8a45e5d58c83b565f2f2921de58ec97b17fbc3a2c7b9e435f1505a49
thread:issue:139313,issue #139313: BFloat16 support for `torch.fft.rfft`,needs maintainer decision,28,needs maintainer decision,1,0,139313,,,,0,f257f483212af8f4bc0751d55983ab507620a9be583571e7ccec18857d13a2e7
thread:issue:139581,issue #139581: Suggestion : enable one to adjust the interval between calls to the pytorch NCCL watchdog thread via env variable,needs maintainer decision,28,needs maintainer decision,1,0,139581,,,,0,d190d6a4a5d0962fdd94cd47d08fe26bdd88133632bd9c07c70116ae4e633a2e
thread:issue:139681,issue #139681: Unexpected type cast from DTensor to Tensor when applying SequenceParallel on RMSNorm.,needs maintainer decision,28,needs maintainer decision,1,0,139681,,,,0,d4ecde584d0ea661a77dd5ed8502cf834bfa775257b2c4d6f600543eeb33c34c
thread:issue:139739,issue #139739: [Distributed][collectiveCoalesced] Why function(check_gpu_tensors_same_device) only works in allreduce_coalesced?,needs maintainer decision,28,needs maintainer decision,1,0,139739,,,,0,8b2139ff42f47e1e71bd0c6eb5098788d770b585b3734cc8910beb961a9641d3
thread:issue:139964,issue #139964: Report issue for torch.nn.Linear when forwarding a 3-dim tensor.,needs maintainer decision,28,needs maintainer decision,1,0,139964,,,,0,f982ce949bf11962f79b4e8c79696c0ad8e5b4c9ed9bd1ea2f9c7cdcdceb2e4c
thread:issue:140107,issue #140107: Support Joinable loss functions with DDP,needs maintainer decision,28,needs maintainer decision,1,0,140107,,,,0,8f9c50a2bf7b688fb20db53a60532ccb0ac4f79fc29b473e2648c5cbb1c676d7
thread:issue:140227,"issue #140227: `Torch.distributed.tensor.parallel.Style.Rowwiseparallel` when send tensor, the use of the underlying operator is `aten.to.dtype_layout`, lack of distributed strategy. But after adding a policy, `AsyncCollectiveTensor` no placements error",needs maintainer decision,28,needs maintainer decision,1,0,140227,,,,0,70173c9af6469ad1bdb0d9dfc0895d6753931123352182f3a10bdabf58501dc8
thread:issue:140458,issue #140458: Making nn.Module generic,needs maintainer decision,28,needs maintainer decision,1,0,140458,,,,0,34376af93738a4b3170755bf27db0960b28a5ebe25b07370c011074f90cfda6c
thread:issue:140463,issue #140463: Ban all operations on sympy expressions in Inductor,needs maintainer decision,28,needs maintainer decision,1,0,140463,,,,0,7f9818a2f04931536ed9de819134182771f8d070bcc6116aed8a35226f1cb7fd
thread:issue:140970,issue #140970: `torch._inductor.cpu_vec_isa.pick_vec_isa` takes ~9 seconds to run,needs maintainer decision,28,needs maintainer decision,1,0,140970,,,,0,7f93a95a758c47ca35a31c7d7241607d40708caebc2630005f8d040266a2040d
thread:issue:141237,issue #141237: test_fsdp_tp_integration fails for non-power-of 2 GPUs,needs maintainer decision,28,needs maintainer decision,1,0,141237,,,,0,05133a116d64c83103aefc7723eff2b155ac45eb3e946ce553178cb7042d45e5
thread:issue:141259,issue #141259: Consider relaxing torch.cond's requirement that the outputs of each branch have matching requires_grad metadata,needs maintainer decision,28,needs maintainer decision,1,0,141259,,,,0,283b6a65c2007b23cf33cde76b438f64f2186ec8094f5bc3eac586895fb57237
thread:issue:141288,issue #141288: Enhancement Request: improve this error message,needs maintainer decision,28,needs maintainer decision,1,0,141288,,,,0,8bf105d16d8938602b368beaa89879656f430a564f0bb74286f5fd9402bcf2e2
thread:issue:141349,issue #141349: Burn down test/functorch/test_aotdispatch.py OpInfo test failures,needs maintainer decision,28,needs maintainer decision,1,0,141349,,,,0,e9c68d61e13cafa02a627e8c5ab7d2bd3ebcf3439d7835da3ef44caa53583b85
thread:issue:14152,issue #14152: Failed to run 'bash ../tools/build_pytorch_libs.sh --use-cuda --use-nnpack --use-mkldnn --use-qnnpack caffe2',needs maintainer decision,28,needs maintainer decision,1,0,14152,,,,0,db9bd04dcef630bb1241aab27a9194846fe36f81b54d4198364748f14eb5767d
thread:issue:141784,issue #141784: [`FlexAttention`] Request for the Support of Dropout,needs maintainer decision,28,needs maintainer decision,1,0,141784,,,,0,8b326982a7719a677aa51d2adeab89f4bf4df4293ae95ba181eaa4b09af11873
thread:issue:141827,issue #141827: Compiled and masked MHA running with DDP and `no_grad()` leads to Graph break due to unsupported builtin torch._C._distributed_c10d.PyCapsule._broadcast_coalesced,needs maintainer decision,28,needs maintainer decision,1,0,141827,,,,0,fc88fbdddba36384a6e904873a82941d60785278984945d4fadab8ed4072ac28
thread:issue:141958,issue #141958: Support SDPA flash attention/ memory efficient attn on ROCm gfx908,needs maintainer decision,28,needs maintainer decision,1,0,141958,,,,0,0f758f2cf3c5404f1f8ce06819f4fb9eb97bf138a23c3cb6bd35b3d5414a9f2e
thread:issue:142095,issue #142095: torch.distributed.new_group got stuck when use backend=mpi,needs maintainer decision,28,needs maintainer decision,1,0,142095,,,,0,a9fcda57f5cf558c9b10afe9ed02635103c32cd933896040a52fa5362d2b46d2
thread:issue:142135,issue #142135: Device assert throws a runtime error in cuda backend and results in a crash in xpu backend,needs maintainer decision,28,needs maintainer decision,1,0,142135,,,,0,e99853ace7bfb205f1892b388e87d866b6d707b0de2c40ec714396ba917de1e6
thread:issue:142356,issue #142356: AllReduce hangs due to no device_id into init_process_group,needs maintainer decision,28,needs maintainer decision,1,0,142356,,,,0,896ed04a1617fafd70cad073daa2323fbd7af4390767dffc4ecf7ba2af7593d5
thread:issue:142388,issue #142388: User-defined way to hook into Inductor autotuning (+ custom op),needs maintainer decision,28,needs maintainer decision,1,0,142388,,,,0,f4ded1595be1a36c4ab47612965cd36835cda97fe506be079ccb20bc84113fa6
thread:issue:143239,"issue #143239: xpu: torch.nn.DataParallel fails on multi-XPU environment with ""module 'torch._C' has no attribute '_scatter'""",needs maintainer decision,28,needs maintainer decision,1,0,143239,,,,0,46ded1faa5f44944a21f59337964da09216e8a5f1e51219d1d6b0341ed1a4244
thread:issue:143281,issue #143281: torch.linalg.qr is significantly slower on GPU compared to CPU and SVD for batched small matrices,needs maintainer decision,28,needs maintainer decision,1,0,143281,,,,0,3debe432f9288579725d06330deb272c5a8239cc54ecd7c91f4a0f1d3faee7ac
thread:issue:143337,issue #143337: An idea to improve Welford implementation in Inductor,needs maintainer decision,28,needs maintainer decision,1,0,143337,,,,0,801cc3e4ae464dde9db56bb660f705344e3fdaf7763aaa96540f8dc4a02250d6
thread:issue:143471,issue #143471: NFS errors during DataLoader shutdown when num_workers > 1 when temporary directory is on NFS,needs maintainer decision,28,needs maintainer decision,1,0,143471,,,,0,60210c3ccd790fc41d8159576e06ad9a6a497bdb1e9c94b82384e65ffd8f2f8c
thread:issue:143647,issue #143647: A very weird bug involving ddp,needs maintainer decision,28,needs maintainer decision,1,0,143647,,,,0,d306da9f3a580668b6a7fd9ac2888d270cc090642bcdce1300430389538c9cad
thread:issue:143721,"issue #143721: add ""enabled=True"" to DistributedDataParallel.no_sync()",needs maintainer decision,28,needs maintainer decision,1,0,143721,,,,0,1674950a5bcfed10e4b0efd8f11f15c107e278ea57e160a82080e4368fed91d3
thread:issue:143748,issue #143748: [DTensor]`Linear` fails on 3D DTensor with `batch size > 1` and `Replicate` input redistributed from `Shard`,needs maintainer decision,28,needs maintainer decision,1,0,143748,,,,0,74584eafa1b5f41ff67e20687a240a48483a13e3336f5024b1083bd2dcda120f
thread:issue:143823,issue #143823: Possible race condition found in TailLogTest.test_tail,needs maintainer decision,28,needs maintainer decision,1,0,143823,,,,0,ef260ff7835ae6234120b04895e2cdde4b43fd6a74aeec7ebb1f7a6982da0af5
thread:issue:143887,issue #143887: `torch.accelerator` cross-device utilities and properties,needs maintainer decision,28,needs maintainer decision,1,0,143887,,,,0,894e54cc05042d38aa3dfeeb85daed8e0e6e97fc0110eab674b66754429d2bbe
thread:issue:143954,"issue #143954: No ONNX function found for <OpOverload(op='quantized_decomposed.dequantize_per_channel', overload='default')>",needs maintainer decision,28,needs maintainer decision,1,0,143954,,,,0,972a92089f77f9ace26d60505487fadb150ffc43d23827076528435a47a1187e
thread:issue:143985,issue #143985: [DTensor] Allow multiple dimensions to be sharded together (as if flattened),needs maintainer decision,28,needs maintainer decision,1,0,143985,,,,0,83a2869770c4525cf3972a4b2dfc7708165bba24c0809f1e59ea53529a1454c0
thread:issue:144239,issue #144239: Integrate ULFM and FTI for OpenMPI (torch.distributed),needs maintainer decision,28,needs maintainer decision,1,0,144239,,,,0,545ec09d56d59ecfcf61d0f5447791ba7e55297278624e6331a19134284e7899
thread:issue:144459,"issue #144459: In AOTI Runtime, acquiring a write lock on model_exec_mutex_ may cause starvation.",needs maintainer decision,28,needs maintainer decision,1,0,144459,,,,0,13466ae7865dd6bfbb7fba030978f6827a62decb701ad904b19326d9918860e7
thread:issue:144608,issue #144608: Build breaks on FreeBSD on arm platforms: Unrecognized CMAKE_SYSTEM_NAME = FreeBSD,needs maintainer decision,28,needs maintainer decision,1,0,144608,,,,0,d6f29e52a5887d56e202b594cd1a38cc72c944a5619f54d0ecc590eb1de44521
thread:issue:144634,issue #144634: [MPS] `torch.mps.synchronize` hangs on error,needs maintainer decision,28,needs maintainer decision,1,0,144634,,,,0,e9e7266169cbb58a73d2df1bb029d51bb47ebbc48b55def6072e34998f585d38
thread:issue:144779,issue #144779: Adding Infiniband to RDZV Backend for optimal torch run training,needs maintainer decision,28,needs maintainer decision,1,0,144779,,,,0,bf11c1437f216ee71c4b0f909deb8cd1024d17eaacd009ec820e516712beb4d0
thread:issue:144805,"issue #144805: ""GenericHOPVariable"" / abstract out Dynamo support for HOPs",needs maintainer decision,28,needs maintainer decision,1,0,144805,,,,0,fdf2c9d2f576fb2e038a859aeffbe09872b9b2901be3deea584d46438f6c35e1
thread:issue:144851,issue #144851: torch.distributed hangs between Linux (X86) and Mac (M2 Pro),needs maintainer decision,28,needs maintainer decision,1,0,144851,,,,0,dc18d5dc7d736b41e184686212716340e61a07d71622dcb829f16655b1db2d91
thread:issue:144856,issue #144856: Connection Limitation in PyTorch Distributed (Vanilla) with c10d Rendezvous Backend,needs maintainer decision,28,needs maintainer decision,1,0,144856,,,,0,9e6ec79adeeb428887b013597708d2ef61a084d6f046f7776292528dd898813e
thread:issue:145170,issue #145170: CUDA initialization error with vLLM 0.5.4 and PyTorch 2.4.0+cu121,needs maintainer decision,28,needs maintainer decision,1,0,145170,,,,0,4eec0855847a57e582407adfb482b1f20318c3967691202d51c03bf32e905de4
thread:issue:145189,issue #145189: DISABLED test_slice_scatter_reinplace_cuda (__main__.GPUTests),needs maintainer decision,28,needs maintainer decision,1,0,145189,,,,0,0644169ee05b32396bdbcfdaaf167cf274c99d5c8d27fd1eb98454c162199298
thread:issue:145218,issue #145218: getting different results when adding `torch.Tensor` or python number to a DTensor - Is that expected?,needs maintainer decision,28,needs maintainer decision,1,0,145218,,,,0,58a935497dd55b4793c147ee71616bbd78ff52fd734c51fdb412462332898f57
thread:issue:145344,issue #145344: DISABLED test_cat (__main__.NumBytesMetricTests),needs maintainer decision,28,needs maintainer decision,1,0,145344,,,,0,0dd2d6c0ab8e7fbaed2bd62ab95b53b17b6851c06e0bc3ae98713262dde5207a
thread:issue:145376,issue #145376: distributed.new_group with backend GLOO hangs when distributed.split_group was called before,needs maintainer decision,28,needs maintainer decision,1,0,145376,,,,0,1f721bff33e232966172fe3b9650155c2723bc961e6aa953b602a63b067e4e57
thread:issue:145401,issue #145401: torch.compile has different numerics for var_mean,needs maintainer decision,28,needs maintainer decision,1,0,145401,,,,0,aab0f23df85f2512c8557bd1bb9d802c9b6a3b50d346a63406b725cb33589447
thread:issue:145522,issue #145522: use_const_ref_for_mutable_tensors doesn't work with out= overloads,needs maintainer decision,28,needs maintainer decision,1,0,145522,,,,0,9187c55f8704ad06b904ebe1190a603a7a79ea5f51554e46dbf2ee8c5e9c610d
thread:issue:145679,issue #145679: Gamma cdf function,needs maintainer decision,28,needs maintainer decision,1,0,145679,,,,0,693423d3f0cff9ba5bf84fba8621a168486afc859da7b1807db2b765a04592d9
thread:issue:145726,issue #145726: Add correct `__repr__` for parallel distributed modules,needs maintainer decision,28,needs maintainer decision,1,0,145726,,,,0,5a7e9c16690eaddb3d7559eacdc0aa22a55b1b8f23017aa889b7928293baa685
thread:issue:146045,issue #146045: Redundant collectives are deduplicated but stick around as dead code,needs maintainer decision,28,needs maintainer decision,1,0,146045,,,,0,26160d6430f7823785ab50f966a17790f826d6491aba7cdeb3f7d7659317ec8f
thread:issue:146094,issue #146094: [CI][Distributed] @skip_if_lt_x_gpu(2) seems to be broken,needs maintainer decision,28,needs maintainer decision,1,0,146094,,,,0,b524e39e01ce9098e61ec45dd6a33cef3917511c6381ec4682e24aaccc4c74e9
thread:issue:146129,issue #146129: torch.compile on Mamba2 model produces NaNs,needs maintainer decision,28,needs maintainer decision,1,0,146129,,,,0,e8dd4560b72d05bffed7d64a38999a16d6c2f5adc979e43eb5d76fba935c35d1
thread:issue:146565,issue #146565: use register_full_backward_hook with c10d in torch.compile and raise error,needs maintainer decision,28,needs maintainer decision,1,0,146565,,,,0,c561e78195e3a01acff9123d30f9e2b073295f5a3dafcd95f998e3df49f2eb38
thread:issue:146601,issue #146601: Enabling efficient model-level redistribution between FSPD-TP,needs maintainer decision,28,needs maintainer decision,1,0,146601,,,,0,54bcc781c06bcd813b01ff5b22c6aa98305760ea97fb052b8010eadddaf56a8f
thread:issue:146691,issue #146691: MPS support `torch.linalg.norm` on complex numbers,needs maintainer decision,28,needs maintainer decision,1,0,146691,,,,0,199aeb2cb89ee566e1c6322230fe7fc09184fbd1cd066ea6defcb875e4db9267
thread:issue:146780,issue #146780: TypeError when using torch.compile with RegionViT under torch.inference_mode(),needs maintainer decision,28,needs maintainer decision,1,0,146780,,,,0,09b44d2919fc623e870ee8fdcd4512f04c7ef4a4d694b0e60f1d6f1109752918
thread:issue:146813,issue #146813: Oneshot AllReduce not being triggered when there's nested intra- and inter-node process groups,needs maintainer decision,28,needs maintainer decision,1,0,146813,,,,0,b6aeb7cd93764fa32cb22f4196fe0a2d27a23a4585a5b3f8444b764e4b57847c
thread:issue:146860,issue #146860: [BE] Put all the HOP tests in one location,needs maintainer decision,28,needs maintainer decision,1,0,146860,,,,0,0a0f9055218b6955d9d8cc82607110a96c011cb1b324c101b0ad49bc81ed7011
thread:issue:146861,issue #146861: [BE] the majority of HOP OpInfo tests should be in one place,needs maintainer decision,28,needs maintainer decision,1,0,146861,,,,0,caebfd60b15f16d345700d360a4043af1ca7ceb07231dd056beb93f0590e1352
thread:issue:146884,issue #146884: [DTensor]Operator aten.mT.default/aten.mv.default/aten.dot.default does not have a sharding strategy registered,needs maintainer decision,28,needs maintainer decision,1,0,146884,,,,0,da993da1d64c58596da577a6ce40a9f0dfb7df66a5354364405d725b1e506529
thread:issue:146936,issue #146936: flexible custom operators: custom operators that accept arbitrary input/output type,needs maintainer decision,28,needs maintainer decision,1,0,146936,,,,0,43507a8d52f1c83eeb46d34e4109216506204d3992038ed408861233c21febe7
thread:issue:147210,issue #147210: Consider relaxing the in-place mutation restriction for torch.cond and torch.while_loop,needs maintainer decision,28,needs maintainer decision,1,0,147210,,,,0,5ddf824125de74015facf283c1243b7113a9a7fbd6eb5b6ea5d4152117cfc989
thread:issue:147478,issue #147478: Blackwell Only Bugs,needs maintainer decision,28,needs maintainer decision,1,0,147478,,,,0,2c26f38801f802e5cd29a0816ece2ec4546f550934c79a1151b9103f09b72757
thread:issue:147971,issue #147971: Fp8 scaled-mm row-wise is substantially slower than tensor-wise,needs maintainer decision,28,needs maintainer decision,1,0,147971,,,,0,250a3be030be75866266fb0e6c0356076c11da929b79f6ce8af09ce913562441
thread:issue:148420,issue #148420: Float8_e4m3fn,needs maintainer decision,28,needs maintainer decision,1,0,148420,,,,0,70ec27e46b68a0fc3453663d00b3146234c1203c84623439ebe753bb7f893a60
thread:issue:149080,issue #149080: DISABLED test_aoti_debug_printer_codegen_cuda (__main__.AOTInductorTestABICompatibleGpu),needs maintainer decision,28,needs maintainer decision,1,0,149080,,,,0,f67e7bf6e5f1070048e314022b438320bf9b5688a780525948aca90bcc35c106
thread:issue:149605,issue #149605: torch.export does not support nested tensor operations,needs maintainer decision,28,needs maintainer decision,1,0,149605,,,,0,8224f59215af86fce0ca465cb2e1765cb667d7030618238f58dfa65c7d310508
thread:issue:149639,issue #149639: Pylint error: ` torch.linalg.vector_norm is not callable`,needs maintainer decision,28,needs maintainer decision,1,0,149639,,,,0,97a25126a896093d2e237407c069b0122225080a13387c2277d41e68d3f18ff7
thread:issue:149809,issue #149809: LoadHIP.cmake should find_package(composable_kernel),needs maintainer decision,28,needs maintainer decision,1,0,149809,,,,0,af6d907db3ee5e42510568078a01964c1e943222b110f2074341812dbdaaf26d
thread:issue:150062,issue #150062: Support HOPs in fx_graph_runnable,needs maintainer decision,28,needs maintainer decision,1,0,150062,,,,0,88358bfbbdaa59dd126f968139baf7ed9c619237c5deadd34a713d4bfa324e19
thread:issue:150121,issue #150121: torch.compile on MPS progress tracker,needs maintainer decision,28,needs maintainer decision,1,0,150121,,,,0,115d733ae1616f093ef2a5db5d0da01f19096276abdd2f7785dd79e129b243b7
thread:issue:150168,issue #150168: [ROCm] PyTorch slow on TTS,needs maintainer decision,28,needs maintainer decision,1,0,150168,,,,0,e66ee8118c394a00cbeb89e5e323bf678c46cef5cc08ba04601a22ad93af0d67
thread:issue:150296,issue #150296: [RFC] zentorch Integration,needs maintainer decision,28,needs maintainer decision,1,0,150296,,,,0,96c28abcc6b63c3b3b9dee3ecdd7d4cf041ef53c40e697c428f491d6ef768c40
thread:issue:150655,issue #150655: Create better alerting for binary size validations and time it takes to build the binary,needs maintainer decision,28,needs maintainer decision,1,0,150655,,,,0,09636120d91091a651ed21294765ff9bc3bdb9c34e01e0db689919497cd5d3ad
thread:issue:150725,issue #150725: Continuous calls to nn.Linear in fp32 on the 5090D cause severe performance degradation,needs maintainer decision,28,needs maintainer decision,1,0,150725,,,,0,fbfe2191992c9b553514c2350dae8341b9dd41f70437be107bf38dd4a3add50f
thread:issue:150877,issue #150877: DISABLED test_multiple_module (__main__.TestInvokeSubgraphExportStrict),needs maintainer decision,28,needs maintainer decision,1,0,150877,,,,0,ac2a2fb95c3bc4f3ed16660e0860a422ac84efb7ccdcdc25da4fb0aa72854156
thread:issue:150974,issue #150974: feature: tlparse summary,needs maintainer decision,28,needs maintainer decision,1,0,150974,,,,0,bdc3143a9795a01ac5d9825f42ba9d99ad103f409a03bc0a60a2477de7aba3ca
thread:issue:151204,issue #151204: [Dynamo] Dynamo fails to trace reduce_scatter_v,needs maintainer decision,28,needs maintainer decision,1,0,151204,,,,0,b3c9f50943b541d7b3eef37b0da8d18672dfe244d2c396eac5fae4893c641d53
thread:issue:151761,issue #151761: Inconsistent `sum`/`dot`/`norm` behavior,needs maintainer decision,28,needs maintainer decision,1,0,151761,,,,0,c36c0aa1c87b9f6d77d6d78e8b654cf68500a9ca9d1f409f05cba2b3a40790b0
thread:issue:151782,issue #151782: JVP: Option to Disable Gradient Caching for Tangents,needs maintainer decision,28,needs maintainer decision,1,0,151782,,,,0,153db6e1bff28e5fb7d1e2cfa3ae353789847c6b12acd831dfede0c20a76c6c9
thread:issue:151997,issue #151997: torch.compile doesnot support index with tensor,needs maintainer decision,28,needs maintainer decision,1,0,151997,,,,0,a11f28164f02915773ca0a97178deefef6c86a1ee1473123a4e54e0e41644f81
thread:issue:152028,issue #152028: [CUDA] MultiheadAttention with masks and dropout produces NaNs,needs maintainer decision,28,needs maintainer decision,1,0,152028,,,,0,e8a8681da8b856422392e0137fcf22802e40a0455274f62c53eb805ea5532cf9
thread:issue:152042,issue #152042: distrubuted: false positive Grad strides vs Bucket strides warning,needs maintainer decision,28,needs maintainer decision,1,0,152042,,,,0,8c85cfa97410d756a54a27590d6a1a7dec371234ad20587d79cf158e792b4713
thread:issue:152213,issue #152213: Outdated install commands,needs maintainer decision,28,needs maintainer decision,1,0,152213,,,,0,b0c7da8ca36cf8fdc287a613dc3b72ec31b5cf530ccef114d20cbc5858faab02
thread:issue:152270,issue #152270: Question about that support of torch.compile for a custom CUDA operator?,needs maintainer decision,28,needs maintainer decision,1,0,152270,,,,0,05f2ac691c4941543b0102c7815caccbab2363700ece34b196132c989c7a89e6
thread:issue:152355,issue #152355: [source.wheel] Pin setuptools runtime dependency,needs maintainer decision,28,needs maintainer decision,1,0,152355,,,,0,175031b3f0e0cf9c6a32d5e17359a65886e3b7f211af1df05d9dafcd6928095f
thread:issue:152427,issue #152427: Add switch to disable truncation to long list print,needs maintainer decision,28,needs maintainer decision,1,0,152427,,,,0,f55b84e04743249187fd343e5ded5f6af16146710d80a6593ce1627fd1c6e002
thread:issue:152442,issue #152442: `torch.compile` causes assertion error in distributed checkpoint wrapper test,needs maintainer decision,28,needs maintainer decision,1,0,152442,,,,0,ed77d1b4ec3a552eb34873d6627fbb667ea8a15df50464761c0f12b535c3a8f8
thread:issue:152632,issue #152632: DISABLED test_torchvision_models_efficientnet_v2_l (__main__.TestVisionTracing),needs maintainer decision,28,needs maintainer decision,1,0,152632,,,,0,44649e8af17ec7cf552f8808d89faefc1b729b9a7a8426c9968838710d6c435b
thread:issue:152710,"issue #152710: [FSDP2] NO_SHARD as fully_shard(mesh=(Replicate, Shard)) with shard of world size 1",needs maintainer decision,28,needs maintainer decision,1,0,152710,,,,0,b04d0f07009e36189a077320329645e8b1021b81c93b455ff5dd86d208aa27fe
thread:issue:152944,issue #152944: DISABLED test_compiler_collectives_automatic_dynamic_tensor (__main__.TestMultiProc),needs maintainer decision,28,needs maintainer decision,1,0,152944,,,,0,a2827a9844da3f90393b0ec40d6059fa77a60e24ddd37dd30dfaadca6e5ac58e
thread:issue:153016,issue #153016: inconsistent grads between two types of `allgather`s,needs maintainer decision,28,needs maintainer decision,1,0,153016,,,,0,bacd5bcb535c809c5ee8665e936fa1194bb9afe107362cdec5eab92628c0b1fc
thread:issue:153122,issue #153122: [CUDA] test_c10d_nccl test_extra_cuda_context failure due to _helper_test_extra_cuda_context_by_memory,needs maintainer decision,28,needs maintainer decision,1,0,153122,,,,0,212c311599feec26238c2fbef8b982fcfeefd006fb9c6b4955593b25c2911935
thread:issue:153221,issue #153221: FlexAttention: Support for post_mod,needs maintainer decision,28,needs maintainer decision,1,0,153221,,,,0,8ed32ec91f8da0f34c621fd783d2d08997dc6241eab5ca84754c02fa4c15c351
thread:issue:153236,issue #153236: DISABLED test_clip_grad_norm_2d (__main__.TestClipGradNormWorldSize4),needs maintainer decision,28,needs maintainer decision,1,0,153236,,,,0,1789b3550023d54108159a31560a02b8dcbe244ac877275e4579577db5a5797a
thread:issue:153350,issue #153350: `distributed.checkpoint.state_dict.set_model_state_dict` returning incorrect `incompatible_keys` when `full_state_dict=True`,needs maintainer decision,28,needs maintainer decision,1,0,153350,,,,0,360beb914ea6f0c8bef1938095e09e88400c380a117e45c7adb169183b73da1b
thread:issue:153369,issue #153369: Rank {rank} has different values for {name}: {scalar_tensor_value}.,needs maintainer decision,28,needs maintainer decision,1,0,153369,,,,0,ffa9e9c61afb85b552cb0d75702c60cda3715fa66f58689e12390e35b89f9fb4
thread:issue:153460,issue #153460: DISABLED test_mempool_ctx_multithread (__main__.TestMemPool),needs maintainer decision,28,needs maintainer decision,1,0,153460,,,,0,ae9b7df0b2141121aafb3e84950d2c3bf312e7eb9a7dab497cd073fee6e3f812
thread:issue:153546,issue #153546: Make implicit packages (PEP420) explicit PyTorch,needs maintainer decision,28,needs maintainer decision,1,0,153546,,,,0,35314f520f0cdbc3dfa7a955cdf6cef56ea8de1c34f60b96fd09583d01505e46
thread:issue:153555,issue #153555: Updated Scaled_mm to support more scaling formats via CuBlas,needs maintainer decision,28,needs maintainer decision,1,0,153555,,,,0,5f021c08420f9de1624c0857e471539d9437ce4f9d0b79fad59048566059883e
thread:issue:153615,issue #153615: [DTensor] Backward Pass Failure with 2D Sharded DTensor After Mean Reduction,needs maintainer decision,28,needs maintainer decision,1,0,153615,,,,0,caeda273d5bf60ebfb8a98eaf428a4ef3b50630ba505b6ec7f72e068b3b2d36d
thread:issue:153809,issue #153809: Optimize rowwise scaled mm CUTLASS kernel for large values of K,needs maintainer decision,28,needs maintainer decision,1,0,153809,,,,0,695247cd9cb863694462ada5e48ccb39d49fb0046ebca2932964d12424a4fc60
thread:issue:153822,issue #153822: Passing string to devices in register_backend causes incorrect behavior,needs maintainer decision,28,needs maintainer decision,1,0,153822,,,,0,d799be47e49f6f4951cccafb4540e49c9fe67bb538c6a3547765f3699463ecb0
thread:issue:154077,"issue #154077: accept 0-d tensor for `fill_value` parameter of `full`, `full_like`",needs maintainer decision,28,needs maintainer decision,1,0,154077,,,,0,6c0a6b4859415ffea9969625c1d403b00400bbdc1c1b8cfd08408142ecfb98c3
thread:issue:154102,issue #154102: dist.init_process_group not work,needs maintainer decision,28,needs maintainer decision,1,0,154102,,,,0,82fd659ea1fc02541b2ed570444704188ffb217fbc30f245c1ddcb0e5254f02c
thread:issue:154116,issue #154116: Timeout in distributed training,needs maintainer decision,28,needs maintainer decision,1,0,154116,,,,0,d6f8bcd6c71491d88657c814d391e85f074ed0eabe2c17c03e1c71fa0641b6d5
thread:issue:154299,issue #154299: ImportError: libcudnn.so.9: cannot open shared object file: No such file or directory,needs maintainer decision,28,needs maintainer decision,1,0,154299,,,,0,580b11cc03d3d7466f5413fbc0640a639d865c41cc04871cfa3c35ed9431ad41
thread:issue:154327,issue #154327: `torch.distributed.checkpoint.state_dict.get_model_state_dict` does not update the state_dict._metadata key,needs maintainer decision,28,needs maintainer decision,1,0,154327,,,,0,c0c9cf49fdf16c7e361137d49dc7f3bcbcbd721a415e5c9b711f3a0fa90471bb
thread:issue:154345,issue #154345: [RFC] : Remove Explicit Backend References from `torch.distributed` (`c10d`),needs maintainer decision,28,needs maintainer decision,1,0,154345,,,,0,a585613b3513aa789721e97a7d4799d92b8910140d329171361041fcec146b7d
thread:issue:154363,issue #154363: `scaled_dot_product_attention` broadcasting (GQA) is a memory footgun,needs maintainer decision,28,needs maintainer decision,1,0,154363,,,,0,e5357076fb0eafa11c43db8bcb2ac726abb94d872d3f15b6d73f64c6bb50609b
thread:issue:154370,issue #154370: [C10D] Blackwell B200: NotImplementedError: Could not run '_c10d_functional_autograd::all_to_all_single' with arguments from the 'CUDA' backend.,needs maintainer decision,28,needs maintainer decision,1,0,154370,,,,0,4d8d7e994923857a8cdf9b5258fd23518ee5b75228647d8ce6b5f25f58e024c6
thread:issue:154444,issue #154444: [RFC] Removing ideep git submodule dependency from PyTorch & add oneDNN git submodule,needs maintainer decision,28,needs maintainer decision,1,0,154444,,,,0,3ffd958e354e736c8ad635c06beaac59c4f9c17ce7e61f12f022739c4f8b7205
thread:issue:154449,issue #154449: 16KB pagination support for PyTorch Torch Vision and PyTorch Lite,needs maintainer decision,28,needs maintainer decision,1,0,154449,,,,0,35cf04833abc62a7a24837c8cb9a3d877a5147b06a002237874b893a89135127
thread:issue:154563,issue #154563: Distributed Breakpoint doesn't exit safely,needs maintainer decision,28,needs maintainer decision,1,0,154563,,,,0,5aef450ae0b78795dea955ba8d89f3c2dee0e56785789958e4a5d75941a3fdc2
thread:issue:154621,issue #154621: functional all_gather does not work with scalar tensor and fail with torch.compile,needs maintainer decision,28,needs maintainer decision,1,0,154621,,,,0,8c08670195b825ea446c2bdeb77573a76b05bd7da36487b11d9204975462503f
thread:issue:154657,issue #154657: [FSDP2] offer public API to share communication context aross fsdp roots,needs maintainer decision,28,needs maintainer decision,1,0,154657,,,,0,5592b8136b8d5e5521da4724d02065f40c472aabe1be0a8e71e2e3b54e7e1f2e
thread:issue:154787,issue #154787: Redistribute DTensor across different DeviceMeshes,needs maintainer decision,28,needs maintainer decision,1,0,154787,,,,0,bdbebfae4240e8a75565fb29c5ee56fcce01f461ed0a8047a92818cdecdcd3a2
thread:issue:155240,issue #155240: Dynamo Fails on torch_scatter.scatter_max with Fake Tensor Allocation Error During Graph Tracing,needs maintainer decision,28,needs maintainer decision,1,0,155240,,,,0,7e38a5acfb9a8cdffe8ae138750facf1380fabb0547584faf6afedb741d4efde
thread:issue:155467,issue #155467: torch.distributed TCP init method binds socket to all interfaces in all cases,needs maintainer decision,28,needs maintainer decision,1,0,155467,,,,0,2ec06bc9afb3228ebde820075db9641c2b78373b70565b2d2add5e27ca803abf
thread:issue:155512,issue #155512: Unable to find caffe2 when building libtorch from the source and then trying to use it in a cmake project,needs maintainer decision,28,needs maintainer decision,1,0,155512,,,,0,15aef82758439554c8abdd6bf075f382e712aceae564fa818c34d71018dace51
thread:issue:155547,"issue #155547: [Misc] The keys of the sac_milp return dict were modified, but the test case was not updated",needs maintainer decision,28,needs maintainer decision,1,0,155547,,,,0,8dca391fcbfe05ef1231e88683cd96adfb6309f8e80bcc5ce42ac6c648bbc580
thread:issue:155562,issue #155562: [Misc] test_foreach_add_different_mesh cannot work on machines with less than 4 GPUs,needs maintainer decision,28,needs maintainer decision,1,0,155562,,,,0,ee8784b7cd195e4b21aadcaeacaf7f10965c01a7c440502c1d53c4c5aaa96064
thread:issue:155589,"issue #155589: pipeline() fails when a sub-module uses ""no_grad()""; impacts RoPE implementation on HF models",needs maintainer decision,28,needs maintainer decision,1,0,155589,,,,0,7f6a845a3e1cdcf2a82e3596e18b8f6aed7d58adfb7ba6c9a0c95f350249e4f0
thread:issue:155632,"issue #155632: Functional all_gather_into_tensor does not support stacking, fails when compiled",needs maintainer decision,28,needs maintainer decision,1,0,155632,,,,0,2bcbfecb46f7c5e61126d6a265cad4fb94559ae3d0091d3fc623999305095d88
thread:issue:155657,"issue #155657: [DTensor] DTensor is not well supported on older versions of GPUs, such as A10",needs maintainer decision,28,needs maintainer decision,1,0,155657,,,,0,2fb82d18f7e32b5d2b3a4d6f666c2d4864dbba2482c7c9c78de5944e0e084878
thread:issue:155664,issue #155664: [Misc] skip_if decorators aborts the test process in distributed/test_composability.py,needs maintainer decision,28,needs maintainer decision,1,0,155664,,,,0,df44177ba5e52bfeceb80578f4ba28b6ed70c2917bdb9481cbe6efc41ebb142f
thread:issue:155679,issue #155679: Torch compile CUDA graphs leads to a large number of CUDA streams,needs maintainer decision,28,needs maintainer decision,1,0,155679,,,,0,7b1b01640cc2a4d317b7119039562e83c4ac0a90454ad476c76096a75cbd342f
thread:issue:155892,issue #155892: [RFC] `DeviceGroup`: a mixin of `ProcessGroup` and `Backend`,needs maintainer decision,28,needs maintainer decision,1,0,155892,,,,0,41968c983ba29f82b8d3c323b319ebbd66bee42511e220ed179feca114f8a323
thread:issue:155916,issue #155916: [`torch.distributed`] Watchdog monitor error handling should match watchdog error handling,needs maintainer decision,28,needs maintainer decision,1,0,155916,,,,0,02ecc40a9cb0f3654810acc1f78230c8cc390ba5ca149e355d75d0eaf3b0c40d
thread:issue:155993,issue #155993: Keep gettting AssertionError: found no DeviceMesh from dtensor args for c10d.broadcast_.default!,needs maintainer decision,28,needs maintainer decision,1,0,155993,,,,0,3ec39b4bcb6de7189738924b30934d5b949ba01149b4ead05c09972a5ff99711
thread:issue:156002,issue #156002: DCP save only saves one shard of tensor parallel model when using DP + TP,needs maintainer decision,28,needs maintainer decision,1,0,156002,,,,0,3c980750a94f82a43e0e47e5b5b83497bcce3d08fd24bd5c52e5167a4f06b96e
thread:issue:156130,issue #156130: [DDP][FSDP2] add unit test to showcase DDP mixed precision with FSDP2 mixed precision,needs maintainer decision,28,needs maintainer decision,1,0,156130,,,,0,4801125a9ea6e3511120c7101f11672a882729b88b52337e18772d6c8eebb1ab
thread:issue:156392,issue #156392: DataParallel gather NaN across multiple gpus,needs maintainer decision,28,needs maintainer decision,1,0,156392,,,,0,8d899ee3b87960222ecfe4e6904cf1efffed72e48ae717655cf943d2bca493c7
thread:issue:156523,issue #156523: [Dtensor] handle dtensor ops that only need to operate on certain shard without all_gather first.,needs maintainer decision,28,needs maintainer decision,1,0,156523,,,,0,143923ecff1d7f10f683dc482c2b226bde45ce32cfcc1deab9a82346b9100dec
thread:issue:156533,issue #156533: `<<` and `>>` operators seem silently broken for DTensor operand 1 and scalar operand 2,needs maintainer decision,28,needs maintainer decision,1,0,156533,,,,0,7699adffd7c4b19e0a6234b244d4d74afcdf71070e7e2b23b561df3afc8cad57
thread:issue:156627,issue #156627: `distributed.all_gather()` support for uneven tensors does not apply to all backends,needs maintainer decision,28,needs maintainer decision,1,0,156627,,,,0,b7cf0ceb2f42faf13a160bb72b7f3f2687cfb5bd8eabbb11224d72fef4e228b4
thread:issue:156771,issue #156771: Support fp16 and fp32 outputs for rowwise scaled_mm,needs maintainer decision,28,needs maintainer decision,1,0,156771,,,,0,ad6d693dd53229c2a8a451b6a41c182244a2b6a52c2b4c51280ad5aed5d65be2
thread:issue:156784,issue #156784: [FSDP2] allow different dtypes for the model params with gradients,needs maintainer decision,28,needs maintainer decision,1,0,156784,,,,0,c3712c945eb2ce8d9fd40efb8b1442e98875fde610ff08c933f145b4ef9f821d
thread:issue:156805,issue #156805: [RFC] Support the unified memory API GetAllocator and getDeviceAllocator for in-tree and out-of-tree backend,needs maintainer decision,28,needs maintainer decision,1,0,156805,,,,0,fcd77ffe21bc9c425caf456dde6c0bd26c58e5d71614b69a1593e4bb315a3890
thread:issue:156818,issue #156818: Tighter invariants for the inductor graph,needs maintainer decision,28,needs maintainer decision,1,0,156818,,,,0,cab52e3cbaf4a38079ce3641063009baff39e43606aac31bb780c489054e8a69
thread:issue:156842,issue #156842: FSDP2 + TP does not work,needs maintainer decision,28,needs maintainer decision,1,0,156842,,,,0,25e8a542594b68b79678bfd7bd5290986bfffe755bee771cf4e8a5e50d1330bc
thread:issue:156942,issue #156942: Can't use pinned_memory to do async save ckpt when using torch.distributed.checkpoint.async_save,needs maintainer decision,28,needs maintainer decision,1,0,156942,,,,0,20483fdb00ff24c00a51fa55c7dc7b4a9837879159ac356436068f8efaaa7c82
thread:issue:156966,issue #156966: Perf drop when running with FSDP and torch.compile,needs maintainer decision,28,needs maintainer decision,1,0,156966,,,,0,128798fe1ca36544718eb06dc3496eef89daca0cebd18fd4b829336eea61354d
thread:issue:157027,issue #157027: [RFC] Remove the FSDP data copy from compute stream critical path,needs maintainer decision,28,needs maintainer decision,1,0,157027,,,,0,4d17fe1c8d948bbbf06190f1f31f3722316b6c18f058176a07f18ecc84d726b4
thread:issue:157318,issue #157318: Torch Elastic Wait timeout increase,needs maintainer decision,28,needs maintainer decision,1,0,157318,,,,0,cbf5da70097ff731004a240aa6ac70cdd21856ace0d22289b26eff14295d9fba
thread:issue:157345,issue #157345: nn.rmsnorm is super slower than nn.layernorm,needs maintainer decision,28,needs maintainer decision,1,0,157345,,,,0,78bd83396ce68b780525711c90eb343fb395307c291a1dec24201209ed382ae8
thread:issue:157395,issue #157395: [FSDP2] figure out the contract for mp_policy and tensor subclass extention,needs maintainer decision,28,needs maintainer decision,1,0,157395,,,,0,75d975f3bee33df9300f39299e3765e25e9a4e3b22dbec907e2edc0e9cf9f3f2
thread:issue:157456,issue #157456: Deprecation of CUTLASS Python interface,needs maintainer decision,28,needs maintainer decision,1,0,157456,,,,0,cafcc8afc44f1e0866a34e19a6864293ade6204d4dd1a88303ebbf9af7832d94
thread:issue:157781,issue #157781: torch.distributed.checkpoint.state_dict.set_state_dict stucks with StateDictOptions(full_state_dict=True),needs maintainer decision,28,needs maintainer decision,1,0,157781,,,,0,05fa6f4cce4b5b3ec773bda94afb184a594b1987e5c694b65268e559878482c7
thread:issue:157862,"issue #157862: Spurious ""Grad strides do not match bucket view strides"" warning for 1x1 convolution",needs maintainer decision,28,needs maintainer decision,1,0,157862,,,,0,b3bdd0ff4faf3959d9c20a123dc91cd9cb46cb43e12af79058be3568c712cc48
thread:issue:157989,issue #157989: [XPU] torch.xpu.mem_get_info() query failed or reports Out-Of-Memory even when there are enough free memory,needs maintainer decision,28,needs maintainer decision,1,0,157989,,,,0,b25520aa82de312c0ba7e5d2cffd3d91d86ea9b1082453751cab4eb41f02c2c5
thread:issue:158016,"issue #158016: dist.all_to_all_single, when input tensors of different shapes result in undefined output behavior",needs maintainer decision,28,needs maintainer decision,1,0,158016,,,,0,2ff55a20f79967db5efe82ec0a31ce9d814d4a158ccdb869a284bb6342a6d063
thread:issue:158032,issue #158032: extension-cpp claims ROCM support with no additional changes; we should test this.,needs maintainer decision,28,needs maintainer decision,1,0,158032,,,,0,9ee3d44d3ad520be8a0715793079ffbd5281e1616a01feb417d896d078556970
thread:issue:158186,issue #158186: Flash Attention 2 + Dynamo + FSDP accelerate plugin + torch.compile error,needs maintainer decision,28,needs maintainer decision,1,0,158186,,,,0,9e0e475e3a2fac7ddc479c5df78a20aca3eba1abecb47641e0a713560885c727
thread:issue:158343,issue #158343: Upgrade to latest XNNPACK,needs maintainer decision,28,needs maintainer decision,1,0,158343,,,,0,fe9cb2b7118ced236c60b3ad7e4fa0099de0c1d226367cd7dce12eac865a89f6
thread:issue:158443,issue #158443: DISABLED test_gradients_synchronizations (__main__.TensorPipeTensorPipeCudaDistAutogradTest),needs maintainer decision,28,needs maintainer decision,1,0,158443,,,,0,4d437860ee6ecb269deab3b2340f5100ac2124d8501ed295e737e036f4611889
thread:issue:158519,issue #158519: `get_model_state_dict` output inconsistent before and after `set_model_state_dict` when using `torch.compile`,needs maintainer decision,28,needs maintainer decision,1,0,158519,,,,0,aa56890cecce613ae572be939fad4ef0ab34238e97ecdffc2cf998f3fd740d46
thread:issue:158551,issue #158551: CUDAGraphs RuntimeError: Accessing Overwritten Tensor Output Despite Clone and cudagraph_mark_step_begin in PyTorch 2.3.0,needs maintainer decision,28,needs maintainer decision,1,0,158551,,,,0,ef0337bba370970951fb854d8f14db91c73ec349d250aa0c74ffcd03ed74eff3
thread:issue:158714,issue #158714: Feature request: `DataLoader` using multithreading instead of multiprocessing,needs maintainer decision,28,needs maintainer decision,1,0,158714,,,,0,03066d56b858b9019256190ae3131ce0e21e4926f088e344d5c4d0779e3e1aa1
thread:issue:158775,issue #158775: Add docs for torchrun numa binding on website,needs maintainer decision,28,needs maintainer decision,1,0,158775,,,,0,e499be3f177eeb387eae3fb0411dfd9cb1b957081103c59109fdc070c595bd8c
thread:issue:158911,issue #158911: [RFC] Refactor test_c10d_nccl.py to Support Backend-Agnostic Testing for PyTorch,needs maintainer decision,28,needs maintainer decision,1,0,158911,,,,0,9ca5b8fae4359078fd91fc5535837deb6344a7b48e8a2e942a6961271476d199
thread:issue:159113,issue #159113: DISABLED test_graph_two_successive (__main__.TestCuda),needs maintainer decision,28,needs maintainer decision,1,0,159113,,,,0,d33ccf241f1b7c1d89e93edef72483044346d82b1d2325a969df971de5999ff7
thread:issue:159150,issue #159150: DISABLED test_fake_crossref_backward_amp_nn_functional_bilinear_cuda_float32 (__main__.TestFakeTensorCUDA),needs maintainer decision,28,needs maintainer decision,1,0,159150,,,,0,bf68cdbbc57ae3caa5bd014b342d3cbbfcdfefb84a75a3d667d91c10d84a0ed0
thread:issue:159151,issue #159151: DISABLED test_fake_crossref_backward_no_amp_nn_functional_bilinear_cuda_float32 (__main__.TestFakeTensorCUDA),needs maintainer decision,28,needs maintainer decision,1,0,159151,,,,0,b8bafef0b8485cfd16a831490a79a681d83f287e7656a9c34c39fb6d1716ba5d
thread:issue:159155,issue #159155: Audit all usages of constrain_to_fx_strides. Many of them might want constrain_to_fake_tensor,needs maintainer decision,28,needs maintainer decision,1,0,159155,,,,0,49522ee4741c738cb49ac6b723e469a6883225c60e3ce01a33b3b45ec375f3ba
thread:issue:159413,issue #159413: Inductor Assertion Error with extremely large input,needs maintainer decision,28,needs maintainer decision,1,0,159413,,,,0,9151c5524c6b0cd1574b60d1c6de1fee42a620293549c6b4d4444d3f58caec3a
thread:issue:159444,issue #159444: #define CHECK causes issues with a bunch of other libraries,needs maintainer decision,28,needs maintainer decision,1,0,159444,,,,0,a7b0abec130dbd507110c38236fbfd89ecb20ce6ed5c2c7d38eb38acbf320641
thread:issue:159516,issue #159516: Validate PyPI Wheel Binary Size failing for >3months,needs maintainer decision,28,needs maintainer decision,1,0,159516,,,,0,1189449b48c6eab63e036df460bbd49ca9bf5940563323b98eef0ae2ebb34dd7
thread:issue:159520,issue #159520: [RFC] Enable native Windows CI/CD on ROCm,needs maintainer decision,28,needs maintainer decision,1,0,159520,,,,0,1a24b4c286995f07cfad6312facf8a9a16148b8411f36a4f38aadada1244d9c7
thread:issue:159547,issue #159547: grouped_mm for float16,needs maintainer decision,28,needs maintainer decision,1,0,159547,,,,0,59c2b4f72283690cfa75906ca72a5a91dee49d73657ce9cb2ddf3d2bc22fd43d
thread:issue:159563,issue #159563: About init_process_group with different backends,needs maintainer decision,28,needs maintainer decision,1,0,159563,,,,0,7bc40a35f4018acd8f390cb415a1ce6bb64b1ec1ddb65d22f60497bd9757f090
thread:issue:159635,issue #159635: [BUG] DTensor (PyTorch native TP) + dynamic shapes + torch.compile failed!,needs maintainer decision,28,needs maintainer decision,1,0,159635,,,,0,b1ad949c8d9f6accb374ebca2035777d5e2632f9689d196200f421a349d9e29f
thread:issue:159686,"issue #159686: c10d broadcast copies at::Tensor, creates footgun for downstream implementations",needs maintainer decision,28,needs maintainer decision,1,0,159686,,,,0,56390c6c5716d33f529f63d3dc5ffeb3ae0b38d87ef27841d7fe2a2ae8c28a18
thread:issue:159711,issue #159711: Support dict as input/output for pipeline parallelism,needs maintainer decision,28,needs maintainer decision,1,0,159711,,,,0,7ed32a1aef9b18f986164d457e0b620afab38083c073434239936bf035338c8f
thread:issue:159770,issue #159770: extract statistics from attention weights in FlexAttention,needs maintainer decision,28,needs maintainer decision,1,0,159770,,,,0,2f0efae598964c91ee57300df7d777b1cd29ce1492af6297cb9f6ab4415ec27f
thread:issue:159836,issue #159836: Missing documentation for device mesh on DDP,needs maintainer decision,28,needs maintainer decision,1,0,159836,,,,0,fb4346d564762de37a1b1585c17c0ab2830102b7f4322266d03a038196d6242c
thread:issue:159959,issue #159959: TensorMetadata not retained during DTensor convolution double backpropagation,needs maintainer decision,28,needs maintainer decision,1,0,159959,,,,0,2ff42e27ecd92c826efe4dfcb50976c33ebca2c95b4f5370144c2586a259442b
thread:issue:160006,issue #160006: Support NUMA Binding for Callable entrypoints to `elastic_launch`,needs maintainer decision,28,needs maintainer decision,1,0,160006,,,,0,2afdfe474ddf44114626348c5ad7c95cb6138f261e10cd41e4892031675a4050
thread:issue:160068,issue #160068: FSDP2 not compatible with transformers >= 4.54.0 GenericForTokenClassification,needs maintainer decision,28,needs maintainer decision,1,0,160068,,,,0,32cf10f62ee25d5dd58bd6b7bbeb629c0a48c5675d6e84569229455f6b839c44
thread:issue:160130,issue #160130: Torchrun breaks with virtual environments,needs maintainer decision,28,needs maintainer decision,1,0,160130,,,,0,3e112ad9bef580b17844485f8c73cb687ef41b168126f10779a7eddab1321476
thread:issue:160169,"issue #160169: When using DP + TP, DP only parameters diverge across TP ranks if using operations with non-deterministic implementations",needs maintainer decision,28,needs maintainer decision,1,0,160169,,,,0,591d03f99cdbe20bc909b3c3e0b13b16551b294f91939ac2cb58497f338575ab
thread:issue:160272,issue #160272: Add Green Context Support for fsdp2,needs maintainer decision,28,needs maintainer decision,1,0,160272,,,,0,73d87826450dcb2666bb138407b8192d828a5bd03ae90316a93cf6782906d891
thread:issue:160333,issue #160333: Tensor subclasses don't work with torch.library.triton_op,needs maintainer decision,28,needs maintainer decision,1,0,160333,,,,0,df56b91f5428aa88fae4c5ce86d188db9d47e781c8168bb12f881ff53a7c209a
thread:issue:160340,issue #160340: Meta device initialization on HF models + FSDP leads to different numerical behaviour,needs maintainer decision,28,needs maintainer decision,1,0,160340,,,,0,864ba7742ecb486ed2065c3eeb00383a69bf1297bf0786ec7ee4c48ea11712c6
thread:issue:160472,issue #160472: PT2 Cache <> CustomOp issue,needs maintainer decision,28,needs maintainer decision,1,0,160472,,,,0,bc4ff5123815232ea83f9d05cb4c17e5b11f15612c08788a6bb36454317d0285
thread:issue:160514,issue #160514: BenchmarkingTest.test_benchmark_on_non_zero_device mock hit failed,needs maintainer decision,28,needs maintainer decision,1,0,160514,,,,0,e65805aca82fc7f754e1265c3a062bae12861dd1adfa856a58c2c2881ccd326f
thread:issue:160648,issue #160648: How to Use Pipeline Parallelism in Multi-input Models,needs maintainer decision,28,needs maintainer decision,1,0,160648,,,,0,fb1e8e45242e2b4e1296d731cb8200bce5b0caf43100efda25b22d48d7beb40d
thread:issue:160731,issue #160731: Cannot initialise process group on CPU device with GLOO backend,needs maintainer decision,28,needs maintainer decision,1,0,160731,,,,0,1f55ed52fd23174026da18feca088fc9d9af0010bea99ee3319eb29bb208f105
thread:issue:160732,issue #160732: GLOO backend should have MPS capability,needs maintainer decision,28,needs maintainer decision,1,0,160732,,,,0,3b0e42d718ad0cfe7dd3683c78297e5683debe2c6d4ea0ce961b06116140f022
thread:issue:160740,issue #160740: MPS torch.argmax/torch.argmin fail for non-contiguous input,needs maintainer decision,28,needs maintainer decision,1,0,160740,,,,0,87676c77a2a12a04f8fb26b1ce7688d4ed01afe5ae6710dcb41b739d4a7cd66f
thread:issue:160911,issue #160911: Severe differences between outputs of Linear with DTensor and Tensor in half precision,needs maintainer decision,28,needs maintainer decision,1,0,160911,,,,0,538a2069fa5c1237f3adf497eaa5af7b38099fbfd8a5f959fd81c17c87bf496a
thread:issue:160950,issue #160950: fx.node._sort_key dones't work right with torch.distributed._functional_collectives,needs maintainer decision,28,needs maintainer decision,1,0,160950,,,,0,c9205b5ab9b7a6b99e819f5aa60fed6d92a4baf523640f5a675eb6b7299e4984
thread:issue:160968,issue #160968: [DTensor] broadcast_to fails with KeyError,needs maintainer decision,28,needs maintainer decision,1,0,160968,,,,0,a90a63df52a8a44efee5443f84b7bd8d9a8edc85b47794933afa31236fddea1a
thread:issue:161091,issue #161091: DTensor view op results in RuntimeError: shape '[38597376]' is invalid for input of size 38598144,needs maintainer decision,28,needs maintainer decision,1,0,161091,,,,0,2cea5f3305b92f7a9fbc975a76f11e589e06677a9bd85a7467c1669e7b04d593
thread:issue:161153,issue #161153: torch.compile fails in FSDP due to .data assignment,needs maintainer decision,28,needs maintainer decision,1,0,161153,,,,0,2f02cc6c1f53113da88645e7771b806aa88348b15c9e7912e37b736339b9d324
thread:issue:161218,issue #161218: [DTensor] Losing TupleStrategy in OpSchema.args_strategy,needs maintainer decision,28,needs maintainer decision,1,0,161218,,,,0,b91f9182919536e21faa17387f7607d0cff76b178aa3bb8f76075254a1d6e1bb
thread:issue:161368,issue #161368: Incorrect offsets for 2D uneven sharding in DTensor in `compute_local_shape_and_global_offsetproduces`,needs maintainer decision,28,needs maintainer decision,1,0,161368,,,,0,3fa589c2e36d9218d5935da9052cac0688e1641a3fcbb0c36aa86a34e4e1bb8b
thread:issue:161609,issue #161609: Bucketize lowering crash when using a sliced computed tensor as boundaries,needs maintainer decision,28,needs maintainer decision,1,0,161609,,,,0,bfde962f111e707beba538cac1584c12e4038dbd767e6bd3009a774dd09b0ef0
thread:issue:161757,issue #161757: [RFC] Flash decoding optimization based on FlexAttention on X86 CPU,needs maintainer decision,28,needs maintainer decision,1,0,161757,,,,0,b89db9bf03f6024bb3cee76d181f6d9e0ff510be66dc7660459ed6aaff556e32
thread:issue:161776,issue #161776: FSDP&device_mesh Mesh_dim_names in init_device_mesh not work,needs maintainer decision,28,needs maintainer decision,1,0,161776,,,,0,0924ec122ac6b570050aa2432e67fcad83d322fa7cba5b4ec80ce4a42788c339
thread:issue:161842,issue #161842: Exhaustive Autotuning causes `cudaErrorIllegalAddress`,needs maintainer decision,28,needs maintainer decision,1,0,161842,,,,0,f536b3430c938043f03971e5e1b1b1f77a1564697931498b5b821bbbaa11392c
thread:issue:161847,issue #161847: Feedback about FSDP Notes,needs maintainer decision,28,needs maintainer decision,1,0,161847,,,,0,e2357cbcad852665785d968808d454915f0c9f9843d7ea569ffe10b67a26287f
thread:issue:161974,issue #161974: DTensor communications cost model is way off for H100,needs maintainer decision,28,needs maintainer decision,1,0,161974,,,,0,a60999575d50d204afb351a9111b0eda021bf3b889229a558c3bd9c977f4c93a
thread:issue:162032,issue #162032: torch thread yield after launch nccl kernel,needs maintainer decision,28,needs maintainer decision,1,0,162032,,,,0,1c75abbce03444ae5885c8efb54cb002390facc489b4e22899471777a7e287f5
thread:issue:162160,issue #162160: [BUG] Experts weight not release after backward when FSDP2 + moe expert parallel,needs maintainer decision,28,needs maintainer decision,1,0,162160,,,,0,5a012ad69c6b38a9248ad0c8c446e6cc6a0bf2914a24d945eca1381b48e60330
thread:issue:162215,issue #162215: consider allowing more dtype in the dtensor opinfo test,needs maintainer decision,28,needs maintainer decision,1,0,162215,,,,0,2884f292509c913e1b4fd3cd304134aeb598b03776e159fd4e517929b30c5aef
thread:issue:162237,issue #162237: Does it support new CPU group after passing `device_id` to `init_process_group`,needs maintainer decision,28,needs maintainer decision,1,0,162237,,,,0,8aff4c3c09f1a61cb2345ee279334740fc1661ba24ef6bc97e8fb8fda73be748
thread:issue:162257,issue #162257: Optimize FSDP1 sharded_state_dict saving performance,needs maintainer decision,28,needs maintainer decision,1,0,162257,,,,0,1560c595fd806f18a8838c72110033e28550de78226e7942eb374edc28b8d9d0
thread:issue:162290,issue #162290: local_output_for_attn: bool = False ( for Col/Row-wiseParallel ),needs maintainer decision,28,needs maintainer decision,1,0,162290,,,,0,41e574960ac5e524c5cf0a82e694aedf958c61f6945df683a3f1cda65877f599
thread:issue:162638,issue #162638: Gradient Clipping in Pipeline Parallelism Schedules,needs maintainer decision,28,needs maintainer decision,1,0,162638,,,,0,b335a7a5fa34ceaabd505b5d8bb308bce7f0a6380c101508a880df63546719e8
thread:issue:162668,issue #162668: ReduceOp has incorrect py::class_ binding,needs maintainer decision,28,needs maintainer decision,1,0,162668,,,,0,3bf78fafa6db4fe96121fa6aa497edbfd0633767a8b60c43d4483e8775662b23
thread:issue:162728,issue #162728: Significant Numerical Inconsistencies After torch.compile() for Local Attention Model,needs maintainer decision,28,needs maintainer decision,1,0,162728,,,,0,ac8565b05c4fd068b19811637e89514c388c91de19bc7dc26b377e35affb7d2f
thread:issue:162741,issue #162741: `torch.distributed.pipelining` Llama 2 and 3 Results Torch Dynamo Compiled Error,needs maintainer decision,28,needs maintainer decision,1,0,162741,,,,0,4e77a479dd11b224e4be9d17453c95933a004240a59eee9ae720ae1706bc0d92
thread:issue:162755,"issue #162755: Ideally, we should not use `require_exact_world_size(4)` if we only have 8-GPU runners",needs maintainer decision,28,needs maintainer decision,1,0,162755,,,,0,5a1ed60c8a017d7fc5f42d4da862bdf0ab8a5188e1e30974794259ddeb3cc25c
thread:issue:162780,issue #162780: Prebuilt Wheels supporting T4G,needs maintainer decision,28,needs maintainer decision,1,0,162780,,,,0,8446923c044c08704e9d9e50e75c142c0bc23024e09de8128c47cbb4e8cbf251
thread:issue:162872,issue #162872: [MPS] dead lock when calling `Event.synchronize()` before `Event.elapsed_time()`,needs maintainer decision,28,needs maintainer decision,1,0,162872,,,,0,7a6be857ca2844344a1cb53afb1010a92b4de5bc6ff157c2a271c754bf5bea5a
thread:issue:163043,issue #163043: `parallelize_module` causing scatter,needs maintainer decision,28,needs maintainer decision,1,0,163043,,,,0,93492575e77927c05cfb17c3462f4aa311e44d70082075d0842b542aa27fef5b
thread:issue:163327,issue #163327: [MPS] scatter_add_ is incorrect on MPS with non-zero offset slices,needs maintainer decision,28,needs maintainer decision,1,0,163327,,,,0,ba7f9e6a2d1cb0e6f7d6bd916fb09bad19de8311a7a9ae44069e47ab724c1d83
thread:issue:163546,issue #163546: Better nccl timeout handling,needs maintainer decision,28,needs maintainer decision,1,0,163546,,,,0,aeae6c68c1e3d23df605e48207eb8f32755db080d464e4b12781973cb87874ba
thread:issue:163650,issue #163650: [DO NOT CLOSE] Autorevert actions shadow mode stream,needs maintainer decision,28,needs maintainer decision,1,0,163650,,,,0,f1310a74a36e8a81deac521487dc1cd749c09e2911c0306953848430b9237012
thread:issue:163678,issue #163678: RuntimeError: NCCL error: unhandled cuda error,needs maintainer decision,28,needs maintainer decision,1,0,163678,,,,0,fdca973e7797007b61c71e9872b933a6010f339ab2570065b3361a482206c20f
thread:issue:163688,issue #163688: [torch.distributed.pipelining] Gradients are None in first training step with ScheduleGPipe,needs maintainer decision,28,needs maintainer decision,1,0,163688,,,,0,9e8a588df425a5a37f5690121d66d1bec547aa7f6ff2e0d55e396ef54040d147
thread:issue:163692,issue #163692: [feature][dtensor] support nn.Module inputs containing DTensor parameters in local_map,needs maintainer decision,28,needs maintainer decision,1,0,163692,,,,0,8675878db5dd6cfb64d6a788f64db0b708da2c2287a63c8d66c5338bc0387cbe
thread:issue:163716,issue #163716: torch.compile bug because aot autograd trace ignores requires_grad,needs maintainer decision,28,needs maintainer decision,1,0,163716,,,,0,b775089def2149c78e81036a2d7837180b2be40d54cb2504cbbc2562b624cac1
thread:issue:163722,issue #163722: AdaptiveLogSoftmaxWithLoss with weight tying,needs maintainer decision,28,needs maintainer decision,1,0,163722,,,,0,ad5a91c71a34208db48e26a635ec0fabcc7aa88a9034ad97fceb032e88977c49
thread:issue:163913,issue #163913: test_triton_wait_until hangs,needs maintainer decision,28,needs maintainer decision,1,0,163913,,,,0,a304f72f9e6c0cdd417e0aa994fad7f5e2c39dc89fb95d900ad1c3f1fa25dd4a
thread:issue:163932,issue #163932: [inductor] a potential numerical instability issue inside a softmax model.,needs maintainer decision,28,needs maintainer decision,1,0,163932,,,,0,de4e87e5f9ac88173acec9195ed492fefa10b224556b744c4cfa28dc8193eb5c
thread:issue:163974,issue #163974: C++ build dependencies,needs maintainer decision,28,needs maintainer decision,1,0,163974,,,,0,b94a326ff89d1f6a8ddec47f6b93506d3a01b938ed9aab721c73e68032bd15db
thread:issue:163997,issue #163997: `F.scaled_dot_product_attention` returns all-`NaN` rows during evaluation but not training,needs maintainer decision,28,needs maintainer decision,1,0,163997,,,,0,d9546e83570d856332a9c2823e5b37657f304c48aaced7253d13cc09e09c24df
thread:issue:164015,issue #164015: [CI][CUDA] test_multimem_one_shot_all_reduce_bfloat16_align_bytes_1 failures on B200 runner,needs maintainer decision,28,needs maintainer decision,1,0,164015,,,,0,955adc5f72912514cdba2d6fc8fe641d77f6c335bf8a62ffb62939b97128be02
thread:issue:164455,issue #164455: AMP + Torch Compile bug,needs maintainer decision,28,needs maintainer decision,1,0,164455,,,,0,5d66550c641186829992e45f7998c885cfb0c1860b7d60347607a8da3c4a6436
thread:issue:164563,issue #164563: [Blackwell][Inductor] Numerical mismatches in test_max_autotune.py::TestMaxAutotune::test_non_contiguous_input_mm_plus_mm,needs maintainer decision,28,needs maintainer decision,1,0,164563,,,,0,0e975cb9f5d803aa93607fd693db2fbe842afa9fb835be01886045d01f63ade5
thread:issue:164593,issue #164593: [CI] Make sure B200 benchmark testing covers regression such as https://github.com/triton-lang/triton/issues/8328,needs maintainer decision,28,needs maintainer decision,1,0,164593,,,,0,d91a3930703613887fcfb315bbd18af5080dc701d7930e3416819c4a99099ea4
thread:issue:164696,issue #164696: Support torch._inductor.config.inplace_buffers for custom_op whenever possible,needs maintainer decision,28,needs maintainer decision,1,0,164696,,,,0,bde07d43706625140f9596f3779a7248e2df2881c0ea6f9c2409ea102d9645a5
thread:issue:165070,issue #165070: Dynamically load CUDSS library using pypi wheel to save wheel space,needs maintainer decision,28,needs maintainer decision,1,0,165070,,,,0,6f9eb5b5bbe4794714a47718b0c2e8f4495f5829742c61c20e46d56ce7bbae58
thread:issue:165280,issue #165280: [inductor] avoid triton kernel when creating zero tensor,needs maintainer decision,28,needs maintainer decision,1,0,165280,,,,0,f430146c0008f26f1968b1c569f94dda31f5bcd00bde486a1e844892cd630998
thread:issue:165398,issue #165398: Skipping combo kernel for large pointwise nodes,needs maintainer decision,28,needs maintainer decision,1,0,165398,,,,0,c1a72e1ea53b697f11ddf821b6677ce3a36e5da89f440248d2ea46803525566d
thread:issue:165501,issue #165501: Inductor custom pass mechanism that maintains config constraints,needs maintainer decision,28,needs maintainer decision,1,0,165501,,,,0,4d91e425f691c319fb7178383e54c1a83205c0154164b1f200bba2d2ce4c83ca
thread:issue:165506,issue #165506: [torch.compile] dynamic shape kernel mismatch,needs maintainer decision,28,needs maintainer decision,1,0,165506,,,,0,20173dfbec0a0c67e7c044aa5aa7be288478838720b782a752a7c9278dae7da5
thread:issue:165628,issue #165628: [Meta] Type Hints in Pytorch,needs maintainer decision,28,needs maintainer decision,1,0,165628,,,,0,b19e273bdc23697dd45c7a3e9e9d97f9b25c14db13baef37b4bdcdcf7c2c1b9d
thread:issue:165653,issue #165653: [distributed] test_symmetric_memory fails on odd-numbered GPU configurations (3 GPUs),needs maintainer decision,28,needs maintainer decision,1,0,165653,,,,0,46d61b7a733f43841c9e86927d9538b3e6158d6ae0692004c9e1368ff4193e7f
thread:issue:165785,issue #165785: Harden CuTeDSL Inductor Path,needs maintainer decision,28,needs maintainer decision,1,0,165785,,,,0,cb12b11439dcf71c71b2cce1248822f83d33ee6984172f97311d2380316c8746
thread:issue:165933,issue #165933: [Distributed] fully_shard: support no_shard (ddp) strategy?,needs maintainer decision,28,needs maintainer decision,1,0,165933,,,,0,a8f8e5d05edce516445277933b8f0ff85a96193474bec1e913c6261441d1b6c4
thread:issue:165987,issue #165987: `triu_/tril_` ignores memory overlaps which can lead to silent correctness issues,needs maintainer decision,28,needs maintainer decision,1,0,165987,,,,0,1daca14270c8bed783573bee923e02dcfe0460eeabd565ef75987f00b71d5d00
thread:issue:166123,"issue #166123: Can't compile torch-cublas-hgemm project in Windows on a torch newer than 2.8.0, I get complains involving compiled_autograd.h",needs maintainer decision,28,needs maintainer decision,1,0,166123,,,,0,81ccae0efee72f06b0cc2b8edc6b4ee806b15b554c8ad21a1259c152f5f12dc7
thread:issue:166166,issue #166166: MultiheadAttention fast_path is substantially less efficient than the alternative,needs maintainer decision,28,needs maintainer decision,1,0,166166,,,,0,794fe35fec7a6372b071f4bd1b4dd116ee745c392a692906b018ea41d348d3fb
thread:issue:166523,issue #166523: DDP + MPI-based distributed backend: missing work/future completion assertion for Broadcast operations,needs maintainer decision,28,needs maintainer decision,1,0,166523,,,,0,9ae5cf815d70731bf233a1f1a434dcfed8be4378d937e69c2d97aeb1fb81db5a
thread:issue:166856,issue #166856: CPU `normal_()` Non-Deterministic Behavior with Non-Contiguous Tensors,needs maintainer decision,28,needs maintainer decision,1,0,166856,,,,0,9e7dde513d65b65e7279e45a8b637a3430a54f35c62a7e63b8b8685b96e4f6be
thread:issue:166930,issue #166930: [DTensor] DTensor randn inconsistent with single device behavior,needs maintainer decision,28,needs maintainer decision,1,0,166930,,,,0,aefbe4624cb354e7694e79ab99c855b29a8e15def417d6c982be7134c6071a12
thread:issue:167036,issue #167036: FSDP2 moves my inputs to device when I don't want it to,needs maintainer decision,28,needs maintainer decision,1,0,167036,,,,0,e3e852ef6a82ae4522824f93027900149d62113c746d995e74b1baa66dcc411b
thread:issue:167062,issue #167062: How to use torch.compile on Windows GPU?,needs maintainer decision,28,needs maintainer decision,1,0,167062,,,,0,ff34b4066a7f0f7486eedf4ce4200e4a716575e90d6679360570bb009c9e57e6
thread:issue:167074,issue #167074: DTensor with `torch.compile` fails with `aten.as_strided.default` error,needs maintainer decision,28,needs maintainer decision,1,0,167074,,,,0,342c94f23b3e90e87e0dcbd591ee699cd49cd8b5cb04eb8311d9b4e290d59ef1
thread:issue:167090,issue #167090: DTensor `F.conv2d` fails with `bias=None` - AssertionError in sharding propagation,needs maintainer decision,28,needs maintainer decision,1,0,167090,,,,0,0c14e0f7968560ba9acfa8877c2a06ca6b7308727a0275187f690d30ff4ecffe
thread:issue:167106,issue #167106: DTensor operations incur unnecessary gathering and memory consumption resulting therefrom,needs maintainer decision,28,needs maintainer decision,1,0,167106,,,,0,060fe29224a4251e6fbad10785b8f81e50c38a81266fcd481e13e1ca0f0df3fa
thread:issue:167219,issue #167219: Are there limitations to dtensor's registration strategy?,needs maintainer decision,28,needs maintainer decision,1,0,167219,,,,0,26a3dd8c350e7ea8158c258da27dd6f6d335d66bb4e5771c24ca1f266a6d1a3d
thread:issue:167272,"issue #167272: DISABLED test_max_autotune_addmm_max_autotune_gemm_backends_ATen,Triton,CK_x_shape2 (__main__.TestCKBackend)",needs maintainer decision,28,needs maintainer decision,1,0,167272,,,,0,162f7bc8c2dcf2bb10547ec03d47152f5e88733273876d6ca9e6dd9b38a519c0
thread:issue:167288,"issue #167288: when use torchrun and set restarts for elastic run, cause NCCL INFO socketPollConnect: connect returned Connection refused",needs maintainer decision,28,needs maintainer decision,1,0,167288,,,,0,be1ea9b7cc111d295a5517569e33e7486df881e58780d8c24ec62dc103b0bdec
thread:issue:167297,"issue #167297: DISABLED test_max_autotune_addmm_max_autotune_gemm_backends_ATen,Triton,CK_x_shape1 (__main__.TestCKBackend)",needs maintainer decision,28,needs maintainer decision,1,0,167297,,,,0,bc107749b33327a947454fc1f1f1286b4cc68aa4cc8b9bb433385dbfd9ba8c3d
thread:issue:167305,issue #167305: Pytorch distributed plugin backend _set_default_backend error when use multi-plugin backends,needs maintainer decision,28,needs maintainer decision,1,0,167305,,,,0,1fd2f10de641afc4ab043e494d3dcdcbcba4962a1bf922b00dd56a78c579b890
thread:issue:167353,issue #167353: `use_mem_pool` is not active during backward,needs maintainer decision,28,needs maintainer decision,1,0,167353,,,,0,5ebc9e8be68323681a49e24596166464d2e572fd9a0538904490699937bcfdb9
thread:issue:167365,issue #167365: PyTorch is no longer importable from `build` folder,needs maintainer decision,28,needs maintainer decision,1,0,167365,,,,0,a1d853d76899ba7bc9c2efc2166c03fa11c17c065a70aa805c02e91ebaf6b0cc
thread:issue:167411,issue #167411: Torch nightly rocm6.4 pytorch-triton-rocm requirement mismatch,needs maintainer decision,28,needs maintainer decision,1,0,167411,,,,0,2d6ca08eac32ec282a87dccbc09586ddc562090292b5a953225b3bd78b3f609b
thread:issue:167425,issue #167425: confusion on pipeline parallelism implementation,needs maintainer decision,28,needs maintainer decision,1,0,167425,,,,0,a4f6f9cc3b4b4809a0703500cc93f0fb4cb90597deee93dec388700a07bb9a70
thread:issue:167540,"issue #167540: [Dtensor]:change the test_mm shape from (12,8) * (8,16) to (512, 512) * (512, 512), throw assert error",needs maintainer decision,28,needs maintainer decision,1,0,167540,,,,0,4b463989532e0f06f9445bcd1f15941efc3cee99dffbd4c932569325ed22a600
thread:issue:167616,issue #167616: UNSTABLE trunk-rocm-sandbox / linux-jammy-rocm-py3.10 / test,needs maintainer decision,28,needs maintainer decision,1,0,167616,,,,0,30ddf5321948cb5ac2b3078674d093ffaa1f574662d6f5a3bb28f9195db721bd
thread:issue:167657,issue #167657: make_fx trick for tracing DTensor compute doesn't work with redistribute,needs maintainer decision,28,needs maintainer decision,1,0,167657,,,,0,9810b5b652c2ae55bb937e17300340796b235fc0a621300ba419e79fe954fb53
thread:issue:167684,issue #167684: [RFC] Access to ncclDevComm in kernel,needs maintainer decision,28,needs maintainer decision,1,0,167684,,,,0,b48e8bc264d658d6a07a1db80cefcebe44d467ee63451a871a39062677bf3d3d
thread:issue:167766,issue #167766: Fix cudagraph flaky tests,needs maintainer decision,28,needs maintainer decision,1,0,167766,,,,0,0b4e4c66663252127222057e3164fde3a0684b2a1a702708df4f69414be17ead
thread:issue:167775,issue #167775: Graceful Ctrl+C handling from torchrun,needs maintainer decision,28,needs maintainer decision,1,0,167775,,,,0,bfa09f5a64e6cde4512a5a22c9fbc38babc5762cccbb17be8b8afdbebec9492f
thread:issue:167822,issue #167822: [RFD] Composable Kernel (CK) moving to C++20 — request for feedback/concerns,needs maintainer decision,28,needs maintainer decision,1,0,167822,,,,0,9259f5e58b6d78552c5cb233ac8edeb5acd1018df8db5a6727cf59a60ad89741
thread:issue:167832,issue #167832: PG splitting does not consider new groups with a different backend,needs maintainer decision,28,needs maintainer decision,1,0,167832,,,,0,0027784b90545801037921067840db98d0fe356219eecd55ff8f6532d9547131
thread:issue:167897,issue #167897: Hooks as extension points for torch.pipelining,needs maintainer decision,28,needs maintainer decision,1,0,167897,,,,0,a73578868ab28245ae9cc988d274f5f0ec23ef81184b65e606e1c8c4e76279de
thread:issue:167917,issue #167917: Operator aten.detach_.default does not have a sharding strategy registered,needs maintainer decision,28,needs maintainer decision,1,0,167917,,,,0,35326ffd0408fdbeeab4bd8d89ca1306c33fef93b1002bd1aceb2c9705c8421d
thread:issue:167994,issue #167994: CI Not Detecting Failing Tests in test/distributed/elastic/*,needs maintainer decision,28,needs maintainer decision,1,0,167994,,,,0,82c9f2f927bf428af817e2d5318dafb5625630aa800af03aa4dc3f06e1822fe2
thread:issue:168099,issue #168099: Unify pointwise DTensor and NestedTensor OP Coverage. Adds over 100 op overloads to DTensor and about to 10 to NestedTensor,needs maintainer decision,28,needs maintainer decision,1,0,168099,,,,0,d8a0e95d656b4b8d342c34f46741affb33a7379bfd55c9884417b43325177b1f
thread:issue:168143,issue #168143: FSDP2 does not work with custom Tensor wrapper objects,needs maintainer decision,28,needs maintainer decision,1,0,168143,,,,0,42acde78de9e704a08e72dd723e4f134e9b67cd66aa490d7c12a6f4cf3840aa9
thread:issue:168168,issue #168168: [CD][ARM][CUDA] PYTORCH Released Docker Images Should Ideally Support Multi-ARCH But Currently It is only x86_64,needs maintainer decision,28,needs maintainer decision,1,0,168168,,,,0,f3f78181f69b788a9bb7914c794133f8fe2a6c8fc03bfb204fc6258dab9dd000
thread:issue:168279,issue #168279: Add flag to error on any Inductor specialization,needs maintainer decision,28,needs maintainer decision,1,0,168279,,,,0,27e7a1faf4cd4d436a1d87028a37420e97e86a73eb639ae67f2e790f7bd0e053
thread:issue:168573,issue #168573: TestModule: test/inductor/test_codecache.py::TestFxGraphCache,needs maintainer decision,28,needs maintainer decision,1,0,168573,,,,0,51d63e1626e0afb80b7244433015ab70f075e1a6abeded87d9950685bf215026
thread:issue:168995,issue #168995: MXFP8 + torch.compile error: NotImplementedError: make_reindexer NYI on DtypeView,needs maintainer decision,28,needs maintainer decision,1,0,168995,,,,0,3f361f8edc4802e5de9f87341252180d7ec3357ee1f54838281f258c313bc438
thread:issue:169015,issue #169015: [CI][CUDA][B200] Test_weight_norm_conv2d_cuda Failure,needs maintainer decision,28,needs maintainer decision,1,0,169015,,,,0,edcd809c1461d82430d6031cbd606a3afc1a763398342231ac347d7d3641587e
thread:issue:169035,"issue #169035: [Question] Why torch.ops.symm_mem.multimem_all_reduce_() don't support e4m3, e5m2, fp16?",needs maintainer decision,28,needs maintainer decision,1,0,169035,,,,0,fc81d56ab40032df1eab769569baa0346388217e1a5a7a45f682f0396e74523f
thread:issue:169182,issue #169182: Kineto library not installed,needs maintainer decision,28,needs maintainer decision,1,0,169182,,,,0,36297f88dd33e9c9c17c020974ecdfd9ea6082a3cecb9d2f2aebf659d358cb5b
thread:issue:169275,issue #169275: GLOO distribute train can not find each ranks,needs maintainer decision,28,needs maintainer decision,1,0,169275,,,,0,96f4a2d2cc54f3d57209ab3a36e6e031ad90ebf5684ba20200f4eeb7cd6d0d68
thread:issue:169320,issue #169320: [DTensor] A New Placement -- RaggedShard,needs maintainer decision,28,needs maintainer decision,1,0,169320,,,,0,ff515c604f7c92176193073566b364322070b2414ab1d3ca6523531cfb7cacc4
thread:issue:169597,issue #169597: Standardize Testing in OpenReg,needs maintainer decision,28,needs maintainer decision,1,0,169597,,,,0,4806e88ca3737809ec886cb2a3f2635d2dd3806d08f7bdd4d2459e38506d2933
thread:issue:169670,"issue #169670: Energy normalization error on torch.fft.fft2 and torch.fft.ifft2 when the matrix size = [2048, 2048]",needs maintainer decision,28,needs maintainer decision,1,0,169670,,,,0,2f02b2d70cb2888025e4373ed5f8747b2a6456586bc0812e88af1829831b59f2
thread:issue:169672,issue #169672: torch while_loop does not avoid CPU synchronization,needs maintainer decision,28,needs maintainer decision,1,0,169672,,,,0,67fb2376e087584fa74a8ba02f591ffd8effdfc8778ffe40471a284b9ae542c1
thread:issue:169752,"issue #169752: if __name__ == ""__main__"": run_tests() cannot be found in the test files under distributed/elastic/rendezvous",needs maintainer decision,28,needs maintainer decision,1,0,169752,,,,0,6fba258ab7f96533c1b956dfb35e48a18579c1f2c3cfca01f8ce232296d3177c
thread:issue:169814,issue #169814: DTensor.detach() retains stale DeviceMesh after destroying and reinitializing process groups,needs maintainer decision,28,needs maintainer decision,1,0,169814,,,,0,043a0598a15879adcab8ac9f55a3a2ac0cc41f78ed3263732f4f465b6c14ac8c
thread:issue:169841,issue #169841: ProcessGroup hash doesn't incorporate pg_options,needs maintainer decision,28,needs maintainer decision,1,0,169841,,,,0,4fa7bf2b3a0e36f261423e39502fd60ff49346ee6af8edf160bbb9567d202da5
thread:issue:169926,issue #169926: Failed to build PyTorch with CUDA by using clang with undefined reference of const_data_ptr in TensorBase,needs maintainer decision,28,needs maintainer decision,1,0,169926,,,,0,e420643f91bb6e8c1aabbcef080c6e245f43e20755c3397a19babdbf03cc5d68
thread:issue:170117,issue #170117: PyTorch Nightly's Torch Compile Feature Does Not Work on THOR due to Triton 3.6.x regression.,needs maintainer decision,28,needs maintainer decision,1,0,170117,,,,0,a29ad30a4fdc3395ad57a3dfc92bbd2e5e06b5043fa12844ef24c3735aa01706
thread:issue:17023,issue #17023: Download speed issues with the pytorch conda channel,needs maintainer decision,28,needs maintainer decision,1,0,17023,,,,0,9a5ce4466d4aec2a8489de2078332e32ff44c69cbd7668778e0d4083d94bc354
thread:issue:170241,issue #170241: Allow for multiple stream captures to a cuda graph cudagraph at once.,needs maintainer decision,28,needs maintainer decision,1,0,170241,,,,0,8788deb6a6fe90017af8c2b5e55631862ac53ed009849e0cb5c91ed9f1355c84
thread:issue:170290,"issue #170290: A lot of ""Failed to check the ""should dump"" flag on TCPStore"" output in ProcessGroupNCCL when rank 0 exit first",needs maintainer decision,28,needs maintainer decision,1,0,170290,,,,0,6e904e5110b1575a90022d4236ad478f1500a651a034851794311e4ba767a894
thread:issue:170361,issue #170361: [Dynamo] Use VariableBuilder/SourcelessBuilder consistently,needs maintainer decision,28,needs maintainer decision,1,0,170361,,,,0,b9a71512d0abe4929895998c05b6775c8ce3ca55eb0313e0d3a0defe963d9901
thread:issue:170419,issue #170419: Cannot set version_counter for inference tensor,needs maintainer decision,28,needs maintainer decision,1,0,170419,,,,0,0fddf3d53f8fe8c968b0dd31bdbd9163f3328822720576b86469c9f2bba90982
thread:issue:170479,issue #170479: nvshmem [GPU 0] Peer GPU 8 is not accessible,needs maintainer decision,28,needs maintainer decision,1,0,170479,,,,0,9e79ebc0de8c41f45edc4e1d5a52c3743953ca2a34025a93636adb0712a6e244
thread:issue:170535,issue #170535: Graphsafe RNG clone_state,needs maintainer decision,28,needs maintainer decision,1,0,170535,,,,0,cdceb2e6dde8c9ef3f81715beab019aa4bec1c7bf3ab95c0fceb807e4e2acbb7
thread:issue:170556,issue #170556: [dynamo] Add linter rule to disallow `raise Unsupported` in torch/_dynamo,needs maintainer decision,28,needs maintainer decision,1,0,170556,,,,0,31e5267580f1f8098bc8a01e7cfd5290f6413795ef5928abde7ea9964447ecf7
thread:issue:170558,issue #170558: [dynamo] Remove all Unsupported subclasses,needs maintainer decision,28,needs maintainer decision,1,0,170558,,,,0,2ba02e3d6aa1777358fbe27db9862bc30d2aba5ddde99a56b55fad8c5efac399
thread:issue:170562,issue #170562: PadMMTest.test_original_aten_preserved_pad_mm fails,needs maintainer decision,28,needs maintainer decision,1,0,170562,,,,0,787240f783ea68b789642ab501ce4a117e263377366c573ef584d3cace21a27c
thread:issue:170743,issue #170743: Reference count modifications led to performance drop,needs maintainer decision,28,needs maintainer decision,1,0,170743,,,,0,72ea03b45be48134847fdabc6915085a2278f09358d8809e74e0221470a2b6bb
thread:issue:170758,issue #170758: bf16 layernorm has wrong answer,needs maintainer decision,28,needs maintainer decision,1,0,170758,,,,0,1b51dc90708945d46bd2339aded2dcad952bdd9be98f4c66e4fc50067191184c
thread:issue:170762,issue #170762: [BUG][Inconsistency] aten.cudnn_batch_norm_backward produces numerical mismatch between eager and inductor,needs maintainer decision,28,needs maintainer decision,1,0,170762,,,,0,2f5682afc29c757ac89a3ace6d821a6cfbfba8c6e7f83056e4508c2d0b9cb0b5
thread:issue:170837,issue #170837: MPS backend - inconsistent results for batched inference on BERT/RoBERTa,needs maintainer decision,28,needs maintainer decision,1,0,170837,,,,0,87690695dfb50645185efb71915c842b9b6c489bae46cc8d63668c68a52ffcf8
thread:issue:171022,issue #171022: In-place real-domain FFT operators for memory-efficient training (bf16 support),needs maintainer decision,28,needs maintainer decision,1,0,171022,,,,0,79f5c45e80b432238612f5389734c01ced4dfb1bcb85706ea8bb0b35f85f18e0
thread:issue:171194,issue #171194: CUDA Smoke tests: Extend smoke tests to include profiler coverage,needs maintainer decision,28,needs maintainer decision,1,0,171194,,,,0,bf1af179da57edd3ea0d83955ff80ae8812aabecf9310a9cb938472d294846a8
thread:issue:171283,issue #171283: aten.uniform cannot be called with keyword argument from,needs maintainer decision,28,needs maintainer decision,1,0,171283,,,,0,24465fbe66030f0047e75c686d931c8ef2e02af8dc94916ffbe17e69071162f3
thread:issue:171375,issue #171375: aten.lift triggers internal assert failure in Inductor functionalization pass,needs maintainer decision,28,needs maintainer decision,1,0,171375,,,,0,6ce811d3f1ec186a84161b75f67189bda62c6b57294c04f0f13009ad4e621f25
thread:issue:171376,issue #171376: PyTorch ThroughputBenchmark.add_input crashes when split_sizes is incompatible with input tensor shape,needs maintainer decision,28,needs maintainer decision,1,0,171376,,,,0,3fa30c1a075e956c30d82a342070ab7332e74dc6791770b5149cb5e4575d39b2
thread:issue:171398,issue #171398: ISAAC inference,needs maintainer decision,28,needs maintainer decision,1,0,171398,,,,0,4a41d0b39a7cc26d6e6c1ea98e62e099fa0f4272909708bbc5114348338cbb3c
thread:issue:171495,issue #171495: [internode] Symmetric mem rendezvous erro,needs maintainer decision,28,needs maintainer decision,1,0,171495,,,,0,d85ee81f8fc00fbfe43a1a534d823578a473819b5a1d93903eeb1b8b36f53aab
thread:issue:171501,issue #171501: Several Windows-related GitHub Actions not running — are they intentionally disabled?,needs maintainer decision,28,needs maintainer decision,1,0,171501,,,,0,e85df107597944dcf43bf3c8adeb48fd86b1526f911e1d012e54a626b5224e0f
thread:issue:171558,issue #171558: `torch.sparse.spdiags` produces incorrect results,needs maintainer decision,28,needs maintainer decision,1,0,171558,,,,0,24442295d1c9dd22361020a864acff9e57d62fbb430824160ba5c9c7cbfc725c
thread:issue:171568,issue #171568: [ROCm][Inductor] Flash/efficient attention meta kernel shape mismatch on CDNA (MI200/MI300),needs maintainer decision,28,needs maintainer decision,1,0,171568,,,,0,354294d585c6a4b986c13c2ab574cabd2b9f1e37d8a352bd2796aaf62f624ac3
thread:issue:171620,issue #171620: Feedback about torch.autograd.graph.Node.next_functions,needs maintainer decision,28,needs maintainer decision,1,0,171620,,,,0,ca45b3137c1b534f8cc14bbf1823d30e2e54624d5db1d0f7e526a55d63a74189
thread:issue:171656,issue #171656: torch.distributed.pipelining fails on models having DynamicCache (esp. Llama),needs maintainer decision,28,needs maintainer decision,1,0,171656,,,,0,3db5bc0a0d05f2e049b44fa76b47bb47b9b5425e5346b7d0d77eff9c5d631b1b
thread:issue:171759,issue #171759: [ONNX] Create a tutorial,needs maintainer decision,28,needs maintainer decision,1,0,171759,,,,0,e5a28d18aa98142af9a10315f9f4666ccf6c38f1673ab8a397d5a2037df74d09
thread:issue:171796,issue #171796: Exception in subprocess causes infinite hang and zombie processes,needs maintainer decision,28,needs maintainer decision,1,0,171796,,,,0,dbbb9e54870c9a3ed5377ef934aa11171f7de6801c0a6d3797ecc9dd5408059f
thread:issue:171866,issue #171866: Why are there three run_repro functions implemented in pytorch?,needs maintainer decision,28,needs maintainer decision,1,0,171866,,,,0,5254b1d55fa37cfd54a45285a6d1c31234eaa6c1070378bf12c6aaab7fae6417
thread:issue:172059,issue #172059: Add Tweedie distribution to torch.distributions,needs maintainer decision,28,needs maintainer decision,1,0,172059,,,,0,770a91d2b4a8e6e6a191e4eed7076630358961af80439af42d208d6c5882a35e
thread:issue:172213,issue #172213: Eager buffer mutation after `torch.compile` + backward + `torch.export.export` hits `_map_assigned_buffer_to_proxy` `FakeTensor` assertion,needs maintainer decision,28,needs maintainer decision,1,0,172213,,,,0,3b9eb111fbfe53532a6cd31156c7d8d4b965e4ea6cd2b584c7c88aea03bba91b
thread:issue:172231,issue #172231: [cuBLAS] setting `cublas/cublaslt` backend seems to be a no-op for `torch.addmm` when dispatched to cuBLASLt,needs maintainer decision,28,needs maintainer decision,1,0,172231,,,,0,c413e034bd72fd153ce608cebb4d6e79162a80b50ba075a71cb1ed327719f736
thread:issue:172273,issue #172273: Numeric difference between div and foreach_div,needs maintainer decision,28,needs maintainer decision,1,0,172273,,,,0,6ca691cca473866d4562430973df0787bcd8a34d1eeeaf27c14429b8d0f5cb19
thread:issue:172319,issue #172319: [OpenReg] Add OpInfo-based operator tests using op_db,needs maintainer decision,28,needs maintainer decision,1,0,172319,,,,0,ec0e9f84952c8c127a5c7ea42267feac8cc211f7e6f2e81d80489bed3d3fd3d4
thread:issue:172320,"issue #172320: [OpenReg] Extend device infrastructure tests (device, event, streams)",needs maintainer decision,28,needs maintainer decision,1,0,172320,,,,0,054e69f0cb5f1cb2c90fcb0159cb6ab6f1f1739ad77d53f3c5528eda5648625b
thread:issue:172322,issue #172322: [OpenReg] Extend test_rng.py and test_misc.py coverage,needs maintainer decision,28,needs maintainer decision,1,0,172322,,,,0,30f67cedbef8cf44c7318fb32640bbd55327a667c64c02d5bbd99f98d1187132
thread:issue:172338,issue #172338: Fix: Support is_causal=True with need_weights=True in Multi-Head Attention,needs maintainer decision,28,needs maintainer decision,1,0,172338,,,,0,001d499e31bc5cd5090073e55a09a9a9ccfa961bd8d6812560465ce5016cab4c
thread:issue:172351,issue #172351: [RFC][Release-2.11] Dropping Volta support from CUDA-12.8,needs maintainer decision,28,needs maintainer decision,1,0,172351,,,,0,97c2bcd3402821a9311394479cf9193618ca7db31174ed4d870b57406ca4b168
thread:issue:172383,issue #172383: torch.func.linearize produces NaNs with autocast + LayerNorm (jvp works),needs maintainer decision,28,needs maintainer decision,1,0,172383,,,,0,46f17c39d970feba9bfed1fc003b1e4b09bbb80a45fae53c8748883eba36546e
thread:issue:172427,issue #172427: [CI] Permission denied while trying to connect to the docker API,needs maintainer decision,28,needs maintainer decision,1,0,172427,,,,0,19a81b7bb018763ca530c8a0c9da3b3eab36028175aacd5e00c6e0461ea83faa
thread:issue:172475,issue #172475: [Inductor] Track deterministic `alloc_id` assignment across cached compilation artifacts,needs maintainer decision,28,needs maintainer decision,1,0,172475,,,,0,3e330553cc33f16d903d5fcb477862a977f01a3637377f1179f415cb70fd6bca
thread:issue:172485,issue #172485: [DTensor] Summary of issues related to `Partial` placement,needs maintainer decision,28,needs maintainer decision,1,0,172485,,,,0,1a6bfa3730a171899a1a938c417eb92f96c6a02b31c31c3b22c8eb11dba1da84
thread:issue:172491,issue #172491: Direct access to DTensor._local_tensor breaks _dynamo_graph_capture_for_export,needs maintainer decision,28,needs maintainer decision,1,0,172491,,,,0,d474cb6fd3cc3fd5c0b285cb1fc2a507d32bf4ebf26c864f40f74f78b71a19e8
thread:issue:172495,issue #172495: Add T5Gemma2 to TorchInductor benchmark HF suite,needs maintainer decision,28,needs maintainer decision,1,0,172495,,,,0,f2778e9ba3885af265a79f18cf6239a22a35a6a3ddbba2e7d8de075ffbc0f67d
thread:issue:172522,issue #172522: Support out-of-graph mutations with side streams,needs maintainer decision,28,needs maintainer decision,1,0,172522,,,,0,4f42703a42d40697573d46a32082fbb66bae7f8f181ae55e16b09f562279a3f9
thread:issue:172549,"issue #172549: AOTAutograd error when compiled graph has two DTensor outputs that alias each other, one differentiable one not",needs maintainer decision,28,needs maintainer decision,1,0,172549,,,,0,64fb93497b09b6d2450c4ac52a19ca892c6ae587e47e3aec1b9a760a5368488f
thread:issue:172623,issue #172623: Nested Tensors should have their own type,needs maintainer decision,28,needs maintainer decision,1,0,172623,,,,0,15387f2c3a9cd98153af2cf12006e2912bb5136b69c716e29f13a8219eade959
thread:issue:172797,issue #172797: autograd.Function.apply yields surprising strides,needs maintainer decision,28,needs maintainer decision,1,0,172797,,,,0,6f114f8d8686a153a42e9c2c44c1d09aeab272a449fc629900c7a361c2622ac8
thread:issue:172803,issue #172803: [OpenReg] PrivateUse1 backend missing distributed backend registration causes KeyError,needs maintainer decision,28,needs maintainer decision,1,0,172803,,,,0,cb2bea3f4f67dfa8e24297a074c85cee2d3623ab1dc3f7e7b3553424f4365db0
thread:issue:172807,"issue #172807: [CUDA] Auto-detected arch flags omit accelerated suffix (sm_120a → sm_120), breaking Blackwell NVFP4/block-scaled MMA",needs maintainer decision,28,needs maintainer decision,1,0,172807,,,,0,8f70edb85b1685841da466884316db34a3add4bb64e9223efd22272e17069234
thread:issue:172854,issue #172854: [test] Add unittest for nanmean output dtype validation (follow-up to #131043),needs maintainer decision,28,needs maintainer decision,1,0,172854,,,,0,0b1a43aa42afcdd07b92568ec287226d31fccb08e72a0174662798fc0c1d293e
thread:issue:172886,issue #172886: libtorch_cpu.so have unresolved symbols,needs maintainer decision,28,needs maintainer decision,1,0,172886,,,,0,254a43983fa267b0b726949c1bf27d402dd1ccb80e48d8235e237a4120c9eda5
thread:issue:173071,issue #173071: Allow privateuse1 devices in FlexAttention device validation,needs maintainer decision,28,needs maintainer decision,1,0,173071,,,,0,48687cf6bc9513e3c41d90174f1b0c81ea97c7c05c701da934f7abeb72a0080b
thread:issue:173104,issue #173104: Crash with `CUDACachingAllocator.cpp` when running Nvidia `nemotron-parse` at load,needs maintainer decision,28,needs maintainer decision,1,0,173104,,,,0,8e83641ce7092ac3b15a0fb92a5349497dd807e3ad14d76149d708b9297c7d8f
thread:issue:173123,"issue #173123: [DTensor] Gradient of unused output becomes plain torch.Tensor in backward, causing type mismatch with torch.compile",needs maintainer decision,28,needs maintainer decision,1,0,173123,,,,0,db412d70e0d2b96d9e05c098378e6ec3622fdebc418dda25a90b4ace1687e2df
thread:issue:173160,issue #173160: Feature Request: load_state_dict should optionally support partial tensor filling in non-strict mode,needs maintainer decision,28,needs maintainer decision,1,0,173160,,,,0,af4a6ad57926136ad6805ec1dee88810c1b2b050ac133f964bde0947d5405429
thread:issue:173255,issue #173255: Torch.compile is decreasing training speed instead increasing it,needs maintainer decision,28,needs maintainer decision,1,0,173255,,,,0,3e28e2f1aca885d8f15c40e06ef646df4809162ffc5432b639b9078e23f43b60
thread:issue:173290,issue #173290: Potential issue with lt_two in DistanceKernel.cu,needs maintainer decision,28,needs maintainer decision,1,0,173290,,,,0,04fb3ceb57e492540caaa20aa1133356bbffac17aca9818af220b8a7fe5b4fef
thread:issue:173369,issue #173369: Exporting ONNX model with captum heatmaps generation,needs maintainer decision,28,needs maintainer decision,1,0,173369,,,,0,57c122ffb5f837c25cafa6ccf86f4e8e25a63894681a1f4498fe5c59cc38d1f5
thread:issue:173452,issue #173452: Feature request: Support forward hooks with torch.compile fullgraph=True for capturing intermediate activations,needs maintainer decision,28,needs maintainer decision,1,0,173452,,,,0,6570f6b832168276f93e818a7d353e8ac86056fbf805a840b3e315f840e9c030
thread:issue:173516,issue #173516: [CUDA] MaxPool2d CUDA kernel lacks int64 support → one-side error,needs maintainer decision,28,needs maintainer decision,1,0,173516,,,,0,a4c967f052723aac9141a92e65477e36761cfbaf78ab20e6396ab5aac9c5cfc6
thread:issue:173525,issue #173525: [MPS] layer_norm backward numerical issues,needs maintainer decision,28,needs maintainer decision,1,0,173525,,,,0,ebe675140f0c5a0d019f1ddd28703ec48c9e55474e0aef9fb633f99db81595b4
thread:issue:173608,issue #173608: The program is stuck at function torch.distributed.new_group when parameter use_local_synchronization is True and len(_world.pg_names) is different between NPUs,needs maintainer decision,28,needs maintainer decision,1,0,173608,,,,0,f0b831dcaf53d6572f48f0b2037365f4e6e66da730d1ae837035bad81a068e77
thread:issue:173621,issue #173621: Feasibility of Persistent Caching for Dynamo Frontend,needs maintainer decision,28,needs maintainer decision,1,0,173621,,,,0,147a2c07c237623a66b1a8f8574a0d269e536baaf670994af69f9a31a1c1ab66
thread:issue:173638,issue #173638: `torch.linalg.slogdet` does not propagate `NaN` in CUDA,needs maintainer decision,28,needs maintainer decision,1,0,173638,,,,0,71a59a7861ee7605f9333ff479325d2dfa084ade6c6a199c9ea033a130d753de
thread:issue:173642,issue #173642: [Feature Request][torch.compile] Improve UX when autotuning repeated layer,needs maintainer decision,28,needs maintainer decision,1,0,173642,,,,0,87613c70ba04c96b873679be9104eb2ef38f7d885721a2f90d9e83c0fb7eccc8
thread:issue:173661,issue #173661: [Perfomance issue] Dataloader has unexpected overhead,needs maintainer decision,28,needs maintainer decision,1,0,173661,,,,0,0f0e1203746fd7649a64ea4e6de30524797cd0d035bffea4d5f3585a9f37796e
thread:issue:173729,issue #173729: Way to request a non-aliased torch.cuda.Stream,needs maintainer decision,28,needs maintainer decision,1,0,173729,,,,0,0a395f80c7e051b11ff8bf56b7c805a2dcbc5cb68314f8e5f1f62a4648267b58
thread:issue:173789,issue #173789: [Devicemesh] Bug: `child_to_root_mapping` overwritten due to hash collision in DeviceMesh submeshes,needs maintainer decision,28,needs maintainer decision,1,0,173789,,,,0,992b6956db2d3d8e3b82962083a5370422e01ba8f1658f813233e5821eab5dbf
thread:issue:173793,issue #173793: Numerical stability discrepancy in torch.compile: LayerNorm produces NaN for large inputs (~1e37) while Eager mode is stable,needs maintainer decision,28,needs maintainer decision,1,0,173793,,,,0,7fafb2f2dd24235cfd0cfad7ce6efc9c97052e0d604939b1d87fb3dca6e97e83
thread:issue:173796,issue #173796: `torch.nn.functional.PoissonNLLLoss` returns `NaN` imaginary part on CPU (vs 0 on GPU) for overflowing complex inputs,needs maintainer decision,28,needs maintainer decision,1,0,173796,,,,0,f17f64f178ef689faeed4ac4cab0b76c926db20430190d69ae4e332fa0c36450
thread:issue:173802,issue #173802: Batching rule of aten::_weight_norm,needs maintainer decision,28,needs maintainer decision,1,0,173802,,,,0,b6229d9760a86ce545357721f6e7346b443a3394596930d1489885139f3204d4
thread:issue:173885,issue #173885: [Inductor][CPU][float16] LayerNorm outputs NaN when input contains Inf,needs maintainer decision,28,needs maintainer decision,1,0,173885,,,,0,fec806e92ca2fe7aaaf03f61cc92cb3ebdbd469b50a8e59aa46718af5178db89
thread:issue:173897,issue #173897: XPU dist.all_reduce is async: torch.xpu.synchronize does not wait for collective completion,needs maintainer decision,28,needs maintainer decision,1,0,173897,,,,0,eabbd0effd561018bf9600a684fe16a721dc3638e0d9871c17115ef0a46211f8
thread:issue:173904,issue #173904: Unexpected cuda memory usage when doing 4d+ matmul,needs maintainer decision,28,needs maintainer decision,1,0,173904,,,,0,abf330e551dabf37a3a93edf907737edd4b5e1840954e863f8aeb9db9cf42b08
thread:issue:173950,issue #173950: [CI] enabling CUDA 13.0 periodic test,needs maintainer decision,28,needs maintainer decision,1,0,173950,,,,0,0d8296a93a0c5fdee290f317e37dbd8202b50808a102dea464974190a1736cb0
thread:issue:174045,issue #174045: Suboptimal `affine_grid` implementation,needs maintainer decision,28,needs maintainer decision,1,0,174045,,,,0,9816d5b46b7b084732a3b589f5366f0c04a3478a7bc1b170a033aab72ed9b45c
thread:issue:174066,issue #174066: How to prevent operator fusion at specific points in TorchInductor?,needs maintainer decision,28,needs maintainer decision,1,0,174066,,,,0,acdcc9f525993c917afc3880c0c6ccce09542d0145e96ced963886bef79bef84
thread:issue:174088,issue #174088: Gold doesn't work with GB200,needs maintainer decision,28,needs maintainer decision,1,0,174088,,,,0,427eb736845f54b8dcd2b1e51be63efa9fd50931d81b2eaa5362a98082bae046
thread:issue:174131,issue #174131: turn on autograd_cache_normalize_inputs in fbcode,needs maintainer decision,28,needs maintainer decision,1,0,174131,,,,0,684bd88d804cb8fde8d32d6537e3772141f87f0e4d6cdf06b8d8b56846b603e9
thread:issue:174165,issue #174165: Fallback to Composite Implicit Autograd for Purely Functional Composite Custom Ops,needs maintainer decision,28,needs maintainer decision,1,0,174165,,,,0,5ba9dca7aac5b4ad83c2a4d9b97b8f4ef92e0f97e1f30f5c4e464208373f20c3
thread:issue:174220,issue #174220: Dr CI produces confusing issue attribution report,needs maintainer decision,28,needs maintainer decision,1,0,174220,,,,0,4ed07f1d21fba5e1523c5361a0995aeee8ec4e6ee75a59ac67e66758db95cc5c
thread:issue:174227,issue #174227: [RFC] Support physical memory accounting for VMM-based pluggable allocators,needs maintainer decision,28,needs maintainer decision,1,0,174227,,,,0,6bf7f1f3a42835bced160f00e35255418622e4f99076fb1313a11f60a1705909
thread:issue:174266,issue #174266: BatchNorm1d CPU outputs differ across devices/threads with identical inputs,needs maintainer decision,28,needs maintainer decision,1,0,174266,,,,0,36ea3668a8b18807e7308f5cf33def0a1ced022d023beaf8e0823c219b0e5f54
thread:issue:174313,issue #174313: [release 2.11][triton] trunk / linux-jammy-rocm-py3.10 / test failures,needs maintainer decision,28,needs maintainer decision,1,0,174313,,,,0,2d2fbdb64da303299bb30052d4122662ad4069d87a799223ff06538ae323d8b5
thread:issue:174355,issue #174355: Forward AD not implemented for GRU cell when using CUDA,needs maintainer decision,28,needs maintainer decision,1,0,174355,,,,0,bdeed6f9ff84dfab0eb4a0e2c8809eaa3c57129d22579c73588acf0b14bd5106
thread:issue:174359,issue #174359: The OpenReg runtime is stuck,needs maintainer decision,28,needs maintainer decision,1,0,174359,,,,0,b8379ecc36ee5f9357567ae8b448e20e76c58c20fab9aa062c7ac3bf24e2c1d4
thread:issue:174373,issue #174373: Increase NUM_THREADS that embedded ARM OpenBLAS is compiled with,needs maintainer decision,28,needs maintainer decision,1,0,174373,,,,0,7ba40656c0af8ae795a5a17272c2caf56bedced3d662590a0d0196d010541722
thread:issue:174417,issue #174417: [release 2.11] [triton] test/inductor/test_loop_ordering.py::LoopOrderingTest::test_fp8_cast_and_t,needs maintainer decision,28,needs maintainer decision,1,0,174417,,,,0,b5b1a4afe73a8eeb47eb59fa07d0c0405e4cf1ceb8cd5729652cf099b5958fe7
thread:issue:174507,issue #174507: PyTorch Stable ABI support,needs maintainer decision,28,needs maintainer decision,1,0,174507,,,,0,648bfc04f7fff199f9a6f7b605e543724a58a1c45ea6d8ab5775d5b07a9f59b1
thread:issue:174602,issue #174602: [nondeterminism][inconsistency] ctc_loss sometimes return incorrect results,needs maintainer decision,28,needs maintainer decision,1,0,174602,,,,0,d28b624947cc8f72a6f4536af1f08020a8820ac72066372b1ca5e11cbac3b60c
thread:issue:174711,issue #174711: add an alternative to compile time benchmarking in inductor for determinism that is reasonably performant,needs maintainer decision,28,needs maintainer decision,1,0,174711,,,,0,2da0ee9c03bb5eaa2a1dda2e5003676151bb573f8cc57a39ba80411ad3222e63
thread:issue:174763,"issue #174763: torch._grouped_mm is not autocast-compatible, causes dtype mismatch",needs maintainer decision,28,needs maintainer decision,1,0,174763,,,,0,0c5a793e03f02b1e82b3136364073950cd679fb6e04c3e101dbf5a9d4a3e6e7d
thread:issue:174792,issue #174792: Provide a way to systematically run all the autotune configs provided by torch.compile,needs maintainer decision,28,needs maintainer decision,1,0,174792,,,,0,422ccd0e872974e0c6487a00ebf2f6437ca0423afc6144c0c617102941bf852f
thread:issue:174930,issue #174930: UNSTABLE inductor / inductor-test-cuda13 / test (inductor_torchbench),needs maintainer decision,28,needs maintainer decision,1,0,174930,,,,0,60e8e4bbcbb5f62ec8c975bb033627828f5ffc46125492829d75374f0b80103d
thread:issue:174949,issue #174949: [vllm] CUBLAS_STATUS_INVALID_VALUE in cublasGemmEx after upgrading to PyTorch 2.10,needs maintainer decision,28,needs maintainer decision,1,0,174949,,,,0,e45c006271d6134380bd29ad4adca7340b05b58d886d7a04bb89b6fe4780ef7a
thread:issue:174970,issue #174970: [RFC] Hooks to trigger recompilation,needs maintainer decision,28,needs maintainer decision,1,0,174970,,,,0,0a268423a6fba75f137ec8c27d06cf34c7423eade6f1b4e3af9e8c2ac8d4e4f0
thread:issue:174983,issue #174983: Make `export CUDA_LOG_FILE=stdout` less noisy with spurious errors.,needs maintainer decision,28,needs maintainer decision,1,0,174983,,,,0,1c425f419565cd0ceb5011e341e16e50f6a7f7de239ec0d1d9c2ff4e18d3e039
thread:issue:175022,issue #175022: [RFC] Cross-Repository CI Relay for PyTorch Out-of-Tree Backends,needs maintainer decision,28,needs maintainer decision,1,0,175022,,,,0,f07be33e6fdf7d81ce21958d7eef9f08a85acbbe6c333c4c2233ed7ea3d87443
thread:issue:175032,issue #175032: Dead code: reduction of post AI tech debt,needs maintainer decision,28,needs maintainer decision,1,0,175032,,,,0,df1d9ac9d17adf1be906b8d376ba7ce73609795d0d8aaa3124056e717fe49f35
thread:issue:175193,issue #175193: ZLib Reference outdated in riscv ci dockerfile,needs maintainer decision,28,needs maintainer decision,1,0,175193,,,,0,0a89c4df49aad4f3e7a3be718002c29817ae31bb826a3a81c2a23bde1de84a27
thread:issue:175229,issue #175229: Idiomatic way of matching against flex attention regions in SAC policy with wrap_inductor_compiled_regions,needs maintainer decision,28,needs maintainer decision,1,0,175229,,,,0,66719b2af8b0c7e4dc508fd272376492af2092e2a0f94e93506518933eb617aa
thread:issue:175252,issue #175252: Slow reading tensor meta-data,needs maintainer decision,28,needs maintainer decision,1,0,175252,,,,0,e92e95a8280c8463ba6edcd11d17415feb94236d23305d5e56f7713b2c88cff4
thread:issue:175269,issue #175269: [dynamo] Refactor variables/builtin.py,needs maintainer decision,28,needs maintainer decision,1,0,175269,,,,0,81819f683e042c8456e9932f18d681a6acf0bfa84b6acb800fffb51c6f9462bd
thread:issue:175306,issue #175306: SAC should treat inductor_compiled_code and other HOPs specially (pull out a more unique name) to prevent mixups,needs maintainer decision,28,needs maintainer decision,1,0,175306,,,,0,3a71eb067106cca846549cbffbd4909b96e4107aadfcfd95bcdf6942ddd87eff
thread:issue:175311,issue #175311: [RFC][Inductor] PyTorch Inductor cuTile Backend,needs maintainer decision,28,needs maintainer decision,1,0,175311,,,,0,502ca3d0d37dcd080761eca01b47789ced4fb9634b2871e8fabba440f49be6ce
thread:issue:175372,issue #175372: python property setter ignored when assigning nn.Parameter() to nn.Module,needs maintainer decision,28,needs maintainer decision,1,0,175372,,,,0,9428e1309bd18f9b1bb99036e52bbfd03ff4e3ffcd704ca201b70f7a9dece828
thread:issue:175409,issue #175409: Choose rounding mode when casting to dtype,needs maintainer decision,28,needs maintainer decision,1,0,175409,,,,0,8adcb2644cecff18a20aad305fe1cf4503dd05feaf10e4a8a4ca53328e6087ea
thread:issue:175478,issue #175478: torch.distributions.Gamma().sample() should take a torch.Generator,needs maintainer decision,28,needs maintainer decision,1,0,175478,,,,0,601c3b5efd078d494fad8f24c8322f65a3eb0584c9cef383f75308f729a9689e
thread:issue:175479,issue #175479: Feedback about LSTM,needs maintainer decision,28,needs maintainer decision,1,0,175479,,,,0,6aa8c9a3f80c56f82c9ed0cfc0086b9795a8a4892821b424ea8cb9123ddec07a
thread:issue:175484,issue #175484: Unable to disable warning about tf32 precision,needs maintainer decision,28,needs maintainer decision,1,0,175484,,,,0,b77e6664a3afd4b6b0561e7037bda0ad72d26536ed839d275dea1507464ec2c4
thread:issue:175538,issue #175538: [Inductor] Need a CustomIRPass Abstraction for Loop‑Level IR Transformations,needs maintainer decision,28,needs maintainer decision,1,0,175538,,,,0,e0f03c79db210b9f4f0d921e27346ef19a7bd55eb05f85bc2ca16b16d784a83b
thread:issue:175636,issue #175636: `[feature request][discussion][linalg]` Polar decomposition,needs maintainer decision,28,needs maintainer decision,1,0,175636,,,,0,a51d2b20240dc0bde78c18520253ca1bdd6c9abe8fb3472b21b58bd5e305140f
thread:issue:175737,issue #175737: GPUDirectStorage Status?,needs maintainer decision,28,needs maintainer decision,1,0,175737,,,,0,853bbd36b8b1aafc0e880678776c4ccb0d721271fc601c3a3de43447d6f7374c
thread:issue:175738,"issue #175738: torch.export + torch.onnx.export(dynamo=True) gives incorrect results for scatter_reduce_(reduce=""mean"")",needs maintainer decision,28,needs maintainer decision,1,0,175738,,,,0,4f84539e4b39bdc6de240e51e1084341f007294f4320629038eaec2f2691c83d
thread:issue:175829,issue #175829: Why the same test is running twice in regular and slow shard,needs maintainer decision,28,needs maintainer decision,1,0,175829,,,,0,2b0aea2324225c827571b3b25f69dd67ad3ffcad6453e50ac941bc397904ea48
thread:issue:175850,issue #175850: RFC: Vmap Extension Points for Custom Container Types,needs maintainer decision,28,needs maintainer decision,1,0,175850,,,,0,935b16de386a10a11130be3755bdf95073e12dea9c8431aca5f050b2fecc27fa
thread:issue:175869,issue #175869: [dynamo] Tracking issue for itertools polyfills,needs maintainer decision,28,needs maintainer decision,1,0,175869,,,,0,bf1a1fef55623f791c40fbdfb3d7a2031d985d7e85451b636c56369eabd920de
thread:issue:175899,issue #175899: [CI][VLLM] Track disabled vLLM tests in PyTorch CI,needs maintainer decision,28,needs maintainer decision,1,0,175899,,,,0,689457d5aa04ee6322f0e907946d4b21bd41f702485cb88064a3ea2fc9a4c4ad
thread:issue:176065,issue #176065: Errors in Activation Checkpointing tutorial,needs maintainer decision,28,needs maintainer decision,1,0,176065,,,,0,aa1fd23067e7bc8044b1431778044240992d5e60a0d4aa220f0c381f094b8cd7
thread:issue:176078,issue #176078: custom_op with mutable optional arguments,needs maintainer decision,28,needs maintainer decision,1,0,176078,,,,0,07cb6b00a54ecd5dfafc3b633b6f836532faf1c9e74bea4228c8440d173e21aa
thread:issue:176091,issue #176091: Windows CPU: SDPA FlashAttention gives wrong results with multiple intra-op threads (mixed OpenMP runtimes),needs maintainer decision,28,needs maintainer decision,1,0,176091,,,,0,fbd58a9b531429cd4aa4dde963317f98303a716bb9567e5a7ebcd6467432d83a
thread:issue:176153,issue #176153: Libtorch - make call reset_parameters during init optional,needs maintainer decision,28,needs maintainer decision,1,0,176153,,,,0,c02744722c08f277a91f0ffbfb5b55d2b1ab99104502ea6a11f5179908fb6b6a
thread:issue:176166,issue #176166: DataLoader: collate_fn leads to high RAM usage,needs maintainer decision,28,needs maintainer decision,1,0,176166,,,,0,c2a871a0eec22594ccd7e1b8fc5f7186c761a45416a432dfddab2068a6ce8b7c
thread:issue:176188,issue #176188: test_structured_trace.py's test_ddp_graphs and test_graph_breaks flaky on all platforms,needs maintainer decision,28,needs maintainer decision,1,0,176188,,,,0,cc060ea0d8faa3a1a3fc8b984f2bf8e553906a8da224a2848711ae530140286d
thread:issue:176274,issue #176274: PyTorch + Math libraries: redundant `copy_` calls when working with buffer views,needs maintainer decision,28,needs maintainer decision,1,0,176274,,,,0,f36b42dbc9117d4e06fb4d46269271e3f54e3c1c15f70ca273353d53ad8c3614
thread:issue:176384,issue #176384: Output difference anomaly under equivalent migration in erfinv operator,needs maintainer decision,28,needs maintainer decision,1,0,176384,,,,0,9553a451a8bbf17dd7e3dce7d2866304f38ffa0e94b22f50e30438107c66bbd3
thread:issue:176385,issue #176385: Inconsistent cross-framework behavior (nan vs +/-inf) for float64 out-of-domain inputs,needs maintainer decision,28,needs maintainer decision,1,0,176385,,,,0,f2bfe6957227539b8d3f1bdfd9b1d42ddf5eac4a0f9966a3f45cff13944b7cd6
thread:issue:176412,issue #176412: Output difference on 2D float32 tensors with negative out-of-domain values,needs maintainer decision,28,needs maintainer decision,1,0,176412,,,,0,53f169899b32d183a1e71bb3ec59b9909fc948c810ec277d9674d2b23ab02370
thread:issue:176413,issue #176413: Output difference anomaly under equivalent migration in logdet operator,needs maintainer decision,28,needs maintainer decision,1,0,176413,,,,0,3d241bb3f6fbf3cd19f9d5885991a79e4b2a4a477c9b8719dc3ad220f7da9ef9
thread:issue:176415,issue #176415: Output difference anomaly under equivalent migration in logdet operator,needs maintainer decision,28,needs maintainer decision,1,0,176415,,,,0,6e84ecd795ed5329d34554eef297fe72d3b4b00d6d72e96f4e43514c5f803edf
thread:issue:176507,issue #176507: Security hardening: Add HMAC integrity verification and authentication support to torch.compile Redis remote cache,needs maintainer decision,28,needs maintainer decision,1,0,176507,,,,0,28149ce08228397a8320e4f705193afe74b65ddd189ca3575187cedf4b7d37e6
thread:issue:176611,issue #176611: [cudagraph][torch.compile] `cudagraph_or_error` can be silently ignored even when the cudagraph is not captured.,needs maintainer decision,28,needs maintainer decision,1,0,176611,,,,0,0c3f7b7212d4e3ba6aa64d29b148312d2dff14c60c30b93729cc9635736516d5
thread:issue:176625,issue #176625: The `padding` type should be set as Scaler for torch.nested.to_padded_tensor (currently set to double),needs maintainer decision,28,needs maintainer decision,1,0,176625,,,,0,beeeed954c82788e5db5add5212b549046e10440f4f597ed4a440f544a6d0736
thread:issue:176628,"issue #176628: [NN] Performance improvement in `nn.Module._call_impl()` by around 9% by optimizing fast-path for `forward_call()`, hook iterations and checks plus some minor changes",needs maintainer decision,28,needs maintainer decision,1,0,176628,,,,0,37ba5c1be993f82c6c560622b36efd88b10b367e35702dc68fdba9c4f4c083ea
thread:issue:176687,issue #176687: Proposal: Pytorch agent harness for external contributors,needs maintainer decision,28,needs maintainer decision,1,0,176687,,,,0,2f5589d11de2f3daa86dd6f66c2b1a0f212c1e991c01b0cbf50796d99fc7541b
thread:issue:176692,issue #176692: Dynamo does not know how to trace the builtin `<unknown module>.TensorBase._make_wrapper_subclass.`,needs maintainer decision,28,needs maintainer decision,1,0,176692,,,,0,e6e0530bdf3890ec87dd35b74d96e77f691c0bdc03b1c3f50418e4594edbbd20
thread:issue:176966,issue #176966: DISABLED test_trace_return_namedtuple (__main__.TestFX),needs maintainer decision,28,needs maintainer decision,1,0,176966,,,,0,00e138340e2519314dba271e332d17563f952b6062956e3fb2697f8a96be7b0e
thread:issue:177014,issue #177014: `cudaErrorInvalidDeviceFunction` error due to use of `__CUDA_ARCH__` in host code,needs maintainer decision,28,needs maintainer decision,1,0,177014,,,,0,96c0e2cf6dd8a97deab13c63656993634b90e7009c672ec1781f5c83a1f602dc
thread:issue:177021,issue #177021: linear attention / mamba2 OP,needs maintainer decision,28,needs maintainer decision,1,0,177021,,,,0,3dcc71813709e2d413e5aaa3c1c71ef88b642574792db1786c84d8b463441c49
thread:issue:177050,issue #177050: PyTorch 2.10-2.11 Linux x86 Binary size increase investigation,needs maintainer decision,28,needs maintainer decision,1,0,177050,,,,0,b3d47480f3a41bf65e01e3f93827034e92939cee561469c368b2c212fc5367bd
thread:issue:177053,issue #177053: [inductor] realize_reads_threshold could be tuned differently with auto-chunking,needs maintainer decision,28,needs maintainer decision,1,0,177053,,,,0,99ca5d75bb51f843efbcb398a00d241e2d56a008ef5b4fc4331def7781576f75
thread:issue:177135,issue #177135: release notes commit script should respect user-specified labels,needs maintainer decision,28,needs maintainer decision,1,0,177135,,,,0,2fb5ba8e1699ecf00c0340882d074b25a3adb9e8e93fc02ca9dcddedde0077e2
thread:issue:177225,issue #177225: mps test got Exception: mean(): could not infer output dtype.,needs maintainer decision,28,needs maintainer decision,1,0,177225,,,,0,d8cdbe58b31f4ce234ddf59fa8d0e35668594e626a25170e03422bf08e304dc7
thread:issue:177245,issue #177245: AArch64 Unit Test Failures - Multiple failures in test/nn/test_convolution.py TestConvolutionNNDeviceTypeCPU for oneDNN,needs maintainer decision,28,needs maintainer decision,1,0,177245,,,,0,b2e073ad71a46a111a580de8a778edc78696c3bbf928c4a4c22af94edff1a17f
thread:issue:177260,issue #177260: Standardize device-agnostic test cases via instantiate_device_type_tests,needs maintainer decision,28,needs maintainer decision,1,0,177260,,,,0,f337e8fdd66dad87ab009eef54ca5a85771d5b1db720d6c5e04ad1d3d36b3337
thread:issue:177268,issue #177268: [OpenReg] Add testing operator framework usage examples,needs maintainer decision,28,needs maintainer decision,1,0,177268,,,,0,ffb1b368b19a9fcac9069aeadd456616ae105e441f9655566a651ff6dac35e36
thread:issue:177277,issue #177277: Conv3D does not preserve dim_order for some inputs,needs maintainer decision,28,needs maintainer decision,1,0,177277,,,,0,f869f260582beff0f6efcbcf25bfb8ba1887b990e3aca9656392e60ca7e2696d
thread:issue:177322,issue #177322: Support TORCH_SHOW_CPP_STACKTRACES=1 on Windows,needs maintainer decision,28,needs maintainer decision,1,0,177322,,,,0,c678388085e9096aab8228e5cef23455ae83b7b42fee9bb4e5258ab67f9fc042
thread:issue:177372,issue #177372: Add torch.narrow_scatter operator to complete the *_copy/*_scatter operator family.,needs maintainer decision,28,needs maintainer decision,1,0,177372,,,,0,e5f2bb61d77ec8376c9f4ff464f2fc041d7ccc44e8492afd824c1b9fcacbf3a9
thread:issue:177386,issue #177386: `CatBackward0` rejects valid gradient after padded round-trip on jagged nested tensors,needs maintainer decision,28,needs maintainer decision,1,0,177386,,,,0,9321cd2045ffcf3afc929ee9d5a5726e9f83d8ba909b6a768379291514ab5ea5
thread:issue:177393,issue #177393: [release 2.11][vllm] Language Models Test (Extended Generation) test_models[True-False-5-32-bigcode/starcoder2-3b],needs maintainer decision,28,needs maintainer decision,1,0,177393,,,,0,a0a6ce935ac5f151e8fa6ca5ec3c9ad643f273b78016f206883be38bc322ebd8
thread:issue:177425,issue #177425: binary_populate_env.sh: PYTORCH_ROOT optional check is inconsistent with later required use,needs maintainer decision,28,needs maintainer decision,1,0,177425,,,,0,e7c682a2a975497d9a77493b59a6be4a91f1e2f57956a0c9a18e446131d0c881
thread:issue:177477,issue #177477: Misleading error message (INTERNAL ASSERT FAILED) in `jiterator._create_multi_output_jit_fn` when passing CPU tensors,needs maintainer decision,28,needs maintainer decision,1,0,177477,,,,0,4b25075ac29d08a214a173fcc04cb4e4b40af5f4104a93ee1b244d614184f70a
thread:issue:177520,"issue #177520: Memory management, memory not returned after CPU -> GPU",needs maintainer decision,28,needs maintainer decision,1,0,177520,,,,0,13a571cce9336003cd9d071433431d5761365787959c073c0a10cbc0078e0ac2
thread:issue:177548,issue #177548: Update Internal fbsource CK,needs maintainer decision,28,needs maintainer decision,1,0,177548,,,,0,2cc0ae88af25f34d180b6f6b08cbca54c3ebba17e712f90c742e7a53d5d5bd9d
thread:issue:177578,issue #177578: [Inductor][NVGEMM] Extend nvMatmulHeuristics config key tuple for stages/split_k,needs maintainer decision,28,needs maintainer decision,1,0,177578,,,,0,d8a3fd1f8f9badc886656f13f1b7a3f23d1675dae74a97e70f13488b27f68348
thread:issue:177599,issue #177599: arange CPU kernel has different precisions for scalar and vectorized executions,needs maintainer decision,28,needs maintainer decision,1,0,177599,,,,0,476d25d0801b69725c67b0cf5ca9cd34d4393ace7673037efd1ee253b999eed8
thread:issue:177654,issue #177654: `import functorch.dim` silently breaks `torch.Tensor.expand(size=...)`,needs maintainer decision,28,needs maintainer decision,1,0,177654,,,,0,6c091337f38b5e103acb8c92fde401123b3eeb618f125b221a8c2b56eb037f4e
thread:issue:177663,issue #177663: [DTensor] greedy redistribution cost favors all_gather over all_to_all,needs maintainer decision,28,needs maintainer decision,1,0,177663,,,,0,8c870c1d7264116938f0b4cf5f665353886282ab1894b39008e89292f16cfaa0
thread:issue:177719,issue #177719: [Discussion] Disable size_asserts by default in inductor for vllm x compile inference,needs maintainer decision,28,needs maintainer decision,1,0,177719,,,,0,ac1fa0c7d3dab13577c41b931d0a933c0be4fccdc0b2da3aea67c9b86ee3377c
thread:issue:177771,issue #177771: Accelerators: torch.Stream.record_event doesn't handle torch.cuda.Events even though torch.cuda.Event is a torch.Event subclass,needs maintainer decision,28,needs maintainer decision,1,0,177771,,,,0,7a9416714cffb113216f6d75aa2c0591a0da2cee6b7fd6d74ff2c0ee7742d00e
thread:issue:177830,issue #177830: Why register_replacement fails to support fuzzy matching for optional operator parameters in search_fn/search_fn_pattern,needs maintainer decision,28,needs maintainer decision,1,0,177830,,,,0,89929562e5998552e1154cfec0a9f840af5bc9f44be2f7491a80318a50846e48
thread:issue:177859,issue #177859: Combining `bfloat16` with complex scalars raises `NotImplementedError`,needs maintainer decision,28,needs maintainer decision,1,0,177859,,,,0,9231f8809e7bf847658b3fc34ed300c66a6eeb8c3987dbe1c9d6a59beb10b2b9
thread:issue:177896,issue #177896: Bot to notify PoC if ci pin updater fails,needs maintainer decision,28,needs maintainer decision,1,0,177896,,,,0,31c2bc9efcdc274417e37f26d2bab1c2d41ec49023a425e18778a8ad6d41f5ba
thread:issue:177910,issue #177910: nvshmem symm memory backend causes stuck sometimes for helion distributed kernel autotuning,needs maintainer decision,28,needs maintainer decision,1,0,177910,,,,0,07a2ed412b0e874345846fbba5ab84cd9ea06a26d5c8c539e24795399cf02a6a
thread:issue:177935,issue #177935: [RFC] Add A Cross-Backend Device Name API: `torch.accelerator.current_device_name()`,needs maintainer decision,28,needs maintainer decision,1,0,177935,,,,0,0165a529307662733c23911649220018f7889507ea688de06c50c92b006e8ff5
thread:issue:177957,issue #177957: api shows unexpected difference for flatten(): numpy vs. pytorch,needs maintainer decision,28,needs maintainer decision,1,0,177957,,,,0,78d2ab194716ad39c772dbd3ee06663cce327a3375c6ee9da99fea11a611b641
thread:issue:177970,issue #177970: `TEST_DEVICE_BIAS` linter scope should be expanded,needs maintainer decision,28,needs maintainer decision,1,0,177970,,,,0,eabe8f1353c348cfff504514937c056deef4618110114ada3c6a55532136c3cf
thread:issue:178052,issue #178052: Deterministic implementation of CTC Loss,needs maintainer decision,28,needs maintainer decision,1,0,178052,,,,0,558bd65bab45c336b54c950a6eff73d6a9a257ee54986b4f3333cf4cc0329b6b
thread:issue:178072,issue #178072: [PERFORMANCE] [NN] Change `torch._C._get_tracing_state()` to `torch._C._is_tracing()` for around 15% improved performance of `module.py` `_call_impl()`,needs maintainer decision,28,needs maintainer decision,1,0,178072,,,,0,eba2552e82135ef6b6108197afd513bdd4a0852eedc5701e6d8afc2ff46e84b7
thread:issue:178091,"issue #178091: propagate_single_input_strategy copies input tensor_meta to output, producing wrong dtype for ops like convert_element_type",needs maintainer decision,28,needs maintainer decision,1,0,178091,,,,0,b9294ab360b2c28a3d403135bceb5831c04149b0383e0d530a50076044555b23
thread:issue:178132,issue #178132: torch.compile streams support in AOTI,needs maintainer decision,28,needs maintainer decision,1,0,178132,,,,0,3e9fabfe3a66210cc43fa32bf1ab0d3b0983693a8866a201f73b9e5f4cec3db3
thread:issue:178134,issue #178134: torch.compile (CUDA fp32) vs eager gives large max relative error for Conv2d,needs maintainer decision,28,needs maintainer decision,1,0,178134,,,,0,63cc7cc5e05fa47b68d007b5e4d89a54810c93255db97443d8017dd138a0c07e
thread:issue:178196,issue #178196: libvnrtc error in torch nightly on cuda 13.0,needs maintainer decision,28,needs maintainer decision,1,0,178196,,,,0,0085fcf8fc4576be79878505610fc145e56747a7b145d9454c08123655cb8e1e
thread:issue:178345,issue #178345: [build env combo] elainewy,needs maintainer decision,28,needs maintainer decision,1,0,178345,,,,0,27b2a37d2d95454398240429c991d2e40f8a079d34737de63cd9a489f3746816
thread:issue:178368,issue #178368: On the future of triton-windows,needs maintainer decision,28,needs maintainer decision,1,0,178368,,,,0,e8f42a065159a9fd03fba3b1c4385a03ff51805f17f1380d2a3fa77b8ba64bd8
thread:issue:178472,issue #178472: conv1d on meta device path skips groups validation and returns an invalid output shape,needs maintainer decision,28,needs maintainer decision,1,0,178472,,,,0,9b542e1844774de04964bb7868d7f04250f7b56513a756e740bf91c777ccec10
thread:issue:178482,issue #178482: torch.compile(dynamic=True) ignores incompatible `out` tensor for `torch.empty` and returns wrong shape,needs maintainer decision,28,needs maintainer decision,1,0,178482,,,,0,a739b8794808f050e0c4a35331753403b3e8bacbef1da5de3514080dd139cf7e
thread:issue:178511,issue #178511: [torch.compile] allow_in_graph called inside compiled function causes Unsupported error,needs maintainer decision,28,needs maintainer decision,1,0,178511,,,,0,0fb1ceef7ff1ce6ac473e651ad2de189f56348f88029c90b3761a8ca0ef492c3
thread:issue:178600,issue #178600: RuntimeError of use_libuv and ReduceOp.AVG with DDP + Gloo on Windows multi-gpu train,needs maintainer decision,28,needs maintainer decision,1,0,178600,,,,0,f9250b89aaab79ed7c10e9b83f17aa9250bc77efd1721eef437953a4f78c1262
thread:issue:178614,issue #178614: bag boundry is not calculated correctly in EmbeddingBag when setting include_last_offset = True,needs maintainer decision,28,needs maintainer decision,1,0,178614,,,,0,148743f5544748403f38ffbf291396fdd780a399f7e3ab9f5cd4c1707f17d5ed
thread:issue:178616,issue #178616: [DTensor] reshape is dispatched to aten.view instead of aten.reshape,needs maintainer decision,28,needs maintainer decision,1,0,178616,,,,0,43855892afbdef1c143ae9404542d9489442cb35f31fad1149caa90e36859ebe
thread:issue:178749,issue #178749: [FSDP] Multi-GPU distributed tests crash with NCCL 2.29.7 NVLS multicast slot exhaustion on H200,needs maintainer decision,28,needs maintainer decision,1,0,178749,,,,0,3e0b61bdec744619abac8619d797c48dae1fc7c8ef1ef484d8478157ba9b0339
thread:issue:178820,issue #178820: Add Dynamo support for dist.record_comm (capture_profiler_record_comm),needs maintainer decision,28,needs maintainer decision,1,0,178820,,,,0,f3b6eca7a39f2a64c6efe2ffe779f9fdeb25d8868d2da81190abcfa66e0df37c
thread:issue:178839,issue #178839: [ROCm] MIOpen Gemm solver receives workspace_size=0 via legacy Find1 API,needs maintainer decision,28,needs maintainer decision,1,0,178839,,,,0,8faa273f73ea0caa3d9f8415574b773d39bb92db9267248e23a239afc18f4c9f
thread:issue:178864,issue #178864: torch.multinomial on CUDA with zero-sum probabilities triggers device-side assert instead of invalid multinomial distribution error,needs maintainer decision,28,needs maintainer decision,1,0,178864,,,,0,2a037307543be98d2ab12db1284181c7c1c75683defe4589e5e89a513a95d695
thread:issue:178883,issue #178883: `torch.compile` produces different output for model with `amax`/`amin` + partial reuse reduction pattern,needs maintainer decision,28,needs maintainer decision,1,0,178883,,,,0,bf01f62f0888deb8e37db943e1c3bfc5c3cb3c76d827ec673ea7311b9f25989f
thread:issue:178930,issue #178930: [torch.compile] Better way to override the default backend,needs maintainer decision,28,needs maintainer decision,1,0,178930,,,,0,1273e6e6541f7caefa80402b7cb754bd927400b4bbdfd3c8fa02bccee090e376
thread:issue:178988,issue #178988: Cannot build documentation (`RecursionError`),needs maintainer decision,28,needs maintainer decision,1,0,178988,,,,0,3915ec174515095944dbea20da4df718feae1ef053155a11c7ecb0a355e77cea
thread:issue:179124,issue #179124: grid_sampler_2d_backward_cuda is not deterministic when torch.use_deterministic_algorithms(True),needs maintainer decision,28,needs maintainer decision,1,0,179124,,,,0,5727c957abd4510d6319944c6996fc980ac7df60ea9794699e6d7139542d8f0a
thread:issue:179143,issue #179143: [RFC] AdamTR: Adam variant for deterministic Token-Routed architectures,needs maintainer decision,28,needs maintainer decision,1,0,179143,,,,0,0bc4be3752689648d3e6702df13a739f014b6ce0f17178accd2339a0fa7ccc47
thread:issue:179152,issue #179152: `linalg_cholesky_ex_out` contains some dubious logic,needs maintainer decision,28,needs maintainer decision,1,0,179152,,,,0,e6ebc34dd89e962501a50906875c9535ace857835abbf93e29017186e60598ea
thread:issue:179382,issue #179382: Support dropout in varlen_attn,needs maintainer decision,28,needs maintainer decision,1,0,179382,,,,0,00f7d80acb9c9861b0047b44c20d78b4c073837fe41a97c7672e3aac2e6f2554
thread:issue:179430,issue #179430: Add Frechet/Inverse Weibull distribution,needs maintainer decision,28,needs maintainer decision,1,0,179430,,,,0,f8fbacb7afba1a8792489f8dd01d1a6fba0b51ad8940c534cbf877c8e9bf32a7
thread:issue:179448,issue #179448: [DTensor] Inconsistent output placements of sharded DTensor indexing,needs maintainer decision,28,needs maintainer decision,1,0,179448,,,,0,a75b9b59c4d95718ead9172a6e8c40129aca87588a8f61f6202b7ac516be7c26
thread:issue:179520,issue #179520: Different results for with and without torch.compile on flux.1-dev model,needs maintainer decision,28,needs maintainer decision,1,0,179520,,,,0,3d30bd51c75642c766d69af68db975f639ad1125185e889acfbac75f4b945e03
thread:issue:179643,issue #179643: Call Hierarchy Metadata,needs maintainer decision,28,needs maintainer decision,1,0,179643,,,,0,5e07dc38634a7381bf4b1d46f09001cc2a281c912b692a1b7371e10e5671d370
thread:issue:179652,issue #179652: `torch.library.custom.register_autograd` incompatible with `jacrev`,needs maintainer decision,28,needs maintainer decision,1,0,179652,,,,0,d810da7a9c5644aa9b0c532946a3a36266c22b881f98c5492cf56bae0f535838
thread:issue:179707,issue #179707: Improve static type hints for dataloader with known `collate_fn`,needs maintainer decision,28,needs maintainer decision,1,0,179707,,,,0,a6c78d6de0e8459fe3df59192079ae07b121d1a400b124323ba4e983a9237db0
thread:issue:179709,issue #179709: Cannot export model with CenterCrop to ONNX,needs maintainer decision,28,needs maintainer decision,1,0,179709,,,,0,e799ba15074ccd941130e5b3d0df45932ad5f577babcfa9fa71814312257ed8a
thread:issue:179710,issue #179710: AArch64 unit test failure TestFX.test_shape_prop_layout_3d,needs maintainer decision,28,needs maintainer decision,1,0,179710,,,,0,a241f69691d70e4825b7df52642b85051c982cca2a1afaf47c75b652ec293a15
thread:issue:179742,issue #179742: DISABLED test_grad_freevar_python_scalar (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),needs maintainer decision,28,needs maintainer decision,1,0,179742,,,,0,39fdf357f7dfcb7f2a4f4b602c0b828c8b1b4b763a3f1ea69d50f3f4ce829c52
thread:issue:179745,issue #179745: DISABLED test_memory_snapshot (__main__.TestCudaAllocator),needs maintainer decision,28,needs maintainer decision,1,0,179745,,,,0,77616c017588be78313a8c3cd24de393f7faaae6add7c1ca956ef5db095dfe13
thread:issue:179784,"issue #179784: torch.xlogy(0, 0) returns NaN, while in many numerical computing libraries (e.g., SciPy) and common machine learning practice, it is defined as 0.",needs maintainer decision,28,needs maintainer decision,1,0,179784,,,,0,3271e0b2a4c373ba0a72fde22fa3e02627cbbabbacc5339f93414204ea147a2f
thread:issue:179798,issue #179798: DISABLED test_memory_plots_free_segment_stack (__main__.TestCudaAllocator),needs maintainer decision,28,needs maintainer decision,1,0,179798,,,,0,e82ee2c63acea8b71b0db453dfe4d61cff3220e7bcccc04b4f97ab43dac6461f
thread:issue:179799,issue #179799: DISABLED test_grad_has_aux (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),needs maintainer decision,28,needs maintainer decision,1,0,179799,,,,0,ccebc972a636114429f557d39414b0d355be931a77444ac38a6811016057cd43
thread:issue:179811,issue #179811: DISABLED test_hessian (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),needs maintainer decision,28,needs maintainer decision,1,0,179811,,,,0,997d74f0930243ac55e4c14dee307cb73dee8b481b12a4acf3193b7b5ec6f827
thread:issue:179853,issue #179853: DISABLED test_hessian_argnums (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),needs maintainer decision,28,needs maintainer decision,1,0,179853,,,,0,35dc5b6ad085a9841ba3a0063f37331e99222da1020fa58e8e53199bb04be866
thread:issue:179858,issue #179858: torch.compile doesn't use async all_to_all,needs maintainer decision,28,needs maintainer decision,1,0,179858,,,,0,be5987481da8fe26251c9adfa1494764180a03e863c35fc703152366664928c6
thread:issue:179874,issue #179874: DISABLED test_lazy_init_with_view (__main__.LazyFuncionalizationTest),needs maintainer decision,28,needs maintainer decision,1,0,179874,,,,0,15b48239eddfbc036926888c9c6103038d6d209ccc3fa8cc91f82871c7f7df88
thread:issue:179879,issue #179879: DISABLED test_vmap_vmap_cpu (__main__.TestComposabilityCPU),needs maintainer decision,28,needs maintainer decision,1,0,179879,,,,0,0dcaa28f091d36d3541b2cdac8ba0454c197e7177881375648c8494fee564344
thread:issue:179897,issue #179897: DISABLED test_jacfwd (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),needs maintainer decision,28,needs maintainer decision,1,0,179897,,,,0,f15a78695d9dd7cd3f8127e9fb74f23cac33abb4af299c9e1ddfb65f983a9a11
thread:issue:179951,issue #179951: DISABLED test_scaled_mm (__main__.DistMatrixOpsTestWithLocalTensor),needs maintainer decision,28,needs maintainer decision,1,0,179951,,,,0,b38efa1c8a3f871927f144902a5f3882835296483355498544245511a4b68f58
thread:issue:179979,issue #179979: DISABLED test_sparse_tensor_multiprocessing_context_spawn_cuda (__main__.TestDataLoaderDeviceTypeCUDA),needs maintainer decision,28,needs maintainer decision,1,0,179979,,,,0,19a73290e1782618ef7f50facebf6ff4eace51d5b3e87a34ef7391e20a9e3843
thread:issue:179990,issue #179990: DISABLED test_jacfwd_has_aux (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),needs maintainer decision,28,needs maintainer decision,1,0,179990,,,,0,95853b9fab422df755e87408a1508b7671a2c30ec62cb2e207d92bac7b47ffec
thread:issue:180168,issue #180168: [RFC] XPUGraph Trees,needs maintainer decision,28,needs maintainer decision,1,0,180168,,,,0,144f91115c819485928e56ac4a4c868a03e1e174a86e2f37f48aa7ea60b4f47a
thread:issue:180194,issue #180194: DISABLED test_vjp_grad_cpu (__main__.TestComposabilityCPU),needs maintainer decision,28,needs maintainer decision,1,0,180194,,,,0,056f0806c23a7956466f7dda4bbf6c63e7e22bb79e4f10f7d59c2ad2bac36467
thread:issue:180195,issue #180195: DISABLED test_vjp_vmap_cpu (__main__.TestComposabilityCPU),needs maintainer decision,28,needs maintainer decision,1,0,180195,,,,0,ba3c6d0771bd62a70bbedd922ea0ed8d88cd99832ad62309fa42e8e7202bd34f
thread:issue:180204,issue #180204: DISABLED test_jacrev_has_aux (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),needs maintainer decision,28,needs maintainer decision,1,0,180204,,,,0,71c23d1ae1ca6ea38ead2b38762b516fccf6f367b757e314a61bc5b892375cb0
thread:issue:180206,issue #180206: DISABLED test_opcheck_opinfo_NumpyMulScalarCustomOp_cpu_float32 (__main__.TestCustomOpTestingCPU),needs maintainer decision,28,needs maintainer decision,1,0,180206,,,,0,8994832626e0e4b3649e0daac39e03e299b25f6c5ab5abd31b759b49b91897e6
thread:issue:180213,issue #180213: DISABLED test_vmap_vjp_cpu (__main__.TestComposabilityCPU),needs maintainer decision,28,needs maintainer decision,1,0,180213,,,,0,387eb1012355fbd08d9930c117aa988f2c45806c8ea7036781c048ff9dbef78e
thread:issue:180322,issue #180322: DISABLED test_vjp_vjp_cpu (__main__.TestComposabilityCPU),needs maintainer decision,28,needs maintainer decision,1,0,180322,,,,0,8ad1fdb20aa2e43a04fe780485fcd68e90e50193474500d6b2a918dc761a6a17
thread:issue:180324,issue #180324: DISABLED test_jvp_two_tensors_has_aux (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),needs maintainer decision,28,needs maintainer decision,1,0,180324,,,,0,c9e5bc5d6a10bcc36c9eeed7a63794c4f7a15071989abb14d1d3c7e3dbc027dd
thread:issue:180337,issue #180337: DISABLED test_vmap_grad_cpu (__main__.TestComposabilityCPU),needs maintainer decision,28,needs maintainer decision,1,0,180337,,,,0,7de2e9adbe3460ab205d8a75bfd04ac375fa0497fda5a60bd9fc3579c4e76b23
thread:issue:180366,issue #180366: DISABLED test_vmap_free_const (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),needs maintainer decision,28,needs maintainer decision,1,0,180366,,,,0,dbc3f683ef3fe2078e942286ad67b1da2f4d46b379acba424734ba8169dc4912
thread:issue:180414,issue #180414: DISABLED test_torchvision_models_efficientnet_b2 (__main__.TestVisionTracing),needs maintainer decision,28,needs maintainer decision,1,0,180414,,,,0,ecb79eeca4a4ef0965997eae42ba52282a1cc6480181a65b635b3683e0b96340
thread:issue:180437,issue #180437: DISABLED test_find_learning_rate_ensembling_FeatureAlphaDropout_mechanism_functional_call_cpu (__main__.TestExamplesCorrectnessCPU),needs maintainer decision,28,needs maintainer decision,1,0,180437,,,,0,f994e22f8de79f1753f10f42d8347b30c4d19ff3c968f8d72a8b651f009758fe
thread:issue:180439,issue #180439: DISABLED test_vmap_over_vmap_captured (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),needs maintainer decision,28,needs maintainer decision,1,0,180439,,,,0,fd06b9a5a78377d8e488715c252a084f4fcfd9e6e3a526701cdb9f0694ce386d
thread:issue:180490,issue #180490: DISABLED test_vmap_two_inputs (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),needs maintainer decision,28,needs maintainer decision,1,0,180490,,,,0,e2a5d6907ead6c5dd0afcfc4e43b609158b3ee2c3e5774a5a3ef931c17b64302
thread:issue:180542,issue #180542: DISABLED test_vmap_two_inputs_tuple_in_dims (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),needs maintainer decision,28,needs maintainer decision,1,0,180542,,,,0,8a42866aded4c148d406bee8a06d87d1ef08c09465407fae0c9242b542b7dda3
thread:issue:180620,"issue #180620: @custom_op doesn't support ""maybe_out semantics""",needs maintainer decision,28,needs maintainer decision,1,0,180620,,,,0,5faaa61bbebf5d3f8e5e3bb3463c82e1158f1885256b06b0755eb5240212d229
thread:issue:180650,issue #180650: `torch.linalg._powsum` is slow on CPU for ord != 2,needs maintainer decision,28,needs maintainer decision,1,0,180650,,,,0,493913f17055659fc32b47864438c02dff2b5016df384b15ef1ec87c80f1e6f6
thread:issue:180811,"issue #180811: [AOTInductor] max_autotune non-deterministically selects cuBLAS over fused Triton kernels for small GEMM, causing significant performance regression",needs maintainer decision,28,needs maintainer decision,1,0,180811,,,,0,c0a53b76e61188a6aa87f95ad31cf3f3111992ff57fa666751f01dbc74f8e740
thread:issue:180824,"issue #180824: `torch.compile fails with RecursionError on nn.Module with custom __getattr__, while eager execution succeeds`",needs maintainer decision,28,needs maintainer decision,1,0,180824,,,,0,769ee10e494da480efdb69c9ae7cb1efcf95ede4e9da53998ae46846e8dcca8d
thread:issue:180839,issue #180839: [Dynamo] Attempted to call function marked as skipped: torch.sparse_csr_tensor cannot be traced,needs maintainer decision,28,needs maintainer decision,1,0,180839,,,,0,cac2dc716cef9dca4ed799ffa4d030c372e58720f96604f8e3be45fd99f84057
thread:issue:180915,issue #180915: accelerate sympy computations using gmpy2,needs maintainer decision,28,needs maintainer decision,1,0,180915,,,,0,a64de859c6fbcea238ce3af268080d019cdd701015eda5e8dc01cadaa77f7b29
thread:issue:180948,issue #180948: DISABLED test_emulate_precision_casts_preserves_explicit_precision_cast_float16 (__main__.CudaReproTests),needs maintainer decision,28,needs maintainer decision,1,0,180948,,,,0,4ab9b41b7e46ade04f9ab19f626fca0ac857fdf90681510dcdc6b6bfd40dfccd
thread:issue:180960,issue #180960: [cudagraphs] torch.cond does not work with cudagraphs when nccl collectives are used,needs maintainer decision,28,needs maintainer decision,1,0,180960,,,,0,5182d5de6920433d00cc9119faace854e818cbc3831c22d512ba0a6b5c913187
thread:issue:180968,issue #180968: Initializing weights after parametrization silently fails,needs maintainer decision,28,needs maintainer decision,1,0,180968,,,,0,926318667879a446d7ae3152b6622fba2616a42cb14e98e2b280068ab6fb31f7
thread:issue:180974,issue #180974: [RFC][inductor] Support pre/post kernel execution hooks in compiled graphs,needs maintainer decision,28,needs maintainer decision,1,0,180974,,,,0,99e06d5e9951f3b3aca1adec78055e6ffe539e3b82782a7950b1416a7b8dc0e7
thread:issue:181006,issue #181006: [inductor] fails at caching stage with UnboundLocalError (bufXX referenced before assignment) when compiling torch.func.grad with MultiheadAttention + LayerNorm,needs maintainer decision,28,needs maintainer decision,1,0,181006,,,,0,f1cadd1b5da1e452bb2f5dea495c413c18df8277d304438453fc48864522bf7b
thread:issue:181076,issue #181076: [Inductor] Improve compile time of `build_fusion_regions`,needs maintainer decision,28,needs maintainer decision,1,0,181076,,,,0,4dd391315fa91ad9231f9a53015eff9bd0040ccf9ec5db2c485ff8d9f13fe982
thread:issue:181094,issue #181094: Optimization: pin memory management optimization based on pre-allocation and buddy system,needs maintainer decision,28,needs maintainer decision,1,0,181094,,,,0,49a2d78948ecd0a6a546f80106d58f54864c8fa9fb79b4f9d52bbd0835d0df36
thread:issue:181108,issue #181108: Use of dynamic shapes impedes Inductor fusion for XSA projection,needs maintainer decision,28,needs maintainer decision,1,0,181108,,,,0,4f913ca32b4988b20a97ff6e98f54a3a5d1d6b566802853456fc907b0bef5663
thread:issue:181224,issue #181224: DISABLED test_cross_stream_stride_copy (__main__.TestUserStreamCompile),needs maintainer decision,28,needs maintainer decision,1,0,181224,,,,0,b292b5f542ba3531838e0894c70c1707ca337397192dd175fc65e29e18fc1c98
thread:issue:181318,issue #181318: DISABLED test_codegen_structure_three_stream_pipeline (__main__.TestUserStreamCompile),needs maintainer decision,28,needs maintainer decision,1,0,181318,,,,0,47f04c4ad86224aabaeab3ee3b441eb536d65d10cb5e58d6c9248ea87a314955
thread:issue:181322,issue #181322: DISABLED test_trace_return_dataclass_nested (__main__.TestFX),needs maintainer decision,28,needs maintainer decision,1,0,181322,,,,0,ffebdce76673dbfddb24996ad5bb79090af182f4fbf09940c77657b7a18ed6c3
thread:issue:181379,"issue #181379: [cuDNN][SDPA] head_dim limit stuck at 128 on sm_120 (RTX 5090), blocking head_dim=256 models",needs maintainer decision,28,needs maintainer decision,1,0,181379,,,,0,c31bb52516b6c809176bd7661f9602b0c4a07866bae2521b8feb3987ee1f8bd1
thread:issue:181428,issue #181428: torch.compile + torch.func.grad: superlinear compile time with MaxPool2d → log/clamp → pow → AdaptiveAvgPool2d,needs maintainer decision,28,needs maintainer decision,1,0,181428,,,,0,df911434fed8bf8c318f4ace2bce4c33aab9e73f4e487de77a6783a562ed9048
thread:issue:181488,issue #181488: [TensorIterator] reorder_dimensions is sensitive to operand order for commutative ops,needs maintainer decision,28,needs maintainer decision,1,0,181488,,,,0,694ee292923521693f965a85dd8ab894cb6b13d174cca7a76baa3a06b255e473
thread:issue:181494,issue #181494: Add support for sparse block-diagonal matrices,needs maintainer decision,28,needs maintainer decision,1,0,181494,,,,0,f402e700b64afbfd1389282fa309ef9d610bc2d40a39965ea1b65dc5365dd0e0
thread:issue:181499,issue #181499: RocM: Assumed realtime clock rate,needs maintainer decision,28,needs maintainer decision,1,0,181499,,,,0,fbb85da9d3c80106c103fccdfbbf5d83d7c1ca227f73c0a09cd45f375d3a6161
thread:issue:181525,issue #181525: explore using gmpy2 to accelerate expression eval in sympy,needs maintainer decision,28,needs maintainer decision,1,0,181525,,,,0,0bbf9b283c515b1588e305af5bd38560eb0064102ef0ade7eb054b7c099f716c
thread:issue:181629,issue #181629: DISABLED test_torchvision_models_efficientnet_b5 (__main__.TestVisionTracing),needs maintainer decision,28,needs maintainer decision,1,0,181629,,,,0,b9074a8a7ecb94f64a17f6c68ebfd6c24260c212ae89d9c8f4520332d08a7ead
thread:issue:181653,issue #181653: torch.compile(torch.func.grad(fn)) hangs indefinitely on models with pooling ops — sympy.simplify exponential recursion in stride_at,needs maintainer decision,28,needs maintainer decision,1,0,181653,,,,0,b0085e1045b50b13c7661122af17c578474eb36b3d80dd4e284ba2b8068edc3b
thread:issue:181663,issue #181663: torch.softmax is not a true alias for torch.nn.functional.softmax,needs maintainer decision,28,needs maintainer decision,1,0,181663,,,,0,66a2f0c86fc3a474cb6511b543d40a3b37e04f0d71af95244fd63e7bdc4d8846
thread:issue:181718,issue #181718: bf16 matmul throughput drops on RTX 4090 for shapes where N%16==8,needs maintainer decision,28,needs maintainer decision,1,0,181718,,,,0,0edc01cde0107abbe3279df6748a4a1786495f8ee3f0dc49d01c215231f6923a
thread:issue:181723,issue #181723: torch.cuda.make_graphed_callables can silently corrupt parameter gradient accumulation by returning static grad buffers,needs maintainer decision,28,needs maintainer decision,1,0,181723,,,,0,936791fc94b2bab03a23cfc93bb48eb38a5ff6ac6676cb543101ae663959bd1e
thread:issue:181737,issue #181737: image diffusion model output changes for the worse after 530ff92655,needs maintainer decision,28,needs maintainer decision,1,0,181737,,,,0,3531c758a1974b0cc620cba1e947011507c66d607068a18ad0da909da901e171
thread:issue:181851,issue #181851: register_fake to throw error if signature doesn't match op's schema,needs maintainer decision,28,needs maintainer decision,1,0,181851,,,,0,13ae4da29830a1bd2df543a4ce6f502f277fd08159e1c01ece380222bbf87e08
thread:issue:181861,issue #181861: nn.Sequential type hints too general for nested lists,needs maintainer decision,28,needs maintainer decision,1,0,181861,,,,0,b88118ada2f4a639510f355e0ae21fd2c86bb0f5dcde1f7990ce3cd62f418ce6
thread:issue:181939,issue #181939: HIP related errors compiling against PyTorch,needs maintainer decision,28,needs maintainer decision,1,0,181939,,,,0,28c4234070d1f92229d84162bdb04e3de40a3629774fdbfc4663bf8a97be0afb
thread:issue:182001,issue #182001: CUDA linalg: use auxiliary streams for independent looped cuSOLVER batch calls,needs maintainer decision,28,needs maintainer decision,1,0,182001,,,,0,c057af710e35b5f268f6d95f9f89d94aff16897f6ca098ec9252e4af2522919a
thread:issue:182008,issue #182008: Windows Application Control (WDAC) blocks loading unsigned torch DLL,needs maintainer decision,28,needs maintainer decision,1,0,182008,,,,0,faebded9a9e53c0220851cae1a8c45e24aa568afeb332fdcf3ab2f979006b31d
thread:issue:182105,issue #182105: Adding a UserWarning when pickling / deepcopying tensor views,needs maintainer decision,28,needs maintainer decision,1,0,182105,,,,0,f8a4b0dc85a29572f753f3d924f53f7e88626979530287e92fb3d8ac30b74265
thread:issue:182156,issue #182156: Add torch.utils.nan_guard for forward-pass NaN/Inf debugging,needs maintainer decision,28,needs maintainer decision,1,0,182156,,,,0,86f790c57495a3d7fef7d0d9b6d27b1fb221893d5db8352b814e85c2585ee130
thread:issue:182174,issue #182174: DISABLED test_torchvision_models_efficientnet_b0 (__main__.TestVisionTracing),needs maintainer decision,28,needs maintainer decision,1,0,182174,,,,0,d92b59787391a5037dec5ea7f91605a856912d14b4580e6e4feb5bd1b59b83e4
thread:issue:182259,issue #182259: DISABLED test_torchvision_models_convnext_base (__main__.TestVisionTracing),needs maintainer decision,28,needs maintainer decision,1,0,182259,,,,0,2cb04594d3ec5785b7c2082aa12fdf291858ddba45b8329aec400a04dc080ca4
thread:issue:182260,issue #182260: DISABLED test_assert_size_stride_guarded_by_env_check_cpu_with_stack_allocation (__main__.AOTInductorTestABICompatibleCpuWithStackAllocation),needs maintainer decision,28,needs maintainer decision,1,0,182260,,,,0,94dc02c804c88535d9eeffccb48f7bec1b1711bc0f4c1797865a31c8c861adae
thread:issue:182272,issue #182272: [Windows] RTX 5070 Ti (Blackwell sm_120) - cuInit error 100 on hybrid graphics laptops,needs maintainer decision,28,needs maintainer decision,1,0,182272,,,,0,a818a8b18846084daf87b19c36605c1c61424980e5e57bca773ed97b8cec3648
thread:issue:182286,issue #182286: [Bug] CUDA caching allocator unable to satisfy 3-18 GiB contiguous allocation despite 50+ GiB free on RTX PRO 6000 Blackwell + WSL2 (16 GiB hidden CUDA overhead),needs maintainer decision,28,needs maintainer decision,1,0,182286,,,,0,ca37add1e313284ba683cb2963dcd122728df54fb839404c636da5380752df70
thread:issue:182336,issue #182336: Adding new lint passes PR CI but not trunk CI,needs maintainer decision,28,needs maintainer decision,1,0,182336,,,,0,9297beb3a73b2971d5175ee9f5f4b2be5a5ddbe063fb983d8b354aeef94d8a8c
thread:issue:182381,issue #182381: [HOP] Backward pass for scan HOP can result in invalid tensor metadata,needs maintainer decision,28,needs maintainer decision,1,0,182381,,,,0,5abef5e767dc941c90a69f9d9ffce578bc565dd1c6cd27d8d0ede96b6cf32c11
thread:issue:182450,issue #182450: Spin codegen run doesn't use self-contained environement,needs maintainer decision,28,needs maintainer decision,1,0,182450,,,,0,7a0a79552737af43db4e593b693b7d7b4aab7ac49e90ae60c2dd52d4b99386ae
thread:issue:182490,issue #182490: [Docathon] Convert `notes/hip.rst` from rST to MyST Markdown,needs maintainer decision,28,needs maintainer decision,1,0,182490,,,,0,1a37a6e0cf6f3b5ce72d0e7be3582a2e7ee2152dff658de86fa2362081182be7
thread:issue:182560,issue #182560: import torch fails with concurrent nested imports despite GIL ('torch._C' is not a package),needs maintainer decision,28,needs maintainer decision,1,0,182560,,,,0,7040a0bdc95492c11bedefef67725e91ae7243eab08435ae2805ef805575a5f0
thread:issue:182684,issue #182684: [`SDPA`] Checking for FA backend eligibility via SDPAParams fails under fullgraph compile,needs maintainer decision,28,needs maintainer decision,1,0,182684,,,,0,2c2b762b2c86e4ca798509459f6d7633e921c3ea92d41f39edb39d0400aff511
thread:issue:182868,issue #182868: [RFC] [Inductor] L2 cache aware welford/two-pass selction,needs maintainer decision,28,needs maintainer decision,1,0,182868,,,,0,05dad62eb7a75dca570c1639709db80b815e3ab5abb6eeacf94a3659a02cea52
thread:issue:182960,"issue #182960: `torch.cuda.ExternalStream(0, device=...)` silently returns a fresh pooled stream instead of wrapping handle `0x0`",needs maintainer decision,28,needs maintainer decision,1,0,182960,,,,0,926d23c065a9a19eca1d28097a4008e28b07fc7f274bb9ff6b106d3b32eb06f5
thread:issue:183036,issue #183036: Adadelta uses SGD in its examples,needs maintainer decision,28,needs maintainer decision,1,0,183036,,,,0,862b5d26a002ed54a2022243ce68b147882cfe67312b12a1480392d8430ab633
thread:issue:183391,"issue #183391: torch.fmin with CUDA out=int16 casts inputs before computation, producing incorrect result",needs maintainer decision,28,needs maintainer decision,1,0,183391,,,,0,9160ac009b1ac6ec7279235eaab751610fcf20275242c19aeb5dd8a45879170a
thread:issue:183603,issue #183603: torch.compile fails on second-order gradient after autograd.grad(create_graph=True),needs maintainer decision,28,needs maintainer decision,1,0,183603,,,,0,747904cd61f6b936e33d950fa0d93ec96ae988eb5140642098bbdba484870e16
thread:issue:183607,issue #183607: torch.compile changes fp16 overflow behavior for cast-to-float16 multiplication,needs maintainer decision,28,needs maintainer decision,1,0,183607,,,,0,db1c988e2e3b1900b5ba9297dbd967ae65a1819eee617c77c583991a50794425
thread:issue:183686,issue #183686: [triton-cpu] All HuggingFace models of the dynamo benchmark suites failed with the latest triton-cpu backend,needs maintainer decision,28,needs maintainer decision,1,0,183686,,,,0,eabec14c5b1412330a1f31f045dc23af84dda046d5b8f6fef3e637d1ec91ea61
thread:issue:183768,issue #183768: torch.compile returns inf for torch.acosh on large finite float32 inputs while eager returns finite values,needs maintainer decision,28,needs maintainer decision,1,0,183768,,,,0,efc16e314b78007571116d2c491a0f0310905cd806c7f9f8ca687b077c9558b1
thread:issue:184394,issue #184394: [inductor] Merge SymPy printer path into typed ops/CSE path for Triton codegen,needs maintainer decision,28,needs maintainer decision,1,0,184394,,,,0,b0a1a6e8a0ec71aff0d5a0672a78548a516947b35557854f12c147fbd05ac313
thread:issue:184397,issue #184397: [ROCm][Windows] Track proper fix for HIP per-config flag injection bypass (follow-up to #183856),needs maintainer decision,28,needs maintainer decision,1,0,184397,,,,0,bc8d373de09c5e7b02e4fd2f62d03caab4d2f9cc8f0d99f5f6774001b2adb85f
thread:issue:184466,issue #184466: Implementing Self-Regulated Kinetic Learning (SRKL) Optimizer using Second-Order Curvature,needs maintainer decision,28,needs maintainer decision,1,0,184466,,,,0,f8cf6d63a270bd6467db74300ca92fa5b28396a62222e054582e38ebfeae9b14
thread:issue:184769,issue #184769: Optimize Metal Kernels with sincos,needs maintainer decision,28,needs maintainer decision,1,0,184769,,,,0,2a0317f8b3736f60dd2d73a3efebdf09a4723cdfa2b829bf38e358af93b985b1
thread:issue:184903,issue #184903: [Inductor] Codegen Gluon,needs maintainer decision,28,needs maintainer decision,1,0,184903,,,,0,ca4b3d573747bbdfebf22fe5fced5fbe50dd513bf12b1230d83bac327d00672a
thread:issue:185198,issue #185198: [xpu] Intel GPU telemetry management APIs may not work on Arc A-Series and data center GPUs,needs maintainer decision,28,needs maintainer decision,1,0,185198,,,,0,9470cd6d46d5ca81dd1540a0c3a1fba7d5da618fbacb5a19dd0b11c24e9736c1
thread:issue:185215,issue #185215: PyTorch with XPU support fails to build on Windows in Debug/RelWithDebInfo configurations,needs maintainer decision,28,needs maintainer decision,1,0,185215,,,,0,890beebfc0712555480dd011b752733c1234f5fe7342675bd9893f9d65d9d470
thread:issue:185295,issue #185295: [Inductor][feature] Block FP8 quantization: fuse scale layout transformation with quant math,needs maintainer decision,28,needs maintainer decision,1,0,185295,,,,0,b3e15541a0c21cfd44c2dc7f433900907e92542356be66bb16c9e77d64748263
thread:issue:185452,issue #185452: DISABLED test_torchvision_models_convnext_large (__main__.TestVisionTracing),needs maintainer decision,28,needs maintainer decision,1,0,185452,,,,0,395f95bc008f3ef43c1b911087c150abe36a7f8f2c24bc3e205170746d42507e
thread:issue:185534,issue #185534: Sub-optimal heuristics in MixOrderReduction prevent fusion,needs maintainer decision,28,needs maintainer decision,1,0,185534,,,,0,2612a91b8989935594622d39e4be02038b71d3e2cb958b78be015a949a1af855
thread:issue:185587,issue #185587: [inductor] Numerical inconsistency: polygamma + cumulative_trapezoid under torch.compile(dynamic=True) produces wrong results after Conv2d + GroupNorm,needs maintainer decision,28,needs maintainer decision,1,0,185587,,,,0,16e1a78226324ee3c0581d39950c7855f3b99b6eea1cb26168296018afda0739
thread:issue:185639,"issue #185639: torch.func.jacfwd gives incorrect Jacobian for torch.pow(torch.abs(complex), complex exponent)",needs maintainer decision,28,needs maintainer decision,1,0,185639,,,,0,3b6cd45d2bf1956e27ab45061f47c6ef7038f964cf52f1e67ac2accae9938741
thread:issue:185684,issue #185684: 🐛 [Bug] Fake Tensor Updater Doesn't Exist on Windows Platform,needs maintainer decision,28,needs maintainer decision,1,0,185684,,,,0,cd8761921750e02a401d6e227eeb5ee10ec0b82ae89f746d4bb8dbf979396d74
thread:issue:185698,issue #185698: CUDA extension build failure on glibc 2.38+ and GCC 14+ due to math header noexcept mismatch - proposed clean fix,needs maintainer decision,28,needs maintainer decision,1,0,185698,,,,0,4e63a9f1b2e6e4601f195521ed72a7e419489e1e49ad5c23b455cb8d549e58fc
thread:issue:185746,issue #185746: CUDNN version issues in 2.13.0.dev20260530+cu132 nightly wheel,needs maintainer decision,28,needs maintainer decision,1,0,185746,,,,0,4d0e9186ad50a8f5c4efaa87078cb1ea8c58a488ff5a83ec5abc1e91fba35183
thread:issue:186004,issue #186004: git fetch error: not our ref a56dcd79c699366e7ac6466792c3025883ff7704,needs maintainer decision,28,needs maintainer decision,1,0,186004,,,,0,1ff6c2f434e0d5c9d3b4e0b635b16238513a5fd4ea6902031464bb71ade707e0
thread:issue:186030,issue #186030: [DTensor][local_map]Support `None` placements in `local_map` placement specs,needs maintainer decision,28,needs maintainer decision,1,0,186030,,,,0,9cdabec0452e73dbb0bfba20d404dee383132558d2a32adbca1057f36ca88edf
thread:issue:186143,issue #186143: AOTInductor Lowering Backward Compatibility,needs maintainer decision,28,needs maintainer decision,1,0,186143,,,,0,3ba14f796079f9886bfe705c09c87e46ff0874d1271867b6e1c75ca826d76a91
thread:issue:186203,issue #186203: DISABLED test_redispatch_nn_functional_linear_cross_entropy_cuda_float16 (__main__.TestTorchFunctionRedispatchOpsCUDA),needs maintainer decision,28,needs maintainer decision,1,0,186203,,,,0,ed48352b171948c2a74b43fe31fd459a3b9ba88eaad1dcfc9ac1c845efd524e6
thread:issue:186213,issue #186213: CUDA expandable IPC receiver reserves ~9/8 of total GPU memory VA per imported handle and can exhaust device virtual address space,needs maintainer decision,28,needs maintainer decision,1,0,186213,,,,0,10a5c434139053f94c95b76ca9e9c69ea97c67a1475066b2cdb6e9e1d2f85f8e
thread:issue:186259,issue #186259: DISABLED test_redispatch_nn_functional_linear_cross_entropy_chunked_cuda_float16 (__main__.TestTorchFunctionRedispatchOpsCUDA),needs maintainer decision,28,needs maintainer decision,1,0,186259,,,,0,25d7c7585ff7b444419ceba901bfb22e99d4b49c3ead8fd053151cdda8efc3ac
thread:issue:186260,issue #186260: [Accelerator] Maybe host memory should be included in torch.accelerator.mem_stats,needs maintainer decision,28,needs maintainer decision,1,0,186260,,,,0,298c213d29cab3cae35bab401bd1b669d28fbb183be9a067b447346f09f6b58a
thread:issue:186381,issue #186381: [export] No-op contiguous() elided during tracing bakes trace-time layout into the graph; downstream view becomes layout-specific,needs maintainer decision,28,needs maintainer decision,1,0,186381,,,,0,e502083c374991e679a6c46feaa097eceb1f90c55884769359e1ea87f44df39f
thread:issue:186405,issue #186405: arch_id.h: error: template constraint not satisfied,needs maintainer decision,28,needs maintainer decision,1,0,186405,,,,0,78284ea2e6b11f637d18c76a52aca44b23964bee18d98284aef431ed1a606619
thread:issue:186496,"issue #186496: Typo in PyTorch 2.12 quantization docs: ""cetralizing"" should be ""centralizing""",needs maintainer decision,28,needs maintainer decision,1,0,186496,,,,0,e2c11eb04f2f8e33763a3a187f69a0ff31df06da0cdf3f1b14835194eafc5991
thread:issue:186521,issue #186521: Feature Request: Add CPU-only PyTorch wheels for Termux (Android ARM64) + Python 3.13 (cp313),needs maintainer decision,28,needs maintainer decision,1,0,186521,,,,0,082acf855e89295844367bc825ba126c4ec1544792b6283f146931761a0b712f
thread:issue:186594,issue #186594: torch.ops.higher_order.associative_scan cannot be torch.export'ed,needs maintainer decision,28,needs maintainer decision,1,0,186594,,,,0,9739222f7723488dada115cd2e4582135de9c2e10a7854c02ee1b9c65e3fb1f5
thread:issue:186618,issue #186618: Inconsistent rounding between CUDA and CPU for division operator,needs maintainer decision,28,needs maintainer decision,1,0,186618,,,,0,aaf1ad40343e8938095511dd37d17c4730842e7e332fc2574fa4e97d52fb626d
thread:issue:186864,issue #186864: DISABLED test_triggers (__main__.CallbackTests),needs maintainer decision,28,needs maintainer decision,1,0,186864,,,,0,1aa2e007a1e8d1eca5db6387eb87245919bf1497b1c36caf9167a474415610e3
thread:issue:186865,issue #186865: DISABLED test_dynamic_shape_backend_eager_device_cpu (__main__.TestPackage),needs maintainer decision,28,needs maintainer decision,1,0,186865,,,,0,3f528abf2fcde0d393cd5b475b6b68be74ef722ba6363e763cd41297199b6fb5
thread:issue:186868,issue #186868: DISABLED test_LSTM_dropout_per_call_randomness_dropout_p_0_5_training_True_cuda (__main__.TestNNDeviceTypeCUDA),needs maintainer decision,28,needs maintainer decision,1,0,186868,,,,0,35c6448f74627ab6780a4f32927d6048f7c81498e45e31914246ba6744104326
thread:issue:186871,issue #186871: DISABLED test_addmm_relu_tunableop_rocm_cuda_float32 (__main__.TestLinalgCudaOnlyCUDA),needs maintainer decision,28,needs maintainer decision,1,0,186871,,,,0,740c28156d74bcc09a68c57f646933e63763cbf99567befdcfbc20bfa5d36eb9
thread:issue:186872,issue #186872: DISABLED test_call_count_tunableop_cuda_float32 (__main__.TestLinalgCudaOnlyCUDA),needs maintainer decision,28,needs maintainer decision,1,0,186872,,,,0,aec825b0b0bd9efa90d76813c9a16f08eaf26c8886109a6ee643d896efdc28c5
thread:issue:186873,issue #186873: DISABLED test_mm_submatrix_offline_tunableop_cuda_float32 (__main__.TestLinalgCudaOnlyCUDA),needs maintainer decision,28,needs maintainer decision,1,0,186873,,,,0,218524b5cd42e0b382a05fc58824905f4be8dfdb57f6e6d6c21e44defffdcad2
thread:issue:186905,issue #186905: download.pytorch.org packages doesn't have release timestamp,needs maintainer decision,28,needs maintainer decision,1,0,186905,,,,0,7fa9451c67eebaf054cc4887b09d0b073e2b63b4a3110562457f138790db02e2
thread:issue:186908,issue #186908: DISABLED test_nested_tensor_multiprocessing_context_forkserver_cuda (__main__.TestDataLoaderDeviceTypeCUDA),needs maintainer decision,28,needs maintainer decision,1,0,186908,,,,0,4d02949b86b2f0bab951c6246de660a2cf9ba074a6b6a8455e7bf945375fe2c4
thread:issue:186909,issue #186909: DISABLED test_trace_only_export_matches_default_use_cuda_True (__main__.TestProfiler),needs maintainer decision,28,needs maintainer decision,1,0,186909,,,,0,64abfdad608e903c2f0197fcc8a5951bbc5e64118fd08efe44fefc3d7cb158ff
thread:issue:186910,issue #186910: DISABLED test_Embedding_discontiguous_cuda (__main__.TestNN),needs maintainer decision,28,needs maintainer decision,1,0,186910,,,,0,742c3984af2988d42e29e2cbaf7ac45f6476e0716daae898f6657df1813a8c40
thread:issue:186911,issue #186911: DISABLED test_launch_elastic_multiple_agents (__main__.ElasticLaunchTest),needs maintainer decision,28,needs maintainer decision,1,0,186911,,,,0,6ae12a4d4553e2d47be4dc2cf9dd805a5b607338e521b7966266a06433e2eec0
thread:issue:186912,issue #186912: DISABLED test_checkpoint_write_future_state_dict (__main__.TestCheckpointProcess),needs maintainer decision,28,needs maintainer decision,1,0,186912,,,,0,dd6c206ec7ed794f8cd47b6cac3c6d161f27cdb75ab3a067046afb2dba80c5d2
thread:issue:186939,issue #186939: pitch: fast development mode where we use pre-build nightly .so files,needs maintainer decision,28,needs maintainer decision,1,0,186939,,,,0,e49078da6a6d85d63990a57509906ff19169f1f073e8c3f779135338ee2b3053
thread:issue:186999,issue #186999: DISABLED test_bmm_large_batch_dynamic_cuda (__main__.AOTInductorTestDualWrapper),needs maintainer decision,28,needs maintainer decision,1,0,186999,,,,0,ee25a7b3013e9d642f5c735a27dab28ee531a8d3b87629d06ea3f5b758a051b4
thread:issue:187000,issue #187000: DISABLED test_upper_bound_i64_cuda (__main__.AOTInductorTestDualWrapper),needs maintainer decision,28,needs maintainer decision,1,0,187000,,,,0,7aa7290bde98997916dcb07fbb278d76451470afbfb5012732271ea1346c0507
thread:issue:187019,issue #187019: DISABLED test_combo_kernel_no_bench_persistent_reduction (__main__.ComboKernelTests),needs maintainer decision,28,needs maintainer decision,1,0,187019,,,,0,3ec64aa1ea41369431f04ceb592341c120cf7b67e8fe621fcb3fa77f8b6fb811
thread:issue:187020,issue #187020: DISABLED test_redistribute_cost_latency (__main__.TestCostModel),needs maintainer decision,28,needs maintainer decision,1,0,187020,,,,0,f6c638eac5730f0fee9e9494cc97467777a84c2f1a8e30cb3be2aed3656afe40
thread:issue:187045,issue #187045: DISABLED test_out_of_memory_retry (__main__.TestCuda),needs maintainer decision,28,needs maintainer decision,1,0,187045,,,,0,67a850b8220c700bb3d1a89fb7f16e74091a52506dc4baf00e6a23e2de6671c2
thread:issue:187046,issue #187046: DISABLED test_tensor_with_unbacked_shape_closure (__main__.HigherOrderOpTestsWithCompiledAutograd),needs maintainer decision,28,needs maintainer decision,1,0,187046,,,,0,d7b9f1d5aa53d33734ba746d86af375a815147c8cc13494884d0591e8be1a60c
thread:issue:187073,issue #187073: torch.compile recompiles when input alternates between AsyncCollectiveTensor and Tensor,needs maintainer decision,28,needs maintainer decision,1,0,187073,,,,0,eda5c7abacd6e38b8be7b15b3f9f94eda4d22b02a67a366aec7aa0cfc026f575
thread:issue:187148,issue #187148: [Proposal] Fix cpu_fallback returning tensors on wrong device for reverse binary ops,needs maintainer decision,28,needs maintainer decision,1,0,187148,,,,0,ce79376bf356c6deac25f14b261d4f93ebe22c81e19825923e1287fcbec38b1c
thread:issue:187156,issue #187156: DISABLED test_load_with_map_location (__main__.TestCheckpointer),needs maintainer decision,28,needs maintainer decision,1,0,187156,,,,0,0a30e68907687465ef552ff25b2366e61f43d881cbc637dc88bef6d555cba1e4
thread:issue:187176,issue #187176: DISABLED test_functional_call (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),needs maintainer decision,28,needs maintainer decision,1,0,187176,,,,0,6d27fb621b2338c4fbaef75cef58f91b3da4be349784d365416f48a18a8c926c
thread:issue:187180,issue #187180: torch.nn.functional.softplus returns +Inf for finite float16 input when beta is very small,needs maintainer decision,28,needs maintainer decision,1,0,187180,,,,0,c10373ca769665c2330e2faff0983693aeb2652f9f8d834ef35a819ac65190fb
thread:issue:187202,issue #187202: DISABLED test_combo_kernel_fusion_within_same_stream (__main__.TestUserStreamCompile),needs maintainer decision,28,needs maintainer decision,1,0,187202,,,,0,4a9a9b78aac2ca751a6e47f6e0986b0a087700f20f25aeef7a35a8cdb5e19c83
thread:issue:187203,issue #187203: DISABLED test_disable_external_correlation (__main__.TestProfilerCUDA),needs maintainer decision,28,needs maintainer decision,1,0,187203,,,,0,6498fe287979e1f7e5ccac4abfc0b1f27fbeae61a68d5f01954bda124ae73922
thread:issue:187227,issue #187227: [SDPA] Fix 32-bit offset overflow in mem-efficient attention forward with attn_bias,needs maintainer decision,28,needs maintainer decision,1,0,187227,,,,0,ff223ffb72b9f28c6de3287d3bc55cb02b4254238aec193559580b5331009e16
thread:issue:187241,"issue #187241: [inductor] TypeError: Too few arguments for CSE in cutedsl_kernel.py (torch 2.10.0, Python 3.11)",needs maintainer decision,28,needs maintainer decision,1,0,187241,,,,0,3e4a2eb9077376d9b661dc9dcef3df7fd87f40b4bcff099c3245ea2e66350f5b
thread:issue:187276,issue #187276: Two-node DDP ResNet-50/Adam run ends with matching rank-1 loss but gradient and parameter-fingerprint divergence across CPU/Gloo and CUDA/NCCL,needs maintainer decision,28,needs maintainer decision,1,0,187276,,,,0,a275e1b04324da5bd540ced032685bab147dfb96704c8b5d6eafa78bac139efb
thread:issue:187279,issue #187279: Two-node DDP shows reproducible FP32 loss divergence between CPU/Gloo and CUDA/NCCL with large lr and resnet18,needs maintainer decision,28,needs maintainer decision,1,0,187279,,,,0,129dc167836788d3705b80e3fdec7c1380e7c3b785afa4c3117e2d802859bce7
thread:issue:187289,issue #187289: DISABLED test_col2im_uint32_overflow (__main__.TestMPS),needs maintainer decision,28,needs maintainer decision,1,0,187289,,,,0,6e040de5cb88ed6c5e80a580895cc7d38c607041846d588cfc1ae38ba415ff05
thread:issue:187290,issue #187290: DISABLED test_Conv2d_depthwise_kernel_flag_cuda_float16 (__main__.TestConvolutionNNCUDACUDA),needs maintainer decision,28,needs maintainer decision,1,0,187290,,,,0,3a49d846d734a612d2e16c1414206c6a8fd3edac6d0d9e40a6381388acfba526
thread:issue:187378,issue #187378: DISABLED test_grad_capture_tensor (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),needs maintainer decision,28,needs maintainer decision,1,0,187378,,,,0,0ab0a3990502c2e0a1f0cde3d78bcb48a0cdcc8351b0015fafeaa7afa0cef613
thread:issue:187424,issue #187424: DISABLED test_range_based_static_shape_no_cond_dispatch (__main__.TestCustomOpAutoTune),needs maintainer decision,28,needs maintainer decision,1,0,187424,,,,0,c421f00d295e3c5651e0c047c23109b6e83626e3a8c0700ac5e4695846c830bd
thread:issue:187425,issue #187425: DISABLED test_grad_closure_scalar (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),needs maintainer decision,28,needs maintainer decision,1,0,187425,,,,0,af736f8b6038ef1c96b6137979bc43adfb03ce040ed5e93065218f548b454260
thread:issue:187470,issue #187470: MemberType LLM readability,needs maintainer decision,28,needs maintainer decision,1,0,187470,,,,0,957e60d6beda6779870555e24036e27bbd294be2e42c758ebb6a07cc7d969ad1
thread:issue:187505,issue #187505: OpaqueObject Fuzzing,needs maintainer decision,28,needs maintainer decision,1,0,187505,,,,0,b2420b49157f5bf9771c2f9c0db46408690a375e2abb067413e6a0c8796c14c3
thread:issue:187561,issue #187561: DISABLED test_view_ops (__main__.TestViewOpsWithLocalTensor),needs maintainer decision,28,needs maintainer decision,1,0,187561,,,,0,611c62821cd85af81021a462173e4117b6c9f0c34463f490d500b9ce1a5b1f6e
thread:issue:187582,issue #187582: torch.nested NJTs with identical layout fail on binary operation,needs maintainer decision,28,needs maintainer decision,1,0,187582,,,,0,f24f557f06f77112014430273f25497fcfa46056ace05f2828a42a44b7454529
thread:issue:187645,issue #187645: DISABLED test_async_future_results (__main__.TestAsyncCheckpointerSpecific),needs maintainer decision,28,needs maintainer decision,1,0,187645,,,,0,8f435eb8bdfbaeb61dc473cd42bec84783837aca157658278f584ee6f69fd7c7
thread:issue:187649,issue #187649: [Inductor] Support fx wrapper with non-triton fused kernel.,needs maintainer decision,28,needs maintainer decision,1,0,187649,,,,0,47b67f4d4ecfd4f5a50e8c865993a0b5c44f20535698ee6e1380a5ce0c39e65b
thread:issue:187658,issue #187658: DISABLED test_grad_non_tensor_input (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),needs maintainer decision,28,needs maintainer decision,1,0,187658,,,,0,1dd5151eef21a722cd8653e326622f9cb82a39f9a1c9a01154b7d75cb8f0f0a8
thread:issue:187715,issue #187715: `torch.compile(mode='reduce-overhead')` silently degrades training on ROCm — `cudagraph_trees` has no HIP guard,needs maintainer decision,28,needs maintainer decision,1,0,187715,,,,0,35ac98751e8a1d37d17b37c4a9294b80d8a19805b05eee1e7df07f7a05127532
thread:issue:187718,issue #187718: DISABLED test_grad_two_tensor_all_grad_has_aux (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),needs maintainer decision,28,needs maintainer decision,1,0,187718,,,,0,5c7d39691e46031e466f6db295b1ca76e7869886fffcbc5c6681afc45367c22e
thread:issue:187738,issue #187738: DISABLED test_grad_two_tensor_has_aux (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),needs maintainer decision,28,needs maintainer decision,1,0,187738,,,,0,10d324bf94efdf3fc0bf01af727a7a2b27bd2e052089e39a4f083bb008dbc3fe
thread:issue:187770,issue #187770: torch.nn.attention.varlen: make backend selection configurable,needs maintainer decision,28,needs maintainer decision,1,0,187770,,,,0,6968bcf973c200dfb01bf165079d97342649fca16bd60ef7f30b2aaaf4c06cc5
thread:issue:187807,issue #187807: [PyTorch 2.13][API-Unstable] Intel XPU Enhancements,needs maintainer decision,28,needs maintainer decision,1,0,187807,,,,0,1e0e7f40c3dbf679551f6446310713cfecdc2b7ff3198760161c2e5e54178830
thread:issue:187827,issue #187827: [pr-review skill] Add MPS/Metal kernel review checks (tracking),needs maintainer decision,28,needs maintainer decision,1,0,187827,,,,0,00aa4e9346dd71c9d43af37723b2a6efa98e1aa8da7cf35435708ed8fd26f582
thread:issue:187834,issue #187834: reminder to check cu129 binaries aren't released with pytorch 2.14,needs maintainer decision,28,needs maintainer decision,1,0,187834,,,,0,f9aa7b6385f53933e0ba58fb9c755dca35b86a160e7cea395fd6540443f0a5d7
thread:issue:187862,issue #187862: `OpOverload.__call__` is roughly 100ns overhead,needs maintainer decision,28,needs maintainer decision,1,0,187862,,,,0,a4e6515b05e75ff3360a8c1d2dfebf13790281c6cd597bced304dc691463a053
thread:issue:187907,issue #187907: AOTInductor: reverse-mode autograd.grad through sqrt/reciprocal bakes a symbolic-shaped constant under a dynamic batch (strict export),needs maintainer decision,28,needs maintainer decision,1,0,187907,,,,0,e1f76bff4f2ebbf5d7cbd4b9c56cc62d2d851f9394d5c1c997196ec1f138c470
thread:issue:187963,issue #187963: Pallas CPU max_unpool2d does not preserve channels-last output layout,needs maintainer decision,28,needs maintainer decision,1,0,187963,,,,0,cb90015e19bf68494cda5f6575d59ab55e660df8b14113bde3c8f24dbe19e938
thread:issue:187974,issue #187974: reserving issue for myself,needs maintainer decision,28,needs maintainer decision,1,0,187974,,,,0,12d5e5cc4d1c9432fd71a8c50e2cf1f38028179f6cee800f8148755d3fe128b8
thread:issue:187996,issue #187996: TestDTensorStateDictStager.test_async_save_no_memory_leak fails on GB200,needs maintainer decision,28,needs maintainer decision,1,0,187996,,,,0,ba552f9aa99e3c906ed48c200e2837e1e8c1ad96cf7a39ca390d305fa27d4993
thread:issue:188001,issue #188001: [RFC] Unify rendezvous caching across SymmetricMemory backends,needs maintainer decision,28,needs maintainer decision,1,0,188001,,,,0,63c4e28b7912d8d4e58b097f2327eacc22aebf7f08f32affa8834898c88ca5f9
thread:issue:188005,issue #188005: [RFC] Protect SymmetricMemory signal pads against cross-stream races,needs maintainer decision,28,needs maintainer decision,1,0,188005,,,,0,52fc935432dde70bfdedcb1f02cc6df270105a9eb286ac69563b60eed1486b70
thread:issue:188041,issue #188041: DISABLED test_jacfwd_randomness (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),needs maintainer decision,28,needs maintainer decision,1,0,188041,,,,0,f86949f4a4f004969096932e8133d283d259311326e7b82de6be6759f2e4382b
thread:issue:188074,issue #188074: torch.compile produces inconsistent outputs for stateful SNN model with custom LIF neurons,needs maintainer decision,28,needs maintainer decision,1,0,188074,,,,0,17bca883e5ada04284f3adda9a8ce04aae96d3ec93538de7cc7a664dd9f96bc8
thread:issue:188077,issue #188077: torch._dynamo.disable seems expensive,needs maintainer decision,28,needs maintainer decision,1,0,188077,,,,0,8bd0241d280116bed832509a750123614bedcae143b0d96cefc11c767a5b9cb4
thread:issue:188087,"issue #188087: torch.nanmedian(..., dim=None) raises a named-dimension RuntimeError while omitting dim works",needs maintainer decision,28,needs maintainer decision,1,0,188087,,,,0,0efa59c100ac4a9128f9782fc2414f2417472ae9011ed5b8b679a49b1b6cc45d
thread:issue:188123,issue #188123: DISABLED test_index_cuda_with_cpu_int64_propagate_real_tensors (__main__.PropagateRealTensorsFakeTensorTest),needs maintainer decision,28,needs maintainer decision,1,0,188123,,,,0,5432add1f7adcba165d5de6ad87b1e932d4a0244cd61b00ea8d533ec5569225b
thread:issue:188135,issue #188135: DISABLED test_compare_cpu_linalg_polar_xpu_float32 (__main__.TestCommonXPU),needs maintainer decision,28,needs maintainer decision,1,0,188135,,,,0,964bbe4a8c0269989265669c8a9af0b01a40502f5dfdc86f7c44d55bddd8cf6d
thread:issue:188144,issue #188144: DISABLED test_custom_op_dce_dynamic_shapes_cuda (__main__.DynamicShapesCodegenGPUTests),needs maintainer decision,28,needs maintainer decision,1,0,188144,,,,0,c8293ea66b599d2cf7004bbbbbf5cc44c7fda34bc2035e84eea7206dcc29ec1f
thread:issue:188151,issue #188151: torch.cond fails in torch.compile with simpliest symbolic shapes (torch 2.12+),needs maintainer decision,28,needs maintainer decision,1,0,188151,,,,0,bc7bcc00e128a4c3d884ddfd527bc89b9fa40780945cda15ed7ea1a58c08215c
thread:issue:188205,issue #188205: Warn if torch.Tensor.backward() or torch.autograd.grad() is called underneath torch.cuda.use_mem_pool() with torch.autograd.grad_mode.set_multithreading_enabled(True),needs maintainer decision,28,needs maintainer decision,1,0,188205,,,,0,bc86b474762050ae11a0123b771f85a8d187583de6e3307580e2b82e5b7406bf
thread:issue:188212,issue #188212: [RFC] Add Gluon Backend To Inductor Backend,needs maintainer decision,28,needs maintainer decision,1,0,188212,,,,0,b143b5f076d500c8ba19782152765126065ea881b178d72d287b8d0b854edba4
thread:issue:188222,issue #188222: [aot_autograd] `RuntimeError: view size is not compatible ... use .reshape()` under `torch.compile`,needs maintainer decision,28,needs maintainer decision,1,0,188222,,,,0,b0d9be1c0f0b97771105991f55e4203590a3f968aaea646806dbdcd815682fa5
thread:issue:188232,issue #188232: [aot_autograd] `linalg.eigh` fails to converge (ill-conditioned) under `torch.compile` but not eager,needs maintainer decision,28,needs maintainer decision,1,0,188232,,,,0,0cd89d8ae6202104346515b766930c30524ed38a4ab459f85aa51654a1bf2e74
thread:issue:188234,issue #188234: DISABLED test_del_attr_multi_stream_sync_dealloc (__main__.TestStreams),needs maintainer decision,28,needs maintainer decision,1,0,188234,,,,0,9faeda23ec250dfd893646748a8d4f472eebc583ef7bbd3674c2f643aadb00c4
thread:issue:188238,issue #188238: DISABLED test_comprehensive_linalg_polar_xpu_float64 (__main__.TestInductorOpInfoXPU),needs maintainer decision,28,needs maintainer decision,1,0,188238,,,,0,e849db80c9d5fa3fe138fceb6655d54c4881cbec25eb811b5d284491583573db
thread:issue:188239,issue #188239: DISABLED test_comprehensive_linalg_polar_xpu_float32 (__main__.TestInductorOpInfoXPU),needs maintainer decision,28,needs maintainer decision,1,0,188239,,,,0,981d075b9808029c736b5c333969f36d7011c6138f03d65998bd560f2267d750
thread:issue:188317,issue #188317: [RFC] Add process group revoke in torch.distributed,needs maintainer decision,28,needs maintainer decision,1,0,188317,,,,0,8f17354b3da099e9999aa9565f21759302ce28dc0eabe6370f0f1495aad05db6
thread:issue:188360,issue #188360: Inconsistent and unclear error messages for `offsets` validation when `input.dim() == 1` in `torch/nn/functional.py`,needs maintainer decision,28,needs maintainer decision,1,0,188360,,,,0,c9f9f4593ff878e5566bf0232faeb2778b82f6c0d87373230cefc117ec69494f
thread:issue:188375,"issue #188375: CPU LayerNorm normalized_shape=(1,) gives nonzero weight.grad although output is independent of weight",needs maintainer decision,28,needs maintainer decision,1,0,188375,,,,0,9ea0b092d01cd9252e5a87b452782d059f1ce68b29260ec8c897f058e8a59596
thread:issue:188382,issue #188382: torch.special.logit with eps > 0.5 has tiny vectorized-vs-scalar exact parity mismatch on non-contiguous float32 input,needs maintainer decision,28,needs maintainer decision,1,0,188382,,,,0,fc5f5303272e38dd9241af1a4e275259c50ffd0ffcfe56b8415ace342e1dbdb8
thread:issue:188389,issue #188389: [Docs] torch.nn.Module.parameters() does not specify whether the ordering is deterministic,needs maintainer decision,28,needs maintainer decision,1,0,188389,,,,0,2365032ed94b23c2860e02e15e6e0a55fcdcacc6d85bcaabf47e557aed07a77b
thread:issue:188391,issue #188391: DISABLED test_compare_cpu_linalg_matrix_sqrth_xpu_float32 (__main__.TestCommonXPU),needs maintainer decision,28,needs maintainer decision,1,0,188391,,,,0,360c61e80ee55a6aee2e625c2c2d5330bba04a56e2b3fba030df02582fcfa02b
thread:issue:188392,issue #188392: DISABLED test_comprehensive_linalg_matrix_sqrth_xpu_float64 (__main__.TestInductorOpInfoXPU),needs maintainer decision,28,needs maintainer decision,1,0,188392,,,,0,f189a2626e88d11abc7f848d5b1a2f85e9e1d5055dc9dae87576688cb44b4153
thread:issue:188393,issue #188393: DISABLED test_comprehensive_linalg_matrix_sqrth_xpu_float32 (__main__.TestInductorOpInfoXPU),needs maintainer decision,28,needs maintainer decision,1,0,188393,,,,0,b2cfc57ef651eed00f333f4a4824d3e45928834fa9e2b4a33624db4dd27199bb
thread:issue:188446,issue #188446: Add issue de-duplication system,needs maintainer decision,28,needs maintainer decision,1,0,188446,,,,0,9d0baf4586df3e9c2a7f9aa4545bb8ca64f611c2876099d04c3dea12ac0ec408
thread:issue:188496,issue #188496: torch.fft.irfft2 ONNX export produces graph that onnxruntime cannot execute.,needs maintainer decision,28,needs maintainer decision,1,0,188496,,,,0,01dbb2fc7e74873edff606c3a2aa433c1aa44c5bb64d2dfef761e6cf07cc17ae
thread:issue:188542,issue #188542: PyProcessGroup is missing several `ProcessGroup` collective overrides,needs maintainer decision,28,needs maintainer decision,1,0,188542,,,,0,0c3e8c3d01398e089b846fc7f6aae7cc0215d94526c9eeca9db1c3292dd81785
thread:issue:188627,issue #188627: DISABLED test_control_deps_passthrough_creates_ordering_barrier (__main__.TestControlDeps),needs maintainer decision,28,needs maintainer decision,1,0,188627,,,,0,7f1a7dde7fddce615c6c63e2fbc176264daeb0bbb447e57896e82ed3f3b84ec2
thread:issue:188628,issue #188628: DISABLED test_rank0_broadcast_meta_device_init (__main__.TestFullyShardMetaDeviceInit),needs maintainer decision,28,needs maintainer decision,1,0,188628,,,,0,95a116071e6da0da44ddb0b6917a32146e97f4373ae0d86004d96e7284c2a4e6
thread:issue:188648,"issue #188648: treespec_dumps does not preserve tuple keys in dict/OrderedDict/defaultdict pytrees, so the round-trip fails with `unhashable type: 'list'`",needs maintainer decision,28,needs maintainer decision,1,0,188648,,,,0,c43e2ed24c83c01cf90013f20c768c872940234fb71c45758eabd81f073b47d3
thread:issue:188668,issue #188668: DISABLED test_extern_kernel_benchmark_valid_timing (__main__.TestExternKernelCaller),needs maintainer decision,28,needs maintainer decision,1,0,188668,,,,0,4570efbc5fd9981af86b58b60ae3a0e94d5d364cef3ecb3eea4220428e0f9c41
thread:issue:188669,issue #188669: fx_graph_cse skipping duplicate element-wise CUDA optimization.,needs maintainer decision,28,needs maintainer decision,1,0,188669,,,,0,deefa3da9f71750f3925942044535791db173a00dda84f975a3e66cc1af757cc
thread:issue:188676,issue #188676: DISABLED test_float8_e4m3fn_uint8_decode_codegen_cuda (__main__.TestFP8TypesCUDA),needs maintainer decision,28,needs maintainer decision,1,0,188676,,,,0,53710fc7ac331d121edcf48a8725fe61ace52cf31c117408dee6f494799f86d9
thread:issue:188677,issue #188677: DISABLED test_graph_partition_view_fallback (__main__.CudaGraphTreeTests),needs maintainer decision,28,needs maintainer decision,1,0,188677,,,,0,c3d6ea68fb2126a4cd40c9e64679cfcd1e83d0f6f9071da4caf0360a9f45aa0a
thread:issue:188684,issue #188684: DISABLED test_redispatch_nn_functional_conv_transpose3d_cuda_bfloat16 (__main__.TestTorchFunctionRedispatchOpsCUDA),needs maintainer decision,28,needs maintainer decision,1,0,188684,,,,0,14765461a5588c9252bcb4ff4c970aebb7cfaa789ce8ab340988c50e17f663fb
thread:issue:188686,issue #188686: DISABLED test_redispatch_nn_functional_conv_transpose3d_cuda_complex32 (__main__.TestTorchFunctionRedispatchOpsCUDA),needs maintainer decision,28,needs maintainer decision,1,0,188686,,,,0,0f59334ab835bb3c8853b417d286354ea5dab14ee706cc2a6d39e00fd928aa65
thread:issue:188687,issue #188687: DISABLED test_linear_block_compile_parity_forward (__main__.TestDropoutAlignRandomEager),needs maintainer decision,28,needs maintainer decision,1,0,188687,,,,0,4ecddf40ce7cd6fcc5264280835e6791c751a45820535ecd8c0b127243d659d1
thread:issue:188688,issue #188688: DISABLED test_backend_simple (__main__.PrivateUse1BackendTest),needs maintainer decision,28,needs maintainer decision,1,0,188688,,,,0,6ca0974d95a50c670cb7659d2af725d8546e92dc0d884a4c22de3b89f1a04bb6
thread:issue:188692,issue #188692: DISABLED test_mkldnn_rnn_backward_gru_bias_shape (__main__.TestMetaKernelRegistrations),needs maintainer decision,28,needs maintainer decision,1,0,188692,,,,0,888417581a50cbf79814537f7b96971be60694ff2b81495005d4f37e8aef9310
thread:issue:188693,issue #188693: DISABLED test_vjp_scalar_logging_tensor (__main__.TestAutogradFunctional),needs maintainer decision,28,needs maintainer decision,1,0,188693,,,,0,05dec98a32236d340243c925b96b0135aecf5d88058a76e86445cb6b85a25e69
thread:issue:188705,issue #188705: DISABLED test_reduction_not_fused (__main__.TestNVUniversalGemmEpilogueFusion),needs maintainer decision,28,needs maintainer decision,1,0,188705,,,,0,74e008ed974173e5fc10fa4876829683232c06ffeae02a8fd1aeb62a189aa2bc
thread:issue:188740,issue #188740: DISABLED test_linear_block_compile_parity_backward (__main__.TestDropoutAlignRandomEager),needs maintainer decision,28,needs maintainer decision,1,0,188740,,,,0,a504da3aab4a9cf7e128b1fd579022ad5291c0965393249e77b03f140a7ba92d
thread:issue:188767,issue #188767: DISABLED test_reconfigure_scale_down_up (__main__.GlooFaultToleranceTest),needs maintainer decision,28,needs maintainer decision,1,0,188767,,,,0,77a6be79bbb457ad537057e02c3d02a614c9fcb9d26c9708cbf711b62658fd97
thread:issue:188798,issue #188798: DISABLED test_linearize_jvp_fn (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),needs maintainer decision,28,needs maintainer decision,1,0,188798,,,,0,f657bc9cae254082c9b90ec599feb3b266097c5b3db681bc5caae5f8e55e3044
thread:issue:188837,issue #188837: DISABLED test_dropout_parity_cudagraphs_reduce_overhead (__main__.TestDropoutAlignRandomEager),needs maintainer decision,28,needs maintainer decision,1,0,188837,,,,0,7af319987129096b245e3926eb8f9201e1af51092790c303f1b0b73439b50ca2
thread:issue:188840,issue #188840: DISABLED test_eager_equivalence_nn_functional_gelu_backend_inductor_default_cuda_float32 (__main__.TestOpInfoPropertiesCUDA),needs maintainer decision,28,needs maintainer decision,1,0,188840,,,,0,733c1590e35bde5470c319101c84c014f17777915487f8e1e16d4cc17efb5c1e
thread:issue:188851,issue #188851: [CI][ROCm] DebertaV2ForMaskedLM fails to import from transformers on gfx950 (dynamic_inductor_huggingface eager_fail_to_run),needs maintainer decision,28,needs maintainer decision,1,0,188851,,,,0,fc62d6e3dbb69c00eb1a87e7c1cfb16b1a6bf99b3162aab5592ee599241da45b
thread:issue:188871,issue #188871: [FlexAttention] FLASH backend: cute.where Int64/Int32 dtype mismatch in captured-buffer negative-index wrap under dynamic shapes,needs maintainer decision,28,needs maintainer decision,1,0,188871,,,,0,8274e24f53af1c17acd1d065b524822a688b253ae4a8698f170ab7b6bb830b31
thread:issue:188895,issue #188895: DISABLED test_conv2d_backward_parametrized_dilation_2_stride_1_padding_1_kernel_3_channels_groups4_nhwc_True_cuda (__main__.GPUTests),needs maintainer decision,28,needs maintainer decision,1,0,188895,,,,0,65520a5e24fff098d96ed49d7ec3a9311225d8cd1122c5d456e22c7581e04d45
thread:issue:188897,issue #188897: DISABLED test_addmm_relu_tunableop_cuda_float32 (__main__.TestLinalgCudaOnlyCUDA),needs maintainer decision,28,needs maintainer decision,1,0,188897,,,,0,406e5676dd11d2c0283b86145c82353246673e38abcd4a84604fb49eba17664f
thread:issue:188899,issue #188899: DISABLED test_cache_hot_load_caching_precompile_dynamic_True_device_cuda_bfloat16 (__main__.TestFxGraphCache),needs maintainer decision,28,needs maintainer decision,1,0,188899,,,,0,fb328c5bdc5503ffd9f50e9d095fbc0a3e677457a5b6c410ad67ddb0acc9dfdd
thread:issue:188926,issue #188926: [CPU] torch.unravel_index raises ZeroDivisionError when shape has a zero dimension,needs maintainer decision,28,needs maintainer decision,1,0,188926,,,,0,63c228eaa674c37d8208d51c67a4a58562785c5c63422b05d79e3552931f610e
thread:issue:188935,issue #188935: find_library(ROCM_ROCTX_LIB roctx64 ...) should be gated behind USE_ROCTRACER,needs maintainer decision,28,needs maintainer decision,1,0,188935,,,,0,392df4995c5009116adc6a5e9c6202a6c4a39491cb1eb7f60d3f809a6ca4d98b
thread:issue:188983,issue #188983: DISABLED test_comprehensive_combinations_cuda_float16 (__main__.TestInductorOpInfoCUDA),needs maintainer decision,28,needs maintainer decision,1,0,188983,,,,0,908bf3d454d8ab8a36e566618e1482fb517264d95a11bec77c5e04a899f039ee
thread:issue:188985,issue #188985: DISABLED test_output_grad_match_combinations_mps_float16 (__main__.TestConsistencyMPS),needs maintainer decision,28,needs maintainer decision,1,0,188985,,,,0,e15371b9354335630189cd9510438f0c807d78c384f7c9f7ef35f112d18b739c
thread:issue:189015,issue #189015: DISABLED test_jacfwd_two_tensors_argnums (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),needs maintainer decision,28,needs maintainer decision,1,0,189015,,,,0,8349f7fe02aba388f3564b4238bd2fd808e94d47adc546c8a076430432a472d6
thread:issue:189032,issue #189032: nightly / update-metamates-rule failing since ~2026-06-29 — UPDATEBOT_TOKEN can't read the pytorch/metamates team,needs maintainer decision,28,needs maintainer decision,1,0,189032,,,,0,31a46249164c8d6e3f686d76ff38e2b570d32b83fdefac724413aaa2a87780db
thread:issue:189046,issue #189046: Treat PrivateUse1 as CUDA-like under env override to reuse CUDA optimization paths in shared `at::native` operators,needs maintainer decision,28,needs maintainer decision,1,0,189046,,,,0,66d99fd150a0881a96aba0e9e831aa4643ff0ef235347fc4cabff67edc72a56f
thread:issue:189066,issue #189066: DISABLED test_jacrev (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),needs maintainer decision,28,needs maintainer decision,1,0,189066,,,,0,2a2233cd405db73379a1b49200f2d1a0800d27a30a3610ea59a7625ebd1b0811
thread:issue:189067,issue #189067: DISABLED test_pyobject_redispatch_requires_keyset (__main__.TestCustomOpWithCompiledAutograd),needs maintainer decision,28,needs maintainer decision,1,0,189067,,,,0,134bbc6174a6cb1c1e003362ec1b2a611156e9381aad0a52b3ebc2a930c91d5d
thread:issue:189093,issue #189093: Misleading full-backward-hook warning when only `register_full_backward_pre_hook` is used,needs maintainer decision,28,needs maintainer decision,1,0,189093,,,,0,c8f5754225a2cc373384bdfd2b7a1d78be99236b93609285fbf99a6466a228fd
thread:issue:189151,issue #189151: DISABLED test_pending_fusion_pro_and_epi_use_async_compile_False (__main__.TestPrologueFusion),needs maintainer decision,28,needs maintainer decision,1,0,189151,,,,0,fcbbd76ffde2df06c8ec073c52e915e3c8e839f22a9d3106c031c05666f848e1
thread:issue:189164,issue #189164: [CI][B200] B200 Runner Maintenance Notification,needs maintainer decision,28,needs maintainer decision,1,0,189164,,,,0,1a7b83a334204fa951c43dbcdcc3e805425248d3fcfe0a681196f6519f450bfa
thread:issue:189189,issue #189189: Subtraction between two `torch.complex64` tensors fails on modules compiled with `aoti_compile_and_package`,needs maintainer decision,28,needs maintainer decision,1,0,189189,,,,0,c5bed89f7a9533ef2b7e8e24976e2db15bd759b082ac632f03175f0d4c671ce6
thread:issue:189221,issue #189221: DISABLED test_nested_dict_partial_load (__main__.TestCheckpointer),needs maintainer decision,28,needs maintainer decision,1,0,189221,,,,0,f2bd0254b457dfce9c80e0126416e7a8a409a20d83a7aeac2c527c2bd82f4dad
thread:issue:189223,issue #189223: DISABLED test_vmap_free_tensor (__main__.FuncTorchHigherOrderOpTestsWithCompiledAutograd),needs maintainer decision,28,needs maintainer decision,1,0,189223,,,,0,10cc43e80a1751b8d093b2d877df10e8a863434febb0c93cf0b29eab8ae6b83d
thread:issue:189236,issue #189236: [Feature Request] Add 'symmetric' padding_mode (repeats boundary sample),needs maintainer decision,28,needs maintainer decision,1,0,189236,,,,0,f64d0714787512ad9f9c8fe8019ca56c40b2e0d2ba5bc6195b8423dde56dcf38
thread:issue:189244,issue #189244: DISABLED test_cat_pointwise (__main__.NumBytesMetricTests),needs maintainer decision,28,needs maintainer decision,1,0,189244,,,,0,5a173fcf28f0ab743f7f5629e51fca8a27f23610de72a91930d30c31f920ca3c
thread:issue:189255,issue #189255: DISABLED test_delayed_optim_step_offload_true_no_shard_cuda (__main__.TestParityWithDDPCUDA),needs maintainer decision,28,needs maintainer decision,1,0,189255,,,,0,5b0a76ac8be57c7bf8be9069c61f90e64e84ee987377da32e23562be04ab9234
thread:issue:189261,issue #189261: DISABLED periodic-rocm-mi300 / linux-noble-rocm-py3.12-mi300 / test,needs maintainer decision,28,needs maintainer decision,1,0,189261,,,,0,2556344269d8cc2485aa9399c83838d14f8cb48a0cf0757a3a15d7a20e7617e6
thread:issue:189262,issue #189262: DISABLED rocm-nightly / linux-noble-rocm-nightly-py3.12-gfx942 / test,needs maintainer decision,28,needs maintainer decision,1,0,189262,,,,0,e3b94d2cc3e91d186b9e2f0747f0e76cd055f7ff58c7e6f3dc512011a67af57e
thread:issue:189279,issue #189279: [CPU] operator_benchmark: eager transcendental unary (cosh/sinh/atan/log2/log10) ~3-5x slower since ~Nov 2025,needs maintainer decision,28,needs maintainer decision,1,0,189279,,,,0,69f43ba665e82bc91b579a9629d041e9a1957b4400370d9d5af1b8b3d1914266
thread:issue:25174,issue #25174: BCEWithLogitsLoss expects wrong shape of weight (#classes instead of batch size),needs maintainer decision,28,needs maintainer decision,1,0,25174,,,,0,82d461d0d5d8ede5d973ed5c29204a33e5a093350c6d15411c6903464f0969c9
thread:issue:26207,issue #26207: Dispatch Tracing/Debugging,needs maintainer decision,28,needs maintainer decision,1,0,26207,,,,0,c1d77081cfef22761d0a5e6fd5b1cc488385942277871c9b006e7b685015cf38
thread:issue:28648,issue #28648: Can't successfully install pytorch with python3.6 on my pi 4,needs maintainer decision,28,needs maintainer decision,1,0,28648,,,,0,54d610905ea0d13d7030a790604d7227912b7519472d4a3d8a0fe9248a0e81b4
thread:issue:34130,issue #34130: ExceptionWrapper cannot handle exceptions with more than one argument,needs maintainer decision,28,needs maintainer decision,1,0,34130,,,,0,e28f6167c90a3efd9c6ec6ed2e25fa43b769871d38d6568bc6daa32b89577b32
thread:issue:34242,issue #34242: RuntimeError with torch.unique: radix_sort: failed on 2nd step: invalid argument,needs maintainer decision,28,needs maintainer decision,1,0,34242,,,,0,6250216366a8092a0316bcce668107faca2f0617a8f1fa0c60369e77d6072df0
thread:issue:38035,issue #38035: Quantile Regression Loss,needs maintainer decision,28,needs maintainer decision,1,0,38035,,,,0,d6d64b6eb94c48250230fc25e50904e991a3d63b95fe047fbc40677359830932
thread:issue:45728,issue #45728: Low shared memory,needs maintainer decision,28,needs maintainer decision,1,0,45728,,,,0,2354620001c00d2e32e8c69c8c2de86de6cfaa91239b7af672c815a459019aff
thread:issue:48972,issue #48972: non-negative least squares solver feature request,needs maintainer decision,28,needs maintainer decision,1,0,48972,,,,0,9bb137e84089c46dd42aa52aef34a4a780b5c80405c8bf7c1b0e9cd316a133d5
thread:issue:50282,issue #50282: Add MultiScheduler,needs maintainer decision,28,needs maintainer decision,1,0,50282,,,,0,4a9430fe2683ae4dbd888d588aeddad852ad588466e8324285dcddf2ca5b8e6c
thread:issue:51002,issue #51002: Consider adding context manager support for torch.set_printoptions,needs maintainer decision,28,needs maintainer decision,1,0,51002,,,,0,121d22b07af62aa51cb143b09a2f602db98b8f58f80a0284b12036993d4056cf
thread:issue:52241,issue #52241: CTCLoss gradient is incorrect,needs maintainer decision,28,needs maintainer decision,1,0,52241,,,,0,22c26655a89112549f64ab378b4f8308616555ca12e6254d1a1b0d95e4e8d40f
thread:issue:53721,issue #53721: PyTorch on Embedded Hardware,needs maintainer decision,28,needs maintainer decision,1,0,53721,,,,0,9431efec8ccce6c3e35c9f78258a7d9d6bcd1dd86288295e1e1b6b1b3d558fc3
thread:issue:54881,issue #54881: Often the documentation does not match the actual signature of the function,needs maintainer decision,28,needs maintainer decision,1,0,54881,,,,0,7cb5e2c3bcf9eca842221bf466fe0f7aacd5c1ec25055ec49c177f1698354232
thread:issue:55777,issue #55777: torch.load is very slow with gzip.open,needs maintainer decision,28,needs maintainer decision,1,0,55777,,,,0,c51afae03468cc5bef120a8e97c13be9fffbf35d5a954ebfc4ea6000b4550c15
thread:issue:55937,issue #55937: test_stream_event_nogil: Is the test making a wrong assumption?,needs maintainer decision,28,needs maintainer decision,1,0,55937,,,,0,add2f15fbb4a30545778e42353e0329b29687f1369a37ad517f29fae8da46285
thread:issue:56340,issue #56340: log_softmax(x) != x - logsumexp(x),needs maintainer decision,28,needs maintainer decision,1,0,56340,,,,0,ca864c629c294209e752bd462defe6257e61cf3dea9f6f32b937b76e81d12873
thread:issue:59194,issue #59194: Add numerically-stable function for angle between vectors,needs maintainer decision,28,needs maintainer decision,1,0,59194,,,,0,cf8177cebee879ead926efe404b459e585f3d405ab62b639f0990dcae7baf3cf
thread:issue:62982,issue #62982: Don't use try-catch to handle overload resolution handling in torch.ops,needs maintainer decision,28,needs maintainer decision,1,0,62982,,,,0,df3d5f458b562a9c705b593e8e2480313e7ed2da4ae5a581962fecf9720d190e
thread:issue:68110,issue #68110: zeta(1/2) returns nan,needs maintainer decision,28,needs maintainer decision,1,0,68110,,,,0,9be5f6fc17656a6410db53f2f3d376c02a2ca7286c628356254d0f834b024b9b
thread:issue:68506,issue #68506: cudnn_convolution_relu is 17x slower for 7x7 convolutions and channels last,needs maintainer decision,28,needs maintainer decision,1,0,68506,,,,0,f2e32faf4c706d7cd40b9f04029adc8cbe5152bcf018123cea9e2618853eebfa
thread:issue:71381,issue #71381: Cannot import PyTorch in Alpine Docker Container,needs maintainer decision,28,needs maintainer decision,1,0,71381,,,,0,7d5d25cd21994916e2bd1bb5c92405eec08024412c434ccf71ed088c64831570
thread:issue:71389,issue #71389: Create a nested directory while saving objects using `torch.save`,needs maintainer decision,28,needs maintainer decision,1,0,71389,,,,0,3c182d71dc9fedff83ab7c60bb022c5007889c0dcfe45b20670b9b3fc3a146cf
thread:issue:72025,issue #72025: [c10d] destruction of Store objects,needs maintainer decision,28,needs maintainer decision,1,0,72025,,,,0,71c957652d76e74150300a8505057088c76fb52f8f74fb199a80ab7f7a4b216e
thread:issue:72034,issue #72034: Bug about distributed launch,needs maintainer decision,28,needs maintainer decision,1,0,72034,,,,0,abcbe582467d1ec06855b72e6b7ce5eeaa40af691306d2b501a0c8554c14791f
thread:issue:72099,issue #72099: halt or exit function implementation,needs maintainer decision,28,needs maintainer decision,1,0,72099,,,,0,ae60a3eec404c39a7e1c435131d658b6bebc129eed49954dd10536aeeda482b1
thread:issue:72107,issue #72107: DistributedDataParallel creates too many threads,needs maintainer decision,28,needs maintainer decision,1,0,72107,,,,0,213fd4be98d1db6197cf9b12d7ab98f890eddf4ca1940492f3bd8e51e6c3c404
thread:issue:72346,issue #72346: torch.distributed.new_group() should consolidate the type of rank list,needs maintainer decision,28,needs maintainer decision,1,0,72346,,,,0,f3d9e3730570195c8e4c25f899b23d38bb832dca2a2e48f2a4949181ee31abad
thread:issue:72366,issue #72366: Custom ProcessGroup Destructor Not Correctly Called in PT 1.10,needs maintainer decision,28,needs maintainer decision,1,0,72366,,,,0,852ee9e1b9bd20b86c95830525d0a2835534a2a7e871f84dab9a2e3825578572
thread:issue:72435,issue #72435: ProcessGroupWrapper: Additional Improvements,needs maintainer decision,28,needs maintainer decision,1,0,72435,,,,0,92c3301e1e92995dcee936453074705c6c50535c8751310f5fbce743625bdce0
thread:issue:72835,issue #72835: torch.distributed hangs at barrier(),needs maintainer decision,28,needs maintainer decision,1,0,72835,,,,0,73a03e411533c2fc112e230622d358a547f26a3cb6cf828295aae8e9a28770ed
thread:issue:73081,issue #73081: Additional out of the box DDP comm hooks,needs maintainer decision,28,needs maintainer decision,1,0,73081,,,,0,da3eedd109454110f94e7a9aed01f0b8d432029d8fc8c700a260cc92fe5e06f5
thread:issue:74162,issue #74162: consume_prefix_in_state_dict_if_present can not remove prefix of _metadata.,needs maintainer decision,28,needs maintainer decision,1,0,74162,,,,0,0f784dd6f48babf662cc81b8e20d1a9b0c7d477ab343807789f44f08b8613e6e
thread:issue:74282,issue #74282: RuntimeError: cuDNN error: CUDNN_STATUS_VERSION_MISMATCH for torchvision models,needs maintainer decision,28,needs maintainer decision,1,0,74282,,,,0,e5e6b56dfcdffa9bcc010cd501f6fdc2c1cdfef43936676316b1c3d803ba0103
thread:issue:74323,issue #74323: [c10d] `gather`/`scatter` inconsistency behavior on non-dst/src rank,needs maintainer decision,28,needs maintainer decision,1,0,74323,,,,0,c36efc773bfe8aff68a4e1a29bfcc56330614807118151497fa52629e20bb5cc
thread:issue:74415,"issue #74415: Warning: ""Specified kernel cache directory could not be created""",needs maintainer decision,28,needs maintainer decision,1,0,74415,,,,0,a8abc9fc1913c5d4fa4ee59fefa32c3671b9e43d76e14fa87dc7f40cd19e614f
thread:issue:74602,issue #74602: My_rank in the implementation of torch.distributed.scatter_object_list should be global,needs maintainer decision,28,needs maintainer decision,1,0,74602,,,,0,c850793367623d4d92675c2ceb88afb498c8b93e0ea7d7c90c2ab6fd5fbe902a
thread:issue:74666,issue #74666: Re-initializing torch.distributed process_group hangs with destroy_process_group,needs maintainer decision,28,needs maintainer decision,1,0,74666,,,,0,8030f0646de919f1171104576bee4ff853768c7668063f78dfd1aa2b2cf1c949
thread:issue:74746,issue #74746: Change the type hint for nn.Module.__call__ to be friendly to overrides.,needs maintainer decision,28,needs maintainer decision,1,0,74746,,,,0,e0cfc954b42478e804e53efa94d8a5d89187f40561c9d8538876923e71e2bc90
thread:issue:74824,issue #74824: C10d Elastic Training master_addr ERROR,needs maintainer decision,28,needs maintainer decision,1,0,74824,,,,0,e8e7f5df9a4c6b176da99d33c8e8aa50b2b67f229212fa67b174634b4a978258
thread:issue:74839,issue #74839: Feature request: Root-finding functionality (like scipy.optimize.root_scalar),needs maintainer decision,28,needs maintainer decision,1,0,74839,,,,0,9f8b7a0ed56692d92f7457f7875f556c5820999021998b6b2f1072b2c55ba9fc
thread:issue:75031,issue #75031: comm hook error in BWD pass,needs maintainer decision,28,needs maintainer decision,1,0,75031,,,,0,0f623d5f2338c2517448790c64987110adc65d6e7bbd103dea655a4b28cf6451
thread:issue:75087,issue #75087: extend torch.distributed's `--tee` to log the nodename,needs maintainer decision,28,needs maintainer decision,1,0,75087,,,,0,16e2649e47039ea9d0f4a4a23207b2e8241d97d07f8b35ece5f2a6a77660f6c1
thread:issue:75225,issue #75225: One of the backends for `lu_factor` fails on windows.,needs maintainer decision,28,needs maintainer decision,1,0,75225,,,,0,0a44d86e7663fed92c783b4aea50df85e5b6a7e73744302c4258215e7962ff0c
thread:issue:75334,issue #75334: ProcessGroupNCCL is relying on UB to support bool data type,needs maintainer decision,28,needs maintainer decision,1,0,75334,,,,0,0b4225c08b500ef1d068593bde292e46bfad5e5fa3d67129d995e93b972388d1
thread:issue:75657,issue #75657: 1.11.0 distribution train different with 1.8.1,needs maintainer decision,28,needs maintainer decision,1,0,75657,,,,0,9e6ae8075e4796313c01e1823261ceed17a7034d91690ee186c4f799ecd5bdb2
thread:issue:75936,issue #75936: AllGather with backward support async_op=True,needs maintainer decision,28,needs maintainer decision,1,0,75936,,,,0,c2e07e11844e98a7e31fa63a5857fffed5632b3bd0921a056c94bcb5d8774c99
thread:issue:76185,issue #76185: Ensure custom Function are correct in double backward setting,needs maintainer decision,28,needs maintainer decision,1,0,76185,,,,0,8f6d772869594679adb0c22515b38332cf2353a2b9795ec937dcfc0fcd96486d
thread:issue:76287,issue #76287: torch.elastic fails to shutdown despite crashed processes,needs maintainer decision,28,needs maintainer decision,1,0,76287,,,,0,3977bcb770233219bd85ed27c27f8738e2d37c2a76e8574ccb8679848e4754d7
thread:issue:76344,"issue #76344: [ONNX] Use topk to export max(dim,keepdim) to onnx",needs maintainer decision,28,needs maintainer decision,1,0,76344,,,,0,39b5676ef545a5aad70af03b7d321703ef8d09d71638702938e7882c7647f78c
thread:issue:76387,issue #76387: [checkpoint] Make prepare_sharded_tensor_read and prepare_sharded_tensor_write public,needs maintainer decision,28,needs maintainer decision,1,0,76387,,,,0,87b6f80daa6837bb3d0e13b035517fa04001ca639646932a1b8ba2fb6e007a13
thread:issue:76389,issue #76389: Error in DistributedDataParallel with 'CPU' device,needs maintainer decision,28,needs maintainer decision,1,0,76389,,,,0,24de27f4cef40c24223f20c526b739546d5eb4e5c5393636e44713f720c746e1
thread:issue:76395,issue #76395: [checkpoint] SPMD distributed checkpoint coordination,needs maintainer decision,28,needs maintainer decision,1,0,76395,,,,0,46df4dd44f75d711b01d75005f5f7dadc3300265d6eb06cd03e7c394c97ba0cf
thread:issue:76527,issue #76527: Non target rank receives result of 'reduce' op when backend is 'gloo',needs maintainer decision,28,needs maintainer decision,1,0,76527,,,,0,7456d5b889b0f904d14035510ebd160ab24ecc9e196aa37800a89285cb7d8a87
thread:issue:76528,issue #76528: Discrepancy in einsum when done in batch vs non-batch,needs maintainer decision,28,needs maintainer decision,1,0,76528,,,,0,c89c0034b2909facfae51d5d80594c1ce5806b61f965f037f5938fbf8caacad8
thread:issue:76632,issue #76632: RestrictPtrTraits in CUDA potentially has no effect.,needs maintainer decision,28,needs maintainer decision,1,0,76632,,,,0,61b8208e8c5840803a70cf6dd5e2761e0615b49f8b389b7214ed04e3c82c901d
thread:issue:76760,issue #76760: hanging process with init_process_group(backend='mpi') cannot be killed,needs maintainer decision,28,needs maintainer decision,1,0,76760,,,,0,59a38ecc66e719badf29d36d3b9c1ab1eb676b1de420f2516c41feea53de10d4
thread:issue:76913,issue #76913: Pytorch return TCPStore( RuntimeError: Connection reset by peer),needs maintainer decision,28,needs maintainer decision,1,0,76913,,,,0,8a37181bb2ebeca3495741299ea9020a23bd19541b471c8e6dd6f7e1a8f7cb1d
thread:issue:76927,issue #76927: Can't pickle model torch._C._distributed_c10d.ProcessGroupNCCL' object,needs maintainer decision,28,needs maintainer decision,1,0,76927,,,,0,a7e10890d338cecdd888d00c1cf65a6186f43115b25c6c8f3da066c992ff396b
thread:issue:77523,issue #77523: DDP window TCP bug [socket.cpp:558] [c10d] The client socket has failed to,needs maintainer decision,28,needs maintainer decision,1,0,77523,,,,0,b3b55ca78078c9794b6439e7ae7ed876e9de06d324efaf830963d73cbbd16e61
thread:issue:78638,"issue #78638: torch.distributed.init_process_group(backend=""nccl"") NCCL version error",needs maintainer decision,28,needs maintainer decision,1,0,78638,,,,0,320279871ac51550a3480cae17dbeb301b8ce5ccefd92fa60c655de0cff39725
thread:issue:78842,issue #78842: Add TORCH_SHOW_CPP_STACKTRACES when TORCH_DISTRIBUTED_DEBUG = detail,needs maintainer decision,28,needs maintainer decision,1,0,78842,,,,0,de1b187b6c69b7037556450b3b4b254a1fddb060ca235310dde42e4ae1cebdab
thread:issue:7887,issue #7887: [Caffe2] convert ONNX to caffe2,needs maintainer decision,28,needs maintainer decision,1,0,7887,,,,0,6649a7bff81dce296c216a09206dff3905aecc935c184e1cc8a39288101d98ef
thread:issue:78885,issue #78885: [distributed_test.py] Improve `test_barrier`,needs maintainer decision,28,needs maintainer decision,1,0,78885,,,,0,07f7fa8f684fde4b863ae55e2b63e9d60852f3e4804de87e00917b3c2750812d
thread:issue:79388,issue #79388: Init connect timeout when use torch.distributed.run,needs maintainer decision,28,needs maintainer decision,1,0,79388,,,,0,0dacf60dc8731a859b3c4d2b456756c78b96adc1b78a245f653b9e851f988cf3
thread:issue:80496,issue #80496: DDP find_unused_parameters=True does not work for Sparse gradients,needs maintainer decision,28,needs maintainer decision,1,0,80496,,,,0,05802967333d5404ab1eec1ce8cb46eb8c2cf89b603aae6bac3fee5225c4eefb
thread:issue:80595,issue #80595: Overlapping Optimizer.step() with DDP backward,needs maintainer decision,28,needs maintainer decision,1,0,80595,,,,0,0ad3f32974f0d8c8eac95c68dc1c4501937e82472833681dae241f7663b0b3bf
thread:issue:80774,issue #80774: RPC: Make RRefProxy callable,needs maintainer decision,28,needs maintainer decision,1,0,80774,,,,0,7a1bc2d13f542f742f087e35a09d36195495c3ed9e04e42cb2fd746eddda8405
thread:issue:80824,"issue #80824: When running GPT trainning with megatron, the program quit due to torch.distributed.elastic.agent.server.api:Received 1 death signal, shutting down workers",needs maintainer decision,28,needs maintainer decision,1,0,80824,,,,0,2a0e3854afef0f155a157ff5e96fe78e560d5751746e043747302f61f19b97ad
thread:issue:81383,issue #81383: Investigate ncclRedOpCreatePreMulSum operator for gradient reduction,needs maintainer decision,28,needs maintainer decision,1,0,81383,,,,0,dbb0a0e453ab47d0e2cc3855f2f9bf6ba0d32031742425b7c2ec2cfb8f9dbed8
thread:issue:81622,issue #81622: [Distributed] test_dynamic_rpc_existing_rank_can_communicate_with_new_rank_cuda fails in caching allocator,needs maintainer decision,28,needs maintainer decision,1,0,81622,,,,0,38d8be7dfe0e634815f243b449d51cdc40c291b30ab79e3521049fd6df6cb929
thread:issue:82153,issue #82153: DistributedDataParallel hangs when not using GPU 0,needs maintainer decision,28,needs maintainer decision,1,0,82153,,,,0,4372c3b9ac7db2b694b9fefda0822d584a5692f4b8c7a1fa4639ea8339e819a5
thread:issue:82616,issue #82616: FakeTensor Support For Pickling,needs maintainer decision,28,needs maintainer decision,1,0,82616,,,,0,2018fa3c112c02a23af389156187a03d10e1e7795d475b59e3ff9f81fce8cc33
thread:issue:83494,issue #83494: `torch.pinverse` produces wrong output!,needs maintainer decision,28,needs maintainer decision,1,0,83494,,,,0,67c7bc4fbb35beb772d4f060788892b89e17745eff71944fd6ba998d029b2973
thread:issue:84925,issue #84925: `is_pinned()` support in PrimTorch and FakeTensor.,needs maintainer decision,28,needs maintainer decision,1,0,84925,,,,0,c6b4de21f8e1234993e090c954e3f4e119bdcd783552f42f676f188d67ee50ee
thread:issue:85366,issue #85366: More windows for filtering and spectral analysis,needs maintainer decision,28,needs maintainer decision,1,0,85366,,,,0,4081e8b4b570e45466a56a294de32894eac9fb13798b76cdff3346a84b8dbb2b
thread:issue:85949,"issue #85949: [Distributed] Loading distributed checkpoint with FSDP fails with varying key errors (pos.embedding, shared.weight)",needs maintainer decision,28,needs maintainer decision,1,0,85949,,,,0,3803b8fdb5fdc2c7c1c356cb8568364e9a0dbbf1226e91100a1b077b891fb16b
thread:issue:86456,issue #86456: `SyncBatchNorm` doesn't work with subclass of `torch.Tensor`,needs maintainer decision,28,needs maintainer decision,1,0,86456,,,,0,358437d1ba92bc23c084daf4f13b533d6e1b0b5db9f5d299f801145b1bb40406
thread:issue:86735,issue #86735: DISABLED test_reentrant_parent_error_on_cpu_cuda (__main__.TestAutogradDeviceTypeCUDA),needs maintainer decision,28,needs maintainer decision,1,0,86735,,,,0,ce06c135f4e3a53ad58f79d0555b911bdab8749f37333458e2be818499e723c1
thread:issue:87131,issue #87131: torch.clamp does not clamp out of -0 from 0 when ran on the CPU,needs maintainer decision,28,needs maintainer decision,1,0,87131,,,,0,0f491a8b99b0a1fe22f3efbb61c7aef68433fabbb7bd0f8ce00297ac6a13a966
thread:issue:87172,issue #87172: Unstable results in sin/arcsin/arccos calls,needs maintainer decision,28,needs maintainer decision,1,0,87172,,,,0,dffbc363aeff51e796d6c41eb5f1ed22ebd4237b8744333bfa327b2f7156430c
thread:issue:87556,issue #87556: The behavior of cast `NaN` is different on cpu and cuda,needs maintainer decision,28,needs maintainer decision,1,0,87556,,,,0,e6d3d675eb32f78c4e0f7fc07dd2c6bc3c167470d5357c2ac6ebdbbbf5f58e6f
thread:issue:88006,issue #88006: C++ Extensions can't import c10d/reducer.hpp,needs maintainer decision,28,needs maintainer decision,1,0,88006,,,,0,6d04159219dec74152fef6c6a37c6f51d310753def769b5d0486ac148d01c2f7
thread:issue:88096,issue #88096: Add nondeterministic alert to `torch.Tensor.scatter()`,needs maintainer decision,28,needs maintainer decision,1,0,88096,,,,0,3e6b86d29f8b47c1669cc81e6d323333d291c08e04d9b9e17d78d478b027363e
thread:issue:88109,issue #88109: Python Dispatcher registrations beyond BackendSelect do nothing,needs maintainer decision,28,needs maintainer decision,1,0,88109,,,,0,d8054085c5c7f3da268ba8cf4cfbab7b681f9549ff38cd401bfd46a59b29700a
thread:issue:88626,issue #88626: Cloud-based rendezvous backend / distributed store?,needs maintainer decision,28,needs maintainer decision,1,0,88626,,,,0,ffbd7c422a7233f0d3260a1df509b705aeb87ff2f9e965641826f1dfda9c3e3a
thread:issue:89197,issue #89197: Collective operations do not work with `torch.BoolTensor`s on `gloo` and raise `Invalid scalar type`,needs maintainer decision,28,needs maintainer decision,1,0,89197,,,,0,aa60257969e43e9e73934cceda26b00dac91e0c41ddb873868279ddecd6702af
thread:issue:89431,issue #89431: The problem caused by the parameter dim of torch.norm,needs maintainer decision,28,needs maintainer decision,1,0,89431,,,,0,4589493b451ffd892c1783374ef4b221cceabaaefad9546fd300919b259b6743
thread:issue:89546,issue #89546: Unable to launch CUDA Graph with DDP model,needs maintainer decision,28,needs maintainer decision,1,0,89546,,,,0,986c324d54e3d7d34e145588cd0124e2e7c5a685018851f66d587c85039f4b01
thread:issue:89716,issue #89716: DDP hangs on forward pass of transformer,needs maintainer decision,28,needs maintainer decision,1,0,89716,,,,0,e759e3944e3bc370d5d64656fc4495db90462acbfcb317c58af920b9ab99369d
thread:issue:89757,issue #89757: third-order gradient of torch.pow with tensor args and certain input returns NaN,needs maintainer decision,28,needs maintainer decision,1,0,89757,,,,0,6399d5718a4261258f7ad1a9f03afcbbe8f3b31685cab5e7339bf9596e162e71
thread:issue:90052,issue #90052: DDP overlapped optimizer: set grads to None enhancements,needs maintainer decision,28,needs maintainer decision,1,0,90052,,,,0,a26b6606a7357fe237de399319d1a1908a7c02d6ad9c926556f43ef6ba6009d1
thread:issue:90171,issue #90171: Option to let DistributedDataParallel know in advance unused parameters at each forward pass,needs maintainer decision,28,needs maintainer decision,1,0,90171,,,,0,cff3b02b01fc13a13c045b9e246618459c442e440d07ceeb45aedbebccf75370
thread:issue:90245,issue #90245: [Distributed] `Invalid scalar type` when `dist.scatter()` boolean tensor,needs maintainer decision,28,needs maintainer decision,1,0,90245,,,,0,c79544ea2490cddebc1130e22bc20b1bd7f6ed17fdd390ae344f370f22fd9a96
thread:issue:90509,issue #90509: Can torchrun have a shell completion?,needs maintainer decision,28,needs maintainer decision,1,0,90509,,,,0,294464f4f3ac63e562ee83ae01b1a18c1302aa8bdc109fc078657b51b5c92141
thread:issue:90574,issue #90574: Bfloat16 tensor .numpy() support,needs maintainer decision,28,needs maintainer decision,1,0,90574,,,,0,0f82ae0b4b097882dfd148eabd6e550fb48b7d704e3a58276e271f54b8d8e239
thread:issue:90688,issue #90688: In distributed get SIGTERM and run crash,needs maintainer decision,28,needs maintainer decision,1,0,90688,,,,0,2690ea4cdf1d94efda4e7f1f0204387c463a0dd424b3e101ce2f0c4d601bb3b6
thread:issue:90900,issue #90900: Check that SymPy semantics match Python semantics,needs maintainer decision,28,needs maintainer decision,1,0,90900,,,,0,d368050231fbdad4fe0661dcf5fc0c06e54c61ca4882bf6af4cc99d27107c91a
thread:issue:91408,issue #91408: `NotImplementedError` when using `torch.distributed.launch`,needs maintainer decision,28,needs maintainer decision,1,0,91408,,,,0,c23e14784889c0eccc35fa214eb541501d3facb1da7c6bf10445426483e6302e
thread:issue:91469,issue #91469: torch.compile with aotautograd does not support double backwards,needs maintainer decision,28,needs maintainer decision,1,0,91469,,,,0,c3c5638c4f93fba7ee8df2c73a33a6a1652e24e39e78395c1bdeb3122aa0045a
thread:issue:91629,issue #91629: vmap + nn.SyncBatchNorm.convert_sync_batchnorm,needs maintainer decision,28,needs maintainer decision,1,0,91629,,,,0,25ad4a62184c106f8f6b288f57e68984c5a8007892cd166e68a4fdb36ef758ce
thread:issue:91655,issue #91655: FakeTensors not moving between device properly on Module.cuda(),needs maintainer decision,28,needs maintainer decision,1,0,91655,,,,0,b67456e02d4f4cb43745d521e52062ba9d2db785b75b9102ed787918b300f718
thread:issue:92694,issue #92694: API to check for errors in c10d.ProcessGroupNCCL,needs maintainer decision,28,needs maintainer decision,1,0,92694,,,,0,d3e7fe17f03ec498fba82a71d10965e6d72eebc9bb3e40d6bb776bf1d70daeb6
thread:issue:93531,issue #93531: TorchInductor CPU Performance Dashboard,needs maintainer decision,28,needs maintainer decision,1,0,93531,,,,0,1f5568882ca157566e6605425f4865c9dbe37943bd54265f98cc072e16fe921b
thread:issue:94792,issue #94792: Compiling libtorch from Source on Mac Beyond v1.11.0,needs maintainer decision,28,needs maintainer decision,1,0,94792,,,,0,cf9385028367275a7530542ee2a65a228d7d834ac36938771afdb4d5a929f586
thread:issue:95487,issue #95487: Torch RPC on multiple nodes with GPU returns a EOF error,needs maintainer decision,28,needs maintainer decision,1,0,95487,,,,0,24ab222d33a3330aceaceef223fa593d3f81a5756856f3f1fb3bcbcfc4b4cbf6
thread:issue:95724,"issue #95724: distributed training: lots of ""Exception ignored"" at the end of each epoch",needs maintainer decision,28,needs maintainer decision,1,0,95724,,,,0,dafeba638408568b9c5a692c91424dd620fbb2edab2d112a0d256f118c463b17
thread:issue:96602,"issue #96602: [MPS] softmax returns NaN attention probabilities for large tensors, in float16 and float32.",needs maintainer decision,28,needs maintainer decision,1,0,96602,,,,0,d6fbeead2541ba61be7fc479f2c91a56201f781406e230a9f4afaf8e4cefd7d2
thread:issue:97668,issue #97668: sparse_csr_tensor matmul wrong output in bfloat16,needs maintainer decision,28,needs maintainer decision,1,0,97668,,,,0,5bc16287e0c70c1cbf52ce9cd780ed5d8724b64ddb1d8a9aaecc3ccb2d03d431
thread:issue:97892,issue #97892: A parameterized fill value for triu and tril functions,needs maintainer decision,28,needs maintainer decision,1,0,97892,,,,0,257f165891f5cfcf5b1c1abdfca71943a82afa926f0291591e86e3be54df8b40
thread:issue:99037,issue #99037: AttributeError: type object 'torch._C._profiler.ProfilerActivity' has no attribute 'MPS',needs maintainer decision,28,needs maintainer decision,1,0,99037,,,,0,780e227b188b114d4d11f9e6b14ba11edc4b203fbb2770124f86ff2f9636d04d
thread:issue:99082,issue #99082: Look into test coverage for `UntypedStorage`,needs maintainer decision,28,needs maintainer decision,1,0,99082,,,,0,a4dd899fad53d599ea7067a0fc843ffd5b22a250b632ad9c779179a2c76167d2
thread:issue:99614,issue #99614: Make broadcast_coalesced to a op for processgroup,needs maintainer decision,28,needs maintainer decision,1,0,99614,,,,0,990bc29edfa87a0d159f84e6cbce22eca48672ef0506c234760427719d9d78bf
thread:issue:99652,issue #99652: DistributedDataParallel doesn't work with complex buffers,needs maintainer decision,28,needs maintainer decision,1,0,99652,,,,0,c6b0ad01d803612e0ab3ed5128e1138e60efa7bfac71b483dc1db4bca515b1ca
thread:issue:9983,issue #9983: [feature request] Add matrix functions,needs maintainer decision,28,needs maintainer decision,1,0,9983,,,,0,b33ba9a7c2929f9632d5fcb12dbf8992871ebdc28072cd60419d7d0d976821e7
thread:issue:99908,issue #99908: Deprecate torch.distributed.algorithms._optimizer_overlap,needs maintainer decision,28,needs maintainer decision,1,0,99908,,,,0,b81042154ce5f4bce0867ce7d394507aa2759aaea1a6d330a6305b12c7021a1e
thread:issue:99978,issue #99978: torch.jit.trace can not trace buffer by Module.register_buffer() when use DDP Module.,needs maintainer decision,28,needs maintainer decision,1,0,99978,,,,0,36008e2b6831037c8348ea97c29f044fce27727e907b61c2ead23e8979d57b24
thread:issue:38019:pr:127702,pr #127702: Define PY_SSIZE_T_CLEAN before #include <Python.h>,has linked issue,27,Review active PRs and linked issues as one maintainer work thread.,2,0,38019,127702,,,1,ff66fe2437cf0f28cc8200a2c8676751a3e7187e33de62a71119b3ec1b254920
thread:pr:102148,pr #102148: Add mps support for maxpool3d,has linked issue,27,has linked issue,1,0,,102148,,,0,22a5d1e75b405bdf92365a3231b10e4d96126330154a85fcc5a469c3daada73f
thread:pr:142160,pr #142160: Introduce int32 index_fill and index_copy indices,has linked issue,27,has linked issue,1,0,,142160,,,0,99e73543c85fb9198527e367d2a3e40aa6d84975769eb427bab9bb581bbfae66
thread:pr:147365,pr #147365: Replace `fw_metadata` info with trace log hint in hint message,has linked issue,27,has linked issue,1,0,,147365,,,0,55dbbc0d03d4c23a19f23606e541aaa828f62000c757e9c15ee5613e4ad2221f
thread:pr:152104,pr #152104: Update _torch_docs.py to Fix torch.bernoulli(),has linked issue,27,has linked issue,1,0,,152104,,,0,40d1cd9e96b07e51202621f18461047246cdd63b7ba369d38d32131234b7fbfa
thread:pr:152828,pr #152828: [MSVC] Remove redundant MEM_EFF_ATTENTION_CUDA_SOURCES definition as /Zc:lambda is set globally for C++20 by default,has linked issue,27,has linked issue,1,0,,152828,,,0,1838969123a3bcf2dfbf329b14869f23d2fa64f802bbf9a7d9df5a992bbbef45
thread:pr:154395,pr #154395: Updated padding validation in max_pool functions to account for dilation,has linked issue,27,has linked issue,1,0,,154395,,,0,2dbd60a33dfa480ee1fa521009cace4af6e8cee793b673e82f157482108ac6cd
thread:pr:159862,pr #159862: Fix skipIfXpu and skipIfHpu and similar skip decorators,has linked issue,27,has linked issue,1,0,,159862,,,0,eb37edbe4e9a6fa37c35da4abe5f425e23527095bf6bfb90433da684b23605be
thread:pr:162130,pr #162130: Raise error when no record on `extra_files`,has linked issue,27,has linked issue,1,0,,162130,,,0,624b7995ec494f5db20fa4fd9c129247c8f7e208d7f62d0cca759553691c9e42
thread:pr:167306,pr #167306: Fix integer overflow in (fake) quantization,has linked issue,27,has linked issue,1,0,,167306,,,0,da6e55b36e78985c39d6ae25d53b851710f951648446a56be546a2e65b04dec3
thread:pr:173644,pr #173644: [TORCH] [JIT] Change emitDefautlArgsWithOutArgs -> emitDefaultArgsWithOutArgs in export_module.cpp,has linked issue,27,has linked issue,1,0,,173644,,,0,dc48e4f27321b0331aeaf1c1c89cebd6a7c9355905fa453cfaeb7c96fa0a3c19
thread:pr:174882,pr #174882: Fix torch.histc float16 precision inconsistency,has linked issue,27,has linked issue,1,0,,174882,,,0,0ed74a4ed566006c235ffd9851389377fa890cd39eaf4ddf3bbd0f624e6cc946
thread:pr:175537,pr #175537: Fix: Exclude third-party headers from PyTorch wheels,has linked issue,27,has linked issue,1,0,,175537,,,0,aa52aef7b20e884779612b6a08d4c5e01e2f92f85d0610148d276b5833d4b4eb
thread:pr:176893,pr #176893: Add save_state_dict and load_state_dict functions (fixes #150857),has linked issue,27,has linked issue,1,0,,176893,,,0,29ab6e26f76814c2c027fb08a2a47a655c1017c7406edc3b836291d2794efd80
thread:pr:178513,pr #178513: Fix interpolate nearest mode with non-integer scale_factor,has linked issue,27,has linked issue,1,0,,178513,,,0,c0e4d4bd2d6833bde3065a21061b5acfa1a9d28fee5b81b6e92e1869831bf242
thread:pr:179936,pr #179936: [MPS] Fix avg_pool2d precision by using Metal kernel instead of MPSGraph,has linked issue,27,has linked issue,1,0,,179936,,,0,497bf0c89042f72d53e07f02b9824df025a5dc346e2364b658513ad3526afa50
thread:pr:180420,pr #180420: linalg.qr: reject overlapping or aliased Q and R in out=,has linked issue,27,has linked issue,1,0,,180420,,,0,adae65be590ae4c58ec3f62c0e8bb9715907427ebecc538e6f935a294b2122d8
thread:pr:180748,pr #180748: Fix NumPy 2 deprecation warning in __array_wrap__,has linked issue,27,has linked issue,1,0,,180748,,,0,55370bddb55dcb53a9e7b1732cfe7f2647bd544e71a34f083b436b746f13f0b3
thread:pr:181052,pr #181052: Add docs for torch version information,has linked issue,27,has linked issue,1,0,,181052,,,0,d3d0904aef3ca7222f72b4bdf5a25c1ed880d7e78643439b89bd2a64eaf1e735
thread:pr:181073,pr #181073: Validate norm_type in lp_pool1d/2d/3d,has linked issue,27,has linked issue,1,0,,181073,,,0,6fe1bc39715c0e57787143b847ce772260c14ac0591815b78e1fa03b75caf458
thread:pr:181367,pr #181367: [CPU] Fix fp16 matmul ~100x slower than fp32 on CPUs without AVX512-FP16,has linked issue,27,has linked issue,1,0,,181367,,,0,259b8e5efe548324f449b85dc53afb6f120a2c46dc8e001ae63ab45926599a6c
thread:pr:182348,pr #182348: [SymmMem] Implement StoreExchange::broadcast for multicast handle exchange,has linked issue,27,has linked issue,1,0,,182348,,,0,9149513a1bd6b81d4f8077ab9192b86007310bf3fc014a1f6e3791e38b39b5aa
thread:pr:182609,pr #182609: Fix inconsistent CPU/CUDA casting for fmax/fmin/maximum/minimum with narrower out dtype,has linked issue,27,has linked issue,1,0,,182609,,,0,69ebb2f22b401ba757a648ee0e43048e50848a1d0de4313612b719b1ebd9b983
thread:pr:182803,pr #182803: Fix torch.trapz documentation signature,has linked issue,27,has linked issue,1,0,,182803,,,0,6810f9c810fa50b50b029649924e115a9ab2d1eebb4634486898c0e7dca2fc0e
thread:pr:183039,pr #183039: [c10d] Remove false-positive DeviceMesh pickle warning in async checkpoint,has linked issue,27,has linked issue,1,0,,183039,,,0,6cb3dabc9b0ab66a4246ddf68d62948054898027a39ad310264bc9dcabd53e8e
thread:pr:183052,pr #183052: Fix OOM error suggesting expandable_segments when already enabled,has linked issue,27,has linked issue,1,0,,183052,,,0,64d9bd87681758552e8db7ce1ba733c7582b78447efa7643baaa9e0cb230ad3b
thread:pr:183367,pr #183367: Avoid initializing CUDA in mem_get_info,has linked issue,27,has linked issue,1,0,,183367,,,0,712af17119a002df9123378fd8d0e5851ce5292bd0e8c3c723c87ea1740bf114
thread:pr:183522,pr #183522: Fix using `torch.cuda.amp. autocast` with JIT on Python 3.13,has linked issue,27,has linked issue,1,0,,183522,,,0,36fb2a4a752585648abb6c8acad5950914c31ff4a8b9ddb536beb2c1a08acd14
thread:pr:184058,pr #184058: nn.functional: raise ValueError for tau <= 0 in gumbel_softmax,has linked issue,27,has linked issue,1,0,,184058,,,0,f4be8cec3862223f849233ca0bd4f663e1c45e7ffa6db7acf255f1ecc77f7505
thread:pr:184349,pr #184349: Improve complex input error for BatchNorm,has linked issue,27,has linked issue,1,0,,184349,,,0,6a6528d6604ab2990b4617abdc9862881cf6924ed6cba7b4cc707758b6a355d6
thread:pr:88998,pr #88998: Fix fake tensor propagation for nvprims,has linked issue,27,has linked issue,1,0,,88998,,,0,bbb610182fcfdc9e31aab410030d2694452c68c6f473d6df2bf2c8ea23abe15b
thread:issue:113750:issue:51455:issue:71465:issue:72341,issue #113750: [docs] Add reference/decomp impl snippets to functions in online docs for hackability and educational purposes / compensate for some unclear language in existing docs,stale/low urgency,26,stale/low urgency,8,0,"51455, 71465, 72341, 75862, 76012, 77731, 78050, 113750",,,,7,c1f1a669531523631f5fdc61096c1b9a0dd7eb8462ef6137b377c40335ae6e17
thread:issue:168563:issue:168566:issue:168567:issue:168568,issue #168563: TestModule: test/inductor/test_aot_inductor.py::AOTInductorTestsTemplate,stale/low urgency,26,stale/low urgency,4,0,"168563, 168566, 168567, 168568",,,,3,51093d702330ff4e7110db25f6bc165be27bc8113a76955c5be378d4ef6a94d3
thread:issue:151558:issue:151559:issue:151561,issue #151558: 2.6: SDPA disallows masked attention under vmap,has linked PR,26,has linked PR,3,0,"151558, 151559, 151561",,,,3,ca915d99be235480448509703dd92cf4ef4c51951e9f3922e2a63e212933a9c0
thread:issue:105318:issue:106664,issue #105318: [docs] torch.sigmoid to make clear equivalence relations to other sigmoid functions,has linked PR,26,has linked PR,2,0,"105318, 106664",,,,1,bfebf41fd367582aa59975eea0834426178961018c43e2f2d1bdfe898a1ee396
thread:issue:120819:issue:144673,issue #120819: Support a context manager to set torch dtype?,has linked PR,26,has linked PR,2,0,"120819, 144673",,,,1,52cd63e632e54f844b4f81084dce06598681e43d157e080554ed5323de6dd1b7
thread:issue:127111:pr:186435,pr #186435: Fix mkldnn FakeTensor error handling,draft/noise,26,Review active PRs and linked issues as one maintainer work thread.,2,0,127111,186435,,,1,ece6e6e3581633884f96c91edb5c8f7cbd2381333a5847cd91ff251b3696c62e
thread:issue:133250:issue:134363,issue #133250: [torch.compile] tighten FX graph restrictions post-functionalization,stale/low urgency,26,stale/low urgency,2,0,"133250, 134363",,,,2,10d02cd4ce41705d57faa009c92d4ef02d501de31b2c44fe464d928979b09cd7
thread:issue:134784:issue:158610,issue #134784: Dynamo x autograd.Function x funtorch transform: intermediate sliced tensor doesn't require grad caused incorrect result,has linked PR,26,has linked PR,2,0,"134784, 158610",,,,1,157241297af2db7018bc76688add2b75484d192c85966ae2f099fd068f11d4f8
thread:issue:38208:issue:78618,issue #38208: nn.Module Abstract Class attribute overrides Child.,stale/low urgency,26,stale/low urgency,2,0,"38208, 78618",,,,1,8c3ac220078c5900f3b97b38c630e0b17715682dd857d8c7862ff30a833fe67f
thread:issue:101249,issue #101249: `einsum` is about 40x slower on CUDA than manually multiplying and summing,has linked PR,26,has linked PR,1,0,101249,,,,0,9b5ec1dd1d8efba9967b483fd976d784d10c1432d2b1d680e8abd9d98d7fbbc8
thread:issue:103584,issue #103584: torch.cuda.mem_get_info to return 0 if CUDA context isn't initialized,has linked PR,26,has linked PR,1,0,103584,,,,0,110ad9fb1bd415b2987a8bdfeff0832ff37556457a47b599a4fdad122a506812
thread:issue:104860,issue #104860: torch.nn.Conv2d/Conv1d's padding mode circular cannot accept 3-dim input,has linked PR,26,has linked PR,1,0,104860,,,,0,fb50381de31a13c347a233806a828a681afb81b9c6904e39bda041c613426aac
thread:issue:105560,issue #105560: aten._trilinear,has linked PR,26,has linked PR,1,0,105560,,,,0,ff45c777b1005602c51915dcb15664365c510e8bfa5aa3a321403cb2996b7f44
thread:issue:105839,"issue #105839: MPS memory issue, MPS backend out of memory, but works if I empty the MPS cache",has linked PR,26,has linked PR,1,0,105839,,,,0,ffb195ae34a2c010d22a5697cc623a0d2089ecc5b48396f753e137c6073e3a63
thread:issue:109652,issue #109652: InstanceNorm does not catch dim mismatch,has linked PR,26,has linked PR,1,0,109652,,,,0,3a1973780932b93956494ba1b5a4e8a1c169043df17fbbd121b2d3db6eeb98a4
thread:issue:110865,issue #110865: BCEWithLogitsLoss: Check if labels / targets are within zero and one,has linked PR,26,has linked PR,1,0,110865,,,,0,4644f80c63173c54ce762a4ad5548c003dfe35f3f01d9da0bbc418777b7c779a
thread:issue:111471,issue #111471: test_learnable_forward_per_channel fails due to integer overflow,has linked PR,26,has linked PR,1,0,111471,,,,0,df8dde6bced3b0861f2728b1710c0b6a012723d7c56f8bd73f2797fc0fa683e8
thread:issue:112586,"issue #112586: Fix docstring errors in _guards.py, _ops.py, _jit_internal.py, functional.py, _tensor_str.py, library.py",has linked PR,26,has linked PR,1,0,112586,,,,0,747ecb262b885150809d2cd494e9c77321437302fbf0c5515542f24fa0e90e90
thread:issue:117016,issue #117016: There is a performance drop because we have not yet implemented the batching rule for aten::_scaled_dot_product_efficient_attention_backward,has linked PR,26,has linked PR,1,0,117016,,,,0,4fe844615d40218df503c3cc850ef62b54dedfb867971f061fd561501bb038ec
thread:issue:117584,issue #117584: [pt2d] module register_pre_forward_hook and register_forward_hook triggered graph break when it's root module,has linked PR,26,has linked PR,1,0,117584,,,,0,7655891d1b8cfb22b01f6d14bab0e20d6a1d89b21ddce717fdbd3b2aa499e412
thread:issue:119300,issue #119300: Optimizer step pre/post hook tests should try harder to ascertain the hook happens before/after the step call,has linked PR,26,has linked PR,1,0,119300,,,,0,da50211d51929aa503751deaa58c8456de2be9ed5f0ce09369cef77b1a47fbd2
thread:issue:122729,issue #122729: torch.func transforms don't yet support saved tensor hooks.,has linked PR,26,has linked PR,1,0,122729,,,,0,a503395c35471c962a47d6f8498713f3afce70cf1d16d670be583c8888ee4317
thread:issue:125388,issue #125388: torch.multinomial raises no error when sampling from zero weight and replacement=False.,has linked PR,26,has linked PR,1,0,125388,,,,0,8b36e333b5fead3031806e59cae9c45e09421e1a39f7fc96d7f869c2a2c83b3e
thread:issue:125438,issue #125438: Fix broken linalg unittests on ARM platform,has linked PR,26,has linked PR,1,0,125438,,,,0,5561176f1e1ccf527d7d133e063ad1ba5f3123af77ae3a4e7d5f631ec719412f
thread:issue:126642,"issue #126642: [optim] add missing 'maximize' parameter to LBFGS, NAdam and RAdam optimizers",has linked PR,26,has linked PR,1,0,126642,,,,0,027f9374e48359396b742ee41577e185242835fc383769cc327f0bf81ee3a5da
thread:issue:126673,issue #126673: torch.nn.AdaptiveAvgPool2d lacks checking of input dimension,has linked PR,26,has linked PR,1,0,126673,,,,0,9a437f58715a0907a786e2b9e194a71d8018ed83cde6630fafc585e2eb568213
thread:issue:127163,issue #127163: BUG: `torch.pow(-0.0)` wrongly returns `-0.0` (should be +0.0),has linked PR,26,has linked PR,1,0,127163,,,,0,d0365487e40cb6608332d727e55183741b3aecbb1a34f3a21266d4f2b9883c81
thread:issue:129319,issue #129319: `int` type for `dims` of `tile()` without `dims=` works with a tensor against the doc,has linked PR,26,has linked PR,1,0,129319,,,,0,cb468f95b23ef041b85e7090ed7388f3fafe77c3f2fbe88b9f8ef860aacb4ef3
thread:issue:129338,issue #129338: `linspace()` can also use `complex` and `bool` type for `start` and `end` argument against the doc,has linked PR,26,has linked PR,1,0,129338,,,,0,738c85754cf6237e1c98bfba6b78856dc1412098075d2897aca8a1f53410f3e3
thread:issue:129970,issue #129970: Setting the 0D tensor of only one element to `fill_value` of `full()`/`full_like()` works against error message,has linked PR,26,has linked PR,1,0,129970,,,,0,fcdc7d9d81ba242f04e237cefc1eef62ad99583f23e874655d5af5defc7c1d97
thread:issue:130641,issue #130641: `dims` of `flip()` without `list` or `tuple` works with a tensor agaist the doc,has linked PR,26,has linked PR,1,0,130641,,,,0,db662c2d0f6b95f4f9849f626c6abf7f64db67dbde7b83ced592d9b05a8a89e7
thread:issue:131280,issue #131280: The shape of weights in lstm seems to be wrong.,has linked PR,26,has linked PR,1,0,131280,,,,0,aecb4c3d0dd6ea12e84568d6055f215258183931a9f1b49367f6cdb5ba703040
thread:issue:132368,issue #132368: Switch to new name `oneDNN` instead of `MKLDNN`,has linked PR,26,has linked PR,1,0,132368,,,,0,2edf5af7897d38d1e28df831759ab04fc80b0aaa1c4cd643ba85940eb026faaa
thread:issue:132929,issue #132929: Compile + torch.autograd.grad returns no gradients,has linked PR,26,has linked PR,1,0,132929,,,,0,3372d4df94f16d1e4b3b34cac47f03139723424f7b9decb190e3846a94dde34b
thread:issue:133078,"issue #133078: `torch.compile` support for `CXX=""zig c++""`",has linked PR,26,has linked PR,1,0,133078,,,,0,0998a6722f79331e053b3a513abdc588fee27c32f7400c6f3e4448cba031e15c
thread:issue:133462,issue #133462: PT2 inference slowdown after updating Huggingface pin on CI,has linked PR,26,has linked PR,1,0,133462,,,,0,99544c73ba617684ab71ba0b4d91dc027667b3c92cc89da5301c71d8a5f1f881
thread:issue:133592,issue #133592: [torchbind x compile] Can't register a torchbind operator that mutates a tensor,has linked PR,26,has linked PR,1,0,133592,,,,0,a54f47cae0bd11944ddfe6e20d68e88f7954544577b8421bd78a2af98710104e
thread:issue:133655,issue #133655: PythonMod vs Mod spookiness,has linked PR,26,has linked PR,1,0,133655,,,,0,ec189a9413b0ef721d65dd9260db2eb3d81e9919defb5f7a6668eda672c3d748
thread:issue:133734,issue #133734: PyTorch `torch.compile` inductor backend shows enormous compile-time regressions.,has linked PR,26,has linked PR,1,0,133734,,,,0,ffab227e6f43225575a91897e73a2480ab3b0938453331b5fbc7cbedcccb2ba9
thread:issue:133924,issue #133924: AOTAutograd not tracing AMP correctly when using PyTorch/XLA.,has linked PR,26,has linked PR,1,0,133924,,,,0,393da4813af45865e7f76c681ffbf72d206cdf10190d0887e593a4ef77a0ca22
thread:issue:134533,issue #134533: torch._assert does not work with torch.export,has linked PR,26,has linked PR,1,0,134533,,,,0,eadb11b55ef94ce3b8b284f0b1dacc562b44d7ba2081bc97a1dcae8cf92c1697
thread:issue:134841,issue #134841: LPPool2d lacks a check for the validity of norm_type,has linked PR,26,has linked PR,1,0,134841,,,,0,0691e86ca72529e9e9a260d7c6d58ebe6536132a2918d17c7d463678060ff0e1
thread:issue:134950,"issue #134950: Bert model training is not running on cpu (X86/ARM) with torch.compile() mode, generating value error.",has linked PR,26,has linked PR,1,0,134950,,,,0,dd47127697b5b4bd4847643c8fc052ef3e087ddeebcaea5615519c0b1006199e
thread:issue:134964,issue #134964: Add the pages of `torch.e` and `torch.pi` to PyTorch doc,has linked PR,26,has linked PR,1,0,134964,,,,0,8cd3309e77b2e4c1302635fe3e5091f24e412df3563188790d27c16309afad2d
thread:issue:136258,issue #136258: `to()` Method Does Not Move Internal Components of Sparse Tensors,has linked PR,26,has linked PR,1,0,136258,,,,0,913481f099f4e4c920eeede5c5696294b7bf2112037f8770adb630c6f222e7bd
thread:issue:137440,"issue #137440: `torch.pow` throws with `RuntimeError: ""reciprocal_cuda"" not implemented for 'Char'`",has linked PR,26,has linked PR,1,0,137440,,,,0,b95499e92d26f172c6e6718aae5f604527e571c0421acf11c92aacaa40ea2a30
thread:issue:137468,issue #137468: overflow in `torch.quantize_per_tensor`,has linked PR,26,has linked PR,1,0,137468,,,,0,b6a17697021564cbb9c4bb4387d9265e64347afa77fafbd36ecdc2079021e136
thread:issue:137546,issue #137546: `torch.linspace` doesn't agree with decomposition.,has linked PR,26,has linked PR,1,0,137546,,,,0,727ae6a4f85f1d0ef3ee9f9d437f81ffed5a3c66d0854a7cb6e86331c02e54db
thread:issue:139077,issue #139077: Add support for binding tensor kwargs in max-autotune templates,has linked PR,26,has linked PR,1,0,139077,,,,0,85cf1d60b2a9729500069a8a486269427cdf77d33e5145f969775d257b96fc23
thread:issue:140909,issue #140909: [2.5.0+] Running a compiled model under FlopCounterMode regresses its performance,has linked PR,26,has linked PR,1,0,140909,,,,0,31e9180f56bf23d559ad743632ebfde3cf5cbb10142d5fd6c991244d04020e00
thread:issue:142378,issue #142378: Support int step for nonfused optimizer implementations,has linked PR,26,has linked PR,1,0,142378,,,,0,1670649b0c8efeaf86cfbdc7c60c4518c20a47d3c4259a908a1a6aff6233b200
thread:issue:147212,issue #147212: dict_tag optimization leads to wrong results with relational guards,has linked PR,26,has linked PR,1,0,147212,,,,0,9207ed878ae22877fd926eeee75405dc51a883bcc9a8aae47bc510761d1760cb
thread:issue:147991,issue #147991: Verifier (in torch.export.export) does not make use of if-condition inside branches,has linked PR,26,has linked PR,1,0,147991,,,,0,f754b5ec20cec6431dce6d87f2dc8a5e471268f6fa628ec072dc1a85b6041d44
thread:issue:147995,issue #147995: Checkpoint doesn't work with torch_function if torch_function change tensor metadata,has linked PR,26,has linked PR,1,0,147995,,,,0,64cca3b2aeb03da326c5efbfc3be10f5fc349d61677e2d66923f41d50873b93b
thread:issue:149097,issue #149097: Aten arange behavior when dtype is int64 and step size is greater than range,has linked PR,26,has linked PR,1,0,149097,,,,0,de5d28e0256fd5a2815de4669f381df92ab0c408e716b3f3787cad34c5bcd908
thread:issue:149989,issue #149989: [dynamo][higher order ops] Make support_aliasing and support_input_mutation default to False,has linked PR,26,has linked PR,1,0,149989,,,,0,86b71d10e4148b84ff89c8ec89bf5c0776db95ca0abdff04b2c9fa11b1704eeb
thread:issue:150321,issue #150321: Masking out Loads from mask_modded out regions,has linked PR,26,has linked PR,1,0,150321,,,,0,641e45ea515e52f2aa058af2e6d8e99a05225169af42920667eeccdef1e367b0
thread:issue:150857,issue #150857: No way to save/load model weights between PyTorch and LibTorch (C++) in a compatible way,has linked PR,26,has linked PR,1,0,150857,,,,0,e2af7febdcce54597732217ec4faaf39b0014a68b7fd7bcdf3402829e04800e8
thread:issue:152095,issue #152095: To fix inconsistency between signature and doc on `torch.bernoulli()`,has linked PR,26,has linked PR,1,0,152095,,,,0,1da31f79e032efba3eb1aa60f796e74ac74e609781345d30ade1ab1316a9cb05
thread:issue:152178,issue #152178: Raise an Error when File Not Found in `torch.jit.load()`,has linked PR,26,has linked PR,1,0,152178,,,,0,c9cf9bb56988c91b7bb2573c9a1146da5a83429b91024c8df65184495b7f8ffa
thread:issue:152750,issue #152750: Error on padding 0-sized tensors,has linked PR,26,has linked PR,1,0,152750,,,,0,9a3e9f7797e7cd3dd6cf176d0132a1646d0cbeb1b907a3f4d9bd1d16560199f9
thread:issue:154262,issue #154262: max_pool2d padding assert incorrect with dilation,has linked PR,26,has linked PR,1,0,154262,,,,0,14000f0ac84b398dc6afc0149e5f1a0c103898733785a26c008a3eb631773cea
thread:issue:154284,issue #154284: Using `compile` on `hessian(hessian)`,has linked PR,26,has linked PR,1,0,154284,,,,0,b20dc18af5dd9c9b556b5a9826ba21f03aebcce41ec93494711e34f51be163e3
thread:issue:157776,issue #157776: Support for _Float16/C++23 std::float16_t,has linked PR,26,has linked PR,1,0,157776,,,,0,f821e41b98297027ba480e9761cfacbfb2baa06a34754040915d902bfe325b20
thread:issue:157832,issue #157832: [FSDP2] should pass args as is instead of creating new ones,has linked PR,26,has linked PR,1,0,157832,,,,0,dddcaee844c342a2cdd5f51eb0f3161673af9dacf34d4b459a9dafed8b4a1061
thread:issue:158561,issue #158561: `torch.compile` In-place operation raises Autograd error even when gradients are not required,has linked PR,26,has linked PR,1,0,158561,,,,0,40e6ba49cd0028c7969d82c8814e47e78f2d14189bce9ec3404cd8afa72b7f16
thread:issue:158975,issue #158975: Handle matmuls gracefully in foreach_map,has linked PR,26,has linked PR,1,0,158975,,,,0,52792faf0ca28e3684e5bf424a853611982842727f7647b5878f6b0cb0cb508d
thread:issue:159259,issue #159259: Use aten/src/ATen/native/native_functions.yaml to automatically generate tests for type hints,has linked PR,26,has linked PR,1,0,159259,,,,0,5c5213995351a800325725be021f6c01043c7eecdea011fa154048c4b8bb5af9
thread:issue:159594,"issue #159594: During CUDA graph capture, CUDACachingAllocator could potentially trigger cudaFree()",has linked PR,26,has linked PR,1,0,159594,,,,0,301729fba1ed539854938fc823bb6e1457b34729e03ade52eb45237fe4a2d45d
thread:issue:160016,issue #160016: Requesting improved torch.auto_grad.detect_anomaly() for NaN detection,has linked PR,26,has linked PR,1,0,160016,,,,0,363d7b30c6359fbb0c8eeaac9b1459484379a2639c1db6adfae3646c7a6beb91
thread:issue:160739,issue #160739: ConvTranspose 3D with BF16 or FP16 types is not supported on MPS,has linked PR,26,has linked PR,1,0,160739,,,,0,52b8f259eee854473ed162cbf703390519eabcffa5f139a95982c8c48abd8a01
thread:issue:160757,issue #160757: Graph break when torch.compiling an autograd.Function,has linked PR,26,has linked PR,1,0,160757,,,,0,caf867967163c3cbc73ef8d5522f8e73b1daf4ded4fa20116ceb987bb0a8f0fb
thread:issue:162350,issue #162350: Inplace modification error with softmax when combining forward and backward AD,has linked PR,26,has linked PR,1,0,162350,,,,0,2d72d3045a8dda9efe3c1d848f54a9c81773815e1d2c7878bc25e965d7cf900e
thread:issue:162373,issue #162373: Pylint E1102: torch.fft.rfft/istft/irfft and more is not callable,has linked PR,26,has linked PR,1,0,162373,,,,0,0f7919a1e12a38beb2111e657fb48b4507e0b395155a7eb34a9c896cd7b4dd4f
thread:issue:164478,issue #164478: Out of date references in CONTRIBUTING.md and Wiki,has linked PR,26,has linked PR,1,0,164478,,,,0,32fe7b78f7fd76e578713e66ac9d263eb098608eda43ef55b3226d61fe8b5382
thread:issue:164883,issue #164883: Wheels include protobuf headers (and headers for other projects),has linked PR,26,has linked PR,1,0,164883,,,,0,1a112f9fbce741c04b8b84c6b3fe669d125d02b19720c9e4fbec73dc86c761c8
thread:issue:165385,issue #165385: [dynamo] Invalid setattr's are not handled properly in general,has linked PR,26,has linked PR,1,0,165385,,,,0,d750cdce93d7fcf450cf30bbd4e78e3a47310a0517d5b96c79c18ec791cbf3ab
thread:issue:166905,issue #166905: Type Annotation for `named_modules()`,has linked PR,26,has linked PR,1,0,166905,,,,0,55b205acecfca9394d55dad9d98728db00bb37615d5ea715590b0eb9fd7b8bd6
thread:issue:167906,issue #167906: Avoid Exception Refcycle Problems,has linked PR,26,has linked PR,1,0,167906,,,,0,96d211bf887e99ff43a4cd140eca9be9c679b1a557fdc3a79f61b41cc3731fd8
thread:issue:168307,issue #168307: torch.compile error when creating nested tensors dynamically in forward pass,has linked PR,26,has linked PR,1,0,168307,,,,0,a0b81734261e1429135e79fc3bd0c824e89e5866b97bd4d08aed7ced022cc60b
thread:issue:169099,issue #169099: [RFC] Implement StoreExchange.broadcast,has linked PR,26,has linked PR,1,0,169099,,,,0,69df6e9712722fbd7a1ccf2700e276d04c6abbb1ddeec31765416655717977fe
thread:issue:169150,issue #169150: [debug mode] do not hash collectives,has linked PR,26,has linked PR,1,0,169150,,,,0,4fb88777b67e62c5836e3a070fbea999bc76f1db348df643e927c7c9f49863d1
thread:issue:169208,issue #169208: Segmentation fault in torch.native_batch_norm with empty running_var tensor,has linked PR,26,has linked PR,1,0,169208,,,,0,c1fafdff3b0f43d34c3ba280b440eebfc5fb8148f89a5083bd1b2093accd80f8
thread:issue:169236,"issue #169236: [MPS] nn.ConvTranspose{1,2,3}d ignore validation check for output_padding < stride (Silent Failure)",has linked PR,26,has linked PR,1,0,169236,,,,0,1aa75c000064a844047efa7939640be3acf8414c901a5570b1040327cb2fd2c0
thread:issue:170298,"issue #170298: `resize_` updates tensor shape metadata even when storage resize fails, creating corrupted ""Zombie"" tensors",has linked PR,26,has linked PR,1,0,170298,,,,0,01d733a9ae99db18d532c3aefcb4aeaa87532f00b2caa7431de21ed455bbb2b3
thread:issue:170370,issue #170370: [MPS][CUDA] EmbeddingBag on CUDA/MPS does not check offsets[0] like CPU,has linked PR,26,has linked PR,1,0,170370,,,,0,577ce9c70dbb0966ef841794bea2bf2c69c5189d9c68cacbc518dd8ca416897d
thread:issue:170955,issue #170955: `grouped_mm` does grouping on CPU on 5060 Ti (Windows),has linked PR,26,has linked PR,1,0,170955,,,,0,ca96944587e5b70111ed013a1ca76ff7ba1ec9b136355dbd4712519815229608
thread:issue:171981,issue #171981: torch.fx.Tracer.trace fails on wrapped forward methods with arbitrary arguments. ValueError: code: co_varnames is too small,has linked PR,26,has linked PR,1,0,171981,,,,0,59f423f60b7d0afccc7fd425aa28960e859022cacfe810bea1935d33b2d913ac
thread:issue:184280,issue #184280: GroupNorm with num_groups=0 raises ZeroDivisionError instead of ValueError,has linked PR,26,has linked PR,1,0,184280,,,,0,665717deb5dcbbac9c0ca99513a7d30665684e30a92064c3538b12d94ac927bf
thread:issue:185650,issue #185650: [Bug] torch.kron fails with RuntimeError on non-contiguous inputs,has linked PR,26,has linked PR,1,0,185650,,,,0,25ed9b12bf8d055e10e8c0c0115d9f6aaf1cb94e51933d47bb0c3b60364cc9f1
thread:issue:32864,issue #32864: [docs] Missing docs for torch.__version__ and torch.version,has linked PR,26,has linked PR,1,0,32864,,,,0,5082301a5ef59c588d143baf13a609aa50caf4d5eab726bfb06bacda2d11a5c3
thread:issue:39383,issue #39383: Pickling a Tensor or a Storage is not deterministic,has linked PR,26,has linked PR,1,0,39383,,,,0,b00b02986b5489fa12f93849655bca7d1312e1cb8dc3bf2fbaa8eddd861a5403
thread:issue:41245,issue #41245: torch.sign is divergent from numpy.sign on NaN,has linked PR,26,has linked PR,1,0,41245,,,,0,1743a2c58d292c64af86350b24243705fa042cc4b09de2cbf1f087d8471bfbf2
thread:issue:67887,issue #67887: Add more explanation on multithreaded graph building of Autograd,has linked PR,26,has linked PR,1,0,67887,,,,0,e93067039cd744fd14657713ac9b4b6fca8510cbd8d9064460568b5c2c844edf
thread:issue:67969,"issue #67969: [docs] F.multilabel_soft_margin_loss does not list arg reduction, but lists outdated size_average/reduce (true for other losses as well)",has linked PR,26,has linked PR,1,0,67969,,,,0,d9e3c9e07efc3ee71a82d737d6e2a2d680c79a72542da1accc127ffee98ced53
thread:issue:71392,issue #71392: Error in `torch.trapz` documentation,has linked PR,26,has linked PR,1,0,71392,,,,0,014fdecd5f597627efebaa71154b3e7829aa9a6dca51a39344f23b3b47f72d1e
thread:issue:77170,issue #77170: Move the MPSGuardImpl to inherit from NoOpDeviceGuardImpl,has linked PR,26,has linked PR,1,0,77170,,,,0,a153cda53868bf83978872c7de85a78b89b5733a17e748642c7b0db7f38ff27e
thread:issue:85960,issue #85960: Discrepancy in output shape for batch_norm inference mode between CUDA and CPU,has linked PR,26,has linked PR,1,0,85960,,,,0,f20afa6c56f2cd40a519384103e37ca34660895d23fbfd84dbcbf4f899217349
thread:issue:92600,issue #92600: Update quantization to make source files complient with /Zc:lambda,has linked PR,26,has linked PR,1,0,92600,,,,0,4cc422647e87e256b9697ea8a92f3290342fbc4fe8d0e4785eebc1061853961c
thread:issue:94336,issue #94336: `scatter` fails the gradient computation in reverse mode for `src` when `index` is empty,has linked PR,26,has linked PR,1,0,94336,,,,0,a16c7695dc6177b7c8ab0e1221bc08dcaa3ee1397e865d5e719fc93c64576808
thread:issue:99421,issue #99421: Question about GRU(RNN/LSTM) outputs shape,has linked PR,26,has linked PR,1,0,99421,,,,0,92a2f510774b829cbb24e7662a0a12f2be8de8e2287eb1c9e3542b5157f7e2c1
thread:pr:140372,pr #140372: [Intel GPU] Enable mkldnn::_convolution.pointwise at XPU backend,draft/noise,23,draft/noise,1,0,,140372,,,0,f8dae484ddd34c1a8b882d4a30883819f3e2e8dee8a7928753767f1349423a4e
thread:pr:179565,pr #179565: [WIP][CI] Add RHEL 9.6 build and test support,draft/noise,23,draft/noise,1,0,,179565,,,0,f3161f84650b56c40092f72660cef109968214af9855327447f978dcb1068b54
thread:pr:180225,pr #180225: Enable ACL operator fusions for aarch64,draft/noise,23,draft/noise,1,0,,180225,,,0,8bab00d10aa43b4aa5ea7357dd3741b05c9530080035db99cca7fc0264fc8367
thread:pr:180857,pr #180857: Add variable-length support for Flash Attention in XPU,draft/noise,23,draft/noise,1,0,,180857,,,0,ff52c0d9103dbc52f265a9da9cb1d7ff96a6488af1b925ff966f9761a30e7438
thread:pr:182147,pr #182147: Support XPU and other accelerators in RNG state handling,draft/noise,23,draft/noise,1,0,,182147,,,0,d3f692749eee71db1ce529da1850a04da8761be190242475983d1a85db5dcd4c
thread:pr:183366,pr #183366: [FSDP] Support replicated shard placement params,draft/noise,23,draft/noise,1,0,,183366,,,0,5586d06668e6567c68195dfc9b1d7a573cb48a070c9c80f537d044b5cafcf518
thread:pr:183530,pr #183530: feat: add XPUCodeCache tests for compilation and execution,draft/noise,23,draft/noise,1,0,,183530,,,0,6940958fd9f32920f8248c14f9cc6d85ea5fdc53aa052fab79b81112d72f86e6
thread:pr:183875,pr #183875: Add claude skill for PrivateUse1 (PU1) third-party device integration,draft/noise,23,draft/noise,1,0,,183875,,,0,f66bed6515001c0a04be3585b6e6fcafa0db5658868640cfa08b4347c6cb0d81
thread:pr:186068,pr #186068: Add skipIfXpu decorator to tests for semi-structured sparsity support,draft/noise,23,draft/noise,1,0,,186068,,,0,ff76797d1ad51f1b4a7ef5c011ee8dbbf200eab37539aae50114f851326c27b7
thread:pr:186089,pr #186089: [XPU] Align XPU unit tests decorators and overrides.,draft/noise,23,draft/noise,1,0,,186089,,,0,8eb084d9e31cbf66de9a1ab1e3c357e69c93fc088c12b8b60db43213ad0b4bde
thread:pr:186778,pr #186778: Reduce fp16 batch-1 Conv2d forward divergence on XPU,draft/noise,23,draft/noise,1,0,,186778,,,0,3365e4f73915a3d0604588ea2eddf343775561df4949c845c90969e710ca504f
thread:pr:187409,pr #187409: [distributed] Add NCCL M2N-backed symm_mem reshard,draft/noise,23,draft/noise,1,0,,187409,,,0,f135d7eea4401d52ab7c9b8f5c53832ac44b9d6f5151324a910a83aed9e72ccb
thread:pr:187554,pr #187554: Fix flaky gradcheck in _gradcheck_log_prob for simplex distributions.,draft/noise,23,draft/noise,1,0,,187554,,,0,59f234833e82767da74d35b113e5e50211003c5cc8889351c8e5e4d83152e88f
thread:pr:187896,pr #187896: Re-land chain-based dispatch for custom_op fast path with override detection,draft/noise,23,draft/noise,1,0,,187896,,,0,fb3921cf80355fb33a317067b0f02d84bd116410e1b8c12e3212603a0330c1d4
thread:pr:187930,pr #187930: Fix torch.compile validation for random ops,draft/noise,23,draft/noise,1,0,,187930,,,0,a21f1f609b9ffbc5b15ccd1e43cf7dcd1b75e1763a22de7982451c2bf85a8012
thread:pr:188143,pr #188143: Add mixed Tensor and Scalar bound support for torch.clamp and torch.clip,draft/noise,23,draft/noise,1,0,,188143,,,0,34e0fcdbfda73d94377e121a6c5aa04039e275b40cf3cb10ac240e043418989f
thread:pr:189006,pr #189006: Ensure correct sign of input for test_laplace in test_distributions.py,draft/noise,23,draft/noise,1,0,,189006,,,0,4b6cbf928cb3a9b46c735a4f0c5023c02b3e7b89d008f7524254acef50528f93
thread:pr:97825,pr #97825: Support sparse COO/CSR/CSC/BSR/BSC return values in gradcheck input function,draft/noise,23,draft/noise,1,0,,97825,,,0,ccd0d8403c127eab2a5a395d6675d17f57b3687d6864f439682b74edeb87a601
thread:issue:12659:issue:32005:issue:39279:issue:45242,issue #12659: Differentiation through Module parameters updates,stale/low urgency,20,stale/low urgency,7,0,"7313, 12659, 32005, 39279, 45242, 46886, 65913",,,,6,0596a81eaf06855b580ad2f2b8ab3cd8f6cc24742d3db34f95d3e8f320821583
thread:issue:153854:issue:28214:issue:29863:issue:46240,issue #153854: Better padding API,needs design decision,20,needs design decision,7,0,"28214, 29863, 46240, 52205, 57911, 60294, 153854",,,,7,89192b93ba92870cc814d7f28cbb699f4dacdcc619dfdf1cd26c3c74e85002ef
thread:issue:36524:issue:64038:issue:71357:issue:83163,"issue #36524: Drop _stacklevel from argspecs of F.softmax, F.softmin, F.log_softmax (for implicit dim has been long deprecated)",stale/low urgency,20,stale/low urgency,6,0,"36524, 64038, 71357, 83163, 84290, 84524",,,,7,5d3a889d6e34b85e3b6eb65904cfe73c5fdadb950531c5aae18986492e1d0d3d
thread:issue:102730:issue:125861:issue:18182:issue:20102,issue #102730: Best practices clarification for initialization strategies,needs design decision,20,needs design decision,4,0,"18182, 20102, 102730, 125861",,,,3,77f64583aebc711ca20da6944df1e81e30096f5c44a2565fa879df73b1f39dfd
thread:issue:130529:issue:30702:issue:62148:issue:94233,issue #130529: Support tuple of `int`s in `torch.unsqueeze`,stale/low urgency,20,stale/low urgency,4,0,"30702, 62148, 94233, 130529",,,,4,c95c7ffecd76f9a0fef4e767c0578ff75a49c4389a432343e833970305b64810
thread:issue:18173:issue:25883:issue:28440:issue:33227,issue #18173: [docs] How to achieve high-order derivation in my .cpp?,stale/low urgency,20,stale/low urgency,4,0,"18173, 25883, 28440, 33227",,,,3,4cbc2686ddb80e0566dd3e9c52878a3f3bc46ebae738577c8f754b954a311cc0
thread:issue:40761:issue:47915:issue:64544:issue:78262,issue #40761: Sparse Convolutional support,stale/low urgency,20,stale/low urgency,4,0,"40761, 47915, 64544, 78262",,,,3,7b429004013004d400ce42ed531ecd5a9dd430ea338f6a6f8679c51c43475cee
thread:issue:51156:issue:54138:issue:54139:issue:58036,issue #51156: [Feature Request] Support tensor creation from objects that implement the __array__ interface,needs design decision,20,needs design decision,4,0,"51156, 54138, 54139, 58036",,,,5,9ae92000439877ac31b9effc425bdb0b1dd5f64b9e1df8fb7634253d7bd20dbf
thread:issue:115852:issue:116506:issue:144012,issue #115852: `groups` parameter for `nn.Linear`,needs design decision,20,needs design decision,3,0,"115852, 116506, 144012",,,,2,1146ff41aff4d7d825834f24d925e90262ca80961364d34e819418b5c2bcb2d0
thread:issue:120139:issue:122410:issue:47055,"issue #120139: On `Dataset`, `IterableDataset` inheritance",stale/low urgency,20,stale/low urgency,3,0,"47055, 120139, 122410",,,,3,5000e569968f57dca1e8923dc1a981a8397684da874926e76bdaa310ffd85425
thread:issue:153378:issue:160553:issue:163504,issue #153378: MPS backend silently ignores dimension mismatch,needs design decision,20,needs design decision,3,0,"153378, 160553, 163504",,,,2,a979b32ccafcb609f59941444b585d21c02b7a1309879ea314915003cdf1d782
thread:issue:157547:issue:31779:issue:35527,issue #157547: Several `torch.*` functions raise uninformative `NotImplementedError`s when called with integer `dtype`,needs design decision,20,needs design decision,3,0,"31779, 35527, 157547",,,,2,ec8657c9c4737b2e2a08cb2d2624d7e5e75ae18d9284a3de02959ce445904856
thread:issue:160322:issue:161713:issue:163990,issue #160322: `torch.Tensor` methods return type annotation too broad for `torch.Tensor` subclasses,needs design decision,20,needs design decision,3,0,"160322, 161713, 163990",,,,2,b39726d210923206901d2a6f59d9aa022d0c76684d608e1cdd5b707072f7f254
thread:issue:110295:issue:125239,issue #110295: Tests modify global state cause later tests to fail,needs design decision,20,needs design decision,2,0,"110295, 125239",,,,1,47f1293b329af44b118817522fcb3f64b6b595d21283646ddee7674af05e825b
thread:issue:112342:issue:112352,issue #112342: [pytree] Pytree node registration hygeine: deprecate global _register_pytree_node; only allow enabling registered pytree extensions locally,stale/low urgency,20,stale/low urgency,2,0,"112342, 112352",,,,2,167b35dbd97a7f759fe088622c8c9730f1ebff49d922dfc3fc313c6ab2c88fdc
thread:issue:116490:issue:141832,issue #116490: Legitimize/harden differentiable optimizers in torch.optim,needs design decision,20,needs design decision,2,0,"116490, 141832",,,,1,c3e359d65728af77acdb48b86e3d92c0ef5f65c557197585bd9ea1b57857f876
thread:issue:126164:issue:72831,issue #126164: Context manager for determinism,stale/low urgency,20,stale/low urgency,2,0,"72831, 126164",,,,1,d03e2eac9cfe88dee85531246b524fdc8a9371855c1efeb6eb7245bde64e6859
thread:issue:130513:issue:159354,issue #130513: RuntimeError Occurred In getBar1SizeOfGpu when initializing torch RPC,needs design decision,20,needs design decision,2,0,"130513, 159354",,,,1,50ab34e71fb57eab9f77b9a667f6f9144f6c7ba9f5e885f649dfcc195c75393a
thread:issue:132378:issue:156616,issue #132378: per sample gradient with torch.autograd.grad,needs design decision,20,needs design decision,2,0,"132378, 156616",,,,1,8718339eeee47bad85580c783917f0f82abc1964715289821e5ee00947307e84
thread:issue:133566:issue:97575,issue #133566: Performance: Autodiff stores inputs of convolution layers with non-trainable weight,needs design decision,20,needs design decision,2,0,"97575, 133566",,,,1,1f4d64854bfed2d37802f45d71e631f6af8548b5cc81e1eb29dffe03804bc8fe
thread:issue:13811:issue:147048,issue #13811: Feature request: von Mises-Fisher distribution,stale/low urgency,20,stale/low urgency,2,0,"13811, 147048",,,,1,6bd6696335325bf4bfaf8dab8442a05331315fbbba6f6514aec96f0a9359ad7d
thread:issue:152877:issue:71117,issue #152877: Motivate Pytorch's forward mode AD APIs with training examples,needs design decision,20,needs design decision,2,0,"71117, 152877",,,,1,c619b71dcaa6486672aa0b4976b96eeda48769ab08e9bab5cb7afebce695e9a7
thread:issue:40316:issue:42779,issue #40316: Refactor the adaptive avg pool code,good volunteer slice,20,good volunteer slice,2,0,"40316, 42779",,,,1,5b2f442ba1c58e834afb9304cf1c8851e8350bbf5125e92c9ffc6a733e17651a
thread:issue:41499:issue:6662,issue #41499: Helping test example code blocks in the docs,stale/low urgency,20,stale/low urgency,2,0,"6662, 41499",,,,1,103c5d5849f9a1f19b787a63479a42b63e7a0cd2031ffdaf880a633b6656a1fd
thread:issue:55152:issue:55159,issue #55152: Better handling of OpInfo.sample_inputs,needs design decision,20,needs design decision,2,0,"55152, 55159",,,,1,0c92243d9d065756174d43a0857074f12eff146b696119b29123d9428860df2a
thread:issue:64067:issue:66824,issue #64067: Execute C++ based GTest unit tests from a python wrapper,good volunteer slice,20,good volunteer slice,2,0,"64067, 66824",,,,1,ee318c96ea238e1c1d93d10bb57587c0734f1a5fc3d8cafec936edb91396f5ac
thread:issue:73513:issue:73537,issue #73513: Wrapper tensor level confusion,needs design decision,20,needs design decision,2,0,"73513, 73537",,,,1,e8b5a194fece83815932af2571c54bc3d69c8a21530eb66799ea0615ec802f81
thread:issue:96319:issue:97500,issue #96319: views created in __torch_dispatch__ share storage but not version_counter,needs design decision,20,needs design decision,2,0,"96319, 97500",,,,1,176c2528ee461a6349f00da5c5575071d21828ceeb77233efc1e2461a143df26
thread:issue:96764:issue:96766,issue #96764: Improve checkpoint thread-safety,needs design decision,20,needs design decision,2,0,"96764, 96766",,,,1,152821cfab0104e542ac3d39b1e2dba7f2420b97bd78063297aa6ef8e81e57d3
thread:issue:100628,issue #100628: Feature request: fast way to approximate the diagonal of the hessian,needs design decision,20,needs design decision,1,0,100628,,,,0,fda7f606ca38cc2cc984f8a06070c5295700786081597358fd47d35370ced893
thread:issue:100990,issue #100990: Extending compatibility of LibTorch,needs design decision,20,needs design decision,1,0,100990,,,,0,8d5c8dd9422e12568682c11c45c3a46363a5339013959baef2e5e4d56ceb2fb9
thread:issue:102201,issue #102201: scipy.ndimage.find_objects,needs design decision,20,needs design decision,1,0,102201,,,,0,5f6996c0e266ff78c01d1df3148e959c8d867bd5be295b0abd5d877b0a0312c9
thread:issue:102337,issue #102337: Tensors that share same underlying storage to also share gradient storage,needs design decision,20,needs design decision,1,0,102337,,,,0,810b1faf29beed2c37e0121bb77264481c9d8a1083a94d5a5df0b774e222e967
thread:issue:102417,issue #102417: _view_func but without keeping original view tensor alive,needs design decision,20,needs design decision,1,0,102417,,,,0,b741002fa17c3bbe86a00d876bb76460e2035c0d8e00162ac79f38ff4a35ce51
thread:issue:102447,"issue #102447: Add additional ""sigmoid"" approximation to GeLu activation?",needs design decision,20,needs design decision,1,0,102447,,,,0,07f2542c4ac0aa4aba6cbd0a1ab61c3f2dab1757b6833c71a408f6e229dcac21
thread:issue:102905,issue #102905: nn.ChannelShuffle1d,needs design decision,20,needs design decision,1,0,102905,,,,0,e6628d9bd989fe230bdb880728827ad83f2baf4a9a5106d041cc87a130225e8d
thread:issue:102938,issue #102938: Support for efficiently processing categorical distributions with varying dimensions,needs design decision,20,needs design decision,1,0,102938,,,,0,f65ac47c03d70402a561137621026338941f89812058a1ceca32e1e3dad08304
thread:issue:103109,issue #103109: Increased / more verbose type aliases for improved readability of user defined content,needs design decision,20,needs design decision,1,0,103109,,,,0,31c5d649b07a057d375deeb826f3b8419624481e20416d8323e5406ae39d03b9
thread:issue:103449,issue #103449: Asynchronous CUDA AveragedModel,needs design decision,20,needs design decision,1,0,103449,,,,0,308cf9f0f8e6495320940a36bc4bd8dbd4431f6451f1a6ceb7702a73799ddab8
thread:issue:105365,issue #105365: Quadric Layer,needs design decision,20,needs design decision,1,0,105365,,,,0,227f132a45c89b2803a5faefd07f007540518956f57310a7d39151039d287089
thread:issue:106360,issue #106360: Improving save_on_cpu's performance by overlapping memory transfers with compute,needs design decision,20,needs design decision,1,0,106360,,,,0,f4f2be01af1f4066691b2b7f8ea5c159a12fea6c8b61fea9432f8e426656b662
thread:issue:106784,"issue #106784: [ux] Suppot torch.tensor(set([1,2,3]))",needs design decision,20,needs design decision,1,0,106784,,,,0,b92951e8dba2b4382008f1fb538abbdf2fcaa8d934ef36e6e70ac469e69063be
thread:issue:107224,"issue #107224: [CPP API] Add Adadelta, Adamax, ASGD, NAdam, RAdam and Rprop",needs design decision,20,needs design decision,1,0,107224,,,,0,6697703d6592e17a833e71c27d424c19e58783ed330edf0d63021910f29c63d7
thread:issue:10756,issue #10756: Unexpected Behavior when Pointwise Operations Write to Expanded Tensors,good volunteer slice,20,good volunteer slice,1,0,10756,,,,0,8cb311464af8ed313f578be7059ffbd45cdae003d474e24753810fe0ddd16518
thread:issue:108090,issue #108090: [Optimizer Perf] Improve speed of _init_group to c++,needs design decision,20,needs design decision,1,0,108090,,,,0,7542fd28d54362e3573b6d493d40878e21404dcded2f64b111eae8d317181b34
thread:issue:108491,issue #108491: Suppport Fused AdamW on CPU,needs design decision,20,needs design decision,1,0,108491,,,,0,07110b69913ac67e6d9465dc86c89a4f3b18808f9d785582c8a7d38fba30d59e
thread:issue:108692,issue #108692: Adding Maximal Update Parametrization (µP) to torch.nn.init,needs design decision,20,needs design decision,1,0,108692,,,,0,750a3dd9c89c8b23260a2c083df8a182090455430fff7a4d3fdf19a8e9fc13cd
thread:issue:108784,issue #108784: Collect more env variables in `collect_env.py`,needs design decision,20,needs design decision,1,0,108784,,,,0,9e6ffab5f01cd98c0d05dad13474070a20d5e38a350e68526890b76a382e8597
thread:issue:109052,issue #109052: Allow reductions to write into pinned memory,needs design decision,20,needs design decision,1,0,109052,,,,0,d767161f49b5ea40cb11bd57d6b050b0d448820f1c3b938429f80eb619fae9dc
thread:issue:109582,issue #109582: Make Dropout take a dim=... argument,needs design decision,20,needs design decision,1,0,109582,,,,0,c9d6d11199d263b8a162168a1c957daa195b0acad57a56fb9cfe9abef9980ea9
thread:issue:109691,issue #109691: Extends the functionality of `nn.BatchNorm1d`.,needs design decision,20,needs design decision,1,0,109691,,,,0,b47bb0e3e437e2ac1d9f30841943b6bb1a305b0d7273cf6452321bf2690da0e1
thread:issue:109948,issue #109948: Simple script segfaulting when grad is enabled,needs design decision,20,needs design decision,1,0,109948,,,,0,201b564e4c1d11860412264c8a1ebab601775bc0fe8dd24557942f6a3872f6e1
thread:issue:110780,issue #110780: backward and grad behave inconsistently w.r.t. set_ on leaf variable,needs design decision,20,needs design decision,1,0,110780,,,,0,161bec9f525927956dd26b2588e24489d2de4c7a34840c41feb88b6838fdae59
thread:issue:110885,issue #110885: Depthwise conv3d slower than normal conv3d,needs design decision,20,needs design decision,1,0,110885,,,,0,7729065d68093d96f04ff82406aad932bfde58006bf4b3f82ca794e0247e6778
thread:issue:111538,issue #111538: Propose to add constant padding mode to the `torch.nn.functional.grid_sample` function,needs design decision,20,needs design decision,1,0,111538,,,,0,32f61a32e7c45c0fbcd6e34ed0e44bc74baa9c39bdd66d575ed7533b5b84bda7
thread:issue:111563,"issue #111563: Higher-order derivatives extremely slow, increasing exponentially",needs design decision,20,needs design decision,1,0,111563,,,,0,99bf8e085c4a9468cd825aedb6bfbea0fac18ba6037c142df4da652a3ae39d4c
thread:issue:113530,"issue #113530: Make ""torch.load"" multi threaded process",needs design decision,20,needs design decision,1,0,113530,,,,0,80f439689f23d28d946298a9c4f6cfa9639f26d779b6a6f4a23e6ae575cba9f6
thread:issue:113730,issue #113730: Can the official provide cpu offload method for large model inference?,needs design decision,20,needs design decision,1,0,113730,,,,0,04652201acdb45a91d5bd1d130b9b76b9304afd7166a618188cf968e6200b437
thread:issue:114462,issue #114462: Transformer with convolutional network,needs design decision,20,needs design decision,1,0,114462,,,,0,6715f5016cdd626c52f3dc3828323d762934a81ede4848f0bddcdf0e6f2704fd
thread:issue:114650,issue #114650: Error when calculating the Jacobian of torch.conj using forward-mode differentiation,needs design decision,20,needs design decision,1,0,114650,,,,0,7f2348443dcadee0f829e15cdd9aa70427646114aa62dacd58f8aac3ee10e906
thread:issue:115273,issue #115273: Consider improving project management (12 pages of HighPriority tickets > 1 year old),needs design decision,20,needs design decision,1,0,115273,,,,0,80f2af7c6892971b454ceb6ac55803cad489bab294cb85d6bf5bbfd77bb3c778
thread:issue:115585,issue #115585: `torch.nn.CTCLoss` doesn't seem to properly support noncontiguous inputs for 1d `target` with `target_dtype=torch.long`,needs design decision,20,needs design decision,1,0,115585,,,,0,51e82e380c363a6ddcef0a8959321a956921d7ceeeae233754ae0902b2b4e5fc
thread:issue:115588,issue #115588: `torch.nn.KLDivLoss` fails `test_grad` and `test_gradgrad`,needs design decision,20,needs design decision,1,0,115588,,,,0,3f6382a59297cd6ebeca1b781833b2b23900f5b9b92f97f5f687f7b2ecac356c
thread:issue:115937,issue #115937: Building extensions with CMake,needs design decision,20,needs design decision,1,0,115937,,,,0,62ef09cad88b458c64daab0d49f9fa7f78699de225f775ed602f935ba606c73f
thread:issue:115993,issue #115993: [FR] Add a way to reserve memory that survives `torch.cuda.empty_cache()`,needs design decision,20,needs design decision,1,0,115993,,,,0,62fbf626cb60222304188707a77bf70007b4433037aa32fb1ee8e8a91ab7cc3c
thread:issue:116354,issue #116354: Abstract torch.device for GPU/NE/TPU computations in the cloud-based agent,needs design decision,20,needs design decision,1,0,116354,,,,0,13de85d8024238f0a7d71e2f9c6e52dfb7b8d90bdb7fe9fedc6f49fc42921fcf
thread:issue:116555,issue #116555: The performance of `torch.index_select` and regular indexing differs dramatically based on the size of the tensor it is indexing,needs design decision,20,needs design decision,1,0,116555,,,,0,b8fa51af3e98a53dadddf5e0ed32dba745d86bd3fc71447c37c11414ec0fa8f9
thread:issue:117217,"issue #117217: BatchNorm1d, running_mean and running_var",needs design decision,20,needs design decision,1,0,117217,,,,0,86c8704587c90acf03bfdfd0d749794eb4e750e2972796e3e36ac3d96f7d7c87
thread:issue:117806,issue #117806: Supporting custom attributes with `__torch_function__` tensor subclasses,good volunteer slice,20,good volunteer slice,1,0,117806,,,,0,ed500dce32cd377e983b7e05a7d906e8ba4e7668981baab2440daf62e45ecfc1
thread:issue:118133,issue #118133: Better test disabling system,needs design decision,20,needs design decision,1,0,118133,,,,0,9d8b0fefe705a39685ea8d424c5f01d3055707c6a942fd47fa706d4d1a1890e6
thread:issue:118225,issue #118225: Support for banded matrix operations,needs design decision,20,needs design decision,1,0,118225,,,,0,eae4a3ce1b42934dc87f4ec17073dfddf156e54474c7e8f85861ad1aa81c2a81
thread:issue:119475,issue #119475: `autograd.Function` leads to `pylint` warnings in valid user code,needs design decision,20,needs design decision,1,0,119475,,,,0,2572af355b2b3ac98bbeb826cce5af396f44062e24097a7822a036408be19ae7
thread:issue:120072,issue #120072: Please consider adding more optimization support for multi-layer recurrent neural networks to improve fitting speed.,needs design decision,20,needs design decision,1,0,120072,,,,0,fcb317d4d2da342ae39d86298fa4b2f69f65f2fd1f6cb0a4e4327233abc8cda8
thread:issue:120107,issue #120107: [feature request] keep gradcheck numerical jacobian from being depeciated,needs design decision,20,needs design decision,1,0,120107,,,,0,ebf6579bb10eb8c490c582e36f3b860dd3cd7b08cae9392f99a99075587d9849
thread:issue:120189,issue #120189: Making Mamba first-class citizen in PyTorch,needs design decision,20,needs design decision,1,0,120189,,,,0,ddf181a1db0c3fd560225616c1624a49ebed04d60a5716156e5fdee74d8e9ba5
thread:issue:120589,issue #120589: Softmax plus one for Flash attention,needs design decision,20,needs design decision,1,0,120589,,,,0,7a9826cfd6a7d84f1aa7c7b71826d6342f60b0ef7f19d135973ee99a07f9b11c
thread:issue:120700,issue #120700: Invert the model architecture from the unknown neural network model (the python object of the model),needs design decision,20,needs design decision,1,0,120700,,,,0,27f56d4d70203bb200ddd2cf601d488d325c8bd35fe9bf1dd03f9b440230f213
thread:issue:121791,issue #121791: [RFC] Set Float32 Precision for CONV/RNN,needs design decision,20,needs design decision,1,0,121791,,,,0,093db45380f81366331932eea8951079c39ef73b1766f186327ea1132f209342
thread:issue:121933,issue #121933: Please provide `total_repeat_length` field to PyTorch `repeat` / `repeat_interleave` API,needs design decision,20,needs design decision,1,0,121933,,,,0,bdb63e95c37d6e31d9e142bab28ba036968912496cec42e3d56654d3a243e569
thread:issue:122099,issue #122099: No reduction of disk usage on slices of torch tensor,needs design decision,20,needs design decision,1,0,122099,,,,0,b86fd86091072b8fa1c75ab18476c20e02c5293d3b95640be2194a180afc6ee3
thread:issue:122551,issue #122551: Global Response Normalization (GRN),needs design decision,20,needs design decision,1,0,122551,,,,0,c5e0635f20fd75cc6908c43e2568ff12eb006671212491603f87fe4b7ba66a12
thread:issue:122680,issue #122680: Please consider adding MIG (MI-rror with G-radient modification) to torch.nn,needs design decision,20,needs design decision,1,0,122680,,,,0,fdcaed65594b6d5b052dd9cc780c377d5cf87eeac4f2c887c9a95971c3b9220f
thread:issue:124902,issue #124902: Add support for IPC features for PrivateUse devices,needs design decision,20,needs design decision,1,0,124902,,,,0,69b18271d5ed1a7e7eac4e92e95727e3d8d522e3303da6ae9ab01fcc0f577204
thread:issue:125223,issue #125223: torch.uniform_() is single-threaded on CPU,needs design decision,20,needs design decision,1,0,125223,,,,0,692e7654b84d5c15b3a48d98b5adb181167c50a8b517d202c2e98579d26858c0
thread:issue:125422,issue #125422: Tensor roll with different shifts,needs design decision,20,needs design decision,1,0,125422,,,,0,a7ef4d0aa6074333c08fd1f543e409585441cf2e17be7e3f4f9441587c93afb6
thread:issue:125767,issue #125767: RFC: Integrate Arm Compute Library (ACL) into PyTorch as a submodule and add build support in setup.py,needs design decision,20,needs design decision,1,0,125767,,,,0,dc2ff1318283424abbcb5d60562c81c049abfe2deab96a2786f82ae772ad10d3
thread:issue:126762,issue #126762: Feature request to add the SReLU activation function,needs design decision,20,needs design decision,1,0,126762,,,,0,d8891adce7442b4af3396ae8c5b3bf58a4db89176ca3950615d2fadf1dd5b2c4
thread:issue:127284,issue #127284: Fused AdamW maybe should accept lr_dict directly?,needs design decision,20,needs design decision,1,0,127284,,,,0,fc539f0ab3be2dbee03b10fcdf2659081b88b9f5712473ec5fc3ecd849eb35cc
thread:issue:127699,issue #127699: torch.Tensor for optimizer parameters,needs design decision,20,needs design decision,1,0,127699,,,,0,e509128105f5f93e5110af5d387e2301a5ff8d91905c9b31c0dacddc25809043
thread:issue:127932,issue #127932: Add SinusoidalPositionalEmbedding module for use in Transformers and Diffusion models,needs design decision,20,needs design decision,1,0,127932,,,,0,0c56fa8f7d724c05db0baccc0b28f9b40d6d2e3ae4a5d74df1d12391da8fe9a5
thread:issue:127996,issue #127996: Tensor subclass constructor via autograd function is error prone re mutation,needs design decision,20,needs design decision,1,0,127996,,,,0,061c27626e94f266e8cdc1304e0cfaa7f61c692b2b3867fbe8072127bd3abf2c
thread:issue:128283,issue #128283: control output type in matmul,needs design decision,20,needs design decision,1,0,128283,,,,0,9e1a5a6c572660cca1d90a7fb497f9ab0ea2133feca3dd4919fbb0d2bdfc8257
thread:issue:128380,issue #128380: Conformal Prediction framework to enhance reliability in risk sensitive industrial applications,needs design decision,20,needs design decision,1,0,128380,,,,0,c79303e488de95210f00c7a56aac15b21eba2ca3af11f1f07e8c31ddbc54a552
thread:issue:128793,"issue #128793: GradType: a subset of dtype that is differentiable, containing all float and complex dtypes",needs design decision,20,needs design decision,1,0,128793,,,,0,799adce05e5bf46b9987088ade0d79c21f98d2d18c9404749e28bd3c3dfad25b
thread:issue:128934,issue #128934: version inquiry,needs design decision,20,needs design decision,1,0,128934,,,,0,60a654d3a9fcd912bfde56290f514cded1a7f438b4655ad475a838ca20e9155a
thread:issue:129867,issue #129867: `saved_tensors_hooks`/`_pop_saved_tensors_default_hooks` API extension,needs design decision,20,needs design decision,1,0,129867,,,,0,64f55e6d3282f8e664469c1622fd0196c31a479188a1b8689110d25b9314bd85
thread:issue:130428,issue #130428: Extend the device context manager to support existing tensors as well,needs design decision,20,needs design decision,1,0,130428,,,,0,9ef4d684ec4d18136d0e0e961b842348d5fe25626d0159684234d640b15a544e
thread:issue:130451,issue #130451: vmap over torch.allclose doesn't work,needs design decision,20,needs design decision,1,0,130451,,,,0,c7a4283f14b5fee91c50c43341192805b7b5748e0f2ec4c5814e4971b9444cff
thread:issue:130480,issue #130480: [Feature Request] Lazy upcasting for mmap'd state dicts,needs design decision,20,needs design decision,1,0,130480,,,,0,6124e519288bc780f8f3a5710d681ee264ea59c8718cfc1eab57ef76d21f0835
thread:issue:131881,issue #131881: Decouple some generic APIs with CUDA building,needs design decision,20,needs design decision,1,0,131881,,,,0,d997ecebe83f28327c1491780aed510240a578ce13aae11b48f408f728262a89
thread:issue:131927,issue #131927: exclude_dim=[...] as a keyword argument to reduce type operations,needs design decision,20,needs design decision,1,0,131927,,,,0,f60b642b49023301c680881f3d722326911e448d75007c978c6c54fc3fb39677
thread:issue:132446,issue #132446: [Feature Request] Support Custom CPU Allocator to Enable torch.cpu.change_current_allocator,needs design decision,20,needs design decision,1,0,132446,,,,0,ac5d039a5ca5f65955267ad070427058b0ffe4d3a41946a3617dee1e268c0909
thread:issue:133288,issue #133288: Give PrivateUse2 and PrivateUse3 the same treatment as PrivateUse1,needs design decision,20,needs design decision,1,0,133288,,,,0,3a23efbdc690c4fdc8e730d92b1fc646b72a04769f62348be8f95888956b33af
thread:issue:133466,"issue #133466: In-place operations should be deprecated and in the future, they should be removed from PyTorch.",needs design decision,20,needs design decision,1,0,133466,,,,0,0ddcfec88a7740d3f220248f4472e33abe83cbc2e2ae4ba5176ab9af1c0a43b3
thread:issue:134325,issue #134325: Proposal: When serializing to state_dict/parameters use different separators for ParameterList and ParameterDict,needs design decision,20,needs design decision,1,0,134325,,,,0,1efbae8fddfda498855e1ae4653b320d4fca1d38602458700f86edc6238bb0ad
thread:issue:135816,issue #135816: Can we remove @torch._dynamo_disable from torch.utils.checkpoint.checkpoint?,needs design decision,20,needs design decision,1,0,135816,,,,0,b36f9e1234d24b4bb7fb36c9f24a7dad8ec879131e848807e89ea9832b794461
thread:issue:136433,issue #136433: SAC doesn't support nesting with different recompute plans,needs design decision,20,needs design decision,1,0,136433,,,,0,a54bd934d1b38201104701dda68263c06d3df35f661ef85ea3b8a580b6344057
thread:issue:137680,issue #137680: [FEA] Add nvsci support in pytorch multiprocessing module for nvidia jetson devices.,needs design decision,20,needs design decision,1,0,137680,,,,0,2bdf4ed468d232485444805f212b78ebc1c9d7fc9a66c06b575ad1631e4b72c4
thread:issue:138118,issue #138118: [autograd] do not clear all variables when retain_graph=False,needs design decision,20,needs design decision,1,0,138118,,,,0,b56d987af2dcec5a726f8a34005990fe653feb7401ee2b298768c1fee3df5b69
thread:issue:138502,issue #138502: Contribution Proposal: ScatterND Implementation for PyTorch,needs design decision,20,needs design decision,1,0,138502,,,,0,1adf29b42b93e6e670d197158800be0abe60ecf461890f53ba2d16430a4730b8
thread:issue:138906,"issue #138906: RuntimeError: ""addmm_cuda"" not implemented for 'ComplexHalf'",needs design decision,20,needs design decision,1,0,138906,,,,0,c3eda90e3bedafa8fd9c17933d209e64f474372626478049df45dc08144bc4dd
thread:issue:138952,issue #138952: [Feature request] Flag to prevent certain parameters from being typecast by nn.module.to(),needs design decision,20,needs design decision,1,0,138952,,,,0,aa23b7421ff49680c79e739172f13963ba67fe98cbffa71c93d4b6c516750a1a
thread:issue:140378,issue #140378: Differentiable quadratic programing solver,needs design decision,20,needs design decision,1,0,140378,,,,0,f9f919887ca48151925c491ec40ca8120b7dc7942bd400c6b8c26d11c17ac181
thread:issue:141627,issue #141627: Support reductions in FlexAttention's score_mod/mask_mod,needs design decision,20,needs design decision,1,0,141627,,,,0,849274de9c5edfcad58a4a18c4c6795f4c3764c34f4c8bf889bd893b9bcd5f3d
thread:issue:141678,issue #141678: torch.cuda adds support for the changing `CUDA_VISIBLE_DEVICES`,needs design decision,20,needs design decision,1,0,141678,,,,0,92bd04a254d21c629e6a624d4431b87902ac72e469da523c4fa10749d23e8322
thread:issue:141994,issue #141994: Add torch.set_default_int_dtype() / extend set_default_dtype() to allow setting default signed integer dtype,needs design decision,20,needs design decision,1,0,141994,,,,0,8be36efb9bb7340a341eb7ac91e6ae4c2d74b9f6399233c6e572c5b942648d08
thread:issue:142320,issue #142320: Proposal for Official Cross-Platform Support for 3D Gaussian Splatting,needs design decision,20,needs design decision,1,0,142320,,,,0,512ada4173420b2e28b174e53435ae4a2f1b4016e920f3a70cdb5a74f4b1b75a
thread:issue:142381,issue #142381: Add a linter for Docker images which are not included in docker-builds.yml,needs design decision,20,needs design decision,1,0,142381,,,,0,b958afb2cf9b5a9e4785f5680d4cdad32be31fd27e97abacce319b4154bcd13d
thread:issue:144159,issue #144159: Allow generic python data structure input for torch.autograd.Function,needs design decision,20,needs design decision,1,0,144159,,,,0,39f5200da3115862eba8aef1049509067a0a9485ed461691e967123a8b040f7a
thread:issue:144228,issue #144228: grad_fn function disobeys broadcast rules,needs design decision,20,needs design decision,1,0,144228,,,,0,90828b22999568f778e7be18633e2ef39136d15f0ff3460bff69fee9f31ab4c7
thread:issue:144536,issue #144536: [bug report template format] Simplify version information with HTML tags,needs design decision,20,needs design decision,1,0,144536,,,,0,65b8b52c1008299ab7ad9ef5415b29896c658c48ac453e498afde7ec1eb573fb
thread:issue:145739,issue #145739: No broadcasting by default.,needs design decision,20,needs design decision,1,0,145739,,,,0,245a7dda75381be972aac3bec8dca25183a13c113c15182cbc082660517aa292
thread:issue:147437,"issue #147437: To avoid default eps ineffective in some precision(float16), dynamically set the default value of eps in optimizers such as Adam based on precision.",needs design decision,20,needs design decision,1,0,147437,,,,0,86e29492595d7f6b953ba37d90cac819e13530513353b95d435c7400ddc5ad61
thread:issue:148345,issue #148345: Installation of `pytorch==2.6.0+cu124` doesn't install `triton` and `nvidia` libraries,needs design decision,20,needs design decision,1,0,148345,,,,0,ef0961b3a226cf8603aa167c27b9032e4ecedbe80ff41208020ce24be64300d1
thread:issue:14844,issue #14844: Negative indexing for nn.Embedding inputs,needs design decision,20,needs design decision,1,0,14844,,,,0,75c504e383c516abc2010f890a40419e27c0edce0adbada47e739410a315aba3
thread:issue:148843,issue #148843: make `to_empty` a no-op if parameter/buffer already on `device`,needs design decision,20,needs design decision,1,0,148843,,,,0,661ac108f4710b36fbc57a991692b8fb6b0bab57310ec6ae22129a4c42bed3bc
thread:issue:149118,issue #149118: Add dim parameter to torch.bucketize,needs design decision,20,needs design decision,1,0,149118,,,,0,879bba3e45be2cf91ebfc40af650387395b1b0c622b211158a2fbf0c1c4293ab
thread:issue:149133,issue #149133: Implement einsum backprop rather than decomposing,needs design decision,20,needs design decision,1,0,149133,,,,0,7be8589bcfeebc4530c9542ef279db6c047a67a5b40a3d92634ad13e84232908
thread:issue:150432,issue #150432: pytorch pip install instructions: always include the cuda index,needs design decision,20,needs design decision,1,0,150432,,,,0,cf0bd6e163db479ab0d9ee160ac41cc15234aba873aac7fc1bd35066e135b873
thread:issue:150698,issue #150698: [Feature Request] Memory optimization for backward propagation in GPU,needs design decision,20,needs design decision,1,0,150698,,,,0,f1a37d7f49cabaccb27758fb77f3c1f58f357792c8a40bac4b6b3d4e224903ee
thread:issue:150742,issue #150742: Install pytorch from pypi using local CUDA build,needs design decision,20,needs design decision,1,0,150742,,,,0,bf10a2fad40128f9023a7f4b169c96e223e24ed98a21e50e1ab64bbfd9cfdea2
thread:issue:151722,issue #151722: Add maxcount Parameter to torch.unique and torch.unique_consecutive,needs design decision,20,needs design decision,1,0,151722,,,,0,c3ec159d19f52e07c555fa221390eeda4acd14c36ea1f1dd3e1d5a2f81ade15f
thread:issue:152875,issue #152875: Incorporate CUDA Memory Trimming Into DeviceCachingAllocator,needs design decision,20,needs design decision,1,0,152875,,,,0,c37cf662333b4054c3f48aeeaf1ee36587d19f42b4efab0ba99ff0eda07bd887
thread:issue:153035,issue #153035: Add Split Softmax,needs design decision,20,needs design decision,1,0,153035,,,,0,839b38df869309f1c4544e25057ad5683af03ca6712e09eda7e83e92c4bcb41e
thread:issue:153566,issue #153566: Improve PyTorch Logging to Distinguish ROCm from CUDA,needs design decision,20,needs design decision,1,0,153566,,,,0,f6808650c4fbec172c232b89460b4624df87220a88908dd6cc5ebef8d53b35c1
thread:issue:153669,issue #153669: Make tlparse able to show a summary of distinct graph breaks,good volunteer slice,20,good volunteer slice,1,0,153669,,,,0,08c854728d818d04e28a94eb6a72557f410677c379d3d3b5c96763ec7ed719cb
thread:issue:154260,issue #154260: Pypi Support for Windows arm64,needs design decision,20,needs design decision,1,0,154260,,,,0,db93c7a012db873ebe8f25775b6f426038a28dcf5b615a1d08111262ec0b6ffd
thread:issue:154767,issue #154767: Turn gradient_accumulate into a separate aten op,needs design decision,20,needs design decision,1,0,154767,,,,0,5a42930d3faad59fac89a1407fb455c0ddf0a5f250d788d3bd9d84ed9eb54728
thread:issue:154906,issue #154906: load_state_dict with strict should be able to remove_duplicate?,needs design decision,20,needs design decision,1,0,154906,,,,0,5a179512c6b7a5655ac72f35597cd424bf3de94fcd9c0486a9008a7f16b8a011
thread:issue:155322,issue #155322: Segmentation fault after manual tensor assignment with autograd enabled on CUDA,needs design decision,20,needs design decision,1,0,155322,,,,0,40cd85c2c0f1d6eb5920495e4a809a63971e8f54ce3fe4ab02ad1f0293bb38f4
thread:issue:155368,issue #155368: Change error to warnning when tensor size mismatch while loading params,needs design decision,20,needs design decision,1,0,155368,,,,0,f388145df66e0252658b319bfeb07d77d4770b942fe0cf232f8a78b47ea2dc80
thread:issue:155878,issue #155878: RFC: Add Adaptive Entropy-Gated Contrastive Fusion (AECF) for Robust Multimodal Attention Pooling,needs design decision,20,needs design decision,1,0,155878,,,,0,dc5ce5f41c60ec5596f56fa9500b548240ca8b5ad25b434b242ecbea6592c72c
thread:issue:155983,issue #155983: as_tensor of list of tensors should keep grad history,needs design decision,20,needs design decision,1,0,155983,,,,0,ea35d7a74f69c4004d17685862aa9229684f69259523e026e21abee0c382187c
thread:issue:155986,issue #155986: Vulkan interoperability,needs design decision,20,needs design decision,1,0,155986,,,,0,c4f7617fefe965eea840a82ac4586ea32652818fb98c96420a508cf6b81b9f5b
thread:issue:156424,issue #156424: research adding cuda-bindings to core,needs design decision,20,needs design decision,1,0,156424,,,,0,1fab4ebe6f69141f002cf3d4f8cc13b4bd8ee8f9fec76b78c23970635284d26f
thread:issue:158043,issue #158043: Add GoLU Activation Function 🚀,needs design decision,20,needs design decision,1,0,158043,,,,0,cb41c2e5d8e7fd362d1adab7b2281624ff2deac9d525ee2542f1bfe160972fc7
thread:issue:158704,issue #158704: Enable DataLoaders to prefetch batches across epochs,needs design decision,20,needs design decision,1,0,158704,,,,0,391c5c0fc52fa13a6c4101199cf252eef40af0e5556f54d8f0eedfb8f6fc1713
thread:issue:158709,issue #158709: autograd precision: float32 gradients of bfloat16 weights,needs design decision,20,needs design decision,1,0,158709,,,,0,43ede6fdd2bff379c0e3dc466d00c338e326780e77893e4f9aae301f16c434fa
thread:issue:160489,issue #160489: [CI] Mergebot tests are no longer updatable,needs design decision,20,needs design decision,1,0,160489,,,,0,71f3cbca0fc9843cd8372b4ee19e47079ab5d0b82feb6cc8df4d05f68b50f082
thread:issue:160637,issue #160637: Support optimisation of sparse parameters via LBFGS,needs design decision,20,needs design decision,1,0,160637,,,,0,db9f77ff8561a8506c54a644d3fc4e0e3550e80db54919a7d1e4f9f0bb2973a1
thread:issue:161129,issue #161129: autograd assertion error in cuda backend when privateuse1 is renamed,needs design decision,20,needs design decision,1,0,161129,,,,0,3dd7e6fe9cbdc09a7531c8b11d15833cb2c3938c4eac8a60c5add3a5da3670f9
thread:issue:165577,issue #165577: CI: What is the purpose of `slow.yml`,needs design decision,20,needs design decision,1,0,165577,,,,0,035b50e723513a573a818e88df47f92b71e1e4f70860d1680ce5744cfcc9eed6
thread:issue:16663,issue #16663: new_zeros is not traced correctly,good volunteer slice,20,good volunteer slice,1,0,16663,,,,0,ec7a2197c2b17fab17472c38e7faec40c089b2c229ca28d3e195435f5c2cdfce
thread:issue:166826,issue #166826: PyTorch Layer for Pre-Summation Nonlinear Feature Transformation,needs design decision,20,needs design decision,1,0,166826,,,,0,ab143e4482a41fa3f86c71ba49d62517ac95bbdb68d18a29e7f44f27039dfa8d
thread:issue:170944,issue #170944: Indexing a tensor with numpy boolean array is very slow,needs design decision,20,needs design decision,1,0,170944,,,,0,06c16d14101ed7c3cdfdebc3c96a8b46dc476a67658b2af29727ce4895815dee
thread:issue:20704,issue #20704: Remove unpack() in torch/csrc/nn/type_checks.h and its caller functions in the codebase,good volunteer slice,20,good volunteer slice,1,0,20704,,,,0,f93eb5def3315cd5b746c659d82bf284d37c35daf8cfa43da8a9b331291cebf2
thread:issue:20822,"issue #20822: [Proposal] Data reading framework for PyTorch (Hive, MySQL, S3 etc.)",needs design decision,20,needs design decision,1,0,20822,,,,0,24e21e6c2972d03507507fe453f45113a563a0da080116f0219b08081f264617
thread:issue:35458,issue #35458: Shared-QK transformer for the transformer (nn.activation.MultiheadAttention) module in PyTorch?,needs design decision,20,needs design decision,1,0,35458,,,,0,50734363462a5899da7060cf51df3310fa8d7368e8370eb10d03d8ac24be4bcf
thread:issue:35801,issue #35801: Add KFAC optimizer,needs design decision,20,needs design decision,1,0,35801,,,,0,9bc0a0ee4a45fa5be5ccf595e29e556f0a80325a2f13768f949a53f5993ee357
thread:issue:37155,issue #37155: Per-cluster biases in AdaptiveLogSoftmaxWithLoss,needs design decision,20,needs design decision,1,0,37155,,,,0,da1793b0f0b331e3999911d2790ed6fc8c46935468f027df83b9f2c8ba33b94d
thread:issue:3904,issue #3904: Implement DE in pytorch.optim,needs design decision,20,needs design decision,1,0,3904,,,,0,889d7cad299d2e03a4ee990d8fb02320761480ccf6c3efe9102a5da1dbd4035d
thread:issue:39765,issue #39765: Support mainstream pruning techniques,needs design decision,20,needs design decision,1,0,39765,,,,0,48932e2545d7bbea1b251dca8d4d0cbd8c955a80e7b8e4a270834fcebb80151f
thread:issue:39780,issue #39780: LRP-based explainability,needs design decision,20,needs design decision,1,0,39780,,,,0,6ba2d3490d31efa777a6cd45dda9cb8bc41bab2da886cad7e228a2261b6dd357
thread:issue:4132,issue #4132: x.grad should be 0 but get NaN after x/0,needs design decision,20,needs design decision,1,0,4132,,,,0,981743cb742b85edc354901830c932684e5ca649a598cba0df373b06a9d8da22
thread:issue:41556,issue #41556: Implement LSH Optimizations for Enhanced CPU-Only Performance,needs design decision,20,needs design decision,1,0,41556,,,,0,403a944a1e5b6b15b3d3570e87d94cbf6f7c295c247c9909baac525227864a1f
thread:issue:41559,issue #41559: momentum in BatchNorm,needs design decision,20,needs design decision,1,0,41559,,,,0,1f4720388d62662c8c8f6a4cb78a49953d216da26e6a88b056dfb9753aa133ef
thread:issue:41803,issue #41803: out-variant for tensor.bitwise_and (exists for torch.bitwise_and) + bitwise_friends,needs design decision,20,needs design decision,1,0,41803,,,,0,cb540774cd0453f6d564a85af60567f69a2fb39220fcfc0c2954e933920b8457
thread:issue:42652,issue #42652: Add non_blocking=True copy of torch.Storage,needs design decision,20,needs design decision,1,0,42652,,,,0,455f682b14cdb4ed562821b0d86b0b0ee097d05bd82303be3388121ed63815ff
thread:issue:43101,issue #43101: New function for CTC decoding,needs design decision,20,needs design decision,1,0,43101,,,,0,fe072f3ac0aef9ab1d07edb62242652d003f0db9e82485049d5458610c05a873
thread:issue:43450,issue #43450: Training Large Neural Networks with Constant Memory using a New Execution Algorithm,needs design decision,20,needs design decision,1,0,43450,,,,0,975ae16220c41a18d1347f5e073c0093c5d8443a411c6c2dae26e7323ba9be32
thread:issue:44591,issue #44591: [Feature] Fused Matmul & Min/Max/Sum/Prod,needs design decision,20,needs design decision,1,0,44591,,,,0,f5a755476f5073b8b215abd10a319c14cb5cabfcfa33ff0a644b90ac8d481195
thread:issue:44777,issue #44777: Hybrid Memory,needs design decision,20,needs design decision,1,0,44777,,,,0,b5d22efa223d54ff51b114b7704c551cd9dfe1c3d703f440ec8d2532f6fc1e6e
thread:issue:44827,issue #44827: [RFC] Pipeline Parallelism in PyTorch,needs design decision,20,needs design decision,1,0,44827,,,,0,a0060d5a39ea3c601d4fbb79d00ff0733c280e6ce1e24d59dbfa758746641897
thread:issue:44945,issue #44945: Fuse softmax and masking in MultiheadAttention,needs design decision,20,needs design decision,1,0,44945,,,,0,10f7e88892d28e7a843ee41835a13e78ed2da5d85168a9b29f3bf8b7c5d1210e
thread:issue:44954,issue #44954: Optimization with constraints for torch.optim,needs design decision,20,needs design decision,1,0,44954,,,,0,41367f11f7d3f6e827d89433585a263a5e8ad4ec15a9c46904c153896eaf05d5
thread:issue:45051,issue #45051: Custom Datatypes in Tensors,needs design decision,20,needs design decision,1,0,45051,,,,0,a50e4b6ec24dd127efb89f966cae9d1013dc8688a2955463a0182752fd4af123
thread:issue:45133,issue #45133: [Feature Request] Add support for Hidden Markov Models in torch.distributions,needs design decision,20,needs design decision,1,0,45133,,,,0,d6efcf355f7a228de9242230cb7eabb2ba2b62e1b4c8b91387aec922c006698a
thread:issue:45295,issue #45295: Adaptive usage of memory during training inference,needs design decision,20,needs design decision,1,0,45295,,,,0,a4a7caf74c2ea7b085522eee797cf4d664eb87f68e1c42dce2529bb2244f2dbb
thread:issue:45808,issue #45808: Support for Locality Sensitive Hashing Optimizations,needs design decision,20,needs design decision,1,0,45808,,,,0,743b07e3b7b41746325aed190208540433688b5e5bf32a5e61395ad926a5ba2f
thread:issue:46516,issue #46516: Batchnorm support for tracking buffer statistics when using gradient accumulation,needs design decision,20,needs design decision,1,0,46516,,,,0,dc0f20be0e0cf7ba6a01f614ca92f867e691fc2fc07260c96c19a0b7e7bcef6d
thread:issue:4660,issue #4660: [Feature proposal] Add MC-derived optimizers,needs design decision,20,needs design decision,1,0,4660,,,,0,c206a0d845a1286b5e01904f23e27c13d6e200c665893ba5b61f8645b420846f
thread:issue:47037,"issue #47037: Change ""_next_index()"" in DataLoader to a public and stable API",needs design decision,20,needs design decision,1,0,47037,,,,0,dac32c9eedfd10aaff8e1ccd7ee654238b14d1f3f6fac15bb3bc974247c83964
thread:issue:48494,issue #48494: Inconsistent type of property stride of Conv1d and MaxPool1d,needs design decision,20,needs design decision,1,0,48494,,,,0,53af15023d8c851a317f1906fe3578b5511a7019ea90e0f69b5bbe7382b4a13d
thread:issue:48747,issue #48747: [bug][libtorch] torch::load seeks to the start of the input stream before reading,needs design decision,20,needs design decision,1,0,48747,,,,0,9c811effed807e02196e9db8e239c096a672c40ea6416d1ef9cd303fb3653e56
thread:issue:48830,"issue #48830: Please add ""dim"" feature for function ""torch.masked_select""",needs design decision,20,needs design decision,1,0,48830,,,,0,4234d4d6e8c5324433da3fbe75c19b8c1fb54c9dedcce6a86d2f65c71f75d64d
thread:issue:49227,issue #49227: Looking for a more convenient way to getter value of the upper/lower triangular matrix into 1D,needs design decision,20,needs design decision,1,0,49227,,,,0,e2813d7737fcd513e67642af450588a4dd0535fddb209ff1b4b8a688f252c755
thread:issue:49338,issue #49338: a time series library,needs design decision,20,needs design decision,1,0,49338,,,,0,3156c28614958645b866cd8227f1e7dfd1a2a385f12327e59a13a192aed02816
thread:issue:49890,issue #49890: GEMM with int8 datatype throws RuntimeError on GPU,needs design decision,20,needs design decision,1,0,49890,,,,0,13718a792d6d48f3b168314b4a97bea40deb4044f412abe8159406e4e8dff6e9
thread:issue:50204,issue #50204: Find the joint eigenvalue of two matrices.,needs design decision,20,needs design decision,1,0,50204,,,,0,fecb51d7470c68152149d24ba711ebf8b29dc89c58e65a2e8c116e6a510eae29
thread:issue:50566,issue #50566: Enhance supported fill value type for constant pad in functional.pad,needs design decision,20,needs design decision,1,0,50566,,,,0,849b2f226bb43df494514633cb9a5fc14d409d98f00a4eb74b6319540b84332a
thread:issue:50712,issue #50712: ignore_index for nn_mse_loss,needs design decision,20,needs design decision,1,0,50712,,,,0,020579f4b4dec5b9810bdc2e147ea568bb04b15a735d580b2d0414f89fc29f3c
thread:issue:52036,issue #52036: Automatic splitting of batch during model feed-forwarding to avoid memory errors caused by the too big batch_size,needs design decision,20,needs design decision,1,0,52036,,,,0,ce686b0b4252718f38214d40909ca4aee0f15cf40b1477f9fa2a115ea22748f5
thread:issue:52207,issue #52207: Wasserstein metric,needs design decision,20,needs design decision,1,0,52207,,,,0,0bdabc6264245aa394d6f4df2bb51e45c607625c50886244f74624ac40a3c4ea
thread:issue:57724,issue #57724: Incorporate Posit Support,needs design decision,20,needs design decision,1,0,57724,,,,0,243322671a71a851aa915aaeccf6f98cb7b3b323fef77f8d215dc3756ce33ae0
thread:issue:58122,issue #58122: Generate special clear error messages for known common misuses in TorchScript,good volunteer slice,20,good volunteer slice,1,0,58122,,,,0,4c50f6ac8e885d2d85a6042af814edae88a0145dcdb4e4c543b7526148f0fc3f
thread:issue:58228,"issue #58228: ""_amp_foreach_non_finite_check_and_unscale_cuda"" not implemented for 'ComplexFloat'",needs design decision,20,needs design decision,1,0,58228,,,,0,fc8e5dce7862735ba284a38dc6ed00988f2e24c06c8df4a1269b0ef2e062895a
thread:issue:60223,issue #60223: [testing] SkipInfo should error if `cls_name` and `test_name` combination is not valid,needs design decision,20,needs design decision,1,0,60223,,,,0,71c57af4a687adaeed1800ee888e1568d34f44cfc0f64cf2ea17c85f10b99b4d
thread:issue:61343,issue #61343: Feature Request: Remove Optimizer Lazy State Initialization,needs design decision,20,needs design decision,1,0,61343,,,,0,1d3290318f0f65f8c636307360bd64e86ffb0906bbe94c7ea30780d1359c1a34
thread:issue:61654,issue #61654: Allow creation of pseudo devices for testing purposes,good volunteer slice,20,good volunteer slice,1,0,61654,,,,0,17bf046140660f679cc48055cd4858bb9b29b64c40079dbd9415101f840272db
thread:issue:61664,issue #61664: Support adding new keyword-only parameters without breaking FC,needs design decision,20,needs design decision,1,0,61664,,,,0,a2145cbbcd5d675d146bbde27d2dafcd958fb7dce7e12b8d623b1df8341c605b
thread:issue:61901,issue #61901: Reducing over empty dimensions for reductions without identity,needs design decision,20,needs design decision,1,0,61901,,,,0,50a2c3d755a33423cf3ab2ec6c56128728f796fa00b2381f121707367c0cfaaa
thread:issue:62155,issue #62155: [docs] Change code snippet formatting to be directly copy-pasteable into REPL or into source files,needs design decision,20,needs design decision,1,0,62155,,,,0,ea2f378c5e2b3cd6eb2e4bb967b01207fee6a83b91e79bb19b9e07bc1c9ff7ac
thread:issue:62902,issue #62902: distributed/elastic/multiprocessing/api_test is flaky on ROCm,needs design decision,20,needs design decision,1,0,62902,,,,0,2f8d33c914e02fa8efe2893d4d584db622af217980f43574d9f9f138c4833193
thread:issue:63038,issue #63038: Use pytorch-probot for PyTorch-specific stuff + MOTD in Dr. CI comment,needs design decision,20,needs design decision,1,0,63038,,,,0,f75b1e0c50bcf321e21ff2b6bc79ae2ea30353f1d6f4c7618ed79160fa9ddfbf
thread:issue:63265,issue #63265: torch.scatter_: support index.size(d) > src.size(d),good volunteer slice,20,good volunteer slice,1,0,63265,,,,0,7753241abfb964bed5501c1d3c0fdf9d98d431e3111b1c8c7cf40a4cbb0fbc57
thread:issue:65392,issue #65392: torch.sparse.sum result has wrong dtype when reducing over all dimensions,needs design decision,20,needs design decision,1,0,65392,,,,0,effb8b4509a975f6bb149aacdb04410a6e0345a7f62c1ac40150268646532df6
thread:issue:6657,issue #6657: [feature request] Stochastic Variance Reduced Gradient (SVRG) optimizer,needs design decision,20,needs design decision,1,0,6657,,,,0,a57ed894af96e3708420f4e8915192c8c2e7c6d5bb36d534d6027210c4424e72
thread:issue:66813,issue #66813: Use `__slots__` for the `nn.Module` class,good volunteer slice,20,good volunteer slice,1,0,66813,,,,0,cc25ce273ff0aa5d1e2b903e14fbf34df74b86c543ac85909f41039cd194da31
thread:issue:67087,issue #67087: Feature Request: 1d grid sample,needs design decision,20,needs design decision,1,0,67087,,,,0,589f7e37d1baaf7d1ea0442c5957cadbbb9c598a40f10db0e46b0e85540f6174
thread:issue:67321,issue #67321: alias to generate tensor with random uniform distribution.,good volunteer slice,20,good volunteer slice,1,0,67321,,,,0,e4e3bb2c4f07625cf0624bdb185572167097ffc3c1cad4ccedfd79b466c109e8
thread:issue:67587,issue #67587: Re-implement `Optimizer.__repr__`,needs design decision,20,needs design decision,1,0,67587,,,,0,64eac3cbcdd46ce29b4e7925b2e452cd4caacf08420d9eb58003bfbf272085ea
thread:issue:67634,issue #67634: Feature request: Grid sample on complex tensors with float grid,needs design decision,20,needs design decision,1,0,67634,,,,0,8e5b726e38598bde7c5658e493d1cb938f43552fa57f82fae2a5921ff2ac48fc
thread:issue:68514,issue #68514: Add common regularizations to PyTorch.,needs design decision,20,needs design decision,1,0,68514,,,,0,ffd5fc4f5a037c8d3cb6e93318a1e5a1f1f00c631ecbfd60553ffd6f2c291bb0
thread:issue:68516,"issue #68516: [ux] [feature request] Arguments to modules(), named_modules(), children(), named_children() to filter modules of specific types",needs design decision,20,needs design decision,1,0,68516,,,,0,ca0f0ec89d38288c3c486c24609f087d52bad168e35ba6a54dd85071948123d0
thread:issue:68699,issue #68699: Transpose of a sparse tensor is not a view operation,needs design decision,20,needs design decision,1,0,68699,,,,0,fa51279841422581d9004a2334256f5e67721d1d9e9d66d1e876487da0ea382f
thread:issue:68798,issue #68798: Clarify variables of BatchNorm*d functions,needs design decision,20,needs design decision,1,0,68798,,,,0,71211513e94f1f99028a73f41c35d2ecfada270ce4fdaf7463a2258308db41f8
thread:issue:68879,issue #68879: Stop gradient option for padding,needs design decision,20,needs design decision,1,0,68879,,,,0,1154b677d0b1a5f56605a5fe560357fa932bb0bcca1927164864c9f80f6dcff7
thread:issue:69757,issue #69757: [Feature request] nn.Model API: Standard model interface,needs design decision,20,needs design decision,1,0,69757,,,,0,acf20d8ce17eefbcf384e5fb1a628f9d0e15dfd7b19b1659112127eba1c440b1
thread:issue:71479,issue #71479: [numpy compat] torch.stack and torch.tensor doesn't support nested list+tensors (NumPy does support) - at least document the difference in the error message,needs design decision,20,needs design decision,1,0,71479,,,,0,748f5e528da0f505e13c5084454901b0bb1ff60536eb4f86353422ef72a8b38f
thread:issue:72712,issue #72712: Better support for pypip packages implementing torch cuda extentions,needs design decision,20,needs design decision,1,0,72712,,,,0,34df9477bfed0cb5d6b01b91ebbbbd1da6c7247210adaebff00c65e4746e33e8
thread:issue:73143,issue #73143: Add nan-safe einsum and bilinear,needs design decision,20,needs design decision,1,0,73143,,,,0,adaa0bf453cae70ba69fa0f00ecbb4e47e0fc9431b7912c52be67e9f89e547e4
thread:issue:73172,issue #73172: LazyLinear with equal in_features and out_features,needs design decision,20,needs design decision,1,0,73172,,,,0,9744f2f4542a21b9d7790f2b41eb51cc847c435c605bd5c8f955b8630880e8f6
thread:issue:74363,"issue #74363: Efficient very large (e.g., 31x31) depth-wise convolution",needs design decision,20,needs design decision,1,0,74363,,,,0,89a3e1b247782152fb845b46c7014821aea511371536414de19707314feea4d9
thread:issue:74904,issue #74904: Adding novel 'AdaFamily' optimizer,needs design decision,20,needs design decision,1,0,74904,,,,0,53417261c5e55a12e92c0ec88d145d741bc40103062153d95dfcc5bc439f3256
thread:issue:76532,issue #76532: Add ability to add custom suffixes to tensor repr,needs design decision,20,needs design decision,1,0,76532,,,,0,712946ba54cbc7727d8043939ee43b7fe751c391d565e4c8ed737f7d0d5a9ba1
thread:issue:79205,issue #79205: out-of-place functional optimizers: functional optimizers may not be composite compliant,needs design decision,20,needs design decision,1,0,79205,,,,0,0d3f8d6f430ab43b6dedc2bfaaf29da02f3c0be26effd31262de7d8530fe85b0
thread:issue:79452,issue #79452: Use c10d broadcast_object in Zero,good volunteer slice,20,good volunteer slice,1,0,79452,,,,0,6a73952a651aa1f107620f048395c0b292d327bb7140d57083974d36bbd45486
thread:issue:79842,issue #79842: Automatically use CUDA,needs design decision,20,needs design decision,1,0,79842,,,,0,77c496fc82cc25e2723f4a4a01b8380ef3a28282eb263014f61e6219d35bd627
thread:issue:79877,issue #79877: test_meta_vstack_cuda_int16 (__main__.TestMetaCUDA) Fails with DEBUG=1,needs design decision,20,needs design decision,1,0,79877,,,,0,fa524173d2d53bdb444aee8d09ffbdfd80b1ca753fa17d0995c008731fa5d8a2
thread:issue:80168,issue #80168: GEGLU activation,needs design decision,20,needs design decision,1,0,80168,,,,0,94d0e4dc61dc502706ebdbcf1816d9102a90edc322837527631539c443650d79
thread:issue:80308,issue #80308: CosineAnnealingWarmRestarts with initial warm up and weight decay applied on consecutive cycles without warm up,needs design decision,20,needs design decision,1,0,80308,,,,0,2b29105b18d9a7fcc942f4a60931ea438da823af7a1a25077a1196038b40de6d
thread:issue:80380,issue #80380: Support for learnable p Values in LPPOOL like Pool,needs design decision,20,needs design decision,1,0,80380,,,,0,20729cfc6bf9ff7b3ee5a006d1e51ca5a25b8c305558a35d6296fc72eda7b228
thread:issue:81333,issue #81333: Reductions on tensors larger than GPU memory,needs design decision,20,needs design decision,1,0,81333,,,,0,291b65a445b74fb17c6ebfba2113e88e34dff783492c627ebd1bf92506bbc651
thread:issue:83492,issue #83492: `Frozen` module for transfer learning.,needs design decision,20,needs design decision,1,0,83492,,,,0,a864c10a8883ef12115f88c3965299cccd537219f0bc5a380cd537c4b7025fed
thread:issue:83529,issue #83529: Adding Levenberg-marquardt optimizer in PyTorch,needs design decision,20,needs design decision,1,0,83529,,,,0,329a2827cd7bf5d8e07c13ee07fd9791577d5a1214f1c81a432c9f50e8e89158
thread:issue:83800,issue #83800: torch.var_mean is slower than layer norm,needs design decision,20,needs design decision,1,0,83800,,,,0,3a6393e0dd79de1f7be2d0f5582b73467fa61d4128fa6b45a608a24d24679cf7
thread:issue:83854,issue #83854: Pytorch/Nova CI should monitor service outages for major dependencies,needs design decision,20,needs design decision,1,0,83854,,,,0,14e2f070507c000c9a3fdb16cc9aa56db9314c83deb5797b095278742fe3e4ae
thread:issue:84347,issue #84347: Lack of newly raised optimizers,needs design decision,20,needs design decision,1,0,84347,,,,0,9941512f2938e0c62ec7ad3a7cf5e97ba0d972b507e62210bef541457426bb58
thread:issue:84593,issue #84593: Uneven and/or Dynamically sized collectives,good volunteer slice,20,good volunteer slice,1,0,84593,,,,0,35eb8b72d6ec42d8e2c534e87255ea28725558442d87fcdd0d42410fd110b4ce
thread:issue:84751,issue #84751: Allow passing dict (as opposed to OrderedDict) to nn.Sequential,needs design decision,20,needs design decision,1,0,84751,,,,0,4b94d4573386e7c13d84e08a07a0102f79a11b64bbfdad50a5321508a4134e96
thread:issue:84782,issue #84782: macOS Pyinstaller: libc++abi: terminating with uncaught exception of type c10::Error: Type c10::intrusive_ptr<ConvPackedParamsBase<2>> could not be converted to any of the known types,needs design decision,20,needs design decision,1,0,84782,,,,0,d65ac95ef4f18d92b643dac14565d1c6965856fcf2aa2e421111bf9cca4f4391
thread:issue:86641,issue #86641: RFC(from users): nn.Module behavior with in-place changes,needs design decision,20,needs design decision,1,0,86641,,,,0,c435ed74f7528f4f5c1f1bcd015375cc974c759ea384ad5a02c9f1658c75cc42
thread:issue:87734,issue #87734: Checkpointing Support for Modularized Optimizers,needs design decision,20,needs design decision,1,0,87734,,,,0,c93edca615b6b0165269a427f1d72ff6a0e02bf31af0e9f9f32a6939f984e155
thread:issue:88325,issue #88325: Group losses in a common namespace,needs design decision,20,needs design decision,1,0,88325,,,,0,177bdaf9a11af2adba01fdc04bbbcb1af80eec9f43f3c5da6093d31398699e9c
thread:issue:8837,issue #8837: Inconsistency in implementation of _LRScheduler,needs design decision,20,needs design decision,1,0,8837,,,,0,51fc67884a811c2bf413ecfaf776c879e9581ea3282806d620a229d697a106ea
thread:issue:89724,issue #89724: Sample Weighted BatchNorm1d,needs design decision,20,needs design decision,1,0,89724,,,,0,c69490468369061575a0c99f38a6076c9d7ca4aef0ce7ef50df5959cd8432596
thread:issue:91073,issue #91073: Implement L1 and L2 gradient as hooks with the option of changing the weight decay value.,needs design decision,20,needs design decision,1,0,91073,,,,0,df9930895215530014ac5fe8692bd8530a02d88328175be0e4ae34f444e029bd
thread:issue:92330,issue #92330: Simplify module backward hooks to use multi-grad hooks instead,needs design decision,20,needs design decision,1,0,92330,,,,0,67ccd210deb4ad31470d065ef6fb44eee798bc79bbee357b6f0f0ad2d82656c7
thread:issue:93154,issue #93154: Support for VeLO optimizer.,needs design decision,20,needs design decision,1,0,93154,,,,0,44bc11986574aaae90f95eb2c581780a11780d3c3072e80ddaa8ec4c51df51e4
thread:issue:93419,"issue #93419: torchdynamo is not properly setting up input tracking (e.g., for symbolic shape guards) for view bases",needs design decision,20,needs design decision,1,0,93419,,,,0,8337884c8cbf734d5b1679da36128a2dc0bc6ae89249e7a828335b13f738092e
thread:issue:93684,issue #93684: Can we rewrite numpy operators to pytorch operators?,good volunteer slice,20,good volunteer slice,1,0,93684,,,,0,2e65a65f1b5f7bd1efaa1b6ff059190eebeb76cbc3a29ce2b88a14ac81b8d37a
thread:issue:94904,"issue #94904: Optimizer ""Lion"" in Symbolic Discovery of Optimization Algorithms",needs design decision,20,needs design decision,1,0,94904,,,,0,607d11bd95cc7c03dae7119af2cf1ff828c6bb6be233916651b4b25e748d346e
thread:issue:95337,issue #95337: Changing behavior of module.to() to better support mixed real- and complex-valued parameters,needs design decision,20,needs design decision,1,0,95337,,,,0,9a1689057fd3f8444f0373fec33eae7cd727878d8c907bcbfc9435ed542b5bd9
thread:issue:95374,issue #95374: Add oscillating activation functions to PyTorch.,needs design decision,20,needs design decision,1,0,95374,,,,0,0604af697fd783818adbb479628f883831bbba7c29362c8a75a9c187bf8070fd
thread:issue:95916,issue #95916: autograd.functional.jacobian : tensor instead of function as input for reverse mode?,needs design decision,20,needs design decision,1,0,95916,,,,0,760d72ce6d4c8deff1f85a00a29d8d62db8fc7f749926e1632041d703d88816f
thread:issue:96555,issue #96555: early stopping,needs design decision,20,needs design decision,1,0,96555,,,,0,f4a8172aab7775c4c57f24d9b0087f5a8a0bef43904b22ff7a497105b4db35a9
thread:issue:96579,issue #96579: Wrong return type from operation on custom tensor inside registered hook,needs design decision,20,needs design decision,1,0,96579,,,,0,c371e5b172df2f64ce368b62602ef312600dd4207c1f975840f110e9036b9071
thread:issue:97248,"issue #97248: A more systematic API for resolving the ""vmap-incompatible in-place operation"" error",needs design decision,20,needs design decision,1,0,97248,,,,0,0e0a096b06ce34e6b8fd4fef2bbd4e1e16a63fd23011adaeb8897be01b2b881b
thread:issue:97469,issue #97469: Improve collectives fingerprinting,good volunteer slice,20,good volunteer slice,1,0,97469,,,,0,f8c6524365d29d9523be876f7788300f5adeaf98d7c705c43c1e8e9cf688d12d
thread:issue:97788,issue #97788: Add SSIM as Loss Function,needs design decision,20,needs design decision,1,0,97788,,,,0,72046fffe4f5b47c4164160623ae3f8f89c202f17d385c5ead2413b23f5294f6
thread:issue:98089,issue #98089: GroupNorm cpu/gpu parity tests fail with pretty large differences,needs design decision,20,needs design decision,1,0,98089,,,,0,c9165b4550c166efc45004360ddfee9d0a66d564f8e315c5f2e11ba01f6d81c8
thread:issue:99278,issue #99278: Build error on libstc++ header stl_alogbase.h on riscv,good volunteer slice,20,good volunteer slice,1,0,99278,,,,0,f94face5808568cac6ed7156e666b1a305051323a21542a87b796d06ee6bfa05
thread:issue:99873,issue #99873: Dynamo config patching in our code is brittle,needs design decision,20,needs design decision,1,0,99873,,,,0,cf9cdf0039369ed7381e061c266370538d9e8f1a9f61a547dcdc97bdce917512
thread:pr:102613,pr #102613: Support building pytorch using MKL ILP64 model.,stale/low urgency,17,stale/low urgency,1,0,,102613,,,0,51b7ba7560d44bb0872676c1253cef2f78a258a313949fe1a8adfac641c655a9
thread:pr:108303,pr #108303: [optim] Make casting to match params a hook (2nd try),stale/low urgency,17,stale/low urgency,1,0,,108303,,,0,35275324c9a9ab4b2ea263501de27000c4b2f7889dd78c4a46b56cc6eaa598f6
thread:pr:110155,pr #110155: [TESTING] Capture scalar/dynamic by default,stale/low urgency,17,stale/low urgency,1,0,,110155,,,0,50b9b061e7f247419d0c07245f2875055684b28ccb6eeefdb69c63eb856b8a18
thread:pr:117905,pr #117905: Factory function and basic .sizes() support for C++ NestedTensor,stale/low urgency,17,stale/low urgency,1,0,,117905,,,0,6837d7d732f2e64c40782785d1b41217dd6768aeaa30f1210390dc40ae8386e4
thread:pr:119496,pr #119496: Switch batch norm stack to consolidated ops,stale/low urgency,17,stale/low urgency,1,0,,119496,,,0,5bc0eef063d747be6af61432388db047d02601142cc8c42c3ce3527050343462
thread:pr:130462,pr #130462: [testing] Allow creating offsets within AC region for NJT without SAC,stale/low urgency,17,stale/low urgency,1,0,,130462,,,0,3c6ec004b94cad2d8d94d5d40e859a41a22fbde726fc093a4a0cbba5cb1ef108
thread:pr:130752,pr #130752: support zb1p and zb2p algorithms,stale/low urgency,17,stale/low urgency,1,0,,130752,,,0,64b2362e3d4529ab6aeac1285bc85ff0f266ddddb88e141a4b4a5c7cd3194066
thread:pr:132414,pr #132414: Ensure FakeTensor post-creation has associated nested int cached,stale/low urgency,17,stale/low urgency,1,0,,132414,,,0,6af4115ddb78d9b98a05bd879f182a557d7c23c696d0ada21d3580794ad372cd
thread:pr:136835,pr #136835: Add back DistributedDataParallel types that were lost when pyi was removed,stale/low urgency,17,stale/low urgency,1,0,,136835,,,0,cbd9cad8b1dd17ffd190527282f6a0533826869c2af0c66176b25d3071c4da81
thread:pr:138519,pr #138519: [Inductor] introduce comm buffer planning,stale/low urgency,17,stale/low urgency,1,0,,138519,,,0,b18a687a4a3f659c93361fba5d661cade7bfbcfa30974c4a82e7907cdb836441
thread:pr:140321,pr #140321: [scan] Refactored testcases,stale/low urgency,17,stale/low urgency,1,0,,140321,,,0,7caa0cbf9ca56a3df47edb00346c9b5408da6499881dbd13ed0bfdba78acd6e3
thread:pr:141842,pr #141842: Refactor NJT to hold metadata on nested int,stale/low urgency,17,stale/low urgency,1,0,,141842,,,0,db689112e3051e2caf9c56c2a27d3da9464d4eecc131c850ffe673b8eec1bd67
thread:pr:142114,pr #142114: [DO NOT MERGE][WIP] CI: Dispatch PR events to the out-of-tree test infra,stale/low urgency,17,stale/low urgency,1,0,,142114,,,0,f9e9fdf094a8a067ff1a5b978c6945b4e28e9a1ff059750cd6f2fd4c0dbd69d9
thread:pr:145922,pr #145922: Update NestedInt equality to take into account all metadata,stale/low urgency,17,stale/low urgency,1,0,,145922,,,0,dfee5a811da7ad3f72e2831338e6a260a113367e7ed2e7f6ebe27472cbd7930b
thread:pr:146101,pr #146101: (WIP) Update NJT ops to check data for raggedness check,stale/low urgency,17,stale/low urgency,1,0,,146101,,,0,d58a51e8f444d52f512ceba91764709f8b913a31cc7f46eb3b8c510bc941ead1
thread:pr:146318,pr #146318: Hack AC to not clear recomputed activations,stale/low urgency,17,stale/low urgency,1,0,,146318,,,0,b495d68315d0cd9b30a74feb9dbea51b15e3e1f03ce779e46d353f63c201ef9b
thread:pr:147927,pr #147927: Custom ops support arbitrary input types by migrating to python dispatcher,stale/low urgency,17,stale/low urgency,1,0,,147927,,,0,3e57c153872941c0f89f3525fc5248ad5a754cfb6c5c09a645f1555c59f4c601
thread:pr:152158,pr #152158: Add AC_TRACER Infra TorchDispatchMode key,stale/low urgency,17,stale/low urgency,1,0,,152158,,,0,c20aa777f398a90019c3df7cdec74a2274c1ca10dc53b7f5e9347139eafb4969
thread:pr:162079,pr #162079: Tune linalg_eigh_cusolver: better heuristic for syevj_batched selection on cuda,stale/low urgency,17,stale/low urgency,1,0,,162079,,,0,d0d88fdd09e4747aca57247f51ef9922498da80f2d3e7b7f4babe76050cea20e
thread:pr:165986,pr #165986: Skip tests requiring NCCL if no GPU is found,stale/low urgency,17,stale/low urgency,1,0,,165986,,,0,a5fc4c685146decde6cddd75a3667c640dab62f930f9ff69bce161bd68d6545b
thread:pr:166625,pr #166625: Fix skip condition in test_overrides,stale/low urgency,17,stale/low urgency,1,0,,166625,,,0,59851036fdff3fe5eb50cdfee2ebd2c6030781793a14c1a358406bed8e6cc522
thread:pr:167541,pr #167541: [MPS] Fix crashes due to a race condition in the MPS backend,stale/low urgency,17,stale/low urgency,1,0,,167541,,,0,f42a99dcb6f2977b7b45bc71571533c0f69914a0a20aa017a1a455ed393d1068
thread:pr:168895,pr #168895: Add `pad_to_multiple_of` to `pad_sequence` (C++ only),stale/low urgency,17,stale/low urgency,1,0,,168895,,,0,85e40859fff9ed003ab707526e39ba8548fb2f9f4dfff1a2f6fda152d3d3ed90
thread:pr:169936,pr #169936: Fix `TestSaveLoad.test_version_error`,stale/low urgency,17,stale/low urgency,1,0,,169936,,,0,5042aeca4f7361ab29154b1429fb981c04ef965e25189c073846b0358a188b22
thread:pr:171686,pr #171686: Add failing test documenting implicit step during `LRScheduler` initialization,stale/low urgency,17,stale/low urgency,1,0,,171686,,,0,e69a29bd9a68047bc2b90bca5d34da63283f3d8570ed2e6ac04e00ade6c684c7
thread:pr:174768,pr #174768: Normalize object names in TreeProfiler test,stale/low urgency,17,stale/low urgency,1,0,,174768,,,0,84919ff950c94c95d6f11ebebec2e61c26a7f7870f87f9b0735dfa9b4df38160
thread:pr:174959,pr #174959: Make server creation in distributed/elastic/test_control_plane more reliable,stale/low urgency,17,stale/low urgency,1,0,,174959,,,0,2f816fce98c2e5b4df434f921f6007a4db8a03552a84271bd3fc16f584dc5433
thread:pr:175320,pr #175320: [MPS] Ensure pool2d inputs are contiguous in target memory format,stale/low urgency,17,stale/low urgency,1,0,,175320,,,0,05df8d62a2c79978ec93273ade51e0299366b112af8746c95165f3c92de639a5
thread:pr:176430,pr #176430: Don't use sharding for tests with a single shard,stale/low urgency,17,stale/low urgency,1,0,,176430,,,0,405d17f242f4f70e695c3851ee371655c4a10940a1a693ef20fd3c532d660375
thread:pr:176709,pr #176709: Remove Eigen d_cnjg and r_cnjg source reference,stale/low urgency,17,stale/low urgency,1,0,,176709,,,0,d29349f357cdd58d24b4cf6b8cc44fb0cd1feb2368f25aaa1d4d9503d3ffd23b
thread:pr:177126,pr #177126: Fix search for OpenMP using e.g. GCC,stale/low urgency,17,stale/low urgency,1,0,,177126,,,0,483fddba2113eabb5c9c24d8dd2cb3b72d98289f0e46bcf0a862f8e0e7502556
thread:pr:177450,pr #177450: DataPipes: prevent failing properties from masking invalid-iterator errors,stale/low urgency,17,stale/low urgency,1,0,,177450,,,0,2b7af1318c4a7f273592770f44e3ece7830d04fb8e571da638eb06a90add2ecd
thread:pr:181262,"pr #181262: Show node.meta[""custom""][""stream""] in Graph.__str__",stale/low urgency,17,stale/low urgency,1,0,,181262,,,0,b986a7601278b2b8f8a11feb9ebab19d08c544518e9f1ed6fc91e496abca7e18
thread:pr:182114,pr #182114: Add torch.utils.nan_guard for forward-pass NaN/Inf debugging,stale/low urgency,17,stale/low urgency,1,0,,182114,,,0,86969195685e176f805805f9f6e33cb0316d1fe4ec5a95b4c280908f443b348f
thread:pr:182269,pr #182269: [MPS] Fix channels_last batch_norm backward,stale/low urgency,17,stale/low urgency,1,0,,182269,,,0,27fefab1efa88d528edc39884220d71100df40d0f52e9ba010f5c65c6a9e26bd
thread:pr:182424,pr #182424: Warning oversized storage copy,stale/low urgency,17,stale/low urgency,1,0,,182424,,,0,9c16ed5e22a066f6793d3826453bbe1093bca8254f21bf92959f82d082c60fc4
thread:pr:182583,pr #182583: Skip RST<->MD source conversions in doc-redirects-check,stale/low urgency,17,stale/low urgency,1,0,,182583,,,0,789d76bbc172c2d11242847b15ee5ae3f30f4223c90bbd3b60b7a32632ca088f
thread:pr:182746,pr #182746: [MPS] Memory-efficient SDPA backward (opt-in),stale/low urgency,17,stale/low urgency,1,0,,182746,,,0,e2f941fe2d62c95d773eb57d79fcd06d8f728cab312926cf558feda4ab1332a7
thread:pr:182788,pr #182788: Fix decorators of `ProcessGroupGlooTest` & `ProcessGroupNCCLTest`,stale/low urgency,17,stale/low urgency,1,0,,182788,,,0,583d7b00d147b69060edfd092dac3d7373ded7fae016c30506d3535bc81da13c
thread:pr:182836,pr #182836: [ROCm][CI] Update mi200 workflows to use arc label,stale/low urgency,17,stale/low urgency,1,0,,182836,,,0,e527bcc24fceeb20d6807be5365234db0b16f8d91124c5fb743d1888775b7964
thread:pr:182857,pr #182857: [inductor][mtia] Skip matvec bmm/addmm decomposition on MTIA; add MTIA matvec bmm lowering (#182857),stale/low urgency,17,stale/low urgency,1,0,,182857,,,0,f353c682c73d89efb80ec16fa566d2fac54d7aac9af04e3e105096d8a7c4f4d7
thread:pr:182873,pr #182873: test: skip some test_non_standard_bool_values test cases,stale/low urgency,17,stale/low urgency,1,0,,182873,,,0,aff3e8884f84683a06364c3f89a45b55e76afbf116bd07ccdee78a4c2916775b
thread:pr:182934,pr #182934: [RFC][distributed] Add parallel_map with collective operation interception…,stale/low urgency,17,stale/low urgency,1,0,,182934,,,0,33de5817c8fd9c2c6076039ecf178000f7af0d323273acc4c104c9e73127c3a8
thread:pr:182975,pr #182975: Document 5 public APIs in docs/source/utils.md,stale/low urgency,17,stale/low urgency,1,0,,182975,,,0,a5b3b168cc170ff80149a307935f1e6714fedb4c7493ae86f587f18bf69c3af6
thread:pr:183037,pr #183037: [Distributed] Add `ranks_only` to `StateDictOptions` for controlling which ranks receive the full state dict,stale/low urgency,17,stale/low urgency,1,0,,183037,,,0,7779221cec4bd530563ebf85ad6b540e21888e9a044867da6395474799c6c945
thread:pr:183203,pr #183203: perf(dataloader): zero-copy contiguous-slice fast path for TensorDataset,stale/low urgency,17,stale/low urgency,1,0,,183203,,,0,8d1360491657e50317c85d3898dfdd1bea87347b30b20005af25cba0a0372e3c
thread:pr:183386,pr #183386: Add tests for OneCycleLR zero divisor initialization behavior,stale/low urgency,17,stale/low urgency,1,0,,183386,,,0,a116106702bd09c72f9354fd5bfed1e5a04a63206daf77887de87c9470e2bfc5
thread:pr:183405,pr #183405: Fix unused groups in test_convs,stale/low urgency,17,stale/low urgency,1,0,,183405,,,0,d43e4f8fa5d18e71a9df42fa6bc138e9ac954a8fe03657840f1c0c593d5e7d66
thread:pr:183502,pr #183502: Remove outdated code for macOS < 12,stale/low urgency,17,stale/low urgency,1,0,,183502,,,0,c320d3c14b12a1624c4b8c01dd74ae8c99e53b36c29a77fa8ce04b7ecdee3c07
thread:pr:183568,pr #183568: Fix timeout in `test_share_memory` on error,stale/low urgency,17,stale/low urgency,1,0,,183568,,,0,70efdcd15adc5e28ef6ce4443dfcb97051788ded28d4515b4dd6853abbe98e61
thread:pr:183726,pr #183726: [poc] yet another nonstrict api,stale/low urgency,17,stale/low urgency,1,0,,183726,,,0,5cb092bdaab30f30b3fb3065089671c66ea31f7bf4cbdaf3d05c78fabee72ee1
thread:pr:183787,pr #183787: [PyTorch] Support selective TorchScript tensor data elision (#183787),stale/low urgency,17,stale/low urgency,1,0,,183787,,,0,059a93085f5535fba93a9982205b401b002be1b4099aa754c360a5798d496d64
thread:pr:183976,pr #183976: [CUDA] Correct inner scope Math.cuh int i -> int j,stale/low urgency,17,stale/low urgency,1,0,,183976,,,0,824d1e3d9ceee8abfd926c6ae2924fc20dafecfa918ffe7827d30996609bb003
thread:pr:184042,pr #184042: Tma scatter add remove software pipelining,stale/low urgency,17,stale/low urgency,1,0,,184042,,,0,7b8ec27e7126a3627d4e0c9ce56d9023f21ad4cfa9c6199406df758e2497ff9e
thread:pr:184067,pr #184067: [submodule] Drop unused third_party/gemmlowp submodule,stale/low urgency,17,stale/low urgency,1,0,,184067,,,0,a59b22ee6719bd1c4a7ed8a5a0cf7edfae5eb2d074b94ee856769ea2cd3c7ee9
thread:pr:184116,pr #184116: [distributed] Add NVFP4 scaled matmul + reduce-scatter fast path #182915,stale/low urgency,17,stale/low urgency,1,0,,184116,,,0,a61d8cc29c8020499c521f99ad0c0c88f6530cce665377dd421ca2bbb087267e
thread:pr:184605,pr #184605: Add CPython Dynamo agentic loop plan,stale/low urgency,17,stale/low urgency,1,0,,184605,,,0,b8fb38a6c4fa588f2a645c1ba9451c8d542baec09547de882b4445b12b260ac0
thread:pr:184753,pr #184753: Fix Dynamo frozenset copy identity semantics,stale/low urgency,17,stale/low urgency,1,0,,184753,,,0,3d58f32d3abaf7464d900a017108667b82d24f8b07dafbe480539019f62d2187
thread:pr:82997,pr #82997: Implement refs.var as a real reference,stale/low urgency,17,stale/low urgency,1,0,,82997,,,0,ff05dc444737558e8d9d866cd057d6d20caa9eb6d7b97d99a89daed59259be9d
thread:issue:126472:pr:124624,pr #124624: [NJT] Allow construction of NJT within graph using offsets from inputs,draft/noise,15,Review active PRs and linked issues as one maintainer work thread.,2,0,126472,124624,,,2,e7cdcd612b61c6e1ae7bf3e0ef6489a6ee30c01e2a99e8cc14c5bf99d8d92290
thread:pr:153893,pr #153893: [PP] Fix double backward error in stage_backward,draft/noise,15,draft/noise,1,0,,153893,,,0,23d35799e8b31e34994cacf6e28136f81524048e35aa29ea4cb26ef466a8ce50
thread:pr:183009,pr #183009: [Inductor] Fall back native_layer_norm; route var_mean through inductor lowering,draft/noise,15,draft/noise,1,0,,183009,,,0,b772dbb675fc9edf9ec7842fec1e72fe43290d9975217ceee25c94c7f3b70389
thread:pr:183507,pr #183507: einsum: fast path for small-K batched-dot to avoid cuBLAS launch overhead,draft/noise,15,draft/noise,1,0,,183507,,,0,6532f5ffb0fc1a2bf1b9d439d118a0374f97b2843525c06d8d639b0e6107eb63
thread:pr:184358,pr #184358: Add XPU to externalId,draft/noise,15,draft/noise,1,0,,184358,,,0,a6309c9e0f4f72fc9592e56ded630ff91f1dc2f2ab85907a37cccbe39e4191c7
thread:issue:55267:issue:60341:issue:61260:issue:68905,issue #55267: Improve CUDA extension building experience,stale/low urgency,14,stale/low urgency,7,0,"55267, 60341, 61260, 68905, 70191, 77184, 81680",,,,6,1579d10d7b9450d6b6fcc73e5b3c0a730335bea70b8dc65732ffcf0a957fa56c
thread:issue:104587:issue:133006:issue:170038,issue #104587: Conv1d step-by-step numerical error,needs reproduction,14,needs reproduction,3,0,"104587, 133006, 170038",,,,2,3b9f6811439fa141ceea9f50da975c18069ba202edc5cc9eba6cf5214d4401de
thread:issue:114455:issue:68114,issue #114455: Torch Cpu Memory Leak with FastApi uvicorn.,needs reproduction,14,needs reproduction,2,0,"68114, 114455",,,,1,139e2a2d1e6d072a47fdca3a39609da9fd215161239bd70c8e18ddf13b7392e9
thread:issue:114465:issue:30896,issue #114465: Installation error: 'CMake Error at third_party/benchmark/CMakeLists.txt:304 (message): Failed to determine the source files for the regular expression backend',needs reproduction,14,needs reproduction,2,0,"30896, 114465",,,,1,5a847a864d79678a25ec370cf71c2be51ccc47fe8831943d1bf6e8304e5242c7
thread:issue:116567:issue:117617,issue #116567: [MPS] `sort` incorrectly handles 'negative' NaNs,stale/low urgency,14,stale/low urgency,2,0,"116567, 117617",,,,1,f6202962f939bff8a75fd12a1d3120694227aacc5f8b2d4613bd02f046a52e07
thread:issue:120237:issue:151740,issue #120237: FFT operations on MPS are less accurate than on CPU,stale/low urgency,14,stale/low urgency,2,0,"120237, 151740",,,,1,c9c64acce30fc1a0ef50b4c7d6f908229a3806c9769f15c19153172dcb18608b
thread:issue:128564:issue:128778,issue #128564: internal asset failure,needs reproduction,14,needs reproduction,2,0,"128564, 128778",,,,1,3f250dc76c0916972d8775f54aa7a186c22a98ac1cd82e4f862a32f8fdc79110
thread:issue:129658:issue:129668,issue #129658: [TensorDict - compile] dynamo generator compatibility,needs reproduction,14,needs reproduction,2,0,"129658, 129668",,,,1,09cb2159775df3dc50b0d15e32ecb0ca42b80495c5e8922446f746c5eca94bdc
thread:issue:152335:issue:159447,issue #152335: pin_memory crashes for big tensors and leaks page locked memory,stale/low urgency,14,stale/low urgency,2,0,"152335, 159447",,,,1,bbfb13e194b9e73922d10f7d832388ad1eca8bf61d4d9b1a07c990417bdfd4c3
thread:issue:152790:issue:155066,issue #152790: [CXX11ABI] torch 2.6.0-cu126 and cu124 have different exported symbols,needs reproduction,14,needs reproduction,2,0,"152790, 155066",,,,1,6d19db35bed56918292a2c01e4ebfc7c3c1d0aceb1e51480556879219da9511c
thread:issue:30900:issue:33248,issue #30900: CUDA error: initialization error (multiprocessing) with Python 3.7,needs reproduction,14,needs reproduction,2,0,"30900, 33248",,,,1,bc5bf7fc835fa65462248fd0cda8169ccfe53105e53be1cef9854958c2fa1f87
thread:issue:101321,issue #101321: Speed when installing from source is very low with CUDA 11,needs reproduction,14,needs reproduction,1,0,101321,,,,0,9a60811cae3e5b8d8e1e442c81faffe782ebf39b40828ef35b2b58e0942b6ec5
thread:issue:101331,issue #101331: Pytorch 2.1.0.dev20230512 cuda not available,needs reproduction,14,needs reproduction,1,0,101331,,,,0,b86f804dbc468c5978af053c7c1992181b02ae31d64ab2220c0368a735ab02b0
thread:issue:101404,issue #101404: problem of compilation for torch2.0,needs reproduction,14,needs reproduction,1,0,101404,,,,0,e66d5c7e8671da88986727af98ad355349171f1212461df98f8eb7894671e675
thread:issue:101880,issue #101880: Crash on Python // PyArrow //,needs reproduction,14,needs reproduction,1,0,101880,,,,0,5eeff47c62b1c4ed9f5d5f86b997674ab5a4b729187bcdba4b1d9c7b20b44108
thread:issue:103423,issue #103423: The document does not emphasize hidden range in nn.MaxPool2d,needs reproduction,14,needs reproduction,1,0,103423,,,,0,ef0df4ca6786dee38412b5eeb17c35bfed9a3af5759b8dfb930879b3246df96f
thread:issue:103424,issue #103424: The document does not emphasize hidden range in nn.Embedding,needs reproduction,14,needs reproduction,1,0,103424,,,,0,4245a3bdc534d7e8d1f01e834e03dbdee29ed34e3ac30ddc729e3b4f58866ea0
thread:issue:103752,issue #103752: Pytorch not calling to C code from a docker container,needs reproduction,14,needs reproduction,1,0,103752,,,,0,14eb7485fb0eaa1df38edd16712b751b04d8766d4c7ad70fb79d8add708895b8
thread:issue:104259,issue #104259: ImportError: libcudnn.so.8: cannot open shared object file: No such file or directory,needs reproduction,14,needs reproduction,1,0,104259,,,,0,4d2d7e33be060f4b413f34d726cf40b491ffc1bac6deeab926b92d7ffabeb062
thread:issue:104284,issue #104284: `F.conv1d` and `F.conv2d` propagate `nan`'s incorrectly when minibatch > 15,needs reproduction,14,needs reproduction,1,0,104284,,,,0,63507a65953571a16c54686da7f41d315fbc5406338808ae350c46c4930fc3a0
thread:issue:104598,issue #104598: ERROR: THESE PACKAGES DO NOT MATCH THE HASHES FROM THE REQUIREMENTS FILE.,needs reproduction,14,needs reproduction,1,0,104598,,,,0,d8281913545ace6a12723f09aa78c67aeada53a4935703241121ce7c5761f014
thread:issue:104857,issue #104857: Torch's `LayerNorm` and Adam optimizer vs those in tensorflow,needs reproduction,14,needs reproduction,1,0,104857,,,,0,80ee4e8486fa42933ea13f2c46ee47535410255fe61d3ff17b3e8209e3af5e41
thread:issue:105700,issue #105700: Don't use weak ref finalization for freeing resources when code objects die,needs reproduction,14,needs reproduction,1,0,105700,,,,0,e6e22d10d6335ec4747894cdb5dea4f2941dbd29665c3c804da707b20eb6dad7
thread:issue:105941,issue #105941: Differences in the results of conv2d calculations in PyTorch 1.8,needs reproduction,14,needs reproduction,1,0,105941,,,,0,fe508e8cac7e265c261e536e81600237aa0ac5b821ca97765aca8deb6cd3c121
thread:issue:106126,issue #106126: Libtorch report C10 error when compiling on my own project,needs reproduction,14,needs reproduction,1,0,106126,,,,0,5a7feb23f1e82046cb6f62359694200d60b4d94ab5949989d5918c82484dd539
thread:issue:107115,issue #107115: H100 works differently than rtx4090 on same model,needs reproduction,14,needs reproduction,1,0,107115,,,,0,e573c13f64c595da7249de0de1cc1d6f970385f12589c9e4d1355fa3dcbe3bc5
thread:issue:107133,issue #107133: torch.inverse throws error when DP but not in DDP or single GPU,needs reproduction,14,needs reproduction,1,0,107133,,,,0,b94777c31599b5111f41928dd14fdebfb048789a0332907d1aff35d76ab6f759
thread:issue:107276,issue #107276: CUBLAS_STATUS_NOT_SUPPORTED,needs reproduction,14,needs reproduction,1,0,107276,,,,0,af0c20da96ef7523701bb9e4076068a80f944cbf0307bddb0b98adc89faa83fa
thread:issue:108013,issue #108013: Installation with rocm5.6 results in error: assert len(weights) == expected_node_count AssertionError,needs reproduction,14,needs reproduction,1,0,108013,,,,0,10375d0a5a681c42f5cba5f2119f4a79c19890202e2028a37b7c5fb22d1cf65a
thread:issue:108493,issue #108493: RWKV + Adam exp_avg_sq will change from positive to negative after loss.backward(),needs reproduction,14,needs reproduction,1,0,108493,,,,0,89882710dc679ff1f3895fbff6f08f0ae7168a069d2f9ed9a1d6e0d6e332e7cd
thread:issue:108676,issue #108676: RuntimeError when calling conv_transpose2d with groups,needs reproduction,14,needs reproduction,1,0,108676,,,,0,3a6efb280e49f9390626dbb0b3ace60e415729c414afc90aacffa1e6911b0fa7
thread:issue:109193,"issue #109193: F.conv2d(input, weight, bias, self.stride, RuntimeError: cuDNN error: CUDNN_STATUS_MAPPING_ERROR",needs reproduction,14,needs reproduction,1,0,109193,,,,0,bb13e9cd4d1dd96ae866cc3dcdf16ba7ba1ca85bea9a8781b4ce19cbcc73e09e
thread:issue:109453,issue #109453: Inconsistent Behavior of `ConvTranspose2d` on CPU and CUDA,needs reproduction,14,needs reproduction,1,0,109453,,,,0,07c7b8273f83ac3027c00a6c7f3e29681c2d34fa16d9b1481129a2ca14e08459
thread:issue:110588,issue #110588: libtorch.so: error adding symbols: file in wrong format,needs reproduction,14,needs reproduction,1,0,110588,,,,0,fd68aa7805843a219e1d38c58a4b7c56a3fdb37eb5fdcbeea669f2347333f7a3
thread:issue:110594,issue #110594: Multiprocessing takes forever after on .get() with mp.Queue() (Possible Deadlock),needs reproduction,14,needs reproduction,1,0,110594,,,,0,3bd18798cbdf45db13647f50048d74150400ef69dab567dac172af01a5b3b836
thread:issue:110701,issue #110701: Error with monai SwinUNETR and checkpointing,needs reproduction,14,needs reproduction,1,0,110701,,,,0,e36c0141eb89edd7f2042ac7da25a8847d84060483edd8295119866989d8982a
thread:issue:110883,issue #110883: MAX_JOBS ignored when compiling pytorch from source,needs reproduction,14,needs reproduction,1,0,110883,,,,0,2b67c9040880d9f36ca0f8c53a6313980148ddb13e78b90aa183651d998b565e
thread:issue:110944,issue #110944: gdb core dump when enable DEBUG mode to compile cpu torch in centos!!!,needs reproduction,14,needs reproduction,1,0,110944,,,,0,a108e8d317ae3d1d92393587daecc971bfb40782721ecdd90d5e5e695c466c68
thread:issue:110946,issue #110946: RuntimeError: out_ptr == out_accessor[thread_count_nonzero[tid + 1]].data() INTERNAL ASSERT FAILED,needs reproduction,14,needs reproduction,1,0,110946,,,,0,371151bd5314caf38166bc31fed4e9033cfc91e2881477ffa98d25326d7238eb
thread:issue:111158,issue #111158: RuntimeError in run_streaming_llama.py When Using Accelerate with Streaming LLMa Model on A4500 GPU,needs reproduction,14,needs reproduction,1,0,111158,,,,0,c9748e75f73c23c836275198cd3594f1948a34d9482d7e53d90287b4ca962d77
thread:issue:111359,issue #111359: I have a trouble with to_symmetric,needs reproduction,14,needs reproduction,1,0,111359,,,,0,0252745352a261e6e436ff967f7dcd5a6334acb53ab82e04332c7e681aa111c5
thread:issue:111419,issue #111419: nonnull error,needs reproduction,14,needs reproduction,1,0,111419,,,,0,cd04c0cf2f4fbf6b14387eee750dcbda296c52458052738dc70c958a6ccc1d9f
thread:issue:111508,issue #111508: LBFGS accuracy difference between CPU and GPU,needs reproduction,14,needs reproduction,1,0,111508,,,,0,376951440434f579fb482ad0a01ce5929606ed21d2f335d5f109e465fae29e5d
thread:issue:111547,issue #111547: Bug with as_strided_tensorimpl for MPS devices,needs reproduction,14,needs reproduction,1,0,111547,,,,0,7c74ad1466dafabfcda0e17542b345b07e2d2dbd8d6b83aed0381e76b9875156
thread:issue:111896,issue #111896: ImportError: cannot import name 'external_utils' from partially initialized module 'torch._dynamo',needs reproduction,14,needs reproduction,1,0,111896,,,,0,cfd7db9bacd16f04e66ccb38d60aa6258b8eed52b2ee620c5117c34afc13437e
thread:issue:112569,"issue #112569: RuntimeError: NVML_SUCCESS == r INTERNAL ASSERT FAILED at ""/opt/conda/conda-bld/pytorch_1695392035891/work/c10/cuda/CUDACachingAllocator.cpp"":1154, please report a bug to PyTorch",needs reproduction,14,needs reproduction,1,0,112569,,,,0,51400e17b1b5e5ef7fe7eaa465168cb9cbc0900aa41657cc0c8ad5c66fc62976
thread:issue:113035,issue #113035: Training a network SUPER slow with Pytorch 2.1,needs reproduction,14,needs reproduction,1,0,113035,,,,0,c378c07e3e2ce8aefe32e305fee5d3ab98f422b8c49e830035823381caa23e53
thread:issue:113150,"issue #113150: Aborted (core dumped) after Run Pytorch2.0.0, which I compiled by myself",needs reproduction,14,needs reproduction,1,0,113150,,,,0,254aaba33f8e303ece297f15bc99396c10d0ec920a6f98bbb78930037ec1754b
thread:issue:113840,issue #113840: torch.einsum is stuck in mp.Process,needs reproduction,14,needs reproduction,1,0,113840,,,,0,d3496717a8aa54b5a2896d0395fb3207d80a93f9eeaada4d20b064029c709914
thread:issue:114534,issue #114534: `PYTORCH_NO_CUDA_MEMORY_CACHING=1` with `torch.multiprocessing` shared tensors seems to perform use-after-free,needs reproduction,14,needs reproduction,1,0,114534,,,,0,7664bc3adcd33a0a67e370b47f95f82aa980bfd7f2068630f9b034bb3671a39d
thread:issue:114572,"issue #114572: In version 2.1, libtorch needs to be woken up every time it is called after the model is initialized, which means that every time the model is called, it is very slow to predict the first picture.",needs reproduction,14,needs reproduction,1,0,114572,,,,0,e87bb46ede2e004c45432e32db4e83811399ba3a396ad7677bd71001617f02db
thread:issue:116589,issue #116589: Cannot pickle torch model in version 2.1 when using `autograd.Function` as a model's attribute,needs reproduction,14,needs reproduction,1,0,116589,,,,0,f61fcee7d99553690ffe5c3d0b36365bb8f121897e3c3cf259ab48bc4e618c29
thread:issue:116766,issue #116766: [Dynamo][DeepSpeed] torch._dynamo.exc.InternalTorchDynamoError: NestedUserFunctionVariable() has no type,needs reproduction,14,needs reproduction,1,0,116766,,,,0,e1b528becbae7152228adbf9e894cee8503954faf5372388265c3c80e8a01b38
thread:issue:116882,issue #116882: RuntimeError: CUDA out of memory.,needs reproduction,14,needs reproduction,1,0,116882,,,,0,9d93b602c70c21bb531f038a4c3b0027e590372c74358e6b3ac7f990eee08e0c
thread:issue:116931,issue #116931: Previous Torch versions,needs reproduction,14,needs reproduction,1,0,116931,,,,0,7b00e1f41108f07ef499a7108d572a4c5e541befd90c23e331530d62549c55c1
thread:issue:117032,issue #117032: losing precision in torch.sum,needs reproduction,14,needs reproduction,1,0,117032,,,,0,fc136a25ca3bd082c29883475fa72f2ed72fb980744a1383357bb04a68da43eb
thread:issue:117210,"issue #117210: When running the code model.to('mps'), the program gets stuck, but it runs fine on the CPU",needs reproduction,14,needs reproduction,1,0,117210,,,,0,793ca1b935a2b344d3cdffa56c2f45cc97edf04e6075d39795900ee27fcf299a
thread:issue:117469,"issue #117469: Cannot import torch on Ubuntu 20.04 from clean pip install (shared library issues, libcufft and libcudart)",needs reproduction,14,needs reproduction,1,0,117469,,,,0,3a0dcb7f9d6fb953c495de4cdfb88638c870105dd01fa105d5b22115faad68b4
thread:issue:117962,issue #117962: runtime error: unknown parameter,needs reproduction,14,needs reproduction,1,0,117962,,,,0,5b9bf788ee64ccfcdb7c9c46df01c2256f98f454c7e542a2e707fbd86f9cafdf
thread:issue:118172,"issue #118172: ""PyTorch installation failed.""",needs reproduction,14,needs reproduction,1,0,118172,,,,0,727833526533ea86b28a4eaa42dc91f0215d0abc312fad569512a90072a65d31
thread:issue:118263,issue #118263: Getting error while loading LLAMA_2_7B,needs reproduction,14,needs reproduction,1,0,118263,,,,0,02901078350c88ddcc41ef7c26459ec574794733f4604574a763687a9e92a7a7
thread:issue:119147,issue #119147: Incorrect Results from cdist_backward Operator with Input Dimensions Exceeding int32_max,needs reproduction,14,needs reproduction,1,0,119147,,,,0,2f04458f93494607258f4fe631ba7659d63cf68f9526b56be074a7c5137ecf5c
thread:issue:119546,issue #119546: Too many spawn process,needs reproduction,14,needs reproduction,1,0,119546,,,,0,97b9e336f8f0e6570236c3ae2f2443571a47003a3d3004f019ea9e46b92873d0
thread:issue:120081,issue #120081: undefined reference to `__atomic_exchange_1',needs reproduction,14,needs reproduction,1,0,120081,,,,0,88cb18c3078d58d1b1c1179de95e488feac172c105d604d1ec34d7a88071960f
thread:issue:120167,"issue #120167: When loading a text file with torch.load, it exhausts the system's physical memory.",needs reproduction,14,needs reproduction,1,0,120167,,,,0,88ee862732da1786c98b57e49aa173817d30b733d63bafac8b5e7d0c9f43634e
thread:issue:120606,issue #120606: “libc10.dylib” can’t be opened because Apple cannot check it for malicious software.,needs reproduction,14,needs reproduction,1,0,120606,,,,0,6c13b0f68602af289402eb66ec5a1b7e39660852e01a9d0f59ba3c07bcc0eecc
thread:issue:120791,issue #120791: about tensor.new(storage) issue,needs reproduction,14,needs reproduction,1,0,120791,,,,0,738084a66e1d42a118bd8342358b6d15a29ce4a267352147b0ec15fdef9a0d58
thread:issue:120905,issue #120905: F.conv2d double precision error,needs reproduction,14,needs reproduction,1,0,120905,,,,0,2a87e9f5ca3cccff11709e92e558dd1ead99631a448faa02b4d37c3c73714878
thread:issue:121067,issue #121067: RuntimeError,needs reproduction,14,needs reproduction,1,0,121067,,,,0,12cf006027b984f46a9cc2b846b1482a4e20005e47f7e78c503024f009fcf63e
thread:issue:121190,issue #121190: The result of BF16 to FP32 is different in class LlamaRMSNorm training twice.,needs reproduction,14,needs reproduction,1,0,121190,,,,0,9cf2f2686186a3ad11962212b07b3588355e0a62a202f204dff7ab6ce411de1c
thread:issue:121222,issue #121222: Segmentation Fault in `torch.nn.Conv1d` starting from torch 2.2.0,needs reproduction,14,needs reproduction,1,0,121222,,,,0,1d8557e8d5053354b5b1d3dafcb85cbe7f68aa74cb9419b4052336ee51c6bfa3
thread:issue:121287,issue #121287: Libtorch - clang: error: unsupported option '-fopenmp',needs reproduction,14,needs reproduction,1,0,121287,,,,0,f588b03f9ac172f8acea97fc5d047792bb026499795fb076b8c2c1b6f3940329
thread:issue:121646,issue #121646: Installation fails on python 3.7-3.12,needs reproduction,14,needs reproduction,1,0,121646,,,,0,f0c02cd9a6f22445a017e86bf1748ffe78ca506580a05cfdb1139db779f23369
thread:issue:122548,issue #122548: Unexpected device casting from CPU-pinned tensor to GPU tensor.,needs reproduction,14,needs reproduction,1,0,122548,,,,0,5c58780fb8fa2e061ad59d5ee2a282d5d1f6d669bf9249d10968ea49e1a7353d
thread:issue:122707,issue #122707: unresovled symbol when compiling a torch c++ app that using libtorch,needs reproduction,14,needs reproduction,1,0,122707,,,,0,6dff4b2c832026c7652660654dc2071ff8392c1a53a532325fe80fefd44815e5
thread:issue:124197,issue #124197: Libtorch crashes docker when included in header file,needs reproduction,14,needs reproduction,1,0,124197,,,,0,74731c97143f20d14d714c977f134f13080db8331c6e73b3bccd71226981f7b8
thread:issue:125754,issue #125754: Excution difference between Mac and Linux,needs reproduction,14,needs reproduction,1,0,125754,,,,0,300b457c78060cb661671e1bf2573184b06323f45147c0802832f3e4bbf43572
thread:issue:125911,"issue #125911: set_device只支持int,但如果碰到k8s分卡vcuda时报错segmentation fault",needs reproduction,14,needs reproduction,1,0,125911,,,,0,44823276557f17a12a389d46ea0e5f5abd971b07f5386541c3bbbe0405b7297b
thread:issue:126141,issue #126141: Inconsistent gradients in reverse mode and forward mode AD when using `MultiheadAttention`,needs reproduction,14,needs reproduction,1,0,126141,,,,0,a1753e1f5ecc0be7f69fa5739a1757775e0dcac1ad1d16805c765858c3f82e76
thread:issue:126388,issue #126388: RAM Not Freed on CPU After Moving Model with Multiple Transformers to CUDA,needs reproduction,14,needs reproduction,1,0,126388,,,,0,67e8cb9934a636ed61927eef19f681368be4da2f3070cc27ba6b3b7027f3b290
thread:issue:12641,issue #12641: Install Jetson TX2 Max Regcount Error,needs reproduction,14,needs reproduction,1,0,12641,,,,0,94555c2df7ff9b5fa11d0db2806ccf3dbe72c6f72efb672dc670502dee64bf21
thread:issue:126760,"issue #126760: ""torch.orgqr"" performs differently on cpu and gpu",needs reproduction,14,needs reproduction,1,0,126760,,,,0,b90f8c3744e7b6cdfb776e8f1890296f155715bfac89af51dd51bd7a5f545315
thread:issue:126775,issue #126775: The significant difference between the outputs of torch.reciprocal running on cpu and gpu,needs reproduction,14,needs reproduction,1,0,126775,,,,0,b32924326ee9026721b32ce65ba137c28efed3d7b904dec735f7bc8346f72cd8
thread:issue:126862,"issue #126862: worse results by using the MPS backend, compared to the CPU",needs reproduction,14,needs reproduction,1,0,126862,,,,0,cd9c5b5fad380b10bda99074691baecaa1d782b7a13f7f7df8ff2b41a5f64812
thread:issue:127560,issue #127560: [libtorch] duplicate symbol of vtable,needs reproduction,14,needs reproduction,1,0,127560,,,,0,02834f7f2c37d78a75093a994d9d218150f8497dc35b00caf2a23c3885a35270
thread:issue:127978,issue #127978: torch.addr outputs NaN when computing on inf vectors,needs reproduction,14,needs reproduction,1,0,127978,,,,0,cf4326b9546b926b42f0377fd5918b92b594c9c0c013ccb4968f5cbcd6d0f13a
thread:issue:128131,issue #128131: Unable to record Memory consumption with `torch.cuda.memory._record_memory_history()`,needs reproduction,14,needs reproduction,1,0,128131,,,,0,63855fdb499f0ad76149c0b9f59519912e79b9f8ecb4191a93abba3212dce8ce
thread:issue:128290,issue #128290: [apple M2]libc++abi: terminating due to uncaught exception of type c10::Error: Could not find schema for aten::randn,needs reproduction,14,needs reproduction,1,0,128290,,,,0,a607a0409bd9ebc38ce392f2de5ee1bd3671a64be1f1d7e24eed5eb3942a677f
thread:issue:128939,issue #128939: Unable to install PyTorch on M1 Macos with Python 3.10.14,needs reproduction,14,needs reproduction,1,0,128939,,,,0,6cd9ab0f8cf35b9924f34e519f0d4ec1fd2bf53e9078581db16af73a4e73f84d
thread:issue:129303,issue #129303: Incorrect index from torch.mode,needs reproduction,14,needs reproduction,1,0,129303,,,,0,d62b129772383c001fd22d424e25410a9ab28c6febb936ef7e1c333c2e0e3715
thread:issue:129643,"issue #129643: There is a bug, only when self",needs reproduction,14,needs reproduction,1,0,129643,,,,0,e2a5ccb2694b2580a4d43423545eec7d7d3e2ced48bad69dc7a71bc8a291507a
thread:issue:129992,issue #129992: Got this error when just running my flask script,needs reproduction,14,needs reproduction,1,0,129992,,,,0,75cad9e4ed9fc971220de6827e81080b97ec58904721b548bfd459066f5db984
thread:issue:130486,"issue #130486: RuntimeError: NVML_SUCCESS == DriverAPI::get()->nvmlInit_v2_() INTERNAL ASSERT FAILED at ""../c10/cuda/CUDACachingAllocator.cpp"":813, please report a bug to PyTorch.",needs reproduction,14,needs reproduction,1,0,130486,,,,0,33df4be094ec1bd383c773de2411f3e7e6959585de7fe8edfb12b0299e0e0766
thread:issue:130530,issue #130530: Fail to offload FSDP model weights and optimizer states without using CPUOffload(offload_params=True),needs reproduction,14,needs reproduction,1,0,130530,,,,0,706a78dc1265e1754ba54b7cf72b6f5a516015e97c56f196dc4c56361c3b9bbe
thread:issue:130728,issue #130728: torch.cuda.empty_cache() doesn't work correctly for the first instance of nn.Module set to cuda (and a potential method to deal with that).,needs reproduction,14,needs reproduction,1,0,130728,,,,0,e379175cf8be75a0bd3409388a461978bc5b82638fe95861a4f771da14152398
thread:issue:131185,"issue #131185: RuntimeError: Expected grad_output.numel() <= std::numeric_limits<int32_t>::max() to be true, but got false.",needs reproduction,14,needs reproduction,1,0,131185,,,,0,df46c56969509fea789e47f93f467416db05fff7fa7325f98dec0a24f2243010
thread:issue:132226,issue #132226: 关于yolov5在mac设备上使用pytorch的mps加速出现的各种问题,needs reproduction,14,needs reproduction,1,0,132226,,,,0,f61cf016dd4b7f452ad9beea9b7b1575a3f8345c4d758c29e14fe4e991dd204b
thread:issue:133093,issue #133093: Inconsistent results of Conv2d in pytorch between CPU and CUDA,needs reproduction,14,needs reproduction,1,0,133093,,,,0,22f95e562e632103a966df29f43d00faf7d7d2478ec236fff978c2a21aed7e6f
thread:issue:133808,issue #133808: pytorch/aten/src/ATen/cuda/CUDABlas.cpp:1422:28: error,needs reproduction,14,needs reproduction,1,0,133808,,,,0,a063189815d3c7f71270405dc5eb67ef9c85cbafdbc0fb63e958a3e7ce72b01d
thread:issue:133860,issue #133860: torch::tensor function does not work,needs reproduction,14,needs reproduction,1,0,133860,,,,0,281eaaccedda8cafedae5365ec52e03301349ac5e518df40accc011e783402a6
thread:issue:134331,"issue #134331: CMake Error: When installing PyTorch from source, CUDA not being detected.",needs reproduction,14,needs reproduction,1,0,134331,,,,0,63ce211546adab019e2bcd9a843b9801f95de2a117249545d65b3d791c1eaaff
thread:issue:134993,issue #134993: [bug] import torch failed after install the cpu version of pytorch,needs reproduction,14,needs reproduction,1,0,134993,,,,0,8fc5cac22ffbbe6886b423b65d98d14e1e474bcd99b5f30758cb09e14a0cfabb
thread:issue:135693,issue #135693: Python's REPL trying to do invoke `.T` / `.mH` / `.mT` on 0-dim tensor when doing tab-completion for some reason (default install of Python on new Ubuntu 24.04),needs reproduction,14,needs reproduction,1,0,135693,,,,0,c50b29947d92526a546e5a88eb003806e213abd360aca7c74de6697559f284e1
thread:issue:137078,"issue #137078: Missing librt.so.1, libdl.so.2, libz.so.1 and libpthread.so.0 in PyTorch Source Build (aarch64)",needs reproduction,14,needs reproduction,1,0,137078,,,,0,bce74852cbc42948473bddc1983f5cc5c0682a66addf40fb489608b3ce288233
thread:issue:137260,issue #137260: `import torch` takes forever,needs reproduction,14,needs reproduction,1,0,137260,,,,0,8c7012623819bce0279f6d8772ebfaa8bc973c3d52b9e0cf83d56cf6b488ee61
thread:issue:137393,issue #137393: Cannot allocate memory for thread-local data: ABORT,needs reproduction,14,needs reproduction,1,0,137393,,,,0,e1e56e64b99a8a3b3360789be0a9ceedcf8bff36ae2f244aea07c19b507081af
thread:issue:137524,issue #137524: FSDP-2 doesnt do overlapping when composed with TP,needs reproduction,14,needs reproduction,1,0,137524,,,,0,3f8d2b8db2118e17f3c0c612e196e2a4d305a4c8c8b9d2b68db93e6600b44874
thread:issue:137964,issue #137964: Network training silently fails with MPS backend,needs reproduction,14,needs reproduction,1,0,137964,,,,0,ea4131cb740db1b776e12c1353a111fb886b3d5f8485ea273c0ea793e4b9db77
thread:issue:138208,issue #138208: NaN value produced by F.conv2d (cuda) with large values in input,needs reproduction,14,needs reproduction,1,0,138208,,,,0,1f0b4226d45dd6c3b7960b5916b1df227c51ff18899c76bd5419366cda3667aa
thread:issue:138455,issue #138455: Tried to instantiate dummy base class,needs reproduction,14,needs reproduction,1,0,138455,,,,0,c988271255f5742673ec5d083be03304d4a62b0181fe1991c13c1369132975a0
thread:issue:138821,"issue #138821: Failed to load image Python extension: libc10_cuda.so: cannot open shared object file: No such file or directory warn(f""Failed to load image Python extension: {e}"")",needs reproduction,14,needs reproduction,1,0,138821,,,,0,6934db75917506500acebddf2cbd6eff7350cb859abb4dae97215ac286e330f6
thread:issue:139709,issue #139709: less GPU memory used in Centos than ubuntu,needs reproduction,14,needs reproduction,1,0,139709,,,,0,1d65802f958798aac0a914e1d9ab2b521c6bafac48fea1a870a2e02a2edf749e
thread:issue:139884,issue #139884: Persistent memory leak from failed pinned memory allocation,needs reproduction,14,needs reproduction,1,0,139884,,,,0,8bccc665d47ba1d2e20a803da5bd029a91f0ade4a1a14a55cf0f4a4bf59960ea
thread:issue:139977,issue #139977: Assigning the weight of the torch.nn.Embedding object to other variables and then performing subsequent operations on it will cause unstable training,needs reproduction,14,needs reproduction,1,0,139977,,,,0,2a90a0cf5cdcebbaa6367d78edcf0809d0b1dc8211e56952f436726d4a6bd00b
thread:issue:140096,issue #140096: upsample_bilinear backward is super slow in bf16 as compared to fp32,needs reproduction,14,needs reproduction,1,0,140096,,,,0,196b47e9ba449db2f95b271e5bbd618634a7e19bf32788c254c38158590c9ddf
thread:issue:140282,issue #140282: torch.norm produces wrong results,needs reproduction,14,needs reproduction,1,0,140282,,,,0,754c4b0dc5265caa26bd860fa2bb346bde9954e961b2a5c12219a60ff2661566
thread:issue:141465,issue #141465: Inconsistent Bias Broadcasting Results in PyTorch GPU `F.linear(...)` and `F.conv2d(...)` Operations,needs reproduction,14,needs reproduction,1,0,141465,,,,0,1ec63566b029dfac22ed696bc6d80b734282d43bfee918e994ec584dbf575f95
thread:issue:141482,issue #141482: Can't use dataparallel,needs reproduction,14,needs reproduction,1,0,141482,,,,0,41ced589ef442a08edbfc5ce070e95215c5f8ec43ee204ca5b44047a5c46ebb2
thread:issue:142302,issue #142302: torch.nn.functional.normalize get nan,needs reproduction,14,needs reproduction,1,0,142302,,,,0,f3798887d68fea2f4adb256710f1388229aa9afbbcfcf969851b3cce50be6c3c
thread:issue:143259,issue #143259: s.isIntegral(false) INTERNAL ASSERT FAILED,needs reproduction,14,needs reproduction,1,0,143259,,,,0,d19ce32dcd03c92cc5b4af4a58602354ae2d33a9749afac1bf46385c26f3dee1
thread:issue:143288,issue #143288: EXCEPTION : /python3.11/distutils/core.py,needs reproduction,14,needs reproduction,1,0,143288,,,,0,03d53e101ba13fb2c5eccf2f791faa2edd73bb4b65860d588c9ce694cfa065da
thread:issue:143338,issue #143338: c10::SmallVector unusable with gcc/g++ 12 and 13 with `-O3`,needs reproduction,14,needs reproduction,1,0,143338,,,,0,55d1990d3ff9e74c580f8883c10f279c4d7a6b08fbd6b6ee6fd7e17f5e7dd17e
thread:issue:143494,issue #143494: sympy.C.ConstantInteger has no method name,needs reproduction,14,needs reproduction,1,0,143494,,,,0,999f4bdbb5ec97b4c277321ef677c5e93d3b17c856c4ab725290bff184a8b30b
thread:issue:14352,issue #14352: [libtorch] Catkin_make compilation error,needs reproduction,14,needs reproduction,1,0,14352,,,,0,0ea895e0092054a94ceb637ec803c349f7852ec5c059f132dd4f8f97c3550c1d
thread:issue:143646,"issue #143646: Getting ""Could not initialize NNPACK! Reason: Unsupported hardware."" warning even though NNPACK is enabled",needs reproduction,14,needs reproduction,1,0,143646,,,,0,64798ae57ad77f77b9933f8e88189548377ef69e84423bd7e478c6548aba82e4
thread:issue:143651,issue #143651: `MemoryDenyWriteExecute` in systemd service causes `RuntimeError: could not create a primitive`,needs reproduction,14,needs reproduction,1,0,143651,,,,0,acd5956f81162d457f505842ebfbd4f8406cb251c424eeb58e23a4a6ad92d921
thread:issue:143800,issue #143800: The tensor-based computation of exponentiation and logarithmic operations is much slower than using NumPy,needs reproduction,14,needs reproduction,1,0,143800,,,,0,8667c8d30640c4311b183f8e4a3f2ffe7a6a5bc6b371475163113fd43f76a0b9
thread:issue:143852,issue #143852: [CPU][Operator] one channel_shuffle test of the operator benchmark has a Performance fluctuation issue,needs reproduction,14,needs reproduction,1,0,143852,,,,0,0a137682fd3d6a5eddca9d7ea745b143446d418875f822371f09301dfda97e70
thread:issue:143905,issue #143905: The special size tensor containing batches has a difference of a few tens of thousands in calculation results between CPU and GPU,needs reproduction,14,needs reproduction,1,0,143905,,,,0,3bbe1438aa67f31b62ce7a5b7769ca674a7756354b24a396291ede4bbd9e95b9
thread:issue:143906,issue #143906: How to correctly asynchronously copy a GPU tensor to a CPU tensor in another process without introducing blocking?,needs reproduction,14,needs reproduction,1,0,143906,,,,0,bfe20c83cf15c12a4243a2332ca558f1bbfa61bd340ec8ac09a4541f05c6aaa6
thread:issue:144071,issue #144071: torch-nightly doesn't support tesla v100,needs reproduction,14,needs reproduction,1,0,144071,,,,0,1fc13676a2754ee475774f5d1acfefd1f97b6011d9c17dcc12a0ef30b619490a
thread:issue:145827,issue #145827: What changed in torch 2.5.0 that cause photomaker to fail?,needs reproduction,14,needs reproduction,1,0,145827,,,,0,d53bd7dd847a6ab008d359076c24d0e194ad16f5f9a8c4584c4d08f105a5e4b4
thread:issue:146579,issue #146579: Pytorch model (.pth format) not loading in KServe environment,needs reproduction,14,needs reproduction,1,0,146579,,,,0,88ce8218a78a8139980011309d46112431d147042d59607547cc4b2e27856589
thread:issue:146769,issue #146769: MPS Error on sequoia 15.3: NDArray dimension length > INT_MAX',needs reproduction,14,needs reproduction,1,0,146769,,,,0,bd5638c84fe319a5910c848d63d6c19f69d54e3c032ea05ae30a722e63c77e72
thread:issue:148169,"issue #148169: PyTorch's nightly version no longer includes the CU118, CU124, and CU121 versions",needs reproduction,14,needs reproduction,1,0,148169,,,,0,30d3807dd4e859f93a12851b8b49bfc9924ed824881ba1eb1afead549b0a450a
thread:issue:149279,issue #149279: CUDA Assertion Error in Scatter Operation During Training (RTX5090 cu128),needs reproduction,14,needs reproduction,1,0,149279,,,,0,686aca21ddaf81924515b0bec13506728857441984406b32b29c862a1d9992b1
thread:issue:149508,issue #149508: Adam optimizer ValueError: beta1 as a Tensor,needs reproduction,14,needs reproduction,1,0,149508,,,,0,38671196e086a35b9e65d2242fd51fe3767ea2f5e41cc01fd62cf3796222d45f
thread:issue:149937,issue #149937: torch.inverse can't export onnx opset16,needs reproduction,14,needs reproduction,1,0,149937,,,,0,9ecf3a2287a2b6dd133f19531aedf329afc1a2ce3c724b4d301a9d4186e3f5a4
thread:issue:150912,issue #150912: using torch.compile with torchao at the same time cause stack overflow error,needs reproduction,14,needs reproduction,1,0,150912,,,,0,422847cae3d7afe4e918adf2523c644d7bf4344463d7c78c54cac9f1f5397815
thread:issue:150921,issue #150921: `M(*[torch.from_numpy(v).to('cpu') for v in inp])` hang when start with `multiprocessing.Process`,needs reproduction,14,needs reproduction,1,0,150921,,,,0,7424dd3162aa9f25628fdf2a00c2f658e01059661a4552cd2c9e5ec8f1032e36
thread:issue:150998,issue #150998: Unsupported operand 118,needs reproduction,14,needs reproduction,1,0,150998,,,,0,569f50bd01a7b9eb9825b64d61a7cd93c9f8e0f37d61204994e8c71c77404af6
thread:issue:151201,issue #151201: I can't compile Pytorch 2.0.0 because ninja: error: build.ninja:10911: multiple rules generate caffe2/CMakeFiles/torch_cpu.dir/__/aten/src/ATen/UfuncCPUKernel_add.cpp.DEFAULT.cpp.o,needs reproduction,14,needs reproduction,1,0,151201,,,,0,cfd10e8a9ad91acd23db1b6824dd74145675b1f9272766122ba6674bf76da165
thread:issue:151310,issue #151310: cpp_extension.load 'sources' does not support Chinese paths,needs reproduction,14,needs reproduction,1,0,151310,,,,0,091be8a503c931962e3959fdb07156b175aac7231557d0026c70c60ad04182b7
thread:issue:151667,issue #151667: [MPS] MultiheadAttention with masks and dropout produces NaNs,needs reproduction,14,needs reproduction,1,0,151667,,,,0,1af10d3d7969cbf01d555939d7db997318a10a8385951622316979fe965e78b0
thread:issue:152135,issue #152135: Unaccaptable OOMs all the time.,needs reproduction,14,needs reproduction,1,0,152135,,,,0,cd5b0ae32e264fe9e57e7d93b884019a41a848fb26947f9713be7ec000b36d60
thread:issue:152592,issue #152592: [ratter-build] Cannot detect CUDA when build from source,needs reproduction,14,needs reproduction,1,0,152592,,,,0,ab061a4e85a8c93cb52a3531498d5eeb3fde80e0b0b00be9fa6acb5ac024fe83
thread:issue:152783,"issue #152783: undefined symbol: __nvJitLinkCreate_12_8, version libnvJitLink.so.12",needs reproduction,14,needs reproduction,1,0,152783,,,,0,4ef829245a8004bd630accaa0218886ff8ea254b49f22518ac8d66d196a59313
thread:issue:153033,issue #153033: missalignment with differenet shape in F.linear with bf16 dtype,needs reproduction,14,needs reproduction,1,0,153033,,,,0,d92106f94f7c9692856d37a27dc36e86f447f0957e6a7ebf4d48590899f81167
thread:issue:153564,issue #153564: Sum difference for equal channels of tensor,needs reproduction,14,needs reproduction,1,0,153564,,,,0,676b31f52775210ecea2d4e6c45f7e232b6a60a7736d1d5dc3631ab05564739e
thread:issue:153574,issue #153574: Documentation Preview and Build Not Reflecting Latest Changes,needs reproduction,14,needs reproduction,1,0,153574,,,,0,3680c665e665e3562ea7547ec745fed7483378745e4fd5cc2b5d5329978e088f
thread:issue:154266,issue #154266: torch dynamo fails on GH200 with world size 5,needs reproduction,14,needs reproduction,1,0,154266,,,,0,7e21f6ad72538022e9a3ce5fc5419baa8a1a6d4c34361e49f6d8f3a4e4170ff3
thread:issue:154302,issue #154302: Inconsistent output of `torch.func.jvp` calculation,needs reproduction,14,needs reproduction,1,0,154302,,,,0,3e35e5d00f97d2203a190a2be059b07ed41833fb632b015b552a9fdad1e1d7d8
thread:issue:155344,issue #155344: PyTorch CPP Extensions fail when same kernel is compiled more than once on ROCm servers,needs reproduction,14,needs reproduction,1,0,155344,,,,0,6a897a568db62a1fbdd70057d69f606683ee5e763e3f07e5108f27fffddf8a0f
thread:issue:156010,issue #156010: `torch.prod` or `torch.special.entr` triggers `CUDA driver error: invalid argument` on GPU unless kernel cache is cleared,needs reproduction,14,needs reproduction,1,0,156010,,,,0,8434fa419919f4cc6c07f64974e460bd28972eb5c228210e8f38e8a74837ca9c
thread:issue:157237,"issue #157237: Unexpected, batch size and device dependent NaN propagation in Conv1d",needs reproduction,14,needs reproduction,1,0,157237,,,,0,37f0e29163969802c40b612b290d81df959cd35bdf8e0fffa86562cac7a9fa6d
thread:issue:157360,issue #157360: torch.Tensor.addmm_ The calculation result is inconsistent with the formula calculation result,needs reproduction,14,needs reproduction,1,0,157360,,,,0,66c1a71dc2b54dd00e58e735c3f3d3bb5ce06460f540499d8d4d90cfd3430dc4
thread:issue:157398,issue #157398: Cannot copy data from one gpu to another using torch,needs reproduction,14,needs reproduction,1,0,157398,,,,0,ecab7128ed35a1be2aa33ce18324cd62b1d0390441d5778431d9c9792d720cc2
thread:issue:157709,issue #157709: /lib/x86_64-linux-gnu/libstdc++.so.6: version `GLIBCXX_3.4.30' not found when IMPORT TORCH,needs reproduction,14,needs reproduction,1,0,157709,,,,0,2ab50a6765c80b421843b0f253416af68ee7e7606bba021fc31128ba8742d5db
thread:issue:158906,issue #158906: The make lint command got stuck.,needs reproduction,14,needs reproduction,1,0,158906,,,,0,6309a1bc6e6fe8645e9ffd268c54476ffbe8b9776dc5f8f476f09d20f56ae404
thread:issue:158986,issue #158986: `nvlink error: Entry function '<redacted for brevity>' uses too much data for compiler-generated constants`,needs reproduction,14,needs reproduction,1,0,158986,,,,0,0f073593ded6c2d95b8613edda1e100cd977071636a0ada569cbafd865fa75f0
thread:issue:159196,issue #159196: torch.compile: device of tensor switches after the next statement is run. no repro,needs reproduction,14,needs reproduction,1,0,159196,,,,0,e2f61347262c981ce24cc8f3da54539dd990bd5d6512201147c05c1143e6df5d
thread:issue:159224,"issue #159224: Import Error: Undefined Symbol, _PyCode_SetExtra: Torch Extensions doesn't seem to support Python 3.12",needs reproduction,14,needs reproduction,1,0,159224,,,,0,d553ef93cdc54194763e9f3d796b56da6921cac861348d6a5d6103e2dbb24534
thread:issue:159309,issue #159309: torch.nn.InstanceNorm2d CPU/GPU Inconsistency,needs reproduction,14,needs reproduction,1,0,159309,,,,0,7174f500f400f591e7ea0da1b2f771f69a9cd50b6da802b552432bdeb73f376f
thread:issue:159669,issue #159669: Torch Compile. CudaGraphs Memory leak,needs reproduction,14,needs reproduction,1,0,159669,,,,0,fb9b95d06a0c0a514f54eecfe3cbbddd707cc683c001a82c82f6e745e17f75ae
thread:issue:160261,issue #160261: CPU: Pytorch >= 2.7.0 is broken,needs reproduction,14,needs reproduction,1,0,160261,,,,0,6149b4eecbd986d3f766fe7759e98a7de3a6b231fab72319b89b9e809480d7c2
thread:issue:160846,issue #160846: Can't run a model trained in MPS-system on CPU-only system.,needs reproduction,14,needs reproduction,1,0,160846,,,,0,55a9e038b10544b0409cd2b8a4a8ad160d7c3e8c9c8055e30edf60a50cc776dd
thread:issue:161104,issue #161104: Eager CUDAGraph + stream performance,needs reproduction,14,needs reproduction,1,0,161104,,,,0,00aaf5e9e48beed41f2eed21a4beee3bb7fc1e435419bcb4dae5b8e77d615cbc
thread:issue:161494,issue #161494: OOM,needs reproduction,14,needs reproduction,1,0,161494,,,,0,488d18e104a5ccafea259bed5f5d82b97db6352404c2536117a62574a973d8b5
thread:issue:163347,issue #163347: Incorrect bf16 results in Conv,needs reproduction,14,needs reproduction,1,0,163347,,,,0,520f8befcff24cc257925e887c1c82b326fe5f40ae42df8fd9b55bba2e4afae3
thread:issue:163510,issue #163510: building pytorch c++ extension with cuda 13 causes unexpected behavior when pytorch is built with cuda 12.8,needs reproduction,14,needs reproduction,1,0,163510,,,,0,f258a62c442240a44b9a884f77008fa75e1b085c7980707330987ec414f559da
thread:issue:163923,issue #163923: Enabling activation checkpoint debug makes throughput gradually slow down to crawl.,needs reproduction,14,needs reproduction,1,0,163923,,,,0,369b150bf20b7cd38a8ba2b5160e8292cedaaf1bd4d4490fa4867b3a54e2ae15
thread:issue:165590,issue #165590: RuntimeError: non-positive groups is not supported,needs reproduction,14,needs reproduction,1,0,165590,,,,0,12f54a3369fab0ae0674a4183fda291320158768884dfc9b50b63e4d7782a5e5
thread:issue:165798,issue #165798: matmul/mm/bmm crashes on CUDA 13.0 + Libtorch 2.9 on Ubuntu 24.04,needs reproduction,14,needs reproduction,1,0,165798,,,,0,a436f582dd29ecb4d8638128cba12dbb860155d2d0659a9c2893fcae986fe093
thread:issue:166303,issue #166303: Pytorch Operators on older pytorch version,needs reproduction,14,needs reproduction,1,0,166303,,,,0,9eace4416632e9ffc6b922cf4f465b2f10ef1af4f6bd33795ec0577304d284ef
thread:issue:16654,issue #16654: pytorch was blocked at loss.backward,needs reproduction,14,needs reproduction,1,0,16654,,,,0,2f84a7c0257811303341f578c4563be9366d58c7acfd4bc9b6531921babc3b37
thread:issue:167820,issue #167820: Why torch==2.9 compile qwen3 model with block ptr will crash?,needs reproduction,14,needs reproduction,1,0,167820,,,,0,86b8357a4504a630fe6f865ab27e981e73bf526262b430f6fa66cd0859c56257
thread:issue:167833,issue #167833: Incorrect SHA256 on index for torchvision==0.24.1+cu130,needs reproduction,14,needs reproduction,1,0,167833,,,,0,addb8f99aea208e6ac9ce99e4f1ac1e294810462430c53c7a973cfccf6c0afd9
thread:issue:168362,issue #168362: Triton 3.4.0 wheel hash mismatch between PyPI metadata and download server,needs reproduction,14,needs reproduction,1,0,168362,,,,0,b574714a766e1cde25a144ecfd3846980c0005adf92c32b3e110cc40d4976451
thread:issue:169082,issue #169082: Reduce Overhead and Max Autotune NaN,needs reproduction,14,needs reproduction,1,0,169082,,,,0,d6c7b177b62adb26ad295e5206d0dfe88551eb051fca0904040cfd78d944d605
thread:issue:169342,issue #169342: [MPS] Batch inference produces incorrect results when chunk() views are passed through conv operations,needs reproduction,14,needs reproduction,1,0,169342,,,,0,a5d24679cf81665610e0b76dff2e4d4973eedc7a2251e8d6258371934960395c
thread:issue:170016,issue #170016: Conv2d produces numerically inconsistent results compared to other DL frameworks (TensorFlow / Paddle / MindSpore),needs reproduction,14,needs reproduction,1,0,170016,,,,0,e64190213f58407627f4a143f67ffe17d8e5c41676317598095233123809236c
thread:issue:170278,issue #170278: macOS Parallel Builds Fail with jobs > 4,needs reproduction,14,needs reproduction,1,0,170278,,,,0,955456e89b43df6b161cb59ee80a043e810ed032b55ec7410acc53896136eda4
thread:issue:170524,issue #170524: RuntimeError occurred when I train “Resnet”,needs reproduction,14,needs reproduction,1,0,170524,,,,0,bc52fa7b9f83c192d3ed5eadbadc8dc3de58c2ed1b24f6ce4801be415c0fae16
thread:issue:171092,issue #171092: Load nVidia libraries in reverse order so that the newest version can take control,needs reproduction,14,needs reproduction,1,0,171092,,,,0,bd2db50a42562523cf8b3b06222b6fbf7696523034e16114962f44172de9e2c2
thread:issue:19029,issue #19029: C++ custom module not thread safe,needs reproduction,14,needs reproduction,1,0,19029,,,,0,23a815e03e649299d1ec4a05f44263d2991cf2610c2e8a0fc7f0fc94bb7f09d8
thread:issue:19128,issue #19128: pytorch.version.cuda is None when compiling with CUDA support,needs reproduction,14,needs reproduction,1,0,19128,,,,0,0907367dde6e970bae6ddd6bda6f569f2a4f215ce94ceac48c78bfe71ff75c4f
thread:issue:20134,"issue #20134: ERROR: Command ""python setup.py egg_info"" when dockerfile build",needs reproduction,14,needs reproduction,1,0,20134,,,,0,4e977426ace03e18cf95aa0c89621cfd921786428171b3adb9d8217318fafc19
thread:issue:20805,issue #20805: LibTorch :About torch.jit.trace generate model.pt,needs reproduction,14,needs reproduction,1,0,20805,,,,0,b281d3d327a579bae8d9b6c8eb4715da13ba4fbe754dc0bd20d936aecfd9649d
thread:issue:21179,issue #21179: Failed to install pytorch from source on ubuntu.,needs reproduction,14,needs reproduction,1,0,21179,,,,0,6b03aadb287d2acd13d085ca29c1fcd4126869a9b5a67e9d52d658633740ebf9
thread:issue:22597,issue #22597: Autograd profiler memory leak when use_cuda=True,needs reproduction,14,needs reproduction,1,0,22597,,,,0,74ea4d8e763890bc65f62b9e94a4e296317acdc8f1103cf7d86f39830f412521
thread:issue:24397,issue #24397: SyncBatchNorm error when using model.eval() with DistributedDataParallel,needs reproduction,14,needs reproduction,1,0,24397,,,,0,0829f79c2b4c9afe6dacb2149fc714b5616416a9b083e63f2c5aaf22135dc465
thread:issue:24398,issue #24398: Significantly slower in latest version than in 0.4.0,needs reproduction,14,needs reproduction,1,0,24398,,,,0,07fbe04ee21b494991f4cda48096b946ffbd36c1520f4ea0b734c4c78b3ae02e
thread:issue:2482,"issue #2482: ""Shared memory manager connection has timed out""",needs reproduction,14,needs reproduction,1,0,2482,,,,0,fff1132458f032b25d6604c7156e3e37dda034b6bf02d561160153548e608133
thread:issue:25190,issue #25190: Error in python3: double free or corruption (fasttop),needs reproduction,14,needs reproduction,1,0,25190,,,,0,6bcdb2b5e6b7e46058a430435e88297be59fc486dde04c2cfb8f01c7c39f10be
thread:issue:2539,"issue #2539: DataLoader converts cuda FloatTensor into cpu DoubleTensor when shape is (n,)",needs reproduction,14,needs reproduction,1,0,2539,,,,0,b9a00222032fba2e2bdea9d7d800732e694a5c5144e722dcf54b6ae998bb8985
thread:issue:25904,issue #25904: Python hang after using torch.exp(),needs reproduction,14,needs reproduction,1,0,25904,,,,0,5a6a9be950f2e2d757ce76580c6094f2f0271ec1f18e79b8976b35cb4fc4f31a
thread:issue:26400,issue #26400: RuntimeError: tensor.ndimension() == static_cast<int64_t>(expected_size.size()) INTERNAL ASSERT FAILED,needs reproduction,14,needs reproduction,1,0,26400,,,,0,60c02134b6e4fa08012ba395c26688c61de517db918b328624d85ab7dba1affb
thread:issue:27694,issue #27694: Deployment training model at C + + end,needs reproduction,14,needs reproduction,1,0,27694,,,,0,1a517ff387926ddbd919c97fb204def979066ec3f2ca4d9315fac67272b4639c
thread:issue:28943,issue #28943: The `@` operation uses too much memory on GPU,needs reproduction,14,needs reproduction,1,0,28943,,,,0,120b8d11225ed7b22ac2b9ea62388d2b3851d2a5eaf80fe28ca96e4fb7fff7cd
thread:issue:28950,issue #28950: ConnectionResetError when using dataLoader with pin_memory=True,needs reproduction,14,needs reproduction,1,0,28950,,,,0,e45cbbfb3ec0b133f6b0d9aebe526be5ae59f70c5f1b66b67d0582bcd2a0ce46
thread:issue:28956,issue #28956: RuntimeError: CUDA error: unknown error,needs reproduction,14,needs reproduction,1,0,28956,,,,0,970f63d32b7859e730676270207795f793b502952703e03a65cea95a6357f9c1
thread:issue:29891,issue #29891: Get wrong precision when multi nodes run in docker,needs reproduction,14,needs reproduction,1,0,29891,,,,0,cda9386abdde59f51647922ca45f03ff580fda4293dac31276696ffa71370032
thread:issue:30507,"issue #30507: LibTorch, Error in 'xxx': free(): invalid pointer",needs reproduction,14,needs reproduction,1,0,30507,,,,0,ae383f5846a15f50f68550881594da4a429b4f70a35639483fa3f69b676fa5dc
thread:issue:30830,issue #30830: Torch getting stuck transfering model to GPU in multiple GPU setting,needs reproduction,14,needs reproduction,1,0,30830,,,,0,c2d445d65f1a07aee5f6481a9fbfcb8c42c283e2599ac728beb0ec17c2b9b2d7
thread:issue:30955,issue #30955: error C3203: “templated_iterator”: 未专用化的 类 模板 不能用作 模板 变量，该变量属于 模板 参数“_Ty1”，应为 real 类型,needs reproduction,14,needs reproduction,1,0,30955,,,,0,56f3ced6423d3dff49709080702eb65315d90c6e2961491e74a571deb16eefb3
thread:issue:31004,issue #31004: The inference speed of the torch compiled manually is slower than the torch build from official binaries?,needs reproduction,14,needs reproduction,1,0,31004,,,,0,8ff83cd70477b55508050fc04b7979d08e2212ce517078ae1212a241a480e04f
thread:issue:31660,issue #31660: Cuda error 59 : device-side assert triggered,needs reproduction,14,needs reproduction,1,0,31660,,,,0,79cce08d04d20dbe03792dee27079dc985f26b24b1e7d28f72dbf97cfbac32a3
thread:issue:32443,issue #32443: Segmentation fault in lazyInitCUDA -> CUDAHooks::initCUDA -> THCMagma_init -> magma_init,needs reproduction,14,needs reproduction,1,0,32443,,,,0,27956b7b66e1b524be915494d5422c4f37191412064385835bb70bfc7fed86d7
thread:issue:32584,issue #32584: Segmentation Fault (core dumped) with 1.4.0,needs reproduction,14,needs reproduction,1,0,32584,,,,0,57aa23bfed37ccdd1d2397733a2f103710db998989fdd339758fb4a3e50ad5f3
thread:issue:32662,issue #32662: Second-order gradient cause segfault over time,needs reproduction,14,needs reproduction,1,0,32662,,,,0,fdfb516dee367512c2baf957746888090c8ec5c74b6e1c458924d6407f0b78e7
thread:issue:34937,issue #34937: [JIT]torch.jit.export invaild for pytorch1.4,needs reproduction,14,needs reproduction,1,0,34937,,,,0,70012ff31fb34838014107aaef33360ffc1702fe1e1a90282363e6b22cd912f6
thread:issue:34948,issue #34948: TracerWarning: Output nr 1. of the traced function does not match the corresponding output of the Python function,needs reproduction,14,needs reproduction,1,0,34948,,,,0,fc699c7c33def7093d0124f359c8dd818f3e46d878e1ea2a6c74cedb5b311c8f
thread:issue:35754,issue #35754: pytorch 1.4.0 hangs when using with CUDA >= 10.1,needs reproduction,14,needs reproduction,1,0,35754,,,,0,c582300a66660552e0eeaad240e5a345e438279eb9f076ad07712299313c1b20
thread:issue:35870,issue #35870: CUDNN_STATUS_INTERNAL_ERROR with GPU RTX 8000,needs reproduction,14,needs reproduction,1,0,35870,,,,0,83018f8214968ebb502b14c994ecc59ddc317481095721eb85fb9575e4460128
thread:issue:35990,"issue #35990: [JIT] Tensor method API behavior discrepancy, Tensor.detach(..)",needs reproduction,14,needs reproduction,1,0,35990,,,,0,50f4ea9440b91abd4a86f7ee5b9d73022f3ffe751e7ef79d1f1f93d42f6dce82
thread:issue:36649,"issue #36649: RuntimeError: Expected cuda::check_device({sparse_, r_, t, dense}) to be true, but got false.",needs reproduction,14,needs reproduction,1,0,36649,,,,0,2576cde5089085262b2e87380fe3f77ae0083c5ce7df88f53f66dc1445e1d7d3
thread:issue:36721,issue #36721: [1.4.1] Cuda build fails,needs reproduction,14,needs reproduction,1,0,36721,,,,0,5a742a46612de8a41f8f8f87dc82bab10618868bc6bfd97a199c052e279f7552
thread:issue:37057,issue #37057: CUDA error: device-side assert triggered @ model.cuda(),needs reproduction,14,needs reproduction,1,0,37057,,,,0,776b00d22a386d9ebce8a5cd199a5485ac85d8535471ca38d1c100ba78837511
thread:issue:37066,issue #37066: Saved model behaves differently on same data,needs reproduction,14,needs reproduction,1,0,37066,,,,0,96ec6ed6b913f279e5a074e7814f9b725300727fa3daf564b95885cb3f96b352
thread:issue:37067,issue #37067: can't wrap two models in the same class,needs reproduction,14,needs reproduction,1,0,37067,,,,0,f2101f027ce9c8f2134009224500d61d27b58b6260b9fe6f3698b7be4d414ff6
thread:issue:38155,"issue #38155: Python script name with ""profile.py"" will run twice",needs reproduction,14,needs reproduction,1,0,38155,,,,0,d7130e220f9b67b94a6663a6576d92f0262a4ccd057576d7ee26f81b94a22704
thread:issue:3863,issue #3863: Considerable slowdown in Adam.step after a number of epochs with multiple losses,needs reproduction,14,needs reproduction,1,0,3863,,,,0,27a05ac8f708887b484052875171b71f24cec056946b051a8649d995d25cced6
thread:issue:39832,issue #39832: Torch Distributed Asynch docs with invalid ENV args,needs reproduction,14,needs reproduction,1,0,39832,,,,0,ae653cecd75715722d87a2ca7240211df51b6cc1138490032bc1d331372c4f16
thread:issue:40133,issue #40133: Why does Libtorch use more memory than Pytorch does in Python？,needs reproduction,14,needs reproduction,1,0,40133,,,,0,4e210abd6b5114143b0c160dfd1a1e473a42594401b84b3ff3d9eeb17b9bb584
thread:issue:42581,issue #42581: Build script complains that gcc is < 6 even though gcc version is 8,needs reproduction,14,needs reproduction,1,0,42581,,,,0,18ec1d3251b4778d98edd9fd2353ffb105acce027e9e7a638649579c05337b0d
thread:issue:4334,issue #4334: torch.cuda.device_count() returns 1 using 4 TitanX setup.,needs reproduction,14,needs reproduction,1,0,4334,,,,0,c748a5f6ff2770619bbd470fe98e72c4c9c62c4c1e1d6059832fafb81a320545
thread:issue:43859,issue #43859: Multi-machine multi-gpu training won't start on CNGrid,needs reproduction,14,needs reproduction,1,0,43859,,,,0,4276c6fa576a07f1a6333be3428e1ca3f7e7715f6ab2bc0ff2286f2ecff2e7ed
thread:issue:44024,"issue #44024: When the pytorch training data reaches a certain epoch, the memory remains unchanged?",needs reproduction,14,needs reproduction,1,0,44024,,,,0,6fe246a3dc01a9e992c8c497a85069159df25ad896495ae2e58925f5ff3a8e41
thread:issue:44855,issue #44855: RuntimeError: Caught RuntimeError in replica 1 on device 1.,needs reproduction,14,needs reproduction,1,0,44855,,,,0,1bff12d37edcf4e46a533eea0510effaa9f35235bd80c9715d45a007009ce8b7
thread:issue:45273,"issue #45273: Training fast with small dataset, slow with large dataset",needs reproduction,14,needs reproduction,1,0,45273,,,,0,1591a2a7a2604ad6a25e34d13f17b4a8eb568e1908215b0ddae18f464f819896
thread:issue:46153,issue #46153: Getting this error while installing torch and torchvision!,needs reproduction,14,needs reproduction,1,0,46153,,,,0,3ef974e87e01daa146836ff81b62e81424faefe291fbea434a59112e2cc5e92e
thread:issue:46806,issue #46806: Compile Pytorch with MAGMA Issue,needs reproduction,14,needs reproduction,1,0,46806,,,,0,38e33031d81619c0709d62981cfa27066ef9953d9d8d4cc17fa856be3c9c0a0b
thread:issue:47128,issue #47128: AMP doesn't gracefully handle optimizers for disabled regions,needs reproduction,14,needs reproduction,1,0,47128,,,,0,f41c6f5adfea55b5626836f7f207c0c0f3a0cc4e9ed4504da4612925c409ee2a
thread:issue:47832,issue #47832: Error during distributed training,needs reproduction,14,needs reproduction,1,0,47832,,,,0,60fab01ddd305846a578df39860ab58f5c6bc8a9ce6470d62ec2ad9c164627d7
thread:issue:47990,issue #47990: Is nn.Conv2d equivalent with Unfold + Matrix Multiplication + Fold ?,needs reproduction,14,needs reproduction,1,0,47990,,,,0,990b8040a4cfadfde40a3855a30862aa31848c5d7c6c5fce3c795f367fa320ff
thread:issue:48579,issue #48579: test_fs_pool fails,needs reproduction,14,needs reproduction,1,0,48579,,,,0,19f0a5d05471924ae48425b56dd55ae3974f91c82fccf3b25d4240b1a0a61cdf
thread:issue:48580,issue #48580: torch.save() fails when attempting to save to mounted drives,needs reproduction,14,needs reproduction,1,0,48580,,,,0,612cf64e18e1354d21d878eac50339f5cf4f08d70ffb66ba45bbdeb4d858d0b8
thread:issue:49669,"issue #49669: While training the model, GPU util reaches 100% but no progress happends.",needs reproduction,14,needs reproduction,1,0,49669,,,,0,2e347462a68063d397670604e71c29a1f900195a9b3a737c9758aafe401a86e4
thread:issue:49832,issue #49832: RuntimeError: CUDA error: unspecified launch failure,needs reproduction,14,needs reproduction,1,0,49832,,,,0,903a2d5e5fed6d5aef962e419252894fffdc284f4c1c303379dab5dcd9e52d19
thread:issue:49870,issue #49870: Tensorboard Error,needs reproduction,14,needs reproduction,1,0,49870,,,,0,0c7248f6d60431b25a8f3fba5c0ed95187bddc54e70c830df13ecfa0977880b1
thread:issue:50261,issue #50261: ModuleNotFoundError: No module named '__torch__',needs reproduction,14,needs reproduction,1,0,50261,,,,0,db8975db2a2d71d34e1f8fe39db23f038302eb9f3b69946be054a68badbc8dc1
thread:issue:50779,issue #50779: pytorch is built without _GLIBCXX_USE_CXX11_ABI and can cause std::regex crashes (probably),needs reproduction,14,needs reproduction,1,0,50779,,,,0,4df19f6843abc9f14500f26f9a1c263fb741e80e80499fdf38a259a4500d554e
thread:issue:50803,issue #50803: pytorch DDP hangs at .backward() call,needs reproduction,14,needs reproduction,1,0,50803,,,,0,d6154354256a592eaf5bb8ddff6bf28bba9f14c21c24cf96b95e9715f1718f83
thread:issue:51275,issue #51275: Error of running,needs reproduction,14,needs reproduction,1,0,51275,,,,0,c29038118be07bd032b96cd1625820f27cc92929e71b68876be8c08e1ba2b89c
thread:issue:52307,issue #52307: Model produces same outputs for different inputs when using GPU,needs reproduction,14,needs reproduction,1,0,52307,,,,0,69636c9ca35d2be3a93125e7b866eac20444ebc852a35db544c0b19d33c8d845
thread:issue:52515,issue #52515: torch.bmm incorrect with pytorch 1.7.1 and cuda 11,needs reproduction,14,needs reproduction,1,0,52515,,,,0,a0c1c4b74d860c37751e3ee23671cb52111814523b26cf1548344934bd0239e3
thread:issue:52989,issue #52989: lr_scheduler _triangular2_scale_fn calculation is overflowed,needs reproduction,14,needs reproduction,1,0,52989,,,,0,d2aab0beee36f828ae8571c400802c6c27adc3c00e92313e8ad1f6ae6ceb218b
thread:issue:53464,issue #53464: libtorch reports bad_alloc when used together with darknet,needs reproduction,14,needs reproduction,1,0,53464,,,,0,085f55ad0657765b14956308584a4401232f34a27a37bce5a30a5623f1dec0bb
thread:issue:54487,"issue #54487: RuntimeError: iter.device(arg).is_cuda() INTERNAL ASSERT FAILED at ""/opt/conda/conda-bld/pytorch_1607370116979/work/aten/src/ATen/native/cuda/Loops.cuh"":94, please report a bug to PyTorch.",needs reproduction,14,needs reproduction,1,0,54487,,,,0,3cdae34a08f11c049363eddf9ae3464741a78eeb80a0d802859ba8bb2527fd65
thread:issue:54577,issue #54577: Question when I use Pytorch Mobile,needs reproduction,14,needs reproduction,1,0,54577,,,,0,678fa18b7b09f2c75d782f43a6ac24a485710889172dee510ae35b8a73e17647
thread:issue:54716,issue #54716: Segmentation fault in PyTorch dataloader,needs reproduction,14,needs reproduction,1,0,54716,,,,0,24c6fa25c95fe18193287401a0b9aae89e633d4c1abd025e636a6038f3599409
thread:issue:54799,"issue #54799: Vertices=torch.matmul(vertices.unsqueeze(0), rotations_init), RuntimeError: CUDA error: CUBLAS_STATUS_EXECUTION_FAILED when calling `cublasSgemmStridedBatched in CentOS",needs reproduction,14,needs reproduction,1,0,54799,,,,0,b6a4920d687c2915e88b22ec4ced5c0c2c89d1e00da664f7519a4bc6daadfeeb
thread:issue:54827,issue #54827: Import error,needs reproduction,14,needs reproduction,1,0,54827,,,,0,6dfc9ae5674f42ef78dfac2fc11640f773f20e7043e7856dfe034765c0f54697
thread:issue:54975,issue #54975: CUBLAS_STATUS_EXECUTION_FAILED error on torch >= 1.8.0 and CUDA 11.1,needs reproduction,14,needs reproduction,1,0,54975,,,,0,62748d3a6ef1c5e3d439b366a3ae0589efa8f7affe41180a9d6257696d10f1b6
thread:issue:55035,issue #55035: DataLoader performance drop on 4-channel images?,needs reproduction,14,needs reproduction,1,0,55035,,,,0,a861c26e674225de36694ded61d655709865bf4065131ed00bbcb3138a69ea32
thread:issue:55663,issue #55663: torch.nn.functional.log_softmax on Arm CPU gives partly NaN results,needs reproduction,14,needs reproduction,1,0,55663,,,,0,ec417bb1f8f5d4a4d241fef9da3868bc8a8eb80bae346a319c27f78be8c1174a
thread:issue:56019,issue #56019: when building apk with libtorch,needs reproduction,14,needs reproduction,1,0,56019,,,,0,b5f428f19821bd29fb3ae9ac1455b15965d460acb77c06cf4bc4cee0da52d6c2
thread:issue:56771,issue #56771: Problems with initial communication between GPUs,needs reproduction,14,needs reproduction,1,0,56771,,,,0,c1015ca7eadfe0fcf7e9bf05021e9ad90469a3252ed2085256e5f04ee8e07f9a
thread:issue:56928,issue #56928: torch-1.8.1 wheel verification fails with distlib,needs reproduction,14,needs reproduction,1,0,56928,,,,0,c054d38c2a492487543f34b52c412abb169805f09145aab48b5b05275077de80
thread:issue:57491,issue #57491: setting all seed still get different result.,needs reproduction,14,needs reproduction,1,0,57491,,,,0,d2d3e00b60e980ed259406e37963e215c639cb9cdf3f9a48c98e799bdc435043
thread:issue:57794,issue #57794: CUDA error: an illegal memory access was encountered,needs reproduction,14,needs reproduction,1,0,57794,,,,0,1f89d5cfde2e05bc5c8307563c096ed75bc1203c726d35a6c5f59aba6c6493ac
thread:issue:57796,issue #57796: BatchNorm grad calculation is imprecise,needs reproduction,14,needs reproduction,1,0,57796,,,,0,311f08f3f457282f687735b8c85550b01c936fdcbad956cf7815b1425f801787
thread:issue:58135,issue #58135: Segmentation fault need help,needs reproduction,14,needs reproduction,1,0,58135,,,,0,0fdb56d2d1440c15c39443ffbd1eb81503e95517bbe1bd539708ec570d97eab3
thread:issue:58392,issue #58392: torch.cuda.amp.autocast nan values when model.eval(),needs reproduction,14,needs reproduction,1,0,58392,,,,0,44967f1169a10a47c15fdf4828a92ba2a4ceda0f23d6e3013705ba4892b0b69e
thread:issue:5953,"issue #5953: LBFGS always give nan results, why",needs reproduction,14,needs reproduction,1,0,5953,,,,0,4f9f6ae4037d770049058e9021b25d8255d7a3d99d051b0e76b85f6a9569ce78
thread:issue:59626,issue #59626: multi scripted model on the same gpu on multi gpu machine,needs reproduction,14,needs reproduction,1,0,59626,,,,0,f011f3d4693773c9154fe22f51c5ca5c705487fab5c285764edc54451dd9142f
thread:issue:59628,issue #59628: what(): result type Float can't be cast to the desired output type Long,needs reproduction,14,needs reproduction,1,0,59628,,,,0,7c84b507652934a69db59d88c28d92d155929605c36d4ffe9cbc03c6d35a7126
thread:issue:59883,"issue #59883: matmul causing cuDNN error: CUDNN_STATUS_INTERNAL_ERROR at future, unrelated Conv2d",needs reproduction,14,needs reproduction,1,0,59883,,,,0,934c96ea31b3ca894acd7bcc01a92366a83b900b99b8ceb9932345a0ee2d9c14
thread:issue:60094,issue #60094: bundled libiomp5 causing segfaults in other libraries that use libomp,needs reproduction,14,needs reproduction,1,0,60094,,,,0,1edf96c77a4125c8303d9fe09c45fc3a689976631ac855fc0b43f2e3ec934ef8
thread:issue:60477,issue #60477: ROCm miopenStatusInternalError /MIOpen/src/sqlite_db.cpp:109: open memvfs: unable to open database file,needs reproduction,14,needs reproduction,1,0,60477,,,,0,fbbc7aa0ab47076f8a08823cd77223b051bc490652ca92695b95c65757d64786
thread:issue:60923,issue #60923: RuntimeError: derivative for im2col_backward is not implemented,needs reproduction,14,needs reproduction,1,0,60923,,,,0,1179c49864a5089a4c6c3165ff6a0da5641f790ba74d0046ef985b9cc4fd9b89
thread:issue:60936,issue #60936: The usage of get_stoi() in 0.10.0 and stoi[] in previous versions,needs reproduction,14,needs reproduction,1,0,60936,,,,0,5a18f9d1374fab99872befb4be08e1f43557363e53317fdb03a097f1ecbcfb37
thread:issue:61243,issue #61243: Pytorch 1.5+ is slower than pytorch 1.3,needs reproduction,14,needs reproduction,1,0,61243,,,,0,ec231f88ec5c8606440386c12eeac70e3025608b7af982fbe34673892d40fc65
thread:issue:61341,issue #61341: inerror: reference to ‘DeviceType’ is ambiguous,needs reproduction,14,needs reproduction,1,0,61341,,,,0,ff2fdf0848659d98d344584f17a1a4094cf58731bc655753ede8554b08d05b88
thread:issue:61660,issue #61660: Pytorch model load failure in Gunicorn with Gevent workers,needs reproduction,14,needs reproduction,1,0,61660,,,,0,895d992d40950f4b366067e7525e4e9780285fee3f4565b3c5cbe1df6b5ceeb1
thread:issue:62152,issue #62152: Runtime Error: Expected a 'cuda' device type for generator but found 'cpu' in google colab,needs reproduction,14,needs reproduction,1,0,62152,,,,0,b96f7b3080fdd44c18a0e3f4b971ed5f3c1bb6f0e86b56c6ec6ceeb8f4c66933
thread:issue:62644,issue #62644: waste too much time in first two pictures,needs reproduction,14,needs reproduction,1,0,62644,,,,0,a8ed1ee2a5c18697a06c75e7f206d62ca7aa52b4aca0633e8a763a43b8936a22
thread:issue:63322,issue #63322: lstm's input h0 and c0 bug,needs reproduction,14,needs reproduction,1,0,63322,,,,0,82c3c76b45e5195c7f40acbc10616e545462e805d790cec8c755f4bdf786dba7
thread:issue:63726,issue #63726: a problem happened in torch.randperm,needs reproduction,14,needs reproduction,1,0,63726,,,,0,9c7cb669a5936dcf0368092984c5432a639980fc0aeb3832320fd5691324bf9e
thread:issue:63847,issue #63847: RuntimError:CUDA error: an illegal memory access was encountered when copying data to CUDA,needs reproduction,14,needs reproduction,1,0,63847,,,,0,885ba99cfbddbd17d6ffcb10719b7529514402c0cb2c1e207adf0690a9196180
thread:issue:64143,issue #64143: AttributeError: module 'torchvision.models' has no attribute 'load_model',needs reproduction,14,needs reproduction,1,0,64143,,,,0,71daf5ead99a61aeaaf28d49e9042dac4c9d4d7e046f4d17a442dc0a12c36205
thread:issue:64156,issue #64156: Get zero when using torch.matmul and torch.dot with 1-D tensor in torch-1.9.0-cp39-none-macosx_11_0_arm64.whl,needs reproduction,14,needs reproduction,1,0,64156,,,,0,4463bc3d3b284190d1e5ef9b358755dcfa4642f6faae5ce1c91346d6e56226ef
thread:issue:64560,issue #64560: Private bytes consumption exception,needs reproduction,14,needs reproduction,1,0,64560,,,,0,d8e44a55c4280fa4c3a03c6650764e8ba77c783ad8a3360af6890fa42b217815
thread:issue:64804,issue #64804: RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation,needs reproduction,14,needs reproduction,1,0,64804,,,,0,0723b400f6c225b1b7d128597256a3e4d9c50dd8d07810a9931f63cbb9182030
thread:issue:64889,"issue #64889: ""RuntimeError: CUDA error: out of memory"" for no reason",needs reproduction,14,needs reproduction,1,0,64889,,,,0,0eb348ffa56d9b616283a1934a19b79a3c9737d99107ef136e1dfbdaaa59194d
thread:issue:65027,issue #65027: CUDA out of memory,needs reproduction,14,needs reproduction,1,0,65027,,,,0,cd839fba25b671fec4726ba6a5e99ee2f350535726f5f28b709e59c7fd145a29
thread:issue:66781,issue #66781: lr_scheduler.py / list index out of range,needs reproduction,14,needs reproduction,1,0,66781,,,,0,59d5e2cff4eb5dc864b467e6cd6c10abc2fc1dc588114abae8712350cb6237eb
thread:issue:67740,issue #67740: `torch.utils.data.random_split` example broken in 1.10,needs reproduction,14,needs reproduction,1,0,67740,,,,0,31343672aa90e1d73211ee0fef2ef1559cc90394371466f51124750c57f3252f
thread:issue:67835,issue #67835: It seems to be a bug！！！,needs reproduction,14,needs reproduction,1,0,67835,,,,0,62e8b37d7c0a6c63f8cb643bd83e3f491bc74da7892f0931ce37b1cfc05edd5a
thread:issue:68144,issue #68144: Unstable buggy calculation.,needs reproduction,14,needs reproduction,1,0,68144,,,,0,8ef281697fe29e0b61f18668461008457b07b75df04307a2d2405f21da28c36b
thread:issue:68864,issue #68864: Incorrect accuracy pytorch(1.0.1) and libtorch because of nn.LSTM,needs reproduction,14,needs reproduction,1,0,68864,,,,0,a9f5cf8e7939d35be838d42f5ce2d5e70a477d3937e98ef8179334f8e3b766c9
thread:issue:69203,issue #69203: Error in SVD cusolver on Linux,needs reproduction,14,needs reproduction,1,0,69203,,,,0,0966bbde75a5aa33586d6f52cad6e9e3d2db33aa40f2bee71673a63c5db9952c
thread:issue:69448,issue #69448: RuntimeError: PyTorch convert function for op 'inverse' not implemented.,needs reproduction,14,needs reproduction,1,0,69448,,,,0,841dd645380d4668cb7bef84950e1d95880a8de0673b2dd07a10fec439251f50
thread:issue:69966,issue #69966: torch.nn.DataParallel caused inference failure with cpu set as device on NV machine,needs reproduction,14,needs reproduction,1,0,69966,,,,0,22cd9e46bf1e67f50fa888ab39e14bc4a8bb4846918fffcd22e828cda6576924
thread:issue:70008,issue #70008: Torch function runtime seemingly dependent on scipy call,needs reproduction,14,needs reproduction,1,0,70008,,,,0,06a31e0903baab28482294954b0482a66d7ac222892734415298fd4186ef3bfa
thread:issue:70038,issue #70038: comile error,needs reproduction,14,needs reproduction,1,0,70038,,,,0,cfcb09330cd868e710801c39845fc15d28685d6adafe41c2d8107e8d4998be87
thread:issue:70048,issue #70048: Tensor transfer between gpus doesnt work,needs reproduction,14,needs reproduction,1,0,70048,,,,0,a6cc264b19935a4e38a03d0f95c1e0da2e8a1979fa19c3e5d08be00264dc71f0
thread:issue:70519,issue #70519: RuntimeError: cublas runtime error,needs reproduction,14,needs reproduction,1,0,70519,,,,0,a49ab4a7f9c5f915dee025aa908c4d5df335a27d2541448c194d40959a99d518
thread:issue:70702,issue #70702: conv3d padding=`same` gradgradcheck fails on CUDA,needs reproduction,14,needs reproduction,1,0,70702,,,,0,ac03f2fa780d3edb9b846446558998bd8de00762baa427c7dce6c9ba176caef6
thread:issue:71222,issue #71222: torch.linalg.lstsq is nondeterministic,needs reproduction,14,needs reproduction,1,0,71222,,,,0,3b1d8b14d140d439138e2d97fa0d80f8718c88cff6a3a9ed4734828546915fdb
thread:issue:71272,"issue #71272: UserWarning: Seems like `optimizer.step()` has been overridden after learning rate scheduler initialization. Please, make sure to call `optimizer.step()` before `lr_scheduler.step()`. See more details at https://pytorch.org/docs/stable/optim.html#how-to-adjust-learning-rate warnings.warn(""Seems like `optimizer.step()` has been overridden after learning rate scheduler",needs reproduction,14,needs reproduction,1,0,71272,,,,0,562408dfd71d75c1f89fb81f1eaea5f3c70dbf3fdacc2f73829de8e48e70ddb1
thread:issue:71545,issue #71545: pip installation SSLError [SSL: CERTIFICATE_VERIFY_FAILED],needs reproduction,14,needs reproduction,1,0,71545,,,,0,c8eff61b186c6991d2ec024927723201688198165a4de72e6921acfb18197b2b
thread:issue:71704,issue #71704: M1 Pro Apple Silicon chip support.,needs reproduction,14,needs reproduction,1,0,71704,,,,0,804a7d83a3b2351de5b3cec10f9d69a95a301e7ab380a119a51e49c909d5be80
thread:issue:72105,issue #72105: ImportError: cannot import name '_VF' from partially initialized module 'torch',needs reproduction,14,needs reproduction,1,0,72105,,,,0,8fff86f053d640627327acf54a2ae02fbc0ec0acc2c7edb9f3a68fb355af829d
thread:issue:72591,"issue #72591: Can't forward pass conv2d with kernel_size=1, and padding=1",needs reproduction,14,needs reproduction,1,0,72591,,,,0,2c81ddc24372ab627f65f21c87987913413e1f17b6ba310bff71a0bc50fe5b12
thread:issue:73568,issue #73568: Bug when using `nn.Linear`,needs reproduction,14,needs reproduction,1,0,73568,,,,0,7217ad05d8ad6829ff105f9e927804895514e35729ea2c1566445b46a9c995fa
thread:issue:74670,issue #74670: Can't install Pytorch,needs reproduction,14,needs reproduction,1,0,74670,,,,0,ae5a6a26c01a9ebe114493120ccb29b61e01d074b7c3de90cf8b8e67c130211b
thread:issue:75265,issue #75265: Kernel Dies while using conv2d layer / function,needs reproduction,14,needs reproduction,1,0,75265,,,,0,8db9854f59c0bba513e89f532642b425cd19badbac2480fecb491c4d20ffb47a
thread:issue:75458,issue #75458: Placing model on bfloat16 on CPU make it freeze/hang,needs reproduction,14,needs reproduction,1,0,75458,,,,0,1ebc5971bc1ca6254fb993e1c83fc4f3b4eda5630f453df3f21ccd7305c6adce
thread:issue:75949,issue #75949: Potential memory leak in Adam optimizer in AMD chips (CPU),needs reproduction,14,needs reproduction,1,0,75949,,,,0,76de75b097b367b3f02dd2c7921f19fa7ce2e97f0cb2a792016d2842d39925b1
thread:issue:75963,issue #75963: Add build support for GCC 11.2,needs reproduction,14,needs reproduction,1,0,75963,,,,0,c73fe2721046e830a80ef0378e9bb42aec110702243a774bdf755baa76fe6989
thread:issue:76024,issue #76024: The prediction results of different equipment are inconsistent,needs reproduction,14,needs reproduction,1,0,76024,,,,0,17c4ef2e71a1c3eedfe5c760d3ca1dc164b2d3277dffb68e9734e5f038908eb0
thread:issue:76643,issue #76643: Pytorch can't process special unicode,needs reproduction,14,needs reproduction,1,0,76643,,,,0,43431d828339fe216c89fc5a54cc82477ca733dd5340e027c1c11d789f157798
thread:issue:76798,issue #76798: `gradcheck` for `torch.solve` may trigger INTERNAL ASSERT FAILED,needs reproduction,14,needs reproduction,1,0,76798,,,,0,b022945f2efa3317125217a5e398ea253a0414537d1898d5351074f52ae7b7e5
thread:issue:76954,issue #76954: ReduceSum causes the inconsistency of output between frameworks,needs reproduction,14,needs reproduction,1,0,76954,,,,0,bc7d109f795977ffc44a5942e117e027d90c0c84be7a91d7ef8ebf7bbddd1779
thread:issue:76955,issue #76955: GaussianNoise causes the inconsistency of output between frameworks,needs reproduction,14,needs reproduction,1,0,76955,,,,0,727a4528d706aef5cf1906613a3505ed4d43fb3766e1aaaa6a4aa8cc9ec52185
thread:issue:76956,issue #76956: GlobalAvgPool2d causes the inconsistency of output between frameworks,needs reproduction,14,needs reproduction,1,0,76956,,,,0,d32ccce259004bbe6bb81d2a3ab15084f66912748187c35a2ea9dfa9413b8f6e
thread:issue:76959,"issue #76959: When using Lambda, the output of the 1/x process is very likely to have nan/inf",needs reproduction,14,needs reproduction,1,0,76959,,,,0,73403a6b378eb25e21f21d2e4ed00ba3e4c30d7b7b94039f4531c6c652f216de
thread:issue:76960,"issue #76960: When using Rsqrt, the output of the 1/x process is very likely to have nan/inf",needs reproduction,14,needs reproduction,1,0,76960,,,,0,ce159337d3d3673d8307f05f75c42992611ba07d33f4addb67a5a5b560f890d0
thread:issue:77893,issue #77893: Segmentation fault in _remove_batch_dim,needs reproduction,14,needs reproduction,1,0,77893,,,,0,d7ef3c27bfaa90a45d228fd323de75001a5e3acd31706babe0d8b7ac37c6ea9b
thread:issue:78831,issue #78831: Mac M1 Build Failure on DEBUG=1,needs reproduction,14,needs reproduction,1,0,78831,,,,0,6c0840c1f591c53e23626ee68ee7c4ad4be3bd20d2a6ee5d5d95d3f55919b758
thread:issue:80302,issue #80302: AttributeError: 'LinearPackedParams' object has no attribute '_modules',needs reproduction,14,needs reproduction,1,0,80302,,,,0,8a9d991509a0dc90c4214ae701766577499b8a0178a8b019cba0770f3ae3ddd5
thread:issue:80588,issue #80588: Semi-reproducible random torch.baddbmm NaNs,needs reproduction,14,needs reproduction,1,0,80588,,,,0,de3b33974968dce137f8db7fd3b54abcdf7208d9a59668d3b61a4ac476004e44
thread:issue:81257,issue #81257: High GPU context memory on Torch 1.11.0 but none on Torch 1.10.0,needs reproduction,14,needs reproduction,1,0,81257,,,,0,753b5ae0d4bf88e05a0485a429be2d7f3c71284f593d362033a320e977caa37a
thread:issue:82217,issue #82217: backward not available for index and mask,needs reproduction,14,needs reproduction,1,0,82217,,,,0,375c5260390bc7cece4a808056adbf707622e5507dd1b7e8afe949f35dc21ff5
thread:issue:82218,issue #82218: OOM during backward() leads to memory leaks,needs reproduction,14,needs reproduction,1,0,82218,,,,0,ff9d93208d60c913774f8aa65bf65f1b012af88b9d8947a681601130d86df217
thread:issue:82546,issue #82546: Libtorch C++ torch::stack error,needs reproduction,14,needs reproduction,1,0,82546,,,,0,14a9021700f39721e625e27ad5453df74b8026585972810665d97cd77d0db578
thread:issue:83901,issue #83901: pytorch 1.12.1 Adam Optimizer Malfunction!!!,needs reproduction,14,needs reproduction,1,0,83901,,,,0,56c48742816fc4797f87703f3f6548d68949cec4ed3cc6ba0e971210abc358cd
thread:issue:84370,issue #84370: Installation prefix is not passed to CMake appropriately,needs reproduction,14,needs reproduction,1,0,84370,,,,0,b004f86f61eba89a1c1ad82deb098f30e275b5e728cc4be531d222003e4343ac
thread:issue:84705,issue #84705: Set dtype if tensor converted to numpy,needs reproduction,14,needs reproduction,1,0,84705,,,,0,1c9de278fefd4301952031dd7ea7bd3cb6e82127e11551bbe6f59b53d790ddbc
thread:issue:85211,issue #85211: OSError libstdc++.so.6 at import,needs reproduction,14,needs reproduction,1,0,85211,,,,0,fa35f78ab7812c088ca7c47377c954eeb8c2eab67f8815fbff3bb0318c6ffc32
thread:issue:86547,issue #86547: Segmentation fault (core dumped) in RTX3090,needs reproduction,14,needs reproduction,1,0,86547,,,,0,2b1efd553300fa9fe3f34d6cfee6b0dc4d53a32dce57933f1cbe5d175247ca6a
thread:issue:86684,issue #86684: [ONNX] AssertionError: A mismatch between the number of arguments (5) and their descriptors (4) was found at symbolic function 'scatter',needs reproduction,14,needs reproduction,1,0,86684,,,,0,c9dcde49bbeba4f4cbd9299d71ce316b5ba4d6a6a184185ba42efbd09800005d
thread:issue:86877,issue #86877: compile torch from source,needs reproduction,14,needs reproduction,1,0,86877,,,,0,90308774435807c9ceff6bd14958eb0aa7433f2de7d28735c5af297fd345c6d1
thread:issue:86919,issue #86919: Importing torch 1.12.0 breaks subprocess module,needs reproduction,14,needs reproduction,1,0,86919,,,,0,15867eb8a6c3ec7827b46ed79b83e0e7feeb36a7f6425f685255e3b5844e961e
thread:issue:87550,issue #87550: [onnx] export repeat_interleave TypeError: z_(): incompatible function arguments,needs reproduction,14,needs reproduction,1,0,87550,,,,0,4d467ce50e17fd2ae58e538f1301cf5bfe7a8e1f63518afe3d459ca098a137e0
thread:issue:87597,issue #87597: ninja: build stopped: subcommand failed,needs reproduction,14,needs reproduction,1,0,87597,,,,0,101859bb7e4b149b3f4789fce7249e39f703ab11226c58a54cda62a0aee93ab5
thread:issue:88948,issue #88948: torch.linalg.matrix_rank memory leak,needs reproduction,14,needs reproduction,1,0,88948,,,,0,1eb77084d2ba7ad367c304effe06d820397dd949d25adbc6c7042d9bc2775e46
thread:issue:89041,"issue #89041: cross compile pytoch using cmake , get an error : protobuf::protoc: command not found",needs reproduction,14,needs reproduction,1,0,89041,,,,0,5c56a1c3aa0535592915e25c1aa602a7671775ef77d6f67ac9fc3a9be5efabba
thread:issue:90115,issue #90115: Couldn't install pytorch 2.0,needs reproduction,14,needs reproduction,1,0,90115,,,,0,7ac6ab21334d15cbcf548df1d3054a43547da1901f6ddc29409aea92adc82e42
thread:issue:90138,issue #90138: p.block != nullptr && p.block->ptr != nullptr INTERNAL ASSERT FAILED,needs reproduction,14,needs reproduction,1,0,90138,,,,0,c24a873ced1c932a76d753ea8ba1273a491a7cf13f5362809aa5803b31ec4f41
thread:issue:90521,issue #90521: [A ERROR in Docker] RuntimeError: CUDA error: no kernel image is available for execution on the device,needs reproduction,14,needs reproduction,1,0,90521,,,,0,13e119a078d0e1e5adb24ff0096e857b72ed8f07fe947e8c719d130ad73f0de9
thread:issue:90752,issue #90752: Adam (fused=True) issues,needs reproduction,14,needs reproduction,1,0,90752,,,,0,6eb0d0e0d445f4949a18f8578852594cd0937a7a4bd5297f40396164f388d460
thread:issue:91072,issue #91072: Unexpected behavior when running torch.max in cuda,needs reproduction,14,needs reproduction,1,0,91072,,,,0,0b98f139e5abb7706ba43ffd2d6219737e867b523c7f160432d1e4da0a321916
thread:issue:91245,issue #91245: Segmentation faults in DataLoader (in latest torch version).,needs reproduction,14,needs reproduction,1,0,91245,,,,0,77cdddb9ae9f312259f17829164e7b909edbb776668bc9fbd97879c1a55ecc8e
thread:issue:91942,issue #91942: jacrev over huber function,needs reproduction,14,needs reproduction,1,0,91942,,,,0,0c17b7257e515a1d33150e877d0ecde0aca0069cd2d06c8ccf8363375d546413
thread:issue:91970,issue #91970: Unknown CUDA graph CaptureStatus21852,needs reproduction,14,needs reproduction,1,0,91970,,,,0,d26c3d1518493b022e6f54e10fb7f83ffc964c656cc18f8239cff830d1aace25
thread:issue:92794,issue #92794: Segmentation fault when running torch.ge,needs reproduction,14,needs reproduction,1,0,92794,,,,0,d7072e68bd29cefcf3859eab519ed45005e61246ff0ad6569ddeaad0f23a78ad
thread:issue:93900,"issue #93900: Why does the torch model have no memory leaks under gpu, but there is a memory leak under cpu, torch version 1.10.1",needs reproduction,14,needs reproduction,1,0,93900,,,,0,5d1070b5dd0d4bc896a6923b6f4ef9f80ede3204384b32a501a8e1bbdaa1a650
thread:issue:94542,issue #94542: torch.lgamma CUDA driver error,needs reproduction,14,needs reproduction,1,0,94542,,,,0,817a898b591a2045d06e6917effcd2d55f556fe430c922c03183b65c2321cc49
thread:issue:94855,issue #94855: torch.compile breaks reproducibility,needs reproduction,14,needs reproduction,1,0,94855,,,,0,7549a3340ed25f81bf99d5b3824bc58d29154bf2e34795de3c1b81811fc13e51
thread:issue:95677,issue #95677: Unable to import ``torch.linalg``,needs reproduction,14,needs reproduction,1,0,95677,,,,0,73cb1c794259ca6d11daec171bf91912045dfbe45927176f7f89b1e78f361652
thread:issue:97135,issue #97135: Incorrect gradient calculation for upsample nearest on CUDA,needs reproduction,14,needs reproduction,1,0,97135,,,,0,ddf9bb4415bd110fe6b8a3139b7730a7141a6a0023eeb179ad86dad07003f1a7
thread:issue:97847,issue #97847: CUDA 10.2 cudnn 8.2.4 run Conv2d error,needs reproduction,14,needs reproduction,1,0,97847,,,,0,edd0b7a421a94a189ab9d1b47c541152dc2d6cfd478866fd782200eac69642d9
thread:issue:97865,issue #97865: Unable to install torch on python 3.8.16,needs reproduction,14,needs reproduction,1,0,97865,,,,0,cd4874a1b0121d7053dad98d661832f2f1279733be563f95ec4c692e65e916d3
thread:issue:97872,issue #97872: nn.linear not support bfloat16,needs reproduction,14,needs reproduction,1,0,97872,,,,0,b9713bf4d774bfd2589f75792c700ed4a5d1a246a507d8851f85920b6f6e7da4
thread:issue:98142,issue #98142: double free or corruption (fasttop),needs reproduction,14,needs reproduction,1,0,98142,,,,0,268d9926ce99695ee449f170c42d17a4dfe6e8589f26276fb3f864811fedd090
thread:issue:98204,issue #98204: NaN appears when initializing tensor,needs reproduction,14,needs reproduction,1,0,98204,,,,0,19b906d6088f528e79930854b93f3031d08e3466fc36d37937d5495e5d045832
thread:issue:99268,issue #99268: Embedding layer tensor shape,needs reproduction,14,needs reproduction,1,0,99268,,,,0,eaf998120cb3d7b4313394a69b7e77bf2360f99c552b3cde4af9a4bef3964162
thread:issue:99573,"issue #99573: Jacfwd become slower after update pytorch (""We’ve integrated functorch into PyTorch---Documentation"")",needs reproduction,14,needs reproduction,1,0,99573,,,,0,18dfa9b1a1b90efda563dd61f0fc8bf778fd0fc7372b43942ec5bc4a36e7ae99
thread:pr:176393,pr #176393: Make avgpool3d support fp16/bf16,draft/noise,13,draft/noise,1,0,,176393,,,0,f09b90033e858250346be0f8f3b0027f7f5e12c3bf75f59a0beee79bbfd51682
thread:pr:177649,pr #177649: BLAS sparse ops: dense arguments -- extend support beyond just dense and non-overlapping inputs.,draft/noise,13,draft/noise,1,0,,177649,,,0,d0d3a64272d72050d44e7f02fab820c46e93d03f67f10704ae9a51f6d95fa683
thread:pr:178305,pr #178305: [XPU][oneDNN] Enforce const/mutable pointer semantics for mkldnn xpu detail,draft/noise,13,draft/noise,1,0,,178305,,,0,829524bcafdf8e5b635e85693cf2b99691a8bb48df58bba5429ea3a8be0e6164
thread:pr:179223,pr #179223: [Test] Refactor test_higher_order_ops.py to be device-generic,draft/noise,13,draft/noise,1,0,,179223,,,0,d26ba5beb61b8faf49670eeddecf62cf47c19bd2ed90e4f0bb6ca3d3c53b7bcb
thread:pr:180087,pr #180087: [DTensor] Error on random ops with device mismatch,draft/noise,13,draft/noise,1,0,,180087,,,0,95710dbe6475ca2c5d339a3ef627879217a4ca91fdf967b71ec422ec87e39421
thread:pr:180317,pr #180317: [xpu][inductor] Add Dual GEMM SiLU-Mul fusion pass for XPU,draft/noise,13,draft/noise,1,0,,180317,,,0,33ecb62dce3de130f27fc9aab9283e55eed4bc7640a144ee79d183d60a62901c
thread:pr:180460,pr #180460: Decompose all_gather+matmul into P2P+tiled mm for fine-grained overlap,draft/noise,13,draft/noise,1,0,,180460,,,0,45f36a3d0e3c79605561d3692b7bad283479c25b511900bcade878db4723606e
thread:pr:180573,pr #180573: Avoid unnecessary refcount bump in inplace op Python bindings,draft/noise,13,draft/noise,1,0,,180573,,,0,53368983b1a9f0b82188f587888a13a208f57bbb2b1520833eb9dd81146d9690
thread:pr:181145,pr #181145: [CuteDSL][Native Ops] group norm kernel,draft/noise,13,draft/noise,1,0,,181145,,,0,55c5d6564d43ccd033b162aded3edb1e68fb805975335dddca7f19e53427a3b7
thread:pr:181187,pr #181187: [spmd types] fully_shard but in DTensor land,draft/noise,13,draft/noise,1,0,,181187,,,0,7f3048f71136d905045cd82468d53a49a618946e2399c936caf3d5572b975473
thread:pr:181528,pr #181528: sdp: warn when enable_gqa=True silently falls back to MATH,draft/noise,13,draft/noise,1,0,,181528,,,0,0599e7f3316d56667d5ea02a0be99578b9077be96d577c8cf143f96662ede6c8
thread:pr:181557,pr #181557: [XPU] Enable grouped_mm and scaled_grouped_mm with OneDNN,draft/noise,13,draft/noise,1,0,,181557,,,0,02b1a9e3356a803da4603596331159852dbe7de97970a762caf03556c0275a9a
thread:pr:181659,pr #181659: [FSDP2] Add recurse= to set_symm_mem_for_comm and set_force_sum_reduction_for_comms,draft/noise,13,draft/noise,1,0,,181659,,,0,eb9bb74f88e4dc7370b125626b5ba887808e18615057b68e3dbaba062a1a8dbe
thread:pr:182838,"pr #182838: [CI] Fix docker builds: pin miniforge, use --extra-index-url",draft/noise,13,draft/noise,1,0,,182838,,,0,ca8a172459ae5605fede31a6aea44919c71628a98dd967663369b41d24f8b855
thread:pr:182841,pr #182841: [CI] No-op docker change to test current build state,draft/noise,13,draft/noise,1,0,,182841,,,0,b1bd79f977a3440915d145723ebfd831a65e9dab8dc925bf4ffde574da674cca
thread:pr:183041,pr #183041: [FSDP1] Test forward overlap with use_orig_params,draft/noise,13,draft/noise,1,0,,183041,,,0,3048082e3abc27609ab833f8d3b2296ad9078433b791b3603fb048c091ff8d56
thread:pr:185178,pr #185178: [FSDP][AC] Bypass SAC tracking for all-gather internals,draft/noise,13,draft/noise,1,0,,185178,,,0,ad5e10e3d5e1a988c1a3a8a2f3f549af75b800d7f1285efd0a23ca263a9d0c49
thread:pr:185217,pr #185217: Fix Debug/RelWithDebInfo PyTorch+XPU on Windows,draft/noise,13,draft/noise,1,0,,185217,,,0,8ca145cfaf6b073e798a0a8ad4868883fc6fe2e247c68605cbff6e955561bdb3
thread:pr:185816,pr #185816: [WIP] [PoC] Enable CPU FlexAttention on AArch64,draft/noise,13,draft/noise,1,0,,185816,,,1,60907e77fffd4870c9888be08b381a0e888aa91225b48250c4e597c79e854ec1
thread:pr:186302,pr #186302: Install Rust toolchain in PyTorch CI/CD build images (#186302),draft/noise,13,draft/noise,1,0,,186302,,,0,8fe3c94356e858560266f9855a89eeeabf42df7e9ea3d3d53d3044be41bf4519
thread:pr:186851,pr #186851: [profiler][cupti] ProfilerObserver + record_function annotation routing,draft/noise,13,draft/noise,1,0,,186851,,,0,081afd046e28b1d39ce280efc0489a378aff3c2acbd98376886864f5bf0828ed
thread:pr:186852,pr #186852: [profiler][cupti] ProfilerObserver + record_function annotation routing,draft/noise,13,draft/noise,1,0,,186852,,,0,8ce575be6fd1f99f0424c32c917fd083a049b040c4a7e14db7e9b533dcaf605d
thread:pr:186980,pr #186980: Move header-only c10/core headers to torch/headeronly,draft/noise,13,draft/noise,1,0,,186980,,,0,2bf5e7fade4e167e29f64d75c364f0358aacd9d32aeaf51bccc48b403e84e2a3
thread:pr:186981,pr #186981: Move header-only c10/util metaprogramming/type utils to torch/headeronly,draft/noise,13,draft/noise,1,0,,186981,,,0,38071836a419fc3f9f5d5cceb9562d92788f860eab7a8cdbdf5f1e6ebfbc3185
thread:pr:186982,pr #186982: Move header-only c10/util concurrency/lifetime utils to torch/headeronly,draft/noise,13,draft/noise,1,0,,186982,,,0,fb4723c6195419f2919e74b6d5ed26560acb1a416e147be9d09840d80ed6f726
thread:pr:186983,pr #186983: Move header-only c10/util math utils to torch/headeronly,draft/noise,13,draft/noise,1,0,,186983,,,0,2e44af9d613cb213deb5a7377ab0055cf363ba0288fc495ca2a984c51deb7e62
thread:pr:186985,pr #186985: Move header-only c10/util low-level utils to torch/headeronly,draft/noise,13,draft/noise,1,0,,186985,,,0,d434e71d765317aa55dcb06850e4a7d47b74f19d478eb2a76fe923bd7e6bfd5d
thread:pr:187213,pr #187213: Issue-106571: Lint Fixes - Rule B007 - gb_registry_linter - Replace …,draft/noise,13,draft/noise,1,0,,187213,,,0,f319578f5441ca2c9718926a097efa0cdf1687b7d0a43f815daa049d88e36c5d
thread:pr:187540,"pr #187540: [DRAFT] Fix forward AD inconsistency for pow(abs(complex), complex)",draft/noise,13,draft/noise,1,0,,187540,,,0,e101e53727b2d4c3ba7e8018f99298e559c9f1b656e7a5b156ddfb767d1baee8
thread:pr:187564,"pr #187564: Register `linalg_qr`, `geqrf`, and `linalg_householder_product` for XPU",draft/noise,13,draft/noise,1,0,,187564,,,0,e35170716b902e134070e32f99367d20e3804560536a346e5e9ed2d6b3922dcc
thread:pr:187612,pr #187612: docker-builds: move to default runner group (regular OSDC pool),draft/noise,13,draft/noise,1,0,,187612,,,0,def44e281488f08cab985ad827724c1a00ac240f920c8e161b9fe8c417d9a6f8
thread:pr:187620,pr #187620: [FSDP2] Add PartialOffloadPolicy for fractional CPU parameter offload,draft/noise,13,draft/noise,1,0,,187620,,,0,d034edb649ac683f7ac82ed2123a5ac591326ded79e401a1edd8561ebfe11e7e
thread:pr:187647,pr #187647: Remove XPU skips on test_partitioner_unbacked_symints and test_sdpa_s…,draft/noise,13,draft/noise,1,0,,187647,,,0,7f8327b5280a86d03cc22327b9c7a88b2c87da337581fa373788f65a6075e1c3
thread:pr:187685,pr #187685: Name AOTAutograd backward gradient outputs grad_<primal> (#110700),draft/noise,13,draft/noise,1,0,,187685,,,0,43d23ebd13071f75fda303da207864d3c74034c3e3c9526a87a92e6098d0e2fc
thread:pr:187835,pr #187835: [profiler] Enable OVERHEAD activity type for the XPU backend,draft/noise,13,draft/noise,1,0,,187835,,,0,cf597ca7edbca82d0530d11b98999b2de0560143e41a9e1c4bea41d8d3e06161
thread:pr:188065,pr #188065: [AOTI] measure performance with AOTI_RUNTIME_CHECK_INPUTS=1,draft/noise,13,draft/noise,1,0,,188065,,,0,48e3770de5c34144d8308c2527baf619ecb1dd66ed5d919d1faf1c738dffb7af
thread:pr:188120,pr #188120: [FSDP2] Per-FQN offload policy map and cpu_offload_by_budget helper,draft/noise,13,draft/noise,1,0,,188120,,,0,b63109dbb0d5101df80ff39bc0211d6cc694687be643b201583c1e28f024966a
thread:pr:188195,pr #188195: Adding OpenReg allocator stats tests,draft/noise,13,draft/noise,1,0,,188195,,,0,bfb1be8986af88c1addc45a4308317926c413b48b75b66f716c6c471660f52da
thread:pr:188257,pr #188257: [RFC] S3-over-RDMA checkpoint storage backends via NVIDIA cuObject,draft/noise,13,draft/noise,1,0,,188257,,,0,1716e490e89558bd80ffa08d6f802311f04b826fa47745f8062aa2411548c5af
thread:pr:188388,pr #188388: [WIP][XPU][Test]Port 12 UT test suites to Intel GPU,draft/noise,13,draft/noise,1,0,,188388,,,0,e7f3825f548e9b7312497549d4f6aa563de3bf599eee60f987f763d56dbc22c0
thread:pr:188447,pr #188447: Add DTensor shard propagation skill,draft/noise,13,draft/noise,1,0,,188447,,,0,df72417f2f44bcbe3020114012ef67df9afc492095c62c04f1e588f02cbfd934
thread:pr:188565,pr #188565: [ROCm] Add initial GFX1250 architecture support,draft/noise,13,draft/noise,1,0,,188565,,,0,f1822ae7c0dff389aba45f0e4b196b53b635323cafd2143ad5b6b6793705e8eb
thread:pr:188593,"pr #188593: [ROCm] Add skipIfRocmVersionAtLeast([7, 14]) skips for ROCm 7.14 known failures",draft/noise,13,draft/noise,1,0,,188593,,,0,24815bc35ca8d3403e466226cd18bf88a887ffd5faed41719f9fc95a5b580d68
thread:pr:188653,pr #188653: Add test decorator for device-type-specific RNG seed setting.,draft/noise,13,draft/noise,1,0,,188653,,,1,6dea2f75e242355ad5dc333a1a13c92988539970a671523de4edaad77b41a20b
thread:pr:188682,pr #188682: Limit ROCm build targets to architectures covered by testing,draft/noise,13,draft/noise,1,0,,188682,,,0,d3542b5ebf289bd4242f3d6e01a6497a5e2f0816b1af6ab28c9748eadd73b969
thread:pr:188781,pr #188781: [CI] Benchmark TorchInductor perf against 2.12.1 release (baseline),draft/noise,13,draft/noise,1,0,,188781,,,0,5c75ba034ebba1b1d5b9bf1b919cdf302e29791c550f4fa234d0165a3f77744f
thread:pr:188782,pr #188782: [CI] Benchmark TorchInductor perf against 2.13.0 RC,draft/noise,13,draft/noise,1,0,,188782,,,0,bf1a72b9a0c42973fa477311c937cf3082018fb9d150eeee245dc549253e37e4
thread:pr:188888,pr #188888: Add Reusable Event support to XPUEvent,draft/noise,13,draft/noise,1,0,,188888,,,0,d6ac1787369f8920955caa5363bd4426d756b751f7d8c704cd56741b2a6b065c
thread:pr:188918,pr #188918: [dynamo] Model list subclass __new__/__init__ arg handling to match CPython,draft/noise,13,draft/noise,1,0,,188918,,,0,7a488b1f1362412f1195cf54bd28fa60361d1e90f109d19325644483fb93dcc4
thread:pr:189020,pr #189020: Reorder condition to handle empty tensor the same for file_system & file_descriptor,draft/noise,13,draft/noise,1,0,,189020,,,0,d4f90cfec61fc2dae8787c7eceb63a47d4af6137dc68fd75a32c0f815f46961e
thread:pr:189081,pr #189081: Add tp_name property to Variable Tracker,draft/noise,13,draft/noise,1,0,,189081,,,0,a6624771e2f23c8c4c3d48a8d7bbbb3f9b7a855930e855316e34687761740368
thread:pr:189099,pr #189099: [Profiler][NOT FOR LAND] CI test debugging,draft/noise,13,draft/noise,1,0,,189099,,,0,e9c9fab47772592736d5b356d162cb22005532bd9e55f31b90afa531b7d9fa34
thread:pr:189238,pr #189238: [XPU] Register linalg._powsum and _foreach_powsum.Scalar for XPU,draft/noise,13,draft/noise,1,0,,189238,,,0,76623cc722edd6ccf23be2453d92c1c63966b4145ce1af7957db54494f4e4495
thread:pr:189258,pr #189258: [inductor] Replace ExternKernel/ComputedBuffer plumbing tuples with dataclasses + Protocol,draft/noise,13,draft/noise,1,0,,189258,,,0,07a7e0f50f7cad5afa290c7b96b7b591a5ccca4964490e2b721298ce67fb6d8c
thread:pr:189290,pr #189290: Harden Inductor SubprocPool against compile-worker death and teardown hangs,draft/noise,13,draft/noise,1,0,,189290,,,0,7df57933b297906c2bc16d4a9be7992865bd965c135bd80f7ff6235338281f0b
thread:pr:189304,pr #189304: Use freeze_rng_state in distribution rsample gradient tests,draft/noise,13,draft/noise,1,0,,189304,,,0,4bd7e6d374145b6dad3fa694168cdf608514b94cb13ceabda22600e2a1d2c746
thread:issue:104193:issue:105319:issue:105329:issue:109946,pr #84843: Support different NSE in batches of CSR and CSC tensors,draft/noise,5,Review active PRs and linked issues as one maintainer work thread.,5,0,"104193, 105319, 105329, 109946",84843,,,4,8ca583710f63e5749157c1a8d9bb2d3d044fe9f365a2ada5cccb87ba34153d9d
thread:issue:143658:pr:139524,pr #139524: Make anomaly mode asynchronous,draft/noise,5,Review active PRs and linked issues as one maintainer work thread.,2,0,143658,139524,,,1,1494351481c253c0fe5dd9d5e1dce27b9f7526951036ae55c7bbfe128f7b94fb
thread:pr:125270,pr #125270: [do not review] Add tensor equality subsystem and get it working with compile,draft/noise,5,draft/noise,1,0,,125270,,,0,6765cd1bf23c254c21674fde80e792296ce3ad7d3b0aa224fe5881469b5086b4
thread:pr:125326,pr #125326: [do not review] Use union find and new Python NestedInt with NJT,draft/noise,5,draft/noise,1,0,,125326,,,0,05cc516e467e8f60017b7c403bcc4835761d665c22ffd5b90b24d6d5f8dd15ff
thread:pr:125428,pr #125428: [do not review] support max_seqlen,draft/noise,5,draft/noise,1,0,,125428,,,0,205e611e3f02bd8fc182481b4c56d14556efd991920973feb803a8c1953c455e
thread:pr:126199,pr #126199: [POC] use a better default for num_threads,draft/noise,5,draft/noise,1,0,,126199,,,0,3f973252ec384a01fe6a2b38fbfab710d8a7ec200b4be37aae1fc1c2cb17c47a
thread:pr:142295,pr #142295: [inductor] Decide cooperative RSPLIT with same algorithm as split reductions,draft/noise,5,draft/noise,1,0,,142295,,,0,4f48159056a72d2c1397c3446d29b58e024a52cfc51a28fa0795dab056312c94
thread:pr:143812,pr #143812: [inductor] Used fixed configs for contiguous reductions,draft/noise,5,draft/noise,1,0,,143812,,,0,bffc1f590105c1b737f40bd195e60d2420ca27e264340a510c97cdfafa521635
thread:pr:146593,pr #146593: [NOT FOR LANDING] experimental NVSHMEM integration,draft/noise,5,draft/noise,1,0,,146593,,,0,54a3b987057617766f73b3495b699ce1d6bd8c18855c991503a209be5b6bf8e6
thread:pr:147653,pr #147653: Bf16 fused adam(W),draft/noise,5,draft/noise,1,0,,147653,,,0,488dd85b10d366a8c5a4ff8ceee7986e4ab5e44ef9f4afc177221d29b2e50407
thread:pr:153317,pr #153317: [associative_scan] Autograd for additional inputs,draft/noise,5,draft/noise,1,0,,153317,,,0,87e32be618ef74f397e4c8d5f42bf82ef5b47b1765470876ac2b51ed4385452c
thread:pr:177636,pr #177636: Refactor platform support utilities,draft/noise,5,draft/noise,1,0,,177636,,,0,8d12bef06a26c1a49345078a438d749945e8a0b4afcc154e53fc76ff447b72a4
thread:pr:178850,pr #178850: [xpu][feature] Add fork-safe is_available() via PYTORCH_ZES_BASED_XPU_CHECK env var,draft/noise,5,draft/noise,1,0,,178850,,,0,71cede526618bfa2dfeb016351049e9c3064bda1dee30ac8eebf2cf6488894c7
thread:pr:180888,pr #180888: Update requirements-docs.txt,draft/noise,5,draft/noise,1,0,,180888,,,0,413343e5cb5c43caa683a6586cf8337f421b35f5c35f0eefff731aa2e347fddf
thread:pr:182811,pr #182811: Reenable test_addcmul_alpha_one_fma_parity dtypes F32 and F64 on XPU.,draft/noise,5,draft/noise,1,0,,182811,,,0,f20ef9f5ce787326f85a5d6f08b79bc2fa74aea5a9a09a68d1ecbbf8126559d6
thread:pr:182850,pr #182850: [inductor][symm_mem] Add low-contention allgather variants,draft/noise,5,draft/noise,1,0,,182850,,,0,f5c6141d9226d7fc1919cc8b58469395957a09070ccd3473bd1979ccebd1163c
thread:pr:183211,pr #183211: [FSDP] Break cross-language ref cycle keeping FlatParameter alive,draft/noise,5,draft/noise,1,0,,183211,,,0,083ca68e284f843a468421586ddb34d571a043eff945094f7615bd36a1c5cc81
thread:pr:183253,pr #183253: [BE]: Improve torch utils checkpoint typing,draft/noise,5,draft/noise,1,0,,183253,,,0,e245d8faa2066bf640151f14fbda3652b32f432dedf0f0c9b8efeaab1a8df097
thread:pr:183354,pr #183354: [CI] Match MI355 ROCm shards with trunk,draft/noise,5,draft/noise,1,0,,183354,,,0,af33e521be86b755eb8b03fce25ea380806ae87a9e1d2b1bb11a4bc0e1c69f20
thread:pr:183359,pr #183359: Prototype lazy autograd.Function backward compile,draft/noise,5,draft/noise,1,0,,183359,,,0,ccd0e7fed462d887e227b614095d8fc6296d8906759255eb786666467d8ae0b6
thread:pr:183360,pr #183360: Prototype lazy activation checkpoint compile,draft/noise,5,draft/noise,1,0,,183360,,,0,09e6f7b4e8bd37e6314f92800df2d30658fa73afd15c8b44f71acfed2e97dfd6
thread:pr:183594,pr #183594: TMA scatter_reduce min/max for sm_90+ (CUDA 12.8+),draft/noise,5,draft/noise,1,0,,183594,,,0,d158b90a33e11e1f3345ce91fcd0bb4fd301de612c460e824bb5254065a6e15a
thread:pr:183903,pr #183903: [XPU] Relax Muon float32 tolerance in test_state_dict_with_cuda_params,draft/noise,5,draft/noise,1,0,,183903,,,0,86e72aaf2687007afe2847d01d0af0b43818cbd98452ef9fa7052c95fb275a57
thread:pr:184318,pr #184318: [opaque obj] Check opaque object schema types at dispatch,draft/noise,5,draft/noise,1,0,,184318,,,0,eaeff35bcaed7486962e256c952b574c77df4f20c5bb09b38e298d7b8d6ecc90
thread:pr:184498,pr #184498: [profiler] Propagate record_funtion_id from TorchOp ancestor to Kinet…,draft/noise,5,draft/noise,1,0,,184498,,,0,9233fa1cf9dafc5e7f0df93c47f467efc421bd118d9d616fab160b013a6b73af
thread:pr:184724,pr #184724: [XPU][SDPA] Add XPU flash_attention_forward dispatch with dropout fallback,draft/noise,5,draft/noise,1,0,,184724,,,0,ab4135c31878a1dacfb61906d72956044797686c40dd557df687b1a4f4d41be9
thread:pr:184777,pr #184777: Refactor test_dlpack for out of tree backend,draft/noise,5,draft/noise,1,0,,184777,,,0,06f2bfbf3adc9e851d11440af658b4f098d0c64319b3741479ac27109d3bfcf5
thread:pr:184929,pr #184929: [inductor] Sink single-use dropout randoms into fused epilogues,draft/noise,5,draft/noise,1,0,,184929,,,0,44c7cad05580cdbe94e67ebc8048bf6355180f736ca448f8debcae543ef8e988
thread:pr:184931,pr #184931: [inductor] Use persistent reduction for R=2048 online softmax,draft/noise,5,draft/noise,1,0,,184931,,,0,d045ab6d9a533579530e2256b1cd88a16369413d23ec446fdce3cafa9f4f2e20
thread:pr:184932,pr #184932: [inductor] Use non-online softmax for huge inner reductions,draft/noise,5,draft/noise,1,0,,184932,,,0,09bfb61de25f337271be8c453f05fe13dc77d0f3e225b0c0707f601cf87f3f24
thread:pr:184935,pr #184935: [inductor] Canonicalize duplicate mean and var reductions,draft/noise,5,draft/noise,1,0,,184935,,,0,3b110b025835e1a2fb082d6a9ebd5501efcdb2439f852e5d7f546d3c92971dcf
thread:pr:184936,pr #184936: [inductor] Avoid masks for zero-stride cat source loads,draft/noise,5,draft/noise,1,0,,184936,,,0,dd86974cb2f307e0425596ff76d289baa8adde021a2dbbedfe86f2d5e3db6879
thread:pr:184940,pr #184940: [inductor] Enable reduction+transpose fusion via loop reordering,draft/noise,5,draft/noise,1,0,,184940,,,0,e6281d9a8059360296d0f52fe4e3a57956afd9fd5ea1616776897cdb24ad031c
thread:pr:184941,pr #184941: [inductor] Enable reduction epilogue fusion by deferring split decision,draft/noise,5,draft/noise,1,0,,184941,,,0,c50421fbc9fd7cf305b84fb4ce8e1092367bf5e8384ac2b332f7c132c8faf246
thread:pr:184998,pr #184998: [inductor] Use one warp for tiny inner persistent reductions,draft/noise,5,draft/noise,1,0,,184998,,,0,882ddfe3ccc09167a3543d67b0664cf28be6ce31dc6277125a000d508601f9ff
thread:pr:185003,pr #185003: [inductor] Tune small multi-output mix-order reductions,draft/noise,5,draft/noise,1,0,,185003,,,0,94b4e38299d7c897bb1549f07c02d4c016141b390a0f30a477e0939661028b1e
thread:pr:185027,pr #185027: [inductor] Fold CE backward one-hot row sums,draft/noise,5,draft/noise,1,0,,185027,,,0,5b2f08dd71634e162c43312bd0bed011c2894b909d1a0faca81195b356da8aa5
thread:pr:185028,pr #185028: [inductor] Reassociate cat under surviving reductions,draft/noise,5,draft/noise,1,0,,185028,,,0,15ace0619275c576545603fee422b70b6cfd04a89b5cdbff2d7db84b5918be16
thread:pr:185030,pr #185030: [inductor] Fold masked invariant expand reductions,draft/noise,5,draft/noise,1,0,,185030,,,0,50c359715213b197412afe603d8335a40b9adef1247ba33113504635f5dcbd90
thread:pr:185032,pr #185032: [inductor] Reuse expensive broadcast factors in pointwise tiling,draft/noise,5,draft/noise,1,0,,185032,,,0,76348e7e68e25c19ccdb87c6563fc13c399353ca5ead42e0ff4a395d5dff6430
thread:pr:185036,pr #185036: [inductor] Prototype channel-resident reduction suffix,draft/noise,5,draft/noise,1,0,,185036,,,0,07a4ea497d79da1ab20d241ba1e7661e0adffd838ee731d2578861d617dd576f
thread:pr:185037,pr #185037: [inductor] Use persistent path for small fused online softmax,draft/noise,5,draft/noise,1,0,,185037,,,0,24c18677e6b520c15a50711c6b0acf5857422007a0087dd627d96cfc9463eda1
thread:pr:84545,pr #84545: Add nvfuser support for prims.copy_to,draft/noise,5,draft/noise,1,0,,84545,,,0,985c2c1c958dabe54c7fa188e68e94f4437b2723b0ae4a4fdba3cdb6d926ab7f
thread:pr:85699,pr #85699: Add vector-Jacobian-products for a subset of nvFuser-supported prims; add backward support for nvprims,draft/noise,5,draft/noise,1,0,,85699,,,0,2b15cad054972e3b15a8d8e7d13ab76f97f4e4709ce1e164b6e44f9f8b35c6a0
thread:pr:88106,pr #88106: WIP: feat: LARS optimizer,draft/noise,5,draft/noise,1,0,,88106,,,0,5d96daa82e656992adfc98d0e09fd4a9656b40c6e0dc834fb19fc135ff5e15dc
thread:issue:62540:issue:71806:issue:71810:issue:71812,issue #62540: [Meta] Change default branch name to `main` for repos in `pytorch` project,stale/low urgency,2,stale/low urgency,33,0,"62540, 71806, 71810, 71812, 71813, 71814, 71815, 71816, 71817, 71818, 71819, 71820, 71821, 71822, 71823, 71824, 71825, 71826, 71828, 71830, 71831, 71832, 71833, 71834, 71835, 71836, 71837, 71838, 71839, 71840, 71841, 71842, 71843",,,,64,573fd76411e0ee69131ce0b9c67fc04f5bc306eed68c0d0657c05a918e8845be
thread:issue:168835:issue:168836:issue:168837:issue:168838,issue #168835: TestModule: test/test_sparse_semi_structured.py::TestSparseSemiStructuredTraining,stale/low urgency,2,stale/low urgency,11,0,"168835, 168836, 168837, 168838, 168839, 168840, 168841, 168842, 168843, 168844, 168845",,,,10,01c09127c0d9be4157c79493b15e7b5b5abf522fe969ae284d3bd5315b8cc2d3
thread:issue:168424:issue:168425:issue:168426:issue:168427,issue #168424: TestModule: test/distributed/_shard/sharded_tensor/test_sharded_tensor.py::TestShardedTensorFromLocalShards,stale/low urgency,2,stale/low urgency,8,0,"168424, 168425, 168426, 168427, 168428, 168429, 168430, 168431",,,,7,765ee7daef24c599dcf023a78adc4debe791f8c808c1e24965d89b342ae7538d
thread:issue:168453:issue:168454:issue:168455:issue:168456,issue #168453: TestModule: test/distributed/test_c10d_gloo.py::ProcessGroupGlooTest,stale/low urgency,2,stale/low urgency,8,0,"168453, 168454, 168455, 168456, 168457, 168458, 168459, 168460",,,,7,2c3e48bd8317b98a20cf18c23196bc03f778b546a39769529d1d75d30b940ce3
thread:issue:168522:issue:168523:issue:168524:issue:168525,issue #168522: TestModule: test/distributed/test_symmetric_memory.py::AsyncTPTest,stale/low urgency,2,stale/low urgency,8,0,"168522, 168523, 168524, 168525, 168526, 168527, 168528, 168529",,,,7,ec308299d19d2269e21ee366a00beb8d2f9987446b81339434b0d3da2abe641c
thread:issue:69687:issue:72345:issue:72347:issue:72358,issue #69687: Performance improvement in Autograd Forward AD using ZeroTensors,stale/low urgency,2,stale/low urgency,8,0,"69687, 72345, 72347, 72358, 72359, 72361, 72362, 75586",,,,7,82af8148d2f927cbab1e822eef3cdb262e8a7abb09f6e86de88039ea264a8fd5
thread:issue:103580:issue:147628:issue:162918:issue:52743,issue #103580: Support ByteTensor and ShortTensor for nn.Embedding and nn.EmbeddingBag,stale/low urgency,2,stale/low urgency,6,0,"52743, 61819, 81413, 103580, 147628, 162918",,,,6,c34b4e4fa2fa4479bb7f659bd9ce203046e1dbf80e553f7ce1758dae0befe230
thread:issue:138696:issue:141764:issue:148335:issue:148336,issue #138696: [CI/CD] Deprecating PyTorch’s official Anaconda channel,stale/low urgency,2,stale/low urgency,6,0,"138696, 141764, 148335, 148336, 148338, 148339",,,,6,f1a25b3474b07ddbb0174c6281a6efabefcec9a2686e0838714bb6491b95c5d7
thread:issue:168416:issue:168417:issue:168418:issue:168419,issue #168416: TestModule: test/distributed/_shard/sharded_tensor/test_sharded_tensor.py::TestShardedTensorEnumerable,stale/low urgency,2,stale/low urgency,6,0,"168416, 168417, 168418, 168419, 168420, 168421",,,,5,b8b8189f0a52054dd536b7615f960c0a3053c2b289a3892b4d69a11d35d83cee
thread:issue:168503:issue:168515:issue:168516:issue:168517,issue #168503: TestModule: test/distributed/test_nvshmem_triton.py::NVSHMEMTritonTest,stale/low urgency,2,stale/low urgency,6,0,"168503, 168515, 168516, 168517, 168518, 168519",,,,5,68ad9980e97c29b4db19b209d15b11829d630f35c3ba9616cca0095a442a8aef
thread:issue:44316:issue:45565:issue:53678:issue:53937,issue #44316: [FX] Schema normalization tooling,stale/low urgency,2,stale/low urgency,6,0,"44316, 45565, 53678, 53937, 55095, 55104",,,,5,a48e6877a48a4fade58eb553f742fd1921f80b042ffbd166e0f5925b84fe857d
thread:issue:103761:issue:103787:issue:103798:issue:103800,issue #103761: Merge type stubs for `torch.nn.functional`,stale/low urgency,2,stale/low urgency,5,0,"103761, 103787, 103798, 103800, 103803",,,,5,4c85e1fb32601a2d96974bbf4779b4ea988b41f09846362e431042336841ca2b
thread:issue:107112:issue:30387:issue:34646:issue:34649,issue #107112: from_blob python api,stale/low urgency,2,stale/low urgency,5,0,"30387, 34646, 34649, 45901, 107112",,,,4,a796ee79793385d1e12c351ecde3d79739e855813c33dfafe68928f3a4a9b054
thread:issue:116175:issue:118972:issue:121193:issue:122660,issue #116175: Performance for `TransformerEncoderLayer` and `TransformerDecoderLayer` drops severely if pytorch version changed from `pt1.10` to `pt2.0`,stale/low urgency,2,stale/low urgency,5,0,"94428, 116175, 118972, 121193, 122660",,,,4,b443253580a026a64dd20b1bec6bbc1588ec126de919e7a31d15017620d6c58b
thread:issue:121596:issue:123157:issue:26950:issue:52439,issue #121596: Memory Usage of Conv2d Rapidly Increases when Params >1G,stale/low urgency,2,stale/low urgency,5,0,"26950, 52439, 94869, 121596, 123157",,,,4,d51cbab855a5aafae4e17aa8cf7cd08bca14d8d29d51f98bdb6af999eca7f4e3
thread:issue:168411:issue:168412:issue:168413:issue:168414,issue #168411: TestModule: test/distributed/_shard/sharded_tensor/test_sharded_tensor.py::TestShardedTensorChunked,stale/low urgency,2,stale/low urgency,5,0,"168411, 168412, 168413, 168414, 168415",,,,4,e354aae26eeb70262190362377fde73afb240cd096edf1d78389434907f9ff7d
thread:issue:168773:issue:168774:issue:168776:issue:168777,issue #168773: TestModule: test/test_matmul_cuda.py::TestMatmulCuda,stale/low urgency,2,stale/low urgency,5,0,"168773, 168774, 168776, 168777, 168778",,,,4,64bf6c25df2833c3dd56eb091a4d615cc952233c2fc5acf081835b974d966be8
thread:issue:168784:issue:168788:issue:168789:issue:168791,issue #168784: TestModule: test/test_nestedtensor.py::TestNestedTensorSubclass,stale/low urgency,2,stale/low urgency,5,0,"168784, 168788, 168789, 168791, 168792",,,,4,4061d79eef80d5904e1df7d2966321b4ec9a812029b5ddd17d030f24998b0228
thread:issue:168846:issue:168847:issue:168848:issue:168849,issue #168846: TestModule: test/test_sparse_semi_structured.py::TestSparseSemiStructuredCUTLASS,stale/low urgency,2,stale/low urgency,5,0,"168846, 168847, 168848, 168849, 168850",,,,4,9e4fba2a3039e5db46b6622747cf922b5585d3d80774327f2483f7404524e7ba
thread:issue:23110:issue:26759:issue:27099:issue:28245,issue #23110: [RFC] RPC Based Distributed Model Parallel,stale/low urgency,2,stale/low urgency,5,0,"23110, 26759, 27099, 28245, 37529",,,,5,38fd3f9415ef364df1e9c01772f34ebf8043c7bba18c8c186a948ec18be37dc5
thread:issue:44989:issue:44991:issue:45009:issue:45851,"issue #44989: Support unfold for integral types (long, byte etc) tensors",stale/low urgency,2,stale/low urgency,5,0,"44989, 44991, 45009, 45851, 60466",,,,4,5fa2fbe83332f5653f35f8e40922eb27b4637c60ed71ace10feaa2f0ddbd36f8
thread:issue:64327:issue:75242:issue:79967:issue:87033,issue #64327: RFC: multi-part `torch.load`/`torch.save` to support huge models and/or low CPU memory,stale/low urgency,2,stale/low urgency,5,0,"64327, 75242, 79967, 87033, 97196",,,,4,adfdfb6988999591b91947d9155c39bcf65f529e8696d674699f3f3bd5248c3e
thread:issue:10007:issue:10254:issue:9853:issue:9854,issue #10007: Upgrade Caffe2 Hypothesis,stale/low urgency,2,stale/low urgency,4,0,"9853, 9854, 10007, 10254",,,,4,164b1f8938c7600ce8d4a99c214d2f0b637f70d5088b784f19c15c867112b334
thread:issue:10386:issue:49440:issue:53982:issue:64881,issue #10386: [feature request] batch_first option in torch.utils.data,stale/low urgency,2,stale/low urgency,4,0,"10386, 49440, 53982, 64881",,,,3,3023661da2a336c4a131ee683018d5659cd76ef330af8a1dbc24e63c50e187c4
thread:issue:114153:issue:132041:issue:132045:issue:132047,issue #114153: [Tracker] Inconsistencies between CPU and GPU computation,stale/low urgency,2,stale/low urgency,4,0,"114153, 132041, 132045, 132047",,,,3,701a1776c9738e8a7b461176b0d62a4c078bebd2641d53b6ef9452052ecdfcba
thread:issue:115821:issue:5405:issue:56975:issue:5740,issue #115821: [BUG] missing automatic dtype promotion for out-of-place `torch.scatter_*()` functions,stale/low urgency,2,stale/low urgency,4,0,"5405, 5740, 56975, 115821",,,,3,6f5e8980f3a464f5a6bc7729f5eec4624060fff5bac681a75d6bfe327b535106
thread:issue:117394:issue:120814:issue:133230:issue:138217,issue #117394: [RFC] Dynamo Single Step Graph,stale/low urgency,2,stale/low urgency,4,0,"117394, 120814, 133230, 138217",,,,3,9fa8a917fbe957d0ff757623b00028cddaf8f68caab27d0151ee4477321dc6e1
thread:issue:168532:issue:168533:issue:168534:issue:168535,issue #168532: TestModule: test/distributed/test_symmetric_memory.py::SymmMemNegativeTest,stale/low urgency,2,stale/low urgency,4,0,"168532, 168533, 168534, 168535",,,,3,add2410c666e87598b802efe71fd12675b8bcb57d8560eff24215ba31edd76ca
thread:issue:168706:issue:168707:issue:168708:issue:168709,issue #168706: TestModule: test/test_cpp_extensions_aot.py::TestCppExtensionAOT,stale/low urgency,2,stale/low urgency,4,0,"168706, 168707, 168708, 168709",,,,3,f1c164772c605c20f48cece2bc3ea50666bff118d8f7f88f0b33b28cf1f739fa
thread:issue:168831:issue:168832:issue:168833:issue:168834,issue #168831: TestModule: test/test_sparse_semi_structured.py::SparseSemiStructuredTensorCompileTest,stale/low urgency,2,stale/low urgency,4,0,"168831, 168832, 168833, 168834",,,,3,2b6f22ccadd8bea5ee38fec7208011117b21a68891c88b82c7d756a443631f3d
thread:issue:168851:issue:168852:issue:168853:issue:168854,issue #168851: TestModule: test/test_sparse_semi_structured.py::TestSparseSemiStructuredCUSPARSELT,stale/low urgency,2,stale/low urgency,4,0,"168851, 168852, 168853, 168854",,,,3,3c539d0dce0fdca153c85ca63bb4fdfd88acea4ac759e41a65953f3a9dc011fe
thread:issue:19092:issue:24090:issue:32078:issue:37837,issue #19092: [RFC] Memory format (aka layout aka NHWC) support,stale/low urgency,2,stale/low urgency,4,0,"19092, 24090, 32078, 37837",,,,3,fc63ab1ca098c65deabd4944615dbf6ac6e859b7e0f3207de53a895edeb8b2a0
thread:issue:19177:issue:41337:issue:53623:issue:55340,issue #19177: 8 tests in test_c10d fail when running all tests in one command,stale/low urgency,2,stale/low urgency,4,0,"19177, 41337, 53623, 55340",,,,6,37dcd559ae9a0a3189e9a1dc55a4a1000fd4d5e81e3c3420c5eb0821009a3e65
thread:issue:21760:issue:57121:issue:60295:issue:88002,issue #21760: [FR] faster reduce sum on expanded/unfolded tensors,stale/low urgency,2,stale/low urgency,4,0,"21760, 57121, 60295, 88002",,,,5,6b97ab43ed366c2eb4921fcc7150874e3260e6fa30550353aad6e15602bd1bbe
thread:issue:30421:issue:31129:issue:32463:issue:33034,issue #30421: [jit] Module references aren't preserved,stale/low urgency,2,stale/low urgency,4,0,"30421, 31129, 32463, 33034",,,,4,69a44cd76e2fb94676bfa6d7e0deb27182d54c01fb84e775c8357762bd9f030e
thread:issue:31394:issue:80439:issue:83980:issue:94083,issue #31394: torch.scatter_logsumexp,stale/low urgency,2,stale/low urgency,4,0,"31394, 80439, 83980, 94083",,,,3,35f1b19a81417d9ea7d29d44608eba4d69376bea131bfce7d640646cb71b7e78
thread:issue:35041:issue:62032:issue:64407:issue:65151,issue #35041: Make operator registrations truly commutative using priority,stale/low urgency,2,stale/low urgency,4,0,"35041, 62032, 64407, 65151",,,,3,0fd77e7f0ffabc61e9acc8636805cca23c6bbeabf1cc8f76e1075115d8aef5ca
thread:issue:44901:issue:51803:issue:54602:issue:78487,issue #44901: for loop can't be symbolically traced,stale/low urgency,2,stale/low urgency,4,0,"44901, 51803, 54602, 78487",,,,3,af5c216a24b5404cfccd3f8c558867dc84149e4985e83099c4030d498fe1f968
thread:issue:46374:issue:46642:issue:47052:issue:81749,issue #46374: Complex Number support for torch.nn modules,stale/low urgency,2,stale/low urgency,4,0,"46374, 46642, 47052, 81749",,,,4,240648dcb41e9e0be8574b8214116c02f9b7aaf58f6dcfaadea6a23a14a74d77
thread:issue:49016:issue:73479:issue:90560:issue:93859,issue #49016: Discrete Cosine Transform,stale/low urgency,2,stale/low urgency,4,0,"49016, 73479, 90560, 93859",,,,3,376920360e72ec8a8c04b73340f5505ebddf837d649f2e384a7ac5486baa8dc2
thread:issue:62614:issue:62616:issue:62619:issue:62675,issue #62614: LazyTensor recomputes intermediate tensor in fragmented computation,stale/low urgency,2,stale/low urgency,4,0,"62614, 62616, 62619, 62675",,,,3,322374d41a40931fc4a5b4cf25bd355d9dc0a0f904a7b3ed74aaa40bb0d6dcfa
thread:issue:100904:issue:104623:issue:105457,issue #100904: Revise glossary,stale/low urgency,2,stale/low urgency,3,0,"100904, 104623, 105457",,,,2,d54934bd45b3a19300033370a35037111a6ddbbaabcea09f3e4eaf094bfe9e6c
thread:issue:101314:issue:117350:issue:121207,issue #101314: Shared library loading logic breaks when CUDA packages are installed in a non-standard location,stale/low urgency,2,stale/low urgency,3,0,"101314, 117350, 121207",,,,2,bd7679e971fcc28ddd64f8015e2dd483bd37f0aa73fd7d7fcdda45c99073c342
thread:issue:102977:issue:16797:issue:43369,issue #102977: raise `RuntimeError` faster when loading an object with a torch CUDA tensor on a CPU-only machine,stale/low urgency,2,stale/low urgency,3,0,"16797, 43369, 102977",,,,2,c0ebcf885135ad2bb53e89a1f424ba0e4b99c7afbb68fb3574a580b3facc6570
thread:issue:104113:issue:144206:issue:70608,issue #104113: Documentation building fails due to torchgen,stale/low urgency,2,stale/low urgency,3,0,"70608, 104113, 144206",,,,2,c69a9b2df2a7940c2f8366b476d88baaad9c5cf2d4152c4fc00e28aedf3a0294
thread:issue:104472:issue:58833:issue:68041,issue #104472: Investigate numerical stability of forward-mode AD of some foreach functions,stale/low urgency,2,stale/low urgency,3,0,"58833, 68041, 104472",,,,2,e61e42a9fcf586a8cc5a38d1c7cae9311068a59776539d8f246e857f643609d7
thread:issue:106469:issue:106485:issue:146817,issue #106469: Extreme slowdown of torch.mm for certain sizes and strides with bfloat16,stale/low urgency,2,stale/low urgency,3,0,"106469, 106485, 146817",,,,3,bf9bd0678b3ef14d74c482cd947681a01499287f91277cd871c006d4ab10d314
thread:issue:106713:issue:106717:issue:106718,issue #106713: Dev Container Support for PyTorch,stale/low urgency,2,stale/low urgency,3,0,"106713, 106717, 106718",,,,2,bf27d13df5144d3a3dd174c82f96ba551c0043401d3706e4663907839cf1c4a0
thread:issue:107909:issue:66335:issue:71404,issue #107909: Provide a `reset_parameters()` method for MultiheadAttention to support FSDP meta device initializtion,stale/low urgency,2,stale/low urgency,3,0,"66335, 71404, 107909",,,,3,14479abbb09e7c7a28f76592b36b2cb68891f377711fae0c3e86e788f87fef1d
thread:issue:108108:issue:110681:issue:110702,issue #108108: [BC BREAKING] Change default behavior of scaled_dot_product_attention's causal masking alignment,stale/low urgency,2,stale/low urgency,3,0,"108108, 110681, 110702",,,,2,6e8a1756c664d9564047c3f7621d4afd13749c9d42e293e3dd5f9e09c3af388f
thread:issue:108174:issue:108175:issue:109770,issue #108174: FlashAttentionV2 will OOM when building on ci/cd with default settings,stale/low urgency,2,stale/low urgency,3,0,"108174, 108175, 109770",,,,2,de9984d02b0a3fd65c6d3aa32252bf2e1b85edf512c60b722a3e7b5124507e20
thread:issue:11071:issue:15004:issue:39947,issue #11071: undefined reference to 'caffe2::Caffe2FlagsRegistry[abi:cxx11]()',stale/low urgency,2,stale/low urgency,3,0,"11071, 15004, 39947",,,,2,cff6d3bd4981c068900eb6a0a2bf9f4e4315d225fe6de4ea77fc5141886a8f25
thread:issue:11202:issue:28119:issue:46169,issue #11202: [pytorch][feature request] Cosine distance / simialrity between samples of own tensor or two tensors,stale/low urgency,2,stale/low urgency,3,0,"11202, 28119, 46169",,,,2,1e52e0f2ee8a51d137e59822ea400fb2a03041957d97f266c96731d48ecfdbcc
thread:issue:11389:issue:18906:issue:30968,issue #11389: [distributions] Torch distribution samplers slow on expanded parameters,stale/low urgency,2,stale/low urgency,3,0,"11389, 18906, 30968",,,,2,c414e528a7cb802cc763c59c18a498e3d38564ddb3bd9629eeec3b644e1aeb91
thread:issue:11578:issue:52984:issue:54006,issue #11578: Request to import pytest in test/*.py,stale/low urgency,2,stale/low urgency,3,0,"11578, 52984, 54006",,,,2,0fed55ad8fa747efaee2923ab258951c0732a5c23ef210ed697634a37db44271
thread:issue:117736:issue:155218:issue:95024,issue #117736: Illegal memory access for nn.Conv2d when intermediate output has size larger than 2**31,stale/low urgency,2,stale/low urgency,3,0,"95024, 117736, 155218",,,,2,c4975fc363befbb7642cc417969827423d13d0875e9b4a3a40c57d9b8a48371e
thread:issue:122094:issue:124245:issue:135927,issue #122094: Investigate torch.compile Windows support.,stale/low urgency,2,stale/low urgency,3,0,"122094, 124245, 135927",,,,2,7f67369e4f9f3336f9bc7acace4d9e3c570526566fb2dbef3c4aeca01f3c3f6f
thread:issue:124016:issue:124222:issue:125626,issue #124016: Expand stale management to Issues,stale/low urgency,2,stale/low urgency,3,0,"124016, 124222, 125626",,,,2,490674c49b03daa454ad45f3e3e64260fcdc300266a8e8a7e9ffe97e287a825b
thread:issue:12983:issue:9484:issue:9604,issue #12983: Fail to Run Caffe2 with Successful Build: This caffe2 python run does not have GPU support,stale/low urgency,2,stale/low urgency,3,0,"9484, 9604, 12983",,,,2,a165387c7ad8f3fbe7b2d43b6df918a48ee951ee826aa32b302df1c8a57be431
thread:issue:131312:issue:138460:issue:141014,"issue #131312: Pytorch Error: ImportError: /home/dp/anaconda3/envs/ComfyUI/lib/python3.11/site-packages/torch/lib/../../nvidia/cusparse/lib/libcusparse.so.12: undefined symbol: __nvJitLinkComplete_12_4, version libnvJitLink.so.12",stale/low urgency,2,stale/low urgency,3,0,"131312, 138460, 141014",,,,2,745c12cb62f9784986bfa7ef2fffed5fd8b2abb64d0acf7aed351ed9927bdfa5
thread:issue:13207:issue:17966:issue:4716,issue #13207: [pytorch] [feature request] Error out if the needed GPU device capability is absent in runtime,stale/low urgency,2,stale/low urgency,3,0,"4716, 13207, 17966",,,,3,d33fea31f1de504eaf8ddaf02c5342a76b9f221bef373f8c9eeb95b72e5eda4e
thread:issue:13222:issue:56891:issue:61650,issue #13222: Memory inefficient in batched matmul when requiring gradients,stale/low urgency,2,stale/low urgency,3,0,"13222, 56891, 61650",,,,2,7c58a74157fc5be54c80185fb5e2a35b08d52853060ec7370c48e7b2540096c4
thread:issue:135860:issue:137767:issue:156152,issue #135860: Lobpcg for complex gives wrong results.,stale/low urgency,2,stale/low urgency,3,0,"135860, 137767, 156152",,,,2,bb2e59cd97a0d257b093b13b80d3cd41c087ebd564982cf406b1a28c793de51d
thread:issue:15070:issue:29814:issue:29816,issue #15070: In-place operations on `.data` or `.detach()` of sparse tensor doesn't update the original tensor,stale/low urgency,2,stale/low urgency,3,0,"15070, 29814, 29816",,,,2,6c6c2ee2d23f0bab043b4fa8c5a29fa7d457fc89929dc8a64d9d0b721e689566
thread:issue:151936:issue:154984:issue:155171,issue #151936: fully_shard() for huggingface 72B model: pytorch caches too much GPU memory,stale/low urgency,2,stale/low urgency,3,0,"151936, 154984, 155171",,,,2,e89e54850af98900e20365b62a28dfcc3f4477cbe10be5af7e10af4def2e5c04
thread:issue:16717:issue:20367:issue:5388,issue #16717: Performance regression on CPU from 0.4.1 to 1.0.0 on ResNet inference,stale/low urgency,2,stale/low urgency,3,0,"5388, 16717, 20367",,,,2,67b8b64178cca5e79502b3a0fc1892745ffefb52b1c2e0e904e8f68973a66c40
thread:issue:168475:issue:168476:issue:168477,issue #168475: TestModule: test/distributed/test_composability.py::ComposabilityTest,stale/low urgency,2,stale/low urgency,3,0,"168475, 168476, 168477",,,,2,e0e3bd461fe27d566cc10a25431135cf7f12652333bc863436cc5f021f62a306
thread:issue:168644:issue:168645:issue:168646,issue #168644: TestModule: test/inductor/test_multi_kernel.py::MultiKernelTest,stale/low urgency,2,stale/low urgency,3,0,"168644, 168645, 168646",,,,2,a8988d6133d463a5089aa32e61ce9eea602c826b31500697e9f565aa33de0341
thread:issue:168720:issue:168721:issue:168722,issue #168720: TestModule: test/test_cuda.py::TestCudaMallocAsync,stale/low urgency,2,stale/low urgency,3,0,"168720, 168721, 168722",,,,2,82f4a5f859c0f0a98face99ea9c1682ea2fe42e989907f566bc1b0a4af41bb5a
thread:issue:168801:issue:168806:issue:168807,issue #168801: TestModule: test/test_nn.py::TestNNDeviceType,stale/low urgency,2,stale/low urgency,3,0,"168801, 168806, 168807",,,,2,80955cbba6463763ef0f8ea56fc518ac7b9a3d3abfa14fda730cb25608baec13
thread:issue:20117:issue:71249:issue:71329,issue #20117: [FR] [RFC] add Sequential.append & .extend,stale/low urgency,2,stale/low urgency,3,0,"20117, 71249, 71329",,,,2,072aa771c7b11787bfa89a4839aeffd08ac7f105a0771cb55b6d200918a7dedc
thread:issue:22281:issue:30386:issue:6564,issue #22281: Box constraints for optimizers,stale/low urgency,2,stale/low urgency,3,0,"6564, 22281, 30386",,,,2,ec046b019a35412fe30b7d26e7dcddf9f82e53cffa7d11c40cac919c9010cdff
thread:issue:25032:issue:25310:issue:59629,issue #25032: [FYI] NestedTensor Project Progress,stale/low urgency,2,stale/low urgency,3,0,"25032, 25310, 59629",,,,2,1a5019f23f4ce61a5bf675599574e0a331f1e350bbb8f6618787d386d17e4746
thread:issue:27610:issue:30373:issue:33354,issue #27610: [FR][RFC] Build a serving framework to host and serve trained PyTorch models,stale/low urgency,2,stale/low urgency,3,0,"27610, 30373, 33354",,,,2,82ed35818e60829c837fe59c35916fd9a743cbf039dd7a3b84cbc390fe77552a
thread:issue:29093:issue:29739:issue:35363,issue #29093: error: 'SO_REUSEPORT' was not declared in this scope,stale/low urgency,2,stale/low urgency,3,0,"29093, 29739, 35363",,,,2,8ac195cd555df0cd189931807e833cf05c641518e390a814e2d9e3594f0a4845
thread:issue:29554:issue:36990:issue:40989,issue #29554: [feature request] Print some measure of fragmentation at CUDA out-of-memory,stale/low urgency,2,stale/low urgency,3,0,"29554, 36990, 40989",,,,2,fd07cb46e281503ec8ed64b58ec96e96e732948e8935249cda3458599d25c579
thread:issue:30532:issue:31007:issue:61411,issue #30532: Version 1.3 no longer supporting Tesla K40m?,stale/low urgency,2,stale/low urgency,3,0,"30532, 31007, 61411",,,,2,6a9d7e5fc20cc470622a11c50a86f378150edebd5894e219800d241ac64a4919
thread:issue:30929:issue:32407:issue:59218,issue #30929: How to set not to build libtorch_cpu.so and libmkl_*.so dependencies?,stale/low urgency,2,stale/low urgency,3,0,"30929, 32407, 59218",,,,2,60fd0977ab56a30a9cecd714666fc5dcabafbda8a0b3d0fbb97fa7d02e9b889f
thread:issue:41186:issue:41400:issue:57617,issue #41186: test_autograd failures on Power,stale/low urgency,2,stale/low urgency,3,0,"41186, 41400, 57617",,,,3,b684141be517ad502c0a3d8a4387080e9231f8cc83a94b8ceceeca62f52fe853
thread:issue:42502:issue:71409:issue:77140,issue #42502: [proposal] batch mode for randperm,stale/low urgency,2,stale/low urgency,3,0,"42502, 71409, 77140",,,,2,a71e2294276a09b2e1987960f5549183c80557ebf4b78c03b79e11e8bafe1694
thread:issue:43947:issue:44103:issue:63618,issue #43947: torch.cuda.synchronize Influence distributed training,stale/low urgency,2,stale/low urgency,3,0,"43947, 44103, 63618",,,,2,8a895c1620f2843931b78eeb22f31ff13da28ec7440e6fb2f32a9756101c9665
thread:issue:49477:issue:49716:issue:74148,issue #49477: Callbacks may not be automatically synchronized in a single NCCL future chain,stale/low urgency,2,stale/low urgency,3,0,"49477, 49716, 74148",,,,2,c92e81244f2e30a535e90f78ea4996058b405fab5c55f237c8fb1f3b784f9b17
thread:issue:49666:issue:52310:issue:54622,issue #49666: Use rlibm for faster and more accurate floating point operations,stale/low urgency,2,stale/low urgency,3,0,"49666, 52310, 54622",,,,2,a42016a3dc494ca441b2865d89a05ea223ed8d66b43ef8638839baf43c68eeec
thread:issue:50097:issue:67324:issue:74537,issue #50097: Cannot print 32-bit complex tensors,stale/low urgency,2,stale/low urgency,3,0,"50097, 67324, 74537",,,,2,3c2c319b5483a5869cc9b9df9926c4ea3c62b32536a9eecab6e0144a5aae2639
thread:issue:52973:issue:69467:issue:69469,issue #52973: Add numerically stable methods to torch.distributions (e.g. logcdf),stale/low urgency,2,stale/low urgency,3,0,"52973, 69467, 69469",,,,2,3d9d6577f545959dddda54c67d0768a4c36e2f49f0aca9865a52d8b7a2b28af3
thread:issue:53712:issue:62475:issue:79888,issue #53712: Poor support of `Optimizer.add_param_group`,stale/low urgency,2,stale/low urgency,3,0,"53712, 62475, 79888",,,,3,1e9bb72761f9d830cf401d93ef88822480b61385c7ec2a72994e2e14794d3cef
thread:issue:65049:issue:73640:issue:98200,issue #65049: Setting a random seed or passing a generator to Modules for deterministic sampling.,stale/low urgency,2,stale/low urgency,3,0,"65049, 73640, 98200",,,,3,c259f55c75b17da655bf6648bf71d4c358b713425783b49324a98b11ef45408b
thread:issue:65760:issue:65788:issue:73050,issue #65760: Codegen issues with Tensor(a!)? arguments in native_functions.yaml schemas,stale/low urgency,2,stale/low urgency,3,0,"65760, 65788, 73050",,,,2,83cd9b6d349caa89b00915e8d5df65062d4725062e17160fce78f2606db2b766
thread:issue:65868:issue:66491:issue:76555,issue #65868: torch.mul is not consistent with torch.multiply,stale/low urgency,2,stale/low urgency,3,0,"65868, 66491, 76555",,,,2,da10ac700404708408656b6f344fac1189c2f6c823d0057083a163ccc655ef3d
thread:issue:68287:issue:68288:issue:68298,issue #68287: [Lazy] Add torch.autograd.Function wrappers for the following ops in order to support dynamic shapes,stale/low urgency,2,stale/low urgency,3,0,"68287, 68288, 68298",,,,2,fbf93b76b987b539211be1a2287965071f6887e152fac581472ea6bcb742194e
thread:issue:69991:issue:72911:issue:84565,issue #69991: Composite Compliance Problems Tracker,stale/low urgency,2,stale/low urgency,3,0,"69991, 72911, 84565",,,,2,a75a8f2aab33d7834e28c4fe8c2d73859abb609373e9de2f1050c9b26e0d591d
thread:issue:72960:issue:73710:issue:73711,issue #72960: Refactor/Cleanup lazy IR codegen,stale/low urgency,2,stale/low urgency,3,0,"72960, 73710, 73711",,,,2,288bbe40318a9d95f16109f044a563d67ba50b296005a91f6ea9e63759d51d46
thread:issue:88144:issue:88464:issue:95116,issue #88144: AvgPool2D output shapes are inconsistent when ceil_mode=True,stale/low urgency,2,stale/low urgency,3,0,"88144, 88464, 95116",,,,3,95ec70919bdea65a902fff0f7fe9c6faaa23c9f6b6387274b04fca2e232dc38e
thread:issue:100626:issue:35749,issue #100626: Will Deep Implicit Models ever become first class citizens in PyTorch?,stale/low urgency,2,stale/low urgency,2,0,"35749, 100626",,,,1,0702346100eb82c70f732b30b7ba1969bfc83262e4b0dd68498ca8cce638e4d7
thread:issue:101110:issue:37334,issue #101110: Tensorboard graph tracing with torch fx API,stale/low urgency,2,stale/low urgency,2,0,"37334, 101110",,,,1,adedb5ddefcd42325b621c7fbd7213ec5ee5d00b1b42adfc3afccb7fa0cb3f22
thread:issue:101159:issue:37250,issue #101159: Inconsistency between GPU memory usage in torch.cuda.memory_summary and nvidia-smi,stale/low urgency,2,stale/low urgency,2,0,"37250, 101159",,,,2,176ca07deeba3c2ef312829812a334dc11f08ff4580fe0c5ba95c42f99e41348
thread:issue:10119:issue:7807,"issue #10119: [Caffe2] MNIST Tutorial LMDB Error ""Cannot open db""",stale/low urgency,2,stale/low urgency,2,0,"7807, 10119",,,,1,134ee04a6c8ff21954ff2940b87acc79bc41a9e9e7862ef4011dfdc16a61fcf9
thread:issue:101359:issue:97068,issue #101359: RuntimeError in Scaled Dot Product Attention Tutorial Code,stale/low urgency,2,stale/low urgency,2,0,"97068, 101359",,,,1,de0818e24f745930e4c49378e1aaa80d09bc62bbaecb9d32606c0c4cce57b36c
thread:issue:102673:issue:99176,issue #102673: Fix dynamo-related debug Python 3.11 failures,stale/low urgency,2,stale/low urgency,2,0,"99176, 102673",,,,1,eb9bc20ecd220efa90d3e5638e12ed690b87f8bd67e3f96d08d5e4ac58579672
thread:issue:103375:issue:103393,issue #103375: Inplace binary ops on tensor subclasses can cause mypy error,stale/low urgency,2,stale/low urgency,2,0,"103375, 103393",,,,1,fdaeb4d0988a49549729b3281824db3d29a3fc23bf0712a22b8a08ba8ae9c79b
thread:issue:103820:issue:150285,issue #103820: Upgrading SpGEMM algorithm to resolve Cusparse SpGEMM insufficient resources problem,stale/low urgency,2,stale/low urgency,2,0,"103820, 150285",,,,1,4d366515a5b0bcf281b75c7f2fc1b60d7afe770d2e2ef4e3da4db4537f4656c6
thread:issue:106206:issue:106297,issue #106206: Build failure due to C++ version mismatch,stale/low urgency,2,stale/low urgency,2,0,"106206, 106297",,,,1,1556c30ff236562461c9e741bcf856023d5d30431b26cb9c608cdc7c9aa1b903
thread:issue:106455:issue:40208,"issue #106455: vmap, jacrev, jacfwd, hessian, etc., in libTorch",stale/low urgency,2,stale/low urgency,2,0,"40208, 106455",,,,1,f611a096d7fc5f507a10820892ff3c8227c8da60d7f95a1d0211f6d3b9710df5
thread:issue:106951:issue:110758,issue #106951: stride of gradient is not same as the corresponding tensor,stale/low urgency,2,stale/low urgency,2,0,"106951, 110758",,,,1,7051ba570b7e13a9eedcb12f7c6d7326baf325a6425430635fa83cd575fb9f24
thread:issue:107896:issue:128785,issue #107896: [feature request] [ux proposal] Min-max linear normalization to be supported in F.normalize (or in a new function),stale/low urgency,2,stale/low urgency,2,0,"107896, 128785",,,,1,8f1f5b99cb75061a1520dc2762f28e21182eda319a4c2b23ab41fa47d3c5bccd
thread:issue:108739:issue:73656,"issue #108739: DDP Elastic ""master_addr"" resolution error in environment variables.",stale/low urgency,2,stale/low urgency,2,0,"73656, 108739",,,,1,ea65a707967ed97f4a352db7865dac66b76514cd94952c7811d11311a878f695
thread:issue:108934:issue:108984,issue #108934: PPC64le: vsx_helpers.h errors,stale/low urgency,2,stale/low urgency,2,0,"108934, 108984",,,,1,9dee4896a2f30da7f762400cdafee4cfd924e58e59af3387f43d51ad0ac2c1a7
thread:issue:110737:issue:144633,issue #110737: pytorch consuming all cpu cores 100% on ARM,stale/low urgency,2,stale/low urgency,2,0,"110737, 144633",,,,1,8a08f2aaef9f0d32f68610722164f8373814b9126ef4ffce8672d1b63bcd8c65
thread:issue:111384:issue:114859,issue #111384: [quant][pt2] Default batchnorm aten op has poor numerics during QAT,stale/low urgency,2,stale/low urgency,2,0,"111384, 114859",,,,1,83b561725594b7c92aa8e959422c03e8063c408aa9019ae26f3dedae9cc6888f
thread:issue:114569:issue:162235,issue #114569: inconsistency on torch.clamp,stale/low urgency,2,stale/low urgency,2,0,"114569, 162235",,,,2,aff30d5744fe4e82958bd3dbee9690a546fc1004feaef82f3f16028aa78ec6d0
thread:issue:114723:issue:114835,issue #114723: [RFC] Intel GPU Upstreaming,stale/low urgency,2,stale/low urgency,2,0,"114723, 114835",,,,1,244810e26f573232e7dd187c021bf16dee5f28cde7671ea184d273c9a7f1fc59
thread:issue:11514:issue:50192,issue #11514: [feature request] - Allow sequences lengths to be 0 in PackSequence,stale/low urgency,2,stale/low urgency,2,0,"11514, 50192",,,,1,47ed44b25464974868e9afcc3ef743c78ea2a48baf5885c0865d8ffdd6b0653e
thread:issue:115434:issue:75599,issue #115434: torch.kthvalue should take a tensor of `k` values.,stale/low urgency,2,stale/low urgency,2,0,"75599, 115434",,,,1,e740c2d537c69e07bb86561b4ffd22f03f6f87ab4aee9e1af49a7634b5d3ca8c
thread:issue:116461:issue:69359,issue #116461: Numpy like Masked/where operations,stale/low urgency,2,stale/low urgency,2,0,"69359, 116461",,,,2,72e2018d5d0804034bd3ead6f25785de57ba9c7142715bf3cd6e1090c83d2737
thread:issue:116626:issue:122287,issue #116626: `test_modules.py` does not test for half precision or complex types,stale/low urgency,2,stale/low urgency,2,0,"116626, 122287",,,,1,345b2a4ca0590da9bd40ef1fd215d2fba621be632216a3e38ffb7428f3bc14d7
thread:issue:117748:issue:118421,issue #117748: [NCCL 2.19.3] Performance reduction when using NVLSTree algorithm when running nccl-tests all-reduce,stale/low urgency,2,stale/low urgency,2,0,"117748, 118421",,,,1,e9f0f55d420eff7955679e692352e27fe316b77931b485129314c88d43ca76d9
thread:issue:118018:issue:118715,issue #118018: Enable all optimizers (except SparseAdam + LBFGS) to be capturable,stale/low urgency,2,stale/low urgency,2,0,"118018, 118715",,,,1,dde3b914c791ce6930debd7ad5b678ae14cd75dedc0bd938832953fa74ec878d
thread:issue:118093:issue:56300,issue #118093: torch.where only accepts boolean tensor as condition deviating from np.where where you can also use binary input condition of dtype float,stale/low urgency,2,stale/low urgency,2,0,"56300, 118093",,,,1,ade1630315b7f60e2a9b6bf9f4a54072acc2fd41407ea8f0eb9dc21e76699a70
thread:issue:118736:issue:87371,issue #118736: [typing] `torch.Tensor` subclass types (eg `nn.Parameter`) broken by 2.2 release,stale/low urgency,2,stale/low urgency,2,0,"87371, 118736",,,,1,d1cec0b3979735e8fff6d5697731994cb96cc6e24efa1f6ba63462d2a585e796
thread:issue:11982:issue:56397,issue #11982: [feature request] Publish wheels with debug symbols,stale/low urgency,2,stale/low urgency,2,0,"11982, 56397",,,,1,790bc50a3de55443510d1c55e45c66378b9cd5532f3b407fc5a818cface2aa83
thread:issue:120930:issue:71631,issue #120930: make_graphed_callables doesn't turn off autocast for backward,stale/low urgency,2,stale/low urgency,2,0,"71631, 120930",,,,1,716b290da360e099fdc8178e06bbbebc297bd70ba74787b883ffb167241348cc
thread:issue:12117:issue:15630,"issue #12117: Error: Internal Compiler Error (codegen): ""there was an error in verifying the lgenfe output!""",stale/low urgency,2,stale/low urgency,2,0,"12117, 15630",,,,2,d4699b012cd262e294429c175e35705769cfef17ddabb05f0a3106125ca76175
thread:issue:121894:issue:121969,issue #121894: TORCH_USE_CUDA_DSA compiler flag not working,stale/low urgency,2,stale/low urgency,2,0,"121894, 121969",,,,1,e00b239a89d85855c805c6aeefc0f7cd16bdc9803fa2483c2c67e38dec0832e6
thread:issue:122641:issue:72775,issue #122641: [vulkan] Vulkan backend fails creating tensor on aarch64 (Raspberry Pi 5),stale/low urgency,2,stale/low urgency,2,0,"72775, 122641",,,,1,12bd28d514703dcc7875fba2c0e0050a221bf852171dbe3dab231db5498519d7
thread:issue:123835:issue:134117,issue #123835: Package manager install on Nvidia Grace Hopper does not make cuda available,stale/low urgency,2,stale/low urgency,2,0,"123835, 134117",,,,1,e277d1f187022424fcb3971d47b4a2cae4a11052d6b08a62cfed24dad10ed0a5
thread:issue:124262:issue:94294,issue #124262: CUBLAS_STATUS_EXECUTION_FAILED when calling cublasGemmEx,stale/low urgency,2,stale/low urgency,2,0,"94294, 124262",,,,1,1bff41fd6009360863ce5ff07b621319e43bbef9bd55fe722b926874895ba93e
thread:issue:12449:issue:37766,"issue #12449: Could not find a package configuration file provided by ""Torch"" with any of the following names:",stale/low urgency,2,stale/low urgency,2,0,"12449, 37766",,,,1,947422276eb55a563b5426644c069c58bff712b9dbac5e8f8319d3a348e13af5
thread:issue:124788:issue:84864,issue #124788: torch.nn.checkpoint.checkpoint ignores default device in backward() call,stale/low urgency,2,stale/low urgency,2,0,"84864, 124788",,,,1,1e747d47170aa0450976d416f3dbbcbd6a5ee922af289ceaadd9895299e90fdc
thread:issue:12484:issue:18536,issue #12484: CuDNN convolution on some CUDA devices will not preserve NaN weights (upstream bug),stale/low urgency,2,stale/low urgency,2,0,"12484, 18536",,,,1,85890f46ce46e76f5dfb5a72af44454500f13ea29eb8c168d1aa25b27708797c
thread:issue:1249:issue:41142,issue #1249: Dice Loss PR,stale/low urgency,2,stale/low urgency,2,0,"1249, 41142",,,,1,055fb17d847286188d20e767643220b5c16411b2200bd18dbb70bc0775e0a604
thread:issue:125837:issue:140090,issue #125837: Support for PyTorch on iOS Following PEP 730,stale/low urgency,2,stale/low urgency,2,0,"125837, 140090",,,,1,b4b7634b735207091e84a612fe61b4b72ffbff0bcb79afded90384a3e8eeece3
thread:issue:126037:issue:126039,issue #126037: A scalar with `input` argument of `pow()` doesn't work,stale/low urgency,2,stale/low urgency,2,0,"126037, 126039",,,,1,7fb18d56708bdbba351b59ce009a57943be177f11adbd86e83790834b652f07f
thread:issue:126921:issue:139429,issue #126921: [ONNX] view(dtype=dtype) is not supported by both onnx.export and onnx.dynamo_export,stale/low urgency,2,stale/low urgency,2,0,"126921, 139429",,,,1,1aaa5f37b39a4ad306a198a5cf8fa625a4121f138fb0cdb92078f840c87086e5
thread:issue:127075:issue:82510,issue #127075: What is the processing principle when the complex64 input tensor contains nan or inf for addition?,stale/low urgency,2,stale/low urgency,2,0,"82510, 127075",,,,1,330714d39ba6ad845d3cca6655f39dd2b88d3b2bd24ea7aeb716ea80c0b61dfd
thread:issue:127164:issue:79703,issue #127164: randn generates different output for 4x4 tensor size sliced to match shape of direct 2x4 or 4x2 and compare output,stale/low urgency,2,stale/low urgency,2,0,"79703, 127164",,,,1,9ad5cd52b4361115e56122f9dee3fcabbc053c0be00a684897b3987c28f3d114
thread:issue:128071:issue:143670,issue #128071: [user empathy day 2][based] torch.compile issues,stale/low urgency,2,stale/low urgency,2,0,"128071, 143670",,,,1,318476817623ecc644f800b96a3e6fe7a58928d2b910493bd52989e0970faa2f
thread:issue:128641:issue:129272,issue #128641: Backward is not supported by `MaskedTensor`'s `torch.maximum`,stale/low urgency,2,stale/low urgency,2,0,"128641, 129272",,,,1,0895117e72535a03d7141239693643af1694ef9de62485294ce0ead75d125cc9
thread:issue:129140:issue:152220,issue #129140: [RFC][C10D] Avoid creating new nccl communicator for each P2P pair,stale/low urgency,2,stale/low urgency,2,0,"129140, 152220",,,,1,ae5eccabbccac9af5f939fdd2d32156a431ccf35af96a64abcfb864ddb7a9fff
thread:issue:129442:issue:129749,issue #129442: Volatile GPU-Util keeps 100% when using torch.distributed.isend/irecv,stale/low urgency,2,stale/low urgency,2,0,"129442, 129749",,,,1,a138bc3e9cd406d9213f2394b4c5a702fc86194ccb6cd2af58bb04db944e1132
thread:issue:129657:issue:135359,issue #129657: Implementation of a numerically stable log(1 - softmax) function in PyTorch,stale/low urgency,2,stale/low urgency,2,0,"129657, 135359",,,,1,cb44b46bd1f2115a08af243c8796707f272653b967003186433d12a42ac1213d
thread:issue:130055:issue:130058,issue #130055: [pipelining] Non-first stage only takes tensors,stale/low urgency,2,stale/low urgency,2,0,"130055, 130058",,,,1,e118f8484433f76f6b0702d6746106a17f1a8aea54eeb2d89001682cce9b45e2
thread:issue:130217:issue:148743,issue #130217: `unique()` is not intuitively understandable at all.,stale/low urgency,2,stale/low urgency,2,0,"130217, 148743",,,,1,bd72c41dcc625920c5a02a087d410c30522bd774d67949335d74136659e550dd
thread:issue:13023:issue:7359,issue #13023: Python dataloader Improvements,stale/low urgency,2,stale/low urgency,2,0,"7359, 13023",,,,1,6304fc011356d932bd9ed0c0cf64f7fc3813256adca453707bf11df22c18d870
thread:issue:13130:issue:166512,"issue #13130: arm64 port for PyTorch, libtorch",stale/low urgency,2,stale/low urgency,2,0,"13130, 166512",,,,1,5d96d26a901b64aa5db5a5dfb0e1c7a0986d00de9fa9454ae35f2661c70438ec
thread:issue:131650:issue:131667,"issue #131650: Pytorch 2.4, Cuda 12.4: RuntimeError: device >= 0 && device < num_gpus INTERNAL ASSERT FAILED at ""/opt/conda/conda-bld/pytorch_1720538439675/work/aten/src/ATen/cuda/CUDAContext.cpp"":49, please report a bug to PyTorch. device=^A, num_gpus=^A",stale/low urgency,2,stale/low urgency,2,0,"131650, 131667",,,,1,e7c1286b0f4ae0e8ca3c429ffbe81c551be04531a66ffa4499007e44635c9f2c
thread:issue:13188:issue:16708,issue #13188: cudnn explicit paths and GCC multilib suffixes prevents detection of good cudnn headers,stale/low urgency,2,stale/low urgency,2,0,"13188, 16708",,,,1,513ac63733efd9cb9bc7d306cb3eba06ebec1e314bcdccd3b47a256409a34de6
thread:issue:13218:issue:5580,issue #13218: Support for integer interpolation (torch.nn.functional.interpolate),stale/low urgency,2,stale/low urgency,2,0,"5580, 13218",,,,1,e2620a672691ea4e70584170f5dbe2d9a02dbd946045a3706bfa2f02b67fd1bf
thread:issue:133823:issue:73008,issue #133823: Issues compiling `torch` with `mkl`,stale/low urgency,2,stale/low urgency,2,0,"73008, 133823",,,,1,585deb92c994e002f4aae64f5f196450e0151324efa71ba26adc4157ae7b5c18
thread:issue:13447:issue:62466,issue #13447: Support gathering nested lists in DataParallel,stale/low urgency,2,stale/low urgency,2,0,"13447, 62466",,,,1,3d5ebed543ba6ac7592ed7ade7416a093fd44083392f657cf9e77996892d1c3f
thread:issue:137367:issue:137373,"issue #137367: `lr`, `momentum`, `weight_decay` and `dampening` parameter of `optim.SGD()` work with `bool` values",stale/low urgency,2,stale/low urgency,2,0,"137367, 137373",,,,1,a1028f526275fe58ee40e96509906c91d7da93b3dc08c4ecf0131e6f8226aa2b
thread:issue:137482:issue:42246,"issue #137482: `torch.ops.aten.divide_` leads to ""RuntimeError: result type Float can't be cast to the desired output type Int""",stale/low urgency,2,stale/low urgency,2,0,"42246, 137482",,,,1,4f8d3fe5165dc670157cb9bc955669e1e2e0eaf9df8916a262c361c0ac4056a5
thread:issue:138396:issue:138399,issue #138396: `out=` meta device support.,stale/low urgency,2,stale/low urgency,2,0,"138396, 138399",,,,1,dd13c9267b674825765fcf859825d7a275174218bcc3514458488215b0639b02
thread:issue:138786:issue:140437,issue #138786: Support nn.Module arguments for the function to torch.compiler.allow_in_graph,stale/low urgency,2,stale/low urgency,2,0,"138786, 140437",,,,1,6f98807bc20f59ba1e3356b1f203a8c062c689c77d424b936cfba8f5df6a2291
thread:issue:138800:issue:160508,issue #138800: Batching rule not defined for `aten::_make_dual`.,stale/low urgency,2,stale/low urgency,2,0,"138800, 160508",,,,1,0f187cdc2d517c5dcdb70c45d121372f0a32626529098a2e416a4a09048e2dc5
thread:issue:139643:issue:141079,issue #139643: [triton 3.2] test_convolution2,stale/low urgency,2,stale/low urgency,2,0,"139643, 141079",,,,1,e0bceee06e251fa32bb9fe8bb9c03a4e982a77f30ad85cbffda8721e33dfa18b
thread:issue:140222:issue:140252,issue #140222: inconsistency in ```torch.nn.functional.log_softmax``` on CPU and GPU,stale/low urgency,2,stale/low urgency,2,0,"140222, 140252",,,,1,d616cd58afe19b35b31135d2f080d668c9a9fac611ba8e9b4477c28de1aaef8f
thread:issue:140563:issue:140879,issue #140563: NCCL hangs cause timeout,stale/low urgency,2,stale/low urgency,2,0,"140563, 140879",,,,1,0e66ce343caf3e97888dce988c975d835cfdd111d858b26a9cf8d1cd165e19b0
thread:issue:140787:issue:145693,issue #140787: Apple Silicon Unified Memory usage,stale/low urgency,2,stale/low urgency,2,0,"140787, 145693",,,,1,a0fc94ae3db2ed39889ea5e060713ff11135f10f68422447a8247113f9c175b3
thread:issue:141177:issue:141822,issue #141177: `CrossEntropyLoss` much slower with class probabilities vs. class indices target,stale/low urgency,2,stale/low urgency,2,0,"141177, 141822",,,,1,ba3f3d982f66ed3c1732c9ae616d320d7022adb8763b95cfebc87a42e67e08a8
thread:issue:142703:issue:143913,issue #142703: We don't have test coverage for aarch64 (ARM64) bfloat16 feature (__ARM_FEATURE_BF16),stale/low urgency,2,stale/low urgency,2,0,"142703, 143913",,,,1,a01fe02f298b48fd2779263468325c7b459d0d244824595dbadf34b71a117042
thread:issue:14489:issue:98675,"issue #14489: Batch matmul with sparse matrix, dense vector",stale/low urgency,2,stale/low urgency,2,0,"14489, 98675",,,,1,91aef99667d66ae583dd478cf4a3871c1f724ac34a7eb8b3ceadfcb0550a06af
thread:issue:145233:issue:90760,issue #145233: Not using set_num_threads results in very slow .all(),stale/low urgency,2,stale/low urgency,2,0,"90760, 145233",,,,1,c80a3bdb993d6173120893928e34ee11dba0b037fd103a6dab3aa57e44c2fc19
thread:issue:145511:issue:155490,issue #145511: Activation Checkpointing composability with split backward computation,stale/low urgency,2,stale/low urgency,2,0,"145511, 155490",,,,1,1a99dad1282297cd6496d1e89aac008491d86219a4e508f2038892275674e339
thread:issue:145710:issue:146201,issue #145710: LBFGS-B with cuda implementation (and CPU one too),stale/low urgency,2,stale/low urgency,2,0,"145710, 146201",,,,1,f8ac9ca8b1c3351bcc01c9a3f68893eab01d3c5f8cb6e7cc792a611cc254320c
thread:issue:146898:issue:149734,issue #146898: [RFC] Test Cases Enabling for Accelerators,stale/low urgency,2,stale/low urgency,2,0,"146898, 149734",,,,1,eb616cd3d42ba516fe59361689fde70822013249a947570900536f61444eaf96
thread:issue:147044:issue:152285,issue #147044: Clean up backend_type_map from distributed_c10d,stale/low urgency,2,stale/low urgency,2,0,"147044, 152285",,,,1,e3e91185d8a5ff1a2170e7156e5a9b533eba91e7c429d7d850ceb0c57e95a49a
thread:issue:150277:issue:152226,"issue #150277: Sparse tensor indexing not implemented, but partially supported by using index_select",stale/low urgency,2,stale/low urgency,2,0,"150277, 152226",,,,1,cf1499691400e44b250431a22408fe3429a6b874e3830d5534c2482180734ab0
thread:issue:150943:issue:151960,"issue #150943: [RFC][TCPStore] advanced store operations (queues, pub/sub)",stale/low urgency,2,stale/low urgency,2,0,"150943, 151960",,,,1,132720ea6eb309012b8bdb48d6f4d61dd20265c08433beff2906fee008d3f46c
thread:issue:152032:issue:163142,issue #152032: RFC: The State of Custom CUDA extensions in PyTorch,stale/low urgency,2,stale/low urgency,2,0,"152032, 163142",,,,1,9d4b8012e8dadbc05217de0701ec0320d0fc03086c4e56cdd5a58b69f3747874
thread:issue:15457:issue:86162,issue #15457: Improve one_hot,stale/low urgency,2,stale/low urgency,2,0,"15457, 86162",,,,1,8338ab04b96ba8e2a7d06b274bbd6ef764af498c77e3e9d449cabca33bce216c
thread:issue:155325:issue:97990,issue #155325: Excessively restrictive dependencies,stale/low urgency,2,stale/low urgency,2,0,"97990, 155325",,,,1,a8331d62bdacdb12caf7c3a7070ab3233751341e8f80da11d9cfac604e2ea2ce
thread:issue:15617:issue:499,issue #15617: Feature request: transposed locally connected layer,stale/low urgency,2,stale/low urgency,2,0,"499, 15617",,,,1,7f7ccbfd7d5558481dfa0f9ff235e5b0c60bb6031539cddc7fcead1415201222
thread:issue:159229:issue:159230,issue #159229: [Precompile] Assertion error when loading package in Nanogpt training with caching precompile fails after second load,stale/low urgency,2,stale/low urgency,2,0,"159229, 159230",,,,1,92516f19c8ae7e323b7a7b23e010f6daa7f51a984824e68434dfc4d7585e6b1c
thread:issue:159353:issue:159852,issue #159353: torch.cond triggers shape mismatch with linear layer when nested in custom model,stale/low urgency,2,stale/low urgency,2,0,"159353, 159852",,,,1,9f13e338cb5f7229b5044339fb3ca9bf1708556aefcf9c7cdfde517fe5b5ebaf
thread:issue:159380:issue:168253,issue #159380: `torch.nested.nested_tensor_from_jagged` does not automatically compute `max_seqlen`,stale/low urgency,2,stale/low urgency,2,0,"159380, 168253",,,,1,b1d4974b5df31f77e16b6fc707c9081e51c7aa7d5917d1e8898d354be729f95e
thread:issue:159906:issue:160420,issue #159906: New at::HostAllocator interface prevents using more than one allocator implementation for a device type,stale/low urgency,2,stale/low urgency,2,0,"159906, 160420",,,,1,05b4caf7f2f49ec94c8ad3abf2fe12801f94758b117ece5061c2898345993c64
thread:issue:165244:issue:38271,issue #165244: torch.tensor slicing with negative step raises ValueError,stale/low urgency,2,stale/low urgency,2,0,"38271, 165244",,,,1,7cf2eb8e6b5a3101c7fbc81f7d9f2143a47254cc06c95c0b009e5a63df2c4283
thread:issue:168402:issue:168403,issue #168402: TestModule: test/distributed/_composable/fsdp/test_fully_shard_overlap.py::TestFullyShardOverlap,stale/low urgency,2,stale/low urgency,2,0,"168402, 168403",,,,1,a0828a508603ef6913c3ea340d7fb0c63b6eaeac93023cfcafa7f2202f169bbd
thread:issue:168404:issue:168405,issue #168404: TestModule: test/distributed/_composable/fsdp/test_fully_shard_training.py::TestFullyShardNDTraining,stale/low urgency,2,stale/low urgency,2,0,"168404, 168405",,,,1,68fa65b517724968c1fb1242409dc9468fda7877458276c38ee839ea61411375
thread:issue:168406:issue:168407,issue #168406: TestModule: test/distributed/_composable/test_composability/test_2d_composability.py::TestFullyShard2DTraining,stale/low urgency,2,stale/low urgency,2,0,"168406, 168407",,,,1,550ffc588fb392a0c7574356b48ef7d5eab0db1be35001c6edc6026bb755638f
thread:issue:168422:issue:168423,issue #168422: TestModule: test/distributed/_shard/sharded_tensor/test_sharded_tensor.py::TestShardedTensorFromLocalTensor,stale/low urgency,2,stale/low urgency,2,0,"168422, 168423",,,,1,826d0c0d79585c4b6e10a60e332375d2d7068de022fc017c0d612b6208433557
thread:issue:168434:issue:168435,issue #168434: TestModule: test/distributed/elastic/utils/distributed_test.py::DistributedUtilTest,stale/low urgency,2,stale/low urgency,2,0,"168434, 168435",,,,1,acb804f2e1fda48581acb3a0d932b8c330fe683e3b4d64f4055cd54dc8b9c188
thread:issue:168445:issue:168446,issue #168445: TestModule: test/distributed/tensor/test_attention.py::RingAttentionTest,stale/low urgency,2,stale/low urgency,2,0,"168445, 168446",,,,1,0cffb501d0fccd7f3f43f58c66c7a47fe751ee5b40c63ad0496aec44da9e9363
thread:issue:168447:issue:168448,issue #168447: TestModule: test/distributed/tensor/test_matrix_ops.py::DistMatrixOpsTest,stale/low urgency,2,stale/low urgency,2,0,"168447, 168448",,,,1,61314385a4fdc736871b24b4d0aac3fb3de7a011cdb1e3b0a4215727d47c0b68
thread:issue:168449:issue:168450,issue #168449: TestModule: test/distributed/tensor/test_tensor_ops.py::DistTensorOpsTest,stale/low urgency,2,stale/low urgency,2,0,"168449, 168450",,,,1,8407875c7a53526b48843f8590d8e71050d3ea2c4cb968d28a48e79362584a52
thread:issue:168451:issue:168452,issue #168451: TestModule: test/distributed/test_aten_comm_compute_reordering.py::TestComputeCommReorderingMultiProc,stale/low urgency,2,stale/low urgency,2,0,"168451, 168452",,,,1,21be95a5e1290ab2516af4020535968f72c48cd5a014303ac740e1cb74e8f387
thread:issue:168480:issue:168481,issue #168480: TestModule: test/distributed/test_nccl.py::NCCLSymmetricMemoryTest,stale/low urgency,2,stale/low urgency,2,0,"168480, 168481",,,,1,23a89cd9c04a4ab859e99f10c3f853286c8e6dc65a9e5e2db5be12720c64b9af
thread:issue:168536:issue:168537,issue #168536: TestModule: test/distributed/test_symmetric_memory.py::LoweringTest,stale/low urgency,2,stale/low urgency,2,0,"168536, 168537",,,,1,249c91e8a7f71a717cf418d9987a3072607871c46ef0f25e439e3d58d24a044f
thread:issue:168623:issue:168624,issue #168623: TestModule: test/inductor/test_minifier_isolate.py::MinifierIsolateTests,stale/low urgency,2,stale/low urgency,2,0,"168623, 168624",,,,1,13557cffd72f472e07ab75bd754d5ecda0bec98eb4c0044781f21baa4534f22c
thread:issue:168692:issue:168696,issue #168692: TestModule: test/nn/test_convolution.py::TestConvolutionNNDeviceType,stale/low urgency,2,stale/low urgency,2,0,"168692, 168696",,,,1,d1dd900a71784127f4ab70f4f7fc114ac253f31528c52282d8028faf0fc3e3cc
thread:issue:168742:issue:168743,issue #168742: TestModule: test/test_cuda_primary_ctx.py::TestCudaPrimaryCtx,stale/low urgency,2,stale/low urgency,2,0,"168742, 168743",,,,1,d5932ed24b69eae2ac7d2fa816474d64b81051a54ad98f8e76d1ae6f28cf346e
thread:issue:168780:issue:168781,issue #168780: TestModule: test/test_matmul_cuda.py::TestMixedDtypesLinearCuda,stale/low urgency,2,stale/low urgency,2,0,"168780, 168781",,,,1,051a2c514f33bce91a90e1becfdb9addb9ef1d6d609dc7445ecea5a74f4f2d23
thread:issue:168798:issue:168800,issue #168798: TestModule: test/test_nn.py::TestNN,stale/low urgency,2,stale/low urgency,2,0,"168798, 168800",,,,1,03cd04b6c89fb30d6dbd25be61bf0aa38b38c750989c6fffb1d116d3585939cb
thread:issue:168855:issue:168858,issue #168855: TestModule: test/test_testing.py::TestTesting,stale/low urgency,2,stale/low urgency,2,0,"168855, 168858",,,,1,c23fdb15ad3d21c9c88134ff8980b4f9995952cb06b1063eaa487e2a21a89b96
thread:issue:172034:issue:71379,issue #172034: Become compatible with CMAKE_CUDA_ARCHITECTURES,stale/low urgency,2,stale/low urgency,2,0,"71379, 172034",,,,1,97d13e40dc08a75b8dee11acbb90487f0e07165e4ba6486c1780f96533b95e95
thread:issue:17234:issue:7795,issue #17234: [feature request] build and move distributions w/ device and/or dtype,stale/low urgency,2,stale/low urgency,2,0,"7795, 17234",,,,1,f960151eeee5cf7c0b89e51bc2db84ba4f57dfd1e2599395ce5314296be63bd9
thread:issue:173225:issue:86076,issue #173225: [RFC] Renaming MPS backend as Metal backend,stale/low urgency,2,stale/low urgency,2,0,"86076, 173225",,,,1,09fcd314e9544720f8abc3b805f6bd523b9a83a684e6678dc0e043400528c546
thread:issue:17425:issue:19120,"issue #17425: improved assert message in the case of ""CUDA error: device-side assert triggered""",stale/low urgency,2,stale/low urgency,2,0,"17425, 19120",,,,1,4459a21b5b72db423556be92eddc01fe4b03abe64dbee4163dd8e20371246c73
thread:issue:17774:issue:21688,issue #17774: RuntimeError: cuDNN error: CUDNN_STATUS_EXECUTION_FAILED when batch size is too large,stale/low urgency,2,stale/low urgency,2,0,"17774, 21688",,,,1,24b7a7cc57c5756c69f8b03afb6eedf4b466454f992331e23a68962045c880c6
thread:issue:17897:issue:31708,issue #17897: CUDA large matrix-vector product (torch.mv) causes illegal memory access,stale/low urgency,2,stale/low urgency,2,0,"17897, 31708",,,,1,f68fcf5ef95bdd56cd50ca448a44514882b86b2a369c58a9ee435e49eda1e202
thread:issue:17901:issue:40988,issue #17901: should disable AVX on 32bit x86 / refine AVX availability tests,stale/low urgency,2,stale/low urgency,2,0,"17901, 40988",,,,1,bc7b600a44a9f448fc946f1497864b306145fc1f2117e12aac397a3c9aaa3481
thread:issue:18634:issue:31474,issue #18634: Speed-up torch.cat on CPU,stale/low urgency,2,stale/low urgency,2,0,"18634, 31474",,,,1,d6381737a892fae55f69676b52a9fc685bc0411729303112ffa9f68a6704d647
thread:issue:19248:issue:91300,issue #19248: add stable distribution in torch.distributions,stale/low urgency,2,stale/low urgency,2,0,"19248, 91300",,,,1,8c5d554e7c35f1976eeaaf3e4bafe9f8cc5fe8848db7551f29bbc1b890a1ba19
thread:issue:19637:issue:19685,"issue #19637: [jit] Traced {zeros,empty}()/{zeros,empty}_like() calls do not respect device args.",stale/low urgency,2,stale/low urgency,2,0,"19637, 19685",,,,1,3bba9a07ab0bc84ec904adbfb4cef832354aa08ff2575d1836a131a07396625a
thread:issue:2001:issue:62009,issue #2001: Implement similar PyTorch function as model.summary() in keras?,stale/low urgency,2,stale/low urgency,2,0,"2001, 62009",,,,1,824ad83e9fe468e5a40dd170b297c1a9975159ffce271ebb37d713bf5ceabc07
thread:issue:20323:issue:69386,issue #20323: Support size to `torch.normal`,stale/low urgency,2,stale/low urgency,2,0,"20323, 69386",,,,1,b1a4e411a0d52c9fd1d0a2c6d16593a3ff1a1de412f957662d2ad7e314414295
thread:issue:21700:issue:43459,issue #21700: (LLD 8.0.0) ld: error: can't create dynamic relocation R_X86_64_DTPOFF32 against symbol: ideep::utils::computation_cache,stale/low urgency,2,stale/low urgency,2,0,"21700, 43459",,,,1,1c68cd71c87af66ad1176eddd87b4c9ccf8411eeb5502ab69eb9c51993e9f024
thread:issue:21731:issue:22414,issue #21731: Improve multithreaded random number generation (RNG),stale/low urgency,2,stale/low urgency,2,0,"21731, 22414",,,,1,05cc2f6e0ebcaba9a591c2043e3fc431b71a46347fd6e9572a4161ce0d16c2a6
thread:issue:22343:issue:3790,issue #22343: New Weight Scheduler Concept for Weight Decay,stale/low urgency,2,stale/low urgency,2,0,"3790, 22343",,,,1,4c1d49475931c174b09e29e512aebb2c82c7d537db2284a926477b4aaecf4fed
thread:issue:23425:issue:49321,issue #23425: Hanging on when one gpu node return zero as loss in the context of distributed data parallel training,stale/low urgency,2,stale/low urgency,2,0,"23425, 49321",,,,1,0f4f6182e6e4f2bb6f0622767e702deb1c5ab4b1c90f0097997f381497bcae38
thread:issue:24234:issue:91887,issue #24234: torch.utils.tensorboard.SummaryWriter fails to flush at program exit,stale/low urgency,2,stale/low urgency,2,0,"24234, 91887",,,,1,3be43b46105f654ef01d13be059aabf4283290deac0b077bcc7eaa4fd818ac8b
thread:issue:24478:issue:24498,issue #24478: Port CPU_tensor_apply functions to TensorIterator (umbrella issue),stale/low urgency,2,stale/low urgency,2,0,"24478, 24498",,,,2,e74d7980f1e7961b3416a4efdad58ea3fff012caca59a90b0421a172e9deae7e
thread:issue:2478:issue:30873,issue #2478: ReduceLROnPlateau with a naive Backtracking,stale/low urgency,2,stale/low urgency,2,0,"2478, 30873",,,,1,0489a4acff781df0be7dd40069079224928b03b9a229a1852122f3df085e3d30
thread:issue:24834:issue:56126,issue #24834: PyTorch 1.2 'module' object has no attribute 'BFloat16StorageBase',stale/low urgency,2,stale/low urgency,2,0,"24834, 56126",,,,1,04e59efffbb9b0b4fc94495ea276aa0b3252075a3a59f0f524e2d8359d1ef3fa
thread:issue:25150:issue:46724,issue #25150: Problems with install python from source,stale/low urgency,2,stale/low urgency,2,0,"25150, 46724",,,,1,efcd4efccbf5391b89bf18ef011919251b0add59e19938ed67df47dfeef6d59c
thread:issue:25416:issue:38995,issue #25416: Add GPU support to c10d allreduce_coalesced,stale/low urgency,2,stale/low urgency,2,0,"25416, 38995",,,,1,f049ce31a395269d946817682709c161f5134a1808293c570335ab82b77b0d07
thread:issue:25478:issue:68070,issue #25478: Delete TensorOptions::operator==,stale/low urgency,2,stale/low urgency,2,0,"25478, 68070",,,,1,f04ae7d8e160e1a14af2015b25d3073024924f74ce843cea903c137ea5df67c5
thread:issue:25591:issue:29987,issue #25591: IValue pickle does not work properly if an empty tensor table is not provided,stale/low urgency,2,stale/low urgency,2,0,"25591, 29987",,,,1,65dfab7f51c95cf4510f1471206e5b985929927a0b478fb8d910cd655121daaa
thread:issue:27144:issue:27475,issue #27144: torch::jit::script::Module has no zero_grad(),stale/low urgency,2,stale/low urgency,2,0,"27144, 27475",,,,1,799f49105849a3adc37941df0fdda1c73008c25c4c7d3b73754e3a741aee3cb7
thread:issue:27647:issue:40107,issue #27647: 🚀 Graceful RPCAgent termination in multi-driver scenario,stale/low urgency,2,stale/low urgency,2,0,"27647, 40107",,,,1,e6e57b1465a93915ead21cfd47065da58fb3703b4a920abae06e8349ae5eccd4
thread:issue:28417:issue:29042,issue #28417: Patch: Fix for using `clang` to compile CUDA,stale/low urgency,2,stale/low urgency,2,0,"28417, 29042",,,,1,d189aaa7b90346b1f9b17c479f6302d2de48f6e4d73264db6f87465b857d9fea
thread:issue:28594:issue:57815,issue #28594: [BUG] Can't Deepcopy module with weightnorm,stale/low urgency,2,stale/low urgency,2,0,"28594, 57815",,,,1,53c177fc54a6d4ebb3c57aa4c976e6c54cde88229f9b15e13e3917671203d996
thread:issue:28619:issue:59168,issue #28619: Channels Last (NHWC) support plan.,stale/low urgency,2,stale/low urgency,2,0,"28619, 59168",,,,1,c2d6ea3162b8b128a445cd3976b35f432dfeeedb0d68adc11f83e871b6cf3bab
thread:issue:28733:issue:34675,issue #28733: CPU MaxPool2d is very slow,stale/low urgency,2,stale/low urgency,2,0,"28733, 34675",,,,1,1d5e0e28acc267d53170c6211eabd82a07e8212d13bd0d5b22648aae8045fe42
thread:issue:29028:issue:29116,issue #29028: Inconsistent documentation for in-place functions at https://pytorch.org/docs/stable/torch.html,stale/low urgency,2,stale/low urgency,2,0,"29028, 29116",,,,1,687253d48de22cb8976beb125d7de78dfaf651959e638011c43381cd5cce10ee
thread:issue:29548:issue:38975,issue #29548: Dispatch key reorganization,stale/low urgency,2,stale/low urgency,2,0,"29548, 38975",,,,1,597e8e3e78b28798d7aad625e49271b8a72fce3f00c3c2e74126062152257e5b
thread:issue:29981:issue:72258,issue #29981: [feature request] Multivariate normal CDF,stale/low urgency,2,stale/low urgency,2,0,"29981, 72258",,,,1,4f62385cfd7714c5330867d9cd6d0d8c4691a507134bffa0fb099efe28ae2022
thread:issue:30574:issue:31913,issue #30574: `index_select` with multidimensional `index`,stale/low urgency,2,stale/low urgency,2,0,"30574, 31913",,,,1,e8108484f74e1553a64b21ca6b8f47043b80664c1ac10a898ec99bd90467705a
thread:issue:30633:issue:31228,issue #30633: Remove RPC internal helper that overrides the default pickler,stale/low urgency,2,stale/low urgency,2,0,"30633, 31228",,,,1,9f50260a97671be6b2102f4a7c4773c64f2639dfd3bc5535464d39fd27dbf5e5
thread:issue:30803:issue:31356,issue #30803: Pytorch openmp thread number tuning option for CPU trainning,stale/low urgency,2,stale/low urgency,2,0,"30803, 31356",,,,1,70a9eceb0142874abf50375748e5413e5a3d2d3a1f011153b262afe22ea0c856
thread:issue:30903:issue:31907,issue #30903: Don't ship protoc in wheels,stale/low urgency,2,stale/low urgency,2,0,"30903, 31907",,,,1,e17fde926310820aa276f5ca346280b3422ff3d7d99a91aab814025cc774d545
thread:issue:30965:issue:32544,issue #30965: JIT breaks with postponed annotations,stale/low urgency,2,stale/low urgency,2,0,"30965, 32544",,,,1,e6b5b3c1282fbacb2d4c2b0e0919fc6c7f46217bcc464f798043fef646510bee
thread:issue:31461:issue:96036,issue #31461: Make RRef.to_here() non-blocking,stale/low urgency,2,stale/low urgency,2,0,"31461, 96036",,,,1,592fe62926009a91898618ffcca3740960381f3299b88512a411d90cf4c778a3
thread:issue:31528:issue:43453,issue #31528: cuCtxGetDevice error and seg fault with DDP and OpenMPI,stale/low urgency,2,stale/low urgency,2,0,"31528, 43453",,,,1,1f897a1ce5f53f7f1e5ab908e46c746f3c2075df2b3981ed2548aef084ac7547
thread:issue:31822:issue:88980,issue #31822: Libtorch's files conflict with glog's file?,stale/low urgency,2,stale/low urgency,2,0,"31822, 88980",,,,1,bc6030caf76914737d8f0c6c4ee2c7694dcf89bd317bd573fd004bf89dd1e0f2
thread:issue:32300:issue:32365,issue #32300: [jit] Various problems calling `@staticmethod`s in 1.4.0,stale/low urgency,2,stale/low urgency,2,0,"32300, 32365",,,,2,037b53b550ce7099eec0be205303491c61e7481e4d3cb4871eb5c250a03506b0
thread:issue:32306:issue:43867,issue #32306: [FR] bincount along arbitrary dimension,stale/low urgency,2,stale/low urgency,2,0,"32306, 43867",,,,1,9425f133cf79ed7b32ff0820fd7bd2050ae879ab48c55ff3595194e7d3187ccb
thread:issue:34306:issue:50962,issue #34306: [docs] Missing docs online for conv_tbc,stale/low urgency,2,stale/low urgency,2,0,"34306, 50962",,,,1,6ad05bff3fbee9e24bc88b78c0a3965e76401e4ca004d27f43dfb4e1df9d4873
thread:issue:34361:issue:35446,issue #34361: [JIT legacy executor] device propagation regression,stale/low urgency,2,stale/low urgency,2,0,"34361, 35446",,,,1,6e2bc7a3d148679ce3365514279b50fd9d7e3a35148620bab0c60e9c0f0941d3
thread:issue:34606:issue:37488,issue #34606: Multiple CMake target errors ever since commit 0e52627358,stale/low urgency,2,stale/low urgency,2,0,"34606, 37488",,,,1,e74686fa652d2985392f967fbbc8b174d8617fd79f5715c8d5914852ba1d7a9a
thread:issue:35678:issue:36999,issue #35678: USE_AVX/USE_AVX2 does not affect __AVX2__ macro defition,stale/low urgency,2,stale/low urgency,2,0,"35678, 36999",,,,1,329ac19b93732730956c46173e6a0ee3dbf496c149642a0900eae1af5b08573d
thread:issue:3619:issue:4358,issue #3619: Multiprocessing with torch.solve hangs,stale/low urgency,2,stale/low urgency,2,0,"3619, 4358",,,,1,5370de33a47ad4084f1597c0de21c1788f8330bc7109f27d8ac4878d6ba2be64
thread:issue:36333:issue:45769,issue #36333: Illegal memory access / CUDNN_STATUS_MAPPING_ERROR on Quadro 8000,stale/low urgency,2,stale/low urgency,2,0,"36333, 45769",,,,1,4136ee48174b06c855093a1e1f39175b53de606322ef728818010c6646b499de
thread:issue:36444:issue:55661,issue #36444: Comparison ops for Complex Tensors,stale/low urgency,2,stale/low urgency,2,0,"36444, 55661",,,,1,6093af09fe09d239bd3a2c847afcc49437304087ed843ef27167314b0c7b7242
thread:issue:3667:issue:4392,issue #3667: Exposing CuDNN benchmark strategy selection,stale/low urgency,2,stale/low urgency,2,0,"3667, 4392",,,,1,fab1149896755bea6c08dbd903bacc2b4b62d5bd4e72728f0a5373146848d4a7
thread:issue:37002:issue:42849,issue #37002: [RFC] Modularize DistributedDataParallel,stale/low urgency,2,stale/low urgency,2,0,"37002, 42849",,,,1,24501dcef68237d305dbea50eaa04e1900bca39498ffc872a163f43b3453333c
thread:issue:37160:issue:78738,issue #37160: Enhanced operator context when reporting errors,stale/low urgency,2,stale/low urgency,2,0,"37160, 78738",,,,1,a7c13aa7b0c0c704393ad834a20b0038199c123384739fa4db0a2ffecbdb879d
thread:issue:38010:issue:39836,issue #38010: Split more pages into sub-topics,stale/low urgency,2,stale/low urgency,2,0,"38010, 39836",,,,1,929dcd228c1fac5f0c239c966b4f911b06d9a7fe04fb4780c48d311f884ee539
thread:issue:38185:issue:634,issue #38185: Feature Request: Sampled softmax loss,stale/low urgency,2,stale/low urgency,2,0,"634, 38185",,,,1,b6e1ab25f893365a9b9adb56a9e309e2c5170aae573eb75ca6e1b1e264318f50
thread:issue:38310:issue:50503,issue #38310: Compiling errors when trying to cross-compile the C++ API for RTOS (QNX),stale/low urgency,2,stale/low urgency,2,0,"38310, 50503",,,,1,78ba636f5ee9affe15e3add6a75098c556efe4a98d298851fb5b12e68eb3ccd3
thread:issue:38622:issue:47513,issue #38622: Clarification for usage of negative loss with optim.lr_scheduler.ReduceLROnPlateau,stale/low urgency,2,stale/low urgency,2,0,"38622, 47513",,,,1,5638e7c3e8da5ae991a7ab380654473fd9c5ed88bcfe79f30fb25ee429be1916
thread:issue:38642:issue:50092,issue #38642: Synchronization problem in torch.distributed with MPI on CUDA,stale/low urgency,2,stale/low urgency,2,0,"38642, 50092",,,,1,e8dad4b7447284b6d5cf9a81dde65a9fa43885c078770049fbd66fc1c26b3cb6
thread:issue:38767:issue:42843,issue #38767: weight_decay in Adam is not an L2 Penalty,stale/low urgency,2,stale/low urgency,2,0,"38767, 42843",,,,1,2eea70cbf2013c7dab28c586d921d9ab21be649304ca24f0ca401125c8ab1def
thread:issue:39224:issue:53391,issue #39224: Negative stride values in `as_strided`,stale/low urgency,2,stale/low urgency,2,0,"39224, 53391",,,,1,d878b7250a485f791a2a71d9a0f5358433a66b2c77ca4d65360ec785ed1af8c2
thread:issue:40581:issue:48959,issue #40581: Libtorch C++ multiple GPU performance slower than single GPU,stale/low urgency,2,stale/low urgency,2,0,"40581, 48959",,,,1,89687cb94880b48eecedc0ce1b4b47af1f37604fa3301e0d68b175a11b9a958c
thread:issue:4145:issue:4181,issue #4145: [Proposal] Consistent `batch_first` effect for RNN modules,stale/low urgency,2,stale/low urgency,2,0,"4145, 4181",,,,1,c898568d8b7c5b4e94112b4a1d89ac6030280d4ffc0016839b88c90786634f1e
thread:issue:41530:issue:53903,issue #41530: Parallel computation of the diagonal of a Jacobian,stale/low urgency,2,stale/low urgency,2,0,"41530, 53903",,,,1,a7fe46048778a5c0c3a6d1a4832b8b2f6fbfabdbda985998e7a080c74922a174
thread:issue:42109:issue:58136,issue #42109: Slow `index_add_` on `torch.long` tensors,stale/low urgency,2,stale/low urgency,2,0,"42109, 58136",,,,1,8dc83044795fd0d434911f7c4d86d8c04e17436725a8d6b58f7d7d967fbcfc12
thread:issue:42295:issue:42316,issue #42295: KeyError: 'track_running_stats' in batchnorm.extra_repr,stale/low urgency,2,stale/low urgency,2,0,"42295, 42316",,,,1,5de97e93212b219b137a50b297a121fd22916c87566036481793add91c0a1827
thread:issue:42812:issue:7617,issue #42812: checkpoint_sequential breaks backpropagation,stale/low urgency,2,stale/low urgency,2,0,"7617, 42812",,,,1,6487f884e03303fd62c3bc036eed97d9bce9e94a45420cd67cc8459d0b2bec3e
thread:issue:43127:issue:66640,issue #43127: `onnxifi_load` symbol undefined in `libtorch_cpu.a` library built from source,stale/low urgency,2,stale/low urgency,2,0,"43127, 66640",,,,1,28c23d9c0cb35d459f9eee0275a4a410aaa6f98d395e72b1d8814e8732c5bdd6
thread:issue:44159:issue:55174,"issue #44159: [Feature Request] Add an option to run GPU tests only, and skip all CPU tests",stale/low urgency,2,stale/low urgency,2,0,"44159, 55174",,,,1,8227dfccd47b512bdc4dbd98ff9677fc2eb90baf2942888c8d49c5088b01215f
thread:issue:44533:issue:48548,issue #44533: “doxygenfunction: Unable to resolve multiple matches for function ...” in C++ documentation,stale/low urgency,2,stale/low urgency,2,0,"44533, 48548",,,,1,281423ef7af08c4eae6e71934bc06771a74e6ef7c16ea0b89e5b68891e616db3
thread:issue:45682:issue:45685,issue #45682: [FX] Allow customization of the behavior of Proxy,stale/low urgency,2,stale/low urgency,2,0,"45682, 45685",,,,1,35be19946c155e99d503dff2f60e1b8cf12a211170bdbdf04bf324aff1bfed13
thread:issue:4632:issue:52211,issue #4632: CUDNN_STATUS_INTERNAL_ERROR when training with conv3d,stale/low urgency,2,stale/low urgency,2,0,"4632, 52211",,,,1,b62544118e0d08a94e6bd80e0f3181d68433c4add9ced5e38bc415f2cfbc83fd
thread:issue:46559:issue:46653,issue #46559: Exception raised in rpc_async context is silently handled,stale/low urgency,2,stale/low urgency,2,0,"46559, 46653",,,,1,516fd3627a04856b93ce1c2160a84244c637376cb48e24268062195b20e1fe19
thread:issue:46902:issue:47649,issue #46902: How to use clang as a cuda compiler instead of nvcc?,stale/low urgency,2,stale/low urgency,2,0,"46902, 47649",,,,1,73c4d2aee68ba0d255d1150f10d3395cddc7be1dc88fdaa788e554529b584f8e
thread:issue:47149:issue:62171,issue #47149: Allow all torch.nn modules to accept arbitrary batch dimensions,stale/low urgency,2,stale/low urgency,2,0,"47149, 62171",,,,1,cb1156b91da152b13673b9f883623b80f0f11fc71517e43df162433b86290c72
thread:issue:47669:issue:48878,issue #47669: Pytorch 1.7.0 with cuda 11.1.1 and cudnn 8.0.5,stale/low urgency,2,stale/low urgency,2,0,"47669, 48878",,,,1,4486e4dd04cd5bc0f254b4fbcc52e6d763abbf1dea1658f31d728a4b2467cd53
thread:issue:47688:issue:47702,issue #47688: ARM Mac 16-core Neural Engine,stale/low urgency,2,stale/low urgency,2,0,"47688, 47702",,,,1,b670e1d99346eed368e1af740596615ee358bdc30434012a349a01ff653fbd45
thread:issue:47908:issue:54408,issue #47908: The speed of pytorch with cudatoolkit 11.0 is slower than cudatoolkit 10.2,stale/low urgency,2,stale/low urgency,2,0,"47908, 54408",,,,2,6095b6a17790f9bee9e2e0d7930afe3fbe9f805c6e3e2568a2c42784aee5c8f6
thread:issue:48246:issue:53707,issue #48246: [RFC] XPU device for PyTorch,stale/low urgency,2,stale/low urgency,2,0,"48246, 53707",,,,1,46309cdeea39d24b16075c2aa94a587883e51153fdcc6709acfabca15e5b1667
thread:issue:48353:issue:63023,issue #48353: [FR] bool tensor should support basic arithmetics,stale/low urgency,2,stale/low urgency,2,0,"48353, 63023",,,,1,294ce6995cd0d60dcbaae1a833c170fac6f00955037cde4cf745f45a37931746
thread:issue:49252:issue:52915,issue #49252: Methods that solve systems of linear equations are memory-inefficient (batch-wise broadcasting),stale/low urgency,2,stale/low urgency,2,0,"49252, 52915",,,,1,590a3cdaf3b4f41e9c73eaf4ede6e739260374ac5fa457a6291192f2585bfb19
thread:issue:49480:issue:64185,issue #49480: Internal symbols are leaking from torch.nn.functional,stale/low urgency,2,stale/low urgency,2,0,"49480, 64185",,,,1,86c90cb6d3c4dc2838ffe3bec76cb888aadad3640190303b74da37556a928249
thread:issue:49642:issue:75419,issue #49642: Turn deprecation warnings into errors in CI,stale/low urgency,2,stale/low urgency,2,0,"49642, 75419",,,,1,51e9193320b8e9daea105e5ac16b3684bf6077872b945417dfb593fef0dc94a5
thread:issue:49726:issue:50292,issue #49726: Wrong error is raised for property of nn.Module (again),stale/low urgency,2,stale/low urgency,2,0,"49726, 50292",,,,1,e88286f738d2c1178cb4a7cd6a5fd829053d41c1903fd1300ef47f1fe82b5672
thread:issue:49949:issue:50802,"issue #49949: RFC: Add a ""see also"" section to operators' documentation",stale/low urgency,2,stale/low urgency,2,0,"49949, 50802",,,,1,38bd5a6fbfcc4fd82a7c0a1daa1314f0ec29d9d28f82f38a26cac21e81b55541
thread:issue:49992:issue:59566,"issue #49992: [docs] nn.modules pages should mention corresponding functional versions, e.g. for nn.Hardshrink",stale/low urgency,2,stale/low urgency,2,0,"49992, 59566",,,,1,856b08487d972b18b76d08372727e86baaa9cc11e1affeb06a218e7697dd5f92
thread:issue:50034:issue:74771,issue #50034: Multinomial without replacement produces samples that have zero probability,stale/low urgency,2,stale/low urgency,2,0,"50034, 74771",,,,1,0a1bd5fcdd7e0d4fd6be8c9172946ab2b43aa6cf167c73ee94d51be42b0e0001
thread:issue:50689:issue:50940,issue #50689: RandomSampler is very slow with huge dataset,stale/low urgency,2,stale/low urgency,2,0,"50689, 50940",,,,1,ff5f097db418ef9b2cd6aaa32218c2b86bfc7206757102913b68a7040fc24e9b
thread:issue:51471:issue:62094,issue #51471: Make TorchBind object appear in state_dict,stale/low urgency,2,stale/low urgency,2,0,"51471, 62094",,,,1,b8e344286b72aa23ff29ecc30b27c12a0b11f83a3770178a006839b3f849b7ce
thread:issue:51782:issue:51896,issue #51782: [discussion] [python] Generic way for generating module-wrappers for stateless parametrized functions,stale/low urgency,2,stale/low urgency,2,0,"51782, 51896",,,,1,fa14e0332d4c4b757f154be611fe15db067f847445b4ff79d5da95e801b8be99
thread:issue:52260:issue:52289,issue #52260: Unlucky batch->device split by DataParallel causes an exception,stale/low urgency,2,stale/low urgency,2,0,"52260, 52289",,,,1,a04a112d3aaa7e66f4f8d9653383c74b30a7e3473e19dfbb5806454d54cbfc58
thread:issue:52265:issue:62147,issue #52265: 'TypeError: expected CPU (got CUDA)' when placing a CUDA tensor on a class that is inheriting torch.Tensor,stale/low urgency,2,stale/low urgency,2,0,"52265, 62147",,,,1,f73873d8e327a79afb4a92087fdf7a298ad771690932ac468ccbb61abadaedaa
thread:issue:52332:issue:76558,issue #52332: Reciprocals of complex tensors with infinities are different from NumPy.,stale/low urgency,2,stale/low urgency,2,0,"52332, 76558",,,,1,4c514b717bea381fb6ba608ecfd4cbf623309e68a7b14625f2c8afb4d8341be2
thread:issue:52648:issue:52717,issue #52648: Fix `torch.linalg.vector_norm`'s backward to avoid needing `at::abs()` call in forward for ord=+/-inf,stale/low urgency,2,stale/low urgency,2,0,"52648, 52717",,,,1,dbba962cc3d880db030caf42762a99bcc169753d4c89a3ea2440772f32694177
thread:issue:53534:issue:62021,issue #53534: [FX] Ability to wrap functions in other modules for symbolic tracing,stale/low urgency,2,stale/low urgency,2,0,"53534, 62021",,,,1,15446ff53c4201d56a706843298d7c16b436932da3baab88515a5024621e10d4
thread:issue:55296:issue:55297,issue #55296: Build errors with USE_VULKAN=ON when cross-compiling for Android,stale/low urgency,2,stale/low urgency,2,0,"55296, 55297",,,,1,15db34319ad6c3ab97a25634132c3d2c6b4444b4d890ed830f747b4d7e486c06
thread:issue:56440:issue:59526,issue #56440: Raise exception when torch.clamp min value underflows the input tensor's dtype,stale/low urgency,2,stale/low urgency,2,0,"56440, 59526",,,,1,e1cc146e1a81774b32507d3b56acb6dbdb08422bc09e282d4958beedbe972546
thread:issue:56480:issue:56921,issue #56480: shared torch.tensor with multiprocesses using python Queue cause coredump,stale/low urgency,2,stale/low urgency,2,0,"56480, 56921",,,,1,a3ef4ce1d8e5561fb3e3551d972ab7f3a9b7a60c68375629c259aed4c9b37c40
thread:issue:56794:issue:70241,issue #56794: No tensor operations allowed inside at::parallel_for,stale/low urgency,2,stale/low urgency,2,0,"56794, 70241",,,,1,d339a763e31af98f25fddc3750f3b161f2ccae28020398e5669d0f4dd0b57987
thread:issue:58128:issue:60234,issue #58128: the same input with different batchsize got different precision output,stale/low urgency,2,stale/low urgency,2,0,"58128, 60234",,,,2,aaafa1068d20689dcca765a701fa456a843cde11f5786c5c2a43cc36e457e3fc
thread:issue:58466:issue:69688,issue #58466: `set_per_process_memory_fraction()` does not ensure max used GPU memory below fraction,stale/low urgency,2,stale/low urgency,2,0,"58466, 69688",,,,1,19ac9f9f5d89be4f7bc0ce570b4ee608cf46d4d0d0be8862c495c49b9b5646f5
thread:issue:58779:issue:59251,issue #58779: C++ Extensions that use ninja should enable colors by default,stale/low urgency,2,stale/low urgency,2,0,"58779, 59251",,,,1,5e0412956cecf5f7f8faee8f6334f40b475d9152e365107614ead80b7d559041
thread:issue:59682:issue:61079,issue #59682: Add a lint rule to recommend TORCH_CHECK(false) over throw std::runtime_error,stale/low urgency,2,stale/low urgency,2,0,"59682, 61079",,,,1,ec21798cfe2840774e830dd8f2da8f25b61d5f36cedadfe0d54e86eb43e49793
thread:issue:59855:issue:63295,issue #59855: pytorch test failed,stale/low urgency,2,stale/low urgency,2,0,"59855, 63295",,,,1,6389d2892162afcc78074564cec477ff13c068cbc1a57fec56fc0299707063bb
thread:issue:60149:issue:60153,issue #60149: backward compatibility - need a way to find out when a certain API was added/modified/etc.,stale/low urgency,2,stale/low urgency,2,0,"60149, 60153",,,,1,5d664c31797a0b6beceb678c26e4b513fab675e7452e7d1bc0e3b1bdea41889c
thread:issue:60832:issue:94586,issue #60832: State of Torch Named Tensors,stale/low urgency,2,stale/low urgency,2,0,"60832, 94586",,,,1,117c8c332237e08494f4a89a83fc78f43fae957955c3ad66d377d1e978fb4b0a
thread:issue:61211:issue:61213,issue #61211: [feature request] Do zero-padding in high-frequency modes in `ifft`,stale/low urgency,2,stale/low urgency,2,0,"61211, 61213",,,,1,337b0dc7c7e1abac0d09f8b14854e19498a02244c082f9c1e99b9d7a134247f0
thread:issue:61658:issue:77166,issue #61658: Move `torch.cholesky_solve` into `torch.linalg`.,stale/low urgency,2,stale/low urgency,2,0,"61658, 77166",,,,1,e07fb07fdf74341c5fc68a3fab8a0827c6bdfe2e5a130547fda99c1e92aca6fd
thread:issue:62396:issue:70559,"issue #62396: Make interpolation output size compatible with opencv, scikit-image and scipy for floating scale factor",stale/low urgency,2,stale/low urgency,2,0,"62396, 70559",,,,1,4336bd95f7bfe79b810b61cec79aadb8003a469ebf58bcbc2b0973aa3a77e2d0
thread:issue:62451:issue:84234,issue #62451: torch.Generator ignores the device argument?,stale/low urgency,2,stale/low urgency,2,0,"62451, 84234",,,,2,e05ff3586083e16cf79599493cb70b062fd7a44f5ab42762086e2cbef12a048f
thread:issue:6273:issue:7127,issue #6273: Multithreading Scaling Issue with MKL,stale/low urgency,2,stale/low urgency,2,0,"6273, 7127",,,,1,1fa200dfc12fccbabb656847a9c707c5035d44ea22edbd66a94ece3603809d4f
thread:issue:63034:issue:74519,issue #63034: [bug] nn.functional.pad (circular) ubsan failure,stale/low urgency,2,stale/low urgency,2,0,"63034, 74519",,,,1,85757e9acdbfad02844fea1d4423f6e8fc5ef14a5105970448cde79d5819aa46
thread:issue:63079:issue:98414,"issue #63079: PyTorch 1.9.0, test_optim fails on Nvidia A10.",stale/low urgency,2,stale/low urgency,2,0,"63079, 98414",,,,1,afce8f3fbd71b6e79c4f3aac45b0725645964e3ef839f55c9e054b138f49afc9
thread:issue:64162:issue:89054,issue #64162: [feature request] torch.Generator constructor to accept seed directly,stale/low urgency,2,stale/low urgency,2,0,"64162, 89054",,,,1,8f33c75fa37b80c08d41945275462f80ac0abf48c11cd1957fabd06806a7ce8c
thread:issue:65131:issue:65132,issue #65131: Adding a device variable to a `torch.Tensor` subclass fails,stale/low urgency,2,stale/low urgency,2,0,"65131, 65132",,,,1,b33e4f492163dc37e9fd291b2bebf5e83e50f296564d7b507fe403b2b0b38c29
thread:issue:66750:issue:66751,issue #66750: torch.nn.EmbeddingBag behave differently in two cases of cpu and cuda,stale/low urgency,2,stale/low urgency,2,0,"66750, 66751",,,,1,567b52964dc011f82fc8219fc0b607e475cd026bbe85216c895f650f56382e57
thread:issue:67153:issue:70931,issue #67153: Build not working with cuda 11.5,stale/low urgency,2,stale/low urgency,2,0,"67153, 70931",,,,1,e65868d593dc431c1c5a13ec586f8fe53e6dd442a13e71bcdf20dfbeb07271fa
thread:issue:67970:issue:70485,issue #67970: Error when using torch.fx on bert,stale/low urgency,2,stale/low urgency,2,0,"67970, 70485",,,,1,5123009a29e625c8af1dc3612106fa80cfe0e375412936375eb378bbd31cdfd9
thread:issue:69353:issue:78109,issue #69353: Pruning `torch.nn.MultiheadAttention` causes RuntimeError,stale/low urgency,2,stale/low urgency,2,0,"69353, 78109",,,,1,ba45269412ba6d1c407bda3fa39a4b3709196b38865a185cf41ba91fbf9594d3
thread:issue:69531:issue:69532,"issue #69531: `(svd|pca)_lowrank`: backward is unstable when for a matrix `A`, the parameter `q` is set to a value `q > rank(A)`.",stale/low urgency,2,stale/low urgency,2,0,"69531, 69532",,,,1,b2390719f5cfdd4ce37607d5041ac6f55026bde8cca060ec54971af11dd7519d
thread:issue:69786:issue:95776,issue #69786: Implement aten::equal for sparse tensors,stale/low urgency,2,stale/low urgency,2,0,"69786, 95776",,,,1,ee26cc9e0f000bc92bd4bae86cf95e3c5d292f69a0db0d24cfaa1dccbeb60d6b
thread:issue:69822:issue:69892,issue #69822: Implement torch.*_like tensor creation functions on sparse inputs,stale/low urgency,2,stale/low urgency,2,0,"69822, 69892",,,,1,77a69e775bc28a7daa8cc5427859e8f5e46cdebbc2185388b867377684f10e63
thread:issue:70242:issue:70678,issue #70242: `TensorIterator`: provide a two-argument version of `set_output`,stale/low urgency,2,stale/low urgency,2,0,"70242, 70678",,,,1,b4633fa15e495859539142a2e8986c733b686d768e662e82879a5c55b439bf4b
thread:issue:71446:issue:82565,issue #71446: Discussion of TorchQuantum and QuantumNAS,stale/low urgency,2,stale/low urgency,2,0,"71446, 82565",,,,1,d36cf632d6fd30cfd693920764ea77095b32c73bc18c4f9318df926e3ad9c9a0
thread:issue:71636:issue:95756,issue #71636: `torch.median` will return -2147483648 when input is an empty tensor,stale/low urgency,2,stale/low urgency,2,0,"71636, 95756",,,,1,ed5331158290a18338e614fe63943e1753749137ef895ae54824b67740e1db90
thread:issue:72556:issue:72558,"issue #72556: Clarify test dependencies (e.g., into a test-requirements.txt file)",stale/low urgency,2,stale/low urgency,2,0,"72556, 72558",,,,1,3b3b7969d7bbed2331284276d6db9bc4bc9a112f2df15fcd4f32cf1fae4fe050
thread:issue:73113:issue:73412,issue #73113: PyTorch fails to compile on gcc 11.2 due to breakpad,stale/low urgency,2,stale/low urgency,2,0,"73113, 73412",,,,1,5d7355c3c65f1af06393ba6911d6b77f8577545e467e1033618a80611dcd963b
thread:issue:73176:issue:99410,"issue #73176: Cuda lacks checking of ""out of bound""",stale/low urgency,2,stale/low urgency,2,0,"73176, 99410",,,,1,fe526d0e8f4125189951a1ba2a25fd8dbf6b51b7855abd58cc8b50176a31b0eb
thread:issue:73190:issue:73870,issue #73190: Segmentation fault in max_pool1d,stale/low urgency,2,stale/low urgency,2,0,"73190, 73870",,,,1,5c9501eaeb4655be67ec496503cc60ecd8d185bac1d4c4a989304465a9aa8176
thread:issue:73487:issue:85773,issue #73487: PyTorch not recognizing GPU on WSL - installed cudnn and cuda,stale/low urgency,2,stale/low urgency,2,0,"73487, 85773",,,,1,97398053d8a0b03de6695cc148e1b1a69c9608048c3f50bf0c850d0a46b114cd
thread:issue:74236:issue:74420,"issue #74236: ""histogram_cpu"" not implemented for 'Int'",stale/low urgency,2,stale/low urgency,2,0,"74236, 74420",,,,1,487acfb976e2bc81f4806d1740237de2513dba3258a7d3f269be9db49c261071
thread:issue:74616:issue:76806,"issue #74616: __rpow__(self, other) OpInfo should not test the case where `other` is a Tensor",stale/low urgency,2,stale/low urgency,2,0,"74616, 76806",,,,1,d2febb0f97c47fa2bc7a5246a9314c36eafc5619d86d8827d3e378ee3b4dd70e
thread:issue:75363:issue:76354,issue #75363: conv3d has numerical issue where same input produces output that are not bit-wise identical,stale/low urgency,2,stale/low urgency,2,0,"75363, 76354",,,,1,8c2a898618a255a34cf2bbb083ffeed897ac8f0b9b2212b740725928b13b2003
thread:issue:76433:issue:76514,issue #76433: Remove _log_softmax/_softmax in favor of log_softmax and softmax respectively.,stale/low urgency,2,stale/low urgency,2,0,"76433, 76514",,,,1,39d8217b16ae0d539cb04499f5a5320ae236b0976d71ced2076893f9e864f3e4
thread:issue:76483:issue:76853,issue #76483: Expand pow and float_pow sampling function for more coverage,stale/low urgency,2,stale/low urgency,2,0,"76483, 76853",,,,1,be045c6d640ecc9af54a90bff69a2ffa93206721e796946418614db30a025bd6
thread:issue:76654:issue:76655,issue #76654: `torch.sort` does not exploit parallelization when invoked without the `dim` parameter.,stale/low urgency,2,stale/low urgency,2,0,"76654, 76655",,,,2,de0fe45fd10639d3a98fb03a1c6da308a1db2de0839d413150d2013f61fcca37
thread:issue:76785:issue:95434,"issue #76785: `addr, baddmm, dist, l1_loss` will backward fail when input tensors have different dtypes",stale/low urgency,2,stale/low urgency,2,0,"76785, 95434",,,,1,6be32e582923d43d8120a4a9cf803f37aabc2f26c8ca44eb35541be00da068d0
thread:issue:77223:issue:77332,issue #77223: torch.ops.aten.ceil(1.5) returns Tensor rather than scalar,stale/low urgency,2,stale/low urgency,2,0,"77223, 77332",,,,1,624d0c733027479827a27efa5015a5bf306ec542b087544001769c11d81b16c0
thread:issue:77415:issue:95590,issue #77415: Suboptimal error message - nn.Linear with double argument,stale/low urgency,2,stale/low urgency,2,0,"77415, 95590",,,,1,a061b7115c552454c97488421ec8dc60c402634d63e51c4e8ab2b57781b8e0c8
thread:issue:78071:issue:85834,issue #78071: [primTorch] item prim can't return a bool properly,stale/low urgency,2,stale/low urgency,2,0,"78071, 85834",,,,1,d272a5435d8f3d47064a58ba94b613e8802955c323c77e5e61079dce2563db8d
thread:issue:78681:issue:79337,"issue #78681: Unable to install Preview (Nightly) on M1 macOS: ""Symbol not found""",stale/low urgency,2,stale/low urgency,2,0,"78681, 79337",,,,1,41ec90e6d5ebd74ea966353d819154a4561d88cbaf2f7af8298f07c0e51a3539
thread:issue:78878:issue:80208,issue #78878: Improving clarity in the docs of different losses,stale/low urgency,2,stale/low urgency,2,0,"78878, 80208",,,,1,0bea69fd5c781a354211082745946b51922975fc3f6cc4249d0739bc6ffa8366
thread:issue:79039:issue:83773,issue #79039: Add Autograd Support for Nested Tensor,stale/low urgency,2,stale/low urgency,2,0,"79039, 83773",,,,1,ef8ff514f720fd7239603057b47e426ce072eab37754c77e0df81568bbc82b3c
thread:issue:80187:issue:80966,issue #80187: [NVFuser] Investigate modules with bad performance relative to eager,stale/low urgency,2,stale/low urgency,2,0,"80187, 80966",,,,2,442057098b2c2794ce99b431934815afea2f123da2dccb7adcb387d6da6bbb3f
thread:issue:80549:issue:81104,issue #80549: Allow parameterization of Layouts,stale/low urgency,2,stale/low urgency,2,0,"80549, 81104",,,,1,2f4ff4f1633dfeeac5a6d4d5393ae1234dd548b8860bf020507a561f7ad7cfcd
thread:issue:81692:issue:90547,issue #81692: Failed to static link latest cuDNN while compiling,stale/low urgency,2,stale/low urgency,2,0,"81692, 90547",,,,1,30363b5209501c4325edfb3bb31d8af5728b03ee88705ed53c19041f5c7012e2
thread:issue:82139:issue:82443,issue #82139: Re-enable DynamicQuantModule in iOS simulator tests,stale/low urgency,2,stale/low urgency,2,0,"82139, 82443",,,,3,dc448afa20bab04fd1b036c68ecc6cb5307dde73bcaeb7e2cc5187c5c65d8e4a
thread:issue:82886:issue:97210,issue #82886: CUDA graph capturing fails for nn.Embedding and large batch sizes,stale/low urgency,2,stale/low urgency,2,0,"82886, 97210",,,,1,cb751acabbd5df8365270cff72325633cd22ecf5dda54712558e568324df0266
thread:issue:82926:issue:86890,"issue #82926: Slice operation on ""ragged"" dimension in NestedTensor",stale/low urgency,2,stale/low urgency,2,0,"82926, 86890",,,,1,df61c44c8ec8e6ab00c3c06f059595f7731d1b79fbcd1e85920ab3fcc44a56b1
thread:issue:84625:issue:90663,issue #84625: Beta distribution behaves incorrectly for small parameters,stale/low urgency,2,stale/low urgency,2,0,"84625, 90663",,,,1,44dd7b91e04178bedfce4299775af42681e8db9dde9b3314c20d54c7794f582f
thread:issue:88658:issue:95108,"issue #88658: nn.Linear allocate too many space which lead to CPUAllocator ""allocate memory failure"" if it's BF16. good for FP32.",stale/low urgency,2,stale/low urgency,2,0,"88658, 95108",,,,1,0fdd5b260be40be8f648eb41aba4cd9f2fcf4241404fe20382b27e3db7ebc6f5
thread:issue:89105:issue:89601,issue #89105: Bad string in GLSL shader,stale/low urgency,2,stale/low urgency,2,0,"89105, 89601",,,,1,62bdcb1583a8ae3c5508183ae45ef4ebf94537be7571687abbb1365693068c69
thread:issue:89491:issue:91524,issue #89491: build: cmake: functorch.so not installed at expected location,stale/low urgency,2,stale/low urgency,2,0,"89491, 91524",,,,1,c2914b9ac28eda1dad0566a6df9993df4ff20b3fb34590a4c67c4620f12ea71b
thread:issue:91156:issue:91374,issue #91156: `quantile` fails for `float16`/`half` inputs,stale/low urgency,2,stale/low urgency,2,0,"91156, 91374",,,,1,5f4918bf927dffa70d02b53763f20284d69dfac7c5bb8ead025975bd8843c89b
thread:issue:9204:issue:9310,issue #9204: /usr/bin/ld: cannot find -lpthreads,stale/low urgency,2,stale/low urgency,2,0,"9204, 9310",,,,1,b1ced5f45f1a125535e33c6f721e54361e32a9d1e2ccc2db253aa855b841f84f
thread:issue:98073:issue:98406,issue #98073: Add PrivateUse1 folder in aten/src/ATen,stale/low urgency,2,stale/low urgency,2,0,"98073, 98406",,,,1,d8121e5ec174ff26ab84df5a575725033c59af58900b2802fcb2fe9b04791a83
thread:issue:98861:issue:98947,issue #98861: Sparse Tensor not working for `torch.cat`,stale/low urgency,2,stale/low urgency,2,0,"98861, 98947",,,,1,d08af6c6f0bb4e625bc76f81d92c16f38bc0587e2c7de17a59aa23ad9d8983f8
thread:issue:99265:issue:99351,issue #99265: the error message of torch.addcmul is wrong,stale/low urgency,2,stale/low urgency,2,0,"99265, 99351",,,,1,cd9a2e23ec9cd13edc18b8b04e552e62f92d162fec3acffee572735e0a7752b7
thread:issue:100012,issue #100012: Dataloader multiprocess loading with num_worker > 0 calls __main__ file to run,stale/low urgency,2,stale/low urgency,1,0,100012,,,,0,e7632180292432807f5d3f1fee12d9a2dd51949bb3664cdc6f87c5cf25bcef59
thread:issue:100051,issue #100051: Issue of HistogramObserver to handle abnormal value,stale/low urgency,2,stale/low urgency,1,0,100051,,,,0,b8ed2f5436bfeb4ea012ded90325eb15d1f615e0cf8ee152a095e70a2967695d
thread:issue:100054,issue #100054: Add compile option -Werror=return-type compile error,stale/low urgency,2,stale/low urgency,1,0,100054,,,,0,45e91ddcb862f06d11f810dfebf9f6648d51d1f99fb90a9827716b8bea98f0d3
thread:issue:10006,issue #10006: RNN gradients in eval mode in pytorch 0.4,stale/low urgency,2,stale/low urgency,1,0,10006,,,,0,08f681d3f6d915f612a44b7b454a3b157a793f08ee9c6d315bf21fa5b66f9928
thread:issue:100062,issue #100062: add github check that diffs generated code,stale/low urgency,2,stale/low urgency,1,0,100062,,,,0,c225f0b396b9ed8b5cc467d94a921112b963690321b6b95ff59032b662fe94a7
thread:issue:100087,issue #100087: TransformerEncoderLayer behavior inconsistent between training and evaluation mode,stale/low urgency,2,stale/low urgency,1,0,100087,,,,0,30ed2ba3710f07d372f90b54ab2431f11736dff410af48aad6ced5a63bd1f054
thread:issue:100123,issue #100123: Error building Pytorch from source,stale/low urgency,2,stale/low urgency,1,0,100123,,,,0,ce6932975b85c3838b9c6a161f1a5d305e541b396d6d82c94e7d70c2f1ec2307
thread:issue:100145,issue #100145: This flag not work : torch.backends.cudnn.allow_tf32 = False,stale/low urgency,2,stale/low urgency,1,0,100145,,,,0,b309e23b8970a564787c70a381f3c427b53a793615e5cc8c97f0833d3f5c596a
thread:issue:100158,issue #100158: `torch.sparse_csc_tensor` matrix multiplication produces MKL error SPARSE_STATUS_ALLOC_FAILED when density is too high,stale/low urgency,2,stale/low urgency,1,0,100158,,,,0,961f5e955040740a42cffc1b49582277a92757eb3750a43bacb45c0964c26a38
thread:issue:100241,issue #100241: Interaction of torch.no_grad and torch.autocast context managers with torch.compile,stale/low urgency,2,stale/low urgency,1,0,100241,,,,0,f0aaaed031ab629d781ae63d4d7ddb6d710cdb7bcc36b23bd2b84d20309a3981
thread:issue:100244,issue #100244: torch.utils._content_store will deduplicate storage with identical contents; may be problematic for mutation,stale/low urgency,2,stale/low urgency,1,0,100244,,,,0,c661e48d8f3a6a25a7f36f671e968fd8fd9e10cd72335b5052d32a7fe7871d0d
thread:issue:100334,issue #100334: MPS device inference all same value,stale/low urgency,2,stale/low urgency,1,0,100334,,,,0,d564c6c864487661e06a0f387867607cad7d38746f7b8c827094e1f700856c62
thread:issue:100366,issue #100366: some of the enteries in the previous version of pytorch section are invalid,stale/low urgency,2,stale/low urgency,1,0,100366,,,,0,c054a18c74fbe5cb24648086b4c0fee21ff5946c1faccffe0bb9781aa06d7136
thread:issue:100411,issue #100411: Sparse Matrix nnz Overflow when casting from COO to CSR,stale/low urgency,2,stale/low urgency,1,0,100411,,,,0,929da418aedf859362c53d3b0d2e64521b3475cc052ce414ffa4578ae36ee95b
thread:issue:100419,issue #100419: Subgraph rewriter: Unable to match constant args,stale/low urgency,2,stale/low urgency,1,0,100419,,,,0,caca804717bef22e11f0f82980ffc59bf6ea48bdaeba20640544d4421339d948
thread:issue:100425,issue #100425: Higher GPU consumption for Lenet-5 and LSTM models when compared to other frameworks,stale/low urgency,2,stale/low urgency,1,0,100425,,,,0,bd32b38e0ff1ae3c7de27e1057bf2e4c148b4027f60a7788dbacd4e9302c9a43
thread:issue:100584,issue #100584: [doc] torch.scalar_tensor doc is missing,stale/low urgency,2,stale/low urgency,1,0,100584,,,,0,2c242fe2f526a88140dc60859fd2bd8cad4ffb479ba5f8a938cdb47d3ebfde13
thread:issue:100617,issue #100617: GPU VRAM usage significantly higher for Lenet5 models when compared to other frameworks,stale/low urgency,2,stale/low urgency,1,0,100617,,,,0,8f93c4932a3f70e3d899ecf45a3877c8498d0b404c403f0de89119808c524095
thread:issue:100637,issue #100637: Optimal Batch Size Selection in Torchdynamo Benchmarks for Different GPU Memory Sizes,stale/low urgency,2,stale/low urgency,1,0,100637,,,,0,df2120e4d2109b84078ae19eb30a3eecddcf23125c80a1553f2724b4f8816940
thread:issue:100656,"issue #100656: On UMA systems, pytorch fails to reserve memory exceeding the initial memory size",stale/low urgency,2,stale/low urgency,1,0,100656,,,,0,3e21cfeb5d9290866c636bac9785913237b169ee2ab6e471a78742fd271cdf9c
thread:issue:10070,issue #10070: UnicodeDecodeError while loading caffe2 model,stale/low urgency,2,stale/low urgency,1,0,10070,,,,0,aadd9f77e92681ec8dc118a0d1e3bdf86217b2dc4eeede35253ca863de029257
thread:issue:100705,issue #100705: torch.cuda.amp.GradScaler initialization,stale/low urgency,2,stale/low urgency,1,0,100705,,,,0,148cc5b22dc26ce63783969d482dd85c4b944966c5d106f4c0eb10aeb24ce849
thread:issue:100785,issue #100785: compile torch2.0 in debug mode,stale/low urgency,2,stale/low urgency,1,0,100785,,,,0,aa96c040e30c679c5dfdf5362b6204a02479a8d280c8fcf20219287b915b0f56
thread:issue:100795,issue #100795: Quickstart notebook fails to train properly with ROCm,stale/low urgency,2,stale/low urgency,1,0,100795,,,,0,9025b740977e840b5deb592ed9c5522bbd2628c5e04e279f6321ac98c7a0739a
thread:issue:100801,issue #100801: There is a performance drop because we have not yet implemented the batching rule for aten::native_dropout_backward,stale/low urgency,2,stale/low urgency,1,0,100801,,,,0,6fafa2540eb5cd8805f0eff7939194a054d79754fc382c0b09f76fcdd6c19273
thread:issue:100807,issue #100807: [torch.compile] returns output with WRONG SHAPE after `cat_slice_cat`,stale/low urgency,2,stale/low urgency,1,0,100807,,,,0,f784ec8a330afdeab4911cfcc5af4626c0c2a65fe4182c5325fe32653539ab5f
thread:issue:100850,issue #100850: [BUG] Poor torch.bmm performance on H100,stale/low urgency,2,stale/low urgency,1,0,100850,,,,0,6d303c28b97ce7344e0dd287f6dc1c81388d6a937398654d8def7156bb218ca3
thread:issue:10088,issue #10088: [Caffe2] Implementing CoordConv layer,stale/low urgency,2,stale/low urgency,1,0,10088,,,,0,75bcd3b9dafb3155b6f7f6a42fe19562e4899dad9dba64f9828a5660ada86666
thread:issue:100914,issue #100914: [MPS] Track failures of test_module.py for MPS backend,stale/low urgency,2,stale/low urgency,1,0,100914,,,,0,fa10b25b87c934aab546b556b1c565ea4e7310d45c72373985821358dcfc9920
thread:issue:100932,issue #100932: torch.concat fails with float16 input in autocast(device_type=cpu) context,stale/low urgency,2,stale/low urgency,1,0,100932,,,,0,5faaf4b30dc19cf52cfe70b10b9e9d82e908c50d57f43f2ef323dbe921912d5a
thread:issue:100957,issue #100957: PyTorch installs the file mkldnn.cmake that looks for the package MKLDNN that doesn't exist,stale/low urgency,2,stale/low urgency,1,0,100957,,,,0,d45453d51fcb551e85321e04bafce0756d07ce39e2bdcac10a5d621588988581
thread:issue:101069,"issue #101069: torch.autograd.detect_anomaly should report the original forward trace as part of the error, rather than as out of band warning",stale/low urgency,2,stale/low urgency,1,0,101069,,,,0,6e01bbaf0e046c3cd85e5acce4b4942f00f73761b6127025e464a142181e9c7d
thread:issue:101070,issue #101070: Lazily format C++ stack trace if it is not used,stale/low urgency,2,stale/low urgency,1,0,101070,,,,0,cf360fefda25758ea95dd358f563e21a974be28cb80825437e40beeb48740c74
thread:issue:101073,issue #101073: Windows fatal exception: stack overflow while using pytorch for computing,stale/low urgency,2,stale/low urgency,1,0,101073,,,,0,ce58c3c52b0cee2182f7364bf5a1e9e3cec3afc8cb18e883be50bba69aef4adb
thread:issue:101080,issue #101080: Sequence annotation in type hints is wrong,stale/low urgency,2,stale/low urgency,1,0,101080,,,,0,df0c00e93cf008d05effa2ac1882e761f67e646ca37de84569f3f94e463178ea
thread:issue:101106,"issue #101106: addmv doesn't do type promotion correctly,",stale/low urgency,2,stale/low urgency,1,0,101106,,,,0,eec99dea8cc45580f11193f1792a35de52ba8943d48e72a9149f3c806bfaf4a9
thread:issue:101107,issue #101107: Make compiled models serializable,stale/low urgency,2,stale/low urgency,1,0,101107,,,,0,52e222555b8e61c2c7bd35088bd09f66ca2a7c19043566e3efd84c5c9c132231
thread:issue:101117,issue #101117: Barriers to using torch.compile directly in PyTorch library code,stale/low urgency,2,stale/low urgency,1,0,101117,,,,0,0491d16c3c6e10d305a994021c2b57438c3b034c4d000ecfe9fbbaff595780a2
thread:issue:101167,issue #101167: round float16 calculation error in mps backend,stale/low urgency,2,stale/low urgency,1,0,101167,,,,0,e3d6dbba9274d3325cc86cc7e277b5cb3d81c55a4104254550c2833d9d2e2590
thread:issue:101177,issue #101177: version 4.26.1 to 4.29.0 has two bugs,stale/low urgency,2,stale/low urgency,1,0,101177,,,,0,381aa623a9d811b7e3763be0c27a8dfa3c4e055dd1f000866ff7f2fa619993ab
thread:issue:101185,issue #101185: Pure virtual function call exception on Python interpreter exit when using debug wheel,stale/low urgency,2,stale/low urgency,1,0,101185,,,,0,c110ec28bda6ae655ba698d68bc996dbb59e4b36c66689f2f11db36fdec8ce2f
thread:issue:101188,issue #101188: Weird dataloader performance degradation caused by torch and numpy import order,stale/low urgency,2,stale/low urgency,1,0,101188,,,,0,108f61b8e9235d934627250d38baa4fe656db237e8b62f204dc6a9f15f9627bd
thread:issue:101189,issue #101189: `dense -> sparse compressed` to work with empty batches.,stale/low urgency,2,stale/low urgency,1,0,101189,,,,0,3c81ca0e4bd7b3c8ad1812632713c33d3b22051145cf6356c2a847d27f8ab652
thread:issue:101191,issue #101191: custom_op API follow-ups,stale/low urgency,2,stale/low urgency,1,0,101191,,,,0,a6c2dea38d2ffcb8b9047d529e9033ad4d0ae63a16813aba9edbcdd9db561248
thread:issue:101210,issue #101210: GLCM implementation in pytorch C++ api and cuda,stale/low urgency,2,stale/low urgency,1,0,101210,,,,0,eab0d66bf8ea0722c5e3598b321fbd2a961e1521dce88f7460a71dc7454afed2
thread:issue:101241,issue #101241: Different results with vmap when using torch.jit.script,stale/low urgency,2,stale/low urgency,1,0,101241,,,,0,2733be336e0bc75a27cb1122a1b4e3f84aac3e97a983dd0ea121197ce09e41b7
thread:issue:101246,issue #101246: Tool for identifying where in eager model an operation is nondeterministic,stale/low urgency,2,stale/low urgency,1,0,101246,,,,0,b3cb1f9cc47afd046585a3e801551bf30bc6104c9363d5fc00e08b6526c229c1
thread:issue:101251,issue #101251: Mac m2 MPSNDArray.mm:78: failed assertion `[MPSNDArrayDescriptor sliceDimension:withSubrange:] error: dimension index (2) not within number of dimensions (2) Dimension indices are 0-based',stale/low urgency,2,stale/low urgency,1,0,101251,,,,0,9b4056e44f9ec299e794ca2f82095ad84f22730f6f7ba8c2b3e376e01b28d053
thread:issue:101273,issue #101273: IPEX as TorchDynamo Backend Performance Dashboard,stale/low urgency,2,stale/low urgency,1,0,101273,,,,0,36d2d676fdac1c611b33b44eaa49ae80e0063b56d7e2d2b92eb1526580b3234a
thread:issue:101291,issue #101291: fused torch.optim.AdamW isn't faster than the unfused version,stale/low urgency,2,stale/low urgency,1,0,101291,,,,0,efdf90c44f8123c6bd59e7575e2b9fbdff75bde40cda5f57c0ace0580b875ba2
thread:issue:101293,issue #101293: Operations to shared tensors in the forked process could lead to silent crash,stale/low urgency,2,stale/low urgency,1,0,101293,,,,0,630392d8fca3a1963a2b543a9a7b90f71b392284a500d1e8824dbf42045264c3
thread:issue:101319,issue #101319: Deprecated File bug,stale/low urgency,2,stale/low urgency,1,0,101319,,,,0,a5a1846649d091e71b129b585a064c00b29dbecde366ec957b17bc1eff79a318
thread:issue:101334,issue #101334: Functions for Calculating Skewness and Kurtosis,stale/low urgency,2,stale/low urgency,1,0,101334,,,,0,6a63c32dfbd565f9b67ba768882357cfa9404b3d54618f2c494f80ae22791f85
thread:issue:101335,issue #101335: fsdp training with the seq2seqTranier module gets stuck during evaluation.,stale/low urgency,2,stale/low urgency,1,0,101335,,,,0,b9e414f40ef81540ca37b677885ad38e7c07aead8a9bbb2e39f36da4259da1c0
thread:issue:101369,issue #101369: Theme update,stale/low urgency,2,stale/low urgency,1,0,101369,,,,0,f8f6a4ba79ec9a9babf515bddb0d8a62a6170e02439607df269213d7138894f9
thread:issue:101370,issue #101370: SparseAdam: working with dense parameters but sparse gradients - usecase,stale/low urgency,2,stale/low urgency,1,0,101370,,,,0,6b3239a763bba4c326c4f22825ad6abef814fa2edfd3fe099c1c8a9019c8c7b1
thread:issue:101385,issue #101385: torch.Tensor.is_sparse returns false for non-COO sparse tensors,stale/low urgency,2,stale/low urgency,1,0,101385,,,,0,f2f9eda7c8dbf080553aaddf86e0f50457126e3642230c7d94b13e97e24b2b77
thread:issue:101407,issue #101407: Delete old vmap prototype,stale/low urgency,2,stale/low urgency,1,0,101407,,,,0,183942e51395b0ad24411114be53e45457316dbb07610a28314037304ed72629
thread:issue:101415,issue #101415: import functorch.dim monkeypatches torch,stale/low urgency,2,stale/low urgency,1,0,101415,,,,0,c83f6bab6eab3d01511eca78db399812aaa759df4199345e2e7bd668d259a18a
thread:issue:101428,issue #101428: 2D inputs to linear layers run up to 25% slower than 4D ones on some Nvidia GPUs,stale/low urgency,2,stale/low urgency,1,0,101428,,,,0,0e5fae1ba7a9b94a357b5a08602fb197c3e9e4ee596a42b2d22a7b8018ecc5aa
thread:issue:101470,issue #101470: [bazel] add inductor to bazel build,stale/low urgency,2,stale/low urgency,1,0,101470,,,,0,ccc71fe17d9fc95095a2059ccbc9cfbe80667be6f36e82a469d4c373607c49c9
thread:issue:101505,issue #101505: Investigate random sequence number broadcast initially incorrect,stale/low urgency,2,stale/low urgency,1,0,101505,,,,0,00fa8e151f799705038ad012700307872603c49f25708ff7f0b30fc819a2e517
thread:issue:101506,issue #101506: Support pipeline parallelism with PyG,stale/low urgency,2,stale/low urgency,1,0,101506,,,,0,4b82dd8214c8d38a1fd65fe219012038cf01c375ac5db8e85962634f661252f7
thread:issue:101536,"issue #101536: torch.cuda.set_device cannot use to set cpu device, but give an ambiguity hint",stale/low urgency,2,stale/low urgency,1,0,101536,,,,0,4b0b11f783676fac68427483ccb980213a35d5270742612bf06cf656dc4a4850
thread:issue:101564,issue #101564: not yet implemented the batching rule for torchaudio::_lfilter,stale/low urgency,2,stale/low urgency,1,0,101564,,,,0,92d52125dda05bd246304dec97db2dee41a3a9da25f970fa3c6fc89b54beda9c
thread:issue:101586,"issue #101586: /Users/davidlaxer/pytorch/third_party/tensorpipe/third_party/libuv/src/unix/getaddrinfo.c:165:10: error: implicit declaration of function 'uv__idna_toascii' [-Werror,-Wimplicit-function-declaration] rc = uv__idna_toascii(hostname,",stale/low urgency,2,stale/low urgency,1,0,101586,,,,0,c168bb5bd520442456fb7001ed9bd31060526b72bf459666fb87dc99c1dde8b8
thread:issue:10161,issue #10161: error when building caffe2 from source with gcc 7 & cuda gcc 5.5,stale/low urgency,2,stale/low urgency,1,0,10161,,,,0,b26458bff50805bcd7bfab0b960d48b9b176865c3f96cac0b42cee28de3171d7
thread:issue:10165,issue #10165: torch.utils.data.random_split() returns dataset index as tensor,stale/low urgency,2,stale/low urgency,1,0,10165,,,,0,8c141c4712ad18daacf2acde5f6dbbe107563bc4962276612ef6f0ec498320d6
thread:issue:101653,issue #101653: Unexpected behavior comparing uint8 tensor to value greater than 255,stale/low urgency,2,stale/low urgency,1,0,101653,,,,0,0b8c79f0f64dd03b1ae8550e1f4f172d91117d91fadb11d469976bd7f502db70
thread:issue:101666,issue #101666: Unexpected behavior of fmod op in some float32 input,stale/low urgency,2,stale/low urgency,1,0,101666,,,,0,746c6bb75a3ec3992c9aef7d411b3c4740950026889bc611fde79e375a8d08d5
thread:issue:101669,issue #101669: [refs] inplace references resize the input to match the broadcasted input shape,stale/low urgency,2,stale/low urgency,1,0,101669,,,,0,489cd8bf1011706159e15e533a054f78bb3b3d2813f0c7ad265fd62bbf067fea
thread:issue:10170,issue #10170: build error,stale/low urgency,2,stale/low urgency,1,0,10170,,,,0,f7d5f42a2cbb33716527e2c16f16f4c992a4606046cfb787a33db80e3caa6e64
thread:issue:101716,"issue #101716: [TorchScript] aten::__and__ with argument types: Tensor, bool not supported",stale/low urgency,2,stale/low urgency,1,0,101716,,,,0,d5fe968eb6323b42e3c33c41d4e6c903b393b28977db1852c5a131dfe83fba62
thread:issue:10172,"issue #10172: [Feature request] Batch eig/symeig functions (for small matrices, with CUDA)",stale/low urgency,2,stale/low urgency,1,0,10172,,,,0,94f3c3a1c9d7feef891988e14fa4dc8455534b6b3bba6f8d9eaa1bdc0da25606
thread:issue:10176,issue #10176: [Caffe2] Modelling using Ops instead of Helper Functions,stale/low urgency,2,stale/low urgency,1,0,10176,,,,0,67504be7eb6e777d3904a8915f8f7a9ec8f88a9377f4f92f68a1389c52f5980c
thread:issue:101781,issue #101781: Add support for bfloat16 in torch.from_numpy(),stale/low urgency,2,stale/low urgency,1,0,101781,,,,0,e00d2399d89aae5d70997369631ba619320fea35919bcfc83d1269dc344fb48b
thread:issue:101798,issue #101798: Fix absolute links in pytorch repository and allow it to be proxied,stale/low urgency,2,stale/low urgency,1,0,101798,,,,0,b5c60ff0e90407a44e8d2c8bd5d50773b764a57942758b95c1cec09f2bf0d760
thread:issue:101871,issue #101871: Can group convolution support other grouping methods?,stale/low urgency,2,stale/low urgency,1,0,101871,,,,0,31b926ac06f8bba1b39fb08557a829cebe9221ac4f80682f4f8fca2d95cb11d9
thread:issue:101878,issue #101878: torch.quantile on MPS doesn't sort values when dim is not None,stale/low urgency,2,stale/low urgency,1,0,101878,,,,0,95b751bbe6d6a3943d0f0a4ec16b288a202b4b5b45f0b5bc9d474d61798158bb
thread:issue:101890,issue #101890: Can't reproduce/non-deterministic results with CUDA,stale/low urgency,2,stale/low urgency,1,0,101890,,,,0,f9e808c9e4b72771798b752229a05937880ca68eeebd47158ac51e1be60bcf19
thread:issue:101930,issue #101930: [feature request] [minor] Inplace torch.flip_,stale/low urgency,2,stale/low urgency,1,0,101930,,,,0,083c3bedefa149e5724bd9d24b67913bec03ea904a1b82f73876bdfa17c7dd6e
thread:issue:101934,issue #101934: Parameter gradient is not moved parameter is moved across devices,stale/low urgency,2,stale/low urgency,1,0,101934,,,,0,758ef975aff01b10f7793411647f8c4e055669187d479aaf0508636970190f49
thread:issue:101968,issue #101968: CrossEntropyLoss output difference on Windows vs. Linux,stale/low urgency,2,stale/low urgency,1,0,101968,,,,0,c848f64e810aeccbd6d37dd90abecc4a428656d0e9a197ea45da463cb76d2f0b
thread:issue:101972,issue #101972: No pytorch_android 2.0.x builds,stale/low urgency,2,stale/low urgency,1,0,101972,,,,0,5a7b7e1022b11ca6cac1cc224ddd8656ebfb3fa62455346494e9fd0fa2b2383c
thread:issue:101999,issue #101999: [FSDP] Ensure full precision checkpoints,stale/low urgency,2,stale/low urgency,1,0,101999,,,,0,2cb0cae6d2e3b2381d00348289ab9d2a8ab1d3b1830cc013720bdbee3af2b40c
thread:issue:102063,issue #102063: [inlined-inbuilt-nn-modules][dynamo][BE] Revisit call_method of NNModuleVariable,stale/low urgency,2,stale/low urgency,1,0,102063,,,,0,0c17a36269c71af0a7b5e10babda81a1d72e4b24879508cd85b32b0caf5b6297
thread:issue:102081,issue #102081: CPU Fallback does not convert Tensor?[],stale/low urgency,2,stale/low urgency,1,0,102081,,,,0,e7b04145bd0104bbb366f2d7cd0c9ac182ce24f56a45c018b3366423ad5e8c5e
thread:issue:102085,issue #102085: BatchNorm can't be symbolically traced with torch.fx as a standalone module,stale/low urgency,2,stale/low urgency,1,0,102085,,,,0,166a1aafd911afa386fb4d69548230bd5d121fef97685cb2c500be2200007f18
thread:issue:102109,issue #102109: Can't vmap over a slice expression,stale/low urgency,2,stale/low urgency,1,0,102109,,,,0,0ede98d07917a3fff2205c77f44e67ed9d3415313319b0875681b7de134ea986
thread:issue:102142,issue #102142: Pytorch CXX11 ABI version,stale/low urgency,2,stale/low urgency,1,0,102142,,,,0,c032aed6539ccef64cde26ea7e94181b931c562480e048b64f47339b511f8f13
thread:issue:10215,issue #10215: [Caffe2] Build failure on Ubuntu 16.04,stale/low urgency,2,stale/low urgency,1,0,10215,,,,0,27a284f51a5f25ab0bcc64aecf9b462183b2228115109b2e5a7726d2863b1945
thread:issue:102162,issue #102162: NotImplementedError in backprop on on dense-sparse matrices,stale/low urgency,2,stale/low urgency,1,0,102162,,,,0,25e4831c261e881e429fe930d87765908a8ce2ea95e8b7fd16011196c807d2db
thread:issue:102227,issue #102227: lintrunner should fail on badly formatted docstrings,stale/low urgency,2,stale/low urgency,1,0,102227,,,,0,cbb6c4a8ca7972e17ba2ff87d412e449bbdff37a838aaa7c17e07c3e5fc23c98
thread:issue:102272,"issue #102272: after add /path_to_libtorch/libtorch/lib to LD_LIBRARY_PATH, I can't import torch_scatter.",stale/low urgency,2,stale/low urgency,1,0,102272,,,,0,b6168bf0082e3ba2de0a7fae319297f87631473f0e89b644d1367b4a43a5b5be
thread:issue:102319,issue #102319: Re-enable Low Memory Dropout,stale/low urgency,2,stale/low urgency,1,0,102319,,,,0,3d6ac20a0e96b462f97ed59c1bf6892152f0108de5e80c9fc70047e280608e09
thread:issue:10232,issue #10232: [Caffe2] AVX not enabled for pytorch/caffe2,stale/low urgency,2,stale/low urgency,1,0,10232,,,,0,51cd92e1dd1c320d620b1e6b203619a39781a3e56f74af0a966f05f9ea52f41a
thread:issue:102333,"issue #102333: transformer encoder-layer, the sample-Independent attn_mask(dim=3) has different behaviors when training and validating",stale/low urgency,2,stale/low urgency,1,0,102333,,,,0,32c6085f13012e4fdbd74232ac4ff5cb9f112dee8a9ee30e7c3673303e082937
thread:issue:102334,issue #102334: There is a memory leak in torch.load,stale/low urgency,2,stale/low urgency,1,0,102334,,,,0,80344a8bba6614fe704515de69e0e095b3305335ee4a6fd69842454521ddb622
thread:issue:102360,issue #102360: PyTorch hangs at import when used together with TensorFlow,stale/low urgency,2,stale/low urgency,1,0,102360,,,,0,997eefc4ab1f872f7218da917b5be27faf5aa12fd29259a01493684aec8abc37
thread:issue:102368,issue #102368: Issue with ShufflerIterDataPipe in torch 1.13.1,stale/low urgency,2,stale/low urgency,1,0,102368,,,,0,c5a0a75c4d630fcd492e2a6af3c2938718f76cd546393a5e72cb473f0c053e90
thread:issue:102374,issue #102374: Hooks not working in version 2.0.1+cu118,stale/low urgency,2,stale/low urgency,1,0,102374,,,,0,0a05b178dd8fb1b1a43922b3864c6849537b257fb0b341fbbf2c4b04034edf13
thread:issue:102400,issue #102400: Unexpected Behavior when using torch.isclose(),stale/low urgency,2,stale/low urgency,1,0,102400,,,,0,edf52f2b9b3f50fd281dab998ee890ca49c18bb926714d7fa94e30d7e92e6d50
thread:issue:102423,issue #102423: Add default device function interface for device-aware apis,stale/low urgency,2,stale/low urgency,1,0,102423,,,,0,63b57e7e4d22fac460f4e22d4269eb54a0bcff1bf95f1011b710d9961bdb96e5
thread:issue:102435,issue #102435: torch.distributed.all_reduce() has inconsistent behavior,stale/low urgency,2,stale/low urgency,1,0,102435,,,,0,bc5c8218242bd8b961944c0ae2bbe7ab46a739b55076aaed59ca696f95d1cd3c
thread:issue:102437,issue #102437: discuss.pytorch.org signup issue,stale/low urgency,2,stale/low urgency,1,0,102437,,,,0,76a3329d865daa61e54152a814fee1c0eca769773e49d203bda2dd7c1f69fbb3
thread:issue:102459,issue #102459: Matrix multiplication performance regression in case of an additional dimension of size 1,stale/low urgency,2,stale/low urgency,1,0,102459,,,,0,2e7a54a01929348b7403171a7943e9089f70e011afdc43fb4e8db2a37fff6096
thread:issue:102501,issue #102501: [Utils][tensorboard]Enhancement: Include 'max_outputs' parameter in torch.utils.tensorboard.summary's 'image' method,stale/low urgency,2,stale/low urgency,1,0,102501,,,,0,7e0f6aa4583718f94ebf02de183dc6766e7172e883a5233259a9c51c00943651
thread:issue:102511,issue #102511: Faster BatchSampler with big batch size,stale/low urgency,2,stale/low urgency,1,0,102511,,,,0,27d6dc7c6c60eec5e4742cd3e0a266c306f3f450563394f6f30184e530f154fa
thread:issue:102515,issue #102515: pytorch java api documentation is not clear and does not cover example,stale/low urgency,2,stale/low urgency,1,0,102515,,,,0,05d691d1e2594fb1d70ecade4beae5e8d4c97fcf19025eb8195f3e0e7310c2b4
thread:issue:102556,issue #102556: test_functional_autograd_benchmark.py::TestFunctionalAutogradBenchmark::test_fast_tasks passes with all NaNs,stale/low urgency,2,stale/low urgency,1,0,102556,,,,0,f90f55870cb031d4fba7ad0cc41b6efd2018e0bae844f5bf9df6cf5b5aab556f
thread:issue:102559,issue #102559: Mergebot should merge non-stacked PR,stale/low urgency,2,stale/low urgency,1,0,102559,,,,0,c1967a6612cc7ea3952d9fe2c746e803e09baab339d713e3c97cf406dc0a1d63
thread:issue:10256,issue #10256: [Caffe 2] How to access/remove Blobs from Workspace?,stale/low urgency,2,stale/low urgency,1,0,10256,,,,0,2e0ba9d614ddb8e17fc67f3cd0ef9a21b223860a494e95d43bfa4c2c2851e3d5
thread:issue:102600,issue #102600: Support for activation checkpoint on demand in custom function,stale/low urgency,2,stale/low urgency,1,0,102600,,,,0,97abcb1c64881820ae6bbf0722ddb8c361b065b114d93079e188257b0a6af2f0
thread:issue:102609,issue #102609: [prims] torch.ops.aten.le decomposition confuses scalars and tensors,stale/low urgency,2,stale/low urgency,1,0,102609,,,,0,3b77dbd8649612ff3b9db8ece204850544ffacbf12c60a806b8faceb24697477
thread:issue:102610,issue #102610: Cannot invoke prims.sum with output_dtype,stale/low urgency,2,stale/low urgency,1,0,102610,,,,0,82323d62a5c88402fb238789f8c45639601fbc183903a81b6d2593d90954229c
thread:issue:102732,issue #102732: Building NCCL with `make -l $MAX_JOBS` slows down builds,stale/low urgency,2,stale/low urgency,1,0,102732,,,,0,c6c548f969ae2ca8d447a9996bc8c5fb3f70029063223c2cfe855173441cf2ba
thread:issue:102740,issue #102740: parameterizations.orthogonal does not work as intended with nn.GRU or nn.LSTM,stale/low urgency,2,stale/low urgency,1,0,102740,,,,0,2d120d7640a07aff771b182f504e186644a29e843c7a76e474769500221c0244
thread:issue:102741,issue #102741: Unlocking PyTorch's Power: README.md in Multiple Languages!,stale/low urgency,2,stale/low urgency,1,0,102741,,,,0,fc3e1b793557eeff9c8690824879cc3f7c28b267566244f034423fafb2618d99
thread:issue:102753,issue #102753: Error: no matching constructor for initialization of 'at::OptionalIntArrayRef',stale/low urgency,2,stale/low urgency,1,0,102753,,,,0,d1237b8aa7208d5f97fc67560943c728dcf07d820c2b7aabd50e64820e51c5c4
thread:issue:102761,issue #102761: Pytorch Build images for RISCV64 Devices in the nightly builds,stale/low urgency,2,stale/low urgency,1,0,102761,,,,0,ae98b5191530c252fe75e132f0c0cb1a18a97fccf135a748af33d68d732336fd
thread:issue:102772,issue #102772: Support In-place Triangular Matrix Multiplication,stale/low urgency,2,stale/low urgency,1,0,102772,,,,0,8410a45d40d190caf7848c360ac316ef85255c9cc1ebb05d0d7969a892c818b6
thread:issue:102811,issue #102811: `torch.poisson(torch.tensor([torch.inf))` returns 0,stale/low urgency,2,stale/low urgency,1,0,102811,,,,0,88264778ec11afe77379329df5e2a89559e8c92d7f79a536c15f0acd71f0dca3
thread:issue:102812,issue #102812: [DTensor] Error in distribute_module with module._apply,stale/low urgency,2,stale/low urgency,1,0,102812,,,,0,563d76af46544039e9900a7560e3acbdc44b62d320ebf336f2ab5574e4c7d23e
thread:issue:102830,issue #102830: Unknow error when using `make_graphed_callables`,stale/low urgency,2,stale/low urgency,1,0,102830,,,,0,bdf6279f7261b01d80206cdc7454d62abf0cd8dd8ddc5775790b0e7be5574347
thread:issue:102833,issue #102833: LibTorch-Lite 1.13.0.1 Crash on iOS 12 on app startup,stale/low urgency,2,stale/low urgency,1,0,102833,,,,0,9332ced77549d7971af74075b08f13b1c217f9c42ef2ec36b6d8990c123394cd
thread:issue:102853,issue #102853: pdb but for dynamo (and time travel debugging),stale/low urgency,2,stale/low urgency,1,0,102853,,,,0,0c859cd155b3234a3076aa5fbf65a035712348b5e0426dd55baac188d5e863a9
thread:issue:102878,issue #102878: [dynamo] Diffusers - Graph break on OrderedDict,stale/low urgency,2,stale/low urgency,1,0,102878,,,,0,02594c5d444550de905deaadbf735d2dc01b89c7635b41a362033adca0b6c0ac
thread:issue:102894,"issue #102894: BCELoss and BCEWithLogitsLoss differ when one of the input logits is float(""inf"")",stale/low urgency,2,stale/low urgency,1,0,102894,,,,0,04fdce9e24aa864a76330577dffdc437d40cf8727c706ef1b231b14ed42d603f
thread:issue:102898,issue #102898: After dynamo minifier generates repros that don't entirely match what we minified over,stale/low urgency,2,stale/low urgency,1,0,102898,,,,0,ae0af08f7f045b5005a0de842790a6121adb4ba65d75fe9e07a8ebf8d3ffe5fb
thread:issue:102921,issue #102921: Unbox expectedml,stale/low urgency,2,stale/low urgency,1,0,102921,,,,0,f5cb76439d2eb2fc05bb084b8cb6135f0d605f6b1fd5362974b9447e4a0909d4
thread:issue:102936,issue #102936: torch.cuda.is_available() returns False on GTX 1650 with cuda 11.7 and torch==2.0.0+cpu,stale/low urgency,2,stale/low urgency,1,0,102936,,,,0,e6593a0cb73b9cfb68839c07782d582b6ad5e6ed26120caf44ae2ac11557e0e4
thread:issue:102953,issue #102953: TypeError: (): incompatible function arguments,stale/low urgency,2,stale/low urgency,1,0,102953,,,,0,cacc8b5e1ab3df92d8504d615c77c0625d6f34bb53d0872ee77beb823674c072
thread:issue:102966,"issue #102966: how to workaround the error ""don't have an op for vulkan_prepack::create_linear_context"" ?",stale/low urgency,2,stale/low urgency,1,0,102966,,,,0,2c09a7956d3d6cb8fe1b51c9933820410ab1dd86dc844b9136412ed1532985fa
thread:issue:10298,issue #10298: [feature request] Runtime warning for inappropriate labels (among others),stale/low urgency,2,stale/low urgency,1,0,10298,,,,0,2dfe64606299731df097c64f99fe08a6ce0b15f3557106122dd0157ebc75e594
thread:issue:102999,issue #102999: Preserve weight_g/weight_v accessors on new weight_norm,stale/low urgency,2,stale/low urgency,1,0,102999,,,,0,9eb604bdf945aa6ab8dce86972f1d5e2894f2234538277455e30dcf640aebdf2
thread:issue:103023,issue #103023: MPS bug: padding_idx in nn.Embedding does not prevent gradient accumulation,stale/low urgency,2,stale/low urgency,1,0,103023,,,,0,f662cf134d6c497c7d40cd6b49c86c45e5aea012d389451f7e14dc00313cd428
thread:issue:103060,issue #103060: Symbolic trace error about torch.nn.functional.pad,stale/low urgency,2,stale/low urgency,1,0,103060,,,,0,eba355fda4a901a987230352a8598d6a8ab893bdcf7808e55a8dfdfda8dd4fb8
thread:issue:103089,"issue #103089: [OOM] Unable to convert 30B model to ONNX, using 4x A100's",stale/low urgency,2,stale/low urgency,1,0,103089,,,,0,b64202e47d911323afe6dd09881c6082c7647dbd217c43fa68024baffd93c02b
thread:issue:103131,"issue #103131: error: ‘aligned_alloc’ was not declared in this scope static_cast<char*>(aligned_alloc(FLATBUFFERS_MAX_ALIGNMENT, size)), free);",stale/low urgency,2,stale/low urgency,1,0,103131,,,,0,cfd5bee9845163cd72055c2de67c46ce89ccf17d8171a80996336f6e1cb0c980
thread:issue:10314,issue #10314: [Caffe2] ConvPoolOpBase operator,stale/low urgency,2,stale/low urgency,1,0,10314,,,,0,cfafe824ac528fb999f879e70a8416e238a2f62cda13ba94ee6ea7da081ef867
thread:issue:103150,issue #103150: Build fails at linking torch_shm_manager on aarch64,stale/low urgency,2,stale/low urgency,1,0,103150,,,,0,c94adfc7caec48ab881357cfa07bb794cf084a3841f821b9e4b3e47920f925b8
thread:issue:103160,issue #103160: Allow overriding __repr__ to call dataclass_repr (infinite recursion right now),stale/low urgency,2,stale/low urgency,1,0,103160,,,,0,f9ac6c18e757687e841e152ffe067fd9630a9866b5029e1013281d7f41e44c9b
thread:issue:103173,issue #103173: [RFC] Emit better Telemetry in PyTorch,stale/low urgency,2,stale/low urgency,1,0,103173,,,,0,37477da099afd127f6735ab1fad75dd858e6b13467880087315e4184a13329ec
thread:issue:103212,issue #103212: PyTorch can not be compiled with MKLDNN if system compiler is clang,stale/low urgency,2,stale/low urgency,1,0,103212,,,,0,ac5337ef2db621045737077437cb1dbba91173a5bc458a18b2105bcb3cfc21f4
thread:issue:103241,issue #103241: Image Processing with Pytorch,stale/low urgency,2,stale/low urgency,1,0,103241,,,,0,ecb8e78580cb3e1786f0ecf25c6f97ea9c2a1d6b71fc6e17f5a91645b3329851
thread:issue:103258,"issue #103258: Warn / deprecate / remove ProcessGroupNCCL._group_start(), _group_end() APIs",stale/low urgency,2,stale/low urgency,1,0,103258,,,,0,7b5f9a55d43d94da450933b71963f76653a8a7999478fb0490182aabb84297f2
thread:issue:103272,issue #103272: oneDNN kernel fails to compile,stale/low urgency,2,stale/low urgency,1,0,103272,,,,0,e47548c7d3240ee0aeab9b0e254e06c457999f9f264a17ef7da4cfeab800bc72
thread:issue:103306,issue #103306: DTensor uneven sharding corner cases.,stale/low urgency,2,stale/low urgency,1,0,103306,,,,0,9892e1be5544e93fab1846751aa2f8a94bde605353605440b3969e7fd051a4d2
thread:issue:103313,issue #103313: Fast kernels for low rank matrix multiplication,stale/low urgency,2,stale/low urgency,1,0,103313,,,,0,cef849c22688ca198fbd40a7028180424f6c90d52dbc47b4203cf1031addae9a
thread:issue:103316,issue #103316: binary_cross_entropy (loss) seems to be giving incorrect values for very negative logits,stale/low urgency,2,stale/low urgency,1,0,103316,,,,0,58a80b3308db9bb81ff44fe37dc2e2a4439b1ed341592f6e13cda6e3c4d01941
thread:issue:103329,"issue #103329: RuntimeError: torch.vmap a function that includes in-place arithmetic operations on a zero-initialized tensor, an error ""vmap: inplace arithmetic(self, *extra_args) is not possible"" is raised.",stale/low urgency,2,stale/low urgency,1,0,103329,,,,0,afc807ab6be84b2eec1d5e350c648cc32c011f8db5519cb4aee0b8039c76ef6b
thread:issue:103354,issue #103354: Add a requirements.txt for windows pip packages,stale/low urgency,2,stale/low urgency,1,0,103354,,,,0,a07c632a310cb716c0f6b741fb11b8ed90fc725de358c9c6896e22c6cd26df3e
thread:issue:103370,issue #103370: torchgen/gen_backend_stubs.py compatibility with DispatchStubs,stale/low urgency,2,stale/low urgency,1,0,103370,,,,0,5e739b8b172c38bbaa422ab21ac976baaf057ec46bc07fd8c5fcbb29beb87443
thread:issue:103372,issue #103372: ImportError: cannot import name 'Store' from 'torch.distributed',stale/low urgency,2,stale/low urgency,1,0,103372,,,,0,37872f59149f9f422ecb83c9a168fe42c18ab436040a12f74490fecc2cd270e6
thread:issue:103382,issue #103382: NotImplementedError Could not run 'c10d::alltoall_' with arguments from the 'Meta' backend.,stale/low urgency,2,stale/low urgency,1,0,103382,,,,0,01dc6f5ff0c7003200643291524b52d50c9e3f2349037775a21c125ff35cb1e0
thread:issue:103397,issue #103397: LayerNorm freeze processes using torch multiprocessing,stale/low urgency,2,stale/low urgency,1,0,103397,,,,0,666f61850fb76154cfecc23021ad82af7950f1d6966975a9cdbde4362a8082e8
thread:issue:10341,issue #10341: [Caffe2] Which document should we follow to install caffe2?,stale/low urgency,2,stale/low urgency,1,0,10341,,,,0,e90512104b0b249b3f24263e9fe477206a17bad2078d79380ecc110abb715795
thread:issue:10342,issue #10342: [Caffe2] which kind of database can I read from caffe2,stale/low urgency,2,stale/low urgency,1,0,10342,,,,0,d067b84640e8e360a85c436f0486cc856b9c3660c53d177cc5e932446922f96e
thread:issue:103422,issue #103422: Possible memory leak when using Torch and Torchvision in conjunction with XGBoost,stale/low urgency,2,stale/low urgency,1,0,103422,,,,0,370ac7e7ad1b79b332b7ca4bf484827e0e50f166d061528f6ddd58e088a4845b
thread:issue:103482,issue #103482: Document CI retry rules,stale/low urgency,2,stale/low urgency,1,0,103482,,,,0,9cd208e46d75badb3c15c249533a1eecc42cc4171a7f4e95f914ac05c7c08ec6
thread:issue:103483,issue #103483: torch._dynamo.exc.Unsupported: Tensor.backward with aten_graph=True,stale/low urgency,2,stale/low urgency,1,0,103483,,,,0,179c86a1708b63a62c56f69f4118296203dd842827b7badd2167324b043b7ba8
thread:issue:103484,issue #103484: No backward implementation for `torch._native_multi_head_attention`,stale/low urgency,2,stale/low urgency,1,0,103484,,,,0,aae54c3d224cab9081d1e715fd2a9411704501f17565d4f27a2d05cdf0430e77
thread:issue:103495,issue #103495: DISABLED test_mem_get_info (__main__.TestCuda),stale/low urgency,2,stale/low urgency,1,0,103495,,,,0,602d9a81d8bce344a84088e8f4bd31ee338454845d423bcb315d3cd95f68d151
thread:issue:103498,issue #103498: ERROR:torch.distributed.elastic.multiprocessing.api:failed (exitcode: 1) local_rank: 0 (pid: 6047) of binary: /home/win10-ubuntu/anaconda3/envs/vicuna-7b/bin/python,stale/low urgency,2,stale/low urgency,1,0,103498,,,,0,9877624079dd19550448711951eb1a510a7fef489e8769123ea4b6483bc09f10
thread:issue:103499,issue #103499: CUBLAS_WORKSPACE_CONFIG can not be parsed,stale/low urgency,2,stale/low urgency,1,0,103499,,,,0,c73475ab8c8e35ee1c757d079805c66e13fc2a425b9de83b4cd4aac6f5412064
thread:issue:103518,issue #103518: `gradcheck` produces false positives with sparse inputs when `masked=False`.,stale/low urgency,2,stale/low urgency,1,0,103518,,,,0,f22288817fc07c22fe3b759bc00c13d0f62dea1ff1c1c205bc4c15d2e417ac1b
thread:issue:103530,issue #103530: Deduplicate the operands passed into torch.cond after dynamo tracing.,stale/low urgency,2,stale/low urgency,1,0,103530,,,,0,755679bde7986873f026f992903992d9313a6692929c824f78dd43e985f76b09
thread:issue:103570,issue #103570: File Missing When i build with C++,stale/low urgency,2,stale/low urgency,1,0,103570,,,,0,574b1610915ff14a893f4519377a42c9dca59347b32e6439597228e03da7bf69
thread:issue:103578,"issue #103578: ImportError: undefined symbol: cublasSetWorkspace_v2, version libcublas.so.11",stale/low urgency,2,stale/low urgency,1,0,103578,,,,0,b336ee7148635a83c8967ca63827907e679a795104602b9d587c5d2646d297c4
thread:issue:103589,issue #103589: `interpolate` with `antialias=True` on CUDA doesn't work if the difference of spatial size is large,stale/low urgency,2,stale/low urgency,1,0,103589,,,,0,4fdd342ee34f68f8998e915834081d2b6e73ff1d43b3dba668c6951cfefeef65
thread:issue:103608,issue #103608: Installing Torch on AMD Platform Leads to Huge Docker Image,stale/low urgency,2,stale/low urgency,1,0,103608,,,,0,cf1649e1b4e038281f2c8acda772264b952df8fe34bb1dfa2741a3f8fbbfdf1e
thread:issue:103640,issue #103640: Disclose C++ ATen ops type promotion rules under OpOverload in Python,stale/low urgency,2,stale/low urgency,1,0,103640,,,,0,a3e0582d5f607193a02e8a5758d5a9f832d7f7960800f2de197172abf1e210dd
thread:issue:103643,"issue #103643: ""addmm_out_sparse_csr_impl_mkl"" not implemented for 'Byte'",stale/low urgency,2,stale/low urgency,1,0,103643,,,,0,fe444287e351778dcc9740f1e8729de71540d78996435f2b0edc025879a63a6b
thread:issue:103652,issue #103652: inductor: support horizontal reduction with vec_transpose to improve TIMM swin_base_patch4_window7_224 dynamic shape performance,stale/low urgency,2,stale/low urgency,1,0,103652,,,,0,9ed2ed5091b81fe186494263cf8b8c4d9c342c1c6599c428eccf0a2451c48b22
thread:issue:103668,issue #103668: MultiheadAttention should split embed_dim into four parameters,stale/low urgency,2,stale/low urgency,1,0,103668,,,,0,5a9c01782acc832dbf154715f63420dff04e583003cd9a3b35f760e757106b9b
thread:issue:103671,"issue #103671: ""Y.getIntrusivePtr()->set_storage(X.getIntrusivePtr()->storage()); "" in C++ is not supported",stale/low urgency,2,stale/low urgency,1,0,103671,,,,0,0fb646e5effc4479404b43ed8033fa42ee23c18b76cec2d9820d64a3226d22ce
thread:issue:10375,"issue #10375: Sending CUDA tensor to process, and then back, does not work",stale/low urgency,2,stale/low urgency,1,0,10375,,,,0,7b9ad7bda328f8719c231b8162bf8642fbe5c3e0fae9dce838989ef22f67cf69
thread:issue:10377,issue #10377: [Docs] Update broadcasting documentation for scalars / n-dimensional empty tensors,stale/low urgency,2,stale/low urgency,1,0,10377,,,,0,b61157933ade8678c07f0ad450b3c9a43595c1698531e92a95ecf94d6da8ab8a
thread:issue:103802,issue #103802: Eager PTDQ Performs Worse Than Non-Quantized Linear Layer on CPU(in Terms of Speed),stale/low urgency,2,stale/low urgency,1,0,103802,,,,0,45f8e08b384d89824a179a1918c6889a1ffc8e61164e5af6b2458e6f22177e69
thread:issue:103805,issue #103805: Error when building with USE_TENSORRT=1,stale/low urgency,2,stale/low urgency,1,0,103805,,,,0,14f1082cf7bc681f02a26210934621047d50ce08f0bfcc125e6923d6244a8034
thread:issue:103829,issue #103829: RPC Framework support custom backend,stale/low urgency,2,stale/low urgency,1,0,103829,,,,0,0190f83299a64176bb0ad42f57a56b140cb716e6328e0363e7de3e660dad67ff
thread:issue:103832,issue #103832: Torch 1.13 for GPU breaks if libcublas is already present.,stale/low urgency,2,stale/low urgency,1,0,103832,,,,0,ccdddd071e05d2bb93f3077fc9a38a0eb9510bf12f810d71e5ba1032ac5349ff
thread:issue:103836,issue #103836: Please consider the SCFA/dynamic flash attention for your implementation of scaled dot product attention,stale/low urgency,2,stale/low urgency,1,0,103836,,,,0,70df947eda7121279d6f2af188c801e2781a4c3f6807c5aba4bf6b0c091f4870
thread:issue:103837,issue #103837: type conflict,stale/low urgency,2,stale/low urgency,1,0,103837,,,,0,58e3fac8fa723a193032395bd1689f1d2e844e8077cc4925adb2f174e0925b24
thread:issue:103846,"issue #103846: The document style is inconsistent with other documents, and the parameter type is not clearly highlight",stale/low urgency,2,stale/low urgency,1,0,103846,,,,0,0e1c820632e43ee325e7402bc18d3747320b56a91b78e58b010ad366b4719ad2
thread:issue:103847,issue #103847: Some parameters are missing type descriptions,stale/low urgency,2,stale/low urgency,1,0,103847,,,,0,e2e3da5a1617c5e8a6ff7c8af1985cb69ad0ae6bdea27539aba4a4ec627731fe
thread:issue:103860,"issue #103860: Long PR description leads to ""Argument list too long"" error from docker",stale/low urgency,2,stale/low urgency,1,0,103860,,,,0,1fa920fdd0aa5c4bb3ffcf8bb9899e5da06576afa280b13e8681141be3297ad6
thread:issue:103868,"issue #103868: row.device().is_cpu() INTERNAL ASSERT FAILED at ""csrc/cpu/diag_cpu.cpp"":7",stale/low urgency,2,stale/low urgency,1,0,103868,,,,0,c64848403e74dd0d152a212a97baaf9a0add1847796943a52f2604462601bbb0
thread:issue:103879,issue #103879: Can ``torch.vmap`` add ``grad_fn``= SelectBackward when maping over some dimension of the inputs?,stale/low urgency,2,stale/low urgency,1,0,103879,,,,0,83ba08d82354354fec0bee9629268e33f51a96180ffb8c1367db47724003243a
thread:issue:103895,issue #103895: Issue with loading similar checkpoints in a distributed fashion,stale/low urgency,2,stale/low urgency,1,0,103895,,,,0,c8542d8d51515629323ddba2b8495f52a909bc657485d6d55913e5ae2acce7af
thread:issue:103959,"issue #103959: Need the full ""Release Compatibility Matrix"" of torch",stale/low urgency,2,stale/low urgency,1,0,103959,,,,0,81675e288d613faabed3b5626548f66626d4c9f4c01bb1cc97b97ad94084151c
thread:issue:10396,"issue #10396: [feature request] Provide a way to redirect shared memory prefix ""/torch_""",stale/low urgency,2,stale/low urgency,1,0,10396,,,,0,400074ebf18f4e791121f93c0f4513e927ba54e91d7b6603c7d8dec1b908815b
thread:issue:103960,issue #103960: CODEOWNERS file has errors due to non existent people being referred to,stale/low urgency,2,stale/low urgency,1,0,103960,,,,0,74d52b1dda8ec1497d483762b7d9fc41eec2674b2f587c2d79a2207adfbdd69e
thread:issue:103962,issue #103962: How to unwrap after auto_wrap in FSDP?,stale/low urgency,2,stale/low urgency,1,0,103962,,,,0,f4aa182155769a2d9324f08037024312970f8220c9651af50b9fc1843dd68e78
thread:issue:103990,issue #103990: [Inductor] Freezing Add support for Caching Parameter Conversions,stale/low urgency,2,stale/low urgency,1,0,103990,,,,0,245bc05b1ecd331ac1cbca9341e2b8fa1cced3833109b951d6697d0e1ffe819d
thread:issue:104029,issue #104029: SummaryWriter.add_embedding not working for RGBA images,stale/low urgency,2,stale/low urgency,1,0,104029,,,,0,fbfc9e42e3aef23784553b9f9208448e88b34f2a2d353f06ad05bfd943a7662a
thread:issue:104033,issue #104033: Reproducibility documentation should be updated,stale/low urgency,2,stale/low urgency,1,0,104033,,,,0,07fe6a2c65294caa0e3c73f59b070f3977e104ead5ca7bcf5f5567a0ff3bc242
thread:issue:104040,issue #104040: Improved error checking for custom Function when saving intermediates,stale/low urgency,2,stale/low urgency,1,0,104040,,,,0,e42a08695645dbe5c862ec4619690a8902780322c8f255686473f46a3afd2ea4
thread:issue:104045,"issue #104045: ""TorchDynamo Deeper Dive"" doc is missing some part",stale/low urgency,2,stale/low urgency,1,0,104045,,,,0,a380ec296ae4ff3fbe1700e946ed56927099555f0fb99ec6c3318aaa67d2a676
thread:issue:104106,issue #104106: batch size unexpectedly affects model inference on Mac M1,stale/low urgency,2,stale/low urgency,1,0,104106,,,,0,5e65b4b97946a24c21e23b85e81fcc2a856dc9951ed430c1f82e6a7c2d5f0a6a
thread:issue:104107,issue #104107: Tensor to_sparse fails on large matrices,stale/low urgency,2,stale/low urgency,1,0,104107,,,,0,f41abbb9398ba0ce7f9ab8f368f896b2c41720d85294994f1b29bf54bf193a1c
thread:issue:104147,issue #104147: PyTorch2.0 ROCM LayerNorm HIP error: invalid configuration,stale/low urgency,2,stale/low urgency,1,0,104147,,,,0,9845d8e0b542ec35d19604ce9092cbb4be9b8c6113797f07d36f1e71110ddfae
thread:issue:104152,issue #104152: Error 101: invalid device ordinal (Triggered internally at ../c10/cuda/CUDAFunctions.cpp:109.),stale/low urgency,2,stale/low urgency,1,0,104152,,,,0,cf9bc17d09742527aa09802aef842b71b6fd13d61e701fd29f7701c0d2aa2905
thread:issue:104154,issue #104154: Numbers bigger than the range should be inf while the implementation just keeps the original.,stale/low urgency,2,stale/low urgency,1,0,104154,,,,0,62b9578467f718769b4475df3264cb3e40741acd0d9d58394f523aa0cb339a03
thread:issue:104163,issue #104163: Nested Tensor with PyG dataset custom class,stale/low urgency,2,stale/low urgency,1,0,104163,,,,0,403e6b0c5c7989b9c73a9a55fe65c9cadb71e19ce2de2c6b4a971e6310878b24
thread:issue:104188,issue #104188: Add memory managemenet information for Apple silicon mps backend,stale/low urgency,2,stale/low urgency,1,0,104188,,,,0,e61414143eb5fe906e6918c7a3c639ee20ae4ecd805a6974104533cd135bdf31
thread:issue:104195,issue #104195: Torchscript with dynamic quantization produces inconsistent model outputs,stale/low urgency,2,stale/low urgency,1,0,104195,,,,0,b576ee9c69dbbf912948ad8e3d50848ed287d3ceab43c89f12ffb6fe16dea18c
thread:issue:104252,issue #104252: [RFC] Make `_HYBRID_SHARD_ZERO2` public as `HYBRID_SHARD_GRAD_OP`,stale/low urgency,2,stale/low urgency,1,0,104252,,,,0,f3dc2905bd41fd2a375124a982032211efaa5a76152ba109ef28312ab5fe1a77
thread:issue:104282,issue #104282: Rename `topic: not user facing` to `release notes: not user facing`,stale/low urgency,2,stale/low urgency,1,0,104282,,,,0,75703ee89594ee7c6ea9fc589e68ed5ef83cf2bbc10f97a77db33a634b9b82ce
thread:issue:104296,issue #104296: affine_grid and grid_sample operators merge/accelleration,stale/low urgency,2,stale/low urgency,1,0,104296,,,,0,75fdc3694836861a1555ef2b7f00c820bb26c0b66d898fc1a50ce1b1a1e14091
thread:issue:104315,issue #104315: Torch randperm with device mps does not sample exactly uniformly from all possible permutations,stale/low urgency,2,stale/low urgency,1,0,104315,,,,0,fe38f45c0d0a9b0e253183a13ea7d79cca17c1bd8b4d290dc78c912e1c432663
thread:issue:104322,issue #104322: Illegal Memory Access on H100 `TestSparseCompressedTritonKernelsCUDA.test_triton_sampled_addmm_block_size_16_cuda_bfloat16`,stale/low urgency,2,stale/low urgency,1,0,104322,,,,0,c607418790e5fe37024543356839464c053c79322a0b4627fb6870eb55b77f05
thread:issue:104389,issue #104389: distributed hooks want to support custom device,stale/low urgency,2,stale/low urgency,1,0,104389,,,,0,8284de8a3197a5da080985c5ca2061025f0ac507f507ec748fa498d5cb298251
thread:issue:104466,issue #104466: `torch.view_as_real(tensor)` should return `nn.identity(tensor)` if its not complex instead of raising an error,stale/low urgency,2,stale/low urgency,1,0,104466,,,,0,dadf506f288178aa643eef0595dd520d07cc44f833925ae35ddd35a49c2e1916
thread:issue:104479,issue #104479: FSDP Optimizer Overlap - follow ups,stale/low urgency,2,stale/low urgency,1,0,104479,,,,0,d45a8682f9f01e572bff19475e9de5d3ace22123e9a67ed35378619f24f35a7e
thread:issue:104502,issue #104502: Get errors after compiling and running PyTorch MINIMAL EXAMPLE for c++ Mac M1 with make,stale/low urgency,2,stale/low urgency,1,0,104502,,,,0,246467b5b6db96dd443a7d34681b80cf001c6d6fee0c16c0a3c31d3c0e7a79e2
thread:issue:104507,issue #104507: [error] while Implementation of pytorch DistributedParallel,stale/low urgency,2,stale/low urgency,1,0,104507,,,,0,3a6501f35835977305661bc3500e8d44ba85e50815386a976243014f547b4539
thread:issue:104510,"issue #104510: Using the latest version of Torch, when the code executes tcpstore, there is no response",stale/low urgency,2,stale/low urgency,1,0,104510,,,,0,fab3e2e7d02888edc11b4bc8ec165fa162e2d6e4a5682b142db99ef943361d70
thread:issue:104516,issue #104516: vec_test_all_types_xxx with dtype c10::complex<float> and c10::complex<double> has failures on division,stale/low urgency,2,stale/low urgency,1,0,104516,,,,0,d9b84152fa9b5ff9f9c5ab77f78df4ca9b42dde5c190042b71480c95aa2e95b3
thread:issue:104531,issue #104531: [RFC] Optional Modular Representation for FX Graph from `torch.compile()`,stale/low urgency,2,stale/low urgency,1,0,104531,,,,0,c73bc6bf5b92177ff66472dd292f6960a389d77b996dd23e860bf5f203ca0938
thread:issue:104591,issue #104591: version libcudnn_ops_infer.so.8 not defined in file libcudnn_ops_infer.so.8 with link time reference,stale/low urgency,2,stale/low urgency,1,0,104591,,,,0,731650292cbfd99179f25b69239aca70dde0bedbafdae09764aa860ba25838b9
thread:issue:104593,issue #104593: Bug in Conv/BN fuser with torch.fx,stale/low urgency,2,stale/low urgency,1,0,104593,,,,0,0678d0514a009af76e8870bababad8a56435091a8efae5f4d3c656d0bf3d54ec
thread:issue:10460,issue #10460: Type name float registered twice. This should not happen. Do you have duplicated CAFFE_KNOWN_TYPE?,stale/low urgency,2,stale/low urgency,1,0,10460,,,,0,a728f78cad171099b2d8a4da45a4ca4ad1dfa35bcecb7d5d74dc1bba7df8821f
thread:issue:104602,"issue #104602: F.adaptive_avg_pool3d(input, 1) returns infinity in half precision",stale/low urgency,2,stale/low urgency,1,0,104602,,,,0,1f68ed6345c91a881db1027e25c5d8ae236360a0fd7da4305b6a6368f15ae250
thread:issue:104606,"issue #104606: Failure in optimize_for_mobile when using conv1d(..., padding='same')",stale/low urgency,2,stale/low urgency,1,0,104606,,,,0,98db87828167155d369e81f07cb94413d65e02aa503c1cb814bd4ae6989033cf
thread:issue:10461,issue #10461: Key already registered. Offending key: caffe2_print_stacktraces.,stale/low urgency,2,stale/low urgency,1,0,10461,,,,0,3f38f10834c786fc290799f20b165f251c0ba87ce998ed812cb373d4a61fb046
thread:issue:104620,issue #104620: `torch.distributed.rpc.backend_registry.register_backend` fails to update `BackendType` enum,stale/low urgency,2,stale/low urgency,1,0,104620,,,,0,017959c3854534278e8c28ce1fa39cff362d989eae655a51a77300d7ff0b4871
thread:issue:104697,issue #104697: LSTM built-in dropout not reproducible on GPU,stale/low urgency,2,stale/low urgency,1,0,104697,,,,0,bed94141ee1980308407335dee3f615843f1e6e7e3d28247ed1e1cef96bf4faa
thread:issue:104698,issue #104698: Incorrect Error Message Ordering for nn.AdaptiveAvgPool2d with Incorrect output_size,stale/low urgency,2,stale/low urgency,1,0,104698,,,,0,a27ff0918661639c03ef556cea39ca8c367307ac0c7735af0df140b44bd283ab
thread:issue:104704,issue #104704: generate_vmap_rule=True sometimes gives batched grad_output,stale/low urgency,2,stale/low urgency,1,0,104704,,,,0,9602c7111febfce999d2feedd19ebdfd8ab1b7ba07ffecdc90e20043aeadda29
thread:issue:104711,issue #104711: New Loss Function Add In Pytorch,stale/low urgency,2,stale/low urgency,1,0,104711,,,,0,78363da0c21655ff85dccbf13254627dddec80933c50723e4c5c5e3c7aed718a
thread:issue:104739,issue #104739: Error reporting uses formal parameter names of downstream C++ function,stale/low urgency,2,stale/low urgency,1,0,104739,,,,0,c029623295b6cdf758e853f6b9a1ffb46608892f29f8391c9a31c819037d9f14
thread:issue:104761,"issue #104761: Torch Filename Storage hangs on ""file_system"" sharing strategy after in-place fill",stale/low urgency,2,stale/low urgency,1,0,104761,,,,0,da4432b09172e0c1c4ada0290ae3b777b7f04cd4d1bd1339c232737c6f6b4eab
thread:issue:104768,issue #104768: Support for `eval` in functional_call,stale/low urgency,2,stale/low urgency,1,0,104768,,,,0,e171af1dcaad313b1ed97bb0e88099ed52e74408a1d9107337e81df0f3b5ff91
thread:issue:104776,issue #104776: torch.nn.CrossEntropyLoss: class weighting changes label_smoothing,stale/low urgency,2,stale/low urgency,1,0,104776,,,,0,eb4d1261a25b1e7569f552c3092868b6c71a249bb88d49a036897dd90a1076d9
thread:issue:104788,issue #104788: Use `isinstance` instead of `type` when checking for `torch.nn.Parameter`,stale/low urgency,2,stale/low urgency,1,0,104788,,,,0,1ceb2258169349c734a26d30a96e1262cacf98bfafc5478742616f87f56eced5
thread:issue:104789,issue #104789: [BE] Evaluate and improve eager for-loop optimizer memory perf,stale/low urgency,2,stale/low urgency,1,0,104789,,,,0,df23c3a51676f9d3b4b344056fd0c127b5a429691e12b7da643e0406033fb639
thread:issue:104808,issue #104808: PyTorch built with CuDNN-8.8.1 crashes if CuDNN-8.9.2 is installed on the system,stale/low urgency,2,stale/low urgency,1,0,104808,,,,0,bc2ff551d0f867e405b93c6cc1779220dfe30826959f638ebf56fa1adbb7cc50
thread:issue:10482,issue #10482: Reduce code duplication in interpolate and make it more generic,stale/low urgency,2,stale/low urgency,1,0,10482,,,,0,51e4ee8aa549da9a9c099288ec10fff8e22c114272e5c9d29403b19b582795df
thread:issue:104823,issue #104823: Unrelated error messages with torch.nn.AdaptiveAvgPool3d,stale/low urgency,2,stale/low urgency,1,0,104823,,,,0,8902107e366ec8706b68e4f4b08c6fd0cca164e943e7ac070d450e0a2f26e507
thread:issue:104832,"issue #104832: MPS matmul with sliced (strided) out argument produces wrong output, may corrupt memory",stale/low urgency,2,stale/low urgency,1,0,104832,,,,0,cd307095195f0485318dabc8e83b42559c9ceeeede03d51d172deb6945172ce3
thread:issue:104845,issue #104845: Implement `diag` method for sparse COO tensors,stale/low urgency,2,stale/low urgency,1,0,104845,,,,0,fe99a5719e78eb677483f3929db914bf6814118cc6568ea050f0890c09c1b68a
thread:issue:104853,issue #104853: torch.norm inconsistency?,stale/low urgency,2,stale/low urgency,1,0,104853,,,,0,2c17803ed51e197228dc5afe66bc2cc775236cecfaa3b455c821b724d484ceaa
thread:issue:104867,issue #104867: rfftn and irfftn operations in pt2 return different results compared to v1.12.1,stale/low urgency,2,stale/low urgency,1,0,104867,,,,0,9022751f406ac08ac650e83478d131574ed0a38f81c8c47252664d2b0fb0a87a
thread:issue:104872,issue #104872: errors in CONTRIBUTING.md,stale/low urgency,2,stale/low urgency,1,0,104872,,,,0,991b59b676a8f9e483457b89ee18f9573f4cead0448fc6480257fdeb358f9aba
thread:issue:104875,issue #104875: torch/testing/_comparison.py: If you are a user and see this message during normal operation please file an issue,stale/low urgency,2,stale/low urgency,1,0,104875,,,,0,fdd4de4373a9b828e1abb910897d8010599adb575f1fb31bd087f7d907e2fe3d
thread:issue:104925,issue #104925: Unnecessary record_stream call for backend:cudaMallocAsync,stale/low urgency,2,stale/low urgency,1,0,104925,,,,0,a2d6488abe08ffda48fe1e33459b07d0e6b9d68e4c136d137fb01d90a938ff04
thread:issue:104933,issue #104933: torch.func.jvp fails with BERT training,stale/low urgency,2,stale/low urgency,1,0,104933,,,,0,b1e824dc9248ff678c8d32b7231b55ed39ae4ab5b85c9eb2cc86e0816b0a0466
thread:issue:104935,issue #104935: torch version compare,stale/low urgency,2,stale/low urgency,1,0,104935,,,,0,b2dafd133c6c878592c8470a375bcb741933f692665c09aea2f013cd938276f0
thread:issue:104962,issue #104962: Add a diagram showing the code structure to CONTRIBUTING.md,stale/low urgency,2,stale/low urgency,1,0,104962,,,,0,b78384934f39ec956aa53c2bf4eb35b6a1cc71069c95cb524fb7d36bbd0a7bb8
thread:issue:104981,issue #104981: NotImplementedError: Could not run 'aten::_spdiags' with arguments from the 'CUDA' backend.,stale/low urgency,2,stale/low urgency,1,0,104981,,,,0,ed3306a45f66d6945be03ca16b23bbd8989487458658acb083129ae6f1fa0ea3
thread:issue:105060,issue #105060: extra information messages for mac in setup.py would help.,stale/low urgency,2,stale/low urgency,1,0,105060,,,,0,94055f9d87bcd8a3d1f632d67968bb52107d051144b4d6e69965fc5148da5dad
thread:issue:105062,issue #105062: [feature request] make the input k in rot90 a list of int to rotate tensors individually in a batch,stale/low urgency,2,stale/low urgency,1,0,105062,,,,0,2a314fa08d69acd65e5f912782fd508eef8386a35379001dbdb49d0ab38e1762
thread:issue:105068,issue #105068: [linalg] test_ops.py::test_python_ref_meta__refs_linalg_svd_cpu_complex failing,stale/low urgency,2,stale/low urgency,1,0,105068,,,,0,60632141fdde30978ff175a13c4a14c2369c3c9859ff759f17c38d90e2aa3c87
thread:issue:105092,issue #105092: [RFC] Proposal to upgrade LLVM version,stale/low urgency,2,stale/low urgency,1,0,105092,,,,0,2f294f267dc82b91455309489805e140f012ab4bd417bec22e26f7d4e0adb3ab
thread:issue:105105,issue #105105: Remaining functions without meta registrations,stale/low urgency,2,stale/low urgency,1,0,105105,,,,0,681f8af4001530d36cbbbedbc5d549d9b2a5bb857d4714976203906ac9f0e084
thread:issue:10515,issue #10515: [Caffe2] Segmentation faults in multithreading Caffe2,stale/low urgency,2,stale/low urgency,1,0,10515,,,,0,1ec6d47251625a4acfa8ad541217bd0c7c161cb76fc5a5d59bfe639d680311a3
thread:issue:105213,issue #105213: [DTensor] Module parallelized with ColwiseParallel should return a sharded tensor,stale/low urgency,2,stale/low urgency,1,0,105213,,,,0,ddfdcc307d340a003c71be0e06277c73b98b33e32932a720ea176eda933b62f9
thread:issue:105220,issue #105220: Significant time difference of calculating Jacobian matrix using jacrev and oracle functions,stale/low urgency,2,stale/low urgency,1,0,105220,,,,0,68591d5c7e02300103ca7ba09ae6cf3fdecf73fd0f604a4f88d75f6d9423b508
thread:issue:105281,issue #105281: Optimize PyTorch C++ part with Profile-Guided Optimization (PGO),stale/low urgency,2,stale/low urgency,1,0,105281,,,,0,33b055c2792d9124d8992621896b52474e982d786cb5b584fea0a7daec14b821
thread:issue:105299,issue #105299: Support ONNX opset 20 to export GELU to one single op,stale/low urgency,2,stale/low urgency,1,0,105299,,,,0,e5fee753afcd9a2b8b552a10c452affe4aec8fbb3985d0fcca1f23c82b75bc0c
thread:issue:105304,"issue #105304: Backward pass with sparse parameters results in error ""Sparse division requires a scalar or zero-dim dense tensor divisor""",stale/low urgency,2,stale/low urgency,1,0,105304,,,,0,c8f0d3b4db0b14a13c34de64f60e8431aa41cd21b0d5b749d90ef37f06124ead
thread:issue:105326,issue #105326: Can't vmap over torch.tensor constructor,stale/low urgency,2,stale/low urgency,1,0,105326,,,,0,85c579a6a74f3c1e28a1a3ba0845802bdd24e89fbbd84a2f2048124755179920
thread:issue:105335,issue #105335: Enable SLEEF on ARM,stale/low urgency,2,stale/low urgency,1,0,105335,,,,0,59f231eb657ba4cd15dc92bb26f01e784d53ac69cc92a28d6f99c089349ef37e
thread:issue:105445,issue #105445: Silent Error of torch.fx.symbolic_trace when forward hooks are registered,stale/low urgency,2,stale/low urgency,1,0,105445,,,,0,e87784a10c88c48467896aa0a6d510002aac73091e5ade0def552988a9abf33e
thread:issue:105448,issue #105448: Will nn.unfold support non-4D-tensor input in future version?,stale/low urgency,2,stale/low urgency,1,0,105448,,,,0,d6c3f7814653f6a8ad6f31e5c0394acd2ad4c329cd49a00d4af966417d928f81
thread:issue:105460,issue #105460: Specify version,stale/low urgency,2,stale/low urgency,1,0,105460,,,,0,572f2b473e7c611a6d1a6b1556a9052570341eb950a3f03e78fa30ba362f95d4
thread:issue:105529,issue #105529: Efficient BMM for sparse-dense tensors,stale/low urgency,2,stale/low urgency,1,0,105529,,,,0,b20d5f38f4f1ef6c3978e6394d6043221053a816a5aa49679ddc58bda3c5852e
thread:issue:105548,issue #105548: torch.sparse.sampled_addmm doesn't compute gradients for 3D tensors,stale/low urgency,2,stale/low urgency,1,0,105548,,,,0,994bfb7c48a24ed41dc46cd8d861787b6348ca2780b89e59a07f5fc39b2e39a7
thread:issue:105629,issue #105629: F.pad will accept 0 and negative values as parameter,stale/low urgency,2,stale/low urgency,1,0,105629,,,,0,d4050717caa6e0e57bed79fe7c290b100c4968fd5da075aa66fab790c4a71c8b
thread:issue:105644,issue #105644: Tensor subclass is not preserved during backward with gradient checkpointing,stale/low urgency,2,stale/low urgency,1,0,105644,,,,0,8e3eb1d09a612029b3e18b41861901ec4f4376538b540b92298801b50ba030ea
thread:issue:105682,issue #105682: `torch.autocast(bfloat16)` runs bwd matmuls in fp16,stale/low urgency,2,stale/low urgency,1,0,105682,,,,0,77537d596a1f947997fd15a76308f54b1467bbe29223472752a80ceeca320fe6
thread:issue:105697,issue #105697: Support symmetry in einsum,stale/low urgency,2,stale/low urgency,1,0,105697,,,,0,034af78cfc652bf9281213d4a2f2af61a9f7ddb28ac728232ac80ea7f8bb500c
thread:issue:105726,issue #105726: There is a big precision error between A100 and 3090 when using torch.matmul with fp16 precision,stale/low urgency,2,stale/low urgency,1,0,105726,,,,0,17ce947cb430bc86197ce36b39d659a683654167e48ce0ba125a0c744807c596
thread:issue:105728,issue #105728: Compile error PyTorch 2.0.1 / GCC 13.1.0,stale/low urgency,2,stale/low urgency,1,0,105728,,,,0,c4a99a5562dc479d614911a5a596b4193add4d9d6fd0d3f329289af353d9fd8d
thread:issue:10577,issue #10577: [BUG]: unstable happend in saving model.,stale/low urgency,2,stale/low urgency,1,0,10577,,,,0,5189f2e1bf698884bb41f0c0294abd1ef26fb14de2f0be5f6f0e170e06a48dae
thread:issue:105778,issue #105778: https://pytorch.org/docs/stable/backends.html does not describe torch.backends.cpu,stale/low urgency,2,stale/low urgency,1,0,105778,,,,0,58c4ebddacf46ef9adfd3d491c1b462d0c836606004c7de8b9525690c7f3dbef
thread:issue:105790,issue #105790: torch.sparse.mm() with reduce operator for GPU support and COO matrices,stale/low urgency,2,stale/low urgency,1,0,105790,,,,0,e35b36873167b01cee1988f9ad5a2a774037b7a630fae0a0f8448591b5ca67a0
thread:issue:105799,issue #105799: inconsistent signature for dataloader in docs/source/data.rst,stale/low urgency,2,stale/low urgency,1,0,105799,,,,0,7ca8711311951609fa9afa44bf72c21e329e2037e9d79928427c51fc1c438f94
thread:issue:10582,issue #10582: [Caffe2] Unable to use MPI rendezvous in Caffe2,stale/low urgency,2,stale/low urgency,1,0,10582,,,,0,bda9640323ab5b6864fb15f1272e22c6736207ec425731f5330ad8db910ba896
thread:issue:105846,issue #105846: Strange backward behavior with sparse tensors,stale/low urgency,2,stale/low urgency,1,0,105846,,,,0,52b007a619def4d01c7b66a00e97028c39bdb0ddc40c520cb9d271b017664361
thread:issue:105859,issue #105859: Report model flop utilization (mfu) in benchmark,stale/low urgency,2,stale/low urgency,1,0,105859,,,,0,2a355f35b75e29173c382b6ab576b6eadb6119ddcb521b7a236c553263f536b2
thread:issue:105860,issue #105860: Programmation error enabling unlegal memory access on gpu,stale/low urgency,2,stale/low urgency,1,0,105860,,,,0,bba5aabb9c641b6abda0c9f973ed7ecc37c3e2459f8cdedceb7ab2321aceaaeb
thread:issue:105904,issue #105904: Libtorch linking Error:undefined reference,stale/low urgency,2,stale/low urgency,1,0,105904,,,,0,d9247674665652b37628c029fa3893e1c387000f889a0cc51776b08abe4f5d39
thread:issue:105912,issue #105912: Batching rule for aten::bincount.,stale/low urgency,2,stale/low urgency,1,0,105912,,,,0,b415366daa2cc6cf44bc8ba5076636fc93fa15d6a5bc4bb404ef47f5b186b25f
thread:issue:105916,issue #105916: Missing coalesced flag from `torch.autograd.Function.backward`,stale/low urgency,2,stale/low urgency,1,0,105916,,,,0,d31c6187c1418e218401cb4ea43e01c4089c3223af822ee27b619d2326e4ddb1
thread:issue:105925,issue #105925: Possible speed up of nn.MultiheadAttention,stale/low urgency,2,stale/low urgency,1,0,105925,,,,0,1454fda5c8053541a76b2a9c8608122b873168a2243cbce7997a5162c605be43
thread:issue:105954,issue #105954: Ensure PRs are rebased on top of a recent commit (CI check),stale/low urgency,2,stale/low urgency,1,0,105954,,,,0,8c3377a787efa5db8dc5854e56f1e27bd5c5d1bef2c6cce01752fe29be0e6351
thread:issue:106011,issue #106011: Misleading error message in multilabel_margin_loss when passing incompatible tensor dimensions,stale/low urgency,2,stale/low urgency,1,0,106011,,,,0,fb4029af26464520f33a64755ecd2c2ecea1766a6517f8c3edb51e6ffb2411e3
thread:issue:10604,"issue #10604: Bilinear interpolation behavior inconsistent with TF, CoreML and Caffe",stale/low urgency,2,stale/low urgency,1,0,10604,,,,0,497a70f36bab148209dd4111a11bc4eabefb2231dc8c74bca5248c8553dccaf2
thread:issue:106073,issue #106073: Potential lack of CI testing on older NVIDIA GPU,stale/low urgency,2,stale/low urgency,1,0,106073,,,,0,d569ebc12a7680d2089743a044502986212245f39684ced514b139f834ffdd8c
thread:issue:106121,issue #106121: Got error when train models with more than one param_group in torch2.0,stale/low urgency,2,stale/low urgency,1,0,106121,,,,0,3ec544e84e68522a166eceb4f3029f2cfb413720a988a2b533d370a037e5f5a6
thread:issue:106124,issue #106124: [feature request] Better argument checks and error messaging for `tensor.repeat`,stale/low urgency,2,stale/low urgency,1,0,106124,,,,0,e3c0e359f05284329d093d97f6e3a29bc3014acc18dd341152c9d8d1d93e653e
thread:issue:106130,issue #106130: torch._subclasses.fake_tensor.DynamicOutputShapeException: aten.nonzero.default,stale/low urgency,2,stale/low urgency,1,0,106130,,,,0,fdf6b791c2301a8d5c7c4fd300377eb1ebcd013c7117b517729b70c7ab0d790a
thread:issue:106137,"issue #106137: `export(..., pre_dispatch=True)` for model in eval mode still inserts autograd ops",stale/low urgency,2,stale/low urgency,1,0,106137,,,,0,86bd402c2a9f92bbf58c1da715b22a8c149bff146df293c21e3d7d6367025ebd
thread:issue:106141,issue #106141: Sparse COO indices are torch.Int64 -- is this necessary?,stale/low urgency,2,stale/low urgency,1,0,106141,,,,0,12761c1b1fa25fba9851948d45ddd5ceadfb9b66a8aed3482e8a50357bb9f0dd
thread:issue:106173,issue #106173: `torch.ops.aten.split.Tensor._schema` return alias annotations are wrong,stale/low urgency,2,stale/low urgency,1,0,106173,,,,0,a7c36bf5630040d63598a1c90b0044af7a4ef31fde829c8faad53f1185a01788
thread:issue:106221,issue #106221: RProp improvement tracker,stale/low urgency,2,stale/low urgency,1,0,106221,,,,0,16b2419e8523a7a5b26255009b3cced94db62a24204f2497060978eec2242ecd
thread:issue:106237,issue #106237: ghstack + mergebot race condition,stale/low urgency,2,stale/low urgency,1,0,106237,,,,0,bf8e625531c826a4ea466bcc78f60e0cbc6644becdd70a50431309e0c20cf5d1
thread:issue:106265,issue #106265: RuntimeError: Expected a proper Tensor but got None (or an undefined Tensor in C++) for argument #0 'grad_y',stale/low urgency,2,stale/low urgency,1,0,106265,,,,0,1fa995493af10b5ea9b2e0c51d51e1bcec5bad2bee09226acc4d22ccdf061b8d
thread:issue:106269,issue #106269: Make our source attribution debug prints more useful for Compiler Explorer,stale/low urgency,2,stale/low urgency,1,0,106269,,,,0,70192ffcd254d381e51cb59a8f82839b7f778c670b1d93c08c3548e669e68209
thread:issue:106302,issue #106302: `torch.nn.modules.MultiheadAttention` yields different graph under pre_dispatch tracing,stale/low urgency,2,stale/low urgency,1,0,106302,,,,0,91fb829654f205f300e8bac33009e3e75035d8e9c145e484f15aeec51d954432
thread:issue:106339,"issue #106339: ReduceLROnPlateau increases learning rate exponentially, causing training to diverge",stale/low urgency,2,stale/low urgency,1,0,106339,,,,0,f3f1a0a566a1212c9a663c7066e016c642e42992fbb41b6e4f7bd300b0345d52
thread:issue:106359,issue #106359: backwards compatibility about class _LRScheduler,stale/low urgency,2,stale/low urgency,1,0,106359,,,,0,61d3f68488222447c5a1dfd260d8a3bdfd13e7463ff76180f57f55ef4bc87494
thread:issue:106375,issue #106375: nll_loss reference shouldn't be registered as a decomposition.,stale/low urgency,2,stale/low urgency,1,0,106375,,,,0,5ff3ddae634755ad91778b27c87e729003bdf9b3f4fc7529a4eb6fe3619f3d02
thread:issue:106382,issue #106382: Installing torchvision for CPU leads to unwanted upgrade of torch + pip would not install nightly as considers that release is the latest (?),stale/low urgency,2,stale/low urgency,1,0,106382,,,,0,c4ad8258ade1e3a9ee738a2bf06506e1831372f018be29e63c60c56d8987d949
thread:issue:106388,issue #106388: Better export story for autograd.Function?,stale/low urgency,2,stale/low urgency,1,0,106388,,,,0,9f1daaa17a3df8e84a83fc6bfd29790decbb26cdb5f36a3523db7d0fdbf328bf
thread:issue:106408,issue #106408: PyTorchMPS not showing up in Instruments for `torch.mps.profiler`,stale/low urgency,2,stale/low urgency,1,0,106408,,,,0,8b16f17d21f0bf2cb005e605b82bcf31dd3048f5e438c37f393996427128fc41
thread:issue:106427,issue #106427: `torch.nn.utils.clip_grad_norm_()` causes H2D sync with foreach ops.,stale/low urgency,2,stale/low urgency,1,0,106427,,,,0,1eb2c53d4e9e2079b7538dbc431a6958c7aa1f370dae14cf09f6923f9933a0a3
thread:issue:106450,issue #106450: Check for output_padding <= stride/dilation in ConvTranspose1d,stale/low urgency,2,stale/low urgency,1,0,106450,,,,0,709baf862805512be3e89f31e2513ca9468828512e93eebc6d4d3c376a62346d
thread:issue:106467,issue #106467: nn.CrossEntropyLoss with invalid target generates corrups memory eventualy leading to CUDA error: an illegal memory access,stale/low urgency,2,stale/low urgency,1,0,106467,,,,0,9ffb0e75ce9361259fdd796b249d74b19aff5ae2ef1f09d4e577a97d877708e5
thread:issue:106520,issue #106520: Distributed torch.linalg.eigh (and other functions) on cuda using cuSOLVERMG,stale/low urgency,2,stale/low urgency,1,0,106520,,,,0,f9be11c89340039d25c76d72ba1a504d8f35fd80f7a84a958f7e7127e41b2558
thread:issue:106529,issue #106529: Pytorch + ROCm+ Windows,stale/low urgency,2,stale/low urgency,1,0,106529,,,,0,85e9d3eeca1fbacc43d7804592a0bcb51ab7b5bb570f621a6817a3d5cbe67f0f
thread:issue:106533,issue #106533: CUDA device support does not register allocator to c10::GetAllocator(...),stale/low urgency,2,stale/low urgency,1,0,106533,,,,0,becc57cd09283228cc1c40d71fb30bfba1fc04742755741510f58cb03e4e50c1
thread:issue:106540,issue #106540: Dataset with Queue issue,stale/low urgency,2,stale/low urgency,1,0,106540,,,,0,4e5ff23bf4b049f6448ed3b5f53ffd747158d70d1d375fa621c414e655d4b600
thread:issue:106544,"issue #106544: Branch name in double quotes """"",stale/low urgency,2,stale/low urgency,1,0,106544,,,,0,8d797ed160c10abca0744d436835db99a221161801d6241315a66952a13a90f8
thread:issue:106545,issue #106545: `softmax` to handle dimensions comprised of `-inf`,stale/low urgency,2,stale/low urgency,1,0,106545,,,,0,364ecc8ac4468ab0dd7221e62d00e5c622eb2f3667e7e9e2589925b1f47b88b2
thread:issue:106549,issue #106549: Can't build PyTorch 1.13.1 with Vulkan support,stale/low urgency,2,stale/low urgency,1,0,106549,,,,0,67cedc5434ec5fa40c8914ab37cfa9abe5c6f56a72df8edda15a3892e86d194f
thread:issue:106563,issue #106563: NotImplementedError: Could not run 'aten::multinomial' with arguments from the 'Meta' backend.,stale/low urgency,2,stale/low urgency,1,0,106563,,,,0,2db1d8fe5a6300d40090d0f21317f40dc41a536a473546b811d5eb52905a7dfc
thread:issue:106566,issue #106566: DTensor Sharding prop cache stats,stale/low urgency,2,stale/low urgency,1,0,106566,,,,0,43b29b7198c9da90e226cf9382a1725eb0b0dd3b86706b302e67896e9f560033
thread:issue:106579,"issue #106579: Boolean valued images loaded from disk, when converted to torch int/float tensor, the True valued pixels gets converted to 255 instead of 1",stale/low urgency,2,stale/low urgency,1,0,106579,,,,0,cf36c8d43aa48c7ab2fb215ec4528ae0b11736e458341e891eb9bc4171a9ee84
thread:issue:106602,issue #106602: onednn ops supported in pytorch,stale/low urgency,2,stale/low urgency,1,0,106602,,,,0,2f152edb132fee9559a4b98bd81024cdeb3233079c1fd79f218f2daf4abe1c04
thread:issue:106604,issue #106604: Relu6 not able to process nan values,stale/low urgency,2,stale/low urgency,1,0,106604,,,,0,b0842bcc176186b573ddf3597e9bd5c00989b0578c36279358c6d10b51e2afec
thread:issue:106606,issue #106606: More Performant CachingHostAllocator for Pinned Memory Allocation,stale/low urgency,2,stale/low urgency,1,0,106606,,,,0,58360f3f002169342d755166280409caed686fb878031565ad9b35b715b4e1f6
thread:issue:106608,issue #106608: ROCm & Windows Support,stale/low urgency,2,stale/low urgency,1,0,106608,,,,0,8074399cca1e85eb33cafb5c3b8a8caef4a81fa5f7e60db252883da841ebf247
thread:issue:106622,issue #106622: FFT Samples Inputs with More than Three Dimensions,stale/low urgency,2,stale/low urgency,1,0,106622,,,,0,5e6e6ec92056cca88592a0f84788adebc890be07492bc1c937634063558268e5
thread:issue:106630,issue #106630: no_grad() changes output of TransformerDecoder module during evaluation,stale/low urgency,2,stale/low urgency,1,0,106630,,,,0,3c1bdb1e28438d04e85abac2a13847ebcd876ae260e1e15f5b22bfdc3b968720
thread:issue:106634,issue #106634: Confusing error message for DataLoader with num_workers=0 and non-zero timeout,stale/low urgency,2,stale/low urgency,1,0,106634,,,,0,b428106362849ad1e7245af6a09864e27f8c30649a9269a5a3bce0f447b594cf
thread:issue:106662,issue #106662: Will torch.sparse.mm support multiplying two boolean matrices?,stale/low urgency,2,stale/low urgency,1,0,106662,,,,0,6d8089516b9a2d85b84b207208a9f3a96c47ef15ff4b3a3cfb32e60f5f3a1935
thread:issue:10675,issue #10675: GenerateProposals CUDA implementation,stale/low urgency,2,stale/low urgency,1,0,10675,,,,0,dc73912d12c409784dbf79f03ae458d8d82457a4eca117816e76cd40b210031d
thread:issue:106780,issue #106780: inf and nan are mapped to quant_min in torch.fake_quantize_per_tensor_affine,stale/low urgency,2,stale/low urgency,1,0,106780,,,,0,a4ddbb45ea36cf3e289318605d80a05bac536b1cc7bc82ce2e643d02d447b0fc
thread:issue:106801,"issue #106801: Timer benchmark stores only one time value, and therefore has broken mean/median/etc metrics",stale/low urgency,2,stale/low urgency,1,0,106801,,,,0,00a91addb388e5e6299db8ac4d488051d8f3d1269fb51a05ebe6031dc8fe1462
thread:issue:106828,issue #106828: Use expect tests for error inputs,stale/low urgency,2,stale/low urgency,1,0,106828,,,,0,86fe313134981f1deb74921a62c3a38532458ca3aaea61b93a11ec6235e0d79a
thread:issue:10683,issue #10683: Tensor.register_hook is not passing the tensor object to the hook function,stale/low urgency,2,stale/low urgency,1,0,10683,,,,0,d58e0e23d8fabb51b939939fa4ee8c93af87144ba81372b19ae8898ce81abd66
thread:issue:10684,issue #10684: make install error： [third_party/gloo/gloo/CMakeFiles/gloo.dir/all] Error 2,stale/low urgency,2,stale/low urgency,1,0,10684,,,,0,bd075f856f6173278c8d63b2307cc8b3d970a5beb471c1b011cb17d6f5e194db
thread:issue:10685,issue #10685: [feature request] convtbc with group convolution,stale/low urgency,2,stale/low urgency,1,0,10685,,,,0,d962ddebe09f73f624ed897e435c450e0be088b516f2bd574c9ef557b5e7c78f
thread:issue:106885,"issue #106885: [bug] Dynamo graph break when using pyton module `heapq` (manipulates with `list`s), although succeeds when placing `heapq.py` near the test script",stale/low urgency,2,stale/low urgency,1,0,106885,,,,0,a1fe1f813412df9465ff582435689ad2974a7f1a6cf0505de3c17541198e1bba
thread:issue:10691,issue #10691: Build system doesn't prevent ATen/core from including non-core files,stale/low urgency,2,stale/low urgency,1,0,10691,,,,0,a3e98fe71e252311d82524bc284e5294c61f69bf8a452d5473d21dbb3903dc3c
thread:issue:106931,issue #106931: Other overloads of `_foreach_clamp`,stale/low urgency,2,stale/low urgency,1,0,106931,,,,0,025ceeea4e2134dfa0ba3aea35e7cb3fbbb61815ea5377a8746f67aa5d692009
thread:issue:106971,issue #106971: Found two conflicting CUDA installs,stale/low urgency,2,stale/low urgency,1,0,106971,,,,0,60f9c2231c6b89aaf58beb7728e8a7056b3b0f6515b623132f921516f18613de
thread:issue:106989,issue #106989: `ray` multiprocessing interference by torch import,stale/low urgency,2,stale/low urgency,1,0,106989,,,,0,92c84468d297d68dba191747459e9df527d681112cc2ff71f4863698f44e3125
thread:issue:107011,issue #107011: max_pool3d_with_indices_backward_cuda does not have a deterministic implementation,stale/low urgency,2,stale/low urgency,1,0,107011,,,,0,ade01c67c37aca67974315ec020719da2ff57ad10196b0546b0397bb962bd2f2
thread:issue:107016,issue #107016: cov onnx error,stale/low urgency,2,stale/low urgency,1,0,107016,,,,0,4c199fd6f54c0fa42bc7e210cce023ffc0bfe60087bcb662a9e6c2d441584f5f
thread:issue:107026,issue #107026: a bug about tensor stride,stale/low urgency,2,stale/low urgency,1,0,107026,,,,0,c5f34e9c120caa6e8d65d91f40d4bd694cbd672deb1a382c59969d058243525b
thread:issue:107041,"issue #107041: I want to calculate the matrix multiplication of two Boolean matrices, but torch.mm will report an error. Is there any more efficient alternative?",stale/low urgency,2,stale/low urgency,1,0,107041,,,,0,1412a5d3caa6e2a1f13de205afa9840eb8354657b7d9a73bcbd25b0f42615f5a
thread:issue:107087,issue #107087: RuntimeError with operations on torch.float8_e5m2 and torch.float_e4m3fn data types,stale/low urgency,2,stale/low urgency,1,0,107087,,,,0,0356bd537c21b6b3cb76f726f66ae916f8c2de86aa5e1e5ab8448cc319b86df4
thread:issue:107102,issue #107102: Error when using sparse_coo tensor with optimizer,stale/low urgency,2,stale/low urgency,1,0,107102,,,,0,ad98e782067295be844b2f6c6fbe814ccb4933f41cca6beea95ebed43ad6fa27
thread:issue:107125,issue #107125: combining `vmap` with NN containing `MaxPool2d' leads to discrepancies in output,stale/low urgency,2,stale/low urgency,1,0,107125,,,,0,c3c29f032d0d065e8dfa4f0e4082d3b47a8845d4bcc160e3ddb521363deee993
thread:issue:107130,issue #107130: [docs] Document dtype conversions dtype.to_complex() dtype.to_real(),stale/low urgency,2,stale/low urgency,1,0,107130,,,,0,5cd9b42d1d1ffb2d1668a05ae407979c61700f7b4ede536ac2e436a29cf29766
thread:issue:10714,issue #10714: [feature request][caffe2] extend FC/FCTranspose op to handle 2d bias.,stale/low urgency,2,stale/low urgency,1,0,10714,,,,0,e6da75ed93799e6bd47682b3bff9711da1671c1b50dca95bd9830e85d21fdee5
thread:issue:10719,issue #10719: size mismatch when trying to reconstruct predifined network,stale/low urgency,2,stale/low urgency,1,0,10719,,,,0,8695c5c75e5cbbcb67601bca5a04f2cc668e9ddb8741cb7d572d1e5e0f771cf7
thread:issue:107201,issue #107201: The difference between channels last backward and channels first backward of AvgPool2d on CUDA is too large,stale/low urgency,2,stale/low urgency,1,0,107201,,,,0,11f01b5eaf3a4c5b75ee587190b9960d59eb4532058c372e24f1970d92bd5eef
thread:issue:107214,issue #107214: The difference between input grad computed by channels last backward and the input grad computed by channels first backward of Hardswish on MPS is too large,stale/low urgency,2,stale/low urgency,1,0,107214,,,,0,02e43534ad25b3d1f851728c7bc430524b6d1705e20a216ab910b5643d92fdb1
thread:issue:107217,issue #107217: Documenting `IterableDataset`'s needing `StopIteration` for finite data,stale/low urgency,2,stale/low urgency,1,0,107217,,,,0,c16fa3963c0020b276616e7e1ecfd53deb0c8be2099cb3f531af2759f0b51a66
thread:issue:107253,issue #107253: Conda configuration shouldn't pollute $PATH variable,stale/low urgency,2,stale/low urgency,1,0,107253,,,,0,32638e078ba5b99ee5eaca141d51caddcdd104a23d1a7073c25dd785c7c7dcd6
thread:issue:107294,"issue #107294: 'MPS' training Issue(s) with NanoGPT: -Inf, NaN's",stale/low urgency,2,stale/low urgency,1,0,107294,,,,0,63ffb4b5a44662493ab3af2e1188f58fc157ccc360297650c4e21c646ed5c3ac
thread:issue:107321,"issue #107321: using Union[str, Tensor] as an argument to a torch.jit.script function",stale/low urgency,2,stale/low urgency,1,0,107321,,,,0,118fed6f7f4216b3bdcc89a4eacbb884f3a4ed65d93347c7ff6cf1aab1dfb8c9
thread:issue:10735,issue #10735: Have all C++ modules expose a __file__ attribute,stale/low urgency,2,stale/low urgency,1,0,10735,,,,0,86a03c10265d1f9116677614cc8903aee677552d013bbdefab6defe86b09f6cb
thread:issue:107352,issue #107352: dist.scatter is incompatible with transpose/permute operation,stale/low urgency,2,stale/low urgency,1,0,107352,,,,0,f55c87e36398883c5cd5fada9858a6d9dfdb53556132ef1b0259c2646ef4ac31
thread:issue:107374,issue #107374: DataParallel scatter method split tensor wrong,stale/low urgency,2,stale/low urgency,1,0,107374,,,,0,0ef5911d3813fbe875ddc22b273ca5bb95d61a8b07edff12ecae6889644eccd2
thread:issue:10738,issue #10738: [caffe2] the caffe2 operators document is too old,stale/low urgency,2,stale/low urgency,1,0,10738,,,,0,fa7c6712bfd344205965f2b2113b1e91289e4e551e0f75ed13e1c0cd59c0cc32
thread:issue:107381,issue #107381: sparse_mask method ignores masked-in elements of sparse compressed input tensors,stale/low urgency,2,stale/low urgency,1,0,107381,,,,0,ff054c5ac806c88f8147d1e437d379136728e6f9f6474b1258c161f9d0cc4d38
thread:issue:107387,issue #107387: Incorrect type hint for `torch.library.Library.define`,stale/low urgency,2,stale/low urgency,1,0,107387,,,,0,1646370bc26e122eb91dcdf4b460ac1d2609ed4ee5e79746ff2173f0b719016c
thread:issue:107389,issue #107389: caffe does not respect CUDNN_LIB_DIR when building from source (cmake),stale/low urgency,2,stale/low urgency,1,0,107389,,,,0,1c719328ba5d5d1acdf21db1389d03984e428bbffc8a138ca3babaf25c70bbc7
thread:issue:107392,issue #107392: Overly strict type hints for `torch.utils.data.random_split`,stale/low urgency,2,stale/low urgency,1,0,107392,,,,0,8da07de5edc23814efda43b50b94d07668b8a08fd29a9be6069f885e18e3b337
thread:issue:107394,issue #107394: [LibTorch/iOS] Unknown custom class type quantized.Conv2dPackedParamsBase. Please ensure it is registered,stale/low urgency,2,stale/low urgency,1,0,107394,,,,0,39e2e692de63167c0334fd3d8a010b9082233ccbea8aff52d4fa096ebdb83ee4
thread:issue:107412,"issue #107412: max_pool1d, max_pool2d, max_pool3d Integers for cpu and cuda",stale/low urgency,2,stale/low urgency,1,0,107412,,,,0,b8d2bca6213fb9e702201604ce7e2bdf2986ccfc7aa09ca8974a90885e0a3e99
thread:issue:107433,issue #107433: No checks when running torch.nn.functional.ctc_loss with bogus inputs,stale/low urgency,2,stale/low urgency,1,0,107433,,,,0,bd5ccac4d0d946c6e466da5d5f8f204010f871f766769e9c7fd57301bd18815e
thread:issue:107443,issue #107443: Determinism by using datapipes shuffle,stale/low urgency,2,stale/low urgency,1,0,107443,,,,0,d1b0f4180df394dacf800f506f7e44e8fadc6a98ebc0c04a1ded377056c8e35f
thread:issue:107451,issue #107451: Conversion from COO with two sparse dimensions to CSR with dense_dim specified fails,stale/low urgency,2,stale/low urgency,1,0,107451,,,,0,d94d5c78b14b8166bd674ced118abe1eb651f521cbb45d210ac12400999b0409
thread:issue:107455,issue #107455: Moving tensor to MPS using .to(torch.device('mps') deletes entries from tensor,stale/low urgency,2,stale/low urgency,1,0,107455,,,,0,d3ab52089e2decfb50c5f30d265cdba9c71fcdd9d102cc0fd50d1f037ac65c9e
thread:issue:107478,issue #107478: Adding batched CSR tensors with different sparsities produces an invalid tensor,stale/low urgency,2,stale/low urgency,1,0,107478,,,,0,3e8f985e60e6aafe40e30c34ef653b5204e422baca34614db374621849357fa9
thread:issue:107494,issue #107494: Mismatch in type of error raised when reducing along empty slice between eager and primtorch,stale/low urgency,2,stale/low urgency,1,0,107494,,,,0,207ef8e9fc2586d5a0b1b20c7949b840852efcc743a85b8cf2e3e8b3f8e02dec
thread:issue:107503,issue #107503: [nightly][jit] bad constant exponent (e+38.f) in default_program fused_mul_div_add,stale/low urgency,2,stale/low urgency,1,0,107503,,,,0,538b21e6af26525b85952c5228edaca35363a90753622e87f00eab58fa718228
thread:issue:107514,issue #107514: conv2d wrong results on 3090/3090ti,stale/low urgency,2,stale/low urgency,1,0,107514,,,,0,66dd3af0450d7501580a491e04237bb96388676e66cec6c93d7e1a3b390c76d6
thread:issue:107528,issue #107528: Inconsistent results when running torch.arctanh,stale/low urgency,2,stale/low urgency,1,0,107528,,,,0,9805d033fdad2d38675b44c48f531fed6895d054f2c4961496a786e0eb3d645b
thread:issue:107534,issue #107534: NNPACK slow down M1/M2 Mac CPU,stale/low urgency,2,stale/low urgency,1,0,107534,,,,0,d763a981358670174b6b2de8f4be143bf44917984c9a125ca807603c509087b6
thread:issue:107536,issue #107536: Appending new logs to existing tbevent files when using tensorboard,stale/low urgency,2,stale/low urgency,1,0,107536,,,,0,042f1429714a01e2e0cba526af7a16ca7135da34964e4bd4528bdc8320560ff5
thread:issue:107538,issue #107538: [fx] tracing function with in-place mutation results in unexpected behaviour due to local vars becoming persisted in `GraphModule(nn.Module)`,stale/low urgency,2,stale/low urgency,1,0,107538,,,,0,9821ec76bfa23c89c6ff642f085a54bb977f36407684300a79c4fb26d4fa5797
thread:issue:107539,issue #107539: Integer overflow when running torch.nn.functional.max_unpool2d,stale/low urgency,2,stale/low urgency,1,0,107539,,,,0,c0a6817054a98331c79c1f0505cc84e12a8e16380e02fc808a8116e89d62a551
thread:issue:107541,issue #107541: Integer overflow when running torch.nn.MaxUnpool2d,stale/low urgency,2,stale/low urgency,1,0,107541,,,,0,45ac91690f7903c00dad766235d3f3dd65d73bbdf6b428590ac67099506a4c8f
thread:issue:107543,issue #107543: Integer overflow when running torch.nn.AdaptiveAvgPool2d,stale/low urgency,2,stale/low urgency,1,0,107543,,,,0,9bf596ea1eedb3c4d44ad51805ea55f6d4e86843f3a7e70efd7c699a372c0012
thread:issue:107544,issue #107544: Integer overflow when running torch.nn.ReplicationPad3d,stale/low urgency,2,stale/low urgency,1,0,107544,,,,0,40979ddde08f21c3efebb425253477881a2f315665a0616e05e73385668bf518
thread:issue:107545,issue #107545: Integer overflow when running torch.nn.functional.upsample,stale/low urgency,2,stale/low urgency,1,0,107545,,,,0,8f74bf7ba08eedc5689cde4b4299fcf04e2860c9dd3aebdd6cdb9b2637cd8058
thread:issue:107546,issue #107546: Integer overflow when running torch.nn.functional.upsample_bilinear,stale/low urgency,2,stale/low urgency,1,0,107546,,,,0,4a36b14f2653fff966dbbd1ba8fde21811964261743a5315a02438ca98a34dfd
thread:issue:107548,issue #107548: Integer calculation overflow when running torch.nn.functional.adaptive_avg_pool2d,stale/low urgency,2,stale/low urgency,1,0,107548,,,,0,1480ece1aa365c0e131b7860bdcfb3b2186b5892c55d0903e6a4c25fc14802c8
thread:issue:107550,issue #107550: Integer multiplication overflow when running torch.eye,stale/low urgency,2,stale/low urgency,1,0,107550,,,,0,c06b78919810ea5b1b91dda2c397e875eef8940bc27c21ec47d0e5facb5af98f
thread:issue:107552,issue #107552: Storage size calculation overflowed when running torch.nn.functional.interpolate,stale/low urgency,2,stale/low urgency,1,0,107552,,,,0,c69841e02c4e720d186941d138e8d979109342318e6b3932edd9d04e1a9373c1
thread:issue:107554,issue #107554: Integer multiplication overflow when running torch.diagflat,stale/low urgency,2,stale/low urgency,1,0,107554,,,,0,160787f0d9be4c75bc70d4d3ee026774c9708a1e2b11e1ee55ccb0df9d5c263e
thread:issue:107555,issue #107555: Integer multiplication overflow when running torch.nn.MaxUnpool3d,stale/low urgency,2,stale/low urgency,1,0,107555,,,,0,897e722a002166e8bcad74901f9690d97a4751afdee43cbf28072c14fd9a97fc
thread:issue:107556,issue #107556: Integer multiplication overflow when running torch.nn.AdaptiveAvgPool2d,stale/low urgency,2,stale/low urgency,1,0,107556,,,,0,1ac78e2555093e2645bcd54b3dc2f4d24f088c0cee5f38970a523aeebb68ebdf
thread:issue:107573,issue #107573: caching keys+values in TransformerDecoderLayer for faster inference,stale/low urgency,2,stale/low urgency,1,0,107573,,,,0,79382fb20cc00cbbd1cd1420a0758bf01bad23ef33373039feab8a307fe69e0c
thread:issue:107575,issue #107575: halo，I continue pretrain llama2-13B model ，but save state_dict is about 50GB file,stale/low urgency,2,stale/low urgency,1,0,107575,,,,0,a0f1eca1c41d67df35ff5b616ad2a84bc2de3e074179c78b4a1bc7a3ee6f6384
thread:issue:107581,issue #107581: [LibTorch/iOS] Building with METAL support script is freezing,stale/low urgency,2,stale/low urgency,1,0,107581,,,,0,02d0a831af0ce1d3fd8eb8e88a022c6407d4e0c291f3cc269ce91e36f459479c
thread:issue:107590,issue #107590: Select on a coalesced COO tensor returns COO tensor with coalesce flag set to False.,stale/low urgency,2,stale/low urgency,1,0,107590,,,,0,d6f55f046112f8d28c1ebfb88c4a2245cb481890ba1306c398541c505460e61b
thread:issue:107611,issue #107611: Previous version not found,stale/low urgency,2,stale/low urgency,1,0,107611,,,,0,f74ae376021b7d73f60a2d4f947ac9ab79477caa43346312f92b2fd7e591fc21
thread:issue:107627,issue #107627: ModuleNotFoundError: No module named 'torchgen.code_template',stale/low urgency,2,stale/low urgency,1,0,107627,,,,0,f92a19a5fff673e94413f998708173415daaddf85f86e4edcf348b50b3decf0d
thread:issue:107663,issue #107663: `RuntimeError: expected scalar type BFloat16 but found Float` with `torch.nn.TransformerEncoder`,stale/low urgency,2,stale/low urgency,1,0,107663,,,,0,8540fe2986557902760c010ba882b8234dbec55663878a415c44480b73f671dd
thread:issue:107678,issue #107678: [Torch.fx] Torch fx failed to trace torch extension library,stale/low urgency,2,stale/low urgency,1,0,107678,,,,0,9216dc2c487f25ff3affc95c1e8f36af42e58583b3a0702358ae0f843f73d756
thread:issue:107680,issue #107680: torch.nn.functional.cross_entropy different loss when providing one_hot_target and class weights,stale/low urgency,2,stale/low urgency,1,0,107680,,,,0,7753ebabb75425dc9debb77bbb4d75a307377678f4cabb0abe21143ef47878b4
thread:issue:107693,issue #107693: Inconsistencies when casting to integral types,stale/low urgency,2,stale/low urgency,1,0,107693,,,,0,856aa88ac2d55523a42de893d0bd8f1427b0a3b81f6941f126db73e2ee570bc3
thread:issue:107694,issue #107694: masked_fill_ outputs incorrect results for 'mps' tensor after transpose,stale/low urgency,2,stale/low urgency,1,0,107694,,,,0,f408ce6d9ad14c64fca4bdae6471cb8fa81bee37d6da7c548c1771925cbfa2c6
thread:issue:107695,issue #107695: New variables in torch._ops.py pollute the torch.ops namespace,stale/low urgency,2,stale/low urgency,1,0,107695,,,,0,01a201c60f40a8abe065c0dfc9c4fbf7ea8708c9c2c171af4514fc37d496548f
thread:issue:107700,issue #107700: where.self_out doesn't fail gracefully when inputs have different dtypes,stale/low urgency,2,stale/low urgency,1,0,107700,,,,0,2809e3c6291d31d07143809489dd7c50a971ba9cf853598793f07be108d4fabb
thread:issue:107701,issue #107701: arange.out produces incorrect output when out tensor has dtype long,stale/low urgency,2,stale/low urgency,1,0,107701,,,,0,6babeec40b29028409ec31e29dbfed161f6c2cd76750ae966b5963281c9833ce
thread:issue:107702,issue #107702: Inconsistencies when handling scalars that are out of the range relative to the input tensor's dtype,stale/low urgency,2,stale/low urgency,1,0,107702,,,,0,0dce0ce026a06367884de419a96e7b7da0afed3453767b4978dc6ba68bbb3cfe
thread:issue:107703,"issue #107703: Hardtanh docs are inaccurate/incomplete, since hardtanh behaves like clamp",stale/low urgency,2,stale/low urgency,1,0,107703,,,,0,4b6b333c1827589f2bb832ed7506d379593af9c8f047f313ea9ec3276b5053bd
thread:issue:107751,issue #107751: conv cudnn support integers,stale/low urgency,2,stale/low urgency,1,0,107751,,,,0,f33d186f437b24b6cfffe09e1f403f0cc30168a4d146031c6d5e1beed2c94aac
thread:issue:107780,issue #107780: Add caffe2 ideep/onednn tests to OSS CI,stale/low urgency,2,stale/low urgency,1,0,107780,,,,0,f7da59cc22fed77df74d98af7e97f61f43a1de2c4262a0b75dbdc355b77630b0
thread:issue:107800,issue #107800: [feature request] [discussion] Include basic `ctypes` bindings for `cudart`/`cublasLt`/`cublas`/`nvrtc`/`cudnn` with stock PyTorch,stale/low urgency,2,stale/low urgency,1,0,107800,,,,0,e029aee3acefec80d1af35489c2cdd1d3212fedeaad304d8b6743315c50f5b57
thread:issue:107820,issue #107820: `add_image_with_boxes` method from `torch.utils.tensorboard.writer.SummaryWriter` is broken,stale/low urgency,2,stale/low urgency,1,0,107820,,,,0,d62d7bd29aecb75aac6c334b093acfe882cea3edc50775e6183ac1ec19800cfd
thread:issue:107821,issue #107821: `torch.distributions.Pareto.sample` sometimes gives `inf`,stale/low urgency,2,stale/low urgency,1,0,107821,,,,0,83034223dd47bcce9392a30ec8511426a2505b3e4e2a3906b44db022c6200bfb
thread:issue:107854,"issue #107854: ""file_descriptor"" multiprocessing sharing strategy works incorrectly in dataloading",stale/low urgency,2,stale/low urgency,1,0,107854,,,,0,6d70e9320cceb59193b6797746622d94a54294764258bac8e85c7d9c7459b3ef
thread:issue:107855,issue #107855: About the multi-node example not working properly,stale/low urgency,2,stale/low urgency,1,0,107855,,,,0,0489789643bbaf38dbf1ccf92b45c9ae23128fb39605cd261d88c46b2f784aa7
thread:issue:107864,issue #107864: `C10_HOST_DEVICE` for `std::isnan(c10::complex<T>)`?,stale/low urgency,2,stale/low urgency,1,0,107864,,,,0,4e08528e69ebc5085dc40276f9ec00667ce5a8e5a6b5bf33605f3b787d71a3c8
thread:issue:107873,"issue #107873: [BE] Consolidation of SymNode methods constant_int, maybe_as_int, etc",stale/low urgency,2,stale/low urgency,1,0,107873,,,,0,98e80fc63cc4fb7df50a82453cfb221e5862b8c29bac48422636b9f90c003384
thread:issue:107961,"issue #107961: aten.lift throws error in dynamo backends -> RuntimeError: !at::functionalization::impl::isFunctionalTensor(self) INTERNAL ASSERT FAILED at ""../aten/src/ATen/FunctionalizeFallbackKernel.cpp"":167",stale/low urgency,2,stale/low urgency,1,0,107961,,,,0,9145a3c23663208f35b2e09d03266954ece0179b80afc809f3a015f3480a7956
thread:issue:107999,issue #107999: `upsample_bilinear2d_backward_out_cuda` is nondeterministic,stale/low urgency,2,stale/low urgency,1,0,107999,,,,0,b7fc93546165c71911531a9cb9de77303b37feb6043071f2c0f452906ef98507
thread:issue:10800,issue #10800: [distributed] Synchronization on CUDA side with MPI backend,stale/low urgency,2,stale/low urgency,1,0,10800,,,,0,6b07ad3c349a7af5084bc5651c1c946428b4e2ebca3edc8a9ff4099eafbeeda4
thread:issue:108016,issue #108016: Failure in Initiating Pyotch DDP-style code ( Multi-machine multi-card environment),stale/low urgency,2,stale/low urgency,1,0,108016,,,,0,393ef83af299e4e4708ecfa50449a096168fa63988ddec525c1a261637d8dec7
thread:issue:108023,issue #108023: [FSDP] Ignored modules on meta device seem to be initialized on CUDA device,stale/low urgency,2,stale/low urgency,1,0,108023,,,,0,44f1bffed50f97c57bc06e6fa3bc383fde51f392dc7b7de0dd3e775e67be47b4
thread:issue:108047,issue #108047: DDP training can not accept subnet address in IPV6,stale/low urgency,2,stale/low urgency,1,0,108047,,,,0,ee47788d8b4538dd9c02adc229f92b8b830c1f7cad5176f6fb0d8fd8e89cdd09
thread:issue:10808,issue #10808: [Feature Request] Crop op or ConvTranspose with output_shape,stale/low urgency,2,stale/low urgency,1,0,10808,,,,0,6d2e27a26a917b34c0957480e6994f7f464ac9afd9b164f72a7cec8bdf0a2b39
thread:issue:108102,issue #108102: _sampled_addmm_kernel cause 'misaligned address' with new triton pin,stale/low urgency,2,stale/low urgency,1,0,108102,,,,0,eb7c90aec879b7726ccf2b0caf7a2be8b7e921d7e34e3aa8ba96385af1a63f68
thread:issue:108152,issue #108152: `Tensor.uniform_` uses illegal argument name `from`,stale/low urgency,2,stale/low urgency,1,0,108152,,,,0,8eda7b5026c94970083d8bdc01712f2bc5d97d4610e44b27b0082fc9cf142886
thread:issue:108190,issue #108190: [FSDP] incorrect backward prefetch order when using BackwardPrefetch.BACKWARD_POST,stale/low urgency,2,stale/low urgency,1,0,108190,,,,0,b28c68fd562db3aec6e9c0d50c70ae2b1e9e080274d5cafb573c59cc136e84b9
thread:issue:108197,issue #108197: AdaptiveMaxPool documentation is not detailed,stale/low urgency,2,stale/low urgency,1,0,108197,,,,0,bf13f6bbd117aa07bdfe75f0ac1e67e7b70adadbebbdb945c0ddf0e412f679f8
thread:issue:108224,issue #108224: qnnpack quantized model can not be traced,stale/low urgency,2,stale/low urgency,1,0,108224,,,,0,12a064da1a11ab0ecb3ab4c47033eef443a5c5e0ff55d37134012927e5f1d24d
thread:issue:108231,issue #108231: DistributedSampler class: Change total_size into num_samples,stale/low urgency,2,stale/low urgency,1,0,108231,,,,0,ed43a4533b153eff0981fb4fc1a7226f716977b7d7ec16457d2a5ed0f0cbe007
thread:issue:108244,issue #108244: Pytorch versions without the abi3 flag,stale/low urgency,2,stale/low urgency,1,0,108244,,,,0,d203802c1b8f6c941286d0d9b044bd2060b8822b7747533974c44e8500e60fc9
thread:issue:108277,issue #108277: Transformer performance drop due to slow PyTorch GEMMs,stale/low urgency,2,stale/low urgency,1,0,108277,,,,0,9414ea753abe181cf485081f03544732ad6783b6cfa598c093d7abd4d47c169f
thread:issue:108341,issue #108341: `adaptive_avg_pool2d_backward_cuda` is not deterministic,stale/low urgency,2,stale/low urgency,1,0,108341,,,,0,cae0cbfd559deea9e610b9604c3359a8b5af1f73271acd08abfbb1ed3a376d18
thread:issue:108342,issue #108342: ONNX export constant folding messes up with shared weight deduplication,stale/low urgency,2,stale/low urgency,1,0,108342,,,,0,532da9b4244422cccefda2bd6b6d2cc600e0aaf46b3f495a1cc7def9a78cc06d
thread:issue:108399,issue #108399: Generalize weight prepacking during quantized model deserialization,stale/low urgency,2,stale/low urgency,1,0,108399,,,,0,3cc4f45a720cdb12af7d318d978a8fed92790354cd01339333e5e438c23e6edf
thread:issue:108407,issue #108407: torch.einsum() computes different results on cpu and cuda on A100 GPU.,stale/low urgency,2,stale/low urgency,1,0,108407,,,,0,88d6a8aeba11d4e413c4614520261960770e189b5cb8d4cd2ecbc96dcb50209d
thread:issue:108432,issue #108432: Wrong result of first run with torch.compile() when backend is using torch.jit.trace() and model has inplace operators,stale/low urgency,2,stale/low urgency,1,0,108432,,,,0,78d61943c161485d2077131edf2b5b7d432aa1f8d3554abfd16d9b9d7b9a700c
thread:issue:108445,issue #108445: _foreach_copy_ with scalar second arg,stale/low urgency,2,stale/low urgency,1,0,108445,,,,0,cb737ab80165401f5b8eb5e3aeb8ac3be9541af13fd8ea59a8b5854143fb0e73
thread:issue:108483,issue #108483: Efficient and robust calculation of diag(sparse @ diag @ sparse),stale/low urgency,2,stale/low urgency,1,0,108483,,,,0,53449fb40bbc23b9b2293031957e8a2834d74b520be6534d7b49da5fe14b5ace
thread:issue:108484,issue #108484: DistributedDataParallel to support __getattr__,stale/low urgency,2,stale/low urgency,1,0,108484,,,,0,598078b13831cd4e84673d2357c55685e1f6df86b444976be93516d2218275be
thread:issue:108494,issue #108494: backend-friendly distributions,stale/low urgency,2,stale/low urgency,1,0,108494,,,,0,413957ac6b4336e2edb6d2e5b5d7500810fdd4bf9681ac66c4a3a0c008a0c557
thread:issue:108496,issue #108496: The CPU version of `torch.cummax` is slow,stale/low urgency,2,stale/low urgency,1,0,108496,,,,0,edd8752c57ff2a7398b962bd7d1a28c925abc60288124f702e86d0186e21b488
thread:issue:108522,issue #108522: nn.Transformer has dropout layers that BERT / GPT-2 do not have,stale/low urgency,2,stale/low urgency,1,0,108522,,,,0,d46f35a2735f02099426fc7f3f04833d1fdc01a589ebf2a1d0c09448450b8b10
thread:issue:108569,issue #108569: Call for a deterministic implementation of scatter_add_cuda_kernel,stale/low urgency,2,stale/low urgency,1,0,108569,,,,0,b7336385762c52f7c9ef689e45c750a87a7881ea55336eb68a58428728ac72c2
thread:issue:108627,issue #108627: autocast not consistent across different GPUs (A100 and RTX A6000),stale/low urgency,2,stale/low urgency,1,0,108627,,,,0,222810802fcd600b8354778a9b3cb93a185c46520e900a0bc9881376a62984fa
thread:issue:108642,issue #108642: torch.topk returned values and indices are reordered if sorted=False,stale/low urgency,2,stale/low urgency,1,0,108642,,,,0,63be0d8e54f8224504e5a51ae0b1478458bd9f58fc3faff70a99bd4e76161fa4
thread:issue:108645,issue #108645: uninformative OOM error,stale/low urgency,2,stale/low urgency,1,0,108645,,,,0,b606ed3000a85f9d445c67068984a84d96a7deeabddae0101cbf07a8c8931034
thread:issue:108651,issue #108651: libtorch: runtime error when iterating batch of dataloader,stale/low urgency,2,stale/low urgency,1,0,108651,,,,0,099aa8f104298aaeb5d397d03ef9ab20c205ec17efff6b143eca010537d218b5
thread:issue:108671,issue #108671: avg_pool3d_backward fails on meta with grad_input parameter,stale/low urgency,2,stale/low urgency,1,0,108671,,,,0,623cf8164eb3bed47ae36011124d3c16b3221da2ccb7bbc82ccb32c8ddaaed3f
thread:issue:10875,issue #10875: how can i run two model in a simple project?,stale/low urgency,2,stale/low urgency,1,0,10875,,,,0,c45ef4fc652fc17e93cf558805f7c7b9e7f0750d182a66b2b3c142f0948f773a
thread:issue:108851,issue #108851: about nccl not work,stale/low urgency,2,stale/low urgency,1,0,108851,,,,0,723ac2e54456277b29119612757abaa73d859116853c86c8da6a3bf440571519
thread:issue:108859,issue #108859: Certain torch functions are not handled by torch func wrapper,stale/low urgency,2,stale/low urgency,1,0,108859,,,,0,029f32b36ab0f6a87fa177317f9915cb66f746ec71a26aed71915f7223d382cf
thread:issue:108861,issue #108861: RuntimeError: DataLoader worker (pid 11011) is killed by signal: Bus error. It is possible that dataloader's workers are out of shared memory. Please try to raise your shared memory limit,stale/low urgency,2,stale/low urgency,1,0,108861,,,,0,94f66b7c05f596061762fb32f73f0be4b0b9ef07a289f8fd2cb2b13a8f598de4
thread:issue:108862,"issue #108862: A100 runners down: apt-get install nvidia-docker2, Could not get lock /var/lib/dpkg/lock-frontend",stale/low urgency,2,stale/low urgency,1,0,108862,,,,0,810d650803239c38a241e28c8a744f1705e10e89520cb3a2f7e67e99f862d316
thread:issue:108870,"issue #108870: Inconsistent, platform-dependent torch.ones_like behavior on metatensors",stale/low urgency,2,stale/low urgency,1,0,108870,,,,0,6b51d2266cc3f048b593d08a4a4c25a68bbce52db39fae1a7d6222da52b8dd2a
thread:issue:108882,issue #108882: Inconsistent any( ) between cuda and cpu - Incorrect complex to bool conversion,stale/low urgency,2,stale/low urgency,1,0,108882,,,,0,402304005ccd02b4cdc89b78384d3f81ca9a7e34c7ad4197bc42a703979e2af6
thread:issue:108888,issue #108888: RuntimeError: Unrecognized tensor type ID: ZeroTensor,stale/low urgency,2,stale/low urgency,1,0,108888,,,,0,edd2a17aa3b3ba702846af128cf8bd7f651322596496fcc575fd1bff2803da6e
thread:issue:108900,issue #108900: [c10d] fix functional collective reduce op naming convention,stale/low urgency,2,stale/low urgency,1,0,108900,,,,0,02a77096a4f0debcebb5ae92564cb3e9d7b379355716be312304df9281bf21ce
thread:issue:108948,issue #108948: Add Lambert W function as torch.special.lambertw,stale/low urgency,2,stale/low urgency,1,0,108948,,,,0,6b38e4279e8aa70f65c0e165e86c8d3d640e9477e75a6eaa643fb4a3e7d848a1
thread:issue:108976,issue #108976: About FSDP,stale/low urgency,2,stale/low urgency,1,0,108976,,,,0,4dc97aad1265c7a74d8834b451c1da76947b91e946bda117456531ba2a312e9f
thread:issue:108983,issue #108983: ZeroTensor (and probably neg/conj) doesn't play well with wrapper tensor subclasses,stale/low urgency,2,stale/low urgency,1,0,108983,,,,0,be221e9881367f3150dc5072ac78e143f46382417dc71ec797028622f3e41795
thread:issue:109002,issue #109002: [FSDP] How can I wrap a model that has both nn.Parameter and nn.Module,stale/low urgency,2,stale/low urgency,1,0,109002,,,,0,9e82f44935278fc455485649c99946ae0755c319b6efccbf98f46e290392a97a
thread:issue:109009,issue #109009: TypeError: mask must have dtype bool,stale/low urgency,2,stale/low urgency,1,0,109009,,,,0,372911b288f219cb85c5a13fdace3c2e4fc4a77da5aab0857f2dcec3eefbab35
thread:issue:109010,issue #109010: The following will always fail on NixOS,stale/low urgency,2,stale/low urgency,1,0,109010,,,,0,a12e56fc326104b9fb8273b1b58e9a679a4cc4a885538369e05e0b949c9b9307
thread:issue:10902,issue #10902: LNK2019 error when linking with MSVC [Caffe2],stale/low urgency,2,stale/low urgency,1,0,10902,,,,0,b7d6a5460c3809b7988b3dab6061bebbdf8421ad36c3747a02dbd25af8b50f6f
thread:issue:109074,issue #109074: torch.compile/triton holding GIL during compilation and CompiledKernel call results in deadlocks during distributed training,stale/low urgency,2,stale/low urgency,1,0,109074,,,,0,b2c4adc68c60a57bcf0746d1ad61058aa92ee4e64d4221405d332692fd34d488
thread:issue:109098,issue #109098: Can't initializa NVML,stale/low urgency,2,stale/low urgency,1,0,109098,,,,0,fa44b9af5e651cbfc873746c2863343f5a6866d3c58310dc67898ac63401f576
thread:issue:109112,issue #109112: Unable to install the latest version of PyTorch using mamba.,stale/low urgency,2,stale/low urgency,1,0,109112,,,,0,c0527488e8a985c857171273ccecdb7921e79e6fef23e56bd94cbe74a403be92
thread:issue:10913,issue #10913: Cannot find operator schema for 'ATen' Caffe2 Ios,stale/low urgency,2,stale/low urgency,1,0,10913,,,,0,df6abbce84b1c3908f2d8be661b2443a6debc3ac33eb1d488271b4f8d37b1672
thread:issue:109191,issue #109191: Gradients across different ranks are not synchronized when using DDP,stale/low urgency,2,stale/low urgency,1,0,109191,,,,0,a918d9435b501342dbf84201e6cf516b8ddb99902af5e192205ae893f3b8b8c5
thread:issue:109194,"issue #109194: `torch.mean` not supported for `torch.sparse_coo_tensor`, but `torch.sum` is supported (`scipy.sparse.coo_matrix` does support both `mean` and `sum`)",stale/low urgency,2,stale/low urgency,1,0,109194,,,,0,262bb5da12719545f62a76826c3a54af4e7e56eab36ca052e8b3881f3539a43d
thread:issue:109213,issue #109213: inductor/test_max_autotune having timeout issues,stale/low urgency,2,stale/low urgency,1,0,109213,,,,0,7cd3888862d0aa1199077f3e3018f615552c9f232dbd55895213bd8c090517de
thread:issue:109292,issue #109292: aten::squeeze exported to ONNX as an `If` node,stale/low urgency,2,stale/low urgency,1,0,109292,,,,0,bbae910f0eaadf4e970d21aa8fa4ff81ce152c9bc4cae264aa3b4bd48ef10813
thread:issue:109309,issue #109309: Support the `ExitStack` context manager (or a simplified version),stale/low urgency,2,stale/low urgency,1,0,109309,,,,0,b9a06959b43882183020f5cff1198d8b41f7d8fc4ddaada6160ecb1eb3150fbe
thread:issue:109383,issue #109383: cuda rng state for 2.0.1 cannot be used for 2.1.0,stale/low urgency,2,stale/low urgency,1,0,109383,,,,0,c31277045ac7ba0e5887d172c8a3b66497840a99d44b4eb3fd763e197f233597
thread:issue:109386,issue #109386: Make Fx Generating Incorrect Graph For GPTQ model,stale/low urgency,2,stale/low urgency,1,0,109386,,,,0,bb1b2b8d612c9bce97fec0a4fb4b71f978867d52f701855dcadce0f4ed823192
thread:issue:10940,issue #10940: [Caffe2] Model work Alright in Python but Failed in C++,stale/low urgency,2,stale/low urgency,1,0,10940,,,,0,6120e881cac2593c05ef1f35768435564ff9b69fba6bc3e81dc7d96ce880712b
thread:issue:109401,issue #109401: Interleaved isend and irecv causes hang,stale/low urgency,2,stale/low urgency,1,0,109401,,,,0,d10fc2f4825a10f6e22a12e042ab6883ab9792f9180a630125fc4d2b6b4654e5
thread:issue:109440,issue #109440: [FSDP] supports QLora finetuning,stale/low urgency,2,stale/low urgency,1,0,109440,,,,0,3fd2b9424ccf7ce935b49a72e59c9bc6d3cbc073b95fa4920fb684aab7aa46bf
thread:issue:109445,issue #109445: Memory usage steadily increasing when using back propagation with sparse CSR parameter matrices on CPU,stale/low urgency,2,stale/low urgency,1,0,109445,,,,0,23b8ca6695fa7f3c65bb2c36868b79daf80faf5877f0a242912ee07eb0aa162d
thread:issue:109460,issue #109460: [BUG][pytree] treespec serialization for locally defined classes and namedtuple types,stale/low urgency,2,stale/low urgency,1,0,109460,,,,0,d9583e1c928401640ade1cf68f6d4659216d543ec10783cdfe03561d88ca0390
thread:issue:109462,issue #109462: Inconsistent behavior for in-place operations on coalesced sparse tensors,stale/low urgency,2,stale/low urgency,1,0,109462,,,,0,cd8962e218c3ede3c65313517a553081f11ff5d5795532f037fd687873a4886d
thread:issue:109494,issue #109494: Performance degradation on AMD + A800 when computation is small,stale/low urgency,2,stale/low urgency,1,0,109494,,,,0,dc49fae8696248f673aaff8e07c37a6a0a459436d0435ed7f964d25b49a6a96a
thread:issue:109497,issue #109497: Very big differences in output of `torch.lobpcg` (values and run-time) compared to SciPy on a very ill-conditioned Laplacian matrix,stale/low urgency,2,stale/low urgency,1,0,109497,,,,0,7b3b2e4b98103e6bd544cb7684a87b4479d617b171ef5443fda36a94a9cfc7d8
thread:issue:109539,issue #109539: Torch FX SubgraphMatcher Any / Oneof Patterns,stale/low urgency,2,stale/low urgency,1,0,109539,,,,0,8b84e8179f192035f80113021b33e069689020bd5c41a1e82f20084f85a17b32
thread:issue:10956,issue #10956: Better dev docs for writing native CPU kernels with Vec256,stale/low urgency,2,stale/low urgency,1,0,10956,,,,0,53e751a2676f63de99a9d01211798a149e57ac3946ca19e241be98530efdff09
thread:issue:109586,issue #109586: Max pool with negative integer inputs and channels_last memory layout gives the wrong values,stale/low urgency,2,stale/low urgency,1,0,109586,,,,0,f0a0793ff2b34f460dcfbf9fa5bd358144297dfc4652f213233d31471a3b2aa0
thread:issue:109724,issue #109724: assert_is_valid_input_type is too weak,stale/low urgency,2,stale/low urgency,1,0,109724,,,,0,97ffafb39b29810752f0b87a31fa06a7b5974fc81eef88a22953acfb94f2521c
thread:issue:109747,issue #109747: [RFC][TorchElastic] topology info in training apps/ranks,stale/low urgency,2,stale/low urgency,1,0,109747,,,,0,c3f15bd0ef04b1056df9a2fced8bdf225f679fabe5f1b6993f71e62f93558865
thread:issue:109753,issue #109753: fp16 parity issue with traced code on GPU,stale/low urgency,2,stale/low urgency,1,0,109753,,,,0,86038f5efde3654975766a7289900a55aefd59db604e4203731e7089707c672a
thread:issue:109774,issue #109774: [DDP + Dynamo] Tracing DDP AllReduce (Compiled DDP),stale/low urgency,2,stale/low urgency,1,0,109774,,,,0,bb13b51595c8ab585b90b9d7207133cb9e4f841009608218691e90896ffe2046
thread:issue:10978,issue #10978: [feature request] padding for torch.cat,stale/low urgency,2,stale/low urgency,1,0,10978,,,,0,45b2ee266691e8bfd080402c866941d2e153cede1e5ae7c1f0558a01360137c2
thread:issue:109781,"issue #109781: `torch.embedding`, `weight[indices]`, `torch.index_select` returns random data with indices on meta device",stale/low urgency,2,stale/low urgency,1,0,109781,,,,0,7348bb55cce29245c70981596e102d66922931fd97015688ffc25ce4afc4b98d
thread:issue:109791,issue #109791: Heap-buffer-overflow during tensor unpickling,stale/low urgency,2,stale/low urgency,1,0,109791,,,,0,b5736d6c7eb6d9de2999bc5936ea0ae2e0d3b1d5af5c6f70703cd3a7ce6f3937
thread:issue:109833,issue #109833: Implement Copy-on-write (COW) tensors,stale/low urgency,2,stale/low urgency,1,0,109833,,,,0,db66a86598a96cb9bdb045b2de20682e810fccc335f425de1edb36e4bc55a184
thread:issue:10984,issue #10984: [caffe2] Implement Gather for any value of `axis`,stale/low urgency,2,stale/low urgency,1,0,10984,,,,0,3309afd6be19c9159aca3c1a07b33e82b32b98ec2f1b61454490a1773af5bdd0
thread:issue:10985,issue #10985: Where could I see all of the caffe2 operators and their arguments in python API?,stale/low urgency,2,stale/low urgency,1,0,10985,,,,0,07e0ca1245b351de0f2fa692fae0a5f63b1478768eb20d6db4f82b7ab360e74b
thread:issue:109850,issue #109850: torch._export has no logging,stale/low urgency,2,stale/low urgency,1,0,109850,,,,0,b6a54b1e9b7815f48741e2d9606b170583f9e3d97fc98f75e5ba309509cd6c52
thread:issue:109863,issue #109863: [BUG?] Why Allocator use stream to manage Block?,stale/low urgency,2,stale/low urgency,1,0,109863,,,,0,ee189012007782a250f3db370a8d9528db2bedf17f5f1901d992c71999f720c2
thread:issue:109870,issue #109870: Wrongly returns nan for vectorized complex numbers division on PPC/ZArch,stale/low urgency,2,stale/low urgency,1,0,109870,,,,0,a6897570229932b06f03c17e01fb621448286d0080d47dd2064d69ecd174c35a
thread:issue:109873,issue #109873: Allow try except check for numpy bfloat16 representation,stale/low urgency,2,stale/low urgency,1,0,109873,,,,0,0e9d4aac986f0297d31c7c5ad3ecd46a677331da83b49916a0834bad17e4b1cf
thread:issue:10990,issue #10990: [Caffe2] [Question] How to disable the bias parameter?,stale/low urgency,2,stale/low urgency,1,0,10990,,,,0,ac41f40d7d23c2d07a5b1003fe96aa6b62fe66a6bbd52ffdadc578f54664ee5b
thread:issue:10991,issue #10991: Error when building,stale/low urgency,2,stale/low urgency,1,0,10991,,,,0,0bc0abe13c8ff56a3bc094d05f4384d52f4ca26caf53d8e23fb81f6ce6a9c386
thread:issue:109934,issue #109934: test test_2d_fsdp_integration_fsdp_nested_param_groups failed,stale/low urgency,2,stale/low urgency,1,0,109934,,,,0,23c5fbdb28bc92c9e7604d3f760644683c661f091965b528b992363f50b47a3b
thread:issue:109938,issue #109938: Batching for is_in,stale/low urgency,2,stale/low urgency,1,0,109938,,,,0,763a434e1f1ae5179f460f4d801cac0c9e89a3db97d3182e53f747a5137beb22
thread:issue:109943,issue #109943: Problems when loading PT files und Linux - Duda which are created under Mac Apple Silicon MPS,stale/low urgency,2,stale/low urgency,1,0,109943,,,,0,5904a76990c6121c48ab0a2256a1a8cf533a408f6d2bf39021dbf325818e8153
thread:issue:109958,issue #109958: How to compile torch 2.0.1 version from source?,stale/low urgency,2,stale/low urgency,1,0,109958,,,,0,18664926cfe4d8d2244d94eddb7dc6ea802285cfaf4febcaf45ad5d286b343f5
thread:issue:10996,issue #10996: Multiprocess Deadlock when using np.transpose and torch.stack,stale/low urgency,2,stale/low urgency,1,0,10996,,,,0,56aaf2f6cdf7007853241fb0e7d06c8139309f23c11dedf8c48bfc9ac266f6ca
thread:issue:109968,issue #109968: Dtype hard-coded in DataLoader (for python floats).,stale/low urgency,2,stale/low urgency,1,0,109968,,,,0,1f5775a8049cc11407f2e7c0de1b597f275c3b40fc65418ceef1bf0717474928
thread:issue:109987,issue #109987: Static quantization for Transformer block : AttributeError 'function' object has no attribute 'is_cuda',stale/low urgency,2,stale/low urgency,1,0,109987,,,,0,3fb4a50936221fe67d4e666f13141066eceebbcdac9dd07262e9323e82ef4f10
thread:issue:110014,issue #110014: tan/tanh discrepancies with complex due to jiterator,stale/low urgency,2,stale/low urgency,1,0,110014,,,,0,ac56b281aa320434d6020bab35f9ae1f015cf2c9a0c6cf6687efac2231796bee
thread:issue:110029,issue #110029: Dataloader resetting with num_workers=1 and persistent_workers=True,stale/low urgency,2,stale/low urgency,1,0,110029,,,,0,3f3d89aa34e4af82e72bb72f53cbffc3e726c4a0afe4e6d42cd1a4da4208f8b4
thread:issue:110032,issue #110032: Race condition on shutdown involving PThreadPool and autograd,stale/low urgency,2,stale/low urgency,1,0,110032,,,,0,ee98cea5e820c4fc93e46b19b471c97b7e6188a70ce7bdc7611526035f2bed6c
thread:issue:110056,issue #110056: torch.onnx.export causes floating point exception with core dump for empty slice assignment,stale/low urgency,2,stale/low urgency,1,0,110056,,,,0,4432db453cbc128ce7a9410357c19689583d0304c85b93118e5fa7147dca1946
thread:issue:110074,issue #110074: ImportError: libc10_cuda.so: cannot open shared object file: No such file or directory,stale/low urgency,2,stale/low urgency,1,0,110074,,,,0,914fabd39b5cb66c596e26a988a7d8cea8ce2d29d35a30391dd85afa9355d20a
thread:issue:110084,issue #110084: scatter_add: Mixing 0-dim and 1-dim tensors,stale/low urgency,2,stale/low urgency,1,0,110084,,,,0,1c2b5c32ce26840f551123c526d752d558b6df679d537dff5d0b127ed10dd2b9
thread:issue:110116,issue #110116: Skip cuda kernel launch with torch.sum when dimension length is 0,stale/low urgency,2,stale/low urgency,1,0,110116,,,,0,73ebd1ec9998cbd61bb833ab45444351d1c29adced36a62e2904e42e960ca448
thread:issue:110130,issue #110130: Create a new heuristic TD rule for failures coming from base commit of the pull requests,stale/low urgency,2,stale/low urgency,1,0,110130,,,,0,f9c0c4f9091798bf83e2c48787191ca1dd8114546c1dea1454f313d2e783a9ce
thread:issue:110131,issue #110131: Some ONNX tests have been disabled because of new tensor.split signature,stale/low urgency,2,stale/low urgency,1,0,110131,,,,0,80abc57d492b5f41c3e291e44c6510a92957df7aa532be0af5ff7a5b53a089ec
thread:issue:110154,"issue #110154: ValueError: args contained 2 None's after flattening. When exporting a ScriptModule or ScriptFunction, no args may be None because that breaks type propagation.",stale/low urgency,2,stale/low urgency,1,0,110154,,,,0,283c584f583db54de9067d0bd02f194c0686f8ede3d5ccf3d061774e2f63a3cd
thread:issue:110156,issue #110156: Add _worker_end_fn_t to the DataLoader,stale/low urgency,2,stale/low urgency,1,0,110156,,,,0,52af17fdca6e445f0753553418604edc398fe29726dbfbdd156e810b9aa91629
thread:issue:110194,issue #110194: cudaMallocAsync cause too much fragmentation.,stale/low urgency,2,stale/low urgency,1,0,110194,,,,0,cb002c771ab891d493181a5acd31d1d5c228060498ddc8a69c7a3598c4eb9ad4
thread:issue:110238,issue #110238: PyTorch with non-shared build (building a single shared lib) is unsupported,stale/low urgency,2,stale/low urgency,1,0,110238,,,,0,e1a9614e4078d6857870b7253075de32fa105bddcf625521d153bcf658992ab9
thread:issue:110249,issue #110249: `torch.func.functional_call` does not work with `__torch_function__ ` Tensor-like objects,stale/low urgency,2,stale/low urgency,1,0,110249,,,,0,90f48a50ade17b6706a06f357ce92683384a7cf69ee8158b98c8dd271439bf47
thread:issue:110250,issue #110250: Accessing Particular Nightly Builds Don't Work,stale/low urgency,2,stale/low urgency,1,0,110250,,,,0,de6ccb0b0a852978fc2a98c811516a99a091f84bd49758cf0f8cc1d52c9a0573
thread:issue:110252,issue #110252: cuda/tf32 docs are outdated,stale/low urgency,2,stale/low urgency,1,0,110252,,,,0,2ed19d091b928af590068f6df7f74796ce931f2e80d0f3fd84b23fdaa1b59bf8
thread:issue:110285,issue #110285: TypeError: Got unsupported ScalarType BFloat16,stale/low urgency,2,stale/low urgency,1,0,110285,,,,0,096081c74dc1dba5b9435a295490023084bfda64378503859fa63fd00dc8bee5
thread:issue:110356,"issue #110356: Dropout signature inconsistent between `torch.dropout`, `torch.nn.Dropout` and `torch.nn.functional.dropout`",stale/low urgency,2,stale/low urgency,1,0,110356,,,,0,0b40086b3f320647e036c6b40184a8aaf27685668f5e3fdaafce3a71cffdb4a9
thread:issue:110363,"issue #110363: nn.BatchNorm2d (track_running_stats = True) causes ""modified by an in-place operation"" error when in torch.nn.parallel.DistributedDataParallel",stale/low urgency,2,stale/low urgency,1,0,110363,,,,0,d8de619e29c12550f6a7d00a02496d46496b7d800e2becc914bd4f9632abd819
thread:issue:110366,issue #110366: Categorical Simplex constraint throws error for valid values,stale/low urgency,2,stale/low urgency,1,0,110366,,,,0,e03a50ec4cf5d5e006a33d1809da9c61e45d046669dd72b1e128aaccd5c54ae3
thread:issue:110387,issue #110387: Change unsqueeze(0) to preserve memory layout contiguity of original tensor,stale/low urgency,2,stale/low urgency,1,0,110387,,,,0,b739926ccd02dd9955f07fa9d8f509add62e91497c6e70ab699b5590e3dd7d56
thread:issue:110448,issue #110448: Explore Hybrid (CPU+GPU) Graphs in Scalar parameters,stale/low urgency,2,stale/low urgency,1,0,110448,,,,0,22299f671cf2d1f355790a19a6c93d3d7c4837e32920a751f772c8126dd85822
thread:issue:110461,issue #110461: Custom tensor attributes not preserved with registered functions,stale/low urgency,2,stale/low urgency,1,0,110461,,,,0,b05355b2a049ae561d99cc5685095ed81289342930e12ffcb30ca488afab4174
thread:issue:110515,issue #110515: DynamicQuantizedLinear shows incorrect qscheme after applying eager mode dynamic quantization,stale/low urgency,2,stale/low urgency,1,0,110515,,,,0,f2a856cc174f1c16b5245846c653c47957530bdd5b57cc5059623b2b5d2e0742
thread:issue:110516,issue #110516: Torch Nested Issue With Backward Pass In Transpose,stale/low urgency,2,stale/low urgency,1,0,110516,,,,0,01482e3db14476c368d93ca232175d7177b3f6a156d9d37e72c4345b1c67a43e
thread:issue:11058,issue #11058: A bug in roi_align.cc? [Caffe2],stale/low urgency,2,stale/low urgency,1,0,11058,,,,0,f8d73ff937ed96984cda3f99f4be17afca90016b7dde20ca611100902d839214
thread:issue:11062,"issue #11062: non-shuffling data loaders can affect random states, thus the results of shuffling data loaders.",stale/low urgency,2,stale/low urgency,1,0,11062,,,,0,43695824a8a52cd6b421042fcf78bc4557bdac2d841649775cf6daa615539f01
thread:issue:110698,issue #110698: max_pool3d_with_indices_backward_cuda and avg_pool3d_backward_cuda does not have a deterministic implementation,stale/low urgency,2,stale/low urgency,1,0,110698,,,,0,488b0bdf1f26383e05579079795a9044a4340eec74ca6402cd422498a6ade69c
thread:issue:11072,issue #11072: [...] operator for masked select does not broadcast anymore,stale/low urgency,2,stale/low urgency,1,0,11072,,,,0,c8b0dc29cd26c21707a18a34007d7169797693da8d8d66696303c4d1a705826b
thread:issue:110786,issue #110786: .lldbinit formatters only work when building with clang,stale/low urgency,2,stale/low urgency,1,0,110786,,,,0,e1421f6a28ec85b27efaceb35059a491a893a1157602506a032cc3d14324df8e
thread:issue:110795,issue #110795: Build process failure with torch_shm_manager,stale/low urgency,2,stale/low urgency,1,0,110795,,,,0,c1ad78c2691f0fdc82f385a31dd263a9e54193b02eadcd9a145251f9f8d130fe
thread:issue:110819,issue #110819: Segmentation fault on aarch64 (Rpi4) using Pytorch 2.1.0 & torchaudio,stale/low urgency,2,stale/low urgency,1,0,110819,,,,0,0c3aded17f7254b8fadea7a30814886e9ec9c69bae06df11597f25ed270f16f2
thread:issue:110820,"issue #110820: ""Expected a 'mps:0' generator device but found 'cpu'"" using shuffle=True on DataLoader",stale/low urgency,2,stale/low urgency,1,0,110820,,,,0,c1c3a86e737d1f9a3a603171ca29a22d985207eec61fc969182c6f3e238117f7
thread:issue:110858,issue #110858: Broadcasting matmul is much slower than corresponding einsum,stale/low urgency,2,stale/low urgency,1,0,110858,,,,0,d705203f87526ce336f899433c8054d90ab8ee37dee6b17281a05656db748c88
thread:issue:110866,issue #110866: Gradients (Jacobian) in inference,stale/low urgency,2,stale/low urgency,1,0,110866,,,,0,56e25e8555e5105e7401acfa5aac7cbf2a20d8a156bf67bd1ca91c1a6a12dd1f
thread:issue:110959,issue #110959: `model.named_buffers()` fails if module not hashable.,stale/low urgency,2,stale/low urgency,1,0,110959,,,,0,de674c24c5b058367ca44b5e5e6c8022c84b7d15c379aaf0ee9f06c680863baa
thread:issue:110966,issue #110966: `torch.is_autocast_enabled()` always False on CPU,stale/low urgency,2,stale/low urgency,1,0,110966,,,,0,d3e20bf48a50de91c43e163ae8e7b383f98fa364d063cd6994ffe0720301f2d2
thread:issue:111024,issue #111024: ncu python conv2d.py runs indefinitely after activating cudnn.benchmark,stale/low urgency,2,stale/low urgency,1,0,111024,,,,0,b11e3b3c43fa736b435bb5713768106dae184be1d079b289de590c11d4341811
thread:issue:111025,issue #111025: Unprompted UserWarning,stale/low urgency,2,stale/low urgency,1,0,111025,,,,0,25804c0aadfea775dbc44ec31019aad812a4e6f7a09f8a3db1384ce2e4223598
thread:issue:111033,issue #111033: `CapabilityBasedPartitioner` returns invalid partitions.,stale/low urgency,2,stale/low urgency,1,0,111033,,,,0,0358c6734075e73c1eb2f031127fe7297bb31f2c8d5b8f2da9ec6eeee8c1f4f7
thread:issue:111126,issue #111126: type promotion test for torch.div variants is broken,stale/low urgency,2,stale/low urgency,1,0,111126,,,,0,fee895eb999fd3a2b70329e8dba1e3fdd8cfa24386114ef54484bd78f6cd5464
thread:issue:11113,issue #11113: use_system_nccl flag does not work?,stale/low urgency,2,stale/low urgency,1,0,11113,,,,0,2ba5b7bfe6284dc45d54a78cd1e76c444d58317d492c8c280182d309657323c2
thread:issue:111157,issue #111157: overloads can perhaps be more performant?,stale/low urgency,2,stale/low urgency,1,0,111157,,,,0,51e4ad280eb7d9c1dbad2cfb86342b1f038cbd59fc34970a594dd27c6f16c045
thread:issue:111159,issue #111159: Wrong onnx model from `torch.onnx.export` when using `index_add_` function with duplicate `index` values.,stale/low urgency,2,stale/low urgency,1,0,111159,,,,0,e0614ff5a84ce75522f8788a26be07582951fd9ef29cb366dee7eed311672346
thread:issue:111186,issue #111186: [ONNX][Exporter] Maintain support for exporter arguments export_params and keep_initializers_as_inputs,stale/low urgency,2,stale/low urgency,1,0,111186,,,,0,80d3e579ea8ab152c1c2a61aa73ac12b9db217af87f4c85ba4609719f66a99c3
thread:issue:111187,"issue #111187: torch.distributed.DistBackendError: NCCL error in: ../torch/csrc/distributed/c10d/ProcessGroupNCCL.cpp:1331, unhandled cuda error (run with NCCL_DEBUG=INFO for details)",stale/low urgency,2,stale/low urgency,1,0,111187,,,,0,7ed1adc5f6d7e0f8484354c7e13d9a07b3deccddae7cfc831f52262f061d4489
thread:issue:11119,issue #11119: [Caffe2] Error C2492: data with thread storage duration may not have dll interface,stale/low urgency,2,stale/low urgency,1,0,11119,,,,0,06ddd42efdf211d7fc45427dce539454568abbd84f4803a396af3ba85e16b39d
thread:issue:111210,"issue #111210: RuntimeError: Expected is_sm80 || is_sm90 to be true, but got false. (Using Google Colab)",stale/low urgency,2,stale/low urgency,1,0,111210,,,,0,87acb87d0a9f7979b9a478d175e9d0edf6fe0a8b22e411d070f628b499d9684e
thread:issue:111311,issue #111311: There is a performance drop because we have not yet implemented the batching rule for aten::mkldnn_rnn_layer_backward.,stale/low urgency,2,stale/low urgency,1,0,111311,,,,0,83c99d1243529bd094346b65d7569d5cefef629d68a92155facb05368e5e7af2
thread:issue:111349,issue #111349: CUDA version 12.2 has differential accuracy when executing CPU and GPU,stale/low urgency,2,stale/low urgency,1,0,111349,,,,0,f9463b9dd9172047583ccb400a3fad84849a56f19fba693377c502e634251f73
thread:issue:111351,issue #111351: _foreach_copy_ supports fast copy between cpu and cuda devices.,stale/low urgency,2,stale/low urgency,1,0,111351,,,,0,443080a35a852f595c3dc20932fc267ecce3822682182c607200d3c2e5653b14
thread:issue:111363,issue #111363: Simulating lower memory on GPU does not indicate simulated memory in error message,stale/low urgency,2,stale/low urgency,1,0,111363,,,,0,c41a7c7c6f1cf7e53f2eae7bbc3561a23da5762300ccbdb13d9d567b7be417f3
thread:issue:111366,issue #111366: RuntimeError: CUDAPluggableAllocator does not yet support cacheInfo,stale/low urgency,2,stale/low urgency,1,0,111366,,,,0,c71744aa7f06ea2ce35ac3521c4f19a0803e107549735eb556c0b28998a59bd4
thread:issue:111368,issue #111368: Implement device parameter in Dropout2d,stale/low urgency,2,stale/low urgency,1,0,111368,,,,0,3da9873f9b240cfdc094a1ed5de11de05ef70c9922f7089d46f64e765626d5e2
thread:issue:111374,issue #111374: Tensor.lerp inconsistent when using -Infinity between MPS and CPU,stale/low urgency,2,stale/low urgency,1,0,111374,,,,0,c7529907f9bce6a773950cc1175efcf8d3ab2a1ea4d550902d4ea17d92096af4
thread:issue:111410,"issue #111410: Use of -Wl,--as-needed in cmake config files can leak into third-party users' code and modify their own private libraries",stale/low urgency,2,stale/low urgency,1,0,111410,,,,0,c7731b905a7df9bea3bca8beb5a2a6cda8262d230651d84ab7caa723a071b23a
thread:issue:111454,issue #111454: [ONNX][dynamo] Failed to export cumsum with dtype=float16,stale/low urgency,2,stale/low urgency,1,0,111454,,,,0,1c80e164c997014c08a50c9ae36476b7c5bbac3e1463e31558fff819996ba1ee
thread:issue:111466,issue #111466: yolov5_train,stale/low urgency,2,stale/low urgency,1,0,111466,,,,0,856a69e8d9f336ed75caadf6d3fc5d1cf537b17a2a258958dbdbc41351240daf
thread:issue:111479,issue #111479: multi_head_attention_forward generates different values on MPS compared to CPU,stale/low urgency,2,stale/low urgency,1,0,111479,,,,0,2281fb6fc4b2bbcbeaa7c034215a16e91892a1d914b6b45f251d0eb2dfcc6190
thread:issue:111509,issue #111509: Sparse Tensor Sum Still Does Not Work for PyTorch Geometric,stale/low urgency,2,stale/low urgency,1,0,111509,,,,0,1ac2f39fee393d58ca809e0dfb3832ac04348713e333b93a28eca33db3934f95
thread:issue:111519,"issue #111519: [pt2+profiler] attach aot_id to CompiledFunction, _compiled_fn number without aot_autograd",stale/low urgency,2,stale/low urgency,1,0,111519,,,,0,6d1aa4b9c451b56a14fa1abf075095bca197ac59e87b6821e2b76a7b7183ad82
thread:issue:111525,issue #111525: Functorch FCD breaks with tensor subclasses,stale/low urgency,2,stale/low urgency,1,0,111525,,,,0,93c414afc61df01f1e13ace8ca46fda1d86585bc9d53338d1eff7ea5f3745ce6
thread:issue:111564,issue #111564: misusing percision value in test_cuda function in torch/testing/_internal/common_nn.py.,stale/low urgency,2,stale/low urgency,1,0,111564,,,,0,a3e6bd46d0e2e334496edff87e55d6ad3fae1da4bc5cf9cdb1a0a8ce0f928685
thread:issue:111566,issue #111566: build: failure when building pytorch with TBB,stale/low urgency,2,stale/low urgency,1,0,111566,,,,0,693fa5d7a34e2e76ecab45c80e5f8368d219cad9de5c5e705f4523bf2980bb99
thread:issue:111621,issue #111621: maximum Python version supported is not indicated,stale/low urgency,2,stale/low urgency,1,0,111621,,,,0,7e4231a1fee3071427c6984026f341240dd234d4a758b6538a01f47b0d147077
thread:issue:111623,issue #111623: Missing `ignored_param` when calling wrapper_cls (FSDP) recursively,stale/low urgency,2,stale/low urgency,1,0,111623,,,,0,67fe0f85fe09e78bdb1c6e3923f671b50b70ce01d9ad7652db59890dffe481a9
thread:issue:111654,"issue #111654: Static Linking C++, Op not available at runtime",stale/low urgency,2,stale/low urgency,1,0,111654,,,,0,4a36c7b01cb9cad7e22d119bfaf48bad0ecb6844059277869c7e1e1082b115de
thread:issue:111674,issue #111674: Dynamo Compile samples should record file/line that raised exception,stale/low urgency,2,stale/low urgency,1,0,111674,,,,0,30e78a8f6e1316ef1b2e75bab6e180435888c3cba0708c60c57c52eace201076
thread:issue:111704,issue #111704: Add more flexibility on print / output console,stale/low urgency,2,stale/low urgency,1,0,111704,,,,0,8260501ee1b72b80e740ba8d4b4acfe3b4f0556b845189b61c43325f7dc853fc
thread:issue:111739,issue #111739: grad is inf/nan when using torch.amp,stale/low urgency,2,stale/low urgency,1,0,111739,,,,0,ab55ac9954079452b6b1160f84c222463ac29b49e4d4fc14bb24799380c32a8b
thread:issue:111789,issue #111789: Precisely monitor the collective communication tasks,stale/low urgency,2,stale/low urgency,1,0,111789,,,,0,992161bfacbacb0eecf98891b30d87ae37ab9630f82d0c760c4845d4d0c4b427
thread:issue:11180,issue #11180: A Caffe2 Implementation of Pose Estimation,stale/low urgency,2,stale/low urgency,1,0,11180,,,,0,9887d3ece64be3d5237a05064e0b0667edde729fdb927cafc97171593af66ae6
thread:issue:111806,issue #111806: Revisit security implications of #31875,stale/low urgency,2,stale/low urgency,1,0,111806,,,,0,51409a62d4efbf1dcabeb830611b9cd6de5bce79a299b95846f3b22f607db9da
thread:issue:111819,issue #111819: __slots__ + inheriting from torch.Tensor,stale/low urgency,2,stale/low urgency,1,0,111819,,,,0,0a0ac6508c9f385ac1de2e63099470bdc85f838e53518217bf644249ed754a2b
thread:issue:111902,issue #111902: torchscript file can not be loaded if its saved form the export model produced by torch.export.export,stale/low urgency,2,stale/low urgency,1,0,111902,,,,0,097fd4b9b7985b992c3c5a5e3d4a1ad0a7317b8e1fa522c59beb45ed757aa991
thread:issue:111907,issue #111907: Migration from c10::variant to std::variant causes undefined symbols when linking against older pytorch,stale/low urgency,2,stale/low urgency,1,0,111907,,,,0,ac5798a313eeb94aa3b3a055b886d33028ee1ad250f139979a2899b6cbe61759
thread:issue:111908,issue #111908: Inconsistent Keyword Arguments behaviors in torch.triangular_solve(),stale/low urgency,2,stale/low urgency,1,0,111908,,,,0,e3668206bd13dfdf53060aa293b2fc669f59cfbf3e016f5fb0ef69d49cb9a0bf
thread:issue:111924,issue #111924: [opcheck] Cannot share failures_dict between multiple tests with differing sets of tests they run,stale/low urgency,2,stale/low urgency,1,0,111924,,,,0,b7abcbf7137d4ae565ab8ce7ef358c2d4642851fe7795304fc6ef2bacf1294b4
thread:issue:111925,issue #111925: Resize warning in two argument torch.logical_* with broadcasting,stale/low urgency,2,stale/low urgency,1,0,111925,,,,0,5525196a6a7571576b6393b7c6a5164297409912f09d69238e5bda5fd8360e99
thread:issue:111930,issue #111930: [opcheck] Way to reduce Hypothesis sampling when running opcheck,stale/low urgency,2,stale/low urgency,1,0,111930,,,,0,dec7fe09a09b297de3eeaf36e82444f5b9d23166861aab72a885657ea798e69c
thread:issue:111931,issue #111931: [opcheck] Faster gradcheck execution,stale/low urgency,2,stale/low urgency,1,0,111931,,,,0,de908c2eaba05a757a97d63b0c743d7962b6956ca1d83d8c0a8dde29ca81d924
thread:issue:111950,issue #111950: Operators that return dynamic-shape outputs that require_grad choke in AOTAutograd / unbacked SymInts are reallocated,stale/low urgency,2,stale/low urgency,1,0,111950,,,,0,6088c496b5a550a3835d49a8b4abf0f85e02db7f8bdd806b83c54dfe3f861ef5
thread:issue:111962,issue #111962: [pytree] `pytree.tree_map` does not respect type of `torch.Size`,stale/low urgency,2,stale/low urgency,1,0,111962,,,,0,d4a3aff45c0f1e26a576f70b042ef14603dbb8e2b3b78fbeddcb67fbec1b8b44
thread:issue:11197,issue #11197: [Caffe2] How to correctly add Caffe2 libraries to Visual Studio to write C++ programs?,stale/low urgency,2,stale/low urgency,1,0,11197,,,,0,cec2f161b2fffaadac2350d55b81b94df1a2af54e4651feaaf136ba3db86ce42
thread:issue:112013,issue #112013: RuntimeError for hessian vector product with jvp,stale/low urgency,2,stale/low urgency,1,0,112013,,,,0,c8138f991da39452bf24719290da6f2d2f4ada6ad6cead2d0fa395e5779ab3dd
thread:issue:112015,issue #112015: op scaled_dot_product_attention case different results,stale/low urgency,2,stale/low urgency,1,0,112015,,,,0,933289c97b06d25557445de47c37abc279db892d46c9c1806ddf4217d763fed6
thread:issue:112079,issue #112079: The error type `UserError` in dynamo is confusing.,stale/low urgency,2,stale/low urgency,1,0,112079,,,,0,870b2f2ea79efa3ccda4e1e47a089c76e5f73c6b33f3b4d9af401d43da885347
thread:issue:112089,"issue #112089: Failed to compile: null in call to `__builtin_memmove(__result, __first, sizeof(_Tp) * _Num);` Debian 12, ppc64le, gcc 12.2",stale/low urgency,2,stale/low urgency,1,0,112089,,,,0,c3d862b2badcae12a854eaf317ed2554a7d39357bcf9108280ba49172104f785
thread:issue:112095,issue #112095: torch.export fails on a model with optional parameter,stale/low urgency,2,stale/low urgency,1,0,112095,,,,0,b2b65c45844665e041ec5e3b6568db056883e01e480dc2f47772d083668bd5c0
thread:issue:11213,issue #11213: Ubuntu 16.04 setup.py error - undefined reference to elfLink_Get_FatBinary_From_Object' /usr/lib/x86_64-linux-gnu/libcuda.so: undefined reference to elf32_section_header',stale/low urgency,2,stale/low urgency,1,0,11213,,,,0,696b9a34f7d718362e144ce9a2866404ee96c59e8851a15e7c83190515ded9fb
thread:issue:112251,issue #112251: Optest is unable to xfail on parametrized tests,stale/low urgency,2,stale/low urgency,1,0,112251,,,,0,2a9b86d6babda196696e28646c9bcf55525542bab7338fdd855e6baf03c887a9
thread:issue:112318,issue #112318: IBM AfroHacks at AfroTech Participation,stale/low urgency,2,stale/low urgency,1,0,112318,,,,0,b0432f89c79188ed2e8f144c3403ee8e15a6839796c114567aa1ea31963fb738
thread:issue:112330,issue #112330: Library is included twice QNNPACK,stale/low urgency,2,stale/low urgency,1,0,112330,,,,0,b62fecea18f3c1be5b0b07519c7347a323844d6031ce5f5b0e508658082c363b
thread:issue:112376,issue #112376: Contradictory Error Message for stride Argument in torch.conv_transpose3d(),stale/low urgency,2,stale/low urgency,1,0,112376,,,,0,0fa30e324c2479a4a359e717e4b2c170d70e330853521c59ce8b2deeeea7d413
thread:issue:112377,"issue #112377: RuntimeError: NVML_SUCCESS == r INTERNAL ASSERT FAILED at ""/opt/conda/conda-bld/pytorch_1695392020201/work/c10/cuda/CUDACachingAllocator.cpp"":1154, please report a bug to PyTorch.",stale/low urgency,2,stale/low urgency,1,0,112377,,,,0,e0785ead8588ec499b14472f8ccc1e06e999b2c1dbaf370fddc796bc0f50a8e3
thread:issue:112395,"issue #112395: KeyError during model export while using ""newer"" data types",stale/low urgency,2,stale/low urgency,1,0,112395,,,,0,ea77d69d7451f1c97f6c4b744df99b40e933cea063655d2565fd6b5bb4ae6b14
thread:issue:112443,issue #112443: torch.export emits node outside of Core ATen IR,stale/low urgency,2,stale/low urgency,1,0,112443,,,,0,67f13b850c0edd01a726419af012b82a326a6f275105fd5e5d969abe2022c067
thread:issue:112575,"issue #112575: RuntimeError: ""grid_sampler_2d_cuda"" not implemented for 'BFloat16'",stale/low urgency,2,stale/low urgency,1,0,112575,,,,0,4988f43341677768a586a4e2b9c55745f3d90440031b710813ac10c02d186073
thread:issue:112585,"issue #112585: Fix docstring errors in _VF.py, _appdirs.py, hub.py, _classes.py, _storage_docs.py, _linalg_utils.py, torch_version.py, quasirandom.py, random.py, __future__.py, _lowrank.py, _vmap_internals.py, _sources.py, __config__.py, _lobpcg.py, _namedtensor_internals.py",stale/low urgency,2,stale/low urgency,1,0,112585,,,,0,a882ee003a71e70fcd48aa827abb35ab8980866179bbae4ad7ec8d0fbca54c3a
thread:issue:112587,"issue #112587: Fix docstring errors in __init__.py, _tensor_docs.py, _meta_registrations.py, _tensor.py",stale/low urgency,2,stale/low urgency,1,0,112587,,,,0,6886c4aa1e4045111770bda880f3e6f024f478b36e555476f9db5578b2a30d67
thread:issue:112588,"issue #112588: Fix docstring errors in _torch_docs.py, serialization.py, overrides.py, _utils.py",stale/low urgency,2,stale/low urgency,1,0,112588,,,,0,cb759b8054786ec8d81097035435d9c33e0687cbec00a89701cd359fa29753c4
thread:issue:112600,issue #112600: Fix docstring errors in loss.py,stale/low urgency,2,stale/low urgency,1,0,112600,,,,0,2ade8e1621a6c97b7947185b4f5f54d43f2993df01a5a2532799f65ae386861f
thread:issue:112608,issue #112608: `grad_fn` is not defined in `__torch_dispatch__` arguments when lazy device is used,stale/low urgency,2,stale/low urgency,1,0,112608,,,,0,26aa7aafcf6aabb4a1f440b92d4de163d578c7f4cbfcd1e16d4a8358b35333eb
thread:issue:112638,"issue #112638: Fix docstring errors in spectral_ops_fuzz_test.py, simple_timeit.py, timer_interface.py, op_benchmark.py, _stubs.py, fuzzer.py, compare.py, compile.py, interp.py, hipify_python.py, common.py, end_to_end.py, timer.py, __init__.py, sparse_fuzzer.py, blas_compare_setup.py",stale/low urgency,2,stale/low urgency,1,0,112638,,,,0,32a75a05d38292442b530136ae7430b656b55f438abe440055dd18870c743bc6
thread:issue:112641,"issue #112641: Fix docstring errors in post_localSGD_optimizer.py, functional_sgd.py, _functional_collectives.py, optimizer.py, utils.py, api.py, server_process_global_profiler.py, functions.py, backend_registry.py, __init__.py, options.py, internal.py, functional_adam.py",stale/low urgency,2,stale/low urgency,1,0,112641,,,,0,b947652acca4a87104c76c55fe9675732890ccc2c852fbdef1e981ecabe811d7
thread:issue:112644,"issue #112644: Fix docstring errors in default_hooks.py, post_localSGD_hook.py, debugging_hooks.py, utils.py, hierarchical_model_averager.py, optimizer_overlap_hooks.py, mixed_precision_hooks.py, quantization_hooks.py, ddp_zero_hook.py, __init__.py, powerSGD_hook.py, averagers.py",stale/low urgency,2,stale/low urgency,1,0,112644,,,,0,39b2266120f3e6c94b3c7a00407e324764a3af395da9dc16af98e04e12c1894a
thread:issue:112646,"issue #112646: Fix docstring errors in stream.py, pipe.py, blockpartition.py, microbatch.py, namespace.py, profile.py, tracker.py, portal.py, layout.py, __init__.py",stale/low urgency,2,stale/low urgency,1,0,112646,,,,0,a99cd7ac938d5df6a52a218880bf5cf02e8fe3df692b9bc6b65b25e6a7002142
thread:issue:112666,issue #112666: dispatcher cannot determine dispatch key on tuple input,stale/low urgency,2,stale/low urgency,1,0,112666,,,,0,bbc781dc8484c5cde6ee859d76857c2dc5519b4da16b73f725573f52b93e5237
thread:issue:11267,issue #11267: How to simulate multi-node using single-node with 8 GPUs,stale/low urgency,2,stale/low urgency,1,0,11267,,,,0,63cde197686f841f9302512923977632d67a0a80ebbe83f17d4b61454a3df5a6
thread:issue:11268,issue #11268: How to re-shuffle lmdb per epoch in the caffe2 training process,stale/low urgency,2,stale/low urgency,1,0,11268,,,,0,3d8af02bdb95902f00b3c2afc30a97fa917a8870510b0ae33d8e061430540f54
thread:issue:112691,issue #112691: when huawei NPU,stale/low urgency,2,stale/low urgency,1,0,112691,,,,0,aaab5938aa196ded318a366f625cac616bb3c175ad430a7c11f4a6dc0d88bc96
thread:issue:112714,issue #112714: log_softmax() on CPU and GPU has expected numerical error when used with low-precision bfloat16,stale/low urgency,2,stale/low urgency,1,0,112714,,,,0,88d8e298c9a4e654a898e581bac309fa880ae5d017023c31f4434e4be873be34
thread:issue:112715,issue #112715: Wrong with code_coverage/readme.md,stale/low urgency,2,stale/low urgency,1,0,112715,,,,0,6cff773955898dbc4b5e68ee59ef2fb8ff4a71f4126be8991c78ccadb5069a8b
thread:issue:112716,issue #112716: Quantized model's size gets doubled on optimizing and saving it for mobile,stale/low urgency,2,stale/low urgency,1,0,112716,,,,0,fcee05cea65c928a6392964437100127aaefe4c63c50a8c1dbb33239d750476d
thread:issue:112727,issue #112727: [dynamo] Implement iter fallback (and possibly all iters/generators) as polyfill,stale/low urgency,2,stale/low urgency,1,0,112727,,,,0,cce123ce304ad14b4dc4d63e5bf42bd77b589bca81a30033282b72ad7e8b12e8
thread:issue:112775,issue #112775: Multi-Threaded GraphModule / torch.fx inference raises an exception,stale/low urgency,2,stale/low urgency,1,0,112775,,,,0,c3f4a2a66c23136ad22c1833f39eab4d48ee17380e7d25308489ab509461dba2
thread:issue:112831,issue #112831: dynamo_export successfully export model but fails at onnx.checker.check_model,stale/low urgency,2,stale/low urgency,1,0,112831,,,,0,b680525f5e36aae02a41b356b43bceddce3be506cefec301a7e50c1fa7d8f1ab
thread:issue:112833,issue #112833: Failed to remove output annotation in Quantizer for PT2 QAT quantization,stale/low urgency,2,stale/low urgency,1,0,112833,,,,0,99064b1ec2f7c930287d31168eb40f61a44b2bc961cbed050cab72b40b785598
thread:issue:112847,issue #112847: High dimensional grid sample,stale/low urgency,2,stale/low urgency,1,0,112847,,,,0,5b741d784b3a73e87d7f1a996fc46a32e4dfed0404c2047cdc6577a6247ea5d1
thread:issue:112849,issue #112849: CSR matrix on MPS,stale/low urgency,2,stale/low urgency,1,0,112849,,,,0,7e71c827aa1b88888ffefa67c8b9919134eb19924e5a93ae34c6a22faa9a6370
thread:issue:112857,issue #112857: Vendored FindCUDAToolkit.cmake deviates from upstream in splayed installation support,stale/low urgency,2,stale/low urgency,1,0,112857,,,,0,7841462bff95099d6c286b6f8adba7629e5b50aec13c7f02d16ef3bad0cda2dc
thread:issue:112872,issue #112872: libtorch exports miniz symbols,stale/low urgency,2,stale/low urgency,1,0,112872,,,,0,5364314679edab6e5e3b737b098567f6695352674bb179a5ebe1b1bc47d948ca
thread:issue:112876,issue #112876: How to handle CVE vulnerabilities in underlying operating system?,stale/low urgency,2,stale/low urgency,1,0,112876,,,,0,2848e015f2b25802ace49a33a93fd402c980d3debc5e0bc3cb4895803a893b83
thread:issue:112942,issue #112942: Can't build PyTorch 2.1 from source by GCC 13.2 on M1 MacOS,stale/low urgency,2,stale/low urgency,1,0,112942,,,,0,b7d200c061697b766a019ca6f5251b7c25c11702a021757904e2c7abf2195580
thread:issue:112956,issue #112956: Optim.step() is significantly SLOW on MPS,stale/low urgency,2,stale/low urgency,1,0,112956,,,,0,3e0e3d4a09a472d35d766613a1caf7df630c5cfc218056e5f2742c3ab2a78c0f
thread:issue:113007,issue #113007: `make_fx` failing for `_scaled_dot_product_flash_attention` decomposition,stale/low urgency,2,stale/low urgency,1,0,113007,,,,0,09bb9516a0163317d269dbfde0a352a643273134d8e344d36c62e72e5f5622a2
thread:issue:113024,issue #113024: Upsample bilinear 2d decomposition does not match native implementation for uint8,stale/low urgency,2,stale/low urgency,1,0,113024,,,,0,e85c67f9016d2326dd579725c0d91c841390de20f3e31e54af1149983891f2c9
thread:issue:113034,issue #113034: make_fx produces incorrect graph when used under FunctionalTensorMode,stale/low urgency,2,stale/low urgency,1,0,113034,,,,0,ed94cc2c81d10953d2a2176c626625f18590467a0ca14b4f04c88a14da04e149
thread:issue:113053,issue #113053: [CUDA-12.2] cuSPARSE deprecated support for sparse BSR,stale/low urgency,2,stale/low urgency,1,0,113053,,,,0,5d80e4f365f158f46d95109cdb4d8e8147db880ac9f7c551f8cfe7bfff3969d3
thread:issue:113122,issue #113122: Request to create a tutorial for loading a model dumped by `torch.export`,stale/low urgency,2,stale/low urgency,1,0,113122,,,,0,1e0977ae4bb9e7603cf239d7194f0eaeebdabb327d9f69951d3d2532dc36ea05
thread:issue:113128,issue #113128: NCCL watchdog thread terminated with exception,stale/low urgency,2,stale/low urgency,1,0,113128,,,,0,d89321cb2cac0790de0f1bcfabc6cabb7daaf92bf9f21499924815ac281f48a9
thread:issue:113186,"issue #113186: Fix docstring errors in shard.py, op_registry_utils.py, local_timer.py, file_based_local_timer.py, api.py, distributed.py, cycling_iterator.py, __init__.py, _utils.py, reshard.py, common_op_utils.py, elastic_distributed_sampler.py, utils.py, log_level.py, store.py, logging.py, sharder.py, metadata.py",stale/low urgency,2,stale/low urgency,1,0,113186,,,,0,aaa6207ae6e3cf7acae5420703a7f7b365c7f2028429c9160f77306e594992b9
thread:issue:113187,"issue #113187: Fix docstring errors in embedding.py, _limiter_utils.py, _dynamo_utils.py, embedding_bag.py, tensor_ops.py, api.py, _internals.py, _common.py, init.py, _exec_order_utils.py, _traversal_utils.py, chunk_sharding_spec.py, _trace_utils.py",stale/low urgency,2,stale/low urgency,1,0,113187,,,,0,4af358a35ceb581ce4e9f31c23c678cdd1520cc35eb91de44fb8bbe60cc31d91
thread:issue:113188,"issue #113188: Fix docstring errors in _common_utils.py, _optim_utils.py, _wrap_utils.py, _unshard_param_utils.py, _fsdp_extensions.py, api.py, _debug_utils.py, _utils.py, wrap.py, sharded_grad_scaler.py",stale/low urgency,2,stale/low urgency,1,0,113188,,,,0,f96d5a3b3d0d956ec08f8513bef7554c73ca7869de1469d063fcbb3c210da6d3
thread:issue:113190,"issue #113190: Fix docstring errors in _state_dict_utils.py, _runtime_utils.py, _shard_utils.py",stale/low urgency,2,stale/low urgency,1,0,113190,,,,0,1f796339f7313679083cde45a554eb1cd66d706366bc15e952ac86e5727cb91c
thread:issue:113192,"issue #113192: Fix docstring errors in checkpoint_example.py, basic_strategy.py, op_schema.py, contract.py, redistribute.py, __init__.py, api.py, device_mesh.py, _utils.py, sharding_prop.py, random.py, checkpoint_activation.py, placement_types.py, fully_shard.py, parallel_mode.py, replicate.py",stale/low urgency,2,stale/low urgency,1,0,113192,,,,0,ef3f062d561c3b10717edc747956ef9fc047daba8f22df15eb47fc2457cd8b42
thread:issue:113222,issue #113222: EPOCH_DEPRECATION_WARNING in ChainedScheduler.step,stale/low urgency,2,stale/low urgency,1,0,113222,,,,0,8b2b180cf11aa836ba1ea837f3e71ed5dd606c2a45101d07a150f827644a08dc
thread:issue:113232,issue #113232: [Torch Inductor] Torch Inductor Better Support for GNN workload and Inductor Sparse Compiler,stale/low urgency,2,stale/low urgency,1,0,113232,,,,0,fe83204299ee745d4b8df89b716da43be51418193beed08ae3e5734f1b7fe5a3
thread:issue:11324,issue #11324: Why does DistributedDataSampler not use default RNG?,stale/low urgency,2,stale/low urgency,1,0,11324,,,,0,cef2e77c3353187abddea01608fb341a5ff432e9773563d63ba09e7fa2940acb
thread:issue:113326,issue #113326: `torch._C._cuda_getDeviceCount` inflates system memory usage,stale/low urgency,2,stale/low urgency,1,0,113326,,,,0,f3428a18b1489983c21ea4fd2eeee7b486cbaf4e91bfa0ceb08612bf947b8f6c
thread:issue:113329,issue #113329: Mark aten.normal as a core aten op,stale/low urgency,2,stale/low urgency,1,0,113329,,,,0,e33025fc0f1bd0414b41d4d3133c84c444a6022155918150e1a404d628e15599
thread:issue:113368,issue #113368: Gradient of image rotation,stale/low urgency,2,stale/low urgency,1,0,113368,,,,0,a030e4b733779dfd6ed483810f2b8c864ebdc8fac617fa7db9d6f14a785625f8
thread:issue:113369,issue #113369: When will Huawei Shengteng atlas be supported,stale/low urgency,2,stale/low urgency,1,0,113369,,,,0,544679f512a9f659f9bca8821d31792711f6bc1d702d80f893088516e271c665
thread:issue:113370,issue #113370: Incorrect stride when permuting shapes where a zero dimension is present.,stale/low urgency,2,stale/low urgency,1,0,113370,,,,0,1c90dbe1d619c3c031d2d3f7a2e6c7a8cc8d1f1a36f43c0cb704a5b1671cbaba
thread:issue:113378,issue #113378: [pytree] implement key path API,stale/low urgency,2,stale/low urgency,1,0,113378,,,,0,3993ecd6e8ee13800ff33f0cf0790e73dc97b948958a82182ec253d7e282cc15
thread:issue:113382,issue #113382: TorchXLA - owner @JackCaoG,stale/low urgency,2,stale/low urgency,1,0,113382,,,,0,d6e63fddacefe218e8ac3324a97b5c81593041924519c8d05599cb2a93ef1fb9
thread:issue:11340,issue #11340: Better user experience for using Generator object,stale/low urgency,2,stale/low urgency,1,0,11340,,,,0,7b688b4e66d4260c4d3e2db0bddbf2a26adfd272d0645631fad2a196a136463e
thread:issue:113408,issue #113408: Early testing stop logic for CUDA error looks wrong for instantiated_test with pytest,stale/low urgency,2,stale/low urgency,1,0,113408,,,,0,357268b31e0a66c531511d0aa26474d872b32e1d9682e05fb3770c1c74726dd3
thread:issue:113449,issue #113449: cuDNN error: CUDNN_STATUS_MAPPING_ERROR on gtx_1080/A10 when conv1d is called,stale/low urgency,2,stale/low urgency,1,0,113449,,,,0,db71a1243143235e755182c18fc11f2f5e1c8ceecfe2ec90d3b0b644f82fde26
thread:issue:113456,issue #113456: Expose FakeTensor and FakeTensorMode as public APIs,stale/low urgency,2,stale/low urgency,1,0,113456,,,,0,3641d0656b91abee8afbed66624ebf01393c701e5e89e9e25d59c9ec5c5c8642
thread:issue:113507,issue #113507: Revisit test_edge_op_registration on PyTorch CI,stale/low urgency,2,stale/low urgency,1,0,113507,,,,0,728a8e724d52e8924178e76efdf76394e8b81d95ec8bdd905f33d42be554e11f
thread:issue:113515,issue #113515: torchrun discarding --rdzv-endpoint when it should not,stale/low urgency,2,stale/low urgency,1,0,113515,,,,0,d4c20c93364f3c311680dd9e5f781a8d3b18cd1965586ea940c7b453f44d8e15
thread:issue:113524,issue #113524: _fused_sdp_choice returns wrong backend in pt2,stale/low urgency,2,stale/low urgency,1,0,113524,,,,0,f213c43cac14e8a0f8372b03bce9f6d664eb7e7a764a342e8e5ec73be5e7bf85
thread:issue:113525,issue #113525: [Feature request] `stft` doesn't have `pad_value` argument,stale/low urgency,2,stale/low urgency,1,0,113525,,,,0,3d5efc59e66dde396dbaf0bcf86190d72f112e9f365f45099b27745d794ba622
thread:issue:11354,issue #11354: caffe2 failed to build from source,stale/low urgency,2,stale/low urgency,1,0,11354,,,,0,705549122d16625315fab7a06febc725fcaf4f9c1acc3a5892497a1a18cd6068
thread:issue:113543,issue #113543: [PTD] Logging Improvements Main Task,stale/low urgency,2,stale/low urgency,1,0,113543,,,,0,eb18572469640cf0c39f13f838f865de3dbee8b60a3bdce584956c63958f6a90
thread:issue:113546,"issue #113546: P0: Improve failure trace back when crashed to identify the cause of a crash and the ranks that the crash, output the real traceback at last.",stale/low urgency,2,stale/low urgency,1,0,113546,,,,0,cec512f53bfb0f2778f6cef8dec0f6103d21476757804da2c391077c1d646640
thread:issue:113550,issue #113550: CUDA extension error message doesn't look correct,stale/low urgency,2,stale/low urgency,1,0,113550,,,,0,008bca8169cba0beb5e975ae401561cd8e4180d386861fba8d11ef12e4eae300
thread:issue:113552,issue #113552: 2.1.0 export module：Why can't export() capture reverse graphs?,stale/low urgency,2,stale/low urgency,1,0,113552,,,,0,957d59caebfb235c43bd55e6e68428828124f76c50db9d13abd31a9432e6a79c
thread:issue:113560,issue #113560: t.contiguous() ~10 slower in eager mode compared to torch.compile,stale/low urgency,2,stale/low urgency,1,0,113560,,,,0,7b000d0417afc460ba8d65466827efdac3f9a84bf8e1b7f1a81b0d494e05e332
thread:issue:113579,issue #113579: Creating Gaussian Mixture Models with MultivariateNormal,stale/low urgency,2,stale/low urgency,1,0,113579,,,,0,9c1b821d0a2306436cf0a4f6a59f4d70ff2187a358edee076a9ec0b15ad85e29
thread:issue:113590,issue #113590: Implement FlashFFTConv algorithm,stale/low urgency,2,stale/low urgency,1,0,113590,,,,0,9805cca57abf98102d941ed9dec46fe9c7711f5960fda8d35271ef77e31687ab
thread:issue:113600,issue #113600: Segmentation fault in RPC worker when DataLoader has num_workers > 0,stale/low urgency,2,stale/low urgency,1,0,113600,,,,0,6beb6d62393033bb7a2deb4040d7722daabf90c0a07818e2da3b5b2d241303d8
thread:issue:113621,issue #113621: App crashes when I attempt to run it with an iOS xcframework that relies on the LibTorch-Lite CocoaPod,stale/low urgency,2,stale/low urgency,1,0,113621,,,,0,b6fde995b4269d8634787c6f40d5a762a8f2d004bede4b7cad71690f33aa64bf
thread:issue:11363,issue #11363: Hypothesis operator tests in Caffe2 generate too many warnings,stale/low urgency,2,stale/low urgency,1,0,11363,,,,0,b13156421b196b67bff4b24447c842fa379a64e16b90b2e77dcbc900bdfbbea4
thread:issue:11372,issue #11372: Suggest: DataLoader add device parameter,stale/low urgency,2,stale/low urgency,1,0,11372,,,,0,3dbfcd8e4de35ac74f85da104e54da8908d84cab1c1e7be00f84f2950e91871e
thread:issue:113743,issue #113743: The cuda batched GEMM has a poor performance for bigger batch size with smaller matrix size,stale/low urgency,2,stale/low urgency,1,0,113743,,,,0,61dc043fd0729fca33cdff52e5329c718d02dfd80070a9462ddb67f67fb021cc
thread:issue:113748,issue #113748: New feature: Balanced Sampler,stale/low urgency,2,stale/low urgency,1,0,113748,,,,0,5da139b8816bfd8c629f6b4a91069862f759bb62581dc8f592606d0c7c109eb7
thread:issue:113761,issue #113761: [DTensor][TP] 1-way TP raises index out of range,stale/low urgency,2,stale/low urgency,1,0,113761,,,,0,aab23c0acacb7392aa4f8874964ee2b1cb3bc395d9074c6c556d118bd977b660
thread:issue:113817,issue #113817: [ONNX] ONNX export of simple quantized model fails,stale/low urgency,2,stale/low urgency,1,0,113817,,,,0,fe937aefead8f458ab4d746801f6a1ac287da6cff709d8faa8219c537017a4c1
thread:issue:113831,issue #113831: Update _scaled_mm to support addmm with bias.size == out.size,stale/low urgency,2,stale/low urgency,1,0,113831,,,,0,9459fdfecbab68e12a2b6553645344f543505ba7127ee3c70b463f2cc573e5b0
thread:issue:113847,issue #113847: RuntimeError in use torch 2.1.0 cuda 11.8,stale/low urgency,2,stale/low urgency,1,0,113847,,,,0,485ba3cf6b4bf0eda531d30976dccc4619726ac4d34e925f6ac06bf92c3b5c7c
thread:issue:113854,issue #113854: Timer: Add Standard Deviation for Single Measurement,stale/low urgency,2,stale/low urgency,1,0,113854,,,,0,3dde975851fd251c884238b8f5347f632869ea2c658f59bc2d28382d3575028a
thread:issue:11390,issue #11390: Views created in no_grad block still have requires_grad=True,stale/low urgency,2,stale/low urgency,1,0,11390,,,,0,8a9c6d7521273493dd5f3ffd4e94f7a204564cd61f207b09917b1b0eb008ed9f
thread:issue:113905,issue #113905: Sympy reasoning with true division is broken,stale/low urgency,2,stale/low urgency,1,0,113905,,,,0,040318c7580168b8527a451021bf49f848b634b55884184fc0e6be175774df73
thread:issue:113941,"issue #113941: libtorch_cpu.so: Undefined symbol ""__assert_fail""",stale/low urgency,2,stale/low urgency,1,0,113941,,,,0,fa08a262e51215580637084c01083e97dbab86a098a6aa0da73e3e40dec97bb2
thread:issue:113942,issue #113942: Tensorboard list of tensors as input,stale/low urgency,2,stale/low urgency,1,0,113942,,,,0,19f77c1c4956d16292d484a11b3d74f684462ccf3a6ad0acb146fcc09ef2378d
thread:issue:113948,issue #113948: Unknown CUDA Architecture Name 9.0a in CUDA_SELECT_NVCC_ARCH_FLAGS (compiling from source),stale/low urgency,2,stale/low urgency,1,0,113948,,,,0,f88123c87db817aa288d035de49709144b0d886d8742be8adc5c5ef47e1767f2
thread:issue:113956,issue #113956: iSTFT gives wrong results for some batched input,stale/low urgency,2,stale/low urgency,1,0,113956,,,,0,a6056d28f0ad0215e7b59482ff23d25b4955b680ed163791d9c2abadd9ac2b72
thread:issue:113985,issue #113985: [export] Helper function for specifying dynamic batch size,stale/low urgency,2,stale/low urgency,1,0,113985,,,,0,6257a5c1d3f452599f81433193b683d5892ea3a0da096d7ad84720ff429a11fb
thread:issue:114040,issue #114040: IndexError: map::at with MPI CUDA collectives,stale/low urgency,2,stale/low urgency,1,0,114040,,,,0,9c47c5aa685dc85f0116e7269bbc2ad6fdb367bee1a4d077ab310aec071fbf75
thread:issue:114044,issue #114044: Nan on torch.corrcoef(x.t()),stale/low urgency,2,stale/low urgency,1,0,114044,,,,0,011abaa3a91e3092dbff0b06b9b1378bdd3f99366d3e8156143830cfbbd22400
thread:issue:114080,issue #114080: torch.matrix_exp(x) get inf and nan,stale/low urgency,2,stale/low urgency,1,0,114080,,,,0,3e10fcffc5488c6dc6375c47e792ec89f9cb959396d91d5354d70a311ea32075
thread:issue:114093,issue #114093: Found nn.LazyBatchNorm1d(0) has inconsistency bug between GPU and CPU testing,stale/low urgency,2,stale/low urgency,1,0,114093,,,,0,8c03a98de841789e501de274e5af661c41d6b0ae86aa3350b27d1e64ca946715
thread:issue:114097,issue #114097: [RFC] Support for Redundant Hosts in TorchElastic,stale/low urgency,2,stale/low urgency,1,0,114097,,,,0,330df57afac0b9f0ccf401b902e1afe7eb457907b945b91313ec65a729f42ff3
thread:issue:114107,issue #114107: Add the possibility to pass a `Generator` to `gumbel_softmax`,stale/low urgency,2,stale/low urgency,1,0,114107,,,,0,51f07a723b8a66e4305d00d06afc9ef0ab31c069d3a8e68fb12b9048a253a765
thread:issue:114131,issue #114131: Inconsistency of state_dict loading across devices,stale/low urgency,2,stale/low urgency,1,0,114131,,,,0,b26d272fa830f7e790739bbf0a58971e55eac74d0eb527f4aa5e1ef39b536ef6
thread:issue:114155,issue #114155: Lintrunner on all files fails locally even though it passes in CI,stale/low urgency,2,stale/low urgency,1,0,114155,,,,0,1d5862a20eb4053d3a63272f48c4065de57b191e400ca0049fa2d8fbb10e1d16
thread:issue:114175,issue #114175: Unable to build on CUDA 11.8 due to cutlass incompatibility,stale/low urgency,2,stale/low urgency,1,0,114175,,,,0,dd4d5a31c1ec327ea21cee7c48c606ceb822cc03e849a5c1e9ce07a1730dece4
thread:issue:114188,issue #114188: Unsupported operator error: `aten::to_mkldnn` export to ONNX not supported,stale/low urgency,2,stale/low urgency,1,0,114188,,,,0,f6136e8ef3ee48645aec8b9b06d043ee12d8011663b90af1fc9080c1fd543b40
thread:issue:114216,issue #114216: torch.nn.functional.max_pool2d outputs inf,stale/low urgency,2,stale/low urgency,1,0,114216,,,,0,090a4b631f5e0587ec0315833b8b2ce15098f6b6d8b519fc6144402d46fbfa24
thread:issue:11426,issue #11426: [Feature request] Intuitive error message when input to Linear is not cudarized,stale/low urgency,2,stale/low urgency,1,0,11426,,,,0,983bb0b5d12ef6aeaea998fc69b33ce8427009039a82d31fce2c47d9fcd6f6c4
thread:issue:11433,issue #11433: Assorted issues in Caffe2's Metal ops,stale/low urgency,2,stale/low urgency,1,0,11433,,,,0,6129dcb80a370f4e5f74207f077f2c0f715650ae749d7c18d39712d0627233b6
thread:issue:114350,issue #114350: The 'out'-parameter in torch.matmul() works for 'cuda' device but not for 'cpu',stale/low urgency,2,stale/low urgency,1,0,114350,,,,0,c1b407acc8ab2a596cdc7d48a781c6fa68a523d0ece676268aafe9d58098d31d
thread:issue:114353,issue #114353: add support for linalg lstsq in dynamo,stale/low urgency,2,stale/low urgency,1,0,114353,,,,0,f357a3b607415f319fd579ac130ee83233076dfb39979fee2c3da8b3c9fac308
thread:issue:114356,"issue #114356: [WARNING] could not load None, generating random data instead is too spammy",stale/low urgency,2,stale/low urgency,1,0,114356,,,,0,8c169874c4f6fbdc4d91c922c0ddf394c171892a9ca8ce40a034450b32342a6a
thread:issue:114357,issue #114357: Pytorch DDP across nodes: self._store = TCPStore( # type: ignore[call-arg] RuntimeError: Stop_waiting response is expected,stale/low urgency,2,stale/low urgency,1,0,114357,,,,0,2d9707ac2706d4666c3b74d5f3406f97d467c96879a6da2083958ca89d81bf78
thread:issue:114364,issue #114364: Improve usability of CUDA package by adding description of CUDA,stale/low urgency,2,stale/low urgency,1,0,114364,,,,0,b17552ef2af257b2ed94bced3ad1fb68311845074a5809a29828edadac77b0ff
thread:issue:114388,issue #114388: Respect user-specified USE_ROCM/USE_CUDA,stale/low urgency,2,stale/low urgency,1,0,114388,,,,0,a634d65a55d116834ba7e67141e08f535b96d207890cea755e4d3cb3d99c5979
thread:issue:11439,issue #11439: High leverage TH operations to port to ATen,stale/low urgency,2,stale/low urgency,1,0,11439,,,,0,5757d462756488c31c274f236f06dc647142bb78ccfbd1473872f94704eb40fa
thread:issue:114440,issue #114440: There seems to be redundant calling of findOp in findSchemaOrThrow method of Dispatcher.cpp,stale/low urgency,2,stale/low urgency,1,0,114440,,,,0,3ca2ff3739508c34263b2bc134aa4d6168272f84c50b7945107570180d56a932
thread:issue:11451,issue #11451: [caffe2] caffe2 openmp linking error with xcode 9.0(AppleClang 9.0) on mac,stale/low urgency,2,stale/low urgency,1,0,11451,,,,0,07cb1c46b8d4fd2bc71086ed28b410ad3477693494ed7bf6016f744ed34c2ac2
thread:issue:114565,issue #114565: Create a reproducible build for LibTorch x64 on VS2022,stale/low urgency,2,stale/low urgency,1,0,114565,,,,0,5709c9b9747a83a4c7709e233cbbe851be7eb9d9939bae2f27d1b3e342c9caa5
thread:issue:114581,issue #114581: Flawed testing of onesidedness in istft,stale/low urgency,2,stale/low urgency,1,0,114581,,,,0,b72482239ca70dc4c381d60a8319d7d78ebc7a3a753bf550a471d88e9ba29451
thread:issue:114604,"issue #114604: lax.cond, lax.switch alternatives",stale/low urgency,2,stale/low urgency,1,0,114604,,,,0,e3f3c4ee361422e5262c53579fed665368fd94f5ead020ff0176dc8583d3c0ff
thread:issue:114605,issue #114605: vmap: Transform single-element tensor to integer,stale/low urgency,2,stale/low urgency,1,0,114605,,,,0,b92c5cc52b22853c371446b44b7e9e153c30419c5ed6bf679d09bcf01ebe2710
thread:issue:114629,issue #114629: [DeviceMesh] `init_device_mesh` does not support CPU-only,stale/low urgency,2,stale/low urgency,1,0,114629,,,,0,17b1ba0f6af828d6f2ec0c8e9d3481bbde14c5160902fbd820e70ddda0cb2150
thread:issue:114632,issue #114632: [DeviceMesh] back DeviceMesh initialization by custom_pg,stale/low urgency,2,stale/low urgency,1,0,114632,,,,0,c18ff45e1bf8e169f7360ea5233edfa4a048bf3419cc16e12c4eed94565f7fcf
thread:issue:114652,issue #114652: __torch_dispatch__ fails on functionalization when lazy device and `out=` arg is used,stale/low urgency,2,stale/low urgency,1,0,114652,,,,0,dae00b80a131a92d86ad3056ebd4181ffc441975a3222e34b55b6aca6247605a
thread:issue:114672,issue #114672: torch.fft.ifft crashes for empty input,stale/low urgency,2,stale/low urgency,1,0,114672,,,,0,8b63e812bd7455cee956316df49c66d6051642d64a36721095f893f184dd0a81
thread:issue:114679,issue #114679: Unexpected Results in PyTorch Tensor Operations with Python Scalars,stale/low urgency,2,stale/low urgency,1,0,114679,,,,0,cd006be8ad61669efabf3521a68ab9f3806fabc28baa63d6dd5dbda432ecc459
thread:issue:114796,issue #114796: Add head_mask for transformers,stale/low urgency,2,stale/low urgency,1,0,114796,,,,0,4d4cc9c25c582abfa20d22b56b6804d548b507daa27c5725a891244ac27ca42c
thread:issue:114805,issue #114805: Unsupported: call_method UserDefinedObjectVariable(_profiler_enabled) __call__ [] {},stale/low urgency,2,stale/low urgency,1,0,114805,,,,0,ed8123c1ab3c5df7e4f7be4549122943c0749d29157f3ce5078bd9af128cea9d
thread:issue:114808,issue #114808: Efficient Cholesky and QR updates,stale/low urgency,2,stale/low urgency,1,0,114808,,,,0,0741b26a87b6e0bc55c458d20829e578c63719e2698404ac0135fe84e2280a09
thread:issue:114871,issue #114871: Sparse block tensors (`torch.sparse`),stale/low urgency,2,stale/low urgency,1,0,114871,,,,0,90c5008bcfac3abeef5ed58c40ad83969c88c892fb70eecf8f57c0011ef76b73
thread:issue:114928,issue #114928: `torch.nn.utils.rnn.pack_padded_sequence` heap-buffer-overflow,stale/low urgency,2,stale/low urgency,1,0,114928,,,,0,6941e0d58e0f9f8a729f027f80fc830a25802df5d1f24a31f2cf6b1f5236c34d
thread:issue:114935,issue #114935: print statements used in torch/utils/cpp_extension.py,stale/low urgency,2,stale/low urgency,1,0,114935,,,,0,e695679d6d2ebf23f145a21573e2c9d53d84dd91d2e0e2d7acc0d9f3a6c70b94
thread:issue:114943,issue #114943: torch.einsum may choose a strategy for which there is not enough memory,stale/low urgency,2,stale/low urgency,1,0,114943,,,,0,ccdef4c2dc1027f3dbd4e734e532a21b51b6657a8723eeffda16d506951ed6f9
thread:issue:114945,issue #114945: torch.multinomial - Unexpected (incorrect) results when replacement=True in version 2.1.1+cpu,stale/low urgency,2,stale/low urgency,1,0,114945,,,,0,ee62cac1c28bd291f9ee9a95dd68971798582ecd74dd861eaef58f4247bbfdb4
thread:issue:114948,issue #114948: conda package _openmp_mutex makes pytorch dataloader slower when set num_workers > 0,stale/low urgency,2,stale/low urgency,1,0,114948,,,,0,c23878ff728c064dbac74be4d4d6284231eadfcb20dec50fcda2c9a983433463
thread:issue:114949,"issue #114949: Nvidia P100, where to disable upcasting? Plus kernel image missing.",stale/low urgency,2,stale/low urgency,1,0,114949,,,,0,7dceecaef4375cd848f61ceaa7254fe78f12b0485278f6d536764fe3ccb9409a
thread:issue:11496,"issue #11496: Tests with ""."" in the name cannot be run standalone",stale/low urgency,2,stale/low urgency,1,0,11496,,,,0,7bbd32e19eaf92b0c30b8b2df6a3b4c0ed7be5c13d1fea4af02b629e994fb85b
thread:issue:114962,issue #114962: `CUDAExtension` no longer works with `ccache`,stale/low urgency,2,stale/low urgency,1,0,114962,,,,0,70c2c75b24a664858c9e5cd2f923560a0c70cc90677c5916a38af129f24a2fcf
thread:issue:115020,issue #115020: [Libtorch/iOS] Build for iOS as dynamic library fails in linker phase.,stale/low urgency,2,stale/low urgency,1,0,115020,,,,0,257779ba4cb7ac758071ee56f5322659cb3f8026f64c32660142ca22527406b4
thread:issue:115024,issue #115024: InstanceNorm1d throws bad warning,stale/low urgency,2,stale/low urgency,1,0,115024,,,,0,c200f0bc6b8d396d2a27717c71a3f0420c3496afe5dbd4725e1e86cf46bb8df7
thread:issue:115034,issue #115034: libtorch can't switch between with and without layers by is_training(),stale/low urgency,2,stale/low urgency,1,0,115034,,,,0,6d6c58e632647e3afaf609e186daf2767101e976ae5b8c6260f113ff5395f122
thread:issue:115058,issue #115058: [DTensor] enforcing foreach for optimizer causes jittering,stale/low urgency,2,stale/low urgency,1,0,115058,,,,0,6db186f19776f86636ddc1e8cf7f07fd2986b69b201ad0230f515509e519bbf5
thread:issue:115064,issue #115064: torch.nn.PairwiseDistance，The results vary widely from version to version,stale/low urgency,2,stale/low urgency,1,0,115064,,,,0,e7ac48f3ea9905e611c64300da89ac14aeb5473cbc0c023016e872ee566ef884
thread:issue:115077,issue #115077: RuntimeError: cuSPARSELT not supported on your machine. When I was calling: torch._cslt_compress(),stale/low urgency,2,stale/low urgency,1,0,115077,,,,0,67154d973952ac26b0eef8a71befdc17b5bcb81a53907b7c12160daddecc47a7
thread:issue:115122,issue #115122: [dynamo] Graph breaks from copy.deepcopy,stale/low urgency,2,stale/low urgency,1,0,115122,,,,0,b0b5846d0c2590d81c6224b48d4a2c83b6ecb4775b6303f933ddd2978f231098
thread:issue:115138,issue #115138: ImportError: /lib64/libstdc++.so.6: version `GLIBCXX_3.4.20' not found,stale/low urgency,2,stale/low urgency,1,0,115138,,,,0,305b600d7a067e433f98e21d623cc3f7c1ca4035d44db6f6befac291541b1be2
thread:issue:115152,issue #115152: scatter_ op after permute op not working correctly on mps device,stale/low urgency,2,stale/low urgency,1,0,115152,,,,0,cc0c2a80f070c192ba097c98fdea03dc770efccd6c0bfbaf31bf07281b8dccfa
thread:issue:11516,issue #11516: at::Device makes it very easy to write buggy code,stale/low urgency,2,stale/low urgency,1,0,11516,,,,0,729d956942ac266f5c1017530d38afc62632cf54319b8c3b8808bbafeea1001f
thread:issue:115164,issue #115164: nansum with complex support,stale/low urgency,2,stale/low urgency,1,0,115164,,,,0,2d60b35b999a86815f9a0b7ef0678c94352fbd1556a3687420bea12973f4472f
thread:issue:115202,issue #115202: Auto skip the entire test when @parametrize takes an empty list,stale/low urgency,2,stale/low urgency,1,0,115202,,,,0,b52d3c4fdffbdf66cd7ab9b865bcf087e656ce65d6334fc62f3ad898ecc95988
thread:issue:115257,issue #115257: Compiling source code reported an error,stale/low urgency,2,stale/low urgency,1,0,115257,,,,0,3c0607d5e6ab37f362fda450d327c59d380a1328a7ef9b1e8495a57487b6834e
thread:issue:115301,issue #115301: inconsistency for enableTiming / FlightRecorder / duration_ms,stale/low urgency,2,stale/low urgency,1,0,115301,,,,0,a06b61e5540c0dced2fa6e587e03e2133c8e1cde50f4743f8cff5d5238d74d43
thread:issue:115309,"issue #115309: [distributed] add a ""small"" transformer model to the test suite for testing usage",stale/low urgency,2,stale/low urgency,1,0,115309,,,,0,3dd227818bd1fe4ea4d52b547b19be7ebf87110b8dd23b0aad615c1817650ba4
thread:issue:115331,"issue #115331: Please increase ""Number of CUDA devices"" recognized by pytorch.",stale/low urgency,2,stale/low urgency,1,0,115331,,,,0,b8bfb632a90d9d53fc3f9261e2496b4dfba51dfc6eda2f5dbcee850bda69098a
thread:issue:115339,issue #115339: Add support for measuring elapsed time in cuda-graphs,stale/low urgency,2,stale/low urgency,1,0,115339,,,,0,ff82be3a642fb4edbd1d375d412143bb27eda1761d5a67b68d9ec0592c92909c
thread:issue:115340,issue #115340: RuntimeError: CUDA unknown error even after sudo modprobe -r nvidia_uvm && sudo modprobe nvidia_uvm,stale/low urgency,2,stale/low urgency,1,0,115340,,,,0,10bcc4f2c461fdd49bacdcb734b8af06069b1a005dc548d8a1bd49a5942973f8
thread:issue:115342,issue #115342: torch.func.grad not working with list comprehension over dictionary,stale/low urgency,2,stale/low urgency,1,0,115342,,,,0,94fcbb1ad05a45a86883991642000bbe28ca18354569df2d0ac203533e250899
thread:issue:115350,issue #115350: Enable skipped MPS OptimizerInfo tests,stale/low urgency,2,stale/low urgency,1,0,115350,,,,0,8c67f227eec7ebd00585f708af765ec242c18f08f5b93508990dc6c32735273a
thread:issue:115362,issue #115362: torch._fused_sdp_choice for fake and real tensors returns different results,stale/low urgency,2,stale/low urgency,1,0,115362,,,,0,836cc07ac65e44eb6bb482535db209bf49611697321f6fdd51a5625151f7ec3c
thread:issue:11537,"issue #11537: [CLEANUP] Context functions should return TypeExtendedInterface, not Type",stale/low urgency,2,stale/low urgency,1,0,11537,,,,0,865bf02c9d2dd96f4bd52596dce104437693cfe242b3aae4c0d26fd6e58f2b23
thread:issue:115418,issue #115418: max_pool2d_with_indices_backward_out_cuda: remove useless code gradInput.zero_();,stale/low urgency,2,stale/low urgency,1,0,115418,,,,0,51a7e8afd3876949ca5f37a21f197fa953daeec45c3d306bb2624f6555d09791
thread:issue:115421,issue #115421: Document PyTorch docker release policies in RELEASE.MD,stale/low urgency,2,stale/low urgency,1,0,115421,,,,0,4063cec0faa65f6608114728c73859a40efe9ab81b9f0ccc0778201e00e08577
thread:issue:115422,issue #115422: string representation method for empty masked tensors fails,stale/low urgency,2,stale/low urgency,1,0,115422,,,,0,106fcec61cd6b3db6502ad1dedd1cf362c0a14aab6f60581c20792ae41509b9e
thread:issue:115472,issue #115472: Potential issue with custom transformer masks when using fast path and batch_first=True,stale/low urgency,2,stale/low urgency,1,0,115472,,,,0,2a5f12e85ad1997d1159b3beeb348539ce5f4caa7e8be0ad3059ad3c0fe3f3af
thread:issue:115500,issue #115500: Batchnorm NAN in amp autocast mode.,stale/low urgency,2,stale/low urgency,1,0,115500,,,,0,222e372610b2f8494f53894b986b094cbb5af00dfb9ef3a7c8cc8fe138027a17
thread:issue:11551,"issue #11551: CrossEntropyLoss, ignore_index does not prevent back-prop if the logits are -inf",stale/low urgency,2,stale/low urgency,1,0,11551,,,,0,c9d3528e488e0cdc031b14130dce74421efddd5ce3ff207b8dd7017d5cdae5bb
thread:issue:115526,issue #115526: Can PyTorch support using custom compilers for model's components and the Inductor compiler for the model?,stale/low urgency,2,stale/low urgency,1,0,115526,,,,0,29dd4a389ef6db75b9e6196d6c459a0421637b336de0f4c7bd2b4e021ce67205
thread:issue:115610,issue #115610: [Bug] Big difference between the output of Conv float precision and double precision,stale/low urgency,2,stale/low urgency,1,0,115610,,,,0,6732bbd8f711a7141d4a2410180ba3e1352b8c8d29af9b2925f7172c339faa78
thread:issue:115614,issue #115614: Make test_public_bindings.py a lint,stale/low urgency,2,stale/low urgency,1,0,115614,,,,0,74b190e55a64817daee81d2b3cba9d6ddbf24dd089c172a3b8e434a58735be4a
thread:issue:115618,"issue #115618: RuntimeError: handle_0 INTERNAL ASSERT FAILED at ""../c10/cuda/driver_api.cpp"":15",stale/low urgency,2,stale/low urgency,1,0,115618,,,,0,161de537389f924670851ffdb2ed1fac82947b9b69a310b74b5c08d0da9b9ff5
thread:issue:115624,issue #115624: `amax` fails on `masked_tensor` with multiple negative dims,stale/low urgency,2,stale/low urgency,1,0,115624,,,,0,32276a2dbc8a21b620fae3c6db59ff8edac3a7dfe4ae21a3e3ef7fe2630cc6f3
thread:issue:115659,issue #115659: Clean up AWS credentials from PyTorch CI and replace them with OIDC,stale/low urgency,2,stale/low urgency,1,0,115659,,,,0,3e414c052ed6c7c6e398102b379cc5d5fdec91ce398418a2d83da4a79c9eca23
thread:issue:115672,issue #115672: Latest PyTorch is not buildable against CUDA-11.2,stale/low urgency,2,stale/low urgency,1,0,115672,,,,0,55fde6eb8f7bfcc3160d83ced70e620c439ac580837b441a9009341372582ff3
thread:issue:115706,issue #115706: [export][functionalize] Complicated subgraph from functionalizing inplace op on sliced tensors,stale/low urgency,2,stale/low urgency,1,0,115706,,,,0,24c16ecef2f238ed407199ef19effee6a41b1f0b0e4150a340b256c7996125b3
thread:issue:115735,issue #115735: Document required semantics of allocation functions in `CUDAPluggableAllocator`,stale/low urgency,2,stale/low urgency,1,0,115735,,,,0,3e88eb6124b0a4f45029e5dcf3cc852deddce3d1748112a031978ec5f122ec31
thread:issue:115785,issue #115785: arctan2 fp16 error when optimising,stale/low urgency,2,stale/low urgency,1,0,115785,,,,0,06b665c22e6e46e9b166f7f4aaa62ae086248e215363ca99aba13ba4380a0948
thread:issue:115806,issue #115806: Merge all 3 mypy lint jobs,stale/low urgency,2,stale/low urgency,1,0,115806,,,,0,7adddefd7ac1983276f2f7f4d5b702a703e545e698d4ff4ffef6c59e17daa5cf
thread:issue:115820,issue #115820: Question on XNNPACK support for PowerPC,stale/low urgency,2,stale/low urgency,1,0,115820,,,,0,a7f49df2d3acf423f53da9e022b88386a27d7ce3a6e53717c90d0a29f0861f06
thread:issue:115966,issue #115966: Adam Eager Benchmark Failures,stale/low urgency,2,stale/low urgency,1,0,115966,,,,0,8c16e07e8549b976d1c069d97e84f96491503dfcfc6ade45e8350621e497682d
thread:issue:116012,issue #116012: Issue with Protoc while building PyTorch for RISC-V,stale/low urgency,2,stale/low urgency,1,0,116012,,,,0,d839ffb90f5871802caff09271b450662c3424463cba4c072342e23398a9372c
thread:issue:116013,issue #116013: [Dynamo] bytecode transformed by Dynamo is not serializable by marshal,stale/low urgency,2,stale/low urgency,1,0,116013,,,,0,53dbb5471169ca66d238fd85701e29c0f46675626f44092835ddd7f855a9c483
thread:issue:116018,issue #116018: Inconsistent behavior of torch.einsum on float32 and complex128,stale/low urgency,2,stale/low urgency,1,0,116018,,,,0,399675c66c76923a6456fa2ab865ad8efdc137de1d2f70e97e2e6abee432aebd
thread:issue:116052,issue #116052: Add a capturable impl to ASGD single tensor,stale/low urgency,2,stale/low urgency,1,0,116052,,,,0,e733179f868f4c30b0ed9117899743a277a3b5177885c571adec271e25f1d0d3
thread:issue:116077,issue #116077: TransformerEncoderLayer and TransformerDecoderLayer do not work,stale/low urgency,2,stale/low urgency,1,0,116077,,,,0,bb91dc6539f8af6ec8803058ef56fd39d366ac82b2f34db04b08e5eb2ac407d8
thread:issue:116101,issue #116101: [DTensor] Support API to shard to parent mesh,stale/low urgency,2,stale/low urgency,1,0,116101,,,,0,71cee79bb747adf6c86155cbc89e7f2f757eaf181bdeb654aebe35a492eb7b4a
thread:issue:116102,issue #116102: torch.func.grad with respect to intermediate outputs,stale/low urgency,2,stale/low urgency,1,0,116102,,,,0,fe7d62de1d550550e366cfdfc1d4f810c8c6f789135dc156d013755f99e486f9
thread:issue:116117,issue #116117: WeightedRandomSampler doesnt keep expected ratio betwen samples,stale/low urgency,2,stale/low urgency,1,0,116117,,,,0,f10a865917129199dd1fd9c2eb19b121cc0aae4fcd9b0ed44b29a5f8722754db
thread:issue:11612,issue #11612: [feature request] Triangular Matrix Representation,stale/low urgency,2,stale/low urgency,1,0,11612,,,,0,30f5b343ab4a20dfde60b3ef1ab586ae90547cf27d1aea5e5e901178741df5e8
thread:issue:116120,issue #116120: bfloat16 on 'grid_sampler_3d_cuda' not implemented,stale/low urgency,2,stale/low urgency,1,0,116120,,,,0,8c3277e8cc3fa4418595db9ec6e810355d18daa572ade354bc8d74e40aee1bcc
thread:issue:116182,issue #116182: [ONNX] GFPGANv1.pth to onnx conversion error,stale/low urgency,2,stale/low urgency,1,0,116182,,,,0,b9a834fe9dfb6ba0e1d643498ad0ce7e173856e9eeca4f87fa2711ee4e76bf5e
thread:issue:116184,issue #116184: Potential cublas handle leaking,stale/low urgency,2,stale/low urgency,1,0,116184,,,,0,c98a2e3a94e0edb7250942fcb2496dec1460926df5cfa1a7bc3ee06c0c6de201
thread:issue:116237,issue #116237: support Tensor type for scale in scaled_dot_product_attention,stale/low urgency,2,stale/low urgency,1,0,116237,,,,0,3b1f600811918e2a7550ebe682df1c9a30d468885177aa26d6a6d6d6dcb2c261
thread:issue:11624,issue #11624: Add min mode to embedding bags,stale/low urgency,2,stale/low urgency,1,0,11624,,,,0,89f656c7e178cff9af0bc296e71391445c4427ff30bc1c15d5ac1a2f624af798
thread:issue:116242,"issue #116242: Building project using libtorch results in ""Failed to find nvToolsExt""",stale/low urgency,2,stale/low urgency,1,0,116242,,,,0,d3ad3b7ebec5db995bdfd7ea6ec31b709acfbd3a182f0b5e1cc60a01d09c9edb
thread:issue:116255,issue #116255: PrivateUseOne backend does't support tensor shollow copy,stale/low urgency,2,stale/low urgency,1,0,116255,,,,0,1631d7036170326044ea8ff3a4c9b33a6222efbec59f357c641ce12192b0d71d
thread:issue:116320,issue #116320: RuntimeError: CUDA error: an illegal memory access was encountered using vmap and model ensembling call for cuda system,stale/low urgency,2,stale/low urgency,1,0,116320,,,,0,aa2dcaccac40248fe2f67735523f531c3afcbfab9449f7aec9e48e663db7e5c9
thread:issue:116322,issue #116322: [RFC] Intel GPU Runtime Upstreaming for Allocator,stale/low urgency,2,stale/low urgency,1,0,116322,,,,0,8659517477c966fedc6f52078851a0717c744f1cbca2f58cb25527a9b586740c
thread:issue:11633,issue #11633: DataLoader: Could not wrapper a exception in threads,stale/low urgency,2,stale/low urgency,1,0,11633,,,,0,a1494568a2fadd27546deab9feca101a5ef38402ba7d9936ed38066889e461d7
thread:issue:11635,issue #11635: [Feature request] Advanced indexing in functions like `expand`,stale/low urgency,2,stale/low urgency,1,0,11635,,,,0,61dbd73c4eda124335eca937e7c9c54252d46713b691a1119fcc847549d08883
thread:issue:116350,issue #116350: RuntimeError: derivative for aten::_scaled_dot_product_flash_attention_backward is not implemented,stale/low urgency,2,stale/low urgency,1,0,116350,,,,0,076b7e08b6e95e48022ee9c1a0447508265ecddf79b5c86cd3e54b6b6a869cba
thread:issue:11636,issue #11636: undefined reference to caffe2,stale/low urgency,2,stale/low urgency,1,0,11636,,,,0,38e05c0615b1431ce63259d06f3c772f0d5d02922835253ef6ecc4c4a979c263
thread:issue:116393,issue #116393: Error: IndexError: map::at When using torch.distributed.all_reduce(tensor),stale/low urgency,2,stale/low urgency,1,0,116393,,,,0,49f2f0726d9de23c49cf2c5054bfa6d7e8f6be52603befcbb49e93a2aa218959
thread:issue:116400,issue #116400: Pytorch Transformer Distributed Training,stale/low urgency,2,stale/low urgency,1,0,116400,,,,0,56d503c3f6fc221bf90379645df63c634cf60387338153c4a6c6af8bc42f6f87
thread:issue:116403,issue #116403: pin_memory() for a custom Python object,stale/low urgency,2,stale/low urgency,1,0,116403,,,,0,de85b9e39da686ebe6527c19a532854130f4b1a4f8689c4ad618a28a62e9b526
thread:issue:116414,issue #116414: Convolution NN for complex numbers and more special functions,stale/low urgency,2,stale/low urgency,1,0,116414,,,,0,7712c09e7eb42c9546b811aa5fb20075d1dc03cac08b64882eb36521d191e1da
thread:issue:116416,"issue #116416: error: cast from 'PyObject *(*)(THPCppFunction *, void *)' (aka '_object *(*)(torch::autograd::THPCppFunction *, void *)') to 'getter' (aka '_object *(*)(_object *, void *)') converts to incompatible function type [-Werror,-Wcast-function-type-strict]",stale/low urgency,2,stale/low urgency,1,0,116416,,,,0,d0f96edc1036935805868a4806b51b07e340a8828700982a4ad1ec73e2a939c7
thread:issue:11645,issue #11645: One GPU is more memory efficient than Multiple GPUs,stale/low urgency,2,stale/low urgency,1,0,11645,,,,0,fa04a26683d2bf3c6c6d6ed5c9c36f185d530ec0a1aa825f4b2a7be2f31f5027
thread:issue:116472,issue #116472: Using a 4D attention mask in MultiheadAttention forward,stale/low urgency,2,stale/low urgency,1,0,116472,,,,0,6f5cd2e777fe593682b7537ee41280b41d70c2f73c8209f5426e0f1d60114250
thread:issue:116478,issue #116478: The Python module installs cmake files and headers under the Python's site-packages directory that can't be used,stale/low urgency,2,stale/low urgency,1,0,116478,,,,0,428e4c67ac9de5057d1ebf907cbcca3eb412206e57a88aea284915cffcf6500f
thread:issue:116479,issue #116479: Address huge memory usage on large dataset,stale/low urgency,2,stale/low urgency,1,0,116479,,,,0,25be39c82a72921cd14e89f139f353f67fb4a9939a5e10ac3f259c3ea0c697c4
thread:issue:116502,issue #116502: Add BC test for load_state_dict on OptimizerInfos and ModuleInfos,stale/low urgency,2,stale/low urgency,1,0,116502,,,,0,15f701e7941b115763fefe9bf601844e43decc3f3f46c299a8817af6ca15644b
thread:issue:116515,issue #116515: mobile model has very different inference result,stale/low urgency,2,stale/low urgency,1,0,116515,,,,0,238832f44ac812a6feec0a054131a5ac386a60481fbc11f8bc52c5d48f58e8a7
thread:issue:116533,issue #116533: `backward` implicit conversion from tuple to `torch.Tensor` results in an indexing error message,stale/low urgency,2,stale/low urgency,1,0,116533,,,,0,071acf39bc37a431a302898f1f370196e7e96cdc254440726a2c1b64687157bf
thread:issue:116539,issue #116539: Auto bisect for broken periodic CI jobs,stale/low urgency,2,stale/low urgency,1,0,116539,,,,0,b5cadbff5c5fd95a0792f236eff7f7785e59fa4a84ad731bf16e68303b60850e
thread:issue:116543,issue #116543: Multiprocessing using Gunicorn causing CUDA Error,stale/low urgency,2,stale/low urgency,1,0,116543,,,,0,63bbecd9176e58f4dc481032178963c9cbdcfcaa87cc12b6782c929975961442
thread:issue:116558,issue #116558: verbose stack trace for default collate function,stale/low urgency,2,stale/low urgency,1,0,116558,,,,0,f6302eea65190e727811fa47fd2e770efc8c08a1c069e4edd28f3b049393196c
thread:issue:116565,issue #116565: `test_autograd_dense_output_addmm` fails if matrix sizes are square,stale/low urgency,2,stale/low urgency,1,0,116565,,,,0,c8621d4657da72e509501b1f1c3258f736b00f88be09f287bd3487da18508431
thread:issue:116569,issue #116569: Python version inconsistency between Linux and Windows in the getting started,stale/low urgency,2,stale/low urgency,1,0,116569,,,,0,c10746b7e92eaf6aa0d4d072aa880ea13ab65c9bb8606380269c8fea1622f908
thread:issue:116575,issue #116575: Reland guarded config,stale/low urgency,2,stale/low urgency,1,0,116575,,,,0,e787369009a90e42b5e724ede5d428e294af41e2dca30b5acd2cc2e1d3b889db
thread:issue:116576,issue #116576: Build with MKL and OpenMP produces libtorch_global_deps.so with undefined symbol omp_get_max_active_levels,stale/low urgency,2,stale/low urgency,1,0,116576,,,,0,d2eef3fad0afc1a9c96e76172bf29dce4538b8e06c62d076313eb4384643a481
thread:issue:116592,issue #116592: Support float64 step in fused optimizers,stale/low urgency,2,stale/low urgency,1,0,116592,,,,0,5742e97855db3de64e6e61c71f7f8459c78df6606434d44d4042a2856a8676c8
thread:issue:116608,issue #116608: unstable test_GroupNorm_general_cpu in test/test_nn.py,stale/low urgency,2,stale/low urgency,1,0,116608,,,,0,19884145826efd7157075e1193e2b70fd8d59352658926626c72a7d180ffd003
thread:issue:116610,issue #116610: Align `torch.sign` with Array API,stale/low urgency,2,stale/low urgency,1,0,116610,,,,0,384abafb586716ca4b904125f69f90af5ae75b9c09076f1dd324a2fcbf927b6d
thread:issue:116633,issue #116633: Sparse tensor element-wise multiplication bug when broadcast needed,stale/low urgency,2,stale/low urgency,1,0,116633,,,,0,f29eac294cbde2a9d6fd65458a856dc5c722ef4d22f2caf82af7cd3eaabaf7e6
thread:issue:116691,issue #116691: torch.distributed doesn't have is_nccl_available() as the documentation suggests,stale/low urgency,2,stale/low urgency,1,0,116691,,,,0,f208fc912e1ea91129141458f7ad15f7ad6dbc059daa5aed4df2fad31c45300b
thread:issue:116707,issue #116707: AttributeError: `Adam` object has no attribute `param_groups` with Python 3.11,stale/low urgency,2,stale/low urgency,1,0,116707,,,,0,0040c568417a4a78fa831711315151764d1403da965139407db59433935fbf49
thread:issue:116762,issue #116762: parametrized should fail to generate tests with weird names,stale/low urgency,2,stale/low urgency,1,0,116762,,,,0,1d45bedfdf62b7215e834d71035798ee1a55ad5673a8ca72feb0de1a9dbbbcd3
thread:issue:116775,issue #116775: CI on the main branch should run with keep-going,stale/low urgency,2,stale/low urgency,1,0,116775,,,,0,d588abdedd871707f804e3fe29ebae0413cee1d25f1aa4c611fdda47d74c3745
thread:issue:11678,issue #11678: [caffe2] adam_op implementation is incorrect.,stale/low urgency,2,stale/low urgency,1,0,11678,,,,0,11fe9c725e66924f7cfedd094eccf3dc784ccf989767f8e65ba474a8bc7032ab
thread:issue:116857,issue #116857: Operator 'aten::_weight_norm_interface' not implemented MPS device.,stale/low urgency,2,stale/low urgency,1,0,116857,,,,0,8a316199c6d80582614d420f4ec12bc17d0b6cbd86ce3d98da2658986efc7971
thread:issue:116861,issue #116861: test_serialization_offset may fail with small $TMP,stale/low urgency,2,stale/low urgency,1,0,116861,,,,0,a468c09e06d73b0f507f43709ccaae1794950d2b4ea57469453362315aa569e8
thread:issue:116870,issue #116870: cuda FP16 default accumulation type differs between TensorIterator vs. torch.mm,stale/low urgency,2,stale/low urgency,1,0,116870,,,,0,4796a123b2f7c8c634f9ea7ca3f13e4883a0ba41e618df36eb3713945004227f
thread:issue:116896,issue #116896: keep-going UX is not good,stale/low urgency,2,stale/low urgency,1,0,116896,,,,0,a7c6b4f6e0571c5f8c67e3b29ce226d50589dbf37adb86b2d60ec045c8c54d90
thread:issue:116965,issue #116965: Support GridSample 5D,stale/low urgency,2,stale/low urgency,1,0,116965,,,,0,244cdb2c63e1c158ca24e61d5852ebbfbd509c4369ce13ae4f39f27141e2bff4
thread:issue:116980,issue #116980: Sharded shape attributes for DTensor,stale/low urgency,2,stale/low urgency,1,0,116980,,,,0,0d34850c5a30112c2aa71a0cdfd554368cd5eeb16e236057d98cfa032197ecc5
thread:issue:116987,issue #116987: [pytree] class based registration api,stale/low urgency,2,stale/low urgency,1,0,116987,,,,0,5898e10f7061c42fcd3a25d77e9172b3fea17f0fb26bc2802f3a088b48d37895
thread:issue:117094,issue #117094: LBFGS MPS test_can_load_older_state_dict fails on older MacOS versions,stale/low urgency,2,stale/low urgency,1,0,117094,,,,0,cdd53c1df9ea15e96b7e8f7aa74ccec52a1c0b31d78f50667707c8b448263e4f
thread:issue:117099,issue #117099: torch-1.11 Cross-Compiling for QNX 7.1.0 with aarch64le issue: no matching function for call to 'isnan(c10::BFloat16&)',stale/low urgency,2,stale/low urgency,1,0,117099,,,,0,9e97b023dd483ba0af386a86e34e535f355262a639b262f7cb4efb3c0af5d839
thread:issue:117130,issue #117130: `torch.cuda.memory_summary()` can give `KeyError`,stale/low urgency,2,stale/low urgency,1,0,117130,,,,0,34f206e652857d30eedd2629f2f3366fa4035a0c1c4d67fb1f40a7a7a2f7bca6
thread:issue:117146,issue #117146: `test_triton_bsr_scatter_mm` fails on V100,stale/low urgency,2,stale/low urgency,1,0,117146,,,,0,84f4c662dbbac8d7dd8136dd0d32ddbb14c24070459c23a1ccef4a30ff162390
thread:issue:117209,"issue #117209: torch.onnx.export the torch,nn.TransformerEncoderLayer, the dynamic_axes work not right",stale/low urgency,2,stale/low urgency,1,0,117209,,,,0,02c2b9f0054453f74e0b7d32f8f31e43b4554f29ff9ee0076d96196f9709b18d
thread:issue:117232,issue #117232: ```torch.Tensor.renorm```: Internal assert failed,stale/low urgency,2,stale/low urgency,1,0,117232,,,,0,2f0721c64e6ee68b08ac5b232b41dce4c04f64dd46a883c7f4e17b64cf726e67
thread:issue:117233,issue #117233: Submit a bug when trying to assign arg 'out=' in api torch._foreach_add_,stale/low urgency,2,stale/low urgency,1,0,117233,,,,0,e0880483a340c26bde40a9200fef688d96401d4a86c500465b649196f2e3f249
thread:issue:117236,issue #117236: ```torch.linalg.lstsq```: Internal assert failed,stale/low urgency,2,stale/low urgency,1,0,117236,,,,0,596a63a321d11da7ca546f069042a9d2597c7e1718271a9c7bacaacb204ee6f4
thread:issue:117242,issue #117242: ```torch.tensordot```: Internal assert failed,stale/low urgency,2,stale/low urgency,1,0,117242,,,,0,370df83c0b69de8866f1d20fb0f3b5cb30c826c343d9c8e0af5cbef3af02e77d
thread:issue:117243,issue #117243: ```torch.Tensor.cov```: Internal assertion failed,stale/low urgency,2,stale/low urgency,1,0,117243,,,,0,22482ec38527b30068e241b42374871189632dcc7bb31c64364ee1775f76d0ea
thread:issue:117244,issue #117244: ```torch.Tensor.index_copy_```: Internal assert failed,stale/low urgency,2,stale/low urgency,1,0,117244,,,,0,c7388afd7bbb17d1ef5e04c2f8320666b5d503f805309c187c38b74335533267
thread:issue:117245,issue #117245: ```at::batch_norm```: Internal assert failed,stale/low urgency,2,stale/low urgency,1,0,117245,,,,0,9c8ea1ec3c754bd97ffb1ffcc78ec0ee3ab4eac7d46c576ecae821d50c88e5ba
thread:issue:117246,issue #117246: ```at::smooth_l1_loss```: Internal assert failed,stale/low urgency,2,stale/low urgency,1,0,117246,,,,0,b7b3f60dbeb2b744446d301f1aa83e836be86824dc1f8e327849028b70e200a9
thread:issue:117247,issue #117247: ```torch.nn.Linear``` class: Internal assertion failed,stale/low urgency,2,stale/low urgency,1,0,117247,,,,0,b13a5ae04d7c25e78232cbfd4a62348f9530302856ba4ce26274114f1732b37f
thread:issue:117248,issue #117248: ```torch.Tensor.matmul```: Internal assert failed,stale/low urgency,2,stale/low urgency,1,0,117248,,,,0,1c83d812a4463d33480c4845d3c1c1e650625e26c11059b28d48a0cb6b29f0b3
thread:issue:117249,issue #117249: ```torch.Tensor.triangular_solve```: Internal assert failed,stale/low urgency,2,stale/low urgency,1,0,117249,,,,0,d0d43e9b735671bf44a319ac996826841a8d2bd9f1141c4a2c427cf21cec15e0
thread:issue:117250,issue #117250: ```torch.nn.functional.normalize```: Internal assert failed,stale/low urgency,2,stale/low urgency,1,0,117250,,,,0,3a0ffb16943e9b8e7e159c2ed6541c71687a58f9fe9c0820c1b3776b968fe858
thread:issue:117251,"issue #117251: After model compilation and run on second gpu card (cuda:1) , 154 MB still goes to the first gpu card (cuda:0).Using pytorch ignite framework",stale/low urgency,2,stale/low urgency,1,0,117251,,,,0,cc921d74691cdacba8cfc09c90d6ba4c63c364ae44f231511b5a4e3141768bd6
thread:issue:117256,issue #117256: ```torch.Tensor.round_```: Internal assert failed,stale/low urgency,2,stale/low urgency,1,0,117256,,,,0,ed57b4f7512d77341525f90177ddbb2280c45d07d755fc372aaa46607cf37a5a
thread:issue:117259,issue #117259: ```torch.Tensor.sum```: Internal assert failed,stale/low urgency,2,stale/low urgency,1,0,117259,,,,0,577bf723137b66e95af52f33a36b7f1e376e81b2e586e2b754a88a764518b032
thread:issue:11727,issue #11727: Semaphore leaks in dataloader,stale/low urgency,2,stale/low urgency,1,0,11727,,,,0,0aeb8b3b90c8cd804b50821631e949ce649c24599f744364f98dee1572adca7f
thread:issue:117275,issue #117275: deepcopy of `nn.Parameter` does not transfer __dict__,stale/low urgency,2,stale/low urgency,1,0,117275,,,,0,ef9a98c88452e8f861faccbd21e409644f9c132c7ff980e079fc03fc6dba10bf
thread:issue:117287,issue #117287: Error computing the norm of MaskedTensor,stale/low urgency,2,stale/low urgency,1,0,117287,,,,0,99647deeec8726d108046abb4377697855f1cb7449e6cd417448204e454385ff
thread:issue:117321,issue #117321: [C++-polyfill] Pt2 - Discussion around user defined type->behavior dispatching,stale/low urgency,2,stale/low urgency,1,0,117321,,,,0,2c36c56df149fa23cc3debe1399f90b0fc15ba735d1cb5f871cb9a3cb8d0b002
thread:issue:117348,issue #117348: The bitwise_XX.Scalar_Tensor should be dispatched to core IR but is currently not,stale/low urgency,2,stale/low urgency,1,0,117348,,,,0,7826d4b6d291f0096f95f6f7e9f887c3748be19eb393d26b23e22dbd43e250f9
thread:issue:11735,issue #11735: Caffe2 installation: libcaffe2_gpu.so: undefined reference to `caffe2::ClipTransformRGB,stale/low urgency,2,stale/low urgency,1,0,11735,,,,0,a76822d780b08089dccf1a0ce09034f90b50fd6ed02c1fd89dd80c3b37dab416
thread:issue:117357,issue #117357: [RFC] CI test stats,stale/low urgency,2,stale/low urgency,1,0,117357,,,,0,8c1cf16fbcf7605dc6c413c8347d18469cd7007b7105f2f7e63752e705b2e23c
thread:issue:117424,issue #117424: Various CI configurations are missing a slow equivalent,stale/low urgency,2,stale/low urgency,1,0,117424,,,,0,9503096ad3c016a72a4f57ccb01fd64840d778f39ba5acbbd64dad85ce46ef03
thread:issue:117429,issue #117429: Memory leak while initializing tensors in child thread,stale/low urgency,2,stale/low urgency,1,0,117429,,,,0,a884dd1a0bd8e6f109e20e0ee62bec041d57046c1024cf32a0f5ecd9d5e89d9a
thread:issue:117437,issue #117437: _register_state_dict_hook docstring is not accurate?,stale/low urgency,2,stale/low urgency,1,0,117437,,,,0,cc1c754df9f4b287d4daaf8e056b16d65ca85e016e2cffd02263e1ee9fe153a5
thread:issue:117449,issue #117449: Deprecate and remove usage of from __future__ import annotations in codebase,stale/low urgency,2,stale/low urgency,1,0,117449,,,,0,36d02e3da8df6d658c770f359bbfba069325a6dbbf49918e617044153a0bb9af
thread:issue:117467,issue #117467: Is this an expected memory profile?,stale/low urgency,2,stale/low urgency,1,0,117467,,,,0,449d9dafbd2023d75bf08ef13ca4c16ae613d8d717334a140eda8452d2c33195
thread:issue:117508,issue #117508: Consistency between torch.ao.quantization _LearnableFakeQuantize PTQ toggle and FakeQuantize PTQ toggle names.,stale/low urgency,2,stale/low urgency,1,0,117508,,,,0,9cf2f80e767fd2e02a89e42ae6fcab06f9e7fe624db03236385bdc6489595f6b
thread:issue:117540,issue #117540: Unwanted Warning in lr_scheduler.step(),stale/low urgency,2,stale/low urgency,1,0,117540,,,,0,40a0f728b53c098522121d013785e79e7b1878d3b055742e86531767908d8bed
thread:issue:117546,issue #117546: interpolate::trilinear has different result between 1.8.0 and the newer version,stale/low urgency,2,stale/low urgency,1,0,117546,,,,0,1892a9b8c3aed61fc003e23155229d5d006b6ca5016551ee8cfb15f467ac8ac7
thread:issue:117576,issue #117576: Warning for unusual log sizes in CI,stale/low urgency,2,stale/low urgency,1,0,117576,,,,0,2d41eaaf632ea264c7634fa51a48964480c1202f63d882afc5809657d3c8874f
thread:issue:117582,issue #117582: Add forward AD support for nn.functional.grid_sample,stale/low urgency,2,stale/low urgency,1,0,117582,,,,0,49e6a6682ad996c0c845e59d3061b63db83446a98a1c45eea39c534a3f0d02fe
thread:issue:117586,issue #117586: PyTorch 2.1.0 High memory utilization compared to PyTorch 2.0.1,stale/low urgency,2,stale/low urgency,1,0,117586,,,,0,d919b0278d92155888e614c066bef5d4b9d3bfab7cfc139067cb3e3ec3781b86
thread:issue:117601,issue #117601: [FSDP] more actionable error logging in meta init,stale/low urgency,2,stale/low urgency,1,0,117601,,,,0,b41540fdbb66c5204c93ee90f285df38f518760e6f2d53f8f382c9148f6ccb8c
thread:issue:117645,issue #117645: Segfault linear_search_fdes when calling c10::get_backtrace,stale/low urgency,2,stale/low urgency,1,0,117645,,,,0,28b4ed5a86d47eabf3bbc80217c301d5426721b6ca0964a0a1338e8ac2fa4aed
thread:issue:117647,issue #117647: Easier way to add a lot of tests,stale/low urgency,2,stale/low urgency,1,0,117647,,,,0,bfea093272ca87d037052e1ae32ce6cc6eb7d3b001fc177e592476bab6f2f8e2
thread:issue:117681,issue #117681: Improved workflow for tracing dynamism through complicated model,stale/low urgency,2,stale/low urgency,1,0,117681,,,,0,b3ba51b3df9d1fa7e7237c0361ff6c4e001cef9ad226281150544568b6d5a465
thread:issue:117705,issue #117705: Add GPU runners for Dynamo shards,stale/low urgency,2,stale/low urgency,1,0,117705,,,,0,4e94d7c72d32b5797c9646ae09204abfbcb995d536f0beabccdf3e77808a63b4
thread:issue:117728,issue #117728: [DTensor] addmm strategy cannot handle 0-element input tensors,stale/low urgency,2,stale/low urgency,1,0,117728,,,,0,bf552ed960131922bfe93c7151e69b187a01e4d68f1cd1fddf786eba580b74c3
thread:issue:117762,issue #117762: PyEval_AcquireThread can crash process if called from another thread at interpreter shutdown,stale/low urgency,2,stale/low urgency,1,0,117762,,,,0,b6a8f412c34759f198f44d77c985a438580925c6b164ba79b83c867e7029974c
thread:issue:117791,issue #117791: [RFC][Export] Capture module attribute/annotation in `torch.export.unflatten`,stale/low urgency,2,stale/low urgency,1,0,117791,,,,0,ecd7f714302b6a8566ee5d0ced67d99d42b43b1ba4e27f30ee2b07ddeaff4c74
thread:issue:1178,issue #1178: Batched sparse QR factorizations and solves with cusolver,stale/low urgency,2,stale/low urgency,1,0,1178,,,,0,fd8dab70cae9ba09800cffb5c3c7f8e5ed8a469af8d7761382f3ab205936319a
thread:issue:117832,issue #117832: amp need a whitelist,stale/low urgency,2,stale/low urgency,1,0,117832,,,,0,3f4e3b8169b689c23add492688f1cdaa7f55de499eb0b0c8ab99d9b281fc70d8
thread:issue:11786,issue #11786: Mysterious error due to num_workers: 1,stale/low urgency,2,stale/low urgency,1,0,11786,,,,0,e7436395b5b856fbaa7a94a492ea885dedf3a7d0df1a2fba5a723a83546c33a5
thread:issue:117883,issue #117883: Flight Recorder - feature requests,stale/low urgency,2,stale/low urgency,1,0,117883,,,,0,a86ea156043d721a3dafe9e29a19ed3d5cacfb3a7587e4ff8b156843cf188466
thread:issue:117884,issue #117884: Foreach Where support,stale/low urgency,2,stale/low urgency,1,0,117884,,,,0,72973104492eea7603d34cc6bf6df05eadc87bf24440fd12d526aed49465fcc9
thread:issue:11790,issue #11790: set num_workers on the dataloader make the jupyter kernel crash at the almost end of the epoch,stale/low urgency,2,stale/low urgency,1,0,11790,,,,0,1ecb404b9ef6a8b7f04542bae2ad63a145d84426400b824354e94b1eb1471089
thread:issue:11791,issue #11791: [caffe2]Can i build caffe2 library only for cpu inference purpose and reduce the binary size?,stale/low urgency,2,stale/low urgency,1,0,11791,,,,0,30680dc3a1e2387238087e2dbed05ee9a98f96a4f0b4716436204ee39909275f
thread:issue:117910,issue #117910: canonical way to normalize torch.device?,stale/low urgency,2,stale/low urgency,1,0,117910,,,,0,d9301436de14ef25b7e43ac01e5f5a6efdbf9f446e5a920e4341a949d52fb991
thread:issue:11793,issue #11793: DataParallel: Parallel_apply assert len(modules) == len(inputs) AssertionError,stale/low urgency,2,stale/low urgency,1,0,11793,,,,0,8db4bbcb58beefa6d06df29a1916316ed53d277a082d36b523847d4f34c4aea1
thread:issue:117930,issue #117930: C++ Tensor.clone() changes max threads in OpenMP,stale/low urgency,2,stale/low urgency,1,0,117930,,,,0,314a2c157d5930ab92bd9354fa60e96af83f42731c46721872288a2b5229a625
thread:issue:117938,issue #117938: Libtorch Don't Support Multi-thread Inference,stale/low urgency,2,stale/low urgency,1,0,117938,,,,0,82169541901887dc1f5e23e1af782f1c8e960ad61ad8d9e8d9cb5171b6045af3
thread:issue:117949,issue #117949: aten::slow_conv3d_forward still missing from MPS,stale/low urgency,2,stale/low urgency,1,0,117949,,,,0,1ce6a4666b92e33e3792462dc55a4e0d683bf95eb2363091c9c15db3bdea235b
thread:issue:117961,issue #117961: ReplicatePad2d cuda not support bfloat16,stale/low urgency,2,stale/low urgency,1,0,117961,,,,0,7228d4f3be94d2751fdddc40277edb2445cac378295605454395662b83f328b7
thread:issue:117963,issue #117963: Does Libtorch c++ convolution operation use optimization?,stale/low urgency,2,stale/low urgency,1,0,117963,,,,0,46188fcc114cc5c4c81ea8bc0c1d050c2b4f62fd5629fc4c1052c10e21a4b858
thread:issue:117967,"issue #117967: Manually modified parameter during trainning, gradients are not being correctly updated. Help",stale/low urgency,2,stale/low urgency,1,0,117967,,,,0,8dd71fb3161e4f86def80cc14bc14639bf13bfa91bff6d94705f00688fd13f95
thread:issue:117976,issue #117976: Unify test_torchinductor.copy_tests and make_test_cls_with_patches,stale/low urgency,2,stale/low urgency,1,0,117976,,,,0,766d5a9afc27bf9d2cdd1ec3332cf6d92c1580a16d94ced8416f1aa0bd7d8d40
thread:issue:117981,issue #117981: [Feature] Foreach view,stale/low urgency,2,stale/low urgency,1,0,117981,,,,0,8778a541d4ccc261a94d3d423f71c0426a830c9433f90410013cb0e985971d2b
thread:issue:118015,issue #118015: Disable tests UX suggestions,stale/low urgency,2,stale/low urgency,1,0,118015,,,,0,f09e624c92d88b3afd1bdb829f66dbba5820443a7f8af78a5d7c23e73984a9d2
thread:issue:118058,issue #118058: torch.export doesn't handle functools.partial correctly.,stale/low urgency,2,stale/low urgency,1,0,118058,,,,0,0db3bc67f5dc4b4baa4bc2d65653f6bf57b9ccf8db9124e3e35db35a2f2741cd
thread:issue:118095,issue #118095: NAN value for truthfulqa_mc2 on full finetuned model TinyLlama,stale/low urgency,2,stale/low urgency,1,0,118095,,,,0,fd7075c6d06351c30d8449af8cfa73a5c1e59d8cd51ab52d72a67e7a5e80b4d9
thread:issue:118173,issue #118173: ModuleNotFoundError: pynvml does not seem to be installed or it can't be imported.,stale/low urgency,2,stale/low urgency,1,0,118173,,,,0,8af8a709e58498e64e111a5b7f5f31b008d95bc12a521e54eb6bfa6b1da91c57
thread:issue:118175,issue #118175: Different behaviors in `torch.nn.functional.hinge_embedding_loss` between eagermode and torch.compile,stale/low urgency,2,stale/low urgency,1,0,118175,,,,0,f193ef8d9c3b12b4ed959b98155732f6bb06a27fee2b92eb7c11d26272396369
thread:issue:118227,issue #118227: TorchDynamo mistranslates end of tensor slice,stale/low urgency,2,stale/low urgency,1,0,118227,,,,0,622c45c3270d8bdafeb59db417d221953756d1029c4c4e7b5923b65bfb74ffa0
thread:issue:118245,issue #118245: [FSDP] NO_SHARD numerics with cpu offloading and orig params,stale/low urgency,2,stale/low urgency,1,0,118245,,,,0,9bc7dc2f47fcb90cfbe06322021e739c0abc38ae92779e7e009d53818a92dc5f
thread:issue:118294,issue #118294: Failures in test_nn depending on test order,stale/low urgency,2,stale/low urgency,1,0,118294,,,,0,225c30c3eecd87cf3caac9e3ca8739fbbd12cf116be990b84d060347e0c3cad4
thread:issue:118352,issue #118352: Unify TEST_WITH_TORCHDYNAMO and PYTORCH_TEST_WITH_DYNAMO,stale/low urgency,2,stale/low urgency,1,0,118352,,,,0,9dba9fca9c2a03ab2a41f58e8648ba822b005a1e53ae400ed6840c18af65674a
thread:issue:118370,issue #118370: Tests aren't cleaning up state in test_optim.py,stale/low urgency,2,stale/low urgency,1,0,118370,,,,0,032ca48ed4e14b8808e31880930294e2ab384b9335fede9f742770615ebd95c2
thread:issue:118373,issue #118373: Torch + CUDA installation,stale/low urgency,2,stale/low urgency,1,0,118373,,,,0,33c38fc4a4a150a76658f2dbee549734fd3d8df854f6fadfc549e6e0e119bef6
thread:issue:118390,issue #118390: Audit all global mutable state in PT2 to see if it should be thread local instead,stale/low urgency,2,stale/low urgency,1,0,118390,,,,0,0b04114fa5fb95b9c1caf4881ebb947619a0bb3ca5d227b9c0563fa749be55e7
thread:issue:118461,issue #118461: [DTensor] DTensor fails to execute backward of index_put op.,stale/low urgency,2,stale/low urgency,1,0,118461,,,,0,9a4c8189e2808e1512ed01a77057ef1c8ccd58b606968663619d0a8204825ad1
thread:issue:118472,issue #118472: [c10d] Avoid busy looping in C10D_NCCL_CHECK_TIMEOUT,stale/low urgency,2,stale/low urgency,1,0,118472,,,,0,fa38097ba54229faea8697331ba311df2721dfd85344a9a9546f5235ced42125
thread:issue:118478,issue #118478: Cannot formatted print at::Tensor in gdb,stale/low urgency,2,stale/low urgency,1,0,118478,,,,0,20d77a622ce2d4645504671f559737fc2286da797f0a74f92b17d67412436f12
thread:issue:118488,issue #118488: RFC: Enable disallow_untyped_defs by default,stale/low urgency,2,stale/low urgency,1,0,118488,,,,0,31f4f2d98ed3cec03af582d69812c5d105ee089413dfd8879c073a43b298972e
thread:issue:118499,issue #118499: Autocast on cpu dramatically slow,stale/low urgency,2,stale/low urgency,1,0,118499,,,,0,6509b00218f83166f840a41ab5ba82145de7668d044d7665a01650bf6a8d0f21
thread:issue:11850,issue #11850: [Caffe2/Bug] Cannot enable MKL-DNN,stale/low urgency,2,stale/low urgency,1,0,11850,,,,0,73ce7b5a24f4bf389c49567d567d8ffc656f6b14946151c30483d6a34c09f576
thread:issue:118508,issue #118508: [Bug] Bugs when installing `torch-sparse` and `torch-scatter` in docker.,stale/low urgency,2,stale/low urgency,1,0,118508,,,,0,32b919ec4c48f7aceca52ed23a8e7bcc9ac3821bd55b6396d13fe8c2036cabba
thread:issue:118517,issue #118517: ValueError when setting tensor type to match MPS tensor type,stale/low urgency,2,stale/low urgency,1,0,118517,,,,0,4191b7db3fbb823c5ae699b44983cf7f47d3b6f3c304033e6d3dba690957e725
thread:issue:11854,issue #11854: error: ‘array_size’ is not a class template,stale/low urgency,2,stale/low urgency,1,0,11854,,,,0,4bc328ee1a4fb158fa55b376b83f8d5e6aa3e2f3ad2952c2bc52180682bc969f
thread:issue:118570,issue #118570: [Inductor] Kill Mutation Layout,stale/low urgency,2,stale/low urgency,1,0,118570,,,,0,54c98ef29a61707cd3b9ed62cf47f22e71504141de6cf1ebf4f0093b9126c3ef
thread:issue:118573,issue #118573: [Inductor] Unite all possible way to track mutation,stale/low urgency,2,stale/low urgency,1,0,118573,,,,0,1d8bb70e7df5f21c6bdbd0776745387131fb363528d27cb13e2ecde65b89dc8b
thread:issue:118580,issue #118580: [NestedTensor] Support for operations found commonly in point cloud analysis architectures,stale/low urgency,2,stale/low urgency,1,0,118580,,,,0,ba8c19df48f419964ebc1241d0065528bfab56e4ab5ca1b8d766556253e99b60
thread:issue:11861,issue #11861: [Caffe2 installation problem],stale/low urgency,2,stale/low urgency,1,0,11861,,,,0,6d2ad667dcf98d79cad1009f6e59a83beb3cbb57286c2f4a378fdd68e1883a63
thread:issue:118628,issue #118628: TransformerEncoderLayer fast path predicts NaN when provided attention bias,stale/low urgency,2,stale/low urgency,1,0,118628,,,,0,cc71ca90aa5a8843303102dd3d8fddbe1f40ab2fdb93feee091f4284e9892ec3
thread:issue:118634,issue #118634: symbolically traced variables cannot be used as inputs to control flow,stale/low urgency,2,stale/low urgency,1,0,118634,,,,0,7d880d7ebb91967de555dfceb3343c439adfd3f0d4d872bbd597f5b233f50671
thread:issue:118642,issue #118642: Different outputs of `div` and `threshold` with a specific input,stale/low urgency,2,stale/low urgency,1,0,118642,,,,0,a854f5eabf44c4d53b9a159a225cfd2dc600dba290a126c214da950dec044414
thread:issue:11865,issue #11865: [Caffe2] Errors occured when running Cpp Predictor,stale/low urgency,2,stale/low urgency,1,0,11865,,,,0,14bd9e1faa5b5044ef8e5f72c57cc65bf3272072e56fb3d5fbd21b1003d9eaea
thread:issue:118723,"issue #118723: Libtorch torch::load(tensor, ""tensor.pt""); error",stale/low urgency,2,stale/low urgency,1,0,118723,,,,0,23cc509882502dede20c8684a7d8ccc5368d68e45edd8fd8344f4524e4bc1a2a
thread:issue:118774,issue #118774: Tune whether to use mm or bmm for matmul in inductor max-autotune,stale/low urgency,2,stale/low urgency,1,0,118774,,,,0,6fbc2187c3493158b85aac586567ce7b0d06191e911699c20ff399d929b8cf66
thread:issue:118829,issue #118829: The inference result is wrong when I apply the torch.quantization.fuse_modules api to resnet50 model.,stale/low urgency,2,stale/low urgency,1,0,118829,,,,0,8025a4b04325b2679272484cc2abdd3d2adb94367a7c553a4cd92979b3ef353e
thread:issue:118842,issue #118842: CUDA Extension Bug for Pytorch 2.2.0,stale/low urgency,2,stale/low urgency,1,0,118842,,,,0,ed2845aaf707c6214bd3e787eea86a4333f0fd74395fb69d697177a7cf4bbed4
thread:issue:11885,issue #11885: [caffe2] Bug for softmaxwithloss operator,stale/low urgency,2,stale/low urgency,1,0,11885,,,,0,c290702bcaa592509ff786e465368da9d2d0ad7ad4015c9ea1c746dd05226224
thread:issue:118853,issue #118853: MultiheadAttention returns different results for binary mask and float mask,stale/low urgency,2,stale/low urgency,1,0,118853,,,,0,43e024ecc0c9aea75c88df9c73fa3731fe6e03b73e48bbe4a610f934286c2aff
thread:issue:118859,"issue #118859: [W CudaIPCTypes.cpp:95] Producer process tried to deallocate over 1000 memory blocks referred by consumer processes. Deallocation might be significantly slowed down. We assume it will never going to be the case, but if it is, please file but to https://github.com/pytorch/pytorch",stale/low urgency,2,stale/low urgency,1,0,118859,,,,0,aeaf83cb7418e67f499df103fdd982647b7352fa18e43f07da93f28f38545d0e
thread:issue:118861,"issue #118861: Wrong typing for `argnums` in `torch.func.jacrev`, `torch.func.jacfwd` and possibly `torch.func.vmap`.",stale/low urgency,2,stale/low urgency,1,0,118861,,,,0,dd4a0d4171d2bbcce86fae5705aa2ea628dab04e0895f864533b224547f310c3
thread:issue:118863,issue #118863: Pytorch with DSA enabled? Are there binaries? Perhaps the nightly?,stale/low urgency,2,stale/low urgency,1,0,118863,,,,0,74715dfe1b8322321caf2fd80211f1458aac3575bb3b5d9bef8aaee439f58618
thread:issue:118883,issue #118883: Torch Sparse Sampled AddMM Backwards Memory Usage Blowup,stale/low urgency,2,stale/low urgency,1,0,118883,,,,0,d178962885efecb8da3f27aa97fe746a5cef82128e5938356c071cc010266db9
thread:issue:118884,issue #118884: [Export] [TorchVision] Export Support for TorchVision Detection Models,stale/low urgency,2,stale/low urgency,1,0,118884,,,,0,6b576688e753c72dd1ea2f208b762d9e64ae0b941602235132223225847abe45
thread:issue:118891,issue #118891: Add an option for torchrun c10d store to be collocated with rank 0 training node,stale/low urgency,2,stale/low urgency,1,0,118891,,,,0,9a61e0e791a028ed983b15628e925b60c3d03afa38d2e98f12beac0ab60ec7d6
thread:issue:11890,issue #11890: caffe2 argmax and argmin documentation incorrect for output type,stale/low urgency,2,stale/low urgency,1,0,11890,,,,0,1e9753cef8f756313ef044c0bb20c5574c81313213c15239df2976364e332bc6
thread:issue:118918,issue #118918: Expectile Loss?,stale/low urgency,2,stale/low urgency,1,0,118918,,,,0,68cfd3c93f83b95766df159cb44a78c7ad394e5cd574a6ea768923725c71b0db
thread:issue:118922,issue #118922: torch.export(strict=False) fails with tensor constants,stale/low urgency,2,stale/low urgency,1,0,118922,,,,0,13078a228888c653360323fdd1cb0930daf2b8daaad5dcc439510a4d5751cb6f
thread:issue:118970,"issue #118970: When concatenating empty variable with non-empty variable, I get CUDA error: illegal memory access encounter",stale/low urgency,2,stale/low urgency,1,0,118970,,,,0,e26bd682449408bdaba6a02ad93fc50425d12a01b68fc68e377bd6f19871a1b7
thread:issue:118991,issue #118991: Memory Leak in v2.0.1,stale/low urgency,2,stale/low urgency,1,0,118991,,,,0,38d456153f11439e182aaf99a8778a7a084c482c3789633d9012a287ab2345ce
thread:issue:119070,issue #119070: [DCP] Review and update examples and docstring to reflect DCP save/load API updates,stale/low urgency,2,stale/low urgency,1,0,119070,,,,0,367d13b55efd8bf7e6f11bfb1de00a0ff8c44973c0948813bd782c9e322a8f68
thread:issue:119072,issue #119072: Importing PyTorch 2.2 fails with undefined symbol error: ncclCommRegister,stale/low urgency,2,stale/low urgency,1,0,119072,,,,0,b97176d6268ab757ceebee509d508013dda6e6eac6c2131aa2547f76bd7251c8
thread:issue:119076,issue #119076: torch.bmm fails when multiplying a batch of sparse matrices by a dense tensor,stale/low urgency,2,stale/low urgency,1,0,119076,,,,0,cc245ec58db7c8f3ea2b4c20c006298d566f1d22139a4c8bf95d7c48c923832b
thread:issue:119108,issue #119108: [MPS] Tracking issue for ModuleInfo failures when enabling testing for `torch.float16`,stale/low urgency,2,stale/low urgency,1,0,119108,,,,0,e5a1717290a6d70c173fff4007c290616d277e627f334d6ba5a5dbea315dfe90
thread:issue:119123,"issue #119123: `DataLoader` generic argument is dropped in `__iter__`, making it non type-safe",stale/low urgency,2,stale/low urgency,1,0,119123,,,,0,fc1db3813b4f7bfdecf2dd6794d62de38baf510463da454a3203b23c1c1bc21e
thread:issue:119128,issue #119128: [dynamo] Refactor switch statements to improve compile times,stale/low urgency,2,stale/low urgency,1,0,119128,,,,0,acbea239286314212bb72951f191665f9825bac5b86f20183bd8d946bb33d698
thread:issue:119131,"issue #119131: `scaled_dot_product_attention` produces NaNs for math backend, but not memory efficient backend when used in mixed precision.",stale/low urgency,2,stale/low urgency,1,0,119131,,,,0,ef3b5305cd0c5eff20a37c55969d77f3f9ed90bcde2d8a89cdede97d84b2bc31
thread:issue:119181,issue #119181: [inductor][cpu]Background_Matting fP32 static default wrapper multiple threads performance regression,stale/low urgency,2,stale/low urgency,1,0,119181,,,,0,dc8f89c2a60afc8cbbb468f953cc0a93f545815c2d293fce4b9c6bb211adecf3
thread:issue:119188,issue #119188: scaled_dot_product_attention: different output between the reference implementation and the one from torch.nn.functional,stale/low urgency,2,stale/low urgency,1,0,119188,,,,0,c23d104b8916c9a3786f805ad90cb5d999006f1e4dfae49938b2e5087dd1054a
thread:issue:119190,issue #119190: TensorList Python binding triggering C++ TypeError,stale/low urgency,2,stale/low urgency,1,0,119190,,,,0,0f2a5512bd976dfeddd7939d0d1300077afb9d8325d5fc6eea669f91916e1a1d
thread:issue:119194,issue #119194: torch_function objects passed as non-Tensor args should trigger overrides,stale/low urgency,2,stale/low urgency,1,0,119194,,,,0,9495c6a635c050c096068ef4c3d7cbdf8881b6e7b55988eb1794d4716c934c3c
thread:issue:119206,issue #119206: load_inline should always recompile a kernel if it failed,stale/low urgency,2,stale/low urgency,1,0,119206,,,,0,ab7453457f2ba8780a43f1fc9cb291b1cd7616e0a3c91439f114bc1a0c2d8673
thread:issue:119279,issue #119279: MODULE.bazel support,stale/low urgency,2,stale/low urgency,1,0,119279,,,,0,05bd6fb6e5660e9c88c8dcd67478bf7deaa918e96914f71fe814168348dfb07d
thread:issue:119281,issue #119281: Support for dynamic DataLoader (DataLoader.__getitem__),stale/low urgency,2,stale/low urgency,1,0,119281,,,,0,81c22028fb230f18b572534c968b05aff3eaf2290a09c4428d3d2220b96defd0
thread:issue:119343,issue #119343: Use Dtensor and ColwiseParallel api to do Tensor Parallel for llama2 model but got communication error.,stale/low urgency,2,stale/low urgency,1,0,119343,,,,0,1d07d351e21c8945291b2e6f161c30985f0820bb93b116061c8d3b1d58a54f52
thread:issue:11936,issue #11936: torch.bmm doesn't support CUDA uint8 (byte) tensor,stale/low urgency,2,stale/low urgency,1,0,11936,,,,0,c24e36e39a78657bc775a2262c4c2f47f06a357d5ddd76c82966678be3607ade
thread:issue:119368,"issue #119368: Support nn.Module.to(..., copy=True)",stale/low urgency,2,stale/low urgency,1,0,119368,,,,0,bd594c38681daea7f4874ca6320a78791fc82bc353f63e3dd9ae905be72f9fed
thread:issue:11937,issue #11937: [feature request] Kumaraswamy distribution,stale/low urgency,2,stale/low urgency,1,0,11937,,,,0,3fe608d92e0eebf5721c59afd20adfa5ccdbb3b850181800c47e1e67ee4a9947
thread:issue:119371,issue #119371: Nan output after masked TransforrmerDecoder,stale/low urgency,2,stale/low urgency,1,0,119371,,,,0,208ee7c3d3ff4d0ab8508be3fed2786433a6b4a8a9bb5094342854ac8d61b121
thread:issue:119377,issue #119377: Asked to report bug by a `UserWarning`,stale/low urgency,2,stale/low urgency,1,0,119377,,,,0,07c2e357dbe7bfdbf8a25c4e4d4a96cdd0a3bdd806a2ec38665d3523694eef75
thread:issue:119387,issue #119387: Lazy Tensor GetBackendDevice is not working at all,stale/low urgency,2,stale/low urgency,1,0,119387,,,,0,b0442cd2f8a28ba32147794a355546370a32558f8276580dd28ba8b4ecd87312
thread:issue:119417,issue #119417: [c10d] accept [] in torch.distributed.all_gather_object,stale/low urgency,2,stale/low urgency,1,0,119417,,,,0,7237719a1928e436690e3dde59ad1d123df5b118d75bf49c1134bd57fc3393bc
thread:issue:119418,"issue #119418: [c10d][NCCL] better error msg than ""Software caused connection abort""",stale/low urgency,2,stale/low urgency,1,0,119418,,,,0,385b6e3694004bb3a5ba883748842665db0399e060ea518991523e6211b9f404
thread:issue:119420,issue #119420: torch.stft sometimes raises RuntimeError: cuFFT error: CUFFT_INTERNAL_ERROR on low free memory,stale/low urgency,2,stale/low urgency,1,0,119420,,,,0,14634e99122dbc3c07d76bda4e995693fa59c619cb4a2d8bb565553f5240350d
thread:issue:119431,issue #119431: FSDP convergence issue with CPU offload,stale/low urgency,2,stale/low urgency,1,0,119431,,,,0,b41b09424f2b94b1690fd4345d0bd51e6bb19bfdf1ad0239e96843bd98a24199
thread:issue:119442,issue #119442: pytest test/onnx/test_pytorch_onnx_onnxruntime_cuda.py -k test_embedding_bag_2d_per_sample_weights is failing,stale/low urgency,2,stale/low urgency,1,0,119442,,,,0,3c56f688f250ca82c0575aa1ef3ed914c8a55deb4b453dde8c050e208842dcc7
thread:issue:119447,issue #119447: Cannot install or import PyTorch Nightly,stale/low urgency,2,stale/low urgency,1,0,119447,,,,0,f07f9998dfb7d4d06c3bc3c0436d6d4eb28465c9f734a946cbd7299f85ddbb16
thread:issue:119452,"issue #119452: RuntimeError: !block->expandable_segment_ INTERNAL ASSERT FAILED at ""../c10/cuda/CUDACachingAllocator.cpp"":2814, please report a bug to PyTorch.",stale/low urgency,2,stale/low urgency,1,0,119452,,,,0,8790c6258cd3ec15b92e172d97782c3927184ff921e38dc85b43066ccd473f6d
thread:issue:119462,issue #119462: Python 3.12 CPU TestBinaryUfuncsCPU.test_cumulative_trapezoid_cpu is failing,stale/low urgency,2,stale/low urgency,1,0,119462,,,,0,b8c977951506a978a0344736e5e66c648f66d23ce9436f230821a52d21b532af
thread:issue:119464,issue #119464: Python 3.12 CPU TestNamedTensor::test_using_unseen_interned/uninterned tests are failing,stale/low urgency,2,stale/low urgency,1,0,119464,,,,0,4d4c6f81b0b2726ccb06a804155949664b40d12af08838e900e9702213c06f53
thread:issue:119473,issue #119473: Document how to override autocast rules properly,stale/low urgency,2,stale/low urgency,1,0,119473,,,,0,22135e9a5fcaeea549f65006022d972e1d49267de08d84a10926e654787ca8e7
thread:issue:119474,issue #119474: Extend functionality of torch.nn.functional.fold to apply to sequences (1D tensors + batch & channels) not just images (2D tensors + batch & channels),stale/low urgency,2,stale/low urgency,1,0,119474,,,,0,cdc421b5f883b6c4a417a6bfa4ca99c367c686721026e315b589192f8a992546
thread:issue:119482,issue #119482: `pylint`: `torch.nn.functional.linear` is not callable,stale/low urgency,2,stale/low urgency,1,0,119482,,,,0,7cde81c852196466ec660a4091205cc8ea19c0198de9ddc96922aa3595b94428
thread:issue:119486,issue #119486: Conversion of 2d and 4d vectors in c++ to torch::tensor doesn't have the right data,stale/low urgency,2,stale/low urgency,1,0,119486,,,,0,bf0d3ffdf7155074c485cd19fb52cd90cdf9f9eb98a2882146b1a75e2ec9cbad
thread:issue:119524,"issue #119524: Consistency in using (mean, std) vs. (loc, scale) terminology in Normal distributions",stale/low urgency,2,stale/low urgency,1,0,119524,,,,0,97e0524665121380c023bb16c649e8a7a4fc32b8f9cf348faf82b728ab01e165
thread:issue:119548,issue #119548: Inconsistent behavior when indexing a Tensor with a list of lists,stale/low urgency,2,stale/low urgency,1,0,119548,,,,0,e8a3e556419d0c209cc4d98a6d62f9a000d1d6db5d6f0fed512c5ba7170bbb78
thread:issue:119549,"issue #119549: Way of mocking collectives, especially data-dependent ones like all2all_single",stale/low urgency,2,stale/low urgency,1,0,119549,,,,0,7e28cd88ba4e4d209f6f045edac67e1dcd3afefc1e31a4175e3cc6cc66dc9bd8
thread:issue:119557,issue #119557: Python 3.12 CPU Build system cannot find MKL libraries,stale/low urgency,2,stale/low urgency,1,0,119557,,,,0,1680e6560d46f3a3f63cd5d37065d9eb93486bf3aa1ebd3bcc7c5a8047744a97
thread:issue:119624,issue #119624: Submatrix multiplication selected by rows or columns,stale/low urgency,2,stale/low urgency,1,0,119624,,,,0,2de598793a54ccbfbb9c45c0f7c6a2ceaa526772866834d961af16f52981f789
thread:issue:119633,issue #119633: CUDA-GCC version mapping for CUDA 11.7+,stale/low urgency,2,stale/low urgency,1,0,119633,,,,0,a94048f70691ad9d54929b11545f0f17cb5385553e88b41a71e69d5985043dac
thread:issue:119653,issue #119653: I hope PyTorch will open the sliding window local attention feature as soon as possible.,stale/low urgency,2,stale/low urgency,1,0,119653,,,,0,f2e1fc3713c58bfd27b5da847ab281ebf8e4ab7b139dae5a47e70e3595a97ac4
thread:issue:119677,issue #119677: Unexpected Gradient Results From conv1d on MacOS,stale/low urgency,2,stale/low urgency,1,0,119677,,,,0,684dcfc274f6a96eea597fbf1be04b06057914aaaf426f051da576bd4b279c56
thread:issue:119711,issue #119711: torch.sparse.spdiags must support integer offsets,stale/low urgency,2,stale/low urgency,1,0,119711,,,,0,8cde4a6945a406a60ca2f96ad20979e54d2acec1a21b7d0a71cf27a0b20ed77f
thread:issue:119714,issue #119714: [NestedSubTensor] add communication operators support,stale/low urgency,2,stale/low urgency,1,0,119714,,,,0,d2a3adb2ff7a9f45c821de91ac34432981d3bb3a744ff4ed56fea4f09ffd3be3
thread:issue:11974,issue #11974: Specify out= argument to convolution,stale/low urgency,2,stale/low urgency,1,0,11974,,,,0,24a86f630e32e269864ec59e14b397e0900e9db938c3c1f17188861c9dbd9346
thread:issue:119740,issue #119740: Devcontainer for the nightly devenv,stale/low urgency,2,stale/low urgency,1,0,119740,,,,0,7676b1a25654dce0c6bc88007bfe6a5951b3e846ff11b337d5255e0dcbfc1d7a
thread:issue:11976,issue #11976: [caffe2] Documentation of Optimizers (Python API),stale/low urgency,2,stale/low urgency,1,0,11976,,,,0,cdb05433ddb74825a1331b49c69a17e088d437ff0da7dcd9fbfb86e9210a2b04
thread:issue:11978,issue #11978: [Caffe2] Attempting to install Caffe2 in Google Colab,stale/low urgency,2,stale/low urgency,1,0,11978,,,,0,4e16fedec6af856ef40ff71f094b657b4dd672d883064d669942fe3d9b9c14e9
thread:issue:11980,issue #11980: [Enhancement] Increase user-friendliness of dataset.random_split,stale/low urgency,2,stale/low urgency,1,0,11980,,,,0,a4ed6044359236a01923d1bde057d2f1e1f32c89c176cf22c9f484d08ff13d1a
thread:issue:119843,"issue #119843: Docker image pytorch/pytorch:2.1.2-cuda12.1-cudnn8-runtime uses python 3.10, but default for Ubuntu 20.04 is python 3.8 => lots of issues",stale/low urgency,2,stale/low urgency,1,0,119843,,,,0,99edcecb58ad10d38c3ce66b01c4e97a1dc13683051b46081df317858f715620
thread:issue:119873,issue #119873: Different numerical results during training with CUDA graphs,stale/low urgency,2,stale/low urgency,1,0,119873,,,,0,56f6c9735d2d9912417acb930fad911acda345fc2f561ea3e4aefb00ddc3f9d5
thread:issue:11996,issue #11996: Test the Caffe2 Installation with GPU erro,stale/low urgency,2,stale/low urgency,1,0,11996,,,,0,014dec3299a2f37444b9869361d188ed2a4b2e978c076e28a60351f2c3da0dde
thread:issue:119968,"issue #119968: Error on trying to run detectron2 on Mac M1 with device set to ""mps"" (Metal Performance Shaders)",stale/low urgency,2,stale/low urgency,1,0,119968,,,,0,a8de07ade86cc89ac0ededaf377c0ea1e155b2ce50098ee71abdf739ee53f74e
thread:issue:119971,issue #119971: Non core ATen ops are getting dispatched as it is instead of mapping/docmposing to other ATen ops.,stale/low urgency,2,stale/low urgency,1,0,119971,,,,0,884bc37299ff640e908aa7c1a1f72c52c4ef5256f9f0a099863b73da64fddba0
thread:issue:120000,issue #120000: Update Protobuf submodule to v3.21.X,stale/low urgency,2,stale/low urgency,1,0,120000,,,,0,06b06c809925b9d458a678672669edbaddfededa1c1af187dd8fcc8f16cbd074
thread:issue:120002,issue #120002: [c10d] torch.distributed.breakpoint stuck with multi-thread process group,stale/low urgency,2,stale/low urgency,1,0,120002,,,,0,a379c3e1495efca2e2db5421e5d6e7e4ec5107e7f46ed5159a32e221a118481d
thread:issue:120037,issue #120037: [FSDP][BE] print kwargs in run_subtests,stale/low urgency,2,stale/low urgency,1,0,120037,,,,0,c0771929a32dbbe6102ae03104654c548bd688de1691761972d8ff2d1ff6288c
thread:issue:120057,issue #120057: Torch compile fx graph is not removing constant propagation,stale/low urgency,2,stale/low urgency,1,0,120057,,,,0,01d961497b8808eb0d633958762ce6a7418c022f8b22e45df4cb0c92aeb6a2da
thread:issue:12007,issue #12007: caffe2's installation is still problematic...,stale/low urgency,2,stale/low urgency,1,0,12007,,,,0,eac956b0d5d62a5055adafd77de83ae058d7ba42119e3f23066095cc5470bc7e
thread:issue:120079,issue #120079: Use sys.settrace or torch function mode to compute how much of a model was not covered by Dynamo,stale/low urgency,2,stale/low urgency,1,0,120079,,,,0,05c46867d3d27b58e0d97078630f8947bea36f56259ceb17e57d29b70d018579
thread:issue:120085,issue #120085: Collate function in libtorch c++ API,stale/low urgency,2,stale/low urgency,1,0,120085,,,,0,9c48c1ae6fd5e597047def515032e12df454e7666b24ee2e2814f21c9a95590c
thread:issue:12010,issue #12010: Caffe2 compiled with MKLDNN doesn't have device_type = MKLDNN,stale/low urgency,2,stale/low urgency,1,0,12010,,,,0,1c43410fbece5c0dcda5f775c3c4c37dff35f8d7883c3f8bc8a8bffe189b56d8
thread:issue:120104,issue #120104: [memory profiler] keep consistent MB/MiB between y-axis and CUDACachingAllocator size,stale/low urgency,2,stale/low urgency,1,0,120104,,,,0,fb70245da430ca904134e5c18d4f16350d0e305626a705c2e1ca56ac2420d295
thread:issue:120116,issue #120116: Use Incremental Fake Tensor Updater more uniformly across torch.compile compilation,stale/low urgency,2,stale/low urgency,1,0,120116,,,,0,7c093346c78e48693e5224d43537741e634f5759b56ecb71a30d15c6f0e11d77
thread:issue:120209,issue #120209: `torch.put_along_dim` and `torch.put_along_dim_` like `np.put_along_axis`,stale/low urgency,2,stale/low urgency,1,0,120209,,,,0,cda596eb6bd24301a210a321e967f020a19097ed8a8d32d8389ac0a3c708c34a
thread:issue:120240,issue #120240: Cannot get deterministic Mask RCNN without running out of CUDA memory,stale/low urgency,2,stale/low urgency,1,0,120240,,,,0,86ad83a5c79e5598ae47a72dcac334cf841698d1f9dfd321678772ba574e8a80
thread:issue:120255,issue #120255: Add documention to the AT_DISPATCH functionality,stale/low urgency,2,stale/low urgency,1,0,120255,,,,0,3c28e1bfec3e5bb78acc6e70700c0e60b2fe0fe6c058217575e568aaf3e6945b
thread:issue:120291,issue #120291: Cannot download older version of libtorch ( libtorch-macos-1.9.0.zip ),stale/low urgency,2,stale/low urgency,1,0,120291,,,,0,0375b99fbae4d3c922478f907b8f588efe86bb619a82eda6db1460b6b496b468
thread:issue:120319,issue #120319: The CUDA download should include cusparseLt.dll since xformers now wants to use it in the context of torch,stale/low urgency,2,stale/low urgency,1,0,120319,,,,0,5cb51355a114cf784a9847b88f6a2fe3360bd33f5aab76fdeaac85e9ec2ba8b0
thread:issue:120354,issue #120354: Performance degradation for `aten::fill_` and `aten::add_` in backward pass,stale/low urgency,2,stale/low urgency,1,0,120354,,,,0,f69a6860fbc6d91819a4fcd847a8e50f94f80e3030a4f452dc982d01cc12df33
thread:issue:120396,issue #120396: Reinitialize parameters in TransformerEncoder,stale/low urgency,2,stale/low urgency,1,0,120396,,,,0,9bbd27dddc8a376a9f6dc75f274644f5ab12ea5d5a544f53f091153bd0914633
thread:issue:120407,issue #120407: Dark mode please 🙏🏻,stale/low urgency,2,stale/low urgency,1,0,120407,,,,0,ff78d16b6c62208726a9d2a51d62ce50c6aba6602d65983ab58dc07e2f019eac
thread:issue:12045,issue #12045: Create a custom function using the functions from math_cpu.cc source code in Caffe2?,stale/low urgency,2,stale/low urgency,1,0,12045,,,,0,4fa832335b8ad56dd351aa2c669860a2f012fa3d7b9bec7801a6fdd9ca8474c8
thread:issue:120463,issue #120463: Make upsample_bicubic2d_out_frame support BFloat16,stale/low urgency,2,stale/low urgency,1,0,120463,,,,0,a77bc244c58439a973dcc69c9bd37be9dc9db29f111bc081408477e61dfecc95
thread:issue:120466,issue #120466: faster communication,stale/low urgency,2,stale/low urgency,1,0,120466,,,,0,50d21c84005bf95ae608d3c06944a6c228ad0c07efe79d49a10f3a632f5b6e0c
thread:issue:120534,issue #120534: FLOP Counter does not handle kwargs,stale/low urgency,2,stale/low urgency,1,0,120534,,,,0,854a4cc58a892da59812860fb022f86c49c5833db7e2ffdcd25ed1561fff4582
thread:issue:120541,issue #120541: [c10d] distributed.breakpoint fallback when pg is not inited,stale/low urgency,2,stale/low urgency,1,0,120541,,,,0,107695795d2465e91983e3eff074a19e29334989bc169bc694d665dc3591b824
thread:issue:120572,issue #120572: FCD and tensor subclasses,stale/low urgency,2,stale/low urgency,1,0,120572,,,,0,ebb422b8bfd1f3579b75ee5737a31a2d5b6244d8960313405378f6d986cea616
thread:issue:120584,issue #120584: `torch.set_default_tensor_type() is deprecated as of PyTorch 2.1` appearing in logs even when not using this function,stale/low urgency,2,stale/low urgency,1,0,120584,,,,0,f1982db997bdfca16cd1e7f4e082d42ff099131f1fac73ca3358ddd8a416796b
thread:issue:120585,issue #120585: Functionalization of `as_strided` disregards its base tensor storage.,stale/low urgency,2,stale/low urgency,1,0,120585,,,,0,85c9e09f7e53984bb24f8a05697e8bf6b07cead3e2ce5175e4799716e1f46666
thread:issue:120587,issue #120587: Type hints on nn.Module.__setattr__ value,stale/low urgency,2,stale/low urgency,1,0,120587,,,,0,6e17e9964fde318ddf81bffca3ea3d0a8130e8f0d65a7efe61aa9b5a85ecfbbd
thread:issue:12060,issue #12060: cuda test hangs if GPUs in Exclusive Process mode,stale/low urgency,2,stale/low urgency,1,0,12060,,,,0,85ebd4f4e6cd359f4107b3874096397b96e07223d2418985838ffdebd7b62f6d
thread:issue:120667,issue #120667: Triton kernel unexpectedly gets 1.35x slower by more specializaiton,stale/low urgency,2,stale/low urgency,1,0,120667,,,,0,6e681874e6ddbd6d2830807847a10eb9227510dcede31cb61da853c0e09a097c
thread:issue:120689,issue #120689: Tracing UpSample get RuntimeError when use_determinitic_algorithms=True,stale/low urgency,2,stale/low urgency,1,0,120689,,,,0,2c7d2710bdd01db29d1bdf7d0e684fd6fc62b9257efd3927b3df25c795d13062
thread:issue:12069,issue #12069: Stop using make_intrusive directly; provide some make_tensor,stale/low urgency,2,stale/low urgency,1,0,12069,,,,0,97d82abe9e8d0a9bf3f3d67fc489930034ebd1cbe021022d2a8ce390f144a658
thread:issue:120708,issue #120708: ELU quantization parameters (scale + zp) missing from model's state_dict,stale/low urgency,2,stale/low urgency,1,0,120708,,,,0,2a3cd7072a0d0faf23c9659830205b7997a00b31143ed9b80ed88519aa6ba7ba
thread:issue:120710,issue #120710: Reduce duplication in number of places where we specify dependency pin,stale/low urgency,2,stale/low urgency,1,0,120710,,,,0,75c7a9b0c20368a9546e783290c94a02b60cc62120ae10190332cb83548ed6f0
thread:issue:120735,"issue #120735: `torch.optim.lr_scheduler.{LRScheduler,*}`: lack of documentation",stale/low urgency,2,stale/low urgency,1,0,120735,,,,0,bbb2803e317f21adc1f8b6acf6a522e195c5920c20ebb3fb8a5f3b0095f17864
thread:issue:120790,issue #120790: MultiheadAttention returns incorrect results for windowed long causal mask for long input,stale/low urgency,2,stale/low urgency,1,0,120790,,,,0,6eef7415366600bedfe8d95b4d7f49093cc9739ff807b92db5b81c1a3cb57388
thread:issue:120793,"issue #120793: libtorch MultiheadAttention doesn't accept different dimensions for Q, K, V",stale/low urgency,2,stale/low urgency,1,0,120793,,,,0,e4600a9384f5ece1b9fbeacf948469bae5a4907673316b13331c718dd1c99e88
thread:issue:120820,issue #120820: Problems differentiating through certain operations when exporting to TorchScript,stale/low urgency,2,stale/low urgency,1,0,120820,,,,0,af74f35617820251857fbe57f0160a0f635da08f17caa59eb76a867daaceea60
thread:issue:120822,issue #120822: Problem differentiating through a torch.layer_norm() call when exporting to torch.jit.trace,stale/low urgency,2,stale/low urgency,1,0,120822,,,,0,d751e652b465591778a41a2b44deee4e0c7a13f83e20eb442420427d8280e6df
thread:issue:120847,issue #120847: [C10D] DeviceMesh + _set_default_timeout doesn't work,stale/low urgency,2,stale/low urgency,1,0,120847,,,,0,8b067650040aa585141e79881f9de5fb0bb62324f02d4f084497eb4032cc88be
thread:issue:12086,issue #12086: Caffe2 issues with using Glog without GFlags,stale/low urgency,2,stale/low urgency,1,0,12086,,,,0,d62c82ded3ccc6e1d18abfe3e28d43f0518980e0853ab5aeda3ddae37ac15d01
thread:issue:120879,issue #120879: FSDP: unsharded gradients accessor,stale/low urgency,2,stale/low urgency,1,0,120879,,,,0,4c4398df37c9d8cb8136ddd3cdaf83748f4002b8422641e51c64c46b273c5956
thread:issue:120897,issue #120897: [memory profiling] about supporting the C++ Call Stack on the ARM Architecture,stale/low urgency,2,stale/low urgency,1,0,120897,,,,0,a6c65edc472c283532111cbe73584b30040d948b7a7df958ba6f51bf56b8ecbf
thread:issue:120973,issue #120973: TORCH_CHECK level changes in flatten,stale/low urgency,2,stale/low urgency,1,0,120973,,,,0,669838ded67fc80c3c08c2ff5878161e207af567c440d3be9805776bd8739423
thread:issue:120986,issue #120986: Intel MKL DFTI ERROR,stale/low urgency,2,stale/low urgency,1,0,120986,,,,0,d534756d42282d38ec1337e6895558480cb9517af9a444029df3098d4d6626be
thread:issue:120989,issue #120989: input.is_sparse() INTERNAL ASSERT FAILED,stale/low urgency,2,stale/low urgency,1,0,120989,,,,0,a3dae15205b34d5233240c47a2dde40b1eba2d0928f84433285df04f20346735
thread:issue:120994,issue #120994: OOM with mixed precision with OPT1.3B Decoder,stale/low urgency,2,stale/low urgency,1,0,120994,,,,0,245a11d3ed42caac348a0529e381482f05596825b04dc04a8da4af1ee23860eb
thread:issue:120996,"issue #120996: Unnecessary CUDA synchronization, or missing warnings",stale/low urgency,2,stale/low urgency,1,0,120996,,,,0,583700c6ad11341a6ecdf0ae44e476a4b4f3078bc0a1388cd4e7aeff9c861dd9
thread:issue:121090,issue #121090: Pytorch nightly docker images set PYTORCH_VERSION variable that doesn't match the installed torch version,stale/low urgency,2,stale/low urgency,1,0,121090,,,,0,429f6983455873a423f392eed5bc884a08d49a3b567902a882791f3db093523c
thread:issue:121131,issue #121131: Add flags for device that does not support fused or foreach ops,stale/low urgency,2,stale/low urgency,1,0,121131,,,,0,8e43f6e0e76fc3721e64027427f88ac8bbb5859e4725851737abef6c36fefafe
thread:issue:121160,issue #121160: RFCS review request: fast Viterbi Decoding,stale/low urgency,2,stale/low urgency,1,0,121160,,,,0,90a95c92a0d77347eced8257b2e075cbb9ecb1bffe6f21bfeab4212c474eae4a
thread:issue:121162,issue #121162: opcheck should support TorchBind custom classes,stale/low urgency,2,stale/low urgency,1,0,121162,,,,0,25ce25241cf03e847365db40da76f058565263de6ae51ff0d02a362e22294a11
thread:issue:121203,issue #121203: How to clear GPU memory without restarting kernel when using a PyTorch model,stale/low urgency,2,stale/low urgency,1,0,121203,,,,0,506951d6f459212abfc4ba17216d5973d3111e9e5705d5ab4d61103f5f168732
thread:issue:121215,"issue #121215: Build torch From Source is failed on Mac, maybe caused by protobuf",stale/low urgency,2,stale/low urgency,1,0,121215,,,,0,68b66bece49b65489b9055c945779b949999f8113a08b91c1a22e05e6cef3045
thread:issue:121220,issue #121220: Misleading `log_abs_det_jacobian` shape for `StackTransform`?,stale/low urgency,2,stale/low urgency,1,0,121220,,,,0,b4ea0253d69f9c5c53339b5763f4a1bbff07471698c6acbf60e7aecbea75e99b
thread:issue:121226,issue #121226: input_data.to produce inconsistent results,stale/low urgency,2,stale/low urgency,1,0,121226,,,,0,a0a0470c6bc3a99c2c6c083b10e3333aa3ff29891c6789dd81ce4145e62db894
thread:issue:121261,issue #121261: Convolution and norm with vmap jacobian failing,stale/low urgency,2,stale/low urgency,1,0,121261,,,,0,a0949975b7745fa225fa46a1579dccea178e664f69dbb5226fadebb0658dfc25
thread:issue:121271,issue #121271: `ReduceLROnPlateau` does not call `super().__init__()`,stale/low urgency,2,stale/low urgency,1,0,121271,,,,0,a284e22eb27505533816204dc026ff9487b0db2362d67b32a775350640f87f80
thread:issue:121286,"issue #121286: Assertion error (`assert op == ""==""`) in symbolic_shapes.prettify_results for operator `<=`",stale/low urgency,2,stale/low urgency,1,0,121286,,,,0,79a5130e9973094f546a382ccb1bb024165e7a719340abc3d0fc17272aa934da
thread:issue:121308,"issue #121308: installed pycuda, loss in pytorch has bug",stale/low urgency,2,stale/low urgency,1,0,121308,,,,0,856a5fdc3796129246d8e624b6cef16e511a3fd9435594bf41a4de5d761b12c9
thread:issue:121311,issue #121311: In-place native functional collectives right after a custom torch op fail to codegen in inductor,stale/low urgency,2,stale/low urgency,1,0,121311,,,,0,56fc787ddad89deadae3aa6a420a45fd0a043551a329ae1c40f1cdbfafb4547c
thread:issue:121333,issue #121333: Automatic graph break bug reporter,stale/low urgency,2,stale/low urgency,1,0,121333,,,,0,e985105381a01184acb3f73f500151b406fea5a77374f8e7a53d27a65e9f37d8
thread:issue:12134,issue #12134: [caffe2] How to set different learning rates for different layers?,stale/low urgency,2,stale/low urgency,1,0,12134,,,,0,ecac7ef8ecab906cec93a5d281f2f55eb5affdc9f71a4a55389dd4d0f7f54c35
thread:issue:12136,issue #12136: [Caffe2] GAN No Gradients in Generator,stale/low urgency,2,stale/low urgency,1,0,12136,,,,0,6c187109fe9b96e63be032208b6ab1d4b19e574ffcc60e58b0b7a15f092cfd38
thread:issue:121371,issue #121371: Match HuggingFace T5 SDPA pattern in Inductor,stale/low urgency,2,stale/low urgency,1,0,121371,,,,0,34d777188017e117022280c7f522bce4d6e29d6ffa5a561ebc6ce4c1df7ca055
thread:issue:121389,issue #121389: `torch.gather` backward performance extremely low on `torch.bfloat16`,stale/low urgency,2,stale/low urgency,1,0,121389,,,,0,12f51584011647ea9cccb8f3c3ea36077c04a8b32e9b64a74898c8fd26c2516f
thread:issue:121430,issue #121430: Getting a CUDA tensor through a Queue from within the same process causes a segfault,stale/low urgency,2,stale/low urgency,1,0,121430,,,,0,323e004acdbfeab0220ccd6c9dd715bb74f722e5e0b5a1e3f5a3328563ba8964
thread:issue:121439,issue #121439: copy_() produces wrong results for boolean tensors on mps,stale/low urgency,2,stale/low urgency,1,0,121439,,,,0,6ffffc5db6862ce25528f54835f9dc1b630db0ead00a71529e939dc1a287cdeb
thread:issue:12146,issue #12146: Misleading step method in lr_scheduler.ReduceLROnPlateau,stale/low urgency,2,stale/low urgency,1,0,12146,,,,0,b0ae642eff2780b704e73ef5cb81f6b23e364121814c70565e1acce720d279a4
thread:issue:121465,issue #121465: [RFC] PagedAttention Support,stale/low urgency,2,stale/low urgency,1,0,121465,,,,0,a3f1c31dceee7200a4e1ef7362e76b5abb6fb91310509e01222ca34b01233a20
thread:issue:121482,issue #121482: torch.linalg.eigh is very slow compared to tf.linalg.eigh,stale/low urgency,2,stale/low urgency,1,0,121482,,,,0,480b64555631c0827b412465a99281167f8a18d6ce0f5b9f83c565d493106fae
thread:issue:121485,issue #121485: libtorch memory leakage when inference by Multi-thread,stale/low urgency,2,stale/low urgency,1,0,121485,,,,0,7b26e17ec3005494f9b1470bdd507d40dc194b07d43d8357ec97bf77745bace6
thread:issue:121519,issue #121519: Avoid shard 0 download on all ranks,stale/low urgency,2,stale/low urgency,1,0,121519,,,,0,1834d670c27176dcb894e194c9a2ba0000772a72ee41eb96db7fb845c2552233
thread:issue:121588,issue #121588: logistic sigmoid function outputs NaN indeterminate on complex input,stale/low urgency,2,stale/low urgency,1,0,121588,,,,0,4fc8ae4b7ca81926151a1637587b0228669d9e86f7cb2c34c379bf1820dc6d69
thread:issue:12159,issue #12159: clip_grad_norm_ does not work on grads of different types,stale/low urgency,2,stale/low urgency,1,0,12159,,,,0,b89e805c9beae7c39d8502fa3078fcc134b6c938c9dc85bcd4aa70acb56b948f
thread:issue:121595,issue #121595: Different Dropout behavior on macOS and Linux,stale/low urgency,2,stale/low urgency,1,0,121595,,,,0,9ea4ec16103bff13e404e762865bb025d07bb82e0a4b4265e98718d959a02f27
thread:issue:121629,issue #121629: Pad operator produces non-nan value with nan input,stale/low urgency,2,stale/low urgency,1,0,121629,,,,0,bade660cf2e4517004670ee2734cc317f090c69190013e2919bf8d1eac24370f
thread:issue:121631,issue #121631: [inductor] Incorrect handle of `autocast` results in type mismatch,stale/low urgency,2,stale/low urgency,1,0,121631,,,,0,53b9af386607a9c4320854863323b1c67299ec64b1b052a27c730035acb8df99
thread:issue:121632,"issue #121632: [BUG] Given boolean tgt_mask, TransformerDecoder produces wrong results with MPS backend",stale/low urgency,2,stale/low urgency,1,0,121632,,,,0,4617e5c02c5b45f925a073f04fde07966b0221f61df282ae9c4fb895df7d998f
thread:issue:121641,issue #121641: torch.isin on GPU vs numpy isin: perf way slower than expected,stale/low urgency,2,stale/low urgency,1,0,121641,,,,0,bf458ade35d0cf98910d6d834cf051c999fdaab1189a535bbdc6b68136c217d4
thread:issue:121650,issue #121650: add scatter to functional collectives,stale/low urgency,2,stale/low urgency,1,0,121650,,,,0,6891d2ce229ac1510383231bc8c184bff61cdb2bc8282abe8eefc8b44a798348
thread:issue:121661,issue #121661: Look at gather + gemv performance.,stale/low urgency,2,stale/low urgency,1,0,121661,,,,0,d08cf7141ebadefe2c0a53c7b8aa8a97181e09c9510d71dae5e2068a73ff2c86
thread:issue:121668,issue #121668: Warning static library kineto_LIBRARY-NOTFOUND not found when trying build tutorial torch.compiler_aot_inductor.html,stale/low urgency,2,stale/low urgency,1,0,121668,,,,0,39f578c2cc07c9808f87c018b81c0fe3d98b8d4d4995eced9aad462e74826ae5
thread:issue:121683,issue #121683: linalg.ldl_factor raises an error with zero tensor,stale/low urgency,2,stale/low urgency,1,0,121683,,,,0,a9b270f72bb77fd10558b733753debe98256cf622d3f615422e8b046a620c4c7
thread:issue:121715,issue #121715: Depthwise conv2d op produces nan values with non-nan input,stale/low urgency,2,stale/low urgency,1,0,121715,,,,0,9126a0070510fe4aa6feb9c625a28ddd0de032fddbd0f21da44539961116f3cd
thread:issue:121719,issue #121719: index_copy change value when use_deterministic (true),stale/low urgency,2,stale/low urgency,1,0,121719,,,,0,75b6f51b693cd35f5380e30bd76ebe65abe06ec88697dd9f00e57297a87d9e3f
thread:issue:121725,issue #121725: torch.logsumexp throws RuntimeError instead of reducing all dimensions when dim=None,stale/low urgency,2,stale/low urgency,1,0,121725,,,,0,cd631f1c87e0aa66ab8203f365a9ee98234bf927b2f79452bca65b504aa0fc87
thread:issue:121737,issue #121737: Unable to install pytorch from source for version 2.0.0,stale/low urgency,2,stale/low urgency,1,0,121737,,,,0,1d0b38ee8e410bef0387b45d129be3372dc6b9064d98917cd81039df3132db8d
thread:issue:121789,issue #121789: Precision Inconsistent for `half` type compared with NumPy and TensorFlow,stale/low urgency,2,stale/low urgency,1,0,121789,,,,0,fa858264ac72b54d85c477acc86e8c29abc749c71e23fb6a8e5ef393ac5b5181
thread:issue:121797,issue #121797: Can’t save SparseTensor for custom device,stale/low urgency,2,stale/low urgency,1,0,121797,,,,0,63b2d984cf42c348fe0575d3447e0368ce95fc87fae7ca4300b5eb2f7f0d23fa
thread:issue:12181,issue #12181: Network surgery for transfer fails,stale/low urgency,2,stale/low urgency,1,0,12181,,,,0,1468e7e681663f53fabcb7646e91c2cfea5ae193bf4c011b5e02ac9c3a0fd01e
thread:issue:121842,issue #121842: gradcheck fails when inputs are dependent on each other,stale/low urgency,2,stale/low urgency,1,0,121842,,,,0,88b812385382474140b70b9901f76be34d6831ce72411cda7e27e42b8bc31a5a
thread:issue:121857,issue #121857: AdamW(fused=True) slower than unfused AdamW,stale/low urgency,2,stale/low urgency,1,0,121857,,,,0,159dbd23f5db742825dc25d02c01849bb609d20e99bc013cdb71f06249405d68
thread:issue:12189,issue #12189: [Feature] Support Adaptive Max Gradient Norm / Clipping,stale/low urgency,2,stale/low urgency,1,0,12189,,,,0,10b67818b464c530d4f9e8097de74b70191ca3e3eafaf89b652ebfeb7f46290e
thread:issue:121917,"issue #121917: [PythonDispatch] when the value of a argument is the same as its default value, __torch_dispatch fails to capture it.",stale/low urgency,2,stale/low urgency,1,0,121917,,,,0,3d1eb3459109855d09ef9270a44a331378fa52c80a3f5298b6d2831da0874c85
thread:issue:121926,"issue #121926: Call `torch.pow(int_tensor, torch.tensor(-1))` does not raise error like NumPy.",stale/low urgency,2,stale/low urgency,1,0,121926,,,,0,9921b601222e0f963da1cb6aece7e7d4829aaf8f7643f79f1097694d982d415e
thread:issue:121960,"issue #121960: RuntimeError: ""max_pool2d_channels_last"" not implemented for 'Half'",stale/low urgency,2,stale/low urgency,1,0,121960,,,,0,a8384a7dce86edfadf4589b4d0ac04b8d829f33ded4ecf99993910035cb22a13
thread:issue:121961,issue #121961: Numerical error when using torch.nn.functional.pad with a large array on MPS device,stale/low urgency,2,stale/low urgency,1,0,121961,,,,0,d599ff832aa4708743d812dc89963ce17308ba6a2e878af05fd0660288902fec
thread:issue:121967,issue #121967: [PT2.2] PyTorch dispacthing non core ATen ops in .compile mode for torch ops.,stale/low urgency,2,stale/low urgency,1,0,121967,,,,0,49fed11802ec8e53b12261ca13ebd3dfdbc86123340f06cf8c54f274296d549b
thread:issue:121968,issue #121968: [RFC] Use CUDA graphs by default on torch.compile,stale/low urgency,2,stale/low urgency,1,0,121968,,,,0,7ee2ae1ef4c9ce74cccd04986ddf73b052b16e8d522d85893333aa2586384f1a
thread:issue:122017,"issue #122017: Got ""KeyError: 'classname'"" while runing pytorch unit tests",stale/low urgency,2,stale/low urgency,1,0,122017,,,,0,454774e5acbe8e612ba726d883c7e58e7b36dfb20e14fec4b4ddd46e585a7035
thread:issue:122021,issue #122021: Wishart: problems with singular samples.,stale/low urgency,2,stale/low urgency,1,0,122021,,,,0,46863b42cafa43f5824a1c389755b2e97bbece89dbdfd89385e9cf46c55e2eb2
thread:issue:122027,issue #122027: PyTorch fails to build with mkl if cmake is installed from pip,stale/low urgency,2,stale/low urgency,1,0,122027,,,,0,03fade6d25b9833b162e442871c976df8d560b7313dc13fe64b4daa2af4ae29e
thread:issue:122030,issue #122030: [MPS] Wrong Calculation on MPS,stale/low urgency,2,stale/low urgency,1,0,122030,,,,0,73966913f29c26449c8ce542dd30ceb698a1d9bd82b4d7ef391541bafb5bdb45
thread:issue:122068,"issue #122068: NVML_SUCCESS == r INTERNAL ASSERT FAILED at ""../c10/cuda/CUDACachingAllocator.cpp"":830, please report a bug to PyTorch.",stale/low urgency,2,stale/low urgency,1,0,122068,,,,0,ab4644d0f0b6d92a876e1f741575eac05314e67af5b0f797996296fa7001ae07
thread:issue:122080,issue #122080: graph.owning_module does not point to the correct module after dataparallel forwarding.,stale/low urgency,2,stale/low urgency,1,0,122080,,,,0,575ea4e16498810e70bfefa5668c9bb1e76c6a70922e9b07f1c7cd1b77644b4c
thread:issue:122081,issue #122081: [Feature Request] add a new argument `storage_offset=None` to `empty_strided`,stale/low urgency,2,stale/low urgency,1,0,122081,,,,0,7ac1cd8a8177036f83f1527bb88d60030a4d097183698e8710f24b21b0847cba
thread:issue:122089,issue #122089: Infinite cycle on BSC tensor: base -> values -> base,stale/low urgency,2,stale/low urgency,1,0,122089,,,,0,f3aba911cb61968363f2452d7adf723730c8a88114ae5a91862c362f9e64a08d
thread:issue:122092,issue #122092: masked_index_add,stale/low urgency,2,stale/low urgency,1,0,122092,,,,0,9d670a72460e2f406320d0a3849ea034a5af249d97feefb864b6863cedb1477e
thread:issue:122107,issue #122107: Can't export an onnx with dynamic axes if it contains backpropagation,stale/low urgency,2,stale/low urgency,1,0,122107,,,,0,a11e6d02c19962046de16b9320ddd2520e79872818fdcc473c9331497bd8a0f8
thread:issue:122170,issue #122170: torch.cumsum() super slow on 2d input,stale/low urgency,2,stale/low urgency,1,0,122170,,,,0,99a88cbdc74030d5095c19a9240ef5339e3a2c41a3e4734b3d22c62c51e958ae
thread:issue:122185,issue #122185: Numerical issues on positive semidefiniteness constraint,stale/low urgency,2,stale/low urgency,1,0,122185,,,,0,be9fd0f08ca553a68f3e7add4947cfd27664ad87bf2a347e7a9ae707dfdd513a
thread:issue:122186,issue #122186: Deep-copied model cannot be sent through multiprocessing `Pipe`,stale/low urgency,2,stale/low urgency,1,0,122186,,,,0,489a5eee3810ce7d6e976aa848d9a6deffff4dee0c72a5e7a0ed37a0dccb28e8
thread:issue:122187,issue #122187: CuDNN version for pre-built package conflicts with the pip installed version (pytorch 2.2.1 + cuda 12.1),stale/low urgency,2,stale/low urgency,1,0,122187,,,,0,480864f485c22894d25b9f94e8a66dc35039399bf8d66052308e99ac5bf011c4
thread:issue:122250,issue #122250: Proposal to make `torch.nn.modules.utils._pair` public,stale/low urgency,2,stale/low urgency,1,0,122250,,,,0,e8034843a1b247ce5535cc925d2101024989d85160f8bf531422bf57597899f7
thread:issue:122274,issue #122274: CUDA: Performance issues with `torch.unique`,stale/low urgency,2,stale/low urgency,1,0,122274,,,,0,f81e9757b6f2f8735881eea0a0343c15ae8df2421e8899fd509c9525188d6d21
thread:issue:122290,"issue #122290: In a simple model, exporting onnx loses model operations",stale/low urgency,2,stale/low urgency,1,0,122290,,,,0,6b9f4ec383f9f261e9ccee4384655fac2d57bc8601282e1a3bd3f0741b89b2f4
thread:issue:122321,issue #122321: Batch size can't be varied if you export self attention to onnx,stale/low urgency,2,stale/low urgency,1,0,122321,,,,0,a97e97ea6fce5d47d4a289c0dfc17370f7c0f36e93a44f146e468ad7881bd21b
thread:issue:122359,issue #122359: Onnx export memory leak with qlora+deepspeed stage 2+ORT,stale/low urgency,2,stale/low urgency,1,0,122359,,,,0,5209692a7754280d38b18dea1f1e7924a81eafc6dede49e39f17cd2f5d74bf81
thread:issue:122454,issue #122454: Triangular matrix storage + matmul,stale/low urgency,2,stale/low urgency,1,0,122454,,,,0,79cfe6bbc591fa18305f7f11ccb5090fc5f1e01a958e5a3ec2128d5b62d6c9c5
thread:issue:122467,issue #122467: RuntimeError: out_ptr == out_accessor[thread_count_nonzero[tid + 1]].data() INTERNAL ASSERT FAILED,stale/low urgency,2,stale/low urgency,1,0,122467,,,,0,3c8b9834d0a4b1aa9d403010468d274e17e47d961dfe1b6a377da9e2a4c59084
thread:issue:12248,issue #12248: .cuda() changes a module's behavior when there are registered buffers with requires_grad=True,stale/low urgency,2,stale/low urgency,1,0,12248,,,,0,d594fcb2903e1ff66d75b0178f9b72d6e85de8be8787aa35dcaf3874b012d5a6
thread:issue:122485,issue #122485: DTensor doesn't support expand_as dispath mode,stale/low urgency,2,stale/low urgency,1,0,122485,,,,0,e02850fb3caa782fd0250f9a71b208314303d79a063f3ae34f124da9d4f90a2d
thread:issue:122529,issue #122529: Distributed checkpoint state_dict load may report nccl error and hide the real root-cause exception,stale/low urgency,2,stale/low urgency,1,0,122529,,,,0,3e411002e23f92f85e2ad25a7370037c9b4936717a36bf3e07c8aff727f9e738
thread:issue:122550,issue #122550: Adam optimizer：RuntimeError: t == DeviceType::CUDA INTERNAL ASSERT FAILED,stale/low urgency,2,stale/low urgency,1,0,122550,,,,0,53862eba78ed41844dbb2eebe222951a473beb28dbbee4d27f62ba95acc55de7
thread:issue:122552,issue #122552: Could not run 'aten::empty_strided' with arguments from the 'CUDA' backend.,stale/low urgency,2,stale/low urgency,1,0,122552,,,,0,a785945d1b3951452d5953ce10ab7d343c6e860578c059467fbdec9242bb3a28
thread:issue:122568,issue #122568: How do I use exec in a Pytorch Module and train with multiple GPUs?,stale/low urgency,2,stale/low urgency,1,0,122568,,,,0,26f7d89ea5421e78d1fdb9c1729094515b901dceaf813c111578071346360e6c
thread:issue:12257,issue #12257: [Caffe2] Relink error after installing Caffe2 from conda,stale/low urgency,2,stale/low urgency,1,0,12257,,,,0,0b9d75ba7de1e50d78201d9f96581e7e7a87a02c7152a5d31e582647031eff7b
thread:issue:122571,issue #122571: some official docker images has mismatch nccl version with pytorch,stale/low urgency,2,stale/low urgency,1,0,122571,,,,0,fc6f9216cd433337eb43a738ae5e4bc7e168232583d71f407897ea43f03278e4
thread:issue:122592,issue #122592: build libtorch in windows with intel oneapi base 2024 error,stale/low urgency,2,stale/low urgency,1,0,122592,,,,0,ce720496b7168232a3d73089327242f256c2372a553c37e5a18095c4084b8e1c
thread:issue:122626,issue #122626: Significant performance degradation with multiprocessing in PyTorch 2.x compared to 1.13.1,stale/low urgency,2,stale/low urgency,1,0,122626,,,,0,c3f3bbfa60073f3522f4ae24f28aab7cab7e3f755b7f53e4f126d8d06ac90bc8
thread:issue:122694,issue #122694: elastic agent kill hang process when other processes timeout,stale/low urgency,2,stale/low urgency,1,0,122694,,,,0,e2428bbade4bd88f3e7d2dbc1ffd9463a009be8fc80ac3e8adc1bf2c06968628
thread:issue:122699,issue #122699: Bug of `svd_lowrank` when passing parameter `M`,stale/low urgency,2,stale/low urgency,1,0,122699,,,,0,e3d0bdd7d0e46c9f0b68fdc354601fdfea0aef47e4e5c098fb6c8aa5158cf1f2
thread:issue:122701,issue #122701: Unable to call len on nested tensor,stale/low urgency,2,stale/low urgency,1,0,122701,,,,0,90e40ed4fdc7594e6f257290e8497c87917e89e79056ac3ef308fa114837be13
thread:issue:122711,issue #122711: `torch.sparse.sum` does not support boolean and int when summing over dense dimensions,stale/low urgency,2,stale/low urgency,1,0,122711,,,,0,504ddb349ed624545ec849cfee4dd4b2feab9224a1ab2d21fb9c1fdc720be7ac
thread:issue:122756,issue #122756: How to reduce memory usage for large matrix calculations？,stale/low urgency,2,stale/low urgency,1,0,122756,,,,0,0ff9d7e085ed46c501a25a7d43af5a797c9d9b6b85f7b47a32f1fd035b5a528a
thread:issue:12280,issue #12280: `GLIBC_2.23' not found on Ubuntu14.04.,stale/low urgency,2,stale/low urgency,1,0,12280,,,,0,a402d8248912a5f79b71d9821e7c879558ed2937102cde7cc2ede8d19c844675
thread:issue:122803,issue #122803: Enabling `swap_tensors` path by default for all tensor subclasses errors on some transformer tests,stale/low urgency,2,stale/low urgency,1,0,122803,,,,0,3f16d0c795cdf5b6412d4314dbdcccaa41d7df05271c125760fa28f984ac4658
thread:issue:122814,issue #122814: torchdim as torch_function subclass exploration,stale/low urgency,2,stale/low urgency,1,0,122814,,,,0,e070e8729670fa15748cb60795eb559497a271a7e85437e23e5464dcda2e1ebf
thread:issue:122931,issue #122931: Stack smashing detected on nccl initialization,stale/low urgency,2,stale/low urgency,1,0,122931,,,,0,4e838125c6979c4294fb3956b1f1988ed9fdffa71bed65429d2c74ebc954eb7e
thread:issue:122957,issue #122957: How to export torch.optim.LBFGS using torch.onnx.export,stale/low urgency,2,stale/low urgency,1,0,122957,,,,0,1006b6e9efca34e1b05c420567df8c9e89a78cb189b8a8bbb750cd217f80b399
thread:issue:12297,issue #12297: [caffe2] Inference using multiple GPU,stale/low urgency,2,stale/low urgency,1,0,12297,,,,0,1bc23abb3fb68291590490593dfcc3642297bf4518be0360aa266ece1b93abbe
thread:issue:122973,issue #122973: mps training is slower than cpu,stale/low urgency,2,stale/low urgency,1,0,122973,,,,0,68cca6d6b91788a20e0e8d431ddca9910b7bbb271316011549eebd2661d9b285
thread:issue:123047,issue #123047: Alternate algorithm for computing MaxPool2D under specific condition.,stale/low urgency,2,stale/low urgency,1,0,123047,,,,0,8a68a4c1c5b607a6526910b825f78e3c403d37719098de06cf43283533c69b8f
thread:issue:123062,issue #123062: Enable UFMT on all files in PyTorch,stale/low urgency,2,stale/low urgency,1,0,123062,,,,0,f11bcc435dc88ee4a345f7f9de38dcb103399fc28899bdd0a5bff3865040354e
thread:issue:123070,issue #123070: CUDA out of memory still exist after using FSDP,stale/low urgency,2,stale/low urgency,1,0,123070,,,,0,d94bcfd7fe2312e6e173b84803b2f70ba057af3b721873443af3412b82b6be69
thread:issue:123072,issue #123072: Saving & Loading Model Across [In-tree and out-of-tree] Devices,stale/low urgency,2,stale/low urgency,1,0,123072,,,,0,e99a11be84d4c5b138035d832f299161680a744a524c5f446302bccc883e0420
thread:issue:123075,issue #123075: LibTorch Release version 2.2.2+cpu loads traced model only in Debug configuration,stale/low urgency,2,stale/low urgency,1,0,123075,,,,0,bb3da2f09212c5f09683df24001cbd81f0f173ee635eea7e8b0d0cfc408c7313
thread:issue:123091,issue #123091: build error: memory_desc.cpp:716:1: internal compiler error: Segmentation fault,stale/low urgency,2,stale/low urgency,1,0,123091,,,,0,4757c61ece419611387e848aa31fc13ab824d438a8faf4ea58db949e7dde7375
thread:issue:123113,"issue #123113: Please add instructions for ""uninstall compiled PyTorch from source"" in the readme.",stale/low urgency,2,stale/low urgency,1,0,123113,,,,0,8cfcae8b8c5ef01791f60f0a3343525307fd33bbe605d4319314ae5c6eb8d8bb
thread:issue:123119,issue #123119: `nn.Embedding` behaves differently on CPU (fails) and MPS (fails not),stale/low urgency,2,stale/low urgency,1,0,123119,,,,0,2dfe8f2176ce193a60e91cefd38c2579bebc08fe81c4ea23868e88187df17623
thread:issue:123120,issue #123120: TD distributed CUDA Rollout list,stale/low urgency,2,stale/low urgency,1,0,123120,,,,0,beb5d8624038c2b20a795a93954ec959b513eafeab3875527011ca164655321a
thread:issue:123169,issue #123169: Need better documentation for running doctest locally,stale/low urgency,2,stale/low urgency,1,0,123169,,,,0,f2c87d1d590593c48fc9d6156bad05fbb143b47c7b6fa998c9b141385cba6f04
thread:issue:123172,issue #123172: `torch.utils.data.Sampler`: constructor's `data_source` argument should be removed,stale/low urgency,2,stale/low urgency,1,0,123172,,,,0,63db934c5bf842b06a10a6d1e7360f0d6bc464394432cceb95fe5261c87c4774
thread:issue:12322,issue #12322: [Caffe2] Segmentation fault (core dumped) while import caffe2.python.core,stale/low urgency,2,stale/low urgency,1,0,12322,,,,0,f65dcc3f4940ce17b9bb9dfb583ad5d089260f1e87b6ca64e2f771996ecd3081
thread:issue:123238,issue #123238: Fused optimizer kernel for CPU,stale/low urgency,2,stale/low urgency,1,0,123238,,,,0,a9aead837f92ab46001d83ed35dee9d45d945de01848269300e6c3bbed1d54d5
thread:issue:123245,issue #123245: torch.overrides.get_overridable_functions list functions that are not overridable via __torch_function__,stale/low urgency,2,stale/low urgency,1,0,123245,,,,0,45d0699bda64693b83cc94cec9e1c1681e3560bb5fb3582805fb9ead456079f8
thread:issue:123262,issue #123262: mmap error - no such device (19),stale/low urgency,2,stale/low urgency,1,0,123262,,,,0,940876388cc28fe6ebad56aa17f9949f6db5652df29dcad71415e9e5c38f0902
thread:issue:123284,issue #123284: We should walk torch._C and require everything have type annotations,stale/low urgency,2,stale/low urgency,1,0,123284,,,,0,fd29eb29e5f61804d37dd268bcb321e7c81746b69ab7425c539ed50028a2c872
thread:issue:123344,issue #123344: document torch.ops.import_module on the website; also make sure to use this in tutorials,stale/low urgency,2,stale/low urgency,1,0,123344,,,,0,47d2b3b051aef7e5a11349698248621c7eb6ad55088912b56b6e7053302edf19
thread:issue:123353,issue #123353: Type parameter (T) of Optype (Where) bound to different types (tensor(int64) and tensor(float) in node (/Where),stale/low urgency,2,stale/low urgency,1,0,123353,,,,0,790e13b94dd6d6b0c480278ffaeb1527ae021a0f279592ea30e1d2924be1c953
thread:issue:123465,issue #123465: lintrunner-noclang job may diverge from `lintrunner` locally,stale/low urgency,2,stale/low urgency,1,0,123465,,,,0,2abdbefab479da1121a6a264453f8b4bbf5a41449b20194d89404147a24717b4
thread:issue:123466,issue #123466: Undefined behavior when an OpOverload's Library is deleted but the OpOverload continues to be used,stale/low urgency,2,stale/low urgency,1,0,123466,,,,0,b58983f6684c4e82e15e121950a35b6838353eb6d9153eb39049b1a69d6b37d0
thread:issue:123500,"issue #123500: Cannot add 2 sparse_csc_tensor directly, raising RuntimeError",stale/low urgency,2,stale/low urgency,1,0,123500,,,,0,fbe217296a8ee435200592ce3f066cdaafd39dd77ced585070869ddff8d7f43e
thread:issue:123505,issue #123505: RuntimeError: Unrecognized CachingAllocator option: expandable_segments,stale/low urgency,2,stale/low urgency,1,0,123505,,,,0,503105e0d16224b3cd4dcc7a3a1ba60a5e73b6fd88d4aa680ef3da3375abfd98
thread:issue:123506,issue #123506: [RFE] Disallow submodule updates without successful minimal binary builds,stale/low urgency,2,stale/low urgency,1,0,123506,,,,0,43da12b8c0cd23fc8b0e53cbc11aa4139596845b4ec9392d89225d512fd737e5
thread:issue:123528,issue #123528: Support integer parameters in FullyShardedDataParallel,stale/low urgency,2,stale/low urgency,1,0,123528,,,,0,2140cb3fb295f903ff5f4a1b268c7dbf08d4d908920818679c1b102bd37f5c69
thread:issue:123532,issue #123532: Request: Restrict use of tags to strictly formal release versions,stale/low urgency,2,stale/low urgency,1,0,123532,,,,0,b98c530c57069685013d0421c02c5b4b1bb51c53cfaad470ddac3285ba92c984
thread:issue:123533,issue #123533: Memory allocation for Conv2d,stale/low urgency,2,stale/low urgency,1,0,123533,,,,0,92a65c7b4bdfc3043e5976a35f27a44401f1021ed879953345a83ccd2dfccf45
thread:issue:123548,issue #123548: An error occurs when ‘max_split_size_mb ’and ‘expandable_segments ’ are enabled at the same time.,stale/low urgency,2,stale/low urgency,1,0,123548,,,,0,b378c1fc32e4f1825cdbdaedffe33d92f6e40ac98b1fed975ea0e05d58016d73
thread:issue:123555,issue #123555: torch.onnx.errors.UnsupportedOperatorError: Exporting the operator 'quantized::linear_dynamic' to ONNX opset version 17 is not supported. Please feel free to request support or submit a pull request on PyTorch GitHub: https://github.com/pytorch/pytorch/issues,stale/low urgency,2,stale/low urgency,1,0,123555,,,,0,90c0ac3757fa7064b5fd89d41a8fa5a21c3ad327e576f8bb0f38e10f0238d863
thread:issue:123604,"issue #123604: Can't initialize NVML, H100 gpus on ubuntu",stale/low urgency,2,stale/low urgency,1,0,123604,,,,0,60163fb7b54f54ba06ede2a779cb431302e8bf9dd86ba070827aea25c23b8453
thread:issue:123628,issue #123628: `torch.onnx.export` doesn't correctly constfold constants and DequantLinear,stale/low urgency,2,stale/low urgency,1,0,123628,,,,0,66e642745266603b01456aca214bbb09b111a9adeab55ac09449ecb0e3bbf513
thread:issue:123642,issue #123642: TransformerDecoderLayer fails when on cuda and batch dimension is zero (on cpu is fine),stale/low urgency,2,stale/low urgency,1,0,123642,,,,0,f3b2db5495392cbe7caa9226a2ddb7fb849aa3d4c74c9a1692c93745d5cecfcb
thread:issue:123702,issue #123702: Issue with Vulkan Device for nn.Conv2d Operation,stale/low urgency,2,stale/low urgency,1,0,123702,,,,0,f1f05173188bed17f7e5ab986d70d3b32824243eb85b941ac13bf8391494ee92
thread:issue:123761,issue #123761: Support for no-frills FP8 matmuls,stale/low urgency,2,stale/low urgency,1,0,123761,,,,0,7d31b6f59242d3165d3ba3c07bf6bf20ae60014af220caf161b20bb2e221f135
thread:issue:123765,issue #123765: Accurately typed ValueRanges / convenient OpsHandler dtype inference,stale/low urgency,2,stale/low urgency,1,0,123765,,,,0,6607bd75cf2ba64f628763eb099df4dede05b4e5230a504a798bcaea18aad692
thread:issue:123834,"issue #123834: RuntimeError: NVML_SUCCESS == r INTERNAL ASSERT FAILED at ""/opt/pytorch/pytorch/c10/cuda/CUDACachingAllocator.cpp"":830, please report a bug to PyTorch.",stale/low urgency,2,stale/low urgency,1,0,123834,,,,0,03afdf6a0fe2a901e44b0382ee12cf3b5a49ff8170c501877a52c561504342a2
thread:issue:123911,"issue #123911: `torch.softmax(inp, dtype=torch.float32).to(torch.float16)` is not equivalent to `torch.softmax(inp)` for fp16 input",stale/low urgency,2,stale/low urgency,1,0,123911,,,,0,2f4dfc8ad9c5af9d410f1d20782d057aa74b08a0402ce62fc8665a44d574fd35
thread:issue:123971,issue #123971: Dynamo Export: Support for mutating module attributes,stale/low urgency,2,stale/low urgency,1,0,123971,,,,0,3985973491cd7215a452f1ee8ccc9ae90b297ea2e9a8881d61aca23220c142cd
thread:issue:123973,issue #123973: Dynamo-based ONNX Export: Failed to produce a graph during tracing as no tensor operations were found.,stale/low urgency,2,stale/low urgency,1,0,123973,,,,0,d377a372581e3d8df4865368af8fea04a2b06bb2c0a2403ba3a6d7ef00e1fbed
thread:issue:123996,issue #123996: MSE loss triggers ZeroTensor immutable error inside `torch.func`,stale/low urgency,2,stale/low urgency,1,0,123996,,,,0,d92ede54ab13d0824acefcef722e3a88f6486d69b1a5b3657bcd99e9584e2a0b
thread:issue:12401,issue #12401: AttributeError: Method VideoInput is not a registered operator,stale/low urgency,2,stale/low urgency,1,0,12401,,,,0,981a05ca695506ec5b67a644974d711e0524bfebe4780832c29d25a367d44d2d
thread:issue:124018,issue #124018: Fix OOMs during parallel builds,stale/low urgency,2,stale/low urgency,1,0,124018,,,,0,1ec5c2ed230e34ea0797da24022062f24a03038fd8abdb245ed7ea51c2ab2d43
thread:issue:124025,issue #124025: Different results for single sample vs. batched,stale/low urgency,2,stale/low urgency,1,0,124025,,,,0,7468fd931974e8714f33a4c59057ce00d03cfa0d3645514f0705f35ba7fc7efe
thread:issue:124029,issue #124029: rand_like on permuted parameter not working on MPS,stale/low urgency,2,stale/low urgency,1,0,124029,,,,0,8651adb1e2e1d83f14a63652803c0048615794739bd7dae8f87650a196ce9e30
thread:issue:124047,issue #124047: dlopen: cannot load any more object with static TLS,stale/low urgency,2,stale/low urgency,1,0,124047,,,,0,917c6c0af487482ca6171e1e56aeb6c9bd4d02bea32b52520779a5a8056e80c8
thread:issue:124056,issue #124056: MPS slows down after sleep,stale/low urgency,2,stale/low urgency,1,0,124056,,,,0,62c95fcb9d868e93367b1d0453e1f7f723c513b2e55f4d4402389793a1eb38e0
thread:issue:124083,issue #124083: Automatically bind all DispatchKey to Python,stale/low urgency,2,stale/low urgency,1,0,124083,,,,0,f0182c497ec8795f14c97d756d12833fe33dd9768a3faf80c14ee0d400085536
thread:issue:124133,issue #124133: ZeroRedundancyOptimizer + AdamW Fused can't load state_dict back,stale/low urgency,2,stale/low urgency,1,0,124133,,,,0,48e0cae2dfb86e47d1bb982427fb149668cfd6fb20fd204abe90bb52d8fb261a
thread:issue:124145,issue #124145: Use online softmax for just softmax by itself,stale/low urgency,2,stale/low urgency,1,0,124145,,,,0,192b61f391459512cc8f680e45940ef4e5dab9c4e7dea6f56d759fa9f2c93297
thread:issue:124194,issue #124194: Initial state of RNNs: h_0 or hx?,stale/low urgency,2,stale/low urgency,1,0,124194,,,,0,48cf923f743a0af200c3f1851eb884b3c642040f817b8f36cd084775c8798443
thread:issue:124195,issue #124195: torch.save and torch.load is slow. Slower than numpy. Slower even than pickle.,stale/low urgency,2,stale/low urgency,1,0,124195,,,,0,8349c417de403700dadc31132a9c63a4ba9e5467de0d1d8ab0a5cd05e1097fbd
thread:issue:124216,issue #124216: [NestedTensor] NJT with lengths set fails torch.compile(),stale/low urgency,2,stale/low urgency,1,0,124216,,,,0,3e94d5cb32e4dcc2e857723c1553074d5ed241ad97d0c00686fba7f5f4c723b8
thread:issue:124228,issue #124228: [DeviceMesh] Current cache for submesh result is not thread save,stale/low urgency,2,stale/low urgency,1,0,124228,,,,0,d472ac65af5a0057d0521c838001660d92312967a9737ecd0c8ae1c0ba1e1184
thread:issue:124270,issue #124270: PyTorch build fail with GCC 13.2.0 due to aten/src/ATen/core/IListRef_inl.h:171:17: error: possibly dangling reference to a temporary [-Werror=dangling-reference],stale/low urgency,2,stale/low urgency,1,0,124270,,,,0,3d92c45168be3045300964aadc88c45cca9ca8aad9d5965d55f0b4995864deb5
thread:issue:12435,issue #12435: `pstrf` on positive semi-definite matrices,stale/low urgency,2,stale/low urgency,1,0,12435,,,,0,1fafcbbb517253080f510a74fe8b366ddd463444aa1814315d69689b6fa82efa
thread:issue:124351,issue #124351: BUG Report using `backend:cudaMallocAsync`,stale/low urgency,2,stale/low urgency,1,0,124351,,,,0,41b7383d3ba1f9b92d5c7ad262b7969f04927e215d4270e236db4b63d4752530
thread:issue:124383,issue #124383: llvm disabled in official builds from 2.1.0 onwards,stale/low urgency,2,stale/low urgency,1,0,124383,,,,0,3a38193c502cda5b88e8a74180f52c15ddb2117fb3591d6e289740483615936d
thread:issue:124395,issue #124395: Provide TORCH_LIBRARY m.def schema types documentation,stale/low urgency,2,stale/low urgency,1,0,124395,,,,0,3466596150e07fe0af2e80a232ffc16bf3b70a5b4d7e2fc15b9e6b7fb2cdab55
thread:issue:124417,issue #124417: ghstack mergeability check seems wrong,stale/low urgency,2,stale/low urgency,1,0,124417,,,,0,822d0d44778fe3800b0889bef70f2cf2e55562ff1b312debacb83260482a2615
thread:issue:124427,"issue #124427: Find a common home for decompositions, perhaps outside of the obliquely named _refs directory",stale/low urgency,2,stale/low urgency,1,0,124427,,,,0,117a2320ba1ccba76b67cace383e0077c216980d2e9cf1f02a7ee72f476208e9
thread:issue:124454,issue #124454: torch.utils.cpp_extension.load recompiling every time,stale/low urgency,2,stale/low urgency,1,0,124454,,,,0,24c445a78466e9302a0e52e4189d8c31702b811687865be42e1940caee5b7217
thread:issue:124456,issue #124456: CudaHostAlloc takes a lot of time during training,stale/low urgency,2,stale/low urgency,1,0,124456,,,,0,d938320f3fd654d56781d386c16a6e90757eb801d3bba76d49ddc238a9153f2f
thread:issue:124465,"issue #124465: Matmul with broadcast makes copy because of reshape to bmm, and it is slow",stale/low urgency,2,stale/low urgency,1,0,124465,,,,0,c9fce8f34479f8ca329c80a19bfca4eaa74a0e74411430f531a0ac90182a2b98
thread:issue:124487,issue #124487: PyTorch Memory Management in GPU-to-CPU Transfers issue,stale/low urgency,2,stale/low urgency,1,0,124487,,,,0,7d167736d1eab62cb8c64e9f1f2345f8bb2c39755b30fab3cef51032a70c7f6f
thread:issue:124523,issue #124523: optim.apply_optimizer_in_backward does not account for gradient accumulation,stale/low urgency,2,stale/low urgency,1,0,124523,,,,0,996165d5559b5dd74bac2b85cdc913168167cebcb154d54aff67fac26f324d00
thread:issue:124566,issue #124566: Can't download pytorch/ timeouterror,stale/low urgency,2,stale/low urgency,1,0,124566,,,,0,cb581d6f30e884abfcef5c59d29c259ce8b2c1eb89c608603d26805033193a91
thread:issue:12460,"issue #12460: input_device, output_device, devices_used properties",stale/low urgency,2,stale/low urgency,1,0,12460,,,,0,8bb23397ee54e80ba330ee947ad7559a93c274a8bc67098630d93420aef964b8
thread:issue:124630,issue #124630: TORCH_SHOW_DISPATCH_TRACE should give information about PythonDispatcher,stale/low urgency,2,stale/low urgency,1,0,124630,,,,0,f6676257d6ab490ae64deb76ff1cf1bdae99a193c7fdc5d241fe412823c578b7
thread:issue:124645,issue #124645: Registering function that takes `const SymInt&` to op that accepts `SymInt` leads to cryptic error,stale/low urgency,2,stale/low urgency,1,0,124645,,,,0,c0f64a4be59ca676fd51d67a82ea56b723d7ee2666c157352c0548a6687e94fe
thread:issue:124686,issue #124686: Unbacked SymInts: Should backwards graph with unbacked SymInts be recompiled with hints,stale/low urgency,2,stale/low urgency,1,0,124686,,,,0,21f383482d4b1412a9cb074e0c0399869fe2feab7c71dc8b0f6e8c89893c3560
thread:issue:124711,issue #124711: Output Discrepancy between PyTorch Model and Converted ONNX Model,stale/low urgency,2,stale/low urgency,1,0,124711,,,,0,d23889c2b52f76c550d09adbda9f9acddeef0a4738aaba243279fd0c0314c97a
thread:issue:124714,issue #124714: `torch.distributed` hangs when using `torch.distributed.barrier` before any other communication primitives.,stale/low urgency,2,stale/low urgency,1,0,124714,,,,0,f7087581a1b92e27204746ad77c41629380bb85b9d7edb8b2ef05b63f55a8c99
thread:issue:124718,"issue #124718: Add scaled_dot_product_attention ""scale"" argument to nn.MultiHeadAttention",stale/low urgency,2,stale/low urgency,1,0,124718,,,,0,656e1d79de5be120fc2980c942a9863ce61180d3c8f156d87f0da0a1d5b190ad
thread:issue:124719,"issue #124719: Questions about parameter initialization, especially with torch.bfloat16 precision",stale/low urgency,2,stale/low urgency,1,0,124719,,,,0,c4e12a5e9a3b4a95ca567732f2439f8a5f2833be87ccb52a98a0ba6736d727be
thread:issue:124726,issue #124726: Foreach tests should xfail on all dtypes that are not supported,stale/low urgency,2,stale/low urgency,1,0,124726,,,,0,0399a3f7e96a9450f8d15ef0147d46f5a048e91fe9578c7921f5a655aa6a2785
thread:issue:124749,issue #124749: Improve Dynamo (and other) flaky tests,stale/low urgency,2,stale/low urgency,1,0,124749,,,,0,1a3a1868cf27d93a76fa082be2a7e76bf3df82bc1f6273b9af3766f460dc21a2
thread:issue:124778,issue #124778: Dense-sparse broadcasted multiplication fails in the backward pass,stale/low urgency,2,stale/low urgency,1,0,124778,,,,0,25ea464b79b590aa7e50b85505cb38d7b5721873bd6b626551a5f2a735847904
thread:issue:124783,issue #124783: Scatter_add limitation when accumulate beyond 2^24 under float32 precision,stale/low urgency,2,stale/low urgency,1,0,124783,,,,0,7002a6a474a23339bbb7c296acb23678a10bd742dc12ea8c8fce3302386f09c9
thread:issue:12482,issue #12482: [Feature request]: add `LayerNormLSTMCell`,stale/low urgency,2,stale/low urgency,1,0,12482,,,,0,e8673b7777f854b9e921910f50193e6ed8ad3ec903bca93cbe0731c67415edd6
thread:issue:124838,issue #124838: Forwardpropagation of high-dimensional tensors through the nn.Linear module becomes multiple times slower since pytorch 2.1.0,stale/low urgency,2,stale/low urgency,1,0,124838,,,,0,951b5d407ac0fa809bccf7e8487bbaed60e9993865942baac527cb18d28544d0
thread:issue:124857,issue #124857: Use lintrunner-adapters for our adapters when possible,stale/low urgency,2,stale/low urgency,1,0,124857,,,,0,c7d72b55f010c87fb5106d91bcf25bd8f0998cdf99de7efd2ee5cb39d66cf5f8
thread:issue:124866,issue #124866: support side effects in HOPs?,stale/low urgency,2,stale/low urgency,1,0,124866,,,,0,c5dab8a854c8d628748e7100885a442624d051c9b2f6429725d6ba9ce15ac452
thread:issue:124906,issue #124906: Many cases in distributed/elastic/multiprocessing/redirects_test.py fails when use pytest,stale/low urgency,2,stale/low urgency,1,0,124906,,,,0,98a6f6c0e28dffb915f28c4e7126945d4b71e7d5ec19c97d12e29c3b614f9f4c
thread:issue:124915,issue #124915: Exporting torch slice_scatter to onnx Identity,stale/low urgency,2,stale/low urgency,1,0,124915,,,,0,388c65d0808e35db9c972857342f2cb29c118b8d165c12d1887227dba10cf455
thread:issue:124934,issue #124934: Tensor's storage changes computation outcome for CPU tensors.,stale/low urgency,2,stale/low urgency,1,0,124934,,,,0,bc923527e7df8b75540ca8a9dbae3632bc72d356d27e2a1a65e76d4292909ea4
thread:issue:124937,issue #124937: Conflict between bias=False and why_not_sparsity_fast_path in Transformer Module,stale/low urgency,2,stale/low urgency,1,0,124937,,,,0,357ecd658f8dbe78f8d9c9132af95f1f6ffd5b5d11ef2f7ecf0dedf1ace7d3da
thread:issue:124953,issue #124953: PRs with unbalanced quotes can not be merged,stale/low urgency,2,stale/low urgency,1,0,124953,,,,0,0b37407949023aa138345874f9e04c6286b6fcbebfc65246dd5ab7e1a2492979
thread:issue:124956,issue #124956: torch.onnx.export generates an incorrect bias shape for certain conv transpose models,stale/low urgency,2,stale/low urgency,1,0,124956,,,,0,47d031f0f4395a95acfc0f7cf4608ec957a9eb0d8585b0d2846270b3f1db21b7
thread:issue:124973,issue #124973: Export llama v3 to ONNX,stale/low urgency,2,stale/low urgency,1,0,124973,,,,0,d17e60775d109cc5ac240339dc90f9543e07398b81a291c84e8393925f129ef0
thread:issue:124974,issue #124974: [Releng] Triton version for minor releases starting with 2.4,stale/low urgency,2,stale/low urgency,1,0,124974,,,,0,11f0e1312cec381f308c6f90ca32b7267b5a7a16fb5113c4c8e346bd5e6085e1
thread:issue:124989,issue #124989: [DTensor] Keep track of data-dependent ops and skip _propagate_tensor_meta to go thru fake tensor,stale/low urgency,2,stale/low urgency,1,0,124989,,,,0,80a18360c992abb065021e554c928b6b6dfc9bdba5ae806eeb36e961469d3fe1
thread:issue:124990,issue #124990: [DTensor] Sharding strategy not implemented error should be thrown earlier,stale/low urgency,2,stale/low urgency,1,0,124990,,,,0,e3ad4c394885fae294f9e2155e50a1c4f98db24223f7b9f10ff2b1ef9974b29a
thread:issue:124998,"issue #124998: When I set the operator weights all to 1, the operator weights change as soon as an input is made to the operator",stale/low urgency,2,stale/low urgency,1,0,124998,,,,0,73b53e13eedc7c1104bcdbc3b185c0fe29f48ca9f02c49c0a620ab11d1cfae40
thread:issue:12501,issue #12501: C++: Calling Workspace::RunNet for a prediction on a different thread each time causes a GPU memory leak,stale/low urgency,2,stale/low urgency,1,0,12501,,,,0,7bb00e1c014116b0cf764b05ee1c1be814e8109e60da3be49afba0529385fc2c
thread:issue:125028,issue #125028: OpInfo testing is inadequate for `nextafter`,stale/low urgency,2,stale/low urgency,1,0,125028,,,,0,10c518c046848c4a6c6dee4ac64fe2c3976e8359a8e0df85e2cbbdcdffd4659a
thread:issue:125051,issue #125051: MPS backend thinks that subnormals are equal to zero,stale/low urgency,2,stale/low urgency,1,0,125051,,,,0,829199b78953e80cce21df2aef44a5f53b0389da9914fb5290bf54e29b9f26b1
thread:issue:125064,issue #125064: ValueError: weight_norm of 'weight' not found in ParametrizedConvTranspose1d,stale/low urgency,2,stale/low urgency,1,0,125064,,,,0,7d8f590eb13d982a0d04f6906b349a66f2b8be3f0f0faa2086df21083a5d499a
thread:issue:125070,issue #125070: 2.2.0+ regresses SDPA performance on Windows,stale/low urgency,2,stale/low urgency,1,0,125070,,,,0,d65f342b72a126e0772ab70c1df3f570946fb6811908326252181f63ef80d40e
thread:issue:12509,"issue #12509: The text design (color, type) makes it hard to read",stale/low urgency,2,stale/low urgency,1,0,12509,,,,0,4e994a8219939f9aa582e4bf5014dd760b8d9f6f8832ffbb0150068810b1a403
thread:issue:125098,issue #125098: RuntimeError: MPS: Unsupported Border padding mode,stale/low urgency,2,stale/low urgency,1,0,125098,,,,0,bd3ca751ab5e4b1ac67ab714efc09a4b6a9d098ed1e799c0eac7b765b8481915
thread:issue:125116,issue #125116: Does torch.nn.Linear check the weights shape before assignment?,stale/low urgency,2,stale/low urgency,1,0,125116,,,,0,dee79b84222be38d74f499cce52e1e3dfb85bdf0ff737d155d4735b4bafe32f0
thread:issue:125134,issue #125134: binaries/dump_operator_names.cc missing iostream include,stale/low urgency,2,stale/low urgency,1,0,125134,,,,0,8e669dabaaab94e6529daf35d44d28f2acd23ace8aac871ecca7dea6fcb49bba
thread:issue:125146,issue #125146: Race condition in FileTimerServerTest.test_expired_timers,stale/low urgency,2,stale/low urgency,1,0,125146,,,,0,8d7979823138e93b47ed3272047e2355d217b0beb9e84cb13e70ed2a6bbc5727
thread:issue:12515,issue #12515: how to use mask-rcnn in caffe2 c++ gpu,stale/low urgency,2,stale/low urgency,1,0,12515,,,,0,118ac979a8776ed81aea9e5fdd0578aa68d470d0476847b93af8762be9083df4
thread:issue:125173,issue #125173: Flight Recorder Sequence IDs are insufficient,stale/low urgency,2,stale/low urgency,1,0,125173,,,,0,96e36a37e3a560a93eebcc9dfa8236711cba48a519e555edfaf9aa525c062707
thread:issue:125182,issue #125182: QR Decomposition for Sparse Matrix,stale/low urgency,2,stale/low urgency,1,0,125182,,,,0,7fd84fcbd5a4cd4de03800c1f4ad359d5b2dc47edda14e6aa47b11f251375284
thread:issue:125183,issue #125183: torch/_refs/__init__.py is not autoreloadable in bento / jupyter notebook,stale/low urgency,2,stale/low urgency,1,0,125183,,,,0,cba6a729f06567ba7cbfc926c54f49aaf83ce24f5de5f63afccdeef7c92ed704
thread:issue:125198,issue #125198: [CUDA][Compex] `test_reference_numerics_large_jiterator_unary_cuda_complex64` broken after updating to `numpy >= 1.25.0`,stale/low urgency,2,stale/low urgency,1,0,125198,,,,0,3e29dbf5e7c19b9450794941b717303deda137d44efe8c21f06c2f7462472d68
thread:issue:125217,issue #125217: MaxPool2D memory leakage on device MPS,stale/low urgency,2,stale/low urgency,1,0,125217,,,,0,7086c32ba59e35eedbe44a2908a464af07a05a46e2ac68c2d8c2a328f32d6d50
thread:issue:125234,issue #125234: torch.Library can easily cause segfault on loading/unloading,stale/low urgency,2,stale/low urgency,1,0,125234,,,,0,4040486daae5010df226d90b1feee6150757f166bf76d86fb063de78391583cd
thread:issue:125243,issue #125243: omp.h not found on macOS during install from source at github repo,stale/low urgency,2,stale/low urgency,1,0,125243,,,,0,e8a26800e2b079bf565743d1c9a2b2cda404c67807c97c9e6a256b9734f3cbdb
thread:issue:125297,issue #125297: Make the `sccache` cache easily available to all pytorch contributors in readonly mode,stale/low urgency,2,stale/low urgency,1,0,125297,,,,0,bb3db3fdda57cb7f869e03eb73684f87ec77f02b7aa2a6e0659fc70b77ac3bdc
thread:issue:125298,issue #125298: Tensorboard add_custom_scalar does not work,stale/low urgency,2,stale/low urgency,1,0,125298,,,,0,f94c3811b7afa660dd1595a829f58dfd2d51ea9aad259e1015c22c1e824cfd05
thread:issue:12530,issue #12530: [caffe2] Memory usage,stale/low urgency,2,stale/low urgency,1,0,12530,,,,0,23729e22f0f464422aa471694fb394380bbecfe20ec2c139c3dd11def10f8543
thread:issue:12532,issue #12532: Provide better documentation for torch.Size,stale/low urgency,2,stale/low urgency,1,0,12532,,,,0,2f20d7903b213461ed06e8282968cacdddc6eb603f0f161f12e0afd5db1d0fe7
thread:issue:12535,issue #12535: [CMake] Linking against Intel OpenMP,stale/low urgency,2,stale/low urgency,1,0,12535,,,,0,e4b5d010829c0bc1c871b594c3ad1cc41d58bb82889aa34b555d5a833bbcb53f
thread:issue:125358,"issue #125358: [DeviceMesh] Add support for `group: Tuple[ProcessGroup, ...]` in `from_group()`",stale/low urgency,2,stale/low urgency,1,0,125358,,,,0,7d2e5a4ca754c4b598047ac62caee9e35e310d34177294876d98c6efd31073b4
thread:issue:125389,issue #125389: Support for Kolmogorov-Arnold Networks (KANs),stale/low urgency,2,stale/low urgency,1,0,125389,,,,0,e4b9ca3eeea3e5613f3f6153f638f8b1d3a1ffcb036d0f12214c557845fdd2df
thread:issue:125390,issue #125390: [typing] Make arguments to `__getitem__`/`__setitem__`/ etc. positional only.,stale/low urgency,2,stale/low urgency,1,0,125390,,,,0,7be570decede22602c107da949aa4129d193de92b95703c09dc048746a91f6ba
thread:issue:125467,issue #125467: Inaccurate filename due to test class metaprogramming that doesn't set __file__ and __name__,stale/low urgency,2,stale/low urgency,1,0,125467,,,,0,97afcf8db13e33e38a7a87db3efad50ab84713cd272ff362439917cc470fb21e
thread:issue:125471,issue #125471: Build failure when using CUDAGraph,stale/low urgency,2,stale/low urgency,1,0,125471,,,,0,5f6b599f8273dfd218adecde499871685fbb9892c1f3930b2526cd3177b45cb7
thread:issue:125493,issue #125493: Error calling to_sparse_coo() on any subclass of Tensor,stale/low urgency,2,stale/low urgency,1,0,125493,,,,0,ea2012298cb96bb9a30dea7a8dd3969daf4cbd626055cba558b4759c055abcd7
thread:issue:125502,issue #125502: C++ libtorch x86_64 Darwin deprecation note missing,stale/low urgency,2,stale/low urgency,1,0,125502,,,,0,a7c21eca32e25f9d78faff0a25d65e93f1f106efcb9f89ddb732d1ddc95498c6
thread:issue:125539,issue #125539: `triton` package that ships with torch is not compatible with `torch.ops.matmul`,stale/low urgency,2,stale/low urgency,1,0,125539,,,,0,576528c7e9e6a2fd7a1feff8679649d70c5954b5de96bb0fe4450be5a4e4ad82
thread:issue:125547,issue #125547: PyTorch model slows down 30x when binding threads with fewer than half the CPU cores.,stale/low urgency,2,stale/low urgency,1,0,125547,,,,0,9cd861b7076be4c9e962ea21c87367366f1989b42ffd4a3417adc1c8b1583465
thread:issue:125581,issue #125581: fx hook failed,stale/low urgency,2,stale/low urgency,1,0,125581,,,,0,fe1324d1c064a001d3c222b6d31a0a7c95b7e71bade15d4f331259947bcc2733
thread:issue:125591,issue #125591: PyTorch gradient checkpointing consumes a lot more memory than JAX,stale/low urgency,2,stale/low urgency,1,0,125591,,,,0,0195247c806ebe5595fb070affa0b634f8664f6cb0a61d9786689703c6357239
thread:issue:125620,issue #125620: [BE] DTensor optimizer to utilize OptimizerInfo Constructs,stale/low urgency,2,stale/low urgency,1,0,125620,,,,0,b21301637441758c10085e637bbbc57cd5961b6485ef165d480a54d1362a53e4
thread:issue:125622,issue #125622: Wrapper subclass serialization should preserve all _make_wrapper_subclass() args,stale/low urgency,2,stale/low urgency,1,0,125622,,,,0,071aa2e5d37a60fef7a7094438d480dafd97e9317edc23135fe1268d211fd8f2
thread:issue:125672,issue #125672: [inductor][cpu]mobilenet_v2_quantized_qat float32 single thread static/dynamic shape CPP/default wrapper performance regression in 2024-04-28 nightly release,stale/low urgency,2,stale/low urgency,1,0,125672,,,,0,5fb358dafde8e9bca55dcbcffefafaf67ed22fe02e02af22c4dbbcd114b3cedc
thread:issue:125683,issue #125683: [RFC] Add Cpp Template for GEMM related ops via max-autotune for Inductor CPU,stale/low urgency,2,stale/low urgency,1,0,125683,,,,0,4537116ed4fbfe3dc8acd203917058b91af365ff9a004b550e59e48f32ad83f3
thread:issue:125718,issue #125718: `torch.compile` and complex numbers,stale/low urgency,2,stale/low urgency,1,0,125718,,,,0,28a0899614f49a52edb12822cf1d821844833d0be54b091ba767f3f217a97b25
thread:issue:125738,issue #125738: [FSDP2] _sharded_param_data is sitll on meta while sharded_param moved to cuda after calling initialize_parameters(),stale/low urgency,2,stale/low urgency,1,0,125738,,,,0,f8f30985e80679f292e555f9e265cb8e0c356fb790d932efcbaf42496eaabe54
thread:issue:125744,issue #125744: Improve oneDNN memory alloction performance for pytorch Windows,stale/low urgency,2,stale/low urgency,1,0,125744,,,,0,90e2f421be8cb5db923db8dbb09d83c24075cbd89bffd7159c4ef3b9327c6d5f
thread:issue:125746,issue #125746: DefaultCPUAllocator: not enough memory in matrix multiplication broadcasting,stale/low urgency,2,stale/low urgency,1,0,125746,,,,0,a7854f0a5f0c749b6662eeee6871b548c545a6505431028934c68970e8c1e667
thread:issue:12576,issue #12576: [feature request] `ignore_label` argument in Caffe2 `SoftmaxWithLoss`,stale/low urgency,2,stale/low urgency,1,0,12576,,,,0,8147040a29ceb62ce9d11d419e092999b77f19d88f27b81bc68c48d763f18c0f
thread:issue:125800,"issue #125800: [BE][DTensor] get rid of @with_comms decorator for tests, and use run_test to init pg",stale/low urgency,2,stale/low urgency,1,0,125800,,,,0,3bb1ce4bd3ec6ef9a1a5847926ea61c494caa7504df534422e70a882236b1d8d
thread:issue:125835,"issue #125835: When using torch.onnx.export, the result of torch.scatter_add may be wrong for opset11",stale/low urgency,2,stale/low urgency,1,0,125835,,,,0,3b8a605405fcecd2dc7459eb24b016106995ecbd739b5cd203868a8fe5760864
thread:issue:125844,issue #125844: `WeightedRandomSampler` convert weights `dtype=torch.double` internally but `MPS` only support `dtype=torch.float32`,stale/low urgency,2,stale/low urgency,1,0,125844,,,,0,cf5dc0ab5ab9ced065e057414dc54c65153e80f7157a6990d7fda1645bd77260
thread:issue:125862,issue #125862: Pytorch nightly docker image invalidated layers,stale/low urgency,2,stale/low urgency,1,0,125862,,,,0,46f9b3a81cda29d329af0cf6ea2c46b7369b89df11e18dca88f1ead2a8890944
thread:issue:125893,issue #125893: dynamo_export fails to produce a graph from nn.GLU : no tensor operations were found,stale/low urgency,2,stale/low urgency,1,0,125893,,,,0,ecf0141ad9eea641102ae2c5c83c24763c0e089b5f10c4f8f3672540e3a88fb4
thread:issue:125895,issue #125895: Support for third-party use of pipeline,stale/low urgency,2,stale/low urgency,1,0,125895,,,,0,c23b35d3b0d995a78899a14ccd637ed0534ebb3dda6a002eea4995de2ae2617b
thread:issue:125902,issue #125902: How to export onnx with fixed shape output ?,stale/low urgency,2,stale/low urgency,1,0,125902,,,,0,d81f0c9771f9842fe3b591f0e7fc2f004fde15777593f4ce9e93fd7c5f8b1e92
thread:issue:125940,issue #125940: [NestedTensor] multiply batch and ragged dimension to get shape of values tensor,stale/low urgency,2,stale/low urgency,1,0,125940,,,,0,021c95534b967408fb21a6bca30229d074ec3612626649be9be98ce90385afa3
thread:issue:125984,issue #125984: `torch.export` code that includes `torch.autograd.grad` RuntimeError: element 0 of tensors does not require grad and does not have a grad_fn,stale/low urgency,2,stale/low urgency,1,0,125984,,,,0,2203ba77968e50b4c3b08d509ecfd86a80004f93d677ef6a835cc5cea2444ea3
thread:issue:126000,issue #126000: Inform when weights loading promotes types,stale/low urgency,2,stale/low urgency,1,0,126000,,,,0,9b601d3c18529df69684697f3a2be8cc3650c3d82c92f0fb5fefdc379c301895
thread:issue:126005,issue #126005: CUDNN not detected in official nightly devel image. We need to use `12.4.1` upstream images,stale/low urgency,2,stale/low urgency,1,0,126005,,,,0,33aa9c789775d2da7e9ad884640cc4eb0aebd60f7b4f707a720153453e8e7732
thread:issue:126006,issue #126006: The arguments `tesnor1` and `tensor2` of `result_type()` don't work but `tensor` and `other` work,stale/low urgency,2,stale/low urgency,1,0,126006,,,,0,303ce32b1a750a09d2634bc4cf697cbde999bcee3516a7ed491dae83fc4e3120
thread:issue:126008,issue #126008: Torch distribution blocking ports and failing to connect,stale/low urgency,2,stale/low urgency,1,0,126008,,,,0,27b29b79b488e2ced1d860d20a2b003cfcf64495f9405201f9bc1803106f9b69
thread:issue:126014,issue #126014: The unexpected behavior of `can_cast()`,stale/low urgency,2,stale/low urgency,1,0,126014,,,,0,c27dcbd1359131874b30a8407e9daf8e5196ccd6ae1eedb1abc000942cdbd0b4
thread:issue:126021,issue #126021: Incorrect output from `torch.histc` when min and max are equal.,stale/low urgency,2,stale/low urgency,1,0,126021,,,,0,0194cbda2cbcc95ce5d4e408c67bf88156bed46eb0b2dcf17011938723dcdbec
thread:issue:126032,issue #126032: all_gather with gloo backend does not work in inference mode,stale/low urgency,2,stale/low urgency,1,0,126032,,,,0,1997ae7a704e5476fb5d7524297919923d36e348898d992516c4934196cab072
thread:issue:126034,issue #126034: A scalar with `tensor` argument of `result_type()` doesn't work,stale/low urgency,2,stale/low urgency,1,0,126034,,,,0,b884be017154f663608c5531fd17927043208f5aa5919970cc1c2972377c5f97
thread:issue:126041,issue #126041: A scalar with `input` argument of `remainder()` doesn't work,stale/low urgency,2,stale/low urgency,1,0,126041,,,,0,a67d6df340da9e9f8a863d62e94d971034b6c21609a719b24b4ff4da849e9d1c
thread:issue:126045,issue #126045: A scalar with `elements` or `test_elements` argument of `isin()` doesn't work,stale/low urgency,2,stale/low urgency,1,0,126045,,,,0,e0591bac905796db556d246b20d93594b5b3c0ac10ddb26b2402078a914c6a39
thread:issue:126047,issue #126047: `torch_save_to_dcp`'s produced dcp checkpoint doesn't fit the load interface for FSDP,stale/low urgency,2,stale/low urgency,1,0,126047,,,,0,9c6448b5df914d609b50301379efefeb516c72a970418d999b7ca5e6e1b01721
thread:issue:126065,issue #126065: [ONNX] `SymbolicValueError` with `onnx::quantized_gather`,stale/low urgency,2,stale/low urgency,1,0,126065,,,,0,1a09f20e3e6e9389d4a799092232043aee13a7cfb396b7d21e923d6e630bc67c
thread:issue:12609,issue #12609: Request for stripped down / inference only pytorch wheels,stale/low urgency,2,stale/low urgency,1,0,12609,,,,0,516ceb85d32117e3178d102629f6f08a5b76659894344f903c56f9645cbdf3cd
thread:issue:126115,issue #126115: filtering on CPU slower in pytorch than scipy,stale/low urgency,2,stale/low urgency,1,0,126115,,,,0,79b359eada10d69620e545053f92fb4898532cfbc20776fca2da1a58849e2fe3
thread:issue:126154,issue #126154: Optimizer classes not `dill` picklable after using `torch.compile`,stale/low urgency,2,stale/low urgency,1,0,126154,,,,0,e27eceebddda5221a9794986a9063ab221693f38bbf315dd2c6f73fd417c75f4
thread:issue:126190,issue #126190: Centralized Mechanism To Pytorch Query Global Flags,stale/low urgency,2,stale/low urgency,1,0,126190,,,,0,7be4e8590ec1d183021c10d49d6dd260fe3e5edbe128b5e8a8a18cda3940ee92
thread:issue:126211,issue #126211: Conflicting OMP implementations linked when building with MKLDNN,stale/low urgency,2,stale/low urgency,1,0,126211,,,,0,69f50b55862d72280bdb0c33f4f871b13e1416aad38e425ec3e606c5d5b9cb70
thread:issue:126216,issue #126216: Allow Lazy Initialization for PipelineStage,stale/low urgency,2,stale/low urgency,1,0,126216,,,,0,53c34f3eed20e19805e5176e51e2635c7af12703f4d2c4d77345ffef14349a87
thread:issue:126244,"issue #126244: `dtype` of `torch.tensor()` also accepts `int()`, `float()` and `bool()` which are python built-in functions.",stale/low urgency,2,stale/low urgency,1,0,126244,,,,0,59aca0e2d8fa490ff015016c3200f96646642675e9a23dcf38447c61119f7661
thread:issue:126247,"issue #126247: `type1` and `type2` of `torch.promote_types()` also accept `int()`, `float()` and `bool()` which are python built-in functions.",stale/low urgency,2,stale/low urgency,1,0,126247,,,,0,1cb8245560d54cac4ee07cda7a00505dc4250a546fb8fe485fa6f9f850b10ad7
thread:issue:126250,"issue #126250: `dtype` of `to()` also accepts `int()`, `float()` and `bool()` which are python built-in functions.",stale/low urgency,2,stale/low urgency,1,0,126250,,,,0,c0650e7986ffa02bd45740911f2d963147538dfb4edb7d75a0f0feb5b8af0fa2
thread:issue:126253,issue #126253: `dtype` of `type()` doesn't accept a string,stale/low urgency,2,stale/low urgency,1,0,126253,,,,0,35b2435f26aa2362bd9e3f770f2efa10c0d727d3cd54957c73c96a16cf7bb793
thread:issue:126286,issue #126286: Adopt inductor-perf-test-nightly.yml to be used for release testing,stale/low urgency,2,stale/low urgency,1,0,126286,,,,0,f5b358f1acf40d2cdc3f37a77b831c3b6ca9b5a356d473dbf7942a179df60afe
thread:issue:126323,issue #126323: Move pytorch_sphinx_theme out of the container build,stale/low urgency,2,stale/low urgency,1,0,126323,,,,0,71523006125235f3e2249f13ef6657234977eb47fab738bbc543fec6d15c8872
thread:issue:126344,issue #126344: PyTorch: RuntimeError: device >= 0 && device < num_gpus INTERNAL ASSERT FAILED,stale/low urgency,2,stale/low urgency,1,0,126344,,,,0,77b094774fd8aa336de471a991e78a80d6fbd54030834e244056ce486933acdb
thread:issue:126347,issue #126347: Break the dependency between torch.nn and torch.distributed,stale/low urgency,2,stale/low urgency,1,0,126347,,,,0,277b47462d312bfe10663fe4eaf5b2375a656316a7fd4b18acefa83662eb2eba
thread:issue:126360,issue #126360: Dynamo ignores frame using yield,stale/low urgency,2,stale/low urgency,1,0,126360,,,,0,80e09dc9a493b73dfdccabbef362f46e5f14b277101225a62381626c6a6c3c65
thread:issue:126381,issue #126381: All processes running torch.distributed.destroy_process_group() create CUDA context on device 0,stale/low urgency,2,stale/low urgency,1,0,126381,,,,0,703429b50fd7d41b5c72c77b8051755086f2abdc3332c2a27e44be1d311ac159
thread:issue:126386,issue #126386: lintrunner failures on main,stale/low urgency,2,stale/low urgency,1,0,126386,,,,0,08a8078dbab638c9441ab0907dfdd48020fed3d20ba0c9f5d27c61073d07b4b3
thread:issue:126394,issue #126394: Building libtorch for android fails,stale/low urgency,2,stale/low urgency,1,0,126394,,,,0,d56bd8512e2d0dd38b0a3fb9b5ce41816b24a3ab78862fc365c4c87c45cc4453
thread:issue:12640,issue #12640: how to store a bounding box in Tensor?,stale/low urgency,2,stale/low urgency,1,0,12640,,,,0,e894f1084731cd4fd91668e3b70d50cc34591719d9eada1eadc8249f5198220b
thread:issue:126404,issue #126404: Add `int` and `str` type to `device` parameter of `torch.tensor()`,stale/low urgency,2,stale/low urgency,1,0,126404,,,,0,4fdfbcd3eb469bea80844efa5c1230cc365ffa76f885bf807b682266c6aa4dec
thread:issue:12642,issue #12642: fail to visualize caffe2 model,stale/low urgency,2,stale/low urgency,1,0,12642,,,,0,e33b2a5e3bff7982ca7681ced224e8d6e004bae1f0dadef95b850e2c5d4703bf
thread:issue:126453,issue #126453: Add testing for NJT serialization,stale/low urgency,2,stale/low urgency,1,0,126453,,,,0,6b92ceaf899160682a4bc8e182442392750864b8b0a8c95413dcedd05fbc22fd
thread:issue:12646,issue #12646: Caffe2 Installation inside Pytorch,stale/low urgency,2,stale/low urgency,1,0,12646,,,,0,6425fb9900b40b8d01cd2f9576b9462673174e5b71fb67fbfd417fa9dbb54d5c
thread:issue:126518,issue #126518: missing docs for torch.Tag,stale/low urgency,2,stale/low urgency,1,0,126518,,,,0,f53e1b335fbfb6c9a3ee70d7e7ee3844c9c0f542a207c1d5b6c1a37634cffff5
thread:issue:126519,issue #126519: ambiguous overload for TORCH_LIBRARY `def`,stale/low urgency,2,stale/low urgency,1,0,126519,,,,0,7ba0f0791f4b38a65dada9f29f03d14323d726585e2f5fa4e2c80699530f36f9
thread:issue:126544,issue #126544: ncclCommWatchdog always terminates the process and prevents error handling if CUDA context is corrupted,stale/low urgency,2,stale/low urgency,1,0,126544,,,,0,b10369673bfde7ba0e556000d9da7a8187475abd8c0286c3c7655fef0b529c07
thread:issue:126549,issue #126549: Error: Exporting the operator 'aten::searchsorted' to ONNX opset version 17,stale/low urgency,2,stale/low urgency,1,0,126549,,,,0,0d0d35beac68e64753912c3592b13de39ab551eb9da101863aabe321781ffcb1
thread:issue:126566,issue #126566: `torch.compiler.allow_in_graph` does not create a `call_module` op in fx.Graph in torch 2.3.0,stale/low urgency,2,stale/low urgency,1,0,126566,,,,0,ea1ce1b1584abd8654bf7fe5f5ab8e36404304d1753979ec728671c2525b738a
thread:issue:12658,issue #12658: Move BigTensorSerialization tests out of default caffe2_cpu_tests,stale/low urgency,2,stale/low urgency,1,0,12658,,,,0,9d4c3ee30204f2f5c4530ac44e0d030404eda91fc84e9b80e8e3226d87641d38
thread:issue:126620,issue #126620: dataloader crashes after several epochs if the trained model contains triton-based operators,stale/low urgency,2,stale/low urgency,1,0,126620,,,,0,0507b54b1000d87d9952271e267fe668a27b31d737c38925f859cca2f0a15bdc
thread:issue:126638,issue #126638: Test failures with new configs in test_grad_scaling_autocast in test_torch.py,stale/low urgency,2,stale/low urgency,1,0,126638,,,,0,5960b54b0011f0681036fc908c771c65e4882ec58476be4c42d0a082dd2b8824
thread:issue:126644,issue #126644: [torch.compile]: Enhanced Error Reporting and Performance Canary Mode,stale/low urgency,2,stale/low urgency,1,0,126644,,,,0,b37f1634468ea21c8e0537937a10e0a618328d543cd634d1f9e24a896d6b3066
thread:issue:126658,issue #126658: `kthvalue()` cannot be used with `complex` or `bool` type of a 1D or more D tensor but `kthvalue()` can be used with `complex` or `bool` type of a 0D tensor,stale/low urgency,2,stale/low urgency,1,0,126658,,,,0,81e9e570bfe286f50411b1828594c23fc30d88367a104949a1e2b4aacebda957
thread:issue:126659,issue #126659: `topk()` cannot be used with `complex` or `bool` type of a 0D(only `cuda`) or more D tensor but `topk()` can be used with `complex` or `bool` type of a 0D(only `cpu`) tensor,stale/low urgency,2,stale/low urgency,1,0,126659,,,,0,6138c2bfddd132b0b578ccc74d1e077cb121b7cdc6de3e0492f7071adb0107b8
thread:issue:126660,issue #126660: UnsupportedOperatorError: aten::scatter_reduce when include_self=False,stale/low urgency,2,stale/low urgency,1,0,126660,,,,0,84b2ce2a2560360afa718a7c2787b6593cc806091f2d4705419a19a28808f51e
thread:issue:126691,issue #126691: TORCH_CHECK used within torch.compile does not throw legible errors,stale/low urgency,2,stale/low urgency,1,0,126691,,,,0,fe123401594a178baf438ea5251d7fe88f2887d7039ebed8b1049b6bc0ce68d2
thread:issue:126708,issue #126708: Inductor Channels-last slowdown,stale/low urgency,2,stale/low urgency,1,0,126708,,,,0,a8ae203f360b208a15365c15743f9c2131d8834f6b69fca28bb7d469f634a589
thread:issue:126738,issue #126738: FakeTensor `forward()` on `bert` attention when using FakeTensor returns a DataDependentOutputException,stale/low urgency,2,stale/low urgency,1,0,126738,,,,0,662bf1fd22acfa02d51a70e1acae4b15c7b69fd248d1eb1b0baae8cfae9a29e6
thread:issue:126748,"issue #126748: ""torch.geqrf"" performs differently on cpu and gpu",stale/low urgency,2,stale/low urgency,1,0,126748,,,,0,231d6cbb39ccc8c37906aaf80e817f47fe50431165783c96bff9accabf1a35fd
thread:issue:12675,issue #12675: [feature request] ignore_index and size_average in nn.AdaptiveLogSoftMaxWithLoss,stale/low urgency,2,stale/low urgency,1,0,12675,,,,0,ed5e3184e871b64f19fd75daecba43d1b79c1d8a406a2f254f89d91a6fae4d88
thread:issue:126753,issue #126753: [cudagraph] simplify usage of how cudagraph dumps debug file,stale/low urgency,2,stale/low urgency,1,0,126753,,,,0,0899f5526480d1fac193a39a5653902a04b57ecb42ebec1a92eb07b3b0af379c
thread:issue:126755,issue #126755: inplace parameter in dropouts should function as expected regardless of the value of training(or train) paramter,stale/low urgency,2,stale/low urgency,1,0,126755,,,,0,5ce55689bf3fde9dc7574ccb1134d01fedeb7d1ba8506944584da3bf8e653f31
thread:issue:126758,issue #126758: The unexpected behavior of `min()`,stale/low urgency,2,stale/low urgency,1,0,126758,,,,0,f1b77c55e8d3bbef374c9cc45bac56d0330bd74da6c530f99f06cd7bbd2d73cb
thread:issue:126773,issue #126773: Give`wait_tensor()` a schema that reflects its side effect,stale/low urgency,2,stale/low urgency,1,0,126773,,,,0,6bcfe8b150a6ba336d8cfd863eab999b84e7e1ece5439a13ebcdced5f8b62224
thread:issue:126774,issue #126774: [ONNX] Implement aten._trilinear and aten.bilinear.default,stale/low urgency,2,stale/low urgency,1,0,126774,,,,0,b9e69a20198859ffb7a4ad9d743f2c72a20a8d6363de4981711933837f7f0007
thread:issue:126776,issue #126776: One complex number or boolean value of a 1D or more D tensor with `argmin()` and `dim=` works,stale/low urgency,2,stale/low urgency,1,0,126776,,,,0,a087dc2e78dca33e437a2161c701566de10bbb92ee9a13842ad19e45ac61b244
thread:issue:126777,issue #126777: One complex number or boolean value of a 1D or more D tensor with `argmax()` and `dim=` works,stale/low urgency,2,stale/low urgency,1,0,126777,,,,0,0124efba8dba0ffa2f0590d2b701c949080e8465cabcabef8aa402e4b955655d
thread:issue:126795,issue #126795: Using DTensor with device meshes that use different devices for input and output,stale/low urgency,2,stale/low urgency,1,0,126795,,,,0,9bf9cfaf8dbc5b080139b8e6c1d4f67147f93b16b6df64c49f187fc94ec9ee23
thread:issue:126846,issue #126846: propagate_real_tensors can cause deadlocks for programs with collectives,stale/low urgency,2,stale/low urgency,1,0,126846,,,,0,f9fe1f030642f96f3bdf8e2a43048a024f567ab8cb9590512b19068b84765aaf
thread:issue:126856,issue #126856: TorchDynamo ONNX Export does not work as expected with masking (ScatterElements),stale/low urgency,2,stale/low urgency,1,0,126856,,,,0,8df154f626df473565e754a1781c11b26cf5a7c2fe333628a34c6ae9e23e9715
thread:issue:126881,issue #126881: [DCP] DCP does not support objects which are lazy initialized.,stale/low urgency,2,stale/low urgency,1,0,126881,,,,0,4140f12241d2e287e8d806302d8e0b65664bb286957ce723d40d25489ddda45a
thread:issue:126899,issue #126899: function signature for multiprocessing.spawn is multiprocessing.spawn.spawn,stale/low urgency,2,stale/low urgency,1,0,126899,,,,0,24c4469d469feb087fc1501a3e3ba37d82ef2c8e068b5e1bc46de5f1c0a9cd20
thread:issue:12690,issue #12690: How can I build caffe2_gtest_main under pytorch/caffe2/test/ folder?,stale/low urgency,2,stale/low urgency,1,0,12690,,,,0,dacc3bd6e9434a75609556a814816aa469462b73519b694f65f90470c706d254
thread:issue:126913,issue #126913: Alternative access order for the same buffer can bring big perf win,stale/low urgency,2,stale/low urgency,1,0,126913,,,,0,3a2574d67558e24e860cfd4b5ffed679c6a5abcdb599b6eca8be1b9ac0335148
thread:issue:126947,issue #126947: `dsplit()` with `indices_or_sections=` doesn't work while `dsplit()` without `indices_or_sections=` works,stale/low urgency,2,stale/low urgency,1,0,126947,,,,0,fca852403437c1e4f474fff4e7bdb813865ae85a1e3e0a13f4e627f2ea6c919e
thread:issue:126952,issue #126952: Add integrity check in torch.save,stale/low urgency,2,stale/low urgency,1,0,126952,,,,0,d967407e1ec73c6dbcbd171b8c738daeba665e29219a222e55795bf0d6243512
thread:issue:126972,issue #126972: ONNX Exporter Fails with Handling Complex Tensors,stale/low urgency,2,stale/low urgency,1,0,126972,,,,0,2bc094cca1335f901fe32118ad8e74925e3dae44ea5bccf1694b59fe44ac81d6
thread:issue:126973,issue #126973: Performance Degradation in F.linear with Batch Size > 1 in Multi-Head Attention,stale/low urgency,2,stale/low urgency,1,0,126973,,,,0,e1a40dfa8519a28ae7dcdb4966d94a386f3ecaa9a88ee6ace884cacf2a890098
thread:issue:126980,issue #126980: Long build compilation time (>5 hours) for CUDA ARM build,stale/low urgency,2,stale/low urgency,1,0,126980,,,,0,8137e09b3d043281bf396d785e4ebea091e6b85ac6d6a4202939166d3bbf2491
thread:issue:127001,issue #127001: Add ability to pass `MAP_NORESERVE` to`mmap(2)` call in `torch.serialization.set_default_mmap_options`,stale/low urgency,2,stale/low urgency,1,0,127001,,,,0,3ff5fc1ae98509e038aa2a2c5ba89fa0d5df3d29f776eb2022bbbe40c773d353
thread:issue:127009,issue #127009: `indices_or_sections` parameter of `dsplit()` doesn't work while `sections` or `indices` parameter works,stale/low urgency,2,stale/low urgency,1,0,127009,,,,0,78b7d95706d9cbeb939843e4a7d21a7c54ee4fb1127a4a24c9dbbe62715600c6
thread:issue:127010,"issue #127010: `indices_or_sections` parameter of `tensor_split()` doesn't work while `sections`, `indices` or `tensor_indices_or_sections` parameter works",stale/low urgency,2,stale/low urgency,1,0,127010,,,,0,db654ba67ce04167bbe4248745672be3232ba24b59a82cd3ef47c9bf9789cd0d
thread:issue:127019,issue #127019: `indices_or_sections` parameter of `vsplit()` doesn't work while `sections` or `indices` parameter works,stale/low urgency,2,stale/low urgency,1,0,127019,,,,0,bafaa3b0f11e38b912a9fb52a158ce57cccefa48aca56d3b17ea2eb614d8fd38
thread:issue:127020,issue #127020: `indices_or_sections` parameter of `hsplit()` doesn't work while `sections` or `indices` parameter works,stale/low urgency,2,stale/low urgency,1,0,127020,,,,0,581152dc1ddc351f6f479ca2e25af5a1e3893cdd8b3cead7b822f76b41610e66
thread:issue:127031,issue #127031: Make grid_sample a bit more intuitive to use,stale/low urgency,2,stale/low urgency,1,0,127031,,,,0,e60dc561c0f9b58e8c23965db2dc02134d08adb0abc83add3cb55d2e8fa1d318
thread:issue:127072,issue #127072: PyTorch DataLoader improvements for Iterable Dataset,stale/low urgency,2,stale/low urgency,1,0,127072,,,,0,b4080c63b7a310161705109b8e1e1a1983cda00541da8c75d230226518349fef
thread:issue:127083,"issue #127083: Setting a `float`, `complex` or `bool` type value to `precision` argument of `set_printoptions()` has problem when creating a `float`, `complex` or `bool` type tensor",stale/low urgency,2,stale/low urgency,1,0,127083,,,,0,205c23e0f585bcb0d9a32af04a9be94fa78f74606c8ab507ccf0fa4ecfb436f0
thread:issue:12715,issue #12715: pack_padded_sequence throws IndexError when only kwargs are specified,stale/low urgency,2,stale/low urgency,1,0,12715,,,,0,9bd793e94acbeb3c3f0b8262690cdc8c9de83734dc3a29e9ef41df04b04870b3
thread:issue:12716,issue #12716: cdf in torch.distributions.bernoulli throws NotImplementedError,stale/low urgency,2,stale/low urgency,1,0,12716,,,,0,21057382f3f30f749bc98c885de142a45c3af0bb86e6fdc210e39a71778a17cf
thread:issue:127189,issue #127189: `store_param_remainders` from Apex DistributedFusedAdam,stale/low urgency,2,stale/low urgency,1,0,127189,,,,0,9676d8dbcd5fc2aecaf3ac660ce6977e25a5820f015c36add79d10b80930c6ed
thread:issue:127194,issue #127194: Should `torch.Size` convert np.ndarrays to lists of ints?,stale/low urgency,2,stale/low urgency,1,0,127194,,,,0,d96618a6742390842635ff649e7295d5ac3b88e71cf8e234c14200736e68cd52
thread:issue:127196,issue #127196: torch.topk results differ on CPU and CUDA,stale/low urgency,2,stale/low urgency,1,0,127196,,,,0,71b35022e0969b5122b9e89039bcf6ce0a18cacf73cc1aed3f6d4234df8cd8ff
thread:issue:127221,issue #127221: lacking checking for ConvTranspose's parameters when running with GPUs,stale/low urgency,2,stale/low urgency,1,0,127221,,,,0,29187d9de903d0f99868df8575b6495ad721e9b8557e37a189dfdcc8f633ed17
thread:issue:127223,issue #127223: [BUG] torch.linalg.lstsq returning wrong result.,stale/low urgency,2,stale/low urgency,1,0,127223,,,,0,afebdf0bc0c469298a8674e173bf92a3812e5b8d38fdd90a345b796575e158d2
thread:issue:127225,issue #127225: make_graphed_callables don't work with FSDP at all even on a simple network,stale/low urgency,2,stale/low urgency,1,0,127225,,,,0,39e31ac5b0a421234dc27aa96213ea65f4f35b7fbf3a47f8a5e8cb109475b4ec
thread:issue:127270,issue #127270: Not supporting higher protobuf versions,stale/low urgency,2,stale/low urgency,1,0,127270,,,,0,2fe4a84518759982f6413036ee5981604b113edd0bce1c3c3b193bf5ab6a8534
thread:issue:127271,issue #127271: ImportError: cannot import name 'triton_key' from 'triton.compiler.compiler' when using triton 2.3.x,stale/low urgency,2,stale/low urgency,1,0,127271,,,,0,e155684c582008b68feb06eae7b3aa671f985cd46f276846af09bf169e51efd8
thread:issue:127282,issue #127282: torch.fx.symbolic_trace doesn't support many Callable types,stale/low urgency,2,stale/low urgency,1,0,127282,,,,0,a23627f27d366dba5c60d855b3c63bde69a87c012eaf9b3f16d6c1ccdee1f8d2
thread:issue:127337,issue #127337: Add option for custom ops to automatically get a FakeTensor kernel (during static shapes),stale/low urgency,2,stale/low urgency,1,0,127337,,,,0,09f647e8ce343891f30ddc763b521f809fc4585e6387a9450eee561dd9dc5a2c
thread:issue:127352,issue #127352: `flake8: noqa` disables flake8 linter for the whole file and it's not obvious,stale/low urgency,2,stale/low urgency,1,0,127352,,,,0,652b6c8b7c02c4b0e419fc17bf7386e8caf8e2e10399a602482aa42f9823c647
thread:issue:127375,issue #127375: Different tensor strides can result in surprisingly large discrepancies in Conv2d outputs,stale/low urgency,2,stale/low urgency,1,0,127375,,,,0,112ed2108c4861884a1604fa535325513e837bbe4ef648e99cbfb7adb9240f5e
thread:issue:127387,"issue #127387: Backwards pass through Beta distribution rsample gives inf for 4 < alpha - 2**16 < 1040, beta = 3/2",stale/low urgency,2,stale/low urgency,1,0,127387,,,,0,6c19415016244c228cdbb5bff68ffea2caa24fafb5832232c9cf7cc912163b71
thread:issue:127388,"issue #127388: torch=2.1, import torch; fails AttributeError: '_OpNamespace' 'aten' object has no attribute 'sym_constrain_range_for_size'",stale/low urgency,2,stale/low urgency,1,0,127388,,,,0,fca77ae2c10f64a23867f4b994fde7256c115d8f1aa6e02deeb647b7e812874f
thread:issue:127391,issue #127391: Unable to quantize resnet50 model with post training static quantization,stale/low urgency,2,stale/low urgency,1,0,127391,,,,0,d425ae847bcb58d82111a9520768763137af77e7bd1aba5da893394d62661134
thread:issue:127422,issue #127422: index_select op not implemented on Vulkan backend,stale/low urgency,2,stale/low urgency,1,0,127422,,,,0,97d6fd14a1c8b87372558ffe7708ca7731878a1ca271db278bcc42402b386f07
thread:issue:127497,issue #127497: torch.compiler docstrings can be derived from _dynamo and _inductor docs,stale/low urgency,2,stale/low urgency,1,0,127497,,,,0,5ec26d2f6d0b6f6c37ba44bdd171392cd2dd28a36a73df28cda2c96a126c33cf
thread:issue:127500,issue #127500: Avoid Having to Register Op For ExternKernelChoice of Aten Refs,stale/low urgency,2,stale/low urgency,1,0,127500,,,,0,c61f368a297bd8bbb894d44b2b87e1a70a530088be27b50f6cdac09ea32f0f7f
thread:issue:127517,"issue #127517: With pytest8.2.0 or later, test cases under test/distributed/ execute will meet issue ""object has no attribute 'runTest'. Did you mean: 'run_test'""",stale/low urgency,2,stale/low urgency,1,0,127517,,,,0,ebe9217a2f16ca1fd2c38c5fd7feb4a1dafec0ff10731c4a0cb64168c6058534
thread:issue:127518,issue #127518: THPVariable_Check(list_elem) INTERNAL ASSERT FAILED,stale/low urgency,2,stale/low urgency,1,0,127518,,,,0,e6a2e489e26b9c7698a0e6fe02382acc08df011a2b24e11a583c8862a1b2f7f8
thread:issue:127525,issue #127525: Multiple unhandled exceptions in weights_only unpickler,stale/low urgency,2,stale/low urgency,1,0,127525,,,,0,f16d042c59429b62089b0fa27e77c82e8076bac464207b3ac5b62046d569f10a
thread:issue:127530,issue #127530: Run rstcheck on modified docstrings and docs as additional linter,stale/low urgency,2,stale/low urgency,1,0,127530,,,,0,a140fc0d3502e8ff141393088fe08b5b9e489cc357f77cc1607e60164071fe11
thread:issue:127535,issue #127535: Can we have a Y-Split module for torch.nn.Sequential?,stale/low urgency,2,stale/low urgency,1,0,127535,,,,0,b701d1a3a89de05bfabcb4eaa43ac32a777d1d9668f1668667a3b8059054ec5a
thread:issue:127541,issue #127541: RuntimeError: derivative for aten::_spdiags is not implemented,stale/low urgency,2,stale/low urgency,1,0,127541,,,,0,370c7e5423f0e7c3a99623e294afcbaa2e4cfee2744ca532ccafb5790b1dd8ea
thread:issue:12755,issue #12755: caffe2 python custom operator could not update parameters,stale/low urgency,2,stale/low urgency,1,0,12755,,,,0,2d7dce1ff467daa593c92998be66d8814d59f5712e07c62e2b9b23b3dcddec40
thread:issue:127555,"issue #127555: Reserving this issue to test something, don't mind me",stale/low urgency,2,stale/low urgency,1,0,127555,,,,0,4802823cb4a26d2273dcdab2a98380a1b3bbd0f98620c64088d5a42e2370a50c
thread:issue:12760,issue #12760: [feature request] Spectral norm support in torch.norm or factor out Power Iteration from spectralnormalization in some other place and orthogonalization from PowerSGD hook,stale/low urgency,2,stale/low urgency,1,0,12760,,,,0,6aed3351302d790147e9362fe74d9a79a2bedf03779379f12445b759b3868aab
thread:issue:127606,issue #127606: ` to be a quantized tensor. Is this likely due to missing support for quantized `onnx::Pad`,stale/low urgency,2,stale/low urgency,1,0,127606,,,,0,ba0a87ad1bd37b3c332aeda5bee4d9794288db48bed712888604ac77238c9ee5
thread:issue:127615,issue #127615: TORCH_LIBRARY breaks when passing (unexpanded) macro as namespace argument,stale/low urgency,2,stale/low urgency,1,0,127615,,,,0,f122ca337e3bde126911d2690aa006d1ac48c94e74e23adea6144a6b26d67941
thread:issue:12763,"issue #12763: Caffe2: Two entries of external_input ""data_0"" in mnist_predict_net.pbtxt file",stale/low urgency,2,stale/low urgency,1,0,12763,,,,0,52c31e89dd8bb014f438a997582e92350a54ce76de76ba7298627ea26a320cb8
thread:issue:12764,issue #12764: [feature request] Operator Overloading,stale/low urgency,2,stale/low urgency,1,0,12764,,,,0,1380afa4f790d43c85e21c2679d04331df3032e38efc61c6ae6fd8906f217784
thread:issue:127676,issue #127676: module 'torch.mps' has no attribute 'device',stale/low urgency,2,stale/low urgency,1,0,127676,,,,0,fd797bc22427307eefdf5a30ea9aa16e150d60bde5112e1d4f21f7428bcd18dd
thread:issue:127701,issue #127701: The type of exception thrown by torch.nn.Conv2d does not match the actual situation,stale/low urgency,2,stale/low urgency,1,0,127701,,,,0,817a9dead7192de5ebc986ae3f892ce3e6468eeee2599c1ea7acaf7c3d97729d
thread:issue:12773,"issue #12773: Build the docker image from source, but torch.cuda.is_available()==false",stale/low urgency,2,stale/low urgency,1,0,12773,,,,0,5f05a7a9f61d5836e227e6c54c70754193ff3f1606dba8754ff8a135a20f6062
thread:issue:127732,issue #127732: torch.nn.ConvTranspose2d/torch.nn.ConvTranspose1d lacks a check for the parameter dilation in the model building process,stale/low urgency,2,stale/low urgency,1,0,127732,,,,0,07ca11eb7b8b44bab6da3e42925e3adf71f91d5e7cc611a0449fa6582f167502
thread:issue:127733,issue #127733: torch.nn.Linear does not check the value of the out_features parameter,stale/low urgency,2,stale/low urgency,1,0,127733,,,,0,f5e42ac1ee6721e1493be5869a0cfe7a830ca4fd7712f319fb2a6a1c2137e958
thread:issue:127734,issue #127734: nn.ConvTranspose2d does not check for parameter combinations resulting in convolution output with dimension 0,stale/low urgency,2,stale/low urgency,1,0,127734,,,,0,260f5f804858f11db5deb56a42f4623ba285af30c33a59a9f1901fb9b731c632
thread:issue:127736,issue #127736: torch.nn.BatchNorm1d missing check on values of parameters eps and momentum,stale/low urgency,2,stale/low urgency,1,0,127736,,,,0,65a7b523dcb25054ae55f28cbf39bb150c74830b0373045d592b7bddfffc549e
thread:issue:127737,"issue #127737: `torch.tensor()` can also be used with a list of the `float`, `complex` or `bool` type tensors which have a single element each",stale/low urgency,2,stale/low urgency,1,0,127737,,,,0,5cbca96e700e05ce806b1e0ba94b20d4c1370207b94c9e09f40ac6f42f3ad10d
thread:issue:127743,issue #127743: RuntimeError using nested tensor in Apple M1 device MPS,stale/low urgency,2,stale/low urgency,1,0,127743,,,,0,1c345dbb8b568b87261486cef832b41b88412270cb7b8d38ffcc91b93416b6cf
thread:issue:127750,issue #127750: aot_function cannot handle index_add,stale/low urgency,2,stale/low urgency,1,0,127750,,,,0,2bc6ba3b10d7a9ce0d682c6327a826d20e9fa5765e11766385ebe42d96a14db8
thread:issue:127754,issue #127754: profile not record at::meta_xxx time,stale/low urgency,2,stale/low urgency,1,0,127754,,,,0,f7c8a841f2cb08f315ad907612059600f15d4621ba6405b2b9746f4a50c4e06a
thread:issue:127788,issue #127788: ghstack and pytorch-bot can race,stale/low urgency,2,stale/low urgency,1,0,127788,,,,0,d01e3f43242ed54e3254816276ffa610735d647a9840f9a4e17bd7de5aba3b43
thread:issue:127873,issue #127873: [libtorch] `CppCommInterface` subclasses cannot be instantiated,stale/low urgency,2,stale/low urgency,1,0,127873,,,,0,f7e26ff9e065edfd0afc29934b96942a3b6291a3315e6a5a761f629409872ca2
thread:issue:127875,issue #127875: [ShapeInferenceError] Attribute dilations has incorrect size,stale/low urgency,2,stale/low urgency,1,0,127875,,,,0,6a4c7093eeeaf4c263d04c238f1393cd1ecb93c725f13458a1e6fe45fff0747b
thread:issue:127886,issue #127886: Document the torch.fx.experimental.proxy_tensor.make_fx function,stale/low urgency,2,stale/low urgency,1,0,127886,,,,0,079dc4e5b37a722121a9339e8f299cc2934231f09027c97c4b4a54139aef1cd5
thread:issue:127915,issue #127915: Document the torch.fx.traceback.preserve_node_meta function,stale/low urgency,2,stale/low urgency,1,0,127915,,,,0,d2e63e22e441f385f9a22142f4b08b6cc55b433de35438cb3a868fe7c949d9c5
thread:issue:127917,issue #127917: Document the torch.cuda.profiler.start function,stale/low urgency,2,stale/low urgency,1,0,127917,,,,0,d81ef1423c7375b99ec2ede70e18104747c06eea437a26508c7135124278bb51
thread:issue:127928,issue #127928: Ability to do aot/inductor compilation from a jit model (or torch.exported model),stale/low urgency,2,stale/low urgency,1,0,127928,,,,0,3b4a16a72a0f7321ea6b1b8d9c04fd7c17dadd840bccea7a07a01dbe95c7cc99
thread:issue:12795,issue #12795: C++ frontend: how to debug nan gradients,stale/low urgency,2,stale/low urgency,1,0,12795,,,,0,e0da77d943b66b08ace20378aea52b39c75c3eb2568c14c75e705f8cc13849a9
thread:issue:127950,issue #127950: Ensuring reliability of PT2 with TorchRec on PRs,stale/low urgency,2,stale/low urgency,1,0,127950,,,,0,b1905f44477bd15528708e55029fc03cee8ee47871d26a957b99e4a8244215b7
thread:issue:127964,issue #127964: [pipelining] Supporting skip connections in 1F1B schedules,stale/low urgency,2,stale/low urgency,1,0,127964,,,,0,c800e0d7a1e15141de2c2d86ed4628b11f6042be44bda67d9c24de5cf6aa3158
thread:issue:12797,issue #12797: Port dragon4_scientific for pretty float tensor print.,stale/low urgency,2,stale/low urgency,1,0,12797,,,,0,67dacf534d6c22686ab916d7908f8a01fabbf7edf35393077c435d2bc968b072
thread:issue:127973,issue #127973: Torch compile OpenMP error with simple nn.Module,stale/low urgency,2,stale/low urgency,1,0,127973,,,,0,0e0cc40fe5eb169c6ecf5f3f48ab0085c44116f7addb1516d0c19b65722b48aa
thread:issue:128012,issue #128012: compile PixArt-sigma error,stale/low urgency,2,stale/low urgency,1,0,128012,,,,0,85f8470184278161c17b5459b4162cec6b69bfff466e69eae08d0f38edb18436
thread:issue:128015,issue #128015: Best approach for dynamically changing batch sizes on iterable datasets,stale/low urgency,2,stale/low urgency,1,0,128015,,,,0,0b63fab1eeeb806f1057abca556cca3b45856d6bf4da2562a861e212428b370e
thread:issue:128020,issue #128020: Make `torch.autograd.Function` support `vmap`,stale/low urgency,2,stale/low urgency,1,0,128020,,,,0,e10123d1dff2f281b73b453c4060376ed5dfa30b2089d75ea063a13cd06d5a10
thread:issue:128033,issue #128033: Network module memory is not released in libtorch,stale/low urgency,2,stale/low urgency,1,0,128033,,,,0,02bb35dc766e17dfa13279c06a85ae2207e70bd0bcd744b1b749755deb35a81b
thread:issue:128044,issue #128044: Request for in-place FFT,stale/low urgency,2,stale/low urgency,1,0,128044,,,,0,5c54ee96121723854ed32ab35b73368c4a1172eb61059c4b3e57cb6b3d0de8e7
thread:issue:128053,issue #128053: Handling reasoning about rationals in symbolic shapes,stale/low urgency,2,stale/low urgency,1,0,128053,,,,0,ac425e66b0d9b7bbc7cc9f109d6a52df60ec4a6739efda487b4243248a4fbc49
thread:issue:128064,"issue #128064: Improve debugability of warnings/errors ""Triggered internally at""",stale/low urgency,2,stale/low urgency,1,0,128064,,,,0,9e97f77e0fb2b6e5383f7414004e9b3c9964884c2c69a4836e885a1e49a2a8c4
thread:issue:128066,issue #128066: reduce_scatter_tensor with strided inputs produces corrupted results,stale/low urgency,2,stale/low urgency,1,0,128066,,,,0,8f413d0ce5c98d2832ee78a8e434a62afa2c434a213c39359b3f58c774884812
thread:issue:12808,issue #12808: tutorial_blob ERROR,stale/low urgency,2,stale/low urgency,1,0,12808,,,,0,c71f0083a1056b18d0e7c6865f9d4c3e28451ab134aac7d33707692faad0ec01
thread:issue:128095,issue #128095: [inductor] Graph breaks in CohereForAI/aya-23-8b,stale/low urgency,2,stale/low urgency,1,0,128095,,,,0,e4f5dc0ca9d5bd1425b07d962b2e95967e32ed06fdd23ae3ad99000207dbc30c
thread:issue:12812,issue #12812: Pytorch BatchNorm2D Unstable,stale/low urgency,2,stale/low urgency,1,0,12812,,,,0,6e5137a1cbeca5bc246b12a396e7a623e6aff7f617a60213d9ddafe06f2a6e0c
thread:issue:128126,issue #128126: Stateless random numbers,stale/low urgency,2,stale/low urgency,1,0,128126,,,,0,c3a52107a720e7265402a7e259cb38026a7b8e41916ff40fb8280ea52740d2f0
thread:issue:128128,issue #128128: Fake tensor support of distributed ops in FX graph tracing,stale/low urgency,2,stale/low urgency,1,0,128128,,,,0,d5bbfa05cd7f078dcf16352d377145de8f7dbb00934d998d675f21283e238f46
thread:issue:128149,issue #128149: torch.compile with Custom tensor subclass doesn't inline the tensor subclass methods,stale/low urgency,2,stale/low urgency,1,0,128149,,,,0,2fe30339a2f925a178f154c9559f17a8b025a398caead4b5a19ce5a27bbde573
thread:issue:128150,issue #128150: More economical use of FX nodes when generating symbolic shapes related manipulations,stale/low urgency,2,stale/low urgency,1,0,128150,,,,0,06cc21255f00c9da80fa252a42bfdd8184b3768a7a1f9d4975410d4a33fae5cc
thread:issue:128153,"issue #128153: torch.compile Jamba: long compilation time with backend=""eager""",stale/low urgency,2,stale/low urgency,1,0,128153,,,,0,1391d7077ceaa5c6425353165b1c4b8fc9b141ed2437c5fb9d2aeb553cee69ba
thread:issue:128165,issue #128165: `swap_tensors` fail when calling `nn.Module.to` on XLA `DDP` wrapped models.,stale/low urgency,2,stale/low urgency,1,0,128165,,,,0,bc011aea9b392d783ce7511570fb60372c535d07cfa9bb905fed6a8a8d16a404
thread:issue:12819,issue #12819: Initialization error when moving data to the GPU,stale/low urgency,2,stale/low urgency,1,0,12819,,,,0,02031095f84512141383bf34e039ae39e8032abccd6634899469f20a5b680ccc
thread:issue:128204,issue #128204: Documentation for DDP-related environment variables,stale/low urgency,2,stale/low urgency,1,0,128204,,,,0,b3dbcd7090f5bfcf1fcc62cce3fb1df6d38ae87bc231f6f9988110bc5260c774
thread:issue:128208,issue #128208: Adam is storing step as cpu tensor,stale/low urgency,2,stale/low urgency,1,0,128208,,,,0,257dc51ae2f56eb542010b75bd0607659ce133b60956948bfadf1c35693fe77b
thread:issue:128257,issue #128257: can not create custom gradient when fp8 type is used,stale/low urgency,2,stale/low urgency,1,0,128257,,,,0,bdb43bc9e1c25f22fe9152a0ba3789cc688b43d27a3261cfbd6022996ad5c95e
thread:issue:12828,issue #12828: BUILD_BINARY is a lie,stale/low urgency,2,stale/low urgency,1,0,12828,,,,0,9ea0ee49d59d7beeca773bd5d29279bd5ab46f4b477e89e19274631f88041ece
thread:issue:128296,issue #128296: Proposal to add a CUDA backend for tensor conversion from BSR/BSC tensors to CSR/CSC tensors,stale/low urgency,2,stale/low urgency,1,0,128296,,,,0,a113b1173b3c1c693cb24d771a2bd6fa806a59ba11c8ea37d6aa97197d6d5579
thread:issue:12830,issue #12830: Test that (cd build && ninja) immediately after build is no-op in CI,stale/low urgency,2,stale/low urgency,1,0,12830,,,,0,ddae51f86e187e73354356941b4c27f7071569a2d5ea1cb3ef5fa2cd6b260883
thread:issue:128314,issue #128314: `docker compose up` fails with `RuntimeError: Unexpected error from cudaGetDeviceCount()`,stale/low urgency,2,stale/low urgency,1,0,128314,,,,0,1fa964eb0066d27f754f1f75749988160dc2c055d4ced291ff3f1976a651a1c2
thread:issue:128323,issue #128323: UnsupportedOperatorError: Exporting the operator 'quantized::batch_norm2d' to ONNX opset version 14 is not supported.,stale/low urgency,2,stale/low urgency,1,0,128323,,,,0,7283605f6a82163dc464e24138bd58c44f84b513baf5d67dd717bc6c8cb1f204
thread:issue:128325,issue #128325: Cannot export onnx model with autograd.functional.jacobian,stale/low urgency,2,stale/low urgency,1,0,128325,,,,0,9a58b9d93a5a869392e8a82e7dee7de858b1e3b64e744f33361ebfa06bd13224
thread:issue:128351,issue #128351: `Tensor.is_shared` behavior is incorrect for tensors created with `torch.from_file`,stale/low urgency,2,stale/low urgency,1,0,128351,,,,0,31617829b9ef66ebe2e360e12820ef7dd692572d706d5eaff4145206c90c936a
thread:issue:128395,issue #128395: _foreach_atan2,stale/low urgency,2,stale/low urgency,1,0,128395,,,,0,aed4aa1ad1cb8c42b0924693c6d95e1c934967a78f274c58db27ee77264cfcd2
thread:issue:128412,issue #128412: A Permute Layer for torch.nn.Sequential,stale/low urgency,2,stale/low urgency,1,0,128412,,,,0,e6a6c010c6a4d47bb9b40e7f081a541d956c6b95775ed95794e8abac5f8fa3f8
thread:issue:128414,issue #128414: How to enable XNNPACK instead of NNPACK/MKLDNN in Windows?,stale/low urgency,2,stale/low urgency,1,0,128414,,,,0,1098e0181958fb6f105d595232f487660a208b22cc497d575d7eaca97ee04e35
thread:issue:128493,issue #128493: Inconsistency `CrossEntropyLoss` vs `BCELoss` regarding logits/probability space,stale/low urgency,2,stale/low urgency,1,0,128493,,,,0,267c7c556811e5f296ed4eae52e728a479f12cbf1ca837b0b9614893fd953a80
thread:issue:128495,issue #128495: 'make html' will print 'duplicate object description' warnings when there are 1~5 CPUs in the running machine,stale/low urgency,2,stale/low urgency,1,0,128495,,,,0,925a471f1ae4a096e6ed1a76bae7ba8edf870565e05ffbeeaaf7443c833b5891
thread:issue:128497,"issue #128497: [bug]When the variable in the dataset is of type cuda, calling the dataloader will result in 0",stale/low urgency,2,stale/low urgency,1,0,128497,,,,0,566747b0c88f79aed7efd3d36f5c0f333912f026e66aa341d1208125df8d0941
thread:issue:128509,issue #128509: elements of STL get converted to an `IValue` and cannot be converted to an STL.,stale/low urgency,2,stale/low urgency,1,0,128509,,,,0,9e1c0fc74c1dae7a1022c41f59e25d46d97b92f441695c767c69d602e2656071
thread:issue:128548,issue #128548: [torch.compile] Llama2 failure using dynamic shapes with Torch 2.4 nightly,stale/low urgency,2,stale/low urgency,1,0,128548,,,,0,fac337fab870da2def23c3a79c6065b283d2e133cf0d7143682cde48e6f76c18
thread:issue:12855,issue #12855: Model with Caffe2 runs much slower than it with pytorch in GPU mode !!!!,stale/low urgency,2,stale/low urgency,1,0,12855,,,,0,11dd8503e4226c8b86d79f42d68efb5a29c0013a343c2ed7a82586b0b8c72ce4
thread:issue:128600,issue #128600: jacrev and jacfwd yield different results if one uses torch.no_grad blocks in module,stale/low urgency,2,stale/low urgency,1,0,128600,,,,0,006549af2099e2b8d079e74846735c2732c788aa7cde778014f30e57e28b38ae
thread:issue:128601,issue #128601: Retry on CUBLAS_STATUS_ALLOC_FAILED,stale/low urgency,2,stale/low urgency,1,0,128601,,,,0,56c5605b31ece5496f56bdf1664af82ec7c9ebf786e87a9af0e33ca0ee4d0e35
thread:issue:128665,issue #128665: [RFC][pipelining] PipelineStage should let user control send/recv endpoints,stale/low urgency,2,stale/low urgency,1,0,128665,,,,0,64513a18c7d828663626aa6840db6d69ba01dbfa2da142920e750b1c802f3110
thread:issue:12868,issue #12868: [CAPI] Increase of memory usage when exporting a Adam optimzer,stale/low urgency,2,stale/low urgency,1,0,12868,,,,0,334eb40fc024389ec367fd80233628e9453637730dec696246d65b4b6be8a4e4
thread:issue:12869,issue #12869: caffe2: Unsupported type of tensor: nullptr (uninitialized),stale/low urgency,2,stale/low urgency,1,0,12869,,,,0,a6bd5b467c7a11215f9ad61cae1c9d26e397984ecd5ca2df9a04d01ee238ef32
thread:issue:128696,issue #128696: [DDP] DDP bucket memory release during fwd step,stale/low urgency,2,stale/low urgency,1,0,128696,,,,0,642a73ca82c94a8b912c3a5675315eb5e468c01bb6fc4636fb1c65f8de905ad5
thread:issue:128699,issue #128699: NotImplementedError: Operator aten.native_layer_norm_backward.default does not have a sharding strategy registered.,stale/low urgency,2,stale/low urgency,1,0,128699,,,,0,ac6db94bfebd41697cc54837453343c82080f285cef1883ffa40e988d83f17f5
thread:issue:128703,issue #128703: autograd with `is_grads_batched=True` fails on GroupNorm,stale/low urgency,2,stale/low urgency,1,0,128703,,,,0,ddf9d1c91c1bdd8303599e96573a4f6fa37cb27bafc8b3310aa1e276e7098574
thread:issue:128706,issue #128706: does FSDP support AMSP (a new DP shard strategy),stale/low urgency,2,stale/low urgency,1,0,128706,,,,0,3d2c8f275c93268732e02777fa93408e16b0f656c19d63c7eed22c686c8d7228
thread:issue:128735,issue #128735: Flaky test page should include retry runs,stale/low urgency,2,stale/low urgency,1,0,128735,,,,0,9883d4dc59f4b88a52d9b34c47a7e9b4c59fb9eceb0715816673f7246d03e98b
thread:issue:128779,issue #128779: The name of the function `nn.L1Loss()` should be `nn.MAE()` or the name of the function `MSELoss()` should be `nn.L2Loss()`,stale/low urgency,2,stale/low urgency,1,0,128779,,,,0,038ae7c8639b1e3b449d39ab855992e472801d4151ba333568f385709f4d893a
thread:issue:12879,issue #12879: Eigen in Caffe2 doesn't produce vectorized instructions,stale/low urgency,2,stale/low urgency,1,0,12879,,,,0,0488808773e70ff046d53ce8cd82336611d7d9f37f9a3406a8116b8a673b4ea5
thread:issue:128791,issue #128791: CUDA error in torch.cdist with compute_mode=donot_use_mm_for_euclid_dist,stale/low urgency,2,stale/low urgency,1,0,128791,,,,0,abe0e2f7113a1c414dbdfa9f67be7cd0fa5334c12892c0859bebe9181c312ebb
thread:issue:128821,issue #128821: Questions about TCPStoreLibUV,stale/low urgency,2,stale/low urgency,1,0,128821,,,,0,447eb3f170c687cc86895ef7b0aedfc6587ab771ad54817d52d8fd1174b563cb
thread:issue:128910,issue #128910: [pipelining] Free memory in stage after use,stale/low urgency,2,stale/low urgency,1,0,128910,,,,0,5c8312d2ec0c50c7066283332d8d7c346a6bd6e603aed58313f10e1ef806b1d0
thread:issue:128936,issue #128936: Performance degradation for certain input using Conv2D,stale/low urgency,2,stale/low urgency,1,0,128936,,,,0,cbfcf27358ebc1c30398033b964e1f7aa4ae340ad20abdc1316e7f4841a70119
thread:issue:12895,issue #12895: RelaxedOneHotCategorical not implementing entropy (and other abstract methods),stale/low urgency,2,stale/low urgency,1,0,12895,,,,0,d47d9e28fc591c9e15b4568384355764b942efa6468f61c48714cbaff122e5a4
thread:issue:128952,issue #128952: Question about torch.lstsq and torch.linalg.lstsq,stale/low urgency,2,stale/low urgency,1,0,128952,,,,0,ff4b1290cfe96d448d7b541091679378fab2e4158d4dd759687b94405071f3a0
thread:issue:128960,issue #128960: [Dynamo x torch_function] magic methods and methods with regular Tensors don't seem to work,stale/low urgency,2,stale/low urgency,1,0,128960,,,,0,94f918d9232e15002312d47adb3259eec94488cfda6239ad2b3de28584c2fb04
thread:issue:128963,issue #128963: RuntimeError: NCCL error: invalid usage when deploy LLM model by vllm. (torch version: 2.3.0+cu118),stale/low urgency,2,stale/low urgency,1,0,128963,,,,0,cadc78aad691b04e7726399e20c395078e6fd7edd46e1722578615681ed4ec45
thread:issue:128971,issue #128971: Trying to use forward AD with _scaled_dot_product_flash_attention that does not support it because it has not been implemented yet.,stale/low urgency,2,stale/low urgency,1,0,128971,,,,0,e56c1af940fd4be61b2fd203068fe82e6fb99173907e2f9b4d2edeee9c019cae
thread:issue:128988,issue #128988: Look up tensor device member inside Tensor is_pinned() implementation instead of accepting an outside input,stale/low urgency,2,stale/low urgency,1,0,128988,,,,0,a58d11871f7a8340277fff707815a8e49bcaaf8b42f0c9645725b958b6bbd889
thread:issue:129027,issue #129027: Extract some public APIs from torch::cuda::initModule(module) to torch::initModule(),stale/low urgency,2,stale/low urgency,1,0,129027,,,,0,3a8f92a666be505f4e0f0e6864812c94552395ed1b0dc885dc1195720840ab09
thread:issue:129050,issue #129050: Add comment for label_smoothing parameter in torch.nn.CrossEntropyLoss,stale/low urgency,2,stale/low urgency,1,0,129050,,,,0,fdcc538ab4c292b1a8cc001016eccd09ab30fdf740b4444c6d1f7159d4c05994
thread:issue:129085,issue #129085: Adding betainc,stale/low urgency,2,stale/low urgency,1,0,129085,,,,0,943dd268c4ca79519902442d3c8329f25be195cf65b3669e60d1be9484fff8b8
thread:issue:129122,issue #129122: Questions about CVE-2024-31583 and CVE-2024-31580,stale/low urgency,2,stale/low urgency,1,0,129122,,,,0,9ac34c501e7f834527eb01b062f63c1366d092c5a060b40584d018c966fe258c
thread:issue:12913,issue #12913: ubuntu16.04 build from source caffe2,stale/low urgency,2,stale/low urgency,1,0,12913,,,,0,98282177884dd8cf004f5e7f47c4616c9b2747bb0c2d2be58ba3dc8260fbd238
thread:issue:129176,issue #129176: extending forward-mode AD docs should really point to an example,stale/low urgency,2,stale/low urgency,1,0,129176,,,,0,166b8fdae0c07bcfd1ce4d571272e6961485faeca7c04dd2c1b1f46ccacc71b7
thread:issue:129191,issue #129191: RuntimeError when using torch.ops.aten._jagged_to_padded_dense_forward with large jagged tensors,stale/low urgency,2,stale/low urgency,1,0,129191,,,,0,3653d35d890357a1a1f7860de4d3279cdfeb03c0466cd94ffb7270b189821780
thread:issue:129200,"issue #129200: Export model using onnx.dynamo_export has bug of ""torch._dynamo.exc.Unsupported: call_method TupleVariable() size [ConstantVariable(int)] {}""",stale/low urgency,2,stale/low urgency,1,0,129200,,,,0,b87891425155250bea3fec62255017afd5a4491504e0daf38336a15efcd36bd4
thread:issue:129222,issue #129222: ignore all target for CrossEntropyLoss 2d，return nan but 1d return 0,stale/low urgency,2,stale/low urgency,1,0,129222,,,,0,2fd942cdf535fb13df62e5d064543d1756cb8c3f0d0a19e9e9ff9c1d86d8f57c
thread:issue:129224,"issue #129224: _If pin_memory_thread is alive but dataqueue is empty, it will fall into a dead loop,",stale/low urgency,2,stale/low urgency,1,0,129224,,,,0,2d05a33651c842c8dcfaf2cf1c5613eb85ef4a317783a812d751ddd3fa437066
thread:issue:129225,issue #129225: Why need to transpose when collate a sequence data in dataloader?,stale/low urgency,2,stale/low urgency,1,0,129225,,,,0,70bdc62cc030b03279149aeab18769de40e462c25f12e0643db7821d1c973ecd
thread:issue:129228,issue #129228: CVE-2024-5480 reported by security analyzers,stale/low urgency,2,stale/low urgency,1,0,129228,,,,0,df2d163fda8fe6d6e579c4e09033788e93a6560b2a470926237549c254d0f5e6
thread:issue:129293,issue #129293: forward_ad ignores checkpoints,stale/low urgency,2,stale/low urgency,1,0,129293,,,,0,c41703d5ea2ad0e6c0d44418ddf0622b2aaf5e386a9aaaaf2b6442f90a86aa82
thread:issue:12931,issue #12931: USE_OPENMP=OFF is ignored [Caffe2],stale/low urgency,2,stale/low urgency,1,0,12931,,,,0,32a31c07c82c880073820006bdf7ed0a8c4256a9afadde9e222d0ac1e0d8f5a7
thread:issue:129310,issue #129310: Upgrade dependencies MKL and Intel OpenMP to 2024.2.0,stale/low urgency,2,stale/low urgency,1,0,129310,,,,0,88eb8cf263570ab9cfc39fb009bdd166b9bbf23fe08fa3863ccdd2c48ffc029a
thread:issue:129329,issue #129329: _foreach_addc_,stale/low urgency,2,stale/low urgency,1,0,129329,,,,0,8338cae073139b45285babcfcb6c837e1f19ef67f6cbb90a9378b1d8ee326509
thread:issue:129330,issue #129330: Fuyou Training Framework Integration for PyTorch,stale/low urgency,2,stale/low urgency,1,0,129330,,,,0,3bb5c5fca48968b0d7b0e244d11f4a2904a1005cfe8622e46a4241896d062f44
thread:issue:129336,issue #129336: pytorch-nightly export KINETO_USE_DAEMON=1 Cannot initialize CUDA without ATen_cuda library,stale/low urgency,2,stale/low urgency,1,0,129336,,,,0,b8ca9e83fcb8b7654fd6efdc6d0b76b68a6f322122af2a7e3e489d4061eb9782
thread:issue:129356,issue #129356: Does it still make sense to prevent saving multiple tensors that have a shared storage but different dtypes?,stale/low urgency,2,stale/low urgency,1,0,129356,,,,0,ee4fae6fed626ff284ade0740384abe39c6b8b995a27c92688eb0e7b8c46902d
thread:issue:129358,issue #129358: Pytorch build fail with GCC 14.1.0 due to third_party/fbgemm/src/UtilsAvx512.cc:970:35: error: ‘r’ may be used uninitialized [-Werror=maybe-uninitialized],stale/low urgency,2,stale/low urgency,1,0,129358,,,,0,ba60e236bf48d029e5c7cf8ad1175db58c03e60a59a8afb3edfd10466a62b066
thread:issue:129386,issue #129386: Support a more flexible way to mark all CI jobs in a similar category as unstable,stale/low urgency,2,stale/low urgency,1,0,129386,,,,0,75e662171da8d424f1e2a2d53ea84abae9733846dad5352569a4ade912b52464
thread:issue:129410,"issue #129410: line-length of ruff, black and isort are not the same in pyproject.toml",stale/low urgency,2,stale/low urgency,1,0,129410,,,,0,d1cf90517193e2c7ffbc5e1bdc5f00a3582d777f7291c0f8c4f93ef2a3f737e0
thread:issue:129447,"issue #129447: Pytorch of 2.3.0 version loads bool type mask in triton, encountering error when use tl.where.",stale/low urgency,2,stale/low urgency,1,0,129447,,,,0,9eacc20e31a0a159caf7a03a272faec2c88822386dce2085d79d697140593180
thread:issue:129459,issue #129459: Add pmap to torch.func for Parallel Function Execution,stale/low urgency,2,stale/low urgency,1,0,129459,,,,0,95adea81cecb2ce67387c358afd58c22379ce8cf6b2103561d776a7024951790
thread:issue:129477,issue #129477: The ProcessGroupNCCL is not being destructed,stale/low urgency,2,stale/low urgency,1,0,129477,,,,0,c99b279dfd15715acca087d8670b422b1807e222bc9ffae5080e1d6703bfa789
thread:issue:129482,issue #129482: Dynamo graph breaks on boxed ProcessGroup,stale/low urgency,2,stale/low urgency,1,0,129482,,,,0,0d70fa9a15efe8f8a75236bd54b11cc5ce35d3f1f64048688d3a870c2cd57b04
thread:issue:129523,"issue #129523: `test_dummy_mha_with_nt_cuda` fails on `sm70`, `sm75`",stale/low urgency,2,stale/low urgency,1,0,129523,,,,0,8be36a600dfab9b5434dbc796753ea25870750d21e3f17d9111320bca0eb5329
thread:issue:129555,issue #129555: Libtorch 2.3.1 requires Glibc_2.28 and can't be used on ubuntu 18.,stale/low urgency,2,stale/low urgency,1,0,129555,,,,0,63e9991fed85aa20d7653d4b54bd3b9bd2032c04e1b347de548bc9daef2d8886
thread:issue:129564,issue #129564: Support for 5-D output image tensors in `Col2Im`.,stale/low urgency,2,stale/low urgency,1,0,129564,,,,0,1644ca368a5741f920493efe45841e2f02f948dc50bf3e27cab0853e8b2d808c
thread:issue:129565,issue #129565: ChainedScheduler fail on CosineAnnealingWarmRestarts,stale/low urgency,2,stale/low urgency,1,0,129565,,,,0,a009c331e279fdd11ba23ca225e451ceba1ef2d50a87f918f1c294ae83f6453b
thread:issue:12962,issue #12962: Come up with a better strategy for noticing BC-breaking attribute additions to serializable classes,stale/low urgency,2,stale/low urgency,1,0,12962,,,,0,87146cdad0e3f26f29638a745befd43765666ac5bcaaaac78357335b0018696a
thread:issue:129625,issue #129625: Triton segfault with compiled flexattention kernel,stale/low urgency,2,stale/low urgency,1,0,129625,,,,0,96bccbee8d5480f62f981119677775fe8c79c93016aea6bfc75510207503c23c
thread:issue:129634,issue #129634: Type Error when using from_numpy,stale/low urgency,2,stale/low urgency,1,0,129634,,,,0,51be5685f3beb4efc58b4d7dc78e4ef22c0cb6837fe963b6a2490d7459584731
thread:issue:129649,issue #129649: Low GPU precision,stale/low urgency,2,stale/low urgency,1,0,129649,,,,0,2f39f7a8ef6b253b496a3dfa54d448aba4b43c604c53132e9ab6ffaa44fb4437
thread:issue:129656,issue #129656: RuntimeError: Shared memory manager connection has timed out,stale/low urgency,2,stale/low urgency,1,0,129656,,,,0,728e79ca4682e941aebe083221caa647bc2b6438188a08cb999b1b46325f4d44
thread:issue:129670,issue #129670: Libtorch does not support SM90,stale/low urgency,2,stale/low urgency,1,0,129670,,,,0,f7eb3d77d53377395442264f8266f88134a3528fc6bdda3ba6e56b79b29ddab9
thread:issue:129672,issue #129672: Distributed training with NCCL hangs on init_process_group when debug is DETAIL,stale/low urgency,2,stale/low urgency,1,0,129672,,,,0,1256c93b1b7ff8527b89459511a6e9387b002ae4106025c2e6916e63d663eb67
thread:issue:129732,issue #129732: Torch Distributed Pipelinging API datatype mismatch,stale/low urgency,2,stale/low urgency,1,0,129732,,,,0,4f7ea2fd1443c1d36a0c7f8bc8e6dc49072e667e68a9ac52ed3b609a30e4b219
thread:issue:129784,issue #129784: `@pytorchbot merge -i` not ignoring currently failing checks,stale/low urgency,2,stale/low urgency,1,0,129784,,,,0,cf19c57d79da5c3ce9bf42e132ba871fcc729e34e7159e3d2234b200c002d6b9
thread:issue:12980,issue #12980: Installing pytorch from source on labs.cognitiveclass.ai,stale/low urgency,2,stale/low urgency,1,0,12980,,,,0,1f7c38b6dfaf867baf65f959eafd53c089f99e7ec0ee733822ba79b0362d82b4
thread:issue:129803,issue #129803: 3D matrix support in _scaled_mm,stale/low urgency,2,stale/low urgency,1,0,129803,,,,0,6270c8272849228f9c26b14c4caa10c6f99fbf48b2ecaa25f25efd1440a3340a
thread:issue:129843,issue #129843: Concatenating CSR matrices fails,stale/low urgency,2,stale/low urgency,1,0,129843,,,,0,331604f107e0803b5d8a1bf8abb466cccfc0527b92a08541982e480787090db8
thread:issue:129937,issue #129937: [tracker] Dynamo guard validation exposed issues,stale/low urgency,2,stale/low urgency,1,0,129937,,,,0,2a565564ab0be52e6e528508fcff5c032214578e36c3a63c3206d9082a29fec7
thread:issue:129942,issue #129942: Torch Compile mode giving Graph Break for Higher Order Op for ProcessGroup param,stale/low urgency,2,stale/low urgency,1,0,129942,,,,0,ad73b4dd89c7efac069ecce28fc503a848128acdc55514617726fceddbd49cc6
thread:issue:129954,"issue #129954: RuntimeError: NVML_SUCCESS == DriverAPI::get()->nvmlInit_v2_() INTERNAL ASSERT FAILED at ""../c10/cuda/CUDACachingAllocator.cpp"":799",stale/low urgency,2,stale/low urgency,1,0,129954,,,,0,7ec803cae9a3172714b92d48054cd25ccd67d864e3295c07359ed9fd574772eb
thread:issue:130021,issue #130021: [Functionalization] The inplace decompose op is meaningless,stale/low urgency,2,stale/low urgency,1,0,130021,,,,0,b226e733f9a47860ff324eda440604c3126390375dd2357ff25d0ee3ec2f79be
thread:issue:130067,issue #130067: Meta device not supported with autocast,stale/low urgency,2,stale/low urgency,1,0,130067,,,,0,9aacf01d4c9049667da952c44d85c1965d558d0d1392b07485420a866c3729d4
thread:issue:130073,issue #130073: ☂️ TORCH_INTERNAL_ASSERT failures while running CI,stale/low urgency,2,stale/low urgency,1,0,130073,,,,0,7b8eec2b0c66ff8c25f20ae13d809e1dc55b355f1b55fd8cb956dde3767a5c17
thread:issue:130078,issue #130078: cuda graph error for torch.isin if the input tensors are too large,stale/low urgency,2,stale/low urgency,1,0,130078,,,,0,a3cecbe8079b869ed2f1e50e3ac8b117f28627aa5d242a298221bfcbd20c99e9
thread:issue:130096,issue #130096: The time consumption gradually increases when activation checkpoint is iterated and run multiple times,stale/low urgency,2,stale/low urgency,1,0,130096,,,,0,20e654e59515d8ba18b48c6e790c1cd39eccfadc0427fd241dbe071940748b69
thread:issue:130114,issue #130114: [feature request] [discussion] Baseline ONNX interpreter / executor in python / PyTorch,stale/low urgency,2,stale/low urgency,1,0,130114,,,,0,3a176aaf05b5ff4322df513fa855c18eb47a1303e7567abd3b57d5ab7b94446a
thread:issue:130119,issue #130119: Compilation time for packing function is quite long,stale/low urgency,2,stale/low urgency,1,0,130119,,,,0,ca0eb20555eae56df497ce7273862b44b5295ae25ac28918f768edc62ad21c05
thread:issue:130124,issue #130124: Pin memory in subprocess,stale/low urgency,2,stale/low urgency,1,0,130124,,,,0,aadc97dee22d05203ec55fcb382191cac463c4d5f73f85c32a8b780d1b67274d
thread:issue:130128,issue #130128: Installation fails with conflicting dependencies (MarkupSafe>=2.0),stale/low urgency,2,stale/low urgency,1,0,130128,,,,0,beb572e02fd50bb58ec2355c3ee212108b985c44809ae3dd2cf97864666bbbef
thread:issue:130137,issue #130137: How to get stream operators in custom backend compiler ?,stale/low urgency,2,stale/low urgency,1,0,130137,,,,0,d05ba185456e81c1f671d93f698a505be06e1b9b79e0728572991e72d0ed931a
thread:issue:130162,issue #130162: caching_allocator_delete does not free up the memory allocated by caching_allocator_alloc,stale/low urgency,2,stale/low urgency,1,0,130162,,,,0,1e17534fd3903314667689c77f1f6bb5abb6f94d33df20c1c0854e6207b1fe9f
thread:issue:13018,issue #13018: Improved performance for torch.multinomial with small batches,stale/low urgency,2,stale/low urgency,1,0,13018,,,,0,97023272c5aef92096faf68dac62144efe4bace2d3d01c77f0c0af70bf7578a8
thread:issue:130193,issue #130193: Inconsistency `CrossEntropyLoss` weights value between in gpu and in cpu.,stale/low urgency,2,stale/low urgency,1,0,130193,,,,0,a52efcb162607c953223e54d8b2dca32af19c99763085b1b23c9dba903d61d53
thread:issue:130238,issue #130238: how to simplify torch.fx like using onnxsim?,stale/low urgency,2,stale/low urgency,1,0,130238,,,,0,594af6fe803a3cfe0e2d1f9fc1ddd4436afcc152e2eda8b66c3a5d80b32ecad0
thread:issue:130254,"issue #130254: Many types for `start`, `end` and `base` argument of `logspace()` work againt what the doc says",stale/low urgency,2,stale/low urgency,1,0,130254,,,,0,f384292f0b91cb82b3d13b4b15a5863e1b2d7f129577c6ef7473071437510947
thread:issue:130282,issue #130282: [custom_ops] We should be able to mark an operator as non-differentiable.,stale/low urgency,2,stale/low urgency,1,0,130282,,,,0,af894d25769237d6bf2647a41ca98ba446ba8447cf0f06c6b34e0ef81467e38c
thread:issue:13034,issue #13034: [caffe2]How can I export init_net.pb and predict_net.pb files on my own?,stale/low urgency,2,stale/low urgency,1,0,13034,,,,0,6e717d0cf3727d96266c3144835c7bad9c5c460a7f12aa531af03e753cb7589a
thread:issue:130342,issue #130342: Rescale to zero torch.nn.CrossEntropyLoss with label_smoothing > 0,stale/low urgency,2,stale/low urgency,1,0,130342,,,,0,a7ee97c3783d658c37d7dddb5a08da76fc6d66e51c1c1d703e38cebed37ebcd2
thread:issue:130359,issue #130359: FP8 Rowwise Scaling does not work on L4 GPU,stale/low urgency,2,stale/low urgency,1,0,130359,,,,0,ad6bce3bb6ddff2444af2ca642ee17b403ad4a26cb0de925746807eccbdf2bf5
thread:issue:130366,issue #130366: Deadcode emitted when using `jit.script` with `onnx.export`.,stale/low urgency,2,stale/low urgency,1,0,130366,,,,0,d414387161ef80fc9066c3a40a3cd5e0824a93ad694f3c9ac9b95e173e62cbd6
thread:issue:130396,"issue #130396: Add some way to automatically opt-out of CI features, like TD.",stale/low urgency,2,stale/low urgency,1,0,130396,,,,0,a40d646b21e657fa9159186da02f5cffa2b9aa639639aad9f212ed5c13ce3599
thread:issue:13041,issue #13041: [caffe2] How to handle multiple inputs and multiple outputs in the network architecture?,stale/low urgency,2,stale/low urgency,1,0,13041,,,,0,9002bc4dd29327c143ea9ea27cd0f94f63d872887d7fa454023227effa2a473e
thread:issue:130432,issue #130432: Deferred runtime asserts triggered during backwards should not trigger assert not self.runtime_asserts_frozen,stale/low urgency,2,stale/low urgency,1,0,130432,,,,0,e49fb5a429c7fbc98e5314c27b6af96878c2fef73698ad9f409a0a51a16ab16b
thread:issue:130448,issue #130448: `n` argument of `randperm()` with a 0D or more D tensor works against the doc,stale/low urgency,2,stale/low urgency,1,0,130448,,,,0,3552d654d1bd968a2e31b948c19ec6bd5595242314f66b5bbcff47a6318c6820
thread:issue:130464,issue #130464: [dynamo] Investigate unexpected success dynamo-wrapped tests from https://github.com/pytorch/pytorch/pull/129185,stale/low urgency,2,stale/low urgency,1,0,130464,,,,0,6f15969b235f546c380fb20cf1c5f62b148f57ba6ded2af60d7ecd755685c574
thread:issue:130470,issue #130470: nested_tensor jagged_from_list doesn't correctly handle corner case of empty list,stale/low urgency,2,stale/low urgency,1,0,130470,,,,0,34981c85a028a192a75a5e9c50a5c346cfb6d1b66693ada9fbed0cabf1f0b2d1
thread:issue:130512,"issue #130512: The argument of `take()`, `take_along_dim()` and `gather()` should be `index` or `indices` for consistency.",stale/low urgency,2,stale/low urgency,1,0,130512,,,,0,270719e088b4231214d302c90a1893ba89e8bffc0148c933e1b66838baa84703
thread:issue:130514,issue #130514: C++ API Does Not Expose Lazy NVRTC,stale/low urgency,2,stale/low urgency,1,0,130514,,,,0,e9cb23b67f0871ccc3a9d0be156f134388eeb3d23e268e51632d913d546f7769
thread:issue:130524,issue #130524: Unable to compile pytorch with conda python3.8 and cuda 11.4 in Nvidia AGX Jetson Orin aarch64,stale/low urgency,2,stale/low urgency,1,0,130524,,,,0,9b4953ef53a213be318cdf2723f6cc975b664962ca6aba2c6580556a58195d66
thread:issue:130525,issue #130525: [Feature request] Extend index_reduce to allow `index` to be of type List[List[int]],stale/low urgency,2,stale/low urgency,1,0,130525,,,,0,9f4314f3291c89f790902ca94f713160dff0b3e97f323f5d19411cf659012b45
thread:issue:13053,issue #13053: Generalized Data Class,stale/low urgency,2,stale/low urgency,1,0,13053,,,,0,cae627af30bad5455dccd36ea46679eed6800e4b8ff0f81b4845b1b3959fc552
thread:issue:130534,issue #130534: Clarify invariants established by torch.fx.passes.runtime_asserts or factor so we can have a pre-pass that reestablishes them,stale/low urgency,2,stale/low urgency,1,0,130534,,,,0,cce1c7efd0577a5d3bd8a21c5b9bdfbd1482122e96e1475ba6b1bc9500caac86
thread:issue:130539,"issue #130539: RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Reshape node. Name:'/Reshape_5' Status Message: /onnxruntime_src/onnxruntime/core/providers/cpu/tensor/reshape_helper.h:45 onnxruntime::ReshapeHelper::ReshapeHelper(const onnxruntime::TensorShape&, onnxruntime::TensorShapeVector&, bool) input_shape_size == size was false. The input tensor cannot be reshaped to the requested shape. Input shape:{1,32,512}, requested shape:{1,1,8,64}",stale/low urgency,2,stale/low urgency,1,0,130539,,,,0,d9f231b67ab2acc43b3e2e15015d0fa9b571bf037aa0ad4338c45b469ac7020c
thread:issue:13058,issue #13058: Backward pass over torch.nn.functional.pad is extremely slow with half tensors,stale/low urgency,2,stale/low urgency,1,0,13058,,,,0,e90812911e02ba2e654375a824d804f30dcf6ef4503a38b6d2b1787e0d8613a9
thread:issue:130608,issue #130608: Setting `shape` and `dtype` without `dtype=` to `view()` works against error message but an `int` tensor is returned instead of a `float` tensor,stale/low urgency,2,stale/low urgency,1,0,130608,,,,0,8ce4d28d1b763c888a7fcd94855d2179afeb2573a8cd79b9048167494e1f6b71
thread:issue:130610,issue #130610: dataloader hangs with fork and pin memory=True,stale/low urgency,2,stale/low urgency,1,0,130610,,,,0,c9178ec82a80c487308cd6c0dc5ab8c2b6f21bf3bc1426fe4b9c06226d04d1a9
thread:issue:130621,issue #130621: [Bug Report] Choosing the network interface to use,stale/low urgency,2,stale/low urgency,1,0,130621,,,,0,8f9939054cae7499ad2824ea7747a5957dd0ab35bf0d41b1ade92f158b086803
thread:issue:130636,issue #130636: `dims` of `permute()` without `tuple` or with `list` works agaist the doc,stale/low urgency,2,stale/low urgency,1,0,130636,,,,0,366e98a512d2c3092dbc85d3c2b2b83e21bf8fef306a3137f07645fe4b3f742f
thread:issue:130644,issue #130644: Add aten::_weight_norm_interface as an ONNX supported TorchScript operator,stale/low urgency,2,stale/low urgency,1,0,130644,,,,0,b602a4cc7dfbd7e8fbf298e37fbc5e25975afefd095e451f3f058598c1564fd3
thread:issue:130656,issue #130656: [torchelastic] torchrun with static backend hangs when used with port '0',stale/low urgency,2,stale/low urgency,1,0,130656,,,,0,02f1538f1c26ff58dc8b149fa5474096b045122e58b55fa711d0682a7cd0c633
thread:issue:130693,issue #130693: The doc of `vstack()` should say `tuple` or `list` of tensors for `tensors` argument,stale/low urgency,2,stale/low urgency,1,0,130693,,,,0,dd05ad0c966dafc37bc79a03258cfedcd5c4c05808516d2fac9b95cf92bb6beb
thread:issue:130701,issue #130701: fx.wrap() doesn't really work for things in torch/*,stale/low urgency,2,stale/low urgency,1,0,130701,,,,0,520080df57cf5f79d90a94aedd1c5d05a779863e4e04fc649c9790c42f0ff58e
thread:issue:130739,issue #130739: Ineffective Buffer Size Adjustment in Multi-Node NCCL Communication,stale/low urgency,2,stale/low urgency,1,0,130739,,,,0,4cddb6c162fa24eae6631e3613e14f7582a1d5eba3e5d4c41d8378610b8a4307
thread:issue:130768,issue #130768: Migrate gradcheck with check_batched_grad=True to use the new torch.vmap,stale/low urgency,2,stale/low urgency,1,0,130768,,,,0,6044569c7426bdc9c1b54883b4a443a3f58b1b36dc2e94813728c8f78a08c356
thread:issue:130769,issue #130769: [Inductor] Add an API to register external callable candidates for inductor's Matmul/Conv tuning choices,stale/low urgency,2,stale/low urgency,1,0,130769,,,,0,8757248ba15b18f598d578de1ffb3d32782dabe57b065cc730533d862fe0c430
thread:issue:130792,issue #130792: `with torch.device()` modifies the `as_tensor` intended behavior,stale/low urgency,2,stale/low urgency,1,0,130792,,,,0,d83d5bf599b13e84f81cb47dcc45c47c53015bc8d999c082be5b07ca94bec244
thread:issue:130807,issue #130807: Generalizing cross product to hodge star operator,stale/low urgency,2,stale/low urgency,1,0,130807,,,,0,24446dae49dfdba95bee605d824a1f6dc2c0731a4bdd7b305b6c3d8467c55cdb
thread:issue:130847,issue #130847: Where does torch check for autocast op availability?,stale/low urgency,2,stale/low urgency,1,0,130847,,,,0,a0ae2c924feed76c0e3e22ded6fc8598e650fc745cb81a3792ff2d6600de128d
thread:issue:130863,issue #130863: CMake cannot detect non-default fp16 installation path,stale/low urgency,2,stale/low urgency,1,0,130863,,,,0,aaf1094513be360e3841ee7599cb0588ed934359f1993bee03bc33d1ec7ae1bc
thread:issue:130932,issue #130932: Error using torch.func transforms with DDP (broadcast_buffers=True),stale/low urgency,2,stale/low urgency,1,0,130932,,,,0,2acf6e32aa809d9f7dba7e0c2ad3de56f99f52e4d1a015ace20a64f170a8a92f
thread:issue:130948,issue #130948: [inductor][cpu]opacus_cifar10 performance regression in 2024-07-15 nightly release,stale/low urgency,2,stale/low urgency,1,0,130948,,,,0,18e0926c79ac7eeb3504c74de76fe9a5a4db2d2e170378bffebde2c6b3c61e43
thread:issue:130950,issue #130950: Improve detail / Create tutorial for torch.distributions,stale/low urgency,2,stale/low urgency,1,0,130950,,,,0,79f94f28ac990314188b57c54f4a68e2ed863c565d46072bdf194878450a37e1
thread:issue:130968,issue #130968: float8 types cannot be used under deterministic mode,stale/low urgency,2,stale/low urgency,1,0,130968,,,,0,7bd1a768396bf7241cfb63dbc3e541b11e3cb8eff1150e04c117f9b18c6cd563
thread:issue:130985,issue #130985: Runtime error when running huggingface pretrained model with torch-xla,stale/low urgency,2,stale/low urgency,1,0,130985,,,,0,1ae4537428cf1f53946ce8c865ec542f6bff0d91aec2df84f0fbe8609ee4e507
thread:issue:131011,issue #131011: xeon.run_cpu will not use numactl when whithin a docker container run with --cpuset-cpus non-0 numa node,stale/low urgency,2,stale/low urgency,1,0,131011,,,,0,e5fae26262dc4cce019297e3ef47f780784037f243f6c5beadc8eef5792e1206
thread:issue:131054,issue #131054: Shape-aware type hinting,stale/low urgency,2,stale/low urgency,1,0,131054,,,,0,ee4fdada6d10bfee028cd3a5b77156f11e86b275517cb46adcd9baf4ee01977f
thread:issue:131062,issue #131062: Add some more testing around checkpoint over compile,stale/low urgency,2,stale/low urgency,1,0,131062,,,,0,93387cb2cec441e8417c42ec6b48d13de938639e0deacc262b9ed6a9d6a8daa3
thread:issue:131110,issue #131110: Proposal for CUDA-Accelerated Dynamic Time Warping (DTW) Implementation in PyTorch,stale/low urgency,2,stale/low urgency,1,0,131110,,,,0,a68d2ea02836c38bbcf8303f249cb9ea01b586e3a6801efc1c9edff21efb07e4
thread:issue:131113,issue #131113: pytorch dataloader has function that raises error,stale/low urgency,2,stale/low urgency,1,0,131113,,,,0,41d81f14951d56d0d75d7cb5775c696ffa27d939e5787b0da03c54c7a6a32caf
thread:issue:131148,"issue #131148: The empty 1D or more D tensor for `median()/nanmedian()` with the deepest `dim` gets errors, not getting NaNs",stale/low urgency,2,stale/low urgency,1,0,131148,,,,0,67e459880360595726f00969794fd88d6e1a69b848023751fcea2a54865ffbde
thread:issue:13118,"issue #13118: warning: attribute namespace ""clang"" is unrecognized; High Sierra / Fedora compilation with clang results in spurious clang errors in nvcc",stale/low urgency,2,stale/low urgency,1,0,13118,,,,0,c2f0f31227e556a7e3692079460f8622c74d4d1098645af75f8b66876800f878
thread:issue:131196,issue #131196: Missing float8 storage,stale/low urgency,2,stale/low urgency,1,0,131196,,,,0,1e18984ef16a6ad6182bb9737c0b0e01517efc8e55d45f53d73f295aef6e8fed
thread:issue:13120,issue #13120: [caffe2] incompatible constructor arguments,stale/low urgency,2,stale/low urgency,1,0,13120,,,,0,a9dd99aab963912716b156d42758b6d760220f9e45cbdf8b91edb43b3bf1a2e8
thread:issue:131213,issue #131213: Clean up coverage_ignore_functions,stale/low urgency,2,stale/low urgency,1,0,131213,,,,0,a679710c7d9324d34953956cad7c7e6b8cf02925261e8016c84274cd8f06bdf9
thread:issue:131254,issue #131254: Masked Attention has no effect in ``TransformerEncoderLayer``,stale/low urgency,2,stale/low urgency,1,0,131254,,,,0,1b1c464b57e55443166de1876f97ac22c26239f8d2939f9c6459f52c7c69fb6f
thread:issue:131263,issue #131263: An API to re-enable dynamo-disabled functions,stale/low urgency,2,stale/low urgency,1,0,131263,,,,0,2aa0b2f350d4411ff11f657626c3525792c19c01790b0455610faa08517fbb45
thread:issue:131273,issue #131273: The empty 0D or more D tensor with exceeding `dim`(`dim=100` or `dim=-100`) for `cummin()` works,stale/low urgency,2,stale/low urgency,1,0,131273,,,,0,61cb8495d8a051ae0b612855c89ff91f2c5bf128eafcfa956b61ab8c9b35d7fa
thread:issue:131274,issue #131274: The empty 0D or more D tensor with exceeding `dim`(`dim=100` or `dim=-100`) for `cummax()` works,stale/low urgency,2,stale/low urgency,1,0,131274,,,,0,8c92308b8ea282315f9662b63e73d19b610035143c4ce6adac0fa560a42d49be
thread:issue:131283,"issue #131283: `input` argument of `nan_to_num()` works with `complex` type but `nan`, `posinf` and `neginf` argument don't work with `complex` type",stale/low urgency,2,stale/low urgency,1,0,131283,,,,0,d88a42c69358b7fcb00b4c580b03dd22c9c590d6c9d87f18aebac2e10d419f1e
thread:issue:131307,issue #131307: allow tuples as keys for ModuleDict,stale/low urgency,2,stale/low urgency,1,0,131307,,,,0,285ab20ad3202fd1122c36ba4891f0e562faa6a84721ff159eee0921045d67ea
thread:issue:131318,"issue #131318: RuntimeError: INTERNAL ASSERT FAILED at ""../aten/src/ATen/native/sparse/SparseCsrTensorMath.cpp"":1306",stale/low urgency,2,stale/low urgency,1,0,131318,,,,0,00cc64df4ced0cf92023273f67b9be51f56b44814a2f90dadddbb0503806557f
thread:issue:131349,issue #131349: [ONNX] opset 11 aten::index_put generating invalid add ops with None args,stale/low urgency,2,stale/low urgency,1,0,131349,,,,0,186232ec8235e70e235b02a5272f80cb8b599d4b820bc932d568285dde2fbbf6
thread:issue:131357,issue #131357: Follow ups from Increase (4x+) in build times #130744,stale/low urgency,2,stale/low urgency,1,0,131357,,,,0,90be3e603c43f163cf23d544e43f06f1ff9f2c0361d206d3b6de99ebb768479f
thread:issue:131432,issue #131432: Remove all # mypy: ignore-errors,stale/low urgency,2,stale/low urgency,1,0,131432,,,,0,95ea6438d52a65b77b4f0402d8a46170babea4b9eb85719e257f5347784a0891
thread:issue:131450,issue #131450: Make torch.compile support inter-op parallelism,stale/low urgency,2,stale/low urgency,1,0,131450,,,,0,662d34c6109895be8d4f9c45eb66a6578e37907ce0ecd061f750544afac56eb7
thread:issue:131463,issue #131463: Cannot subclass _TensorBase directly error,stale/low urgency,2,stale/low urgency,1,0,131463,,,,0,9c6f655767743feeaedb6f1825cc9a6fef96cceb6394c9bb8e3108b09e1034f0
thread:issue:131491,issue #131491: `torch._assert_async` ignores `assert_msg`,stale/low urgency,2,stale/low urgency,1,0,131491,,,,0,5148c3d5b4cb3a6b220e34e321ff6d42fcdbd84deb60cfa89637ebef6799dd03
thread:issue:13162,issue #13162: circular module reference raises RecursionError,stale/low urgency,2,stale/low urgency,1,0,13162,,,,0,d6b1b3437beddb2ae4c4cc604bca2ff69274057d9e90fd52240680794e043568
thread:issue:131631,issue #131631: DataLoader worker exited unexpectedly,stale/low urgency,2,stale/low urgency,1,0,131631,,,,0,e002d38068646e046768d82f92651c5d2d309f9739df025fc3892e104090f6eb
thread:issue:131655,issue #131655: tensor.share_memory_ causes temporarily increased memory consumption,stale/low urgency,2,stale/low urgency,1,0,131655,,,,0,7dea1c225586235303fd43b8e276576b58d67ffccced38b5ab15a84685f63395
thread:issue:131656,issue #131656: Remove CTCLoss limitation of maximum target sequence length 256 (to satisify CuDNN),stale/low urgency,2,stale/low urgency,1,0,131656,,,,0,52ba5980999b46657e1ff03791c03265df30363dad619676470f430c056e3b09
thread:issue:131688,issue #131688: Drop Connect Layer,stale/low urgency,2,stale/low urgency,1,0,131688,,,,0,a9926c54fbca4616130337105625db647ad23235add2b5b5842d88b1d18334af
thread:issue:131691,issue #131691: TCP socket keeps trying IPv6 address and doesn't try IPv4,stale/low urgency,2,stale/low urgency,1,0,131691,,,,0,f355a8bec5e842a6e7938ec5d1ac0baf1bc651d22e0f3be4d5e2ed4e780d427e
thread:issue:131695,issue #131695: Keep pytorch/pytorch-canary's main branch in sync with pytorch/pytorch's viable strict,stale/low urgency,2,stale/low urgency,1,0,131695,,,,0,52092ae13081f0398886fea927b1758c5549cc068608229ac3632eab245f7a8a
thread:issue:131736,issue #131736: MPS gives incorrect result when torch.nn.functional.softplus follows moveaxis,stale/low urgency,2,stale/low urgency,1,0,131736,,,,0,54032650a635d143c13d98bac1d558b98e42c95676734459070401dcbcabd74e
thread:issue:131753,issue #131753: PyTorch complains ```A leaf node was passed to _will_engine_execute_node``` when calculating the gradient of a model warpped by ```FlopCounterMode```.,stale/low urgency,2,stale/low urgency,1,0,131753,,,,0,64b3e95d845d57a3049acb12866c13529335f61cc1eec0ac88d2db5c10a728dd
thread:issue:131755,issue #131755: CUDA error: an illegal memory access was encountered when use TORCH_NCCL_NAN_CHECK.,stale/low urgency,2,stale/low urgency,1,0,131755,,,,0,7db1cbd5899f329567e45954ee585a4e583cce1aa7b3a7af151242163d84c15d
thread:issue:131769,issue #131769: FlopCounterMode requires `tabulate` dependency,stale/low urgency,2,stale/low urgency,1,0,131769,,,,0,918b6cff95ac6b42499e4bf76d9491c09fb02a15af51fd8e66cf1b71778baa1e
thread:issue:131772,issue #131772: Default argument removal for python dispatch fails for pin_memory on factory functions,stale/low urgency,2,stale/low urgency,1,0,131772,,,,0,b5ad39d566d741065d6d2e09915c27f6db1ed653a78c4833d20975242ef0fcdc
thread:issue:131774,issue #131774: Different Cholesky results between Windows & Linux,stale/low urgency,2,stale/low urgency,1,0,131774,,,,0,1364fc30f37e15b2d82bd421f995f202fc050248e23c85fa841845bd5ad11888
thread:issue:131883,issue #131883: Iterating dataloader fails on sliced dataset,stale/low urgency,2,stale/low urgency,1,0,131883,,,,0,7b28866f11096f386716007dae7f9bbb3a1cd486547ba3641aaf5dd01df94e57
thread:issue:131931,issue #131931: Start processes in _MultiProcessingDataLoaderIter on a ThreadPool,stale/low urgency,2,stale/low urgency,1,0,131931,,,,0,b936bb1e103e0e12bcf4357a083a2f1da1466d14530bd3a4a4f694c3140c1c51
thread:issue:131947,issue #131947: `__torch_function__` does not work for functions that called within other overrided functions,stale/low urgency,2,stale/low urgency,1,0,131947,,,,0,303f127e2e2f9459113fb8c3ec4b70b8cac4296f896065230a24e23f95b063c9
thread:issue:131962,issue #131962: Distributed tests failing on Amazon2023 AMI,stale/low urgency,2,stale/low urgency,1,0,131962,,,,0,053d24d6572b1cf92d61a66a97629fb1d2c58a2c8780af7d6285195e36f92f20
thread:issue:131967,issue #131967: Layer normalization on Nested Tensor ragged dimension fails when `lengths is not None`,stale/low urgency,2,stale/low urgency,1,0,131967,,,,0,07ff69f287f767e21a7637b0130a09d999997344ca3f39ce3eca71afc68fab13
thread:issue:132013,issue #132013: failed to build v2.3.0 docker image,stale/low urgency,2,stale/low urgency,1,0,132013,,,,0,513b7bfb0ca4425a91b9ef0aced951337bc0695aaf8b04923f3a152cdc24b384
thread:issue:132024,issue #132024: Refactor serialization IO infrastructure and support HDFS/HTTP,stale/low urgency,2,stale/low urgency,1,0,132024,,,,0,cc3483dcb95b4051e02ecca90b23905bd0d23ed2cdb635f28ef0d6e9d517fa20
thread:issue:132026,"issue #132026: Incompatibility with Numpy,JAX and MLX for int and bool type inputs",stale/low urgency,2,stale/low urgency,1,0,132026,,,,0,5fe80c739a96c0a40e065fc64d9a05a406010302a916d2a385f69dddf4864216
thread:issue:132028,issue #132028: torch.sspaddmm: self.is_sparse() INTERNAL ASSERT FAILED on torch 2.5.0.dev20240708+cu121,stale/low urgency,2,stale/low urgency,1,0,132028,,,,0,f3afe126e175cc423d25db9a3b1603d4964a268246e1e97e60141204f79da02b
thread:issue:132034,"issue #132034: Implementation of a Q-RNN layer nn.QRNN(filters,kernel,stride,n_cnn,)",stale/low urgency,2,stale/low urgency,1,0,132034,,,,0,c2c49e4f0e38d3ce0d3637b3facc1fae4821d39338c2fc639bac0f73231bd0fe
thread:issue:132052,issue #132052: Make C++ torch::autograd::Function support functorch,stale/low urgency,2,stale/low urgency,1,0,132052,,,,0,63f0d65aa945d401aa643182dc7f3adfcef5725f4f3f904751c58959451a6daa
thread:issue:132067,issue #132067: Pickle loads torch tensor memory leak,stale/low urgency,2,stale/low urgency,1,0,132067,,,,0,1309dd16db3593ee90a2c5d0d91d751d5283f4b7f32d4890bd050d375803a155
thread:issue:13208,issue #13208: [caffe2] How to use the operators that are not included in brew through the python API?,stale/low urgency,2,stale/low urgency,1,0,13208,,,,0,174a733d1201b3e6760047900da55c1ba0eefcba7bf36f440941dc5417abb544
thread:issue:132130,issue #132130: BuildExtension breaks build because of absolute path,stale/low urgency,2,stale/low urgency,1,0,132130,,,,0,8a4e5e09879b0376d54e96db77d82cf56000771ff159df09f08b9bbb55bd5d43
thread:issue:132132,issue #132132: Tests stopped working after merging #126905 on Arm64,stale/low urgency,2,stale/low urgency,1,0,132132,,,,0,cfb76f3982604dd911b83f6c02f6d0500e13f6e832bf8410c4ddd21ad0235432
thread:issue:132136,issue #132136: `torch.nn.transformer.forward` returns incorrect value inside `torch.no_grad()` blocks.,stale/low urgency,2,stale/low urgency,1,0,132136,,,,0,a8fcbfdf27ccb921766575e3a6d8ed43220a4ba397254b923ab5938bc20852f1
thread:issue:132218,issue #132218: [profiler] CUDA runtime op has wrong device time because of lazy init,stale/low urgency,2,stale/low urgency,1,0,132218,,,,0,3436edb40fc02ac42205ac5695370fd0a06a6a5a9abac99e410c3ef11aef4cba
thread:issue:132231,issue #132231: torch.nn.Bilinear weights initialization reason,stale/low urgency,2,stale/low urgency,1,0,132231,,,,0,a2bcd25c819f4eddcc2a9995f41c132b3216e530ed28a665c6507b507a347c43
thread:issue:13224,issue #13224: Caffe2: Causes error when using flag remove_legacy_pad while converting from caffe to caffe2,stale/low urgency,2,stale/low urgency,1,0,13224,,,,0,f0efd32e902611bc92ce85f578cffae968e289b1a8f11f0e44c455b6da181112
thread:issue:13226,issue #13226: [Caffe2] cmake3 detection error?,stale/low urgency,2,stale/low urgency,1,0,13226,,,,0,3b4fa0af12b9c776c63ef67a12d57bf2d436a20c3377d0e3ea2ea905f58ed5ae
thread:issue:132292,issue #132292: Can we avoid RecordStream for P2P ops?,stale/low urgency,2,stale/low urgency,1,0,132292,,,,0,7dd51c63773ff668e17083b065cfa143ea8186f63af2e0657d091bb57afaa3c8
thread:issue:132300,issue #132300: CUDA Memory Explosion -- Conv3D,stale/low urgency,2,stale/low urgency,1,0,132300,,,,0,5b8649c288162d4f81e7731fbb07f6856d0f4b5248e0ee06cc90d5d5f7562404
thread:issue:13234,issue #13234: torch.utils.checkpoint is not compatible with nn.DataParallel,stale/low urgency,2,stale/low urgency,1,0,13234,,,,0,e511935f8b707f0f858c3e01630b1c6a6f73828707a5a546d84f3ff428c48a3c
thread:issue:132386,issue #132386: torch cpu float16 range is not aligned with scipy on polygamma.,stale/low urgency,2,stale/low urgency,1,0,132386,,,,0,be13da277120a81c87ade4321768b2efccf7079ed4f13d7be556b08b52aa6450
thread:issue:132388,issue #132388: `amp.custom_fwd` interacts poorly with custom `autograd.Function` with `setup_context` separated from `forward`,stale/low urgency,2,stale/low urgency,1,0,132388,,,,0,320d6129528e8ab82561a85bbd889fc65e11cb42a55ba4b369b8f572ae617ff1
thread:issue:132395,issue #132395: `torch.multinomial` generates incorrect distribution,stale/low urgency,2,stale/low urgency,1,0,132395,,,,0,6d66441dbba408a81ce0adf80bc36be7059c2d3bef594da064b38acaeabd80fb
thread:issue:132412,issue #132412: [CI] PyTorch eager is slow in pytorch-linux-jammy-aarch64-py3.10-gcc11-inductor-benchmarks docker,stale/low urgency,2,stale/low urgency,1,0,132412,,,,0,fa93b8172319e6cfb1ce9243d131aa74a95cb10b750599bbcdd4950d3e90d898
thread:issue:132428,"issue #132428: MultiProcTest Exception ""None is not -6""",stale/low urgency,2,stale/low urgency,1,0,132428,,,,0,595a544cf3e515fad17bcc0bb0276174305a88f8c9467af3a43e9db905e22ce4
thread:issue:132483,issue #132483: Non-equivalent behaviour of deprecated set_default_tensor_type and set_default_dtype/set_default_device,stale/low urgency,2,stale/low urgency,1,0,132483,,,,0,24aeb50f139e0dfa8bc9363f7ea33368d1e8a5be96dd2a845aeb582c82f63f2c
thread:issue:132485,issue #132485: lazy_clone does not work for pinned memory,stale/low urgency,2,stale/low urgency,1,0,132485,,,,0,af1580038801f893101f010647f5e98aac74fe609bee2144d1d3fb9f3c91037d
thread:issue:132492,issue #132492: Wonder why _MultiProcessingDataLoaderIter.__next__ too slow？,stale/low urgency,2,stale/low urgency,1,0,132492,,,,0,b3d07a9e16bd5d9fe9d2838a9cd7a53b7197388eff79c1f4bd33307bf0c1a6f8
thread:issue:132534,issue #132534: triton import inside function doesn't work,stale/low urgency,2,stale/low urgency,1,0,132534,,,,0,37fefb36305b528ff36b5a6488dedcd106d1d301d7d34e1a69afbed1acdc1cc8
thread:issue:132579,issue #132579: Enable F16C instructions in CUTLASS? At least for the inductor backend?,stale/low urgency,2,stale/low urgency,1,0,132579,,,,0,be0899e46c2c58342124e889cb9b87f416e100f037d998ba1cc9871303608a11
thread:issue:132596,issue #132596: MPS backend leaks memory when input sizes vary,stale/low urgency,2,stale/low urgency,1,0,132596,,,,0,31f771320e4c9b656ae1efdd549a1bd642a40b7186952dfa2126a6150e08410e
thread:issue:132599,issue #132599: [BUG+FIX] LibTorch silently fails to find cublas_v2.h for NVHPC,stale/low urgency,2,stale/low urgency,1,0,132599,,,,0,03c7115be559dcd1e1c584e4c0f48bcdf1c541dacbb0a02c4d7761477c5c8407
thread:issue:132600,issue #132600: Inquiry about Test Error after Static Compilation,stale/low urgency,2,stale/low urgency,1,0,132600,,,,0,e6d7ebe3f2dbee5eb4fd2aae13920321c654f13de7d6a65c904709e3e1d474cc
thread:issue:132602,"issue #132602: Is torch.compile imprecise, as this 4-year-old comment in amp.grad_scaler claims?",stale/low urgency,2,stale/low urgency,1,0,132602,,,,0,31994e7745c6cd6dcba903f32a12a036964875f468d2eb13fe87d6602418ff98
thread:issue:132605,issue #132605: MPS lgamma function changes results when using broadcasting,stale/low urgency,2,stale/low urgency,1,0,132605,,,,0,21c092a733b1110e605276fbb4319de40a7ba95109a37f885e3c650037d7147f
thread:issue:132608,issue #132608: topk with mask - efficent,stale/low urgency,2,stale/low urgency,1,0,132608,,,,0,316568419b3d415014d86e6fc942f8dfb11e9cae296982d479f7f110f597f45c
thread:issue:132612,issue #132612: bitwise_xor seems to be not optimized,stale/low urgency,2,stale/low urgency,1,0,132612,,,,0,cd421a1ec930b54c86d41685fb9aeef1c8328a3236a190f60a0e776791f46b5d
thread:issue:132614,issue #132614: BUG: `torch.special.gammainc`: returns finite results with NaN input,stale/low urgency,2,stale/low urgency,1,0,132614,,,,0,6fdb455d6b9a06d33364aa59e46f7fde07834fa57dd367d64f5a0db73f155b10
thread:issue:132617,issue #132617: conda pytorch does not ship libnccl.so,stale/low urgency,2,stale/low urgency,1,0,132617,,,,0,be8ec36fb9b1e452faa7fa782d9d4d34f23c834167cbb94250450f1541d2915c
thread:issue:132624,issue #132624: Add support for float8 dtypes for the MPS backend,stale/low urgency,2,stale/low urgency,1,0,132624,,,,0,99be29d93e12eb497bccb4793a3f79ad67dd7f065b2eea63f1858ee2056f6748
thread:issue:132649,issue #132649: nn.Parallel,stale/low urgency,2,stale/low urgency,1,0,132649,,,,0,797d943ed32c187736ac5a3e302653f9e4a718e25d0922368f7c30405464cacf
thread:issue:13268,issue #13268: Non-Zero Padding in Convolution Module,stale/low urgency,2,stale/low urgency,1,0,13268,,,,0,a7eca28214caea90086fa31dd0d5a64c84fb604845035f1ef981d9654bb7ca18
thread:issue:132693,issue #132693: INTERNAL ASSERT FAILED in BatchLinearAlgebra.cpp,stale/low urgency,2,stale/low urgency,1,0,132693,,,,0,059395af2ed4c80ef0f31d22a9e274bf91b276ae7ff755d0dea4eb0e72658587
thread:issue:132695,issue #132695: Backward on nested jagged tensor vector multiplication throws error on 0 dim (but not last dim),stale/low urgency,2,stale/low urgency,1,0,132695,,,,0,941f6e4487875e628b78c5a8c5f054cd75003db4110d90d18be485c47f022832
thread:issue:132696,issue #132696: Hanging at waitForFutureOrTimeout when actively destroying ProcessGroups,stale/low urgency,2,stale/low urgency,1,0,132696,,,,0,83d49736a53d184f85c44d72a4fe4ff6c9ac3138df4839550a91400afd9a4a21
thread:issue:132700,issue #132700: Forward Hooks in Transformer Not Working When `model.eval()` and `torch.no_grad()`,stale/low urgency,2,stale/low urgency,1,0,132700,,,,0,0e6fb986e0babdd7fe86b64fbd1ea69bbf16ef63af20c62160fe25accbc9472d
thread:issue:132747,issue #132747: Why is kvcache transfer slow using peer-to-peer communication (two GPUs on one machine).,stale/low urgency,2,stale/low urgency,1,0,132747,,,,0,83f7aea5f0f0dde6cba52fb5b6a468a4a879137538683ae96420027ecb02e265
thread:issue:132759,issue #132759: Delete ProxyTorchDispatchMode.enable_tracing,stale/low urgency,2,stale/low urgency,1,0,132759,,,,0,9c800dadb06664d8bd946322c71bd8d701a3fc0939a491315ea6f8f411cd06da
thread:issue:1328,issue #1328: Unhelpful CrossEntropyLoss dimension error message,stale/low urgency,2,stale/low urgency,1,0,1328,,,,0,b16e7aeb83e166e59b0191854e0a41e49cfe915f18faa92d6e39f50e2089ed17
thread:issue:132856,issue #132856: Enhance the compiler module to use cudagraph with PrivateUse1,stale/low urgency,2,stale/low urgency,1,0,132856,,,,0,404cb21bcf5c7eb025945aab088b634e70769bafa1e14ed589482764d6367cfc
thread:issue:1329,issue #1329: [proposed feature] Eve: Improving Stochastic Gradient Descent with Feedback,stale/low urgency,2,stale/low urgency,1,0,1329,,,,0,b70cec804e8f2cfc22cb056c87d61b9f10e7fd29929f91bbb70984f523cf4236
thread:issue:132919,issue #132919: Add a warning for `drop_last` argument in `DataLoader`,stale/low urgency,2,stale/low urgency,1,0,132919,,,,0,f58433673baa967d80384451ebc9fc7ced01a293a8c8ad282b174a5b34e07cac
thread:issue:132940,issue #132940: Enabling GPU mixed precision algorithms,stale/low urgency,2,stale/low urgency,1,0,132940,,,,0,1b4085d994a13c5ab845968885e4df39cc712b19a884da266779fa75b301f466
thread:issue:132951,issue #132951: torchelastic multiprocessing start_processes documentation out of date,stale/low urgency,2,stale/low urgency,1,0,132951,,,,0,a94f7da1e432ba7cdc5d5d9b412ffbce0e73cfab806a2c030f46567c4ef68804
thread:issue:13297,issue #13297: [Caffe2]How to accelerate group convolution in Caffe2?,stale/low urgency,2,stale/low urgency,1,0,13297,,,,0,c089712c05e1e51c6acd61143fc1632eb4c63a912a869fde285b61488b70817a
thread:issue:132991,"issue #132991: _pickle.UnpicklingError: invalid load key, '\x00'. during torch load",stale/low urgency,2,stale/low urgency,1,0,132991,,,,0,11d3bcd7e42a611a6eb3e8f00f3aa22ec5b1eb4559bd689b0b2f4b42c9e58e02
thread:issue:13300,issue #13300: Use a dill-based multiprocessing library and serialization,stale/low urgency,2,stale/low urgency,1,0,13300,,,,0,8c6a3536e3fa03cd75f745f02e76a3e9f16660fae7d130267b4a6c28172f4974
thread:issue:133010,issue #133010: torch.multiprocessing.start_processes is blocking with large input arguments,stale/low urgency,2,stale/low urgency,1,0,133010,,,,0,8a8e5c3626242a26643a703747d33bb353046332d8e916343b1c1c16ac30d95e
thread:issue:13304,issue #13304: ASSERT FAILED at /opt/conda/conda-bld/pytorch-nightly_1539602533843/work/aten/src/ATen/core/blob.h:79,stale/low urgency,2,stale/low urgency,1,0,13304,,,,0,7ab16ead3c19847acbc11c555c6d1a15c02270350cff548486998ef4620a176e
thread:issue:133040,issue #133040: Add Test for Inductor and Dynamo Config BC Breakages,stale/low urgency,2,stale/low urgency,1,0,133040,,,,0,9ae8d6360e8097558dee6764b0bdb2ce427053f3bdb9ac9812e3d4600dc3a7a7
thread:issue:133084,issue #133084: Addition of Visual Information Fidelity loss,stale/low urgency,2,stale/low urgency,1,0,133084,,,,0,bba9ef52ad8d9e273b21429c8ce9936d53ff4b92e40b5283407df93b0a8f9c9c
thread:issue:133092,issue #133092: Umbrella issue for Dynamo tracing through ProcessGroup related torch.distributed APIs,stale/low urgency,2,stale/low urgency,1,0,133092,,,,0,7d5a1f1b074c75f305d856afd551e8656129433783bb989afe6cf8cc632f8b80
thread:issue:133100,issue #133100: MPS clamp_min on older MacOS,stale/low urgency,2,stale/low urgency,1,0,133100,,,,0,5f45bfe02d006a5bd934577942f948ee734d21eae7e3e89f678d8315a3aa45df
thread:issue:133114,issue #133114: Conv1D: `CUDNN_BACKEND_OPERATION: cudnnFinalize Failed cudnn_status: CUDNN_STATUS_BAD_PARAM`,stale/low urgency,2,stale/low urgency,1,0,133114,,,,0,fc7aac6fdd0deb09c611ebc95e160fe94c1b8723ef3437ccc2f2b64d818d7e55
thread:issue:133117,"issue #133117: [dynamo] Improve support for subclasses of popular builtin types like dict, list, set",stale/low urgency,2,stale/low urgency,1,0,133117,,,,0,2a4be4a60602ba065e144ececfc510c67641c3e6a5d244a4ca0ef5cf68f437dd
thread:issue:133119,issue #133119: MultiProcContinuousTest doesn't work with pytest/buck,stale/low urgency,2,stale/low urgency,1,0,133119,,,,0,aef64106f530248680c883974277f84f2fe10e1c4acbc633ee3f78c3f1c0536e
thread:issue:133138,issue #133138: Layer normalization for Nested Tensors does not return the correct tensor types,stale/low urgency,2,stale/low urgency,1,0,133138,,,,0,3f7b34bded4e0b93280e508c4ad8632fa40290a202ffc11358a807acfeb95292
thread:issue:133183,issue #133183: convolution_overrideable not implemented when executing VAEEncode in ComfyUI,stale/low urgency,2,stale/low urgency,1,0,133183,,,,0,33cf5f77c18deea535db023d21998d887ab4a6b4235d8874b7e6cf3f1bf20e11
thread:issue:133210,issue #133210: IPC feature supports third-party devices,stale/low urgency,2,stale/low urgency,1,0,133210,,,,0,c8ab1ed7ec0cb94711ee95f20369a575621136afdcfb312ddc2a19caea2a4176
thread:issue:133221,issue #133221: Tensor parallel for convolutions and groupnorm,stale/low urgency,2,stale/low urgency,1,0,133221,,,,0,6ca0b89dbb71b64c8855520f438a0822d5bfb6d4beb2ee6cbfc8f60bb463480a
thread:issue:133224,"issue #133224: When using parallelize_module to do tp parallel and pp parallel mixing, the SendRecv time is too long",stale/low urgency,2,stale/low urgency,1,0,133224,,,,0,ee200e1dea597a555671e0f3dd55bbd343b5e92c9f41c47745342882e4b0a7cd
thread:issue:133262,issue #133262: export error when tracing tensor subclass tensors: `'FakeTensor' object has no attribute 'get_plain'`,stale/low urgency,2,stale/low urgency,1,0,133262,,,,0,776db9144ce8efd464d366d709dee8a3f652587ebed577aa1ef7aa8ce1598b5c
thread:issue:133268,issue #133268: Adafactor compile support tracker,stale/low urgency,2,stale/low urgency,1,0,133268,,,,0,f7c12db2f02796696c7ae7448886527e6294833d3ccb05b20e9558bea6349205
thread:issue:133280,issue #133280: Pluggable Allocator Does not raise torch.cuda.OutOfMemoryError,stale/low urgency,2,stale/low urgency,1,0,133280,,,,0,8d04c7f87d6d1b351374ed931f6b5d631063c0587bb993af6ed6a4ac89f80a4c
thread:issue:133281,"issue #133281: Make CUDAPluggableAllocator support getDeviceStats, resetPeakStats",stale/low urgency,2,stale/low urgency,1,0,133281,,,,0,2855fa6988286f6c664d8ed4f1527466272947075def54f5e50d2809c164eca7
thread:issue:133304,issue #133304: The performance regression of `conv_transpose1d` on CPU with variable-shaped inputs,stale/low urgency,2,stale/low urgency,1,0,133304,,,,0,ffed88ea8396482ccebecaaf191584c8f983d1e29c989fb66ec27beaf12ec314
thread:issue:133380,issue #133380: AttributeError: 'NoneType' object has no attribute 'is_failed',stale/low urgency,2,stale/low urgency,1,0,133380,,,,0,f82e7e35c9c465acde6f5e93b60d0614b0aeb375007fe191597d0a3300648d98
thread:issue:133383,issue #133383: [Inductor] Remove dead code from triton kernels,stale/low urgency,2,stale/low urgency,1,0,133383,,,,0,b08531c0afddab5ff51d608b5f69ed70cedd4b16bf5ef7f88f7b629575efd67c
thread:issue:133453,issue #133453: [randperm] Add argument to disallow identiy mappings,stale/low urgency,2,stale/low urgency,1,0,133453,,,,0,6f2c1a2f772ba881a41302a63ba2a4fac5cd8d6825716b48a93e7b3dc9120fc9
thread:issue:133458,issue #133458: [CPU] jx_nest_base AMP both inductor and eager performance regression in 2024-08-10 nightly release,stale/low urgency,2,stale/low urgency,1,0,133458,,,,0,96cf22b91a1f99f3e15f387d26b544248139334508cadd4c0a26466b75b377b9
thread:issue:133522,"issue #133522: `init` of `PReLU()` works with `int`, `complex` and `bool` against what the doc says.",stale/low urgency,2,stale/low urgency,1,0,133522,,,,0,6dca6544256d8c034c24f5379f9e9d162c80cd29e44fbf32ebcb6518f1cc21b6
thread:issue:133527,issue #133527: Using `PReLU()` with a `complex` tensor and without `dtype=torch.complex64` should return a simple error message,stale/low urgency,2,stale/low urgency,1,0,133527,,,,0,c88027954a3acbd9af7be12fcea6b615067c3eff2ef26d3dca270dc326be2c35
thread:issue:133535,"issue #133535: When using a wrong value for `PReLU()`, a PReLU's error message should be returned directly instead of `empty()` error.",stale/low urgency,2,stale/low urgency,1,0,133535,,,,0,39402c2f1ae4d7df7a3855b2e008ccf520d72d73c55bc269b72699b953ab0eb9
thread:issue:133564,issue #133564: torch.cond example fails in PyTorch2.4,stale/low urgency,2,stale/low urgency,1,0,133564,,,,0,034701554c8cc7f33059b4dec5759c2fa828c0123490b2d263f8128a6ee43385
thread:issue:133586,issue #133586: No way for low-overhead total norm in native PyTorch with large number of tensors,stale/low urgency,2,stale/low urgency,1,0,133586,,,,0,360f991b74a6d6bd1495068d91bd7f71760f9b6616cb187bb509aaa2dcab9df2
thread:issue:133629,issue #133629: RuntimeError: free_upper_bound + pytorch_used_bytes[device],stale/low urgency,2,stale/low urgency,1,0,133629,,,,0,bb406281fd45bb9165e1f324f9949796547bbed2091d202b1289abcfc1b22dae
thread:issue:133665,issue #133665: [PixelShuffle] PixelShuffle doesn't support channels_last feature for 5D inputs,stale/low urgency,2,stale/low urgency,1,0,133665,,,,0,d7c6187dffb681cdde9090c0639bcb6b4af709b607d0ce9fb449654c844587fd
thread:issue:133689,issue #133689: [torchbind x PT2] Handle vllm's ScalarType,stale/low urgency,2,stale/low urgency,1,0,133689,,,,0,59e6191618eec888f78328424b04e6b3881ea1639d3d1f1b31d88a7d906b5e67
thread:issue:13373,issue #13373: cannot run mobilenet_v2_quantized on pytorch/caffe2,stale/low urgency,2,stale/low urgency,1,0,13373,,,,0,37e006df2f64e2bd8653894ffbbd49e0bd8c323bb2b18d3dfa50d0dbfb6c3b05
thread:issue:133751,"issue #133751: `torch.nn.LSTM` using cuDNN fails with sequences longer than 65,535",stale/low urgency,2,stale/low urgency,1,0,133751,,,,0,f81372a4893cf722546081b1f681f25d78c43149283a2346ca0beedb0841db19
thread:issue:133754,issue #133754: sample of Binomial should return int instead of float,stale/low urgency,2,stale/low urgency,1,0,133754,,,,0,2a7321ff7a3538cbacfcdbdbf15dba24fd4a5b256e0a21ead07423843bb0d2b1
thread:issue:133761,issue #133761: Incorrect GPU management and deadlocks without torch.cuda.set_device,stale/low urgency,2,stale/low urgency,1,0,133761,,,,0,a174ee7c568e9f63054ee9a7b43bb42c65c6ad012f736a9bd96d1eb3dc089071
thread:issue:133780,issue #133780: 2.4.0: sdpa disallows batch-of-0 in Flash (default backend) and cuDNN.,stale/low urgency,2,stale/low urgency,1,0,133780,,,,0,d3fd25e65130265b2225dbb3555fcd47bff6c96c487aa18e6c9b4b8f7faedc6d
thread:issue:133818,"issue #133818: Warning when initialize `torch.nn.Linear(0, d)`",stale/low urgency,2,stale/low urgency,1,0,133818,,,,0,fc899d47a2e379c558b839314448fe9e005d8c7876f17bcc0b4aaba09244e7a1
thread:issue:133833,issue #133833: Unexpected behavior of `CrossEntropyLoss()`,stale/low urgency,2,stale/low urgency,1,0,133833,,,,0,18ee3fed079f961128e34d8c3d6299702e60205ce6422735982ec166cc4ce617
thread:issue:133849,issue #133849: CatArrayBatchedCopy and AllGather don't overlap during FSDP backward,stale/low urgency,2,stale/low urgency,1,0,133849,,,,0,2e3e48401b5b42ba87473f60eab3e23297cfcbcb84ad4c781e6358a3d1c30e56
thread:issue:133941,"issue #133941: Why core aten IR doesn’t contain scalar_tensor overload version of some ops, like bitwise-like ops, remainder and so on?",stale/low urgency,2,stale/low urgency,1,0,133941,,,,0,0081711e5de848fe15d3fc45dad2edbcc3e795ebe08c83fe7504434c2728edcf
thread:issue:133944,"issue #133944: torch.compile ""run at compile time"" decorator.",stale/low urgency,2,stale/low urgency,1,0,133944,,,,0,42e8d596cc547cdad8454614cfaf2603baec4cdd2f3db0369b694386bc71274b
thread:issue:133984,issue #133984: cudagraph capture p2p ops fail,stale/low urgency,2,stale/low urgency,1,0,133984,,,,0,75b0f3780ab7e3684a8be1802ca23d45a4affdde765c0a6b95773cbbad4f9eed
thread:issue:133985,issue #133985: Feat: output dtype for count_nonzero,stale/low urgency,2,stale/low urgency,1,0,133985,,,,0,f3f111346308adcbad822e026835cd4dbf834bb171f4e69709a44ce02b6fefc6
thread:issue:133998,issue #133998: `torch.quantile`: `nan` instead of `inf`,stale/low urgency,2,stale/low urgency,1,0,133998,,,,0,24ea683da67bbf0e4bd5396577de84147b4ef45c306f0a92a72b86390b229e00
thread:issue:134038,issue #134038: vmap tests aren't erroring on fallback usage,stale/low urgency,2,stale/low urgency,1,0,134038,,,,0,06ca93a7a28dc608dc93e671b5d65ce906b76fd271f7ec104bc0d824983d0d5d
thread:issue:13405,issue #13405: [caffe2] How to use Split operator?,stale/low urgency,2,stale/low urgency,1,0,13405,,,,0,9689e24df99808cbd423b150f51e1a77900b3141c589e7ce89d0b24feeb79126
thread:issue:134051,issue #134051: FSDP activation CPU offload memory not cleaned up,stale/low urgency,2,stale/low urgency,1,0,134051,,,,0,9b8032c61f42cff4703319f3bd1fa8cd09cf0846129ca04fe09f13a1450ff7be
thread:issue:134067,issue #134067: subprocess tensor creation gets stuck,stale/low urgency,2,stale/low urgency,1,0,134067,,,,0,176979b3d877a9ab25469bdff6aa1e98c2ca090a3f97727dbf5b2ec5faf80995
thread:issue:134069,issue #134069: Kron with gradient for sparse tensors,stale/low urgency,2,stale/low urgency,1,0,134069,,,,0,8517e7d4985b4e28b08781b9ba506deeea2f2c78a750c5f8fe2cd901c7619275
thread:issue:134079,issue #134079: operator benchmark diag test cuda problem,stale/low urgency,2,stale/low urgency,1,0,134079,,,,0,7ec465519f49365cd585503326a3a9b056d7e79980b010345795701369a467e8
thread:issue:134098,issue #134098: libtorch reasoned that the crnn model was out of speed the first few times,stale/low urgency,2,stale/low urgency,1,0,134098,,,,0,39e18f9b4af710ae821b65c6b358a6ecaeadc3ed3851059b451ebe42d503c668
thread:issue:134180,issue #134180: [export] Incorrect behavior with torch.LongTensor and torch.Tensor,stale/low urgency,2,stale/low urgency,1,0,134180,,,,0,b370c710aa0afcbd834befe90c07edb18686c314abd1c94ca131cf171a97aec7
thread:issue:134183,issue #134183: storage_module && PyModule_Check(storage_module) INTERNAL ASSERT FAILED at DynamicTypes.cpp,stale/low urgency,2,stale/low urgency,1,0,134183,,,,0,705122934fc75b0f3d149276c11bd547cadc219a4795d60e609392bbc09a1a35
thread:issue:134190,issue #134190: ProcessGroupNCCL::WorkNCCL only records seq_id which is not enough to be matched with comm op in Execution Trace replay,stale/low urgency,2,stale/low urgency,1,0,134190,,,,0,35c6c8d9852ca675be85ed481892c15c006e04682001447d92b4d6771ad0e4ad
thread:issue:134201,issue #134201: Python310 trying to run GFPGAN comes: ImportError: cannot import name 'sym_float' from 'torch' (unknown location),stale/low urgency,2,stale/low urgency,1,0,134201,,,,0,1e3b8584f5d25134ef98b87ff9a3392290cc684b500ed97adb730beb44fd069c
thread:issue:134236,issue #134236: [CI] Binary Test Workflows are failing when updating triton hash,stale/low urgency,2,stale/low urgency,1,0,134236,,,,0,f05fc263676897f614158bec00b604dd9a5eaccf4f427ad85f8180939de787c8
thread:issue:134244,issue #134244: Setting a 1D tensor as `input` and a 0D tensor as `index` and vice versa for `gather()` works against the error message,stale/low urgency,2,stale/low urgency,1,0,134244,,,,0,1d7d62613f947fddd43f32e760de8483b70eef5391db168a21fa93b1d0ce5962
thread:issue:134249,issue #134249: Footgun: Dynamo x tensors created inside or outside of inference_mode,stale/low urgency,2,stale/low urgency,1,0,134249,,,,0,d31b2153efab7ec47b3eedc0f96621a9e28dfd0ff13a4f9847a4d59628207981
thread:issue:134293,issue #134293: [PrivateUse1 Backend] Unable to register backend to PrivateUse1 dispatch key: ModuleNotFoundError: No module named 'torch.privateuseone',stale/low urgency,2,stale/low urgency,1,0,134293,,,,0,f13540bbc4adf91319cb3a819233c25e89dfc75e5d481faffe1811219fe67ea1
thread:issue:134326,issue #134326: Very large memory allocation by torch.linalg.qr,stale/low urgency,2,stale/low urgency,1,0,134326,,,,0,56e6b48ca8e964338e6c365093d1436e5d6bdcd5a756ed042bc22d20b78cef42
thread:issue:134388,issue #134388: Remove redundant copy in functional collectives,stale/low urgency,2,stale/low urgency,1,0,134388,,,,0,63f33a508dd2ecb3501c232c1c98a2caf4edf5e5b8d1161348e7a114875416e4
thread:issue:134396,issue #134396: torch.sparse.softmax missing support for sparse CSR,stale/low urgency,2,stale/low urgency,1,0,134396,,,,0,1e1a3afa9dfcbdc2b58103f8af46739ed8a5b1d26f268a0d7ce957fd0dce7331
thread:issue:134399,issue #134399: Build errors while building PyTorch with BLIS,stale/low urgency,2,stale/low urgency,1,0,134399,,,,0,d1f4e70fde3b9eedb63ae7ad934d5431d833486fbe793da53b283b7801a3c02d
thread:issue:134403,issue #134403: RuntimeError: cuFFT error: CUFFT_INTERNAL_ERROR,stale/low urgency,2,stale/low urgency,1,0,134403,,,,0,517ef9878efb51b571aa679e479faf0dacc682533f5eec42a4d36ea56fb29f27
thread:issue:134430,"issue #134430: [MPS] `F.interpolate` with `mode=""nearest-exact""` differs from expected output",stale/low urgency,2,stale/low urgency,1,0,134430,,,,0,3efd463e905021eec6e1dd47a604b8cb2059decdaeb4f12d1561c1f218885434
thread:issue:134439,issue #134439: configure ignores USE_QNNPACK=0 and still looks for third_party/QNNPACK,stale/low urgency,2,stale/low urgency,1,0,134439,,,,0,0e780a997a6c0f4230963e4bd964cb3b0208c667e6d6502d01fd0072e40d1fcd
thread:issue:134441,issue #134441: Cuda memory error when testing GroupedGemm on a multi-Gpu environment,stale/low urgency,2,stale/low urgency,1,0,134441,,,,0,862ba3860571d2a3a83778b80b1f5cc6bd7fa11e3b79288f5a0e143c83459ba4
thread:issue:134459,issue #134459: linalg.lu_factor: LU without pivoting is not implemented on the CPU,stale/low urgency,2,stale/low urgency,1,0,134459,,,,0,4ce495f10547569603736a8178974f675d982073105397919e896e58ecfe6b0d
thread:issue:134462,issue #134462: Cannot create and distribute array in torch.func.grad,stale/low urgency,2,stale/low urgency,1,0,134462,,,,0,21bb53d97da27c59c374458d2355902f74ce1f65ad13a3bd12beac3cc9205eae
thread:issue:134534,issue #134534: MiniCPM-V 2.6 transformer model returns nonsense tokens when running on MPS backend vs CPU backend.,stale/low urgency,2,stale/low urgency,1,0,134534,,,,0,ff43ef1e528762f8ebeccd327786aed41d62df980a5c248f7a8fcc95aa83165f
thread:issue:134576,issue #134576: Would it be possible to have a CMAKE option USE_SYSTEM_FMT,stale/low urgency,2,stale/low urgency,1,0,134576,,,,0,170700da59eb734dd93d5f7fc5cdf077a5453f9481942cc7f3a05de1a0d3cd7e
thread:issue:134607,issue #134607: torch.angle cuda version not implemented for bfloat16 and half (float16),stale/low urgency,2,stale/low urgency,1,0,134607,,,,0,624be626a321769d59f231fe109b6a301a1f9f0358518d511156d6a8ce9e864f
thread:issue:134612,issue #134612: torch.func.grad uses more memory than expected,stale/low urgency,2,stale/low urgency,1,0,134612,,,,0,cec6c9b64d2d3fe6c84a6f2876f74c7b447b4bbf99d327acfb563226d1949145
thread:issue:134637,issue #134637: LSTM inference c++ threads block on DropoutState,stale/low urgency,2,stale/low urgency,1,0,134637,,,,0,416ab519f1d625a118e48511be04e3b7e8b0632a705cd73824485f14b59834cc
thread:issue:134641,issue #134641: nn.Module.to(memory_format= channels_last format) failed if containing 5D parameters,stale/low urgency,2,stale/low urgency,1,0,134641,,,,0,0b7ad4927b93257cefe355c730b059b9bfb3314b6cbf2d45c2ddecfa910d327a
thread:issue:134664,"issue #134664: Questions about CVE-2022-3171, CVE-2022-3509 and CVE-2022-3510",stale/low urgency,2,stale/low urgency,1,0,134664,,,,0,11fec3025887e6aef0b3cee5ba0358bb137ac4b17887faace4d1d6eb046350d2
thread:issue:134668,"issue #134668: Whether tensor parallelism supports the overlap of communication calculations for gradient computation, and how to implement it",stale/low urgency,2,stale/low urgency,1,0,134668,,,,0,dbceb09418b984588f82f53a2c16b78041c1e5b4f48ea5cadade3fabb6b1d7de
thread:issue:134680,issue #134680: Use `torch._C._stash_obj_in_tls` for global state in serialization,stale/low urgency,2,stale/low urgency,1,0,134680,,,,0,c6abff32a9d028794593d5bc937428203a80df95e48f41932411bba3a0d9a9d0
thread:issue:134682,issue #134682: RuntimeError: cuDNN version incompatibility: happened when I have lstm layer,stale/low urgency,2,stale/low urgency,1,0,134682,,,,0,71d34107865291a123b09e0f522ee56549facc0d1b121c120c43d4b4787ff9fb
thread:issue:134691,issue #134691: [docs] We should add a lint for bullet points and multiple lines,stale/low urgency,2,stale/low urgency,1,0,134691,,,,0,9217c41d4cb6d799e8d5e40c1a5f85e2dbcb070fc01c6d880f6bcd956d108118
thread:issue:134700,"issue #134700: [c10d][MPI] Attempting to create a new group after MPI causes ""RuntimeError: Underlying Non-PrefixStore shouldn't be null""",stale/low urgency,2,stale/low urgency,1,0,134700,,,,0,c6530c49bc7a9655c67890bcdf41875aa273942d06197a60d3efade30ff119a9
thread:issue:134755,"issue #134755: ""Attempted to send CUDA tensor received from another process"" even though that was not the case",stale/low urgency,2,stale/low urgency,1,0,134755,,,,0,56519aeea27a5c5e7586573f05785ecb618381826af2e220dabbaeb80e3e527a
thread:issue:134803,issue #134803: enhanced debugging prints in torch.distributed,stale/low urgency,2,stale/low urgency,1,0,134803,,,,0,925212bdb08994a970cf0a02b94bd492930607223a7b83f3836dfc36d1cc857e
thread:issue:134821,issue #134821: Add decompositions for nanmedian and median,stale/low urgency,2,stale/low urgency,1,0,134821,,,,0,6af85a5d4492c2b094f4322c7f6a93237699703a0b6a18ef7aaefad29dd987a0
thread:issue:134827,issue #134827: ONNX NMS Operator Output Different From Torchvision Implementation,stale/low urgency,2,stale/low urgency,1,0,134827,,,,0,6db6e90beac2029712c5951da0e30d769eff7a3db94141cf6e33c1ef3d94b035
thread:issue:134851,issue #134851: GPU Precision drops significantly when kernel size is increased from 3 to 5 in Conv3D,stale/low urgency,2,stale/low urgency,1,0,134851,,,,0,da154fa223421fcb731b76d8eab48c1283e04b9297de50e4a00c87dfad0d9007
thread:issue:134860,issue #134860: JIT tracing a quantized model with hooks is broken,stale/low urgency,2,stale/low urgency,1,0,134860,,,,0,4c57045483f4a15fae58a05e43c3617773cd5be9934db655f69c564c6c52c768
thread:issue:134901,issue #134901: How to calculate second derivative using PyTorch with GPU (cuda),stale/low urgency,2,stale/low urgency,1,0,134901,,,,0,d88c5fe172777e1bbc5e5e146de7f6b68a9708165fe18f5d702a2a1969116223
thread:issue:134926,issue #134926: register_after_fork fails on python built without multiprocessing support,stale/low urgency,2,stale/low urgency,1,0,134926,,,,0,cfc5fd83503313c55c2d44687668f068cdc5e836a335edd6a4b7b60b7c89a30a
thread:issue:13493,issue #13493: How to do inference with float16 (or half) in caffe2?,stale/low urgency,2,stale/low urgency,1,0,13493,,,,0,eeaecb1d2aad145976397615a25152914ebf50ac9ccbd92884708615e6acb3b2
thread:issue:134945,issue #134945: Broadcasting Parameter destroys hooks in PyTorch 2.4,stale/low urgency,2,stale/low urgency,1,0,134945,,,,0,67c182b4c98559ff09a8bee176a938666222437831042365b87c4ad842ee17a3
thread:issue:134947,issue #134947: torch.hub.load will raise RuntimeError: operator torchvision::nms does not exist,stale/low urgency,2,stale/low urgency,1,0,134947,,,,0,2ee18f77146ff40efab333fa5a48afb7e2b6dca4d8da1f8ebf45b094fefef345
thread:issue:134948,issue #134948: Add Validity Checks for Parameter Combinations and Input Shapes in torch.nn.MaxPool2d to Prevent Unexpected Outputs,stale/low urgency,2,stale/low urgency,1,0,134948,,,,0,8fca468303672b78730d82cc305d1f2948759c8e2ec98a756fcd3d98c0520d9e
thread:issue:134953,issue #134953: _rebuild_sparse_tensor,stale/low urgency,2,stale/low urgency,1,0,134953,,,,0,70f2a28710d31a3fc7f1a2a220c30f49f57fddcd1407c138d5efed2dbe564078
thread:issue:134960,issue #134960: Issue with Weight Synchronization When Using Consecutive DDP Instances in the Same torch.distributed Environment,stale/low urgency,2,stale/low urgency,1,0,134960,,,,0,db57262274232ee764eabcc5644446ec6e463f01de0a35092704d5be9f5e46e2
thread:issue:134977,issue #134977: [BUG] found a weird problem of nn.conv2d in pytorch 2.2.0,stale/low urgency,2,stale/low urgency,1,0,134977,,,,0,268a3f6ec4c951122583d31aff5af126165c1e49da9cf666af4b67a739dab6ed
thread:issue:134978,issue #134978: [RFC] A device-agnostic Python device memory related API design for stream-based accelerators,stale/low urgency,2,stale/low urgency,1,0,134978,,,,0,f56180082528a04ce17cc05f205122acb5a1e45898f658070bd85bc2c4f6d2d9
thread:issue:13498,issue #13498: Caffe2 missing headers in docker/conda,stale/low urgency,2,stale/low urgency,1,0,13498,,,,0,515f308a3f4777345b9a0ebd9fd26086fa10b33115adba514f906d7effcf8a90
thread:issue:134996,issue #134996: Setting CPU affinity doesn't work for torch.softmax when using a thread,stale/low urgency,2,stale/low urgency,1,0,134996,,,,0,9ded2b93dd1302368c4c0c3c9374c75d206455182782571ddf45f089e8992701
thread:issue:135048,issue #135048: Setting a `bool` tensor both to `min` and `max` argument of `clamp()` gets error while setting a `bool` tensor only to `min` or `max` argument of `clamp()` works,stale/low urgency,2,stale/low urgency,1,0,135048,,,,0,d0ed2e2b8ac9b5a5daa4618755ac1a384410c2767c661386e22a652452a3442e
thread:issue:13505,issue #13505: [feature request] Singular values and spectral norm for convolutional layers,stale/low urgency,2,stale/low urgency,1,0,13505,,,,0,3465415f61ee054387465fa262b66c465addc82ca93e8412e0d5aa407428333b
thread:issue:135050,issue #135050: Setting a scalar and a 0D tensor or a 0D tensor and a scalar to `min` and `max` argument of `clamp()` respectively works,stale/low urgency,2,stale/low urgency,1,0,135050,,,,0,4c8c00393581531b6ff4a79e114187a47ed9ca762f1eb58376222ec75a4e6694
thread:issue:13508,issue #13508: OpenCL support for smartphones,stale/low urgency,2,stale/low urgency,1,0,13508,,,,0,9810442e7f87c0f058ba2da1b4503ecbb256506d22e74bb152ff9b380794c1ba
thread:issue:135091,issue #135091: Any support for sparse linear layer in the future?,stale/low urgency,2,stale/low urgency,1,0,135091,,,,0,a3cd7effee2a0b42d73efc7b42435dac4b228a6ccd8ca1f54fb7aa7ee3c3508b
thread:issue:135095,issue #135095: Add MSCCL++ as a communication backend for PyTorch,stale/low urgency,2,stale/low urgency,1,0,135095,,,,0,ce8806a7e2df1da1886e6d5fb05340c4e68fd85f8ed32d6c4988903981fc0993
thread:issue:135126,"issue #135126: ""RuntimeError: CUDA error: operation not supported"" fixed by downgrading toolkit version",stale/low urgency,2,stale/low urgency,1,0,135126,,,,0,1434ae5da805bae343f4e56543701d6406b63287f51d0467f7278908e555193c
thread:issue:135135,issue #135135: TORCH_CPP_LOG_LEVEL cannot enable VLOG,stale/low urgency,2,stale/low urgency,1,0,135135,,,,0,9167ab6f914ca8e9c6eadb02ac0b84f86e4081ddbea83b05304a28ac7803f25a
thread:issue:135150,issue #135150: Undocumented fast pass behavior in nn.TransformerEncoderLayer causes failures in test_transformerencoderlayer_cuda_float32 (__main__.TestNNDeviceTypeCUDA),stale/low urgency,2,stale/low urgency,1,0,135150,,,,0,5cf6944fd78025b2b37d0d09369d817de3dcb273102734f078edfe0023fbedad
thread:issue:135153,issue #135153: TestSDPAPrivateUse1Only tests causes irreversible changes to PyTorch internal state,stale/low urgency,2,stale/low urgency,1,0,135153,,,,0,daa606995abf728d113eb029050c30c6f0cf4e44533ffdf576e66d54929a1158
thread:issue:135198,issue #135198: lstm quantize issues,stale/low urgency,2,stale/low urgency,1,0,135198,,,,0,485248c2648d8fc444ade34785296b7b23c61dcdcdf4e1532852fb93e8c762b0
thread:issue:135210,issue #135210: Incorrect calculation of standard deviation when using complex numbers on MPS backend,stale/low urgency,2,stale/low urgency,1,0,135210,,,,0,fca01c515999870701256ef4a4f122c25c4b970b21a46e6d7d8dd8c0a1b7bdfa
thread:issue:135229,issue #135229: Disable `retain_grad` on non-leaf tensors?,stale/low urgency,2,stale/low urgency,1,0,135229,,,,0,ee08bdaa0311425985d9210926f8603102709cd21fbffc83a462aa9ef0819eb3
thread:issue:135255,issue #135255: Log missed optimization opportunities due to dynamic shapes in Inductor,stale/low urgency,2,stale/low urgency,1,0,135255,,,,0,1d398ba46c65d683e0906327f34fa9258c199398471eeb7a5d2fd5bd5abc88b1
thread:issue:135267,issue #135267: First all_to_all_single / barrier takes a large amount of time.,stale/low urgency,2,stale/low urgency,1,0,135267,,,,0,d780c007f43aa78675957fec31a55dcb6f0e28ae41a1433a1e8df01544d93f8e
thread:issue:135305,issue #135305: Add restrictions so that Inductor never uses more memory than eager,stale/low urgency,2,stale/low urgency,1,0,135305,,,,0,5b0d6168ec7d11f930c4007543391c3c20aa43ceff4c73ffefc09ce9301cf72e
thread:issue:135330,issue #135330: torch.floor_divide not triggering ZeroDivisionError on GPU,stale/low urgency,2,stale/low urgency,1,0,135330,,,,0,13021d15e4eee2ac3db3316ebf9bb9d1c699f634a220617aa9a8c3d9c4c7494e
thread:issue:135362,issue #135362: MacOS runner fails at `Complete runner step`,stale/low urgency,2,stale/low urgency,1,0,135362,,,,0,bd83f64394b096092df7300aa7b2cd144c9509de06782a9e1777782fe8171cfe
thread:issue:135382,issue #135382: Stop tracking FunctionalTensor bases in Python,stale/low urgency,2,stale/low urgency,1,0,135382,,,,0,772662ed9c901a83c3fc56931deb5e22877ded2c6420cf733dd41f28de97e820
thread:issue:135398,"issue #135398: Models that have non-tensor elements in state_dict, are not ONNX-exportable (and not JIT-traceable)",stale/low urgency,2,stale/low urgency,1,0,135398,,,,0,7086c2dc4192d1c8a4890d431646af47920b43e4c978823ddc9bb8f46c9dc4b2
thread:issue:135409,issue #135409: DataLoader + IterableDataset held up by slowest worker!?,stale/low urgency,2,stale/low urgency,1,0,135409,,,,0,7512384e4793746b93e849a9a2c2e0d0881f43f87b7dda82ff091074625a9897
thread:issue:135447,issue #135447: [MPS] Correctness issue in backward pass of `nn.ReplicationPad1d` and `nn.ReplicationPad2d`,stale/low urgency,2,stale/low urgency,1,0,135447,,,,0,c909c6e7b959276ca73e107ad64babdb63fc2b46fd60c3f4783bf16ff7241dd3
thread:issue:135484,"issue #135484: Boolean indexing not working correctly for a numpy array indexed with a boolean pytorch tensor in the case of 2d singletons (shape of (1, 1))",stale/low urgency,2,stale/low urgency,1,0,135484,,,,0,dbd3cb26228d7cf8761bc54f01b8ad9a08327d396d2bb0076ea9cb331d6f9b6e
thread:issue:135487,issue #135487: No error raised when trying in-place operation on tensor of shape[0] > 1 (single memory location),stale/low urgency,2,stale/low urgency,1,0,135487,,,,0,964ea911e5c6a91577ec316ce6ecda835bb0ba999bf04ecb9211bfe3a6335bae
thread:issue:135489,issue #135489: Sympy bottleneck in eval_relation/assumptions getit,stale/low urgency,2,stale/low urgency,1,0,135489,,,,0,d66c5d93dd210b87c24d2535a4328a9992b1d0037785e4a3555394764a74ba6b
thread:issue:135514,issue #135514: add torch.linalg stubs,stale/low urgency,2,stale/low urgency,1,0,135514,,,,0,9df901243dd4acf8f9bb0d38c7b62e70668da367efcf598795c2d547bc06c62b
thread:issue:135542,issue #135542: torch.nn.InstanceNorm2d and torch.nn.InstanceNorm3d returns nan with tensors of float16 dtype on cpu,stale/low urgency,2,stale/low urgency,1,0,135542,,,,0,5c5051d8c8be8d415a828a7d25f49ceaa7fb38551dcca985fd9a450890c891ef
thread:issue:135590,issue #135590: Discussion about the testing methods in test_transformers.py after math backend always uses fp32,stale/low urgency,2,stale/low urgency,1,0,135590,,,,0,2f6fe6e2b46c2fa9c6cdf1bd563bafb5a5ee98ba1dbfcc21805d9eb583d1d95f
thread:issue:135604,"issue #135604: Generic BLAS provider incorrectly considered ""found""",stale/low urgency,2,stale/low urgency,1,0,135604,,,,0,5e1727062ea812940f92e82dd1e8e85d6857678ddf4e9bdc211c000a1f125a4a
thread:issue:135611,issue #135611: Indexing Numpy array with Tensor Gives Unexpected Results,stale/low urgency,2,stale/low urgency,1,0,135611,,,,0,1be06bc00982ac7e18c6852941143a01cacf3555c99f64eb42e5b52b628c2e7b
thread:issue:135640,issue #135640: Issue UserWarning for missing ModuleList/ModuleDict in Module classes,stale/low urgency,2,stale/low urgency,1,0,135640,,,,0,bb9f2b622166d86dd98d26654b81d47f9729fe46d06ccd6ec35234f84c5f1d1f
thread:issue:135648,"issue #135648: Setting a `float` value to `in_channels`, `out_channels` and `kernel_size` argument of `nn.Conv1d()` gets indirect errors and in_channels and kernel_size with with non-bools",stale/low urgency,2,stale/low urgency,1,0,135648,,,,0,6e755b6ddbefa6478b137899fd1e227f24197969114f1a1569509c099f76db04
thread:issue:135660,issue #135660: torch.cat and torch.stack does not invoke `__torch_function__` API,stale/low urgency,2,stale/low urgency,1,0,135660,,,,0,ffb5b5d4c1f853e48420d3f38ac3dae4df0eb56dc4e03d1427434ca2a2ecc06f
thread:issue:135705,issue #135705: Multiprocessing with workers returning `torch.Tensor` and limiting of the number of tasks per worker (`maxtasksperchild=1`) hangs,stale/low urgency,2,stale/low urgency,1,0,135705,,,,0,e864dc323ccb3c359ccd09d029136045ebb142ed47d1c4047d00690923513d2b
thread:issue:135837,issue #135837: cudaFree will be called when the reserved memory > 70G for A100 80G,stale/low urgency,2,stale/low urgency,1,0,135837,,,,0,b3bfde8a4675eaab4e149b3c94a80f238551da75f94d302754f9eb66ce2fc74a
thread:issue:135930,issue #135930: Integer overflow while creating nested tensors,stale/low urgency,2,stale/low urgency,1,0,135930,,,,0,dc92a21e82281fc5600027a84a88d21e0628d1d2858a744cb60054f3fdd21ceb
thread:issue:135969,issue #135969: Add AT_DISPATCH for layernorm gamma and beta,stale/low urgency,2,stale/low urgency,1,0,135969,,,,0,b28d4607f5a34b7f3f0f4055ba63236578f96708b1144919e3e8440b10034999
thread:issue:13598,issue #13598: caffe2: RuntimeError: [enforce fail at reshape_op.h:110] with Alexnet onnx test with cuda,stale/low urgency,2,stale/low urgency,1,0,13598,,,,0,7c49ee946b4100afe46275bb16007b8411823029504a2806fe90304f883caa67
thread:issue:135992,"issue #135992: When calculating the loss, the input data does not contain NaN, but the output contains NaN.",stale/low urgency,2,stale/low urgency,1,0,135992,,,,0,d145b98f541b26cf60690f946c98b99ab9e4e7867aa972ee8f4ffe8ce712a2cd
thread:issue:136031,issue #136031: Conv vs. Scipy tests fail on MacOS,stale/low urgency,2,stale/low urgency,1,0,136031,,,,0,85d8ca7831f1942f400d0f8a760baacbb64e6bd8dee59a03c684f14b8a49a763
thread:issue:136071,"issue #136071: Setting wrong type values to `stride`, `padding`, `output_padding` and `dilation` argument of `nn.ConvTranspose1d()` gets the wrong error messages saying only `tuple of ints`",stale/low urgency,2,stale/low urgency,1,0,136071,,,,0,8e1861f5022d51e21bc5c2533f428ed93dbf750b3e3e84a724586dffe18b5586
thread:issue:136072,"issue #136072: `padding_mode` parameter of `nn.ConvTranspose1d()`, `nn.ConvTranspose2d()` and `nn.ConvTranspose3d()` is not explained in the docs",stale/low urgency,2,stale/low urgency,1,0,136072,,,,0,cc20dc2d3b8792ead1816358f53d6218eda9d1f7df17b60db0632fe15d99f35f
thread:issue:136073,issue #136073: The real AttributeError information,stale/low urgency,2,stale/low urgency,1,0,136073,,,,0,bf5f46ffb843f39f8b68b71c29ea9a7c608e78ad2a4bcc587db5c8e16e6b69a6
thread:issue:136083,issue #136083: ONNX Export Fails with Dynamic Slicing on Data-Dependent Value,stale/low urgency,2,stale/low urgency,1,0,136083,,,,0,f26d458ae790e9daa8b7e8975761fadff63ff7e4e8cefc13fad6e3191d8550a0
thread:issue:136086,issue #136086: Same token different output from `Conv1d`,stale/low urgency,2,stale/low urgency,1,0,136086,,,,0,2870e7895a4afbf2ce2066b65f564725068fc181834d048d21cca1a5f6b274c9
thread:issue:136096,"issue #136096: Each parameter of `nn.MaxPool1d()`, `nn.MaxPool2d()` and `nn.MaxPool3d()` should have `required` or `optional`",stale/low urgency,2,stale/low urgency,1,0,136096,,,,0,0cd19372b596d7379afba5ee862e5facf6cef7e2fd61f3585c436ba2dd923b21
thread:issue:136098,"issue #136098: `int`, `float` and `complex` type with `return_indices` of `nn.MaxPool1d()` also work",stale/low urgency,2,stale/low urgency,1,0,136098,,,,0,d83c52c3a85a8995bc7bee40bfdc376a5c6e1f29fe140856989394d568751843
thread:issue:136110,issue #136110: Setting `True` to `return_indices` argument of `nn.MaxPool1d()` with the tensor of `int` works,stale/low urgency,2,stale/low urgency,1,0,136110,,,,0,198548a67c177d3bb190adcfa898afe9bae018621611c1b167afd8653cb34be4
thread:issue:136118,issue #136118: [RFC] Default torch.compile backend customization,stale/low urgency,2,stale/low urgency,1,0,136118,,,,0,70fcce2092cc18298b6ce8555638b3a24db0bfb6f649daabab70c7ca1a614739
thread:issue:136119,issue #136119: [torch.export] Dim discrete/finite set,stale/low urgency,2,stale/low urgency,1,0,136119,,,,0,8f6bdabd83b094b1d9708e15921a1b62aa0aa0bcfa4c7efa0b8e9d416d4bb98b
thread:issue:136138,issue #136138: `num_features` of `nn.BatchNorm1d()` with a tuple of `int` works,stale/low urgency,2,stale/low urgency,1,0,136138,,,,0,ba7b79afcb7080631c1c1fda347eab01086596377b9ef772bbe4d8b6dc5e7d1a
thread:issue:136145,issue #136145: MultiLabelMarginLoss with customized margin.,stale/low urgency,2,stale/low urgency,1,0,136145,,,,0,c7f2e5eebbad788a38027c213ba485a9fad56b09cf982cc1797da0e0a669d32f
thread:issue:136186,"issue #136186: ""bmm_sparse_cuda"" not implemented for 'BFloat16'",stale/low urgency,2,stale/low urgency,1,0,136186,,,,0,d7bac106b1f01bf6eb5a8bd3208b75ee47f9e2e15a6b75557e45172e55f8ee6c
thread:issue:1362,issue #1362: Feature Request: noise contrastive estimation/negative sampling,stale/low urgency,2,stale/low urgency,1,0,1362,,,,0,d52fe8aab318ed677f323d1a6571914875528b5afcd549202c16852e90fb42dd
thread:issue:136201,issue #136201: Compile with USE_CPP_CODE_COVERAGE=ON throw erros,stale/low urgency,2,stale/low urgency,1,0,136201,,,,0,52e638c0466b1202d51c282ee576303f0190f436ead2f871649549ca3c52fb23
thread:issue:136250,issue #136250: A question about `getDLDevice`,stale/low urgency,2,stale/low urgency,1,0,136250,,,,0,a2e4b04e3c1a1aa1c697bb3815cf7d9a8a1b915d6e26ec513520acb6fccd2281
thread:issue:136305,issue #136305: Numpy related failure of gradient and sobolengine distribution testcases on POWER and x86,stale/low urgency,2,stale/low urgency,1,0,136305,,,,0,70412630f6c6d986088f01a3f028b3697307e36cb7a4a309fbe2afa4c9496a6b
thread:issue:136357,"issue #136357: torch.func.vjp and sparse tensors: ""Sparse CSR tensors do not have strides""",stale/low urgency,2,stale/low urgency,1,0,136357,,,,0,af667c31872f310b4dd72e424c08b21d4444b45d58b939f5fc16ead3c284c9ee
thread:issue:136381,issue #136381: [Meta] torch.export readiness for Recommendation models,stale/low urgency,2,stale/low urgency,1,0,136381,,,,0,d2189dbbe5f21f0bd95fd5fab70bc8f9c88aa4297011ad7a00483a0c17a88468
thread:issue:136409,issue #136409: Linear Sum Assignment,stale/low urgency,2,stale/low urgency,1,0,136409,,,,0,b20840f446f5811abad92b9fe9b34eb5b56b1e59d440bf75200f58a083f347b0
thread:issue:136410,issue #136410: Implement kaiming_uniform_ for sparse tensors.,stale/low urgency,2,stale/low urgency,1,0,136410,,,,0,e28534013a2926c5f669d2501960f8dd6332f14dea86a8f27c762d9ad138d652
thread:issue:136452,issue #136452: Setting a `complex` tensor to `linalg.norm()` returns a `float` tensor,stale/low urgency,2,stale/low urgency,1,0,136452,,,,0,82d799d90950e288045bfc39d8028bbf0bba9e89b17285618e8df4b6d4b84ea0
thread:issue:136479,issue #136479: The small shape change of input tensor leads to a significant increase in GPU memory usage in Conv3D,stale/low urgency,2,stale/low urgency,1,0,136479,,,,0,f17ed978d5d7819424f1d129d43b45e578148b41ad78423f86ba539ec934fa40
thread:issue:13655,issue #13655: [caffe2] How to export onnx model trained on Detectron in Caffe2?,stale/low urgency,2,stale/low urgency,1,0,13655,,,,0,ed18641bd3fc590b3165521e4ed219be9bba73670bcf4a34f9ba27ea4b9e6543
thread:issue:136568,issue #136568: Setting a `complex` tensor to `linalg.vector_norm()` returns a `float` tensor,stale/low urgency,2,stale/low urgency,1,0,136568,,,,0,eee0bcd432f1af42b1230c5da0bac98fd70d77dd5ccb92f6f79aeccae2160202
thread:issue:136578,issue #136578: Bool convolutions (and other integral types),stale/low urgency,2,stale/low urgency,1,0,136578,,,,0,a4141c232a87af3c9d283c56d217b7f50fbefd50e36c94d0dedf25ed94081292
thread:issue:136588,issue #136588: DistributedSampler shuffle option doesn't work as expected,stale/low urgency,2,stale/low urgency,1,0,136588,,,,0,b239fee79e148992bfd3dd51fd42067d60a56099460a070087ff9bd695909077
thread:issue:136592,issue #136592: [ONNX] Single model export for HF models prompt and token phase,stale/low urgency,2,stale/low urgency,1,0,136592,,,,0,12cd66dbc464fc83ed97683e081bf641ff118f277feaab167f887b725ff57ad5
thread:issue:136605,issue #136605: automatic_dynamic_shapes for mark_unbacked,stale/low urgency,2,stale/low urgency,1,0,136605,,,,0,bb554ae4da2c90fd27646605cc8c9d3ed2d42449e961bf0840a6dee0db040cad
thread:issue:136621,issue #136621: A `complex` tensor with `linalg.matrix_norm()` returns a `float` tensor is returned even if I set `dtype=torch.complex64`,stale/low urgency,2,stale/low urgency,1,0,136621,,,,0,4c79b257c37c442a78c3495e1e56ee82bcfe972a8cb2817635cbbec78dfb18d5
thread:issue:136622,issue #136622: torch.onnx.errors.UnsupportedOperatorError: Exporting the operator 'aten::linalg_inv' to ONNX opset version 18 is not supported,stale/low urgency,2,stale/low urgency,1,0,136622,,,,0,56a63bcfdda719ffbcbf20edee81dd7d437c97ecb64c1895ae2b1b9a1b7fb27a
thread:issue:136623,issue #136623: The MPS Backend sometimes samples outside of distribution domain with `multinomial`,stale/low urgency,2,stale/low urgency,1,0,136623,,,,0,0ca7c562f6bc867b48d82110f1cddbce336dcbf40dd488420a0f9ce212654a6e
thread:issue:13664,issue #13664: Caffe2 install,stale/low urgency,2,stale/low urgency,1,0,13664,,,,0,6fd875c406bd5570cc8131ac811cabe1d9ef90e99054ca5d3e630c8418c8f312
thread:issue:136647,issue #136647: torch.compile HUD dashboard should have repro commands,stale/low urgency,2,stale/low urgency,1,0,136647,,,,0,c8db37401b4076d423ff1ad5a31d66eaa2cf02b54d7a649d70ae74ec14772073
thread:issue:136656,issue #136656: [NJT] Dropout(0.0) with NJT increments cuda rng_state (only for no-compile),stale/low urgency,2,stale/low urgency,1,0,136656,,,,0,cf1072776b45a8b151176f68f8a6b7b1d38e07d21a2312d39cbaa774b43199af
thread:issue:136664,issue #136664: [BUG] torch/extension.h: undefined symbol,stale/low urgency,2,stale/low urgency,1,0,136664,,,,0,ff8d256b793fd91148d8c3c890ec6b4cb651094a8faee8a4221fcfd81c3a2483
thread:issue:13667,issue #13667: [caffe2] Modify models in model zoo,stale/low urgency,2,stale/low urgency,1,0,13667,,,,0,1137d7fca24a4833db8405eead605a2b7b7df74d0a2534f114c86fbc3b5c21af
thread:issue:136680,issue #136680: torch.nn.InstanceNorm3d producing inconsistent output for float16 tensors on CPU and GPU,stale/low urgency,2,stale/low urgency,1,0,136680,,,,0,5f59d4dde83f924c9e56a66185a00aeee0fa0ea81d00258df04580f7b92d113e
thread:issue:136681,issue #136681: [triton_op] Automatically `tl.constexpr` user-written kernel params when they are static integers,stale/low urgency,2,stale/low urgency,1,0,136681,,,,0,30fc9e4b6ec249ee05a8fa6d739471c1e90ae49e79575dc6b8e4b77908e84f6e
thread:issue:136691,"issue #136691: Segfaulting/aborting unit tests do not show up in ""Show Additional Test Info"" section",stale/low urgency,2,stale/low urgency,1,0,136691,,,,0,58e5379cfb44d7cf3db7be974ec91ed723207ea3798fbf90d5a52bff0b0216fa
thread:issue:136699,issue #136699: false INTERNAL ASSERT FAILED in `torch._add_batch_dim`,stale/low urgency,2,stale/low urgency,1,0,136699,,,,0,83ce5cf56eee8fa7e095e9eb1002b5763afd42efcdc9d4e80668d973b7191283
thread:issue:13671,issue #13671: Caffe2 Build Static,stale/low urgency,2,stale/low urgency,1,0,13671,,,,0,7faf6bce255fa3a81d3a1bf0d641b7a8b17cb28d3b120715601a1bc88d48d62f
thread:issue:136717,issue #136717: Floating point exception (core dumped) in `torch.ao.nn.intrinsic.quantized.ConvReLU1d/ConvReLU2d/ConvReLU3d` when stride=0,stale/low urgency,2,stale/low urgency,1,0,136717,,,,0,22fbf5b0a46da46d16a32c867e0de5d890d1160663cbe7ecfd4ce7b335309c0f
thread:issue:136733,"issue #136733: Be smart about autograd formulas saving either the input or output, depending on context",stale/low urgency,2,stale/low urgency,1,0,136733,,,,0,3c88f1ecafe20e5137fbfb1ee5c77e51bd600f88318705b55032b113a1709d2b
thread:issue:136736,issue #136736: torch.unravel_index does not check out-of-bounds,stale/low urgency,2,stale/low urgency,1,0,136736,,,,0,b83dcf6730f89c9c148369aba5f36102d37239bee1c2c3beee312ac068deac1f
thread:issue:136747,issue #136747: Provide `gather_mm` functionality and/or expand nested tensor support,stale/low urgency,2,stale/low urgency,1,0,136747,,,,0,200879aa93c7bca3fac7f6ed14c830eb04d598f833f8defb9a0f62874c11c360
thread:issue:136750,issue #136750: Poor-quality random numbers generated by torch.poisson on gpus,stale/low urgency,2,stale/low urgency,1,0,136750,,,,0,02c804d447dc1d33a5096584d547d079fce205aa338a40d66df729eba6989300
thread:issue:136812,"issue #136812: `CUDNN_BACKEND_OPERATION: cudnnFinalize Failed cudnn_status: CUDNN_STATUS_BAD_PARAM` when using `float16`, works for `bfloat16` / `float32`",stale/low urgency,2,stale/low urgency,1,0,136812,,,,0,79e14ad41d9a96fe5a6df6e1f0d9962600d01223b8a42553d64410962f0034e2
thread:issue:136815,issue #136815: Aborted (core dumped) in `torch.cuda.caching_allocator_delete`,stale/low urgency,2,stale/low urgency,1,0,136815,,,,0,180aba0d57a9efa5ab71bfab2c3be88010431643047278cc26156933863c3bed
thread:issue:136833,issue #136833: Thread safety issue with torch.compile(),stale/low urgency,2,stale/low urgency,1,0,136833,,,,0,84bc8686f940d0d6bd5cb0129cb1ccbd40eb2fb7a3afaec07088b63b6ad0a301
thread:issue:13684,issue #13684: [caffe2] Adding CUDA operators for generate proposals and NMS layers,stale/low urgency,2,stale/low urgency,1,0,13684,,,,0,f839858bd51d971a1210c07d036b04167a50d71d865e4f2f10678cbab930a7ed
thread:issue:136913,issue #136913: [BUG] torch.tril return ZERO Tensor when tensor is large and in CUDA,stale/low urgency,2,stale/low urgency,1,0,136913,,,,0,e426e660c26d462f2c1df0f5743070bde38b87f37d9a07bf43b4dcd709cf2db4
thread:issue:136918,"issue #136918: ONNX export: w=tensor.shape() in torch.arange(end=w...) causes ""tensor does not have a device"" error",stale/low urgency,2,stale/low urgency,1,0,136918,,,,0,299f1f886e9eb953dafacbe83b78eef6e36e90786b30598c1a672672f7ba9530
thread:issue:136926,issue #136926: Python implementation of RNN in docstring is broken,stale/low urgency,2,stale/low urgency,1,0,136926,,,,0,8c170163548cc3d2caa84c14fbecb378f4cdb16f99b424c100dd7e10c03ddd61
thread:issue:136937,issue #136937: Setting the boolean value `True` to `hidden_size` argument of `nn.RNN()` gets a indirect error message,stale/low urgency,2,stale/low urgency,1,0,136937,,,,0,0d0ec7a274ee6e9f3609163976d448a47c0713f1d823fa230337e11c85e7772a
thread:issue:136938,issue #136938: Setting wrong values to `num_layers` of `nn.RNN()` gets indirect error messages,stale/low urgency,2,stale/low urgency,1,0,136938,,,,0,82202d4b9231f65780f571e6bc792c66c16fbbc5661213af4781bb2bc820229f
thread:issue:136946,issue #136946: DTensor does not support `nn.init.eye_`,stale/low urgency,2,stale/low urgency,1,0,136946,,,,0,8c45ff90c7d2687ad023b068a451ed34fe36fa976aae177f1536dc556a812c71
thread:issue:136955,issue #136955: Setting a wrong value to `dropout` argument of `nn.RNN()` gets an indirect error message,stale/low urgency,2,stale/low urgency,1,0,136955,,,,0,de957ca5c55bded95f43f5e6b1b0c1f7dc21d879d781db784d724c5ee0e6e500
thread:issue:136994,issue #136994: Add Mcdropout layer,stale/low urgency,2,stale/low urgency,1,0,136994,,,,0,9c14333b0287a813422fb19c53c84cf685678507827810d54802d272ca4d19af
thread:issue:137006,issue #137006: [RFC] Flight Precheck: ahead-of-hang collective mismatch detection,stale/low urgency,2,stale/low urgency,1,0,137006,,,,0,ece1a20c8071d90e585b7f76729236512d74d47a30fad2bb58bc7f4acbf8a109
thread:issue:137012,issue #137012: [RFC] Use NCCL topology detection for Inductor communication time estimation,stale/low urgency,2,stale/low urgency,1,0,137012,,,,0,2d82ec4e6e42b807549f431ed1f4a91e1175f153dc996138541852165423842d
thread:issue:137045,issue #137045: Namespace error when using torch::unravel_index to implement a custom function in cuda.,stale/low urgency,2,stale/low urgency,1,0,137045,,,,0,b7a02bc70611993ec39869a29d366318490811e0082a9cfd32a3bb63a76ef250
thread:issue:137080,issue #137080: Discrepancy between torch.nn.functional.linear and manual computation with bias under torch.autocast,stale/low urgency,2,stale/low urgency,1,0,137080,,,,0,525c0803964e6e6c3761bdfb0d7cf7af37813fa9dbb13c9bb4b05f9febcca122
thread:issue:137086,issue #137086: conv2d boundaries,stale/low urgency,2,stale/low urgency,1,0,137086,,,,0,89dc60009921ffe6ffeee9973018acf50494f5c407499a047ad70b5f077df22d
thread:issue:137089,issue #137089: Select parameters and buffers by pattern (using fnmatch or re),stale/low urgency,2,stale/low urgency,1,0,137089,,,,0,75e4bef2676e156518c6a0f60b1637e56983c65979b45eee95c1cb3ce9c31d13
thread:issue:137128,issue #137128: Context manager to skip checks inside decompositions,stale/low urgency,2,stale/low urgency,1,0,137128,,,,0,e811cb4eafa89e0552d47ce641010acdee712723666f473fec6a6fea2e205f75
thread:issue:13715,issue #13715: caffe2 c++ load pb model fail,stale/low urgency,2,stale/low urgency,1,0,13715,,,,0,c9a32e3099da23765afd9094df16559d5df0c97ddc504cb24fc902f3140bea7b
thread:issue:137167,issue #137167: `empty_like` with NJT on a different device fails because the offsets are not cast appropriately,stale/low urgency,2,stale/low urgency,1,0,137167,,,,0,4561e59369b90d66881ba8a8080e67998e0491d3fe098af67812a351a0d841ce
thread:issue:137172,issue #137172: `torch._batch_norm_impl_index` performs inconsistently on CPU and CUDA,stale/low urgency,2,stale/low urgency,1,0,137172,,,,0,6d82b54bfa9f64f964b5129a28b4c9f5ed12ffe9e9b7f2aef3c64a5288a97e1d
thread:issue:137173,issue #137173: `batch_first` argument of `nn.Transformer()` should be `True` by default because it's not convenient to get a warning with the default settings,stale/low urgency,2,stale/low urgency,1,0,137173,,,,0,6f6f0f1dec1dfcb9172fbe1a2549f06922b743eea4f74ce031fa048e3e9d473a
thread:issue:137174,issue #137174: `torch.vdot` performs inconsistently on CPU and CUDA,stale/low urgency,2,stale/low urgency,1,0,137174,,,,0,27a2caaf840840463a42eb271bba184ee5cdf4349a8b94d36da9ff28b035fb68
thread:issue:137182,issue #137182: torch_np tests failing with Numpy 2.1.1,stale/low urgency,2,stale/low urgency,1,0,137182,,,,0,2edef050ca062e1de58755a2597b6fab6d2dce64e5df2f8e2a8e7a63ddb5e72d
thread:issue:137192,issue #137192: Setting `nn.TransformerEncoder()` and `nn.TransformerDecoder()` to `nn.Transformer()` doesn't get any errors with `nn.Transformer()`'s arguments having wrong type values,stale/low urgency,2,stale/low urgency,1,0,137192,,,,0,8c845222004a722d125f82f5b235da41eb27740efed210d9e02f7034a79e605f
thread:issue:137217,issue #137217: Automagically lift assertions / shape checks into torch._check,stale/low urgency,2,stale/low urgency,1,0,137217,,,,0,61bd1bc6c4e19f6d9d74de12552662f25045182712661098f3103cc4be5ca8eb
thread:issue:137245,issue #137245: [RFC] Retiring ProcessGroupGloo's CUDA support,stale/low urgency,2,stale/low urgency,1,0,137245,,,,0,cfdd0f09092efc0a53fa0347736c7a6fd4e28ae595faefd537a8a505702c9ce1
thread:issue:137261,issue #137261: list index out of range in cpp_extension.py,stale/low urgency,2,stale/low urgency,1,0,137261,,,,0,3d2957312450f0d3d81327aca257e20c8ca6f6d78d8a0e03738545fb4247494d
thread:issue:137268,issue #137268: DDP deadlock ProcessGroupNCCL's watchdog got stuck,stale/low urgency,2,stale/low urgency,1,0,137268,,,,0,de2a14fd4043bf9587077466addc6c0f1626f1816d20264a66e84d6632340e53
thread:issue:137274,issue #137274: Not decomposing CompositeImplicitAutograd ops in post-dispatch export changes autocast (and possibly more) behavior when compared with eager mode,stale/low urgency,2,stale/low urgency,1,0,137274,,,,0,1adff454d2a8faa05d32372ede719ebfb888f4790247133d524213754a305224
thread:issue:137327,issue #137327: [torch2.4][Distributed Checkpoint] new flatten logic is error-prone,stale/low urgency,2,stale/low urgency,1,0,137327,,,,0,d5abe65a2b02a503a869b325cb86b924e588652bbe969eab4d6215e0b27b577c
thread:issue:137336,issue #137336: NJTs with length doesn't support H2D transfer,stale/low urgency,2,stale/low urgency,1,0,137336,,,,0,501a5b922a597a9d9f00e5b7b81976af4a50d29c49c687fa7cc31cfeff77a91c
thread:issue:137368,issue #137368: Fused AdamW causing Illegal memory access on H100 for large models,stale/low urgency,2,stale/low urgency,1,0,137368,,,,0,1dcc2cd9c9cc6af8bf39675124542f3215fee3a092813821dfe223d69d4892ce
thread:issue:137370,issue #137370: [CD] Implement a fix for: add +PTX to CUDA 12.4 nightly binaries,stale/low urgency,2,stale/low urgency,1,0,137370,,,,0,31943506ee36f13041466fde5504dd8b90f54070a011749930a8a210b9f0569a
thread:issue:137387,issue #137387: Simple explanation should be added to `capturable` parameter of `optim.RMSprop()`,stale/low urgency,2,stale/low urgency,1,0,137387,,,,0,bbbcb1850f7514484e2b5c45640332e571572c1c6c90fd6f619094dbbe0b50c7
thread:issue:137389,issue #137389: non-deterministic issue of torch.einsum function on different GPU.,stale/low urgency,2,stale/low urgency,1,0,137389,,,,0,2914f3149a18d50e243ed1334ea1b974fd6ce37a96d27483a79b7a3c9a03e034
thread:issue:137392,issue #137392: segfault when using DTensor with nonblocking nccl comm,stale/low urgency,2,stale/low urgency,1,0,137392,,,,0,95f00310251539871218256ffb71592977ba8a0d1f6ed5db21a34223aa14b37d
thread:issue:137409,issue #137409: Unable to save model after removing weight normalization due to remaining hooks,stale/low urgency,2,stale/low urgency,1,0,137409,,,,0,b0c9ccdf985a3f0acc738b609145bcb2e3cb93866bc0579b3fd67845901b3863
thread:issue:137412,issue #137412: Public API comparison between v2.4.1 and v2.5 release branch,stale/low urgency,2,stale/low urgency,1,0,137412,,,,0,b79a3cdec0afc516a4b94d982fbd24ea85fdca01a95e181851791362160d7dce
thread:issue:137419,issue #137419: Convert LSTM quantized by QAT to onnx,stale/low urgency,2,stale/low urgency,1,0,137419,,,,0,fb1a3a522e462a83507f91b094a079df660682ef1447ed4b6bd874fdbe1981d6
thread:issue:13746,issue #13746: [caffe2] test depthwise3x3_conv_op_test fails to run,stale/low urgency,2,stale/low urgency,1,0,13746,,,,0,cef489d3fa2e658186299e0b3bcdfdd60c9923f7c4d78d78a7c6b7af2a38932d
thread:issue:137464,issue #137464: Add support for using NestedTensors with linalg operations,stale/low urgency,2,stale/low urgency,1,0,137464,,,,0,d322b5fbb444bde66119301f54d0c6645b75ee4f2b9dc7d7f78fcb78867a5fe2
thread:issue:137484,issue #137484: Better error message in `torch.linalg.vector_norm`,stale/low urgency,2,stale/low urgency,1,0,137484,,,,0,cd13785e2f0405f5365fc6c6721fbbc24d474d3f6375b625ac849f37b32a3bd6
thread:issue:137486,"issue #137486: [feature request] [ux] Introduce special value `SDPBackend.ALL` (or similar functionality) to mean ""all available backends"" for `torch.nn.attention.sdpa_kernel(...)` (some special value `SDPBackend.ERROR` already exists)",stale/low urgency,2,stale/low urgency,1,0,137486,,,,0,81576c3d5ad607b3128c07d1d8169d0a6f143e2cf8d1f6fda8dde1752d2cc8ca
thread:issue:137514,issue #137514: Support multiple ragged dims for NJT,stale/low urgency,2,stale/low urgency,1,0,137514,,,,0,4c02858613ca2754483e9bb31fdcbcc7d041a91d0e844f54ab70115efcf2aaf9
thread:issue:137542,issue #137542: Sparse BSR matrix returns `False` is `.is_sparse()`,stale/low urgency,2,stale/low urgency,1,0,137542,,,,0,86939e40bc34cdce8d9381f3b10407a7ceab62eb97cb0056d7dfc30cf0db842a
thread:issue:137581,issue #137581: Expose underlying cudastream_t for torch.cuda.Stream,stale/low urgency,2,stale/low urgency,1,0,137581,,,,0,477689b545455bbdb4ccfc458c16f09289f246829cf82cbaa1a56e923a098ae3
thread:issue:137610,issue #137610: [RFC] Make Flight Recorder device generic,stale/low urgency,2,stale/low urgency,1,0,137610,,,,0,32ec8dfde5a206e2f2a9ebdfbbf7012104e1b900bbf8c7b69c2d5cd3434e581f
thread:issue:13772,issue #13772: Error in building Caffe2 on Windows (experimental operators),stale/low urgency,2,stale/low urgency,1,0,13772,,,,0,5efdf0e59444128fc4e8c4c689e5d03b9c2557a5975345e339600dc2ebe58c25
thread:issue:137777,issue #137777: Using torch from multiple processes is slow,stale/low urgency,2,stale/low urgency,1,0,137777,,,,0,dfe7e9fadaf0ee1b00f4b124c071015ec58d04f37aae27cf68cdc5f824cd338a
thread:issue:137811,issue #137811: Mechanism to attempt to create a VariableTracker but graph break and turn Dynamo off if we fail,stale/low urgency,2,stale/low urgency,1,0,137811,,,,0,fcd79d7fc8a1c845b0df2193d3d29e5ed18f810ccb36621fd568d78740f198f2
thread:issue:13782,issue #13782: torch.linspace does not check for infinity and nan,stale/low urgency,2,stale/low urgency,1,0,13782,,,,0,d360d4eb374c757a1c3aea11969e143f12656cd458281aeb2177762c3216d58d
thread:issue:13786,issue #13786: PyTorch streams are not cuda-memcheck clean,stale/low urgency,2,stale/low urgency,1,0,13786,,,,0,35685b73805d99a2abd05c934077794c045d52e6b3f4b2dbc3d5978f3b3ecc70
thread:issue:137872,issue #137872: [Perf]The delay between inference request impact the performance,stale/low urgency,2,stale/low urgency,1,0,137872,,,,0,236efec2e151b5d946e662ee6295e6d1f6b57577dcedf31355813e6443d65727
thread:issue:137882,issue #137882: Take device of input tensor into consideration for torch::data::transforms::Normalize and possibly other transformations,stale/low urgency,2,stale/low urgency,1,0,137882,,,,0,b690b2c0c56e90fe8f15384e6f638dbdcbbe30d8a29b5df9b6c62849edc10de8
thread:issue:137888,issue #137888: Inconsistent digamma result with other frameworks (1e-6 difference),stale/low urgency,2,stale/low urgency,1,0,137888,,,,0,3b2dbfac02b1d2499b8b65d979c2d5ba8cc14abad579c87d73f5b51659adbef5
thread:issue:137931,issue #137931: TD skips relevant tests,stale/low urgency,2,stale/low urgency,1,0,137931,,,,0,b3b0a12378752860f53833f46b60f97fefde79ef4fe3e67155a6fbd856e721de
thread:issue:137941,issue #137941: Bazel builds intermittently fail while trying to download mkl from anaconda %),stale/low urgency,2,stale/low urgency,1,0,137941,,,,0,6821981ccf6deac2656b706a3489fa9e328e4d33d88c663ae5b0c65bfbd30b62
thread:issue:137951,issue #137951: Better document that non-blocking GPU -> CPU memory requires device sync,stale/low urgency,2,stale/low urgency,1,0,137951,,,,0,065794f1e196eb436b72bc105e4d2afcba879ca6ff85a773665220b98b37b8d2
thread:issue:137968,issue #137968: inconsistency in ```torch.special.entr``` on CPU and GPU,stale/low urgency,2,stale/low urgency,1,0,137968,,,,0,484e8a8b13f6e0e5a4d7a7ce4dd29093d499dd3303196dd69cf299d87840b99b
thread:issue:137969,issue #137969: Flight recorder cannot mark the state of the batch P2P OP entry as completed.,stale/low urgency,2,stale/low urgency,1,0,137969,,,,0,143427e39d5dbb651c9903ed15a44e97f7f6df522731ba580614644058b86bf3
thread:issue:137974,issue #137974: Can FlightRecorder set the stream key of collective and P2P OP into the entry.,stale/low urgency,2,stale/low urgency,1,0,137974,,,,0,4a034e437ddc4c72660267131d249dd76d3ab14d73b067be859c54f49a3656fb
thread:issue:138017,issue #138017: `_disable_byteorder_record` parameter should be added to the doc of `save()`,stale/low urgency,2,stale/low urgency,1,0,138017,,,,0,8a63b620c7c082a9a7f375d36981cb9a2b70f906e1a52a753452be872cafdc4d
thread:issue:138042,issue #138042: `Calculate Docker Image` step times out if run on LF runners,stale/low urgency,2,stale/low urgency,1,0,138042,,,,0,667a9ccc582af5aef3d050f7eea8a0e8691c2f67ce1f2e094ca227b91eabda03
thread:issue:138168,issue #138168: nan in sparse mv result when I use cpu,stale/low urgency,2,stale/low urgency,1,0,138168,,,,0,0cd3183470d03f5385378cee3c27de020e94e352dda9e00d97c097eeecefdb81
thread:issue:13818,issue #13818: test_spectral_norm: Backward is not reentrant,stale/low urgency,2,stale/low urgency,1,0,13818,,,,0,7acb63509c9467d7416059c84ee5dd71b930c190f08fe1a39607477d7b22c0e4
thread:issue:138180,issue #138180: NestedTensor doesn't support elementwise operations?,stale/low urgency,2,stale/low urgency,1,0,138180,,,,0,b7ac25d3617a5ee3a99dec977a2cb9de30792f78bbdc6af0693234098aeeb26f
thread:issue:138192,"issue #138192: Inconsistent behavior of `torch.tensor` in converting `NaN` to int32 on normal list input and numpy array, and different architectures.",stale/low urgency,2,stale/low urgency,1,0,138192,,,,0,62c7311418bce56dbf918a75958622a31c1cb77405da4b056cf2ccecb509ba9b
thread:issue:138197,issue #138197: Compile error with -WERROR on clang-12,stale/low urgency,2,stale/low urgency,1,0,138197,,,,0,f722b3d8f38cfe39a6d87332b8109088fdd7be2899055a4428e06dcb7b9859e4
thread:issue:138199,issue #138199: Faster dropout (aten.bernoulli) on cpu,stale/low urgency,2,stale/low urgency,1,0,138199,,,,0,f30264867f173d94cd3d7ba6db95dcff6bbf987b3ff3ea6c67ff6b7bfbb3ad1c
thread:issue:138215,issue #138215: Dynamo x autograd.Function: leverage compiled autograd on graph break in backwards,stale/low urgency,2,stale/low urgency,1,0,138215,,,,0,e0c0ee92c11223f7fcd7add6c7176d9601f4146e80d66219b20e2c9f233ff228
thread:issue:138226,issue #138226: No guards for frozen dataclasses and readonly configs,stale/low urgency,2,stale/low urgency,1,0,138226,,,,0,572318c972c847337586065ac78c26d8bf782e7caec7e199e6112a72d0f53f6d
thread:issue:138248,"issue #138248: Adam optimizer without first moment estimate, for less vram",stale/low urgency,2,stale/low urgency,1,0,138248,,,,0,04ab2f78cd1281fc860c89bbbd8b95c3c854fe0988faae1c1695c38e4691eda5
thread:issue:138280,"issue #138280: Refactor FlexibleLayout to separate out ""this stride can be changed"" and ""how this buffer is allocated can be changed""",stale/low urgency,2,stale/low urgency,1,0,138280,,,,0,e91e0269613bbbe7031c7ac925cedab46fe97f58791abbcf450d376f02f978d3
thread:issue:138299,"issue #138299: Cache enablement should be tested inside of cache load() function, not at call site",stale/low urgency,2,stale/low urgency,1,0,138299,,,,0,799007f9e344ac907b4c91ee947bcc38591d04d8960f6cbf0f2cc60c4f49f3c6
thread:issue:138320,issue #138320: TeLU activation function,stale/low urgency,2,stale/low urgency,1,0,138320,,,,0,2c5d364733f6dc6f4115adb3f24e1f8eefd98fc64a11c76be018df1f986b8ed2
thread:issue:138329,issue #138329: DTensor RNG state for non CUDA backends,stale/low urgency,2,stale/low urgency,1,0,138329,,,,0,990876055e89c206afc476309f9b336fac01827e0357b60755cb909ca6e2b0a5
thread:issue:138337,issue #138337: quantile does not work with float16: throws RuntimeError: quantile() input tensor must be either float or double dtype,stale/low urgency,2,stale/low urgency,1,0,138337,,,,0,bd05cebdffedc21434bb30a9aaf644e1ae7416d0ecf12bb7dab30ed10eb7de9d
thread:issue:138353,issue #138353: Empty Dimensions returned in `aten::_trilinear`,stale/low urgency,2,stale/low urgency,1,0,138353,,,,0,9caa51d36c2f5043233ead437034e60b85665d6dfd170c379967ad7328e11c43
thread:issue:13836,"issue #13836: Pytorch-Caffe2 export: ""Arrays are not almost equal to 3 decimals""",stale/low urgency,2,stale/low urgency,1,0,13836,,,,0,3fb690b5d87c80c718372fe2f26317f90cdd03f0e7944f82c4e51d7df44483da
thread:issue:138360,issue #138360: [ROCm] PyTorch Profiler Seg Fault on PyTorch Nightly,stale/low urgency,2,stale/low urgency,1,0,138360,,,,0,fc7cf7d839a47a364ff4bc074d6a57abf3a0ddc33e36f2ab6f9b6677e5f31cf6
thread:issue:138387,issue #138387: torch::distributions module for C++ frontend API,stale/low urgency,2,stale/low urgency,1,0,138387,,,,0,609889f5803b6b5c30732057b642b8596ed6a6de4f404401a1b861bff9ca924a
thread:issue:138418,"issue #138418: Internal assert triggered at ivalue_inl.h:1966, Expected IntList but got Int",stale/low urgency,2,stale/low urgency,1,0,138418,,,,0,9fbb10c16518b104a78c8219a7cdda6a1a79810cbf6bbaeebfcd78eddad7b25a
thread:issue:138420,issue #138420: batch_first support for multi_head_attention_forward,stale/low urgency,2,stale/low urgency,1,0,138420,,,,0,9170357c2484a2ecb9d4c534079b736f4218fc2aebff7d2cf02dd4ad6859d32b
thread:issue:138452,issue #138452: Cholesky factorization of sparse tensors,stale/low urgency,2,stale/low urgency,1,0,138452,,,,0,1ab9ee5b0d6a5f08ad651c3452ad78bad4259a819a347899bc298b1646efb9a6
thread:issue:138453,issue #138453: `torch.__config__.show()` silently initialises CUDA (?); forked processes fail with uninitialised CUDA,stale/low urgency,2,stale/low urgency,1,0,138453,,,,0,d1c0c82e3155162fd685fb4ba9ddffc76a12ca2bdec0217a0e6d4059866fb2f8
thread:issue:138468,issue #138468: os.environ modification is not thread safe,stale/low urgency,2,stale/low urgency,1,0,138468,,,,0,0bf39c6899167abac672884a7a4779d99618573a656abdfd52c63d06e209b0d6
thread:issue:138552,issue #138552: The default setting of PyTorch should show warning always.,stale/low urgency,2,stale/low urgency,1,0,138552,,,,0,f486154d15596337d4922c7333b6cf8436018527c7181024706b17622b86197b
thread:issue:138564,issue #138564: [CI] Continuous spam of PyTorch CI build status notifications in fork,stale/low urgency,2,stale/low urgency,1,0,138564,,,,0,09a5f044115e6888a0ff9a0d8eec8be44b35746e4042b06e22be3f9b341a91b1
thread:issue:138570,issue #138570: `torch.std_mean` returns `NaN` as mean of an `inf` array.,stale/low urgency,2,stale/low urgency,1,0,138570,,,,0,fa9f6aed18de80332c556d587f09d7f4f74996fd469aac03a9302261e4e9bfc5
thread:issue:13865,issue #13865: [c10d] Coordinated file truncation for FileStore,stale/low urgency,2,stale/low urgency,1,0,13865,,,,0,2082bbd2521997ed9c23ff2af9cabb76ff4b27a487d611da29461485f7e2b764
thread:issue:138673,issue #138673: dead code in test that can be removed,stale/low urgency,2,stale/low urgency,1,0,138673,,,,0,7d5b710299cb749aa0050e266fd446e9b2dc82ac82a4db4cbe6ab641fe590b8d
thread:issue:138726,issue #138726: [PP] Add unit tests to check for memory regressions,stale/low urgency,2,stale/low urgency,1,0,138726,,,,0,360cec3054166d0bf8da10344c5c53fa0a1dd283d06743e4c42e9e3dec5d3329
thread:issue:138743,issue #138743: Cleanup the scaling logic in runtime.triton_heuristics.triton_config,stale/low urgency,2,stale/low urgency,1,0,138743,,,,0,d80f9edfff681637c9b5ec00986f91f5f70c76da4f51e19f38faa09ea5a0d721
thread:issue:138773,issue #138773: [funcol] functional collectives are 67% slower than torch.distributed collectives,stale/low urgency,2,stale/low urgency,1,0,138773,,,,0,0f7c7795e2d12d27517f7f3bbc52fcb559d2835f144282f02cde56ad638bf192
thread:issue:138811,"issue #138811: same data all reduce on H20, but results are different",stale/low urgency,2,stale/low urgency,1,0,138811,,,,0,030797fdb4efbb2fec1a00f73dd54914157474fd36ad5b82a546a25e0e0a765c
thread:issue:138813,issue #138813: Strange recompilations on torch 2.5 + FSDP + UNet,stale/low urgency,2,stale/low urgency,1,0,138813,,,,0,02db35bc768985272917742197899f322dfcfb9d35dd7623206182601e4c79f0
thread:issue:138829,issue #138829: torch.dot has inconsistent support for int64 (long),stale/low urgency,2,stale/low urgency,1,0,138829,,,,0,1aad2a427e4f417bff8548e97fbb143705eab0033c604b635a3ce64b410ebd71
thread:issue:138831,issue #138831: Add mean and var operation for Nested Tensors,stale/low urgency,2,stale/low urgency,1,0,138831,,,,0,72f6239d031be0cfd88c725940983ab6609cec5e990bff948687cdfb49c579a4
thread:issue:138898,issue #138898: cpu faster than mps for both GRUCell and LSTMCell on apple silicon M3 max: MPS Metal,stale/low urgency,2,stale/low urgency,1,0,138898,,,,0,d13912cac8dd2305862ac485ad16535710bbf1234bb075a45b08ba25e02a4eec
thread:issue:138901,issue #138901: Runtime error: retains_grad_hooks not implemented for compiled autograd,stale/low urgency,2,stale/low urgency,1,0,138901,,,,0,2119f63d69eac743f461886cb8fe1b35adff6e94719f43fbf7da531857d5e379
thread:issue:138907,issue #138907: Select SDPA backend smartly by `sdp_params` and benchmark results,stale/low urgency,2,stale/low urgency,1,0,138907,,,,0,b6fe816a83f221a5604532c183a7614c3fe60ea732f7c7b0580ac9b1dc45f168
thread:issue:138940,issue #138940: Libtorch torch::Tensor default constructor does not initialize on CPU like in python,stale/low urgency,2,stale/low urgency,1,0,138940,,,,0,b8b0e78aadc152d498c2508af1fa9f4bd1cd256ab76a7c6b94028d66a51e3dfa
thread:issue:138957,issue #138957: `DataLoader` probably shouldn't use `fork` by default,stale/low urgency,2,stale/low urgency,1,0,138957,,,,0,b42d299ec8683a2630f6724fa48ad2fb2ad3fb9206ed8e7971f5d27469cf5e6e
thread:issue:138981,issue #138981: The error compiling problem of Qt and libtorch(my solution to eliminate the bug),stale/low urgency,2,stale/low urgency,1,0,138981,,,,0,d9a6260ace1fe5868836f52a9d0d3e84d3fae0c26b15c39037ea10a08a3cd1f3
thread:issue:138993,issue #138993: Support `attn_mask` in `jagged_scaled_dot_product_attention`,stale/low urgency,2,stale/low urgency,1,0,138993,,,,0,210908a3e0055163de03bfae983bbb6a637ce8f137b9e749d7a3b973b89e7757
thread:issue:139005,issue #139005: Dependency management corrupted in pytorch 2.5.0,stale/low urgency,2,stale/low urgency,1,0,139005,,,,0,610fea0ca1072706f2d809fe5888cc3e80f8db6c844f8d15c922d03ce6a192cd
thread:issue:139065,issue #139065: Cannot compile with latest LLVM-19,stale/low urgency,2,stale/low urgency,1,0,139065,,,,0,b8dd6a14bf060b3e7e4c8ce1a15c2473044a3da3adbacde69e2b81b87498b725
thread:issue:139066,"issue #139066: Covert Conv3d to Conv2d, But output different",stale/low urgency,2,stale/low urgency,1,0,139066,,,,0,57bc15b05d0c05540c7af3705f93247fbfe3d66fef368dc8398a41f62de870f8
thread:issue:139108,issue #139108: LibTorch build error on Windows for CUDA version (debug/release),stale/low urgency,2,stale/low urgency,1,0,139108,,,,0,e881dd425b1bba06c18e15ccf7938fb67b203312794513e51403b317fd9f79fb
thread:issue:139115,issue #139115: torch.nn.AvgPool2d works with long on cpu but not gpu,stale/low urgency,2,stale/low urgency,1,0,139115,,,,0,25f1b94db6a22e700a9bf970fe41956dcd1090380a5158ef80143f04103156a1
thread:issue:139131,issue #139131: [FlexAttention] Update the way we generate kernel options,stale/low urgency,2,stale/low urgency,1,0,139131,,,,0,90b51915808beed5878e55d0d98f3a8320a24ccd7b3cba736afbf8f59d787254
thread:issue:139163,issue #139163: SubgraphMatcher may fail to match when the matching pattern having call_module IR,stale/low urgency,2,stale/low urgency,1,0,139163,,,,0,3853932c75cf1dd9555b1ebbe78600300e896607396702ee7c63e143b9c51274
thread:issue:139170,issue #139170: `pkg_resources` module is deprecated by setuptools,stale/low urgency,2,stale/low urgency,1,0,139170,,,,0,f2b6e5e89f02b974f1553eb088229472b6ffc204f5433d71b85e018acb30fbe2
thread:issue:139178,"issue #139178: the mpi backend not support to transfer float16/bfloat16 tensor and would post ""IndexError: map::at"" when transferring float16/bfloat16 tensor",stale/low urgency,2,stale/low urgency,1,0,139178,,,,0,a4d8d062acacdf58c49710aa0328776c2293134e60cb163f3bd941274ebc6696
thread:issue:139185,issue #139185: Operators being traced as method calls in torch.fx,stale/low urgency,2,stale/low urgency,1,0,139185,,,,0,84e4c8407dda1ede1a61d862abc19d66a57f7d1bf4b343ac429598e354b63a51
thread:issue:139188,issue #139188: PCH build fail with sccache-v0.8.2,stale/low urgency,2,stale/low urgency,1,0,139188,,,,0,d7d0462e5dc992b9a8e93ad6149a2bee563666c8667204bfc699aeae28bdc547
thread:issue:139191,"issue #139191: When I use register_full_backward_hook and register_forward_hook in nn.MultiheadAttention, The MultiheadAttention executed wrong conditional branch and get wrong backward grads.",stale/low urgency,2,stale/low urgency,1,0,139191,,,,0,690a02d5d8af6f755243b926ec5a9d9ae1673e86757e9760882aee286584f9b3
thread:issue:139200,issue #139200: find_spec does something weird to Python Path when loading modules,stale/low urgency,2,stale/low urgency,1,0,139200,,,,0,49e6777dec4e9334394c9e2f6fc4c0046909c5252e75334fbbe9c7388e7b372b
thread:issue:139201,issue #139201: First run lint on just the changes in the PR before running it over the entire PR,stale/low urgency,2,stale/low urgency,1,0,139201,,,,0,bc39e955b326392a0d8bd1aeaa96dc0c745367c861aaeeb79323fc92a9a53b34
thread:issue:139210,issue #139210: [Runtime Error] Build PyTorch with cuda12.2 on Jetson AGX Orin with jetpack 5.1.4,stale/low urgency,2,stale/low urgency,1,0,139210,,,,0,0e5cbab3fe85265b51aa9d3a3dff04347570b01f0b6f66576930745bc1aa378b
thread:issue:139241,issue #139241: Please implement batching rule for torch.nn.functional.multi_margin_loss,stale/low urgency,2,stale/low urgency,1,0,139241,,,,0,1a0461fc9027eb8f3f6b4c0ac81eaa1902b22e046cfa50e4df9daf1f8ff217df
thread:issue:139245,issue #139245: ncclInternalError: Internal check failed,stale/low urgency,2,stale/low urgency,1,0,139245,,,,0,06cfe239bb70f3d563fa9a6ffcc980177f433c0427a7b9a5b41d60b9cfb0c1d5
thread:issue:139252,issue #139252: Ban relative imports in test/,stale/low urgency,2,stale/low urgency,1,0,139252,,,,0,b152e244f5bdb05e178a47fed46573e6ecabb8c0ec97572492ba2fd1a608a158
thread:issue:139268,issue #139268: Torch Inductor should have a way for new backend to provide build options,stale/low urgency,2,stale/low urgency,1,0,139268,,,,0,4e14962bda51cc4e79963faac97c70119d41660ebf961a0e7b0a4726ed67b217
thread:issue:139269,issue #139269: TensorBoard images loading error,stale/low urgency,2,stale/low urgency,1,0,139269,,,,0,879efe0442ec604411cd4b330c54a59eeac80fdbe4c29b6d4d585343a203b76b
thread:issue:139281,issue #139281: `torch._numpy.ndarray.astype()` does not accept Numpy Dtypes correctly,stale/low urgency,2,stale/low urgency,1,0,139281,,,,0,d242237c5c844659c0ba567b88776730af05f75e3a649998ef8948f206ae48e5
thread:issue:139290,issue #139290: Using xpu module in the cuda version Pytorch,stale/low urgency,2,stale/low urgency,1,0,139290,,,,0,af52844ccafb014af38ccd748f55d6bbdae729b0bcd46a88d54fe0a8dcc39b16
thread:issue:139304,"issue #139304: [RFC][Pipelining] RNG state communication, avoid seed checkpoint",stale/low urgency,2,stale/low urgency,1,0,139304,,,,0,de72ef307b6ff24df0d7174b385ccd09936c1ecd5a686ea3350272420e9d9630
thread:issue:139305,issue #139305: torch.export torchaudio kaldi module,stale/low urgency,2,stale/low urgency,1,0,139305,,,,0,3476142f67b04100e9b71a4476a1d858dceddb20b164cde2da75cd4b11f74fbe
thread:issue:139324,issue #139324: [FR] Support sub-group partition ProcessGroup,stale/low urgency,2,stale/low urgency,1,0,139324,,,,0,2d83ba42ab2bd4be704214d297e7aa319a0432df6e78521ee6480a8d44da5ff3
thread:issue:139333,issue #139333: torch.nn.ReflectionPad2d throws CUDA error on large tensors,stale/low urgency,2,stale/low urgency,1,0,139333,,,,0,b530f78b0c159632d0215a32886860327f39373c3070712b3ec75f9601392ffa
thread:issue:139342,issue #139342: torch.logcumsumexp runtime error inconsistency on CPU and CUDA (also documentation missing constraint),stale/low urgency,2,stale/low urgency,1,0,139342,,,,0,662416855aeb27a3a8774710bcccb11236c1ebcf5ff6c73ab04ac6f5cd1bd43e
thread:issue:139349,issue #139349: torch.nansum does not work with complex numbers on CPU,stale/low urgency,2,stale/low urgency,1,0,139349,,,,0,1f3f3ffeab0d8ec0a8b4edf3d8f82aad35a5873fe91400565c7aae350820cc64
thread:issue:139381,issue #139381: Improved definition for Hardsigmoid,stale/low urgency,2,stale/low urgency,1,0,139381,,,,0,f25bb631fe5138ba97799627dba1106f7554ad3e1f1db75d106ada877bb16b99
thread:issue:139382,"issue #139382: RuntimeError: free_upper_bound + pytorch_used_bytes[device] <= device_total INTERNAL ASSERT FAILED at ""../c10/cuda/CUDAMallocAsyncAllocator.cpp"":527",stale/low urgency,2,stale/low urgency,1,0,139382,,,,0,3dec84fc10ea7b681c6cf7b021223f2a26d3211e2d8180a355f298f1d0ac9354
thread:issue:139401,issue #139401: [export] Failure to export Mixtral-8x7B-Instruct-v0.1,stale/low urgency,2,stale/low urgency,1,0,139401,,,,0,cad72e9d6b2170c45a9a08b6e72ac1174b41ca20f4f11b99162f875db7baf5a0
thread:issue:139472,issue #139472: _refs.div.floor_rounding returns NaN instead of +- inf when a divide by 0 occurs,stale/low urgency,2,stale/low urgency,1,0,139472,,,,0,e95d3b7c1d6f76d9ecdb633b861f53b5863d865e4312ad159fa1f8daef1f96af
thread:issue:139494,issue #139494: Mutating custom ops slower than non-mutating custom ops.,stale/low urgency,2,stale/low urgency,1,0,139494,,,,0,9c012a04846cf50ac920e54da5f234626b099759b5a9e705f1fc9fedd5b02079
thread:issue:139520,issue #139520: inductor/test_move_constructors_to_cuda.py::TestMoveConstructorsToCuda::test_multi_gpu unit test failure,stale/low urgency,2,stale/low urgency,1,0,139520,,,,0,f3374240d07282977bba760cb4e3baea95820ed865001fb548723f89ec63e4fc
thread:issue:139582,issue #139582: nn.LSTM documentation,stale/low urgency,2,stale/low urgency,1,0,139582,,,,0,d83b74b70067baa09aee636a74e2eee6f8698203efd83ba9f01675b0586dca58
thread:issue:139602,issue #139602: [RFC]: Adding Triton Backend for Aten operators,stale/low urgency,2,stale/low urgency,1,0,139602,,,,0,056b11000dbb3735f2caf73d94ef98a09758b8b258cf39cddbc5848568339216
thread:issue:139609,issue #139609: There is a bug in torch.cat for non-standard boolean,stale/low urgency,2,stale/low urgency,1,0,139609,,,,0,558e99faeb9b0fe0dea8b385e10299270cbf42621fccb34d126756ccaa80e3e5
thread:issue:139648,issue #139648: [Tracker] complex128 tests are disabled for all foreach internally,stale/low urgency,2,stale/low urgency,1,0,139648,,,,0,3e471e55e02f9debe0d53076be643d8f28d5bb26df554edc2174c72be2bc8436
thread:issue:139701,issue #139701: `softmax` fails to commute with `permute` / `transpose`,stale/low urgency,2,stale/low urgency,1,0,139701,,,,0,cc6a78f632088269b48339b4553883cfc305b1e776f2425ab96d736fe19adb92
thread:issue:139761,issue #139761: [RFC] PyTorch - PyPi PEP-759 proposal (wheel-next),stale/low urgency,2,stale/low urgency,1,0,139761,,,,0,6333a2720b30298532e7cc8365f9fca99792f39e0ce4e5c0bd0e3671fc41f847
thread:issue:139795,issue #139795: Support scalar tensor in `value` arg of foreach op,stale/low urgency,2,stale/low urgency,1,0,139795,,,,0,8ed6024c858d09295ca9ca2242e78fca5ce1e094567f28800ea5b6cbe0d8e002
thread:issue:139874,issue #139874: Modify CMake files to support Clang for Windows when building Libtorch-dependent apps,stale/low urgency,2,stale/low urgency,1,0,139874,,,,0,540b532d84fcbb20db2d4d30185015810c0e6984735ade7865b6403d30dc9c97
thread:issue:139876,issue #139876: Allow using context managers in torch.fx,stale/low urgency,2,stale/low urgency,1,0,139876,,,,0,a64ccc925251b8fb3ece740a6f5202d593f16323c306275f3ea8b47205544ab5
thread:issue:139910,issue #139910: Setting a `str` value to `weight` and `pos_weight` argument of `BCEWithLogitsLoss()` gets an indirect error message,stale/low urgency,2,stale/low urgency,1,0,139910,,,,0,35547d0201187fd042ca1efe3180cc2f4935a8758e0588156fcc5a96d67071c6
thread:issue:139916,issue #139916: Both `input` and `target` argument or at least `target` argument of `BCELoss()` shoud accept the values out of `0 <= x <= 1` so that `BCELoss()` with `torch.sigmoid()` and `nn.Sigmoid()` gets the same results as `BCEWithLogitsLoss()`.,stale/low urgency,2,stale/low urgency,1,0,139916,,,,0,3380ce179878b666b0bb4c77a85bc1da9ac2d8f815cb1c501553885981b5ce38
thread:issue:139958,issue #139958: Can't seem to run tip of tree on Linux,stale/low urgency,2,stale/low urgency,1,0,139958,,,,0,fc1bd4b8006adc29414cfaf9755679e97f0b58eec05e3a07b49e393857806bde
thread:issue:139962,issue #139962: fbgemm build errors on warning that VLAs are a clang extension when building with clang18,stale/low urgency,2,stale/low urgency,1,0,139962,,,,0,33a52c3b1a82b4ab6f792de360dc48583436d27a397b5ae0e2e7e9b2ddfe1755
thread:issue:140012,issue #140012: Default value for get_num_threads,stale/low urgency,2,stale/low urgency,1,0,140012,,,,0,1bbe61e4715b1dc4353d4bcfd1d7d8dc308139365464298390fe280f780780a1
thread:issue:140015,issue #140015: Find ninja from local environment in `torch.utils.cpp_extension.load`,stale/low urgency,2,stale/low urgency,1,0,140015,,,,0,b01191037bf0163e176138251c89055b2190fb88cd066b7022e44b50a0444294
thread:issue:140017,issue #140017: ghstack-mergebility-check often gives false positives,stale/low urgency,2,stale/low urgency,1,0,140017,,,,0,da9da75c9ed17110b21e9ccb257c6f276af266e607ac656c18754247a7e3802c
thread:issue:140018,issue #140018: TargetDeterminator skips op_info tests for operator directly changed there,stale/low urgency,2,stale/low urgency,1,0,140018,,,,0,af5a9e5691da480e20e9735b3152931fe4871a8fda504a6432f21f7763370b36
thread:issue:140088,issue #140088: Building PyTorch from source fails when magma.h is in /usr/include,stale/low urgency,2,stale/low urgency,1,0,140088,,,,0,b1b40080f1827f6da8a6b759a456bd8a5e7b9611bd476dd1cff07e8f761e5697
thread:issue:140112,issue #140112: inconsistency in ```torch.nn.BatchNorm1d``` on CPU and GPU,stale/low urgency,2,stale/low urgency,1,0,140112,,,,0,f7d67f0cf13958bc93c81c1686121b7baef77de016523329c8707e3ee4d58233
thread:issue:140130,issue #140130: [onnx] [njt] [feature request] Export NJT-enabled SDPA / MHA ops to ORT's PackingMode Attention,stale/low urgency,2,stale/low urgency,1,0,140130,,,,0,9b01d23090a487a7d81a9871ac2189d098c40be042b09a50065e52126995204a
thread:issue:140214,issue #140214: torch.svd_lowrank - non reproducible,stale/low urgency,2,stale/low urgency,1,0,140214,,,,0,8f59633653695680c88ec431c8ca226738f4200fc931245bf08231049b0218e4
thread:issue:140237,issue #140237: Cannot export FeaStConv-based network to onnx via torch.jit.script,stale/low urgency,2,stale/low urgency,1,0,140237,,,,0,b5d46e2f21a25c71d2d60f5052c2ea4ac413ded973559f81f6ab9d9e0abcc944
thread:issue:140296,issue #140296: asynchronous copies from accelerator to cpu: what should be the expected behaviour?,stale/low urgency,2,stale/low urgency,1,0,140296,,,,0,e2ab561e253865fc2168da49b7e936682390643455895f58912c1ff54665775f
thread:issue:14030,issue #14030: Cannot import caffe2_pybind11_state_gpu,stale/low urgency,2,stale/low urgency,1,0,14030,,,,0,67d75e72cb5ff631af5d1ba4fdd92028737cd38e16a0cf6c793c8dfd099e2ecb
thread:issue:140301,issue #140301: [RFC] Distributed manual_seed API for N-D parallelism,stale/low urgency,2,stale/low urgency,1,0,140301,,,,0,c5e4b5c6c6998e051c7ccf9b31cf8e223b0ff34bb76aa62b985191ca8567c317
thread:issue:140322,issue #140322: torch_dispatch x HOP handling,stale/low urgency,2,stale/low urgency,1,0,140322,,,,0,0e80c76ba510f6e74faae4a8e4924fde9d387797a25aa2b4328067ba0e34d782
thread:issue:140345,issue #140345: Flaky bot might be missing some data when making a judgement,stale/low urgency,2,stale/low urgency,1,0,140345,,,,0,6f97666d811316908db38852f2e587d3c7a2c554ee4579bbe3aa745ce48104cb
thread:issue:140348,issue #140348: inconsistency in ```torch.special.i0e``` on CPU and GPU,stale/low urgency,2,stale/low urgency,1,0,140348,,,,0,249298f0e08e1801f6a9c8326ed9e9d970d95e7fcbf698ca4ecaf7ef36ac8819
thread:issue:140396,issue #140396: Optional ```clamp_k``` Argument for Handling Out-of-Range Values in ```torch.topk```,stale/low urgency,2,stale/low urgency,1,0,140396,,,,0,c2a5b59b8e69b94fa74839413274ff489f0b5489736edc3a05f9b561f1e1f288
thread:issue:140416,issue #140416: [DDP + Dynamo] Failed to get compiled graphs,stale/low urgency,2,stale/low urgency,1,0,140416,,,,0,161b325c368fcb4251be1652ceef70fc3f66f3442eb6d1a348eeea6525bd0704
thread:issue:14047,issue #14047: Make torch.multiprocessing.SpawnContext usable,stale/low urgency,2,stale/low urgency,1,0,14047,,,,0,c1ca4cfea970d4ca295749e83d29c8684b3b52f6b06a4c95e1fa2b34ea43387d
thread:issue:140513,"issue #140513: Using @torch.compile, there are slight differences in output when run on different GPUs on the same machine.",stale/low urgency,2,stale/low urgency,1,0,140513,,,,0,7e083451748102886b69b972f8c8430e62e969f4618e243bab95a0fd180d36d1
thread:issue:140514,issue #140514: Error with fused AdamW and FSDP,stale/low urgency,2,stale/low urgency,1,0,140514,,,,0,8edc447b85fc2a1d9ac289c7d0b92819024d765efedf1d5778f8c4457829a244
thread:issue:140516,issue #140516: [autograd engine && compiled autograd] support for partial backward compilation,stale/low urgency,2,stale/low urgency,1,0,140516,,,,0,211ce1bf71cade7ab7ccba6d55e8f36cbf8310e39f0d5afbd454dcf0c12abd66
thread:issue:140520,issue #140520: Enable autograd for padded dense forward / backward operators.,stale/low urgency,2,stale/low urgency,1,0,140520,,,,0,3e0e966c4a9eeb09bd29f35039c415596bb36bd6defc7e09069850c3822ea263
thread:issue:140560,issue #140560: `skipIfXYZ` decorators will skip CPU tests as well,stale/low urgency,2,stale/low urgency,1,0,140560,,,,0,f1b660773dab8e7b1210042f398a21f4c794bdecae8a9a693ca1a60b50b26e19
thread:issue:140592,issue #140592: Inductor pattern matcher - missing val,stale/low urgency,2,stale/low urgency,1,0,140592,,,,0,aa63a7070d5eeb65a5adf46bfa18e27b0e7d924609de63a9886c99ea2eac435f
thread:issue:140596,issue #140596: [export][non-strict] Passing in kwargs torch.export fails at non_strict_utils.fakify(),stale/low urgency,2,stale/low urgency,1,0,140596,,,,0,a8d40e011b78eefaf2dd6dd87e49dd11d96779accdcfc374cef8ffd1b6b7b76f
thread:issue:140602,"issue #140602: torch.from_numpy() returns a cpu tensor, even under `with torch.device(""meta"")`",stale/low urgency,2,stale/low urgency,1,0,140602,,,,0,8a6a57f777a82514ac49e4ad4579b3bbda2d18c3098adfc7566610f6a9216268
thread:issue:140640,"issue #140640: Identical Formulas for BatchNorm, LayerNorm, InstanceNorm, GroupNorm",stale/low urgency,2,stale/low urgency,1,0,140640,,,,0,7fae111cc2077aeb0d4d203f894fc01df13329c290c4e24e8064c897fff69a42
thread:issue:140687,issue #140687: “Can't swap an already initialized allocator” when calling torch.cuda.memory.change_current_allocator,stale/low urgency,2,stale/low urgency,1,0,140687,,,,0,3d28f2f8d2f0b4612e7ed1fe94a2ac10617e09302bc43722c015287e0c34bfaa
thread:issue:140696,issue #140696: inconsistency in ```torch.nn.functional.adaptive_avg_pool3d``` on CPU and GPU,stale/low urgency,2,stale/low urgency,1,0,140696,,,,0,8e53ac31d6ebe451e66c7669a9adf9d33dca8ccb5884a814690f2a07861c67c2
thread:issue:140701,issue #140701: inconsistency in ```torch.special.polygamma``` on CPU and GPU,stale/low urgency,2,stale/low urgency,1,0,140701,,,,0,3575e070c0aec0fdebede42a4db0a70c88646fb556abc0b871b56361b7345be0
thread:issue:140703,issue #140703: [Feature Request] The `.forward()`/etc. shape API,stale/low urgency,2,stale/low urgency,1,0,140703,,,,0,781d81895f02fb64b7b0475f8fcdd4f8788643de0db443cda4975042d40e3239
thread:issue:140704,issue #140704: vmap with mps tensor fails to recognized batched shape,stale/low urgency,2,stale/low urgency,1,0,140704,,,,0,e0502427e7c37a7b5abe2be1ee0fa5416f7e105ff9af3b77435445adecdf682c
thread:issue:140711,"issue #140711: after around 13 iteration, tensor img_syn contains many ""NaN"" value when trained on cifari10, but on minist dataset, it works fine",stale/low urgency,2,stale/low urgency,1,0,140711,,,,0,7c9cc4f9bf6964ebc4e0ee29319e166d3f3005e0be60a751865241bd6e0afd85
thread:issue:140712,issue #140712: No fake impl or Meta kernel for Communication Operator,stale/low urgency,2,stale/low urgency,1,0,140712,,,,0,9713f3b31841520657b89b3428def643365f1f4936698664b458c13be1b3dc64
thread:issue:140778,issue #140778: Floating point exception (core dumped) in `torch.sparse.sampled_addmm`,stale/low urgency,2,stale/low urgency,1,0,140778,,,,0,680ae1f034fefc6cb260ba3d57c62987b399c54dd8a1a0a0971c502eee3b29e9
thread:issue:140796,issue #140796: inconsistency in ```torch.special.xlog1py``` on CPU and GPU,stale/low urgency,2,stale/low urgency,1,0,140796,,,,0,e601603b18c3778a05c6fa17364903015929bc9d7f4d8084d8467ca46b336977
thread:issue:140881,issue #140881: Implement Cut Cross Entropy,stale/low urgency,2,stale/low urgency,1,0,140881,,,,0,262a6b39a741c160f573ef8927da6f017a36a7cd1b6377397834d6cfb4d15dce
thread:issue:140892,issue #140892: inconsistency in ```torch.special.xlogy``` on CPU and GPU,stale/low urgency,2,stale/low urgency,1,0,140892,,,,0,2eed2f7139a2127f59d79f507a226c62315b08b9c026b5e6071b915e86b7e031
thread:issue:140930,issue #140930: [cudnn_frontend] Error: No execution plans support the graph when using CUDNN_ATTENTION in SDPA,stale/low urgency,2,stale/low urgency,1,0,140930,,,,0,a37bef8690a0dbc771bdaa0ad453edb4b26c457b05b163ae28d5f4a96d8cc00b
thread:issue:1410,issue #1410: BatchNorm should use Bessel's correction consistently,stale/low urgency,2,stale/low urgency,1,0,1410,,,,0,5b93243a0e0e1868fb15fd1308137b595e1ef56a98ba661fdd1b75d1a87c78f6
thread:issue:141018,issue #141018: The description of key_padding_mask in float mode is incorrect,stale/low urgency,2,stale/low urgency,1,0,141018,,,,0,54e15e18fb67cdb6ccaf7a28970eaae16bf481e90ae61594892f7719052db4e4
thread:issue:141094,issue #141094: plus prefix-less wheels not being published on https://download.pytorch.org/whl/torch/ for linux x86_64,stale/low urgency,2,stale/low urgency,1,0,141094,,,,0,19588ce1dd921828f84d41bee73b9e13d7a2119cc0320f40d7aa42cea65c57f4
thread:issue:14112,issue #14112: [sparse] add descriptions and examples for methods at torch.sparse doc page,stale/low urgency,2,stale/low urgency,1,0,14112,,,,0,0ba71362b203ead1930ef02fd9c97d0d976f7facec423baa908b1804b1758850
thread:issue:141125,issue #141125: [Dynamo] Dynamo is not aware of the function users override,stale/low urgency,2,stale/low urgency,1,0,141125,,,,0,93c487a1f12111e7faa53ab3a843d5533f11d86bad41ffe728412dd99d40e998
thread:issue:141139,issue #141139: Consider noting the PyTorch version for BC-breaking error messages,stale/low urgency,2,stale/low urgency,1,0,141139,,,,0,c8efadac098791c44e2f7c8268737ebf871d1849cecf78f277555913535bfa47
thread:issue:141144,"issue #141144: Spsolve for sparse, complex matrices",stale/low urgency,2,stale/low urgency,1,0,141144,,,,0,2b7822108a5fefc3a882016f552871bf87aa68aec8996fccedf08a84f7da4c8e
thread:issue:141203,issue #141203: SEGV in `torch.native_batch_norm`,stale/low urgency,2,stale/low urgency,1,0,141203,,,,0,dc2f279e3504358ef862c16c828da7b7583a9c976e920ac7427367ee32e98edc
thread:issue:141204,issue #141204: Stack trace from pytest is very far away and far too find on some tests,stale/low urgency,2,stale/low urgency,1,0,141204,,,,0,4e11efdad19b76468dcd8f4d9692f60f55b9866fbfa94b8e71eadccdf31918dd
thread:issue:141210,issue #141210: Addmm is not always better than add + mm,stale/low urgency,2,stale/low urgency,1,0,141210,,,,0,dddd4e1aeda411724adbc53158d31c4a78dba8bb0580ec741300dcc010d99367
thread:issue:141216,issue #141216: Implement ADOPT Optimizer into torch.optim module,stale/low urgency,2,stale/low urgency,1,0,141216,,,,0,1bfe838bfbe12f99770f10a276b299c989f50c78bb6b91f43eb19ba0e4ec060b
thread:issue:141235,issue #141235: Enable CI and Build Support for PyTorch on PPC64LE Architecture,stale/low urgency,2,stale/low urgency,1,0,141235,,,,0,5aca4245dbcef7956c1d2a2dde04b4500cc37d79725397982909617c2bec6851
thread:issue:141277,issue #141277: Multiple torch.{*}_copy variants are not in generated docs/docs skiplist,stale/low urgency,2,stale/low urgency,1,0,141277,,,,0,7450135f048347aacbe1e98f09452313cf66ce54f6e0936ca7398f7d35afdfb4
thread:issue:141295,issue #141295: Memory efficient reshape for stride-0 tensors,stale/low urgency,2,stale/low urgency,1,0,141295,,,,0,a26285718c765197c9b87a6aaafccb7f50b327066e9e9441019e83c4b7643e7d
thread:issue:141299,issue #141299: Pytorch notifications,stale/low urgency,2,stale/low urgency,1,0,141299,,,,0,a81f653346ee855bfebb63b56183af488300bc0de62c078e2e360b9a869a46eb
thread:issue:14131,issue #14131: running caffe2 float16 tensors results in aten runtime error,stale/low urgency,2,stale/low urgency,1,0,14131,,,,0,48843ea4e45de36bd262beaa7e840f8720e2a88c1b3f249b33d2852b5d10a79b
thread:issue:141335,issue #141335: Rounding issue of tensor sizes when sharding,stale/low urgency,2,stale/low urgency,1,0,141335,,,,0,412ef68f4f4be5789ca4e64b4492a4d1c148698ed8d43d6518ae23f880ffe113
thread:issue:141337,issue #141337: `torch.det` with python 3.10 and torch 2.5.0+cu121 raises RuntimeError,stale/low urgency,2,stale/low urgency,1,0,141337,,,,0,6d2c01f3a0394c0d2b5e29254778f7a44c0239e1f6a20dfc4f42280a56028bd7
thread:issue:141343,issue #141343: numpy's division compiled but returning Different results,stale/low urgency,2,stale/low urgency,1,0,141343,,,,0,7d55ab856c555699aa504852b6314d61fe2d64d18e998de21abe81d7896ac44a
thread:issue:141350,issue #141350: Support saving pytorch models in OCI registries,stale/low urgency,2,stale/low urgency,1,0,141350,,,,0,ef5e4e556a56d0b51737092ba42e6187cfc7c09bd68f9f5dbe3f30bb5139bc9b
thread:issue:141417,"issue #141417: Could you provide a standardized interface or documentation for mapping all versions of torch, torchvision, and torchaudio?",stale/low urgency,2,stale/low urgency,1,0,141417,,,,0,197b1286312018bc1dab5a29fc01d506be4b88c4ec7d4eecc1e50faabaf84004
thread:issue:141422,"issue #141422: What is ""recompilation profiler"" in doc? (Seems to have a dangling link)",stale/low urgency,2,stale/low urgency,1,0,141422,,,,0,1f8ae586c2de675ab776819de7835523254c5fc4aa8ed601f40aaa8ec20aaa87
thread:issue:141449,issue #141449: Segmentation fault (core dumped) in `_fft_r2c`,stale/low urgency,2,stale/low urgency,1,0,141449,,,,0,bbfe0c7b54b1cf1c98f600c45f7b485e3faeb74ac7732b81d7f709aab1edc128
thread:issue:141450,issue #141450: Segmentation fault (core dumped) in `_fft_c2c`,stale/low urgency,2,stale/low urgency,1,0,141450,,,,0,4fe6c9bf94a06dd1e86fd3b02b61f8ca1efe70e0dde075ff1b7d4c6071737bf9
thread:issue:141467,issue #141467: import torch failed after installing nightly wheel,stale/low urgency,2,stale/low urgency,1,0,141467,,,,0,9de6240bffdeef41f23e283687b32977cec61bfe0798de536f15876cfa9a7d20
thread:issue:141472,issue #141472: [sparse] `mul`: missing support with CSC arguments,stale/low urgency,2,stale/low urgency,1,0,141472,,,,0,16982062d6f233788fe8d85c3de1530e1eea715f2a31001b65a225033000ec05
thread:issue:141507,issue #141507: [AudioLM] Inductor `LoweringException: NotImplementedError: View`,stale/low urgency,2,stale/low urgency,1,0,141507,,,,0,80dfaf366da8d49e4d73cce844783d18e61c139392bc2aa965dc5d44f42ac378
thread:issue:141558,issue #141558: Incremental CI testing,stale/low urgency,2,stale/low urgency,1,0,141558,,,,0,3df9982eca4bad82028470117dc8233f40ed129fc6cbf5ab0a55a3e2b66c514e
thread:issue:141560,issue #141560: [Feature Request] Add source Inductor Python file path to each Triton kernel,stale/low urgency,2,stale/low urgency,1,0,141560,,,,0,943499db1e8c1b196ee8b8711a595ad8fdd4990c1cf1bcd319d5a1e7545a49c6
thread:issue:141589,issue #141589: Minimizing recompiles when compiling multiple inlined `nn.Module`s,stale/low urgency,2,stale/low urgency,1,0,141589,,,,0,db6f8ba0ce9a17c93e6ee1d1659fa5aea2d0cfaac033b6721acb2cef8a9a93f8
thread:issue:141594,issue #141594: Marking integer inputs to a compiled region requires `torch._dynamo.assume_static_by_default = True`,stale/low urgency,2,stale/low urgency,1,0,141594,,,,0,28f68f10ffc747cb5ea7395f457d5c37ea097be0bd1c3fe648da42b772c62ff3
thread:issue:141616,issue #141616: Clarify that `compile(module)` only affects the `forward` method.,stale/low urgency,2,stale/low urgency,1,0,141616,,,,0,22008e8d2466ce67d113925a7587000130fd18be34b7b11831ee5286b8b23bd1
thread:issue:141675,issue #141675: Stuck at dynamo with dynamic shape enabled in PT2.5.1 which was not observed PT2.4.0.,stale/low urgency,2,stale/low urgency,1,0,141675,,,,0,79381f4f0a69a4d19c696c0e182157cefa6dad753378ff68cf9a42a7701e877e
thread:issue:141686,issue #141686: [PT2] Fine-grained custom triton kernels support,stale/low urgency,2,stale/low urgency,1,0,141686,,,,0,f4a0a85119daff657eaf79bb923b144eb0c65fdb670618e9df09c08c6d62cbbf
thread:issue:141690,issue #141690: Reworking is_big_gpu check for inductor GEMMs,stale/low urgency,2,stale/low urgency,1,0,141690,,,,0,3eafc55fe7ac7457798e6724bfa1fbf7b51a1282cdcabd545d6cfdcef18ab357
thread:issue:141727,"issue #141727: inductor and eager gives slightly different results for rms_norm, with `emulate_precision_casts=True`",stale/low urgency,2,stale/low urgency,1,0,141727,,,,0,4c5c3e6e96b96c38d554190e9ca47c9070a4d30f97114493e6599ed341ff9815
thread:issue:141746,issue #141746: How to specify the port for processes with rank > 1 in the Gloo communication backend?,stale/low urgency,2,stale/low urgency,1,0,141746,,,,0,35cab397164f40b6637256035ae2f6c9d8421d055a373b3747507caf70e7a711
thread:issue:141749,issue #141749: Incorrect error message on incorrect byte boundries on _scaled_mm on fp8,stale/low urgency,2,stale/low urgency,1,0,141749,,,,0,36446043892e326e61c478dc190bf715caa10728fd53344319fab66cffe13d8f
thread:issue:141758,issue #141758: Test in `test/test_overrides.py` failed with AssertionError,stale/low urgency,2,stale/low urgency,1,0,141758,,,,0,1d477edad0d226462292b23e663a6b2f7d0340ae64bd536b9273f3ab88f2f03a
thread:issue:141789,issue #141789: torch.roll runs too slow at MPS backend,stale/low urgency,2,stale/low urgency,1,0,141789,,,,0,892aef416c495d0843a1e24c62a768b5051b3d73b76b79213525307bc0a968e2
thread:issue:141801,issue #141801: [typing] add `nn.Module` overload type hint to `torch.compile`.,stale/low urgency,2,stale/low urgency,1,0,141801,,,,0,b9e79a98dfd6c78bf74b6f1662e7718307eddb4638e393ff433a5acc91223cd0
thread:issue:141802,issue #141802: inconsistency in ```torch.Tensor.logcumsumexp``` on CPU and GPU,stale/low urgency,2,stale/low urgency,1,0,141802,,,,0,cae7d3bc9cdd5f2efd9188d654af245034cf7a821242951803c624695a8b389d
thread:issue:141828,issue #141828: Adding to Private Type Stubs,stale/low urgency,2,stale/low urgency,1,0,141828,,,,0,fad7256985dc1bee3163de3e38189499feaf01dc5113f936202e8a7d23a60a6c
thread:issue:141845,issue #141845: Using torch.func with saved tensor hooks,stale/low urgency,2,stale/low urgency,1,0,141845,,,,0,f21ee8a42ae896acbfbbd64000f40450b2f25a142031f08f6ce0b497cabdaa07
thread:issue:14185,issue #14185: [Caffe2] Non-spatial batchnorm optimization,stale/low urgency,2,stale/low urgency,1,0,14185,,,,0,1eb6e99f622d93cce4a47c5930532cf909edb4474e0708f84a301fb886f08025
thread:issue:141869,issue #141869: Use Python 3.9 type annotations,stale/low urgency,2,stale/low urgency,1,0,141869,,,,0,d0e1e2aa90a45a9e62b1f9e1b683efd7a46344feff148b16505916e2528c731b
thread:issue:141890,issue #141890: Does vmap accept None out_dims?,stale/low urgency,2,stale/low urgency,1,0,141890,,,,0,e0dd222ada1af5b6e28b262f9c4a72b44870da9d892fb1037ee5f25993c379a6
thread:issue:141915,issue #141915: Test failures when migrating to SM89,stale/low urgency,2,stale/low urgency,1,0,141915,,,,0,c07fa7f0243eed59095bf6762e6dac4425f5563f38504edbbea0c0cdc129f25f
thread:issue:141934,issue #141934: [Dynamo] Symbolic variables are reduced to Ints in compiled graph inputs,stale/low urgency,2,stale/low urgency,1,0,141934,,,,0,394db2acc5f55f67deed3c9a04a5db0889bcd2a94df116c5c8b3df067da89642
thread:issue:141963,issue #141963: Do an audit on skipfiles and mark more files as inline,stale/low urgency,2,stale/low urgency,1,0,141963,,,,0,12f206e4486b12b60540c49984ef9acb12c56b171f7f8fc25406cc0be8d22d21
thread:issue:141964,issue #141964: Handling custom context managers,stale/low urgency,2,stale/low urgency,1,0,141964,,,,0,362e0ffc56779f9c00a41df27d5b1fd98a8e830ddc6b866bed7caa1e72a61fcd
thread:issue:141977,issue #141977: frozen modules,stale/low urgency,2,stale/low urgency,1,0,141977,,,,0,173e4830ad78a9a56ae3ebc26c346492689e2da114191506a352f577c2cb4f07
thread:issue:141984,issue #141984: CI feature mirroring or caching for Docker Hub images,stale/low urgency,2,stale/low urgency,1,0,141984,,,,0,ad9d67553714bcf3fbef714be60295da94d1064d29d4c4e6cd0c2a9d1bd83ce3
thread:issue:14200,issue #14200: Failed to run 'bash ../tools/build_pytorch_libs.sh --use-nnpack --use-mkldnn --use-qnnpack caffe2',stale/low urgency,2,stale/low urgency,1,0,14200,,,,0,0cb9b74524713e6e6fab4c9482a2d8b2a9cdc403f2c3df9e88dc065f715757b9
thread:issue:142025,issue #142025: [export] NotImplementedError: No registered serialization name for <class 'diffusers.models.modeling_outputs.Transformer2DModelOutput'>,stale/low urgency,2,stale/low urgency,1,0,142025,,,,0,6818dd83d582d9821fb9beebdeda641f14066d52c4c5478dadf6c8a9efa6e055
thread:issue:142029,issue #142029: [RFC] Add Intel GPU Support to Torch Test Cases,stale/low urgency,2,stale/low urgency,1,0,142029,,,,0,4c60d9f0185ee8207918b4c9c62e03978112f837f44bc8a0d05c6af95f744a33
thread:issue:142104,issue #142104: [compiled autograd] Does compiled autograd under torch 2.4 support using selective checkpoint?,stale/low urgency,2,stale/low urgency,1,0,142104,,,,0,e34b30da85481e6e4cb65f67fef79084bd7ddca41f7f608902a81acda35c3e16
thread:issue:142127,issue #142127: [Bug]need check for duplicate malloc calls in CUDAPluggableAllocator,stale/low urgency,2,stale/low urgency,1,0,142127,,,,0,6fbefab15a6fe7061d3bf40a9ff9806b6ee3d45b6ed7ee6b838d6adc22b38a51
thread:issue:142129,issue #142129: Feature Request - Backport torch.special.gammainc for Backward Compatibility,stale/low urgency,2,stale/low urgency,1,0,142129,,,,0,855ff095077e60b38453bb1345210d916eef98f531e62d388d9a8b06f6f2069f
thread:issue:142145,issue #142145: Fix disabled 3.13 cpp_extension test failing due to registration error,stale/low urgency,2,stale/low urgency,1,0,142145,,,,0,4baaff33630a33746799299b934a16b5791f831449094c29fa8983f114a2ab77
thread:issue:142146,issue #142146: Inconsistent Presence of 'out' Arg in cross_entropy_loss vs Other Loss Terms in Python Bindings,stale/low urgency,2,stale/low urgency,1,0,142146,,,,0,81e78cf0e3d1824da4ed4fa794bd455266aa218c885ac1d391d4c235eec96857
thread:issue:142171,issue #142171: [3.13] Fix failing module tracker dynamo-wrapped test due to recursion error,stale/low urgency,2,stale/low urgency,1,0,142171,,,,0,2546e66b41343af9b329d0ed46037b0e14d15a7f59f4d2cd4519714da2e26df8
thread:issue:142174,issue #142174: [3.13] Fix skipped numpy 2.0 dynamo-wrapped tests,stale/low urgency,2,stale/low urgency,1,0,142174,,,,0,610ea7304c2c0ac92546be51f0055af70dafd1aae393c9874ed856a0c15f23da
thread:issue:142176,issue #142176: torch.fx.GraphModule constructor doesn't copy all parameters/buffers from source graph module,stale/low urgency,2,stale/low urgency,1,0,142176,,,,0,429c9eba69988dcf645059f0be2e80d777e7d1484d568f2daaaaeb949a8989b3
thread:issue:14219,issue #14219: C++ model load error,stale/low urgency,2,stale/low urgency,1,0,14219,,,,0,93b1eb56bd8fdae5d77aae717642a9b426baf5a9f4fea5941c65f2449f10e1b2
thread:issue:14222,issue #14222: Caffe2->ONNX conversion issue due to spatial-bn,stale/low urgency,2,stale/low urgency,1,0,14222,,,,0,f7e92ecc0a0782e22fc31d68663b8a1c73c52ebb9f7e55aaa4e5dcc2ec37906d
thread:issue:142229,issue #142229: [Pipelining] Backward pass tries to detach view tensor in-place,stale/low urgency,2,stale/low urgency,1,0,142229,,,,0,78fee3d64249936873a3afbd02251807b95cd61528ea2c02d0d336ec40432a91
thread:issue:142232,issue #142232: torch.tensor.to(‘cuda’) so slow in jetson orin,stale/low urgency,2,stale/low urgency,1,0,142232,,,,0,3451e781283e819eae5fb21ce8a09845f0158bbf345b0f95c923170d4d8b4187
thread:issue:142290,issue #142290: RFC: Display Supported Versions in API Documentation#,stale/low urgency,2,stale/low urgency,1,0,142290,,,,0,d8dda52117bf2fbfbf2344dc2d9edbd355ac9b3e1a3e17f82808b2c99193bdfc
thread:issue:142301,issue #142301: Unfold,stale/low urgency,2,stale/low urgency,1,0,142301,,,,0,e4c96382e1b07b09d81447c2ae50d0b6e966c8348494bb17fef846c470d7e90e
thread:issue:142331,issue #142331: torch.is_vulkan_available() false in windows,stale/low urgency,2,stale/low urgency,1,0,142331,,,,0,dfc777144c4955c1739cd2ff1503f5c9d6fd4ae9ee7c84d058405e697245350a
thread:issue:142353,issue #142353: Extend custom op fwd/bwd functions to support data types beyond the basic set,stale/low urgency,2,stale/low urgency,1,0,142353,,,,0,4d9f917be1461dcbd1c2e308631200b8f840a55fd658a18c500d425be66a6f8d
thread:issue:142395,issue #142395: [Dynamo] Do an audit on skipfiles and mark more files as inline,stale/low urgency,2,stale/low urgency,1,0,142395,,,,0,d47694b5a7ceabc9fb51d0622e34c2e1e26bd7c5a59a40b3b47db294cdb446fa
thread:issue:142414,issue #142414: [dynamo] Tracking issue for multiple inheritance support,stale/low urgency,2,stale/low urgency,1,0,142414,,,,0,727a8e4c9705282b0047b8e20efb99be9c37cb58fd7e9e66b959700c4d259f35
thread:issue:142467,issue #142467: Wrong error message,stale/low urgency,2,stale/low urgency,1,0,142467,,,,0,fec77db67d9f4608f538edbef4f3d2f4bf2059a9fba1f0b25405e935c67e9f59
thread:issue:142469,issue #142469: Using DeviceMesh Slicing with torch.compile(),stale/low urgency,2,stale/low urgency,1,0,142469,,,,0,795c058e01b4340fadc37d975025f7bbec023f9b43e7e7fdd89675141f693650
thread:issue:142484,issue #142484: `RuntimeError: MKL FFT error: Intel oneMKL DFTI ERROR: Inconsistent configuration parameters` in `TestJitCPU.test_variant_consistency_jit_fft_fft2_cpu_complex64` against mkl 2024*,stale/low urgency,2,stale/low urgency,1,0,142484,,,,0,0bb0af29a637dfc6d5f9659c9bdc286f0f09c4fad0fce1c13e502f33262f6eb0
thread:issue:142493,issue #142493: Custom ops are not logged to dynamo_compile/TORCH_TRACE when an error occurs,stale/low urgency,2,stale/low urgency,1,0,142493,,,,0,7ba0b29daf0f83e54445323a76c2b4a1b3e026873bb832fbaf4b1469cf03568d
thread:issue:14262,issue #14262: [Caffe2] How can I use detectron with pytorch?,stale/low urgency,2,stale/low urgency,1,0,14262,,,,0,87b5a8977d751a9ff264f081c441cfac8303a3e5ac954d343bceeec428aa6f66
thread:issue:14277,issue #14277: Provide Protobuf library if libtorch was built with included version,stale/low urgency,2,stale/low urgency,1,0,14277,,,,0,703115762fb11f6f6590f23b625873aadd030b8ad527d40c50e3b10e167e75e4
thread:issue:142815,issue #142815: torch.linalg.solve fails on CPU with multiple threads and batch dimension,stale/low urgency,2,stale/low urgency,1,0,142815,,,,0,6b8a5e483e1c2ba7a4df0ae8da0a746ef8636a54f671735103febe9f87e7a78d
thread:issue:142825,"issue #142825: Add [low, high] interval for torch.distributions.beta",stale/low urgency,2,stale/low urgency,1,0,142825,,,,0,5058946ec9f97e39e833d17aff8d25b5481be84b401fc7b9bd10909908f8b451
thread:issue:142840,issue #142840: Document spectral norm non-support on non-diagonalizable inputs.,stale/low urgency,2,stale/low urgency,1,0,142840,,,,0,a255118300d623cf6453247eb2dae9625ff1d256cb5d2444f8630add1777d388
thread:issue:142852,issue #142852: Deprecate `torch.squeeze()` with unspecified `dim`,stale/low urgency,2,stale/low urgency,1,0,142852,,,,0,55278c845ac59af24ba6c363aa67b431850bdde4d27fe7e83403a82d864cfc6c
thread:issue:142875,issue #142875: [user empathy day] No speedups for SAM2 model,stale/low urgency,2,stale/low urgency,1,0,142875,,,,0,294c64b1adb4582249fd2c8029191605461c4e3c564994294defdc13599554ca
thread:issue:142884,issue #142884: DataLoader hangs when object fails during pickling,stale/low urgency,2,stale/low urgency,1,0,142884,,,,0,98ac510c0cdd1933a9c3a65947f8bbe884ac9970559ea826d7e70637e2d5682f
thread:issue:143041,issue #143041: Use labels to create dedicated pools of non-scalable runners for trunk jobs,stale/low urgency,2,stale/low urgency,1,0,143041,,,,0,d3e78d85736de69f626395a08a5493e9692cfe7ca5cc9cd126fdcbefef1c3890
thread:issue:143079,issue #143079: PyTorch depends on custom actions stored in contributor personal repos,stale/low urgency,2,stale/low urgency,1,0,143079,,,,0,590a55a9463df951c51335a5a7cc5f12c8c514a2f7775477dbd4c17749f674f5
thread:issue:143109,issue #143109: Failed to build Pytorch in Alpine Linux based image,stale/low urgency,2,stale/low urgency,1,0,143109,,,,0,fce14de03bdfc98cdb4cc4465125404ebda79a52516d825f0e0e71cb23123a9d
thread:issue:14312,issue #14312: [Caffe2] Cannot get repeated argument in custom operator in CUDA context,stale/low urgency,2,stale/low urgency,1,0,14312,,,,0,738f3a0faa2f215696cffec985b3ed29d14f4533bed6261213935f64de688b55
thread:issue:143176,issue #143176: Compiled Autograd + Activation Checkpointing/Offloading,stale/low urgency,2,stale/low urgency,1,0,143176,,,,0,7ebcb004c57ea41e5ef13c15e60b94cb764ee77adeadc2341753dcb7537b765c
thread:issue:14318,issue #14318: No code example for AdaptiveLogSoftmaxWithLoss,stale/low urgency,2,stale/low urgency,1,0,14318,,,,0,d5fafe82719c36c91cff5438b2dfac9523e7ccc27889a4b93886e942774b191b
thread:issue:143189,issue #143189: `torch.linalg.qr_ex` and `torch.linalg.lstsq_ex`,stale/low urgency,2,stale/low urgency,1,0,143189,,,,0,c31470cdad7faebf009809f64bbb87dec107870c0d9a64e157a2f5680fc9f68a
thread:issue:14320,issue #14320: [caffe2] Fails to build with fbgemm enabled,stale/low urgency,2,stale/low urgency,1,0,14320,,,,0,8933bc4c05fee9b0d849ac436adfb42c256c8f5ef88f178b53dd4fdc051f418b
thread:issue:143206,issue #143206: Define release branch creation readiness process,stale/low urgency,2,stale/low urgency,1,0,143206,,,,0,ec4cba1ebccaf3b44ab6f64776393471dcb9465f25b69ae3197d0a282bd9f522
thread:issue:143208,issue #143208: We should switch Tensor object (and other perf-critical objects) to use the new managed dict feature from CPython,stale/low urgency,2,stale/low urgency,1,0,143208,,,,0,545c10ee33adff34a18e0c52bf0d48473f260e538718f6d0caf23c4d61e0dfe0
thread:issue:143215,"issue #143215: Runners, torchbench, & the future",stale/low urgency,2,stale/low urgency,1,0,143215,,,,0,418f24fe00d1f8653bbeeecee4ca4b35a20fe912e620310ad69b01a6218261f7
thread:issue:143274,issue #143274: torch._logging.set_logs kind of sucks for Jupyter notebooks,stale/low urgency,2,stale/low urgency,1,0,143274,,,,0,9a3ec4a11b3c4c8f6f1e6733f2be20512516c934e93d8149b32a5536ef58b761
thread:issue:14328,issue #14328: [Caffe2] How to use the euclidean loss (L2) as output of a CNN model?,stale/low urgency,2,stale/low urgency,1,0,14328,,,,0,5c6234848185a23cdc7bac015f81a6e2ab8430cd85fe2dc5c825c9b448dbe0ee
thread:issue:143293,issue #143293: `bias=False` fails in `Transformer` when `batch_first=True` and in eval mode,stale/low urgency,2,stale/low urgency,1,0,143293,,,,0,b54059b0812343c6907789b54d7e62ee1ac1075b09208528108b75d3e55ae0c8
thread:issue:143306,issue #143306: Add CPU scalar support in addcdiv,stale/low urgency,2,stale/low urgency,1,0,143306,,,,0,0222f6fb1ebd3b6629c9859d9a888a5f6cb7fbde53de542b54ec31be34f74f82
thread:issue:143334,issue #143334: Applying python 'next' directly on a tensor is painfully slow,stale/low urgency,2,stale/low urgency,1,0,143334,,,,0,4f000d62109921f2c520c3da1fc0d93cceceafdd79bb3d9aa687e25ece464ba4
thread:issue:14337,issue #14337: Caffe2 C++ tutorial is not working,stale/low urgency,2,stale/low urgency,1,0,14337,,,,0,34fa3ae8ddc35a94dccb590f05bb17af14b5bc6967471522e278b686e34783ee
thread:issue:143424,issue #143424: higher rank convolution,stale/low urgency,2,stale/low urgency,1,0,143424,,,,0,b4d9d635e27b3f7694cee2fc07f75fc83c4ebc3eaca26bd6b6e5151e58d3bdf5
thread:issue:143434,issue #143434: Missing nightly 20241217 on x86_64,stale/low urgency,2,stale/low urgency,1,0,143434,,,,0,c9b1a03219883435482e8715674ae21e48840acb499c3cd21e2a6a82996b106c
thread:issue:143440,issue #143440: Locale issues in colab: after tensor(1j).cuda().abs() !commands cannot be executed.,stale/low urgency,2,stale/low urgency,1,0,143440,,,,0,7c025e0120e53eb556f435c6b16abdf7b564eafbcc0e2715bc5b6104bfb8f014
thread:issue:143453,issue #143453: Compiler Bisector Improvements,stale/low urgency,2,stale/low urgency,1,0,143453,,,,0,177c822e8643b797f72641f31993523aa80292ad1d7c1a04f8bf47774cfaaa13
thread:issue:143466,issue #143466: Support Dict Parameter Type for custom_op,stale/low urgency,2,stale/low urgency,1,0,143466,,,,0,b2debcd2fe1a90e7ef944cd807045df3af4ae7c3b1163ee288164df2e36adff1
thread:issue:143474,issue #143474: [ONNX] Failed to export PyTorch-2-Export-Quantized model to onnx,stale/low urgency,2,stale/low urgency,1,0,143474,,,,0,115a0956d3a9bbbf92e8099f65b26ef05856e09db86bb302548b2110803044fe
thread:issue:14348,issue #14348: [Caffe2] Error when loading a leveldb dataset using brew.db_input (Error protos.protos_size() == OutputSize().),stale/low urgency,2,stale/low urgency,1,0,14348,,,,0,ae83b221cdf60335a4bef79eacb550d2587e688ac6ee3a77e1266f8ac7ef65d8
thread:issue:143497,issue #143497: No reproducibility after ONNX export of fully converted QAT model,stale/low urgency,2,stale/low urgency,1,0,143497,,,,0,17cba0d7ab8492eedcd44d09851abf4b4f9fabbffc31cc1f2e8e4f06f2eaebaf
thread:issue:143561,"issue #143561: ""aten::any"" and ""aten::all"" behavior for unsigned tensors different than signed",stale/low urgency,2,stale/low urgency,1,0,143561,,,,0,9f8113bb461c1e5669901a3282dcd9ae59c4e0d534dcce761aa7ef0ef70965eb
thread:issue:143564,issue #143564: [Bug] Memory leak in C++ libtorch,stale/low urgency,2,stale/low urgency,1,0,143564,,,,0,cc6c9a6225b79ee5ac7975669bf8fc0f6fc505af738efc2dc4d07854757f4b6d
thread:issue:143571,issue #143571: Issues linking to libtorch on M2 mac,stale/low urgency,2,stale/low urgency,1,0,143571,,,,0,ab5f3cca1217e8be1fa3ae537bd839e70d583df47c96bfddff05b12f89c0e150
thread:issue:143655,issue #143655: 反射填充的反向传播没有确定性实现,stale/low urgency,2,stale/low urgency,1,0,143655,,,,0,fe534f23d246f8b7e1b4e60b46c0d3cd47028554a2a0d9fb8e531ebb82b13d32
thread:issue:14366,issue #14366: How to use model.net.Clip?,stale/low urgency,2,stale/low urgency,1,0,14366,,,,0,ee9770672967a05bfd4b71f2a5ddb97f0f9a05370b13a1e8098d72a1e9d83746
thread:issue:143669,issue #143669: String representation of nn.MultiheadAttention should contain arguments,stale/low urgency,2,stale/low urgency,1,0,143669,,,,0,e58788087bcebb1901e1b3f4ec4fced520dace57562bfd6136f714ff50152fbe
thread:issue:14367,issue #14367: Error: DeviceGuardImpl for cpu is not available (static linking PyTorch),stale/low urgency,2,stale/low urgency,1,0,14367,,,,0,f55ba1ef45d3e1759603ae6117c2f47d81bb1e1dd5ba4408d4f90996df2ebb2e
thread:issue:14370,issue #14370: [Caffe2] Check failed: output->size() == values_.size() output size: 1 given size: 1563551,stale/low urgency,2,stale/low urgency,1,0,14370,,,,0,d439531e719083d5cb40d7f99dd0cfce72bf0598feea5b67dddeb34f350c75bd
thread:issue:14372,issue #14372: [c10d] Configurable timeout per operation for MPI backend,stale/low urgency,2,stale/low urgency,1,0,14372,,,,0,6ddc4c586bec853f0c90067a5d71c107e9028a0a25d27648df582f94238038af
thread:issue:143754,issue #143754: nn.LayerNorm is slower than naive implementation when dimension is low,stale/low urgency,2,stale/low urgency,1,0,143754,,,,0,a13c5064ea9cf63e4199198e66f237e6ccf1d7ebded209f8e03e283cb3062f65
thread:issue:143759,issue #143759: compiled autograd tests should use expecttest,stale/low urgency,2,stale/low urgency,1,0,143759,,,,0,de52deb04d8b01e4c548bf82f6f2b313410ec95fcd89a26ec152d8f0913bc15b
thread:issue:143774,issue #143774: CUDA error when compiling loss function,stale/low urgency,2,stale/low urgency,1,0,143774,,,,0,4bb2583329492ac5bc94f5704f67cf23f99409d0596d5041f9b22cb7c6f33def
thread:issue:14380,issue #14380: [caffe2] Caffe2 GlobalInit should be run before any other API calls,stale/low urgency,2,stale/low urgency,1,0,14380,,,,0,637f9a93e99138d85dfeb7db37c71107791f0b084f99cc40c0b7cdbede383f69
thread:issue:143836,issue #143836: Integration of AdamCPR Optimizer into PyTorch,stale/low urgency,2,stale/low urgency,1,0,143836,,,,0,ea3aa3a6533e06048954709031183ce6d8fbe172db0ba21169aa5ce1d16cefdb
thread:issue:143860,issue #143860: _transform_bias_rescale_qkv cpu op get error on debug build,stale/low urgency,2,stale/low urgency,1,0,143860,,,,0,c909f954c65a5651a61a9a804b494b84ddd91cdab4159459b2034bab4dd12fe6
thread:issue:143864,issue #143864: Adaptive pool MPS,stale/low urgency,2,stale/low urgency,1,0,143864,,,,0,e48e5f4171da3d6baad3c5aebb6fef76958ccca2cedb13ace282883fe7f4c5ba
thread:issue:143931,issue #143931: The model compiled with torch.compile encounters an error when run.,stale/low urgency,2,stale/low urgency,1,0,143931,,,,0,59d35b20c90f4ff7dabe2eaa7f798e134324c80266df2a7cdfffc1af2017aec1
thread:issue:143947,issue #143947: Add differentiable flag to SGD,stale/low urgency,2,stale/low urgency,1,0,143947,,,,0,262117822a91dbb76a0540ae9daa981a3d81d0e507f22f56742d4ef2c2934ce0
thread:issue:143955,issue #143955: Issue with aten::_sparse_coo_tensor_with_dims_and_tensors on Apple Silicon GPU (MPS) Backend when Using Whisper Model,stale/low urgency,2,stale/low urgency,1,0,143955,,,,0,dd5a945d6b2b1098ee89c5276d6cfe67e79621a90641500c66c8e75e8826d72c
thread:issue:143965,issue #143965: test case got killed file_based_local_timer_test.py test_get_timer_recursive,stale/low urgency,2,stale/low urgency,1,0,143965,,,,0,701e7a6ac8239c9d3de7764e5dbe747ca996c760e7c8c6ce75752343ef1e9f0e
thread:issue:143969,issue #143969: [distributed] parallelize_module error with `SequenceParallel`,stale/low urgency,2,stale/low urgency,1,0,143969,,,,0,f1b62426065fbcf67a7a09f320ee6b55c26e7de06d90a0d0283cc9375ab6fbe6
thread:issue:144001,issue #144001: torch.utils.flop_counter.FlopCounterMode,stale/low urgency,2,stale/low urgency,1,0,144001,,,,0,48bbeace507cecd95ab435db0675392705d72861572b5694b61e2ad1aafa525d
thread:issue:144025,issue #144025: torch.cuda.empty_cache() causes extra memory usage on 'cuda:0',stale/low urgency,2,stale/low urgency,1,0,144025,,,,0,10e95a4b31e0aea547505db04bd26fc62e1b6d432a889c1be33f329081fcf13b
thread:issue:144033,issue #144033: Training fails with Torch 2.1.0 on Nvidia Jetpack 5.1.2,stale/low urgency,2,stale/low urgency,1,0,144033,,,,0,c15c5e4e7730c86eac7850e869ed42ebc55acb4fbfd073033481c9a6fd239815
thread:issue:144068,issue #144068: redundant recompilation caused by duplicated Sym(),stale/low urgency,2,stale/low urgency,1,0,144068,,,,0,41d9527367cca365996d9528709106419e09a26bfe0085d9c75f6ec813d104db
thread:issue:14408,issue #14408: DeviceOption::set_device_type() doesn't accept caffe2::CPU anymore,stale/low urgency,2,stale/low urgency,1,0,14408,,,,0,1244c6917b6eca74ac3e7e7ff4f80d17265604d5e7c53ab908ba8b85176c02ec
thread:issue:144109,issue #144109: Uneven Sharding in DTensor Leads to unexpected tensor resolution with `full_tensor`,stale/low urgency,2,stale/low urgency,1,0,144109,,,,0,c40f9677a3b2652458fdae72ca1a1f6f970186ba035f9ab39eca820a28c97815
thread:issue:144116,issue #144116: RNN batch_first argument only works on the input not h_0 when if should work on both,stale/low urgency,2,stale/low urgency,1,0,144116,,,,0,7ec99ebf6b5d578d4ed665e52b4ec1a07c65d8d5702a1bb0e004e6442893e3b0
thread:issue:1442,issue #1442: Optimizers can't be moved to a different GPU,stale/low urgency,2,stale/low urgency,1,0,1442,,,,0,252c0d19c77f85c393ec0f1e3f9fc8bc09aff045ca1e40795a3adf18ba5b8449
thread:issue:144210,issue #144210: Add inheritance possibilities,stale/low urgency,2,stale/low urgency,1,0,144210,,,,0,5f9777be70e2d36ab8586f8b291dc3cecee632887509b92c63226d82a8de9e00
thread:issue:144217,issue #144217: PyTorch Enum PYI stubs are invalid. Fix typing of PYI stubs,stale/low urgency,2,stale/low urgency,1,0,144217,,,,0,1f08685daa48b0544e7dd71d8fd235f2a19160f359056031fba8ec0c7ffde8e2
thread:issue:144223,issue #144223: Support LayerNorm2d,stale/low urgency,2,stale/low urgency,1,0,144223,,,,0,e72f099eb9bfc19bb2c47c717d017f0a950383b17d13adc05e6b9d9d3d3a165b
thread:issue:14431,issue #14431: BUILD_CAFFE2_OPS=OFF is not tested in CI,stale/low urgency,2,stale/low urgency,1,0,14431,,,,0,0a5ecdebe2ddd818f4cd346e2e49374ebc373cccf6edf8025cc33ce505cd0e0b
thread:issue:144328,issue #144328: Debug build fails to compile on x86 with WERROR=1,stale/low urgency,2,stale/low urgency,1,0,144328,,,,0,cdabe4215f3bc52c696ea63a9ad7e5c79184bce06bb02646c2fd57954bd7111f
thread:issue:144344,issue #144344: custom_op's backward changes can't invalidate `torch.compile` cache for backward,stale/low urgency,2,stale/low urgency,1,0,144344,,,,0,c44fc46c2535dd13f75bf28b66b1ff1df761d7e0f370e318e474187f2772cc29
thread:issue:144375,issue #144375: Unable to compile models using tensorrt backend: CUDNN_STATUS_BAD_PARAM_STREAM_MISMATCH,stale/low urgency,2,stale/low urgency,1,0,144375,,,,0,db10993161a1654fcfc1710ea944ae2638f75c49481245721ed3a6f28152ba61
thread:issue:144403,issue #144403: Extended functionality for torch.quantization.fuse_modules,stale/low urgency,2,stale/low urgency,1,0,144403,,,,0,602dd618271f331c2b2f83763584533f2d8ef01963ff8daa577fa74dd804688f
thread:issue:144445,issue #144445: NotImplementedError: Output channels > 65536 not supported at the MPS device.,stale/low urgency,2,stale/low urgency,1,0,144445,,,,0,1fe17d71be008929d75dffa153b2f3d59ab873072f15daa45909d58fd2e90ead
thread:issue:144464,"issue #144464: [ONNX] Trying to export QAT model with torch.onnx.export, running into Problems that workarounds cant fix. Is ONNX export planned for QAT?",stale/low urgency,2,stale/low urgency,1,0,144464,,,,0,14220ebcc893e98c87ca98cb028094eb7518ed0683473e460cd43e6484efe10c
thread:issue:144470,issue #144470: Partitioner's auto-AC misbehaves with mixed dtypes,stale/low urgency,2,stale/low urgency,1,0,144470,,,,0,aa7157ec9fd1536f4ee5562bfd6247d16956458f7b78d5fb32a3b3ac3edd370c
thread:issue:144478,issue #144478: [dynamo] Support using UserDefinedFunction as argument (as_proxy).,stale/low urgency,2,stale/low urgency,1,0,144478,,,,0,be6ec19df67bbb9f84173cd12e8c91214cc401acb26c765e1c862d976e8662ac
thread:issue:144495,issue #144495: c10/util/BFloat16-math.h has undefined behavior,stale/low urgency,2,stale/low urgency,1,0,144495,,,,0,03b3dfd1ea4b0bd0030c8a6f07e85c84717b6fd140ccf7cc196bd1aedc6accfe
thread:issue:144530,issue #144530: [Pipelining] PP+DDP does not work for Zero Bubble,stale/low urgency,2,stale/low urgency,1,0,144530,,,,0,a02d7017fb152abd26860c640ab55c237a2593f7fe9792f76c4f017997a49b18
thread:issue:144564,issue #144564: Tabulate not official dependency of PyTorch but needed by features like FlopCounterMode,stale/low urgency,2,stale/low urgency,1,0,144564,,,,0,0a9be9a2a6a354b2dea8d6eda47c4a6076b884a9eede29a458a51b4cdb4e329f
thread:issue:14459,issue #14459: [Caffe2] Error protos.protos_size() == OutputSize() when loading dataset created by regular Caffe (datum),stale/low urgency,2,stale/low urgency,1,0,14459,,,,0,da57ef0190d4c57b1589db5d25aaf90715e0d7b9a6b4a65ae5e13d0a7bc14cc5
thread:issue:14461,issue #14461: The speed of scatter is influenced by the data size while using nn.DataParallel,stale/low urgency,2,stale/low urgency,1,0,14461,,,,0,8561afa8155502e48a4f26bd20d929299069e368c732b02f46b67b8a0d29da52
thread:issue:144611,issue #144611: [CUDA] Illegal Memory Access with `ConvTranspose2d`,stale/low urgency,2,stale/low urgency,1,0,144611,,,,0,ff529621b625060b870d1a018853e7d44c608ea5ecc089e91e6e7e73dab99384
thread:issue:144612,issue #144612: [CUDA] Illegal Memory Access with `torch.bmm`,stale/low urgency,2,stale/low urgency,1,0,144612,,,,0,fd76c5f43c70cd93281cdd7af9f7d29cd994c0ced053701fc19bb48e1f12e8c2
thread:issue:144615,issue #144615: `collect_env.py` fails with `'NoneType' object has no attribute 'splitlines'` if pytorch is installed without pip,stale/low urgency,2,stale/low urgency,1,0,144615,,,,0,b1ab205671ad8b0dfe0c43049d3ccecc3d663855acb331b6ecb729711395f484
thread:issue:144639,issue #144639: torch.vmap + autograd.Function + current_level bug,stale/low urgency,2,stale/low urgency,1,0,144639,,,,0,1c91efca8d9d75bcc840c407b7671a552936ef56ee7f76d68b2ec69407849144
thread:issue:144643,issue #144643: nn.functional.interpolate doesn't work correctly with NCW tensors when output W exceeds 2^23,stale/low urgency,2,stale/low urgency,1,0,144643,,,,0,1548dd25a0d4ebe0c61754c04dd959b47dac886fe14343f7e34882bc8bff37e5
thread:issue:144660,issue #144660: Implement the `mode` property for transformed distributions,stale/low urgency,2,stale/low urgency,1,0,144660,,,,0,5dc1577bbc65e067ce93076f5f515c571f5cd61a635c73f4fd0767fb03843078
thread:issue:14467,"issue #14467: Optimizer warning when parameters ""change""",stale/low urgency,2,stale/low urgency,1,0,14467,,,,0,5d9683872914aaad6c5a57aa395ffe6c87cd9b01f89be5d9bd8d53d68ecd4491
thread:issue:144671,issue #144671: torch.stack for sequences,stale/low urgency,2,stale/low urgency,1,0,144671,,,,0,5a348be18e09e75c1aab4152e35ee9e3148c05923fe94c24304e06becbd9ed10
thread:issue:144677,issue #144677: Tensorboard SummaryWriter.add_hparams doesn't log hparam metrics dictionary,stale/low urgency,2,stale/low urgency,1,0,144677,,,,0,594b3691fb42a84048d8a805570fb95adc3f761e54c6b520f6eda0202cab03fa
thread:issue:144688,issue #144688: topK for sparse Vectors,stale/low urgency,2,stale/low urgency,1,0,144688,,,,0,19bcaa363a7cd9ffa3f67640fcbb760b8f1c71e8795be33ce4eab5f0eb7e25cc
thread:issue:144690,issue #144690: Support new CUDA conda package layout natively in cpp_extension.CUDAExtension,stale/low urgency,2,stale/low urgency,1,0,144690,,,,0,7f84e6fd38c4030eb446f95fe4b18e45cf9d4a8ac15c9268ad86cf71f561e184
thread:issue:144711,issue #144711: Tensor Stride Inconsistent (?) Behavior When One of the Dimension is 1,stale/low urgency,2,stale/low urgency,1,0,144711,,,,0,10ff2a6a5e537f7163a117c18f715cf7b0cefabf9a540ed3cf041b6af8003db5
thread:issue:144766,issue #144766: Matmul with int32 parameters on Intel GPU leads to errors,stale/low urgency,2,stale/low urgency,1,0,144766,,,,0,690ffd18753790246dc4c39d73ed71294ee11bb976182952d19aac0265329962
thread:issue:144767,issue #144767: Unable to build with ATEN_THREADING=TBB option,stale/low urgency,2,stale/low urgency,1,0,144767,,,,0,8f8789dfa299d4daafd255d7ee8e691e18c91961821f7666d287c184159a8c1b
thread:issue:14477,"issue #14477: Add a debug mode which is -O0 for framework code, but -O for kernels",stale/low urgency,2,stale/low urgency,1,0,14477,,,,0,18752764b2a5e07d5cee90c1ba954b365c9638bcb2d7e1759f5d0d20b6b586e2
thread:issue:14478,issue #14478: Flaky download from files.pythonhosted.org when installing botocore,stale/low urgency,2,stale/low urgency,1,0,14478,,,,0,4fde4bf9b6d6bfc736582eff58f31fcda82449179e39a6d5893fe89bab16ca5e
thread:issue:144810,issue #144810: Support torch.func.grad for Flex Attention,stale/low urgency,2,stale/low urgency,1,0,144810,,,,0,4e7dd4b4b7270c9148d238aa8b23295c2a5132e9b0422c1154831df6b2cd2b9b
thread:issue:144850,issue #144850: Batching rule for aten::_thnn_fused_gru_cell,stale/low urgency,2,stale/low urgency,1,0,144850,,,,0,93ca19de1dddf871b0911095ca634c9f846ed92a3b97b679b9bc3a854e07f6d6
thread:issue:14486,"issue #14486: building bundled nccl fails in (caffe2, cuda 8, cudnn 7) CI environment",stale/low urgency,2,stale/low urgency,1,0,14486,,,,0,8c511091207f338d230816060b30069ca6e4b70a9864c00ed7cdb54a1d335a1d
thread:issue:144861,issue #144861: Region check for in-place read and write does not always work,stale/low urgency,2,stale/low urgency,1,0,144861,,,,0,251bcd473c5839a5aeb3272c436d1046098cb47fd0c9198b45f5c8eed0a77f03
thread:issue:14487,issue #14487: TestAdagrad.test_row_wise_sparse_adagrad intermittently fails health check,stale/low urgency,2,stale/low urgency,1,0,14487,,,,0,c61ba1e676587f8f168c502fc9c816abccf38673a524b3b9b2c4713ecbb023aa
thread:issue:144875,issue #144875: [Monitoring] Display on HUD the information about runners that failed to be created (which cause jobs to queue),stale/low urgency,2,stale/low urgency,1,0,144875,,,,0,079c5ad915c31f778c2bcb14b0ace8d7326324116265d2ed8afa8e194bb28501
thread:issue:144890,issue #144890: Ways the HUD compilers dashboard could be better,stale/low urgency,2,stale/low urgency,1,0,144890,,,,0,285cb0d8c601ba8c75bfd82f3ffed8c0ae8e0d2e39b161a0071620dcf7dba165
thread:issue:144913,issue #144913: UserWarning: cuDNN SDPA backward got grad_output.strides() != output.strides(),stale/low urgency,2,stale/low urgency,1,0,144913,,,,0,aafa8457ee2f12fbd81effb6aa323881055de8a32ec67e8bb964c514f38b853b
thread:issue:144928,issue #144928: Add API to detect if activation checkpointing is enabled in the current region or not,stale/low urgency,2,stale/low urgency,1,0,144928,,,,0,6e69a1177697b455be4b8fed16f775ca4ef2e5342eafc4c846fb81b09cb94a86
thread:issue:144931,"issue #144931: Error loading ""torch\lib\aoti_custom_ops.dll"" or one of its dependencies, when importing Torch, when building from Source on Windows 11 with cuDNN.",stale/low urgency,2,stale/low urgency,1,0,144931,,,,0,e6d6e8247d16518ee17c22b00964de247a063f2f491c55893884ba4e28d964f9
thread:issue:144955,issue #144955: ModuleNotFoundError: No module named 'torch.privateuseone',stale/low urgency,2,stale/low urgency,1,0,144955,,,,0,e5edbda33c043aa072e06f9927fa6d24694f9e4483ba95119e690bed5cf86a50
thread:issue:144960,issue #144960: Memory-efficient attention is not selected if inputs's ndim != 4,stale/low urgency,2,stale/low urgency,1,0,144960,,,,0,f974f0016b6e017db22a02f1e01e0bd8ff4af965f6558af14b1763798999bd71
thread:issue:144969,issue #144969: Support something similar to export dynamic dims for torch.compile with fullgraph=True,stale/low urgency,2,stale/low urgency,1,0,144969,,,,0,9fd5b84ddf1335f1e9f4e814e6c8cb5ea7965b13c3086ba8ed690571c46313ff
thread:issue:144982,issue #144982: [compiled autograd] It would be nice if the compiiled autograd graph was actually runnable,stale/low urgency,2,stale/low urgency,1,0,144982,,,,0,518e3249e06d89d058f560180670bc321739f9f01347d38a2add960418a9a529
thread:issue:145040,issue #145040: Issue: Illegal Memory Access in Backward Pass of `scaled_dot_product_attention` with Custom Attention Mask,stale/low urgency,2,stale/low urgency,1,0,145040,,,,0,c5572763910fdb89bfbe87e5aee89aa03f1f01c2f2ad396208e8f9374236bdc2
thread:issue:145043,issue #145043: Bug when using reparameterized model evaluating with DDP,stale/low urgency,2,stale/low urgency,1,0,145043,,,,0,f5fc7c60e727b55c853ec127eb53a02011e31082b07cf3e96a3919da5d2e5f41
thread:issue:14508,issue #14508: [Caffe2] GPU test passed. Cannot see on nvidia-smi,stale/low urgency,2,stale/low urgency,1,0,14508,,,,0,d3da8f898553999d7dae19be6616f5557663d86706179196556a512c241a0c3a
thread:issue:145094,issue #145094: Tracking issue: Incorrect Meta Strides / Turn On PyDispatcher in FakeTensor Mode,stale/low urgency,2,stale/low urgency,1,0,145094,,,,0,ab92e621fa5c38898cc5260a1138a68fb976aa1be59622eb6466624195dfe487
thread:issue:145145,issue #145145: [RFC] Improve performance for layer_norm op for cuda with revectorized,stale/low urgency,2,stale/low urgency,1,0,145145,,,,0,07913ce225c5db0f1ad157cc809da44575a85fa69ebfe34b92eb9565a4b5de65
thread:issue:145158,issue #145158: Pytorch matmul for nested 4D tensors in jagged layout doesn't work,stale/low urgency,2,stale/low urgency,1,0,145158,,,,0,52c45546efa8877490671ff625ab73ffb6cd5e4d94d9227213fd0e1e2020f90e
thread:issue:145214,issue #145214: Nested tensor support for pointwise matrix multiplication of nested tensor and normal tensor,stale/low urgency,2,stale/low urgency,1,0,145214,,,,0,fe7c0f1b202cf2e29a920ecb7893eb49ebe9ddbf07dd90414f00b83ce01f0b3e
thread:issue:145242,issue #145242: Expose configurable path instead of using fixed path in the inductor module for serialized pattern generation,stale/low urgency,2,stale/low urgency,1,0,145242,,,,0,6ee666cb2cc5afe5b9e3d3f232ff53c3e9214b09db559175da5f83ad8c8a833e
thread:issue:145262,issue #145262: Missing create_graph arguments in torch.func apis,stale/low urgency,2,stale/low urgency,1,0,145262,,,,0,11865c93b725935b73e01c70ea88ace15b915729565f2a330bb11747a30c0b68
thread:issue:145263,issue #145263: [Pipelining] Problem using `torch.distributed.pipelining` on `Gemma2ForCausalLM`,stale/low urgency,2,stale/low urgency,1,0,145263,,,,0,baff9265c2bf9614593fb74273b2fe50e317f0fc91b9e88b7c2c880973bd8a6b
thread:issue:145326,issue #145326: [inductor][triton] refactor ASTSource.make_ir integration,stale/low urgency,2,stale/low urgency,1,0,145326,,,,0,3cc765fad03a0184395f5409a4ea14ae0d318e0cecfd434761af6cb107cd5a05
thread:issue:145337,issue #145337: [libTorch] Model initialization on multi-device is slow. It seems to run sequentially in multi-thread,stale/low urgency,2,stale/low urgency,1,0,145337,,,,0,b2f80e4b940a3cf0f42f315858babbd865d9010572a7dcab9d50a70a36d95e3d
thread:issue:145349,issue #145349: [CUDA] Illegal Memory Access with `AdaptiveAvgPool2d`,stale/low urgency,2,stale/low urgency,1,0,145349,,,,0,a024f72ee970548064f9f94d1edcc670f2e8982b5aaffa4e6367ff8123321271
thread:issue:145350,issue #145350: [CUDA] Illegal Memory Access with `ReplicationPad2D`,stale/low urgency,2,stale/low urgency,1,0,145350,,,,0,3d90df2e086ee24b70a43740497a69ce8c319baf8b3e7a8d74a77bf4a484567d
thread:issue:145356,issue #145356: [autograd] inconsistent jvp results,stale/low urgency,2,stale/low urgency,1,0,145356,,,,0,8e256b5887c9aff6dc6838015b8d28739b845134bc17c9159b4792fb148f1ee7
thread:issue:145357,issue #145357: FP8: E5M2: The FP8 E5M2 result is not `inf` when casting a FP32 value larger than max normal value of FP8 E5M2 (57344),stale/low urgency,2,stale/low urgency,1,0,145357,,,,0,d918823c3e657b3677c672e64ebc83000ef4b95a7a52f1b7349a24efd3765cfc
thread:issue:145368,issue #145368: The possible error in the pytorch documentation of RNN.,stale/low urgency,2,stale/low urgency,1,0,145368,,,,0,54f25c831e4f1b21ee12c67c3ccec154c71c3b1304ac33f64e3cec109d13aea7
thread:issue:14538,issue #14538: [Caffe2] How to fetch trainable parameters?,stale/low urgency,2,stale/low urgency,1,0,14538,,,,0,ba909dbfa627804e3f90a2f9785a157c94c2bc55606914aa88272fce56e377b5
thread:issue:145388,issue #145388: create DISABLED issues for specific runner labels,stale/low urgency,2,stale/low urgency,1,0,145388,,,,0,66bd37034f9a73f885182d0d0ff64ddb94c3e1225fdb16fef81bb85508254f58
thread:issue:145453,issue #145453: [CUDA] Illegal Memory Access with `AdaptiveMaxPool2d`,stale/low urgency,2,stale/low urgency,1,0,145453,,,,0,5a1c164ab0580bb9fa18632fd1674ef5945a56528f2cf56f701f22962a38fd8b
thread:issue:145468,issue #145468: [inductor][torchbench] Unsupported operator issue when running the torch_multimodal_clip model with batch size 4.,stale/low urgency,2,stale/low urgency,1,0,145468,,,,0,75527f49ae2308579febac9b8768ef0e1906768cf55b7fbdc8b97a76c48e036e
thread:issue:145471,issue #145471: Is there a PyTorch version that can work properly on the Thor platform based on the Blackwell architecture?,stale/low urgency,2,stale/low urgency,1,0,145471,,,,0,03cc65efd8e182598f0d625d7ba3fc79f75ba378abee7a9efce383ec53916a7a
thread:issue:145493,issue #145493: Add a lint rule to avoid the word `interface` in C++,stale/low urgency,2,stale/low urgency,1,0,145493,,,,0,5f68424f14c290b15a26a7cc5ba2f3b7d0fd93db865e3cdf5a6d38f44112dd69
thread:issue:145592,"issue #145592: _pickle.UnpicklingError: invalid load key, ''.",stale/low urgency,2,stale/low urgency,1,0,145592,,,,0,94b667d13dcf4adcd725bad05a02c2f223aa9812bfc5974a8f5a22f51e45569a
thread:issue:145608,issue #145608: ROCm+gcc 15 asserts,stale/low urgency,2,stale/low urgency,1,0,145608,,,,0,9b235202a7816a6928f40ae5a2be398476a71b678387889e16b5ec6913b447c9
thread:issue:145656,issue #145656: Stable C bindings for libtorch,stale/low urgency,2,stale/low urgency,1,0,145656,,,,0,2469f48b9d22de2562a14412de4fb42d9b13002c7e093358b528fe6b6d34cbc8
thread:issue:145665,issue #145665: Pickling duplicates Storage,stale/low urgency,2,stale/low urgency,1,0,145665,,,,0,6cc7dab9c32258daaee9b4da99af257c8a51fc10ac5aa4421238c04fb3be4ded
thread:issue:145716,issue #145716: support arm architectures in your Docker images on docker hub? (,stale/low urgency,2,stale/low urgency,1,0,145716,,,,0,4d170eae60a4856646630c10a30b6216509079af62c4cb98d745e66cb2f5c364
thread:issue:145729,issue #145729: KL div between t-distribution,stale/low urgency,2,stale/low urgency,1,0,145729,,,,0,b9aae280b1f7ffa51341ada12365850ce2ac52cbfa2fe89831f133f53d73e179
thread:issue:145837,issue #145837: torch.nested.narrow() or torch.nested.to_padded_tensor() breaks backwards pass - invalid gradient,stale/low urgency,2,stale/low urgency,1,0,145837,,,,0,9030080c6ef249372dc74377d9036d8968f3804877c215f4710be53102b4d9e7
thread:issue:145851,issue #145851: FIRE Relative Positional Encodings,stale/low urgency,2,stale/low urgency,1,0,145851,,,,0,54904a8dcf1454e03a1eb7b2b8446d3c1fcf66d2b5d24cf6523eb8a6d4877a6c
thread:issue:145907,issue #145907: There may be a documentation Error in torch.nn.CrossEntropyLoss Formula,stale/low urgency,2,stale/low urgency,1,0,145907,,,,0,062b0bd3f0236443247585d7f8af76d8c39ae21fc50e06bfc75523ab06795cce
thread:issue:145943,issue #145943: Add center and padding options to onnx version of torch.stft,stale/low urgency,2,stale/low urgency,1,0,145943,,,,0,1acdd278ca708a55a1132afd51179b4044fc40ade9698f9b6aeb1a7dbea2ec9f
thread:issue:145944,issue #145944: Add align_to_window option to onnx version of torch.stft,stale/low urgency,2,stale/low urgency,1,0,145944,,,,0,0ff7025fe62e46e5095d48171d35400aa1138b49849171009552fd36b7954b11
thread:issue:145947,issue #145947: flops counter error with PyTorch2.5 and 2.6,stale/low urgency,2,stale/low urgency,1,0,145947,,,,0,c2e9832b58ca014466d7addbef9438f2eb8dd96a7b4eb37d5ed97ab7bc862739
thread:issue:145978,issue #145978: What is the recommended way to use Distributed Checkpointing Save/Load with HSDP?,stale/low urgency,2,stale/low urgency,1,0,145978,,,,0,cf56ee14946856fad2fddc169b67bc89b4a09a77dca36955c774d3ef67773097
thread:issue:146021,issue #146021: multi node Error when dist.destroy_process_group,stale/low urgency,2,stale/low urgency,1,0,146021,,,,0,02c2d0235a6fb9e9f2ac18d3bbd5344d8d9bfe85bfe6da22645f8424aad4da16
thread:issue:146034,issue #146034: This lib does not have new types of loss functions like focal Loss,stale/low urgency,2,stale/low urgency,1,0,146034,,,,0,80d7d3d04b8d06dfef98a23e3086ca0e45cd663952943b934f8d9d79ed8fd092
thread:issue:146036,"issue #146036: torch/autograd/graph.py"" torch.func.{grad, vjp, jacrev, hessian} don't yet support saved tensor hooks",stale/low urgency,2,stale/low urgency,1,0,146036,,,,0,61c93d15326be0a1642e512410a1d9814014cc27127aedb164347a2c4d79ee3e
thread:issue:146040,issue #146040: Poor performance of any() on tensors even worse on cuda vs cpu,stale/low urgency,2,stale/low urgency,1,0,146040,,,,0,3017a9e901742f51e9a7beea721852b06dc003246aa3726a071cd129cb52bcd9
thread:issue:146056,issue #146056: RuntimeError: CUDA error: the provided PTX was compiled with an unsupported toolchain. Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions.,stale/low urgency,2,stale/low urgency,1,0,146056,,,,0,2d93daf72a27a5a9508736e1958cd32496c7e0e95fb7ccc83d67fabd767b77a9
thread:issue:146086,issue #146086: Enhancements to dim_order API for Ambiguity Detection,stale/low urgency,2,stale/low urgency,1,0,146086,,,,0,72b3c04e6101b446c70af43f15cb189561421a8f097cb7b5de3742934fdac23f
thread:issue:146096,issue #146096: `torch.nn.functional.conv2d` 8 times slower in torch 2.5.1 compared to 2.3.1,stale/low urgency,2,stale/low urgency,1,0,146096,,,,0,7ae518c1e5d4e29d89ca2c275fd2388a16174707477257d7bd75ff3a3bc6dd23
thread:issue:146118,issue #146118: [NCCL] Segfault in topology path selection when calculating max bandwidth on nightly,stale/low urgency,2,stale/low urgency,1,0,146118,,,,0,0e6b55cc5f8f1cd3d7f8d5b8684a8e194ef277e68bbe658efc37b014f45d854a
thread:issue:146119,issue #146119: File based multi gpu,stale/low urgency,2,stale/low urgency,1,0,146119,,,,0,2de036a151381df4ba2d9b607c35303b0a717730b5401038212106befa9da3c2
thread:issue:146157,issue #146157: AdamW refactoring broke checkpoint reloading with DCP,stale/low urgency,2,stale/low urgency,1,0,146157,,,,0,c92b824bd2616ad505bdde13fe8f5c13cd60addf0c57b2e84750c661c5920f6c
thread:issue:1462,"issue #1462: Change sparse_mask to take indexing mask, rather than entire sparse tensor",stale/low urgency,2,stale/low urgency,1,0,1462,,,,0,e63a36561d380316d510704bbec0c0f064c95210b92dc4c7d7ad0c8ec2f3a157
thread:issue:146221,"issue #146221: Incompatible Torch and Torchvision while building from source for 2.6.0 and CUDA 12.6, RuntimeError: operator torchvision::nms does not exist",stale/low urgency,2,stale/low urgency,1,0,146221,,,,0,a81587b8855bb3620a166f64a794382f6e63f5eadd4a61e007ea314160f1f2dd
thread:issue:146307,issue #146307: Intermittent false positives in lintrunner's mypy,stale/low urgency,2,stale/low urgency,1,0,146307,,,,0,f7886aa1bbc42453ed5570320c2a2eadd524cc2de0a8eccb64160a278705707a
thread:issue:14632,issue #14632: [c10d] Check that allgather/gather output tensors point to different storage,stale/low urgency,2,stale/low urgency,1,0,14632,,,,0,920d37a28b3fe062ac19f920bc6b1c3ebfe590964a5dc9f2d365cc4b4ab44e01
thread:issue:146371,issue #146371: Torchrun does not handle worker failure gracefully,stale/low urgency,2,stale/low urgency,1,0,146371,,,,0,cb4850a0a1735042219c942adf8316061829c2cd8b723535188928368a01078c
thread:issue:14652,issue #14652: fc_without_bias / FCWithoutBias,stale/low urgency,2,stale/low urgency,1,0,14652,,,,0,e522b70adfd92654bb50d2742a0ba2485c05be356e54a0f73311ae68bf7f7186
thread:issue:146566,issue #146566: [DCP] Allow for rank-specific tensors with duplicate keys,stale/low urgency,2,stale/low urgency,1,0,146566,,,,0,d01125e2f0efb3b692c53c69b5f32d9694e4a38d20799695023a4098abb3b899
thread:issue:146570,issue #146570: inconsistency in geometric_ on CPU and GPU,stale/low urgency,2,stale/low urgency,1,0,146570,,,,0,de468e655d6e9b7084d830ec1db36b0094143ff2c76ec5e782745c5de26b277f
thread:issue:146629,issue #146629: [inductor][idea] Defer realize/inline decisions,stale/low urgency,2,stale/low urgency,1,0,146629,,,,0,be1af5a86694a628af090ffa7b284ab63d5a782831a2d5da9b15c5af81916ec4
thread:issue:146693,issue #146693: Partitioner moves useless and memory-heavy op from bwd to fwd,stale/low urgency,2,stale/low urgency,1,0,146693,,,,0,3b4618e5794d53fa429492fdfb01d0a924574284cf2595781ebf6f909da4e41e
thread:issue:146697,issue #146697: There is a performance drop because we have not yet implemented the batching rule for aten::mse_loss_backward.,stale/low urgency,2,stale/low urgency,1,0,146697,,,,0,d4aaec78bd1538e9037718ef018c90544f67f84d99eedf3f1cdb48569303dac5
thread:issue:146699,issue #146699: `#type: ignore` linter,stale/low urgency,2,stale/low urgency,1,0,146699,,,,0,463ba4ba7faad1f35d4b289419f349f38562e430bb7b5bd37aa7a7c1a3fc6a95
thread:issue:146701,issue #146701: Unable to Import PyTorch After Upgrade in Docker Environment,stale/low urgency,2,stale/low urgency,1,0,146701,,,,0,adf23953c733252180d8dd7dd7755e5a9e128fc0b2d7727c246bad78f816ae89
thread:issue:146707,issue #146707: AutoAC can cause deadlocks with tensor parallelism and data-dependent flop formulas,stale/low urgency,2,stale/low urgency,1,0,146707,,,,0,3f98c723f3206dba09d9d565bd77fd78c77462863eb57e11649a5bf73ce79bb4
thread:issue:146708,issue #146708: Periodic Activations Module,stale/low urgency,2,stale/low urgency,1,0,146708,,,,0,5e94ccf576d262dc753e228e91de53f8f0e65dc5be276a3e58c53a2658413640
thread:issue:146711,issue #146711: Add different padding modes to torch.nn.utils.rnn.pad_sequence,stale/low urgency,2,stale/low urgency,1,0,146711,,,,0,106672f3152bfb8fa334f9b3b131f22629d56fa9fbab2ed46d382f51f8adf5a2
thread:issue:146732,issue #146732: increase lwork/rwork sizes for all float->int conversions,stale/low urgency,2,stale/low urgency,1,0,146732,,,,0,1b1bf49d0bee2a1df64df346db77ecc745843cc7f5ff6467d576a3569a141ecf
thread:issue:146765,"issue #146765: Automatically resolve tensor mismatch issues, tensor conversion, and moving tensors to devices",stale/low urgency,2,stale/low urgency,1,0,146765,,,,0,4dfc8c3f7a4fbeb96ec5a7f5182d54782d1a3f9d9d9c146e94897364baa0744e
thread:issue:1468,issue #1468: avg_pool functions hold input for backward,stale/low urgency,2,stale/low urgency,1,0,1468,,,,0,2083b0eb3ca59e95e4125f296e66405be982ab7c14ed4e848363ec2e0e9cbc27
thread:issue:146831,issue #146831: Missing Typing for Loss Functions Causes Poor Code Completion in VSCode,stale/low urgency,2,stale/low urgency,1,0,146831,,,,0,94bf47f804b9d2837677a9fb26c96d2681f852602afabeb22ce30894d507f757
thread:issue:146848,issue #146848: _scale_dot_product_efficient_attention eager/meta computation mismatch on output tensor shape,stale/low urgency,2,stale/low urgency,1,0,146848,,,,0,368b3501e548fbe7627c7db66952610b41aeafaf753b558f1187240a2dfc6e43
thread:issue:14685,issue #14685: [Caffe2] Exception when creating gradient for [Cast] SquaredL2Distance as output layer of CNN network,stale/low urgency,2,stale/low urgency,1,0,14685,,,,0,527abe7718fc7ab47530c2e76530eb0afd3d474fb91eea52266a9c5cecade1a7
thread:issue:146885,issue #146885: Add numpy.core.multiarray.scalar to default safe globals,stale/low urgency,2,stale/low urgency,1,0,146885,,,,0,cec30e786215c4cc33e5141573a3eaf018f0751ab24eeb148c343d47e07d7dc3
thread:issue:146889,issue #146889: How to customize a torch.Tensor() method to access the underlying data structure of a PyTorch tensor.,stale/low urgency,2,stale/low urgency,1,0,146889,,,,0,0762ca8c543c50933a2553ee1b972b237e7794b5753740fbb5584860e012da3f
thread:issue:146901,issue #146901: Trying to use forward AD with _scaled_dot_product_efficient_attention,stale/low urgency,2,stale/low urgency,1,0,146901,,,,0,66370dfb4507aada7d3b6cd65f1ecc177b3698ad62831eb08ad4e3fd0fcb4343
thread:issue:146910,"issue #146910: INTERNAL ASSERT FAILED at ""/pytorch/aten/src/ATen/native/sparse/SparseCsrTensorMath.cpp"":817, please report a bug to PyTorch.",stale/low urgency,2,stale/low urgency,1,0,146910,,,,0,468f29f487d44cb6d4d2caa698e4b4f930769cf08d2e40da7d97ee0894c38f42
thread:issue:146913,issue #146913: Add a `TORCH_LOGS_RANK=0` env var that integrates with `TORCH_LOGS`,stale/low urgency,2,stale/low urgency,1,0,146913,,,,0,760ebe1b924d187e3d5257f6a735356087708d3211a29c83a237851b7c14add9
thread:issue:146930,issue #146930: pytest test/dynamo fails,stale/low urgency,2,stale/low urgency,1,0,146930,,,,0,be4508647534ea80cf59f97059d826a4165c9e47f5b5f706e0597bc5703583bf
thread:issue:146944,issue #146944: win-vs2022-cpu-py3 test failures in test-default-2-3-lf.windows.4xlarge.nonephemeral_37054642004,stale/low urgency,2,stale/low urgency,1,0,146944,,,,0,0baa236ccf28617eda1be8cb10bbb70da16e35c4cbcce9f45e2ed68a0100f9b5
thread:issue:146945,issue #146945: New CachingAutotuner pickling logic may be brittle to triton upgrades,stale/low urgency,2,stale/low urgency,1,0,146945,,,,0,d8a064093bd6391defa10141124139d1ab23f1e647f3c7b98f59869dc7aada01
thread:issue:146956,issue #146956: [c10d] Consolidate watchdog threads,stale/low urgency,2,stale/low urgency,1,0,146956,,,,0,1cf9e00092fb8258d122e564c58218494eee86312e763b2c3d574a7148653a34
thread:issue:146964,issue #146964: Feature Request: Interface to Check Compilation Status Inside Compiled Module,stale/low urgency,2,stale/low urgency,1,0,146964,,,,0,d13e9e30ec39afd95ab841dabc7fa6c3e9bc2d86fedbd012f65170a1bb31e93b
thread:issue:146976,issue #146976: Adding sparse_coo_tensor objects together behaves differently with CPU and CUDA,stale/low urgency,2,stale/low urgency,1,0,146976,,,,0,fa6d148b5a75be07e064cc165eee2ea961e2a75bef846b5d2496b2fa7525041e
thread:issue:146982,issue #146982: Porting Pytorch to AIX Operating System.,stale/low urgency,2,stale/low urgency,1,0,146982,,,,0,011cd254c6d374c03154722e0749ddd7d0bcf045358af6c8d44be82d0ac00b72
thread:issue:14699,issue #14699: [Feature Request] linux distribution friendly build system,stale/low urgency,2,stale/low urgency,1,0,14699,,,,0,75791dc127c480889fe342be492a46080b8e633d57529ddc9238f1fda51c1cc6
thread:issue:147001,issue #147001: Preload CUDA fails if CUDA libs in different PYTHONPATH,stale/low urgency,2,stale/low urgency,1,0,147001,,,,0,9a3853ebd58ccde0178ee0fbfdd5f3689f4353a0c65903aa0781ca0210547607
thread:issue:14701,issue #14701: CI: Flaky download from download.pytorch.org,stale/low urgency,2,stale/low urgency,1,0,14701,,,,0,7eb3bf211ede34c62f79671891dd520c9d1246b0224af068eba6d76b71c0b36f
thread:issue:147011,issue #147011: Deprecation of NVTX 2 (`nvToolsExt`): Recommended to move to NVTX 3,stale/low urgency,2,stale/low urgency,1,0,147011,,,,0,57028ef8ad9cfa0f66a6e209d29eddccbdbbb41089619e823f48e8ff501e5a95
thread:issue:14702,issue #14702: pytorch_doc_push is racing with itself,stale/low urgency,2,stale/low urgency,1,0,14702,,,,0,22e5432ef450789c91626c42c0d73daa1a1e9a9dfba8f9bd5121d141345cd549
thread:issue:147062,issue #147062: [Feature Request] Release original parameters by layer when turning on `freezing_discard_parameters`,stale/low urgency,2,stale/low urgency,1,0,147062,,,,0,2dead3c16b91e3c8f12bb6843caa2d628d23ae2b14f3576a0263247937499f11
thread:issue:147118,issue #147118: padding fails on view from large tensor,stale/low urgency,2,stale/low urgency,1,0,147118,,,,0,c6d2b91c21a69f7182cccb526f7b1d83df517b5519e6676500fdc2f492393103
thread:issue:147134,issue #147134: `torch.mul` uses `OpMathType` for computation.,stale/low urgency,2,stale/low urgency,1,0,147134,,,,0,aeaf8b846f7d7b42e8832088cc0fd7972d2ef387f774b31b9b6eb83371106858
thread:issue:147239,issue #147239: `DeviceCopy in input program` source hint,stale/low urgency,2,stale/low urgency,1,0,147239,,,,0,ffcdd634967cfa06aaad8722078b170bab3e3ed18f5912983671062a4beb8466
thread:issue:14726,issue #14726: [caffe2] Corresponding C++ API for prepare_prediction_net,stale/low urgency,2,stale/low urgency,1,0,14726,,,,0,986a234822aa440046aa25d40b604b38424a93314e16a978b5ce033d0dfbc305
thread:issue:147263,issue #147263: How to trigger several independent communications simultaneously?,stale/low urgency,2,stale/low urgency,1,0,147263,,,,0,d2ffe6ae9f4668ec2f547717ab6bcc1df0dfde57a14bc7f855389ad2ec2998a8
thread:issue:147264,issue #147264: `Dim.DYNAMIC` inferred to be constant,stale/low urgency,2,stale/low urgency,1,0,147264,,,,0,b5ca8ab1ca94caf9ba0a25b32ffe129b815e50300f192ec92fb90c8da1c66340
thread:issue:147274,issue #147274: PaddleOCR PyTorch conflicts,stale/low urgency,2,stale/low urgency,1,0,147274,,,,0,05b7069893a5dbc847affe8c2e166ef7da778cb313e16399265d8f1c97f9da09
thread:issue:147284,issue #147284: torch.softmax gives wierd result under specific condition on cuda double tensors with multiple rows and 781 columns.,stale/low urgency,2,stale/low urgency,1,0,147284,,,,0,0b8e85da3d935c3567d0450f0b13c65ec88125927d0224a5686236fa866771c0
thread:issue:147287,issue #147287: [inductor][fuzzer] legacy `torch.native_batch_norm` should be removed,stale/low urgency,2,stale/low urgency,1,0,147287,,,,0,794624e05167b7edfddfe4d50eb95f28a43e9d1542359c63c1e4b65b227f5062
thread:issue:147288,issue #147288: Calling optimizer.load_state_dict causes Tensors lost,stale/low urgency,2,stale/low urgency,1,0,147288,,,,0,bf03ff3e29b572e79f1464ff96c03f09d969eca681d3982829efdaf296245f14
thread:issue:14731,issue #14731: Caffe2 C++ runs single threaded,stale/low urgency,2,stale/low urgency,1,0,14731,,,,0,d8acc0882ae4cbe502f0aa3e5de800bc6dba2775cfbde4a3276203c1dac1f490
thread:issue:147317,issue #147317: Tensorboard `add_video()` broken for `moviepy>=2.0`,stale/low urgency,2,stale/low urgency,1,0,147317,,,,0,db9fb0f83aede6b3d7bb7e1ff29a0fab9cb650409b51af0a102b73ecaa9efd7a
thread:issue:147348,issue #147348: [pipelining] AttributeError: 'InterpreterModule' object has no attribute,stale/low urgency,2,stale/low urgency,1,0,147348,,,,0,647d59c20345148035c50edb1cc12b5f86b6d1631b581b8371c095863aff6491
thread:issue:147387,issue #147387: Track test count regressions,stale/low urgency,2,stale/low urgency,1,0,147387,,,,0,d7fb92b4640a3e7c33e88b3ff697a5344eb51d5faec57678fec3d3382194ad09
thread:issue:147389,issue #147389: Fix representation of `Lazy*` modules after loading parameters,stale/low urgency,2,stale/low urgency,1,0,147389,,,,0,f720e8416f0b82b78ee63ee8fec743e8740319dbae1e5b9315f79ea02ebc83b4
thread:issue:147428,issue #147428: Add a config to allow print and generate all recompile reasons and not stop at first.,stale/low urgency,2,stale/low urgency,1,0,147428,,,,0,d7b0e0e7f8f82cfff1d9da8c05fd72faf16b57d370a85c6ac9767e3e8898989f
thread:issue:147449,issue #147449: gradient checkpointing with use_reentrant=False cannot reduce peak memory,stale/low urgency,2,stale/low urgency,1,0,147449,,,,0,1309ff4fc5d66795c39159a588c315497cfeb7e914f436cadfdaa7bd65962eb3
thread:issue:147475,issue #147475: [Export AOTI] dynamic_shapes export and compile degraded output,stale/low urgency,2,stale/low urgency,1,0,147475,,,,0,ee0d708e895bd614bec9c15bc9b7021b0d5082959091d57306bb504f8cf88d06
thread:issue:147495,issue #147495: Native Node Sleep/Wake Functionality for Neural Network Layers,stale/low urgency,2,stale/low urgency,1,0,147495,,,,0,bd578782adefd0ea57fcebc1b9fbe84cba681345ada89b83666d6bb89e58428f
thread:issue:147502,issue #147502: Slow first time torch.distributed op with ray,stale/low urgency,2,stale/low urgency,1,0,147502,,,,0,26d30a502e21500848e80616486175c0fc5ff1c347640b16a6655b1279b66db9
thread:issue:147506,issue #147506: Unintuitive behavior and errors in F.pad,stale/low urgency,2,stale/low urgency,1,0,147506,,,,0,0e88e91c22690f04276199486166a973c1510346ea05ee3448aff7eb26ef1cf7
thread:issue:147510,issue #147510: `clamp_` and `clamp` behave differently on MPS device.,stale/low urgency,2,stale/low urgency,1,0,147510,,,,0,5af8a013feed1aa26ea6715916053b052ae81157ab17929ffc0e0382d8694727
thread:issue:147563,issue #147563: torch.distributed.elastic.multiprocessing.start_process description does not reflect API,stale/low urgency,2,stale/low urgency,1,0,147563,,,,0,4e61ce1abb931bf9d2f4909ab886dd7ef49fcba69f6367b97ebe8a49ca0185f5
thread:issue:147570,issue #147570: `view()` + modify-in-place fails silently with DTensor,stale/low urgency,2,stale/low urgency,1,0,147570,,,,0,0914dd198c03aceecc2a6076a52259d24c2705dddfb556cef454dd9767594b70
thread:issue:147581,issue #147581: cpp libtorch transformerimpl lack some parameter between with python pytorch,stale/low urgency,2,stale/low urgency,1,0,147581,,,,0,c5df2dc4ad5dd4746e103e8aab6c5ef3ff8a0bcaff1dc09e9e6dd9a8f8b1310d
thread:issue:147585,issue #147585: [SDPA backward error]Error detected in ScaledDotProductEfficientAttentionBackward0 when input seqlen is very long and with attn_mask,stale/low urgency,2,stale/low urgency,1,0,147585,,,,0,eca4488692a4a72ae9eccb1fff5d5027b4dd9eff2f706ef563d98ee147f536ac
thread:issue:147595,issue #147595: Non-Determinism in Faster R-CNN Despite Setting All Deterministic Flags,stale/low urgency,2,stale/low urgency,1,0,147595,,,,0,540bfe611238800c01db0cc7f5fbe4fccff6f5081749cbc01d1cd0566a016476
thread:issue:147597,issue #147597: [Dynamo] WeakRefVariable doesn't use the most updated python referent when call_function is executed,stale/low urgency,2,stale/low urgency,1,0,147597,,,,0,79c92d996c2e386e1af7950b9103c3d80747619f6a44a73baa94a5596e2afb6b
thread:issue:147613,issue #147613: torch.nn.AvgPool2d fails with stride >= 2^31 on CUDA,stale/low urgency,2,stale/low urgency,1,0,147613,,,,0,727469e0a7c35e67d791fea57c540c9562423f8eb0e74e9764d3a8fc4351243a
thread:issue:147631,issue #147631: torch.distributed.init_process_group with unspecified backend out of date,stale/low urgency,2,stale/low urgency,1,0,147631,,,,0,6a4ba1ad409758e98708787b3e8fcdf86ea2195bc5e47ddfd60c6a85f54ffccc
thread:issue:147643,issue #147643: Representation string of a meta tensor is not a valid `tensor` call,stale/low urgency,2,stale/low urgency,1,0,147643,,,,0,533ff1a3a98523d3d23ed172f02543378e5c55948cd8ef0599bf4e143e4b98a1
thread:issue:147661,issue #147661: Add shape fn for einsum,stale/low urgency,2,stale/low urgency,1,0,147661,,,,0,165e555f7ca4c5b4b2e9b9b71c542ab31ab20261b4ff444fefea67ee702e06de
thread:issue:147677,issue #147677: distributed_c10d.broadcast causing unexpected CUDA oom error,stale/low urgency,2,stale/low urgency,1,0,147677,,,,0,fb62005ef36d0f8d07752d5c7e4f17cb4616a96bf76d64f430576a40e73fc006
thread:issue:147682,issue #147682: Error in parsing arch_string in _extract_arch_version,stale/low urgency,2,stale/low urgency,1,0,147682,,,,0,606004866075cd3c2da5bcf28bbf4b70a3c1649e2432c9ef50a8bef71f4b1bd6
thread:issue:147720,issue #147720: [ONNX Convert] Error when input to nn.AdaptiveAvgPool2d size is variable,stale/low urgency,2,stale/low urgency,1,0,147720,,,,0,6d745c3c37af568a967a545cac2ab5ce98fdedf94cbd4586ff89daeaa753c08b
thread:issue:147732,issue #147732: [Distribute] len(input_specs) == len(input_args_strategy) AssertionError,stale/low urgency,2,stale/low urgency,1,0,147732,,,,0,0160f5197eca9490ff8c59d26dcb995e49d27875f44d124630888fa55a2056ef
thread:issue:147793,issue #147793: context_parallel fails with plain sdpa kernel SDPBackend.MATH,stale/low urgency,2,stale/low urgency,1,0,147793,,,,0,cee73d31aa1b53a1b78d1170b88dcbb03acf97750b4dcda68806b2bf0684ca44
thread:issue:147814,issue #147814: unbind_copy opinformation cause exception while running test_dtensor_ops.py,stale/low urgency,2,stale/low urgency,1,0,147814,,,,0,3afa5b5de0632afad61ef87e0db07a42b03a2ec2ac5db20772a541b1ce30d84f
thread:issue:147859,issue #147859: Exporting a PyTorch Model with Dynamic ModuleList Indexing to ONNX,stale/low urgency,2,stale/low urgency,1,0,147859,,,,0,346a358f48e23514e0e07da99df471163fe56af6244152184e9ebb1ab45027d8
thread:issue:147867,issue #147867: Track follow ups to #147354,stale/low urgency,2,stale/low urgency,1,0,147867,,,,0,e21b927c93e3a98225468812476a0008e7ec6d051bb69b8c6362dae180afde67
thread:issue:147874,issue #147874: Does CUDACachingAllocator.cpp still require deferred event creation?,stale/low urgency,2,stale/low urgency,1,0,147874,,,,0,42fe58040946e9364389f1d742fd17f3bed61f783c1666ed9855c06f0cf55b36
thread:issue:14788,issue #14788: [caffe2] [NNPACK] how to set thread pool for NNPACK?,stale/low urgency,2,stale/low urgency,1,0,14788,,,,0,4fcbf9d7ab131b56cbab094907017fbce630f0f97e25ab27532837d53f999b1a
thread:issue:147884,issue #147884: `FxGraphDrawer` fails on `einsum` nodes,stale/low urgency,2,stale/low urgency,1,0,147884,,,,0,6324f425e9da41b30ebc51927ec3ca0553dd4daa09eebac5bb9828e5eca10493
thread:issue:14790,issue #14790: Tensor.copy_() seems to work improperly with numpy/list indices,stale/low urgency,2,stale/low urgency,1,0,14790,,,,0,5a11f577fd1428c1a1208ea591bf0221bb41b9dbf4238c316e9b44748bd7c365
thread:issue:14796,issue #14796: Cannot build Caffe2 with TensorRT,stale/low urgency,2,stale/low urgency,1,0,14796,,,,0,6e0ee5285242362a8622d66f27f9aede17ab982f165fe8af426fe7638ff84bed
thread:issue:14797,issue #14797: Modern interface for Storage,stale/low urgency,2,stale/low urgency,1,0,14797,,,,0,f50e6bbb3235b5329420655a622ea872006e739a48b58b9801d59dde6efed73b
thread:issue:147980,issue #147980: Torch distribute module when local dtensor is already on the correct device,stale/low urgency,2,stale/low urgency,1,0,147980,,,,0,13ae76ab9d7176dcffe245001167ad30dd3db797d849f28ff565c1c4ea5f1d7b
thread:issue:14799,issue #14799: which operator of caffe2 have the same function as torch.nn.Parameter,stale/low urgency,2,stale/low urgency,1,0,14799,,,,0,a4fac9578450ead8c361cf99b91ab51e2e7b318953116c689e027f3f29917111
thread:issue:147999,issue #147999: Adjust test_mm_triton_kernel_benchmark for unpadded tensors,stale/low urgency,2,stale/low urgency,1,0,147999,,,,0,76ae52782a8573fe049050aaec9020ed5b0ab0725f750b7b67563d4c02c8060c
thread:issue:148029,issue #148029: abs and arange Meta impls are non-standard and thus not working with other backends,stale/low urgency,2,stale/low urgency,1,0,148029,,,,0,7bf40fff92792985eee27fb2b96654bac13ee4b73322b53bd6b65155f1227303
thread:issue:148044,issue #148044: Sweep through Potentially BC Breaking Commits in Triton,stale/low urgency,2,stale/low urgency,1,0,148044,,,,0,acfa0df4ae11b279243edde9f421a27f9028b3a9e80725c984c0477b24e3b901
thread:issue:148045,issue #148045: Run Performance Regression Tests on new Version of Triton,stale/low urgency,2,stale/low urgency,1,0,148045,,,,0,c65bdc1cd90ba52f547966703405f5a0c76f3cbf219559ff7aed84f6710942ae
thread:issue:148052,issue #148052: [inductor][user triton] support on-device TMA / tensor descriptor API,stale/low urgency,2,stale/low urgency,1,0,148052,,,,0,593576603d3bd0daaf04b768b43abd1d2eac15d5081db900e8ccc5c6271bb986
thread:issue:148065,issue #148065: Support alpha=inf consistently for torch.celu,stale/low urgency,2,stale/low urgency,1,0,148065,,,,0,8a6bd2f997f704a5a8c8a43c0d12c9bba51946fe10303236483c5877cddf4852
thread:issue:148073,issue #148073: torch.compile with the inductor backend slows down (exponentially?) for certain graphs,stale/low urgency,2,stale/low urgency,1,0,148073,,,,0,762dd3793504558d17991505ced23d673d447f12be909344212fec9b5651db8b
thread:issue:148078,issue #148078: Video-Llama (version 1) runs much slower using Float16 than Float32 on Kunpeng CPU,stale/low urgency,2,stale/low urgency,1,0,148078,,,,0,abdafa582bca6037107d8861a2ccaa7e0bb3c9fe52f716ee8ac4b1d46d484683
thread:issue:148086,issue #148086: FSDP2 without sharding works slower than DDP,stale/low urgency,2,stale/low urgency,1,0,148086,,,,0,2205eef2265f5a3c6b7669f33f52f5077bd8419d86ab534b54861f06a048f31c
thread:issue:148098,issue #148098: onnx dynamo export does not support aten bucketize,stale/low urgency,2,stale/low urgency,1,0,148098,,,,0,e137c57792e7a94677ffc94d23c00ec2b11a8b8b9f2441e7716d0c75a7082501
thread:issue:148103,issue #148103: Initial investigation for removing MOD_SKIPLIST,stale/low urgency,2,stale/low urgency,1,0,148103,,,,0,e4f82a32b5d161aaaf95df24cbdbc79ea03a44bb8ffcd521bfe5b4b373139348
thread:issue:148106,issue #148106: NotImplementedError: FlexAttentionHigherOrderVariable() has no type,stale/low urgency,2,stale/low urgency,1,0,148106,,,,0,e20a5bcba5dd823404705448548b21806bfdea650df502fa8cd300995afb7e81
thread:issue:148113,"issue #148113: [inductor][triton] introduce better ""APIs"" in triton that can clean up our triton/inductor integration",stale/low urgency,2,stale/low urgency,1,0,148113,,,,0,4ae680e0164e4fe04aa8153216e63a60a32b4bfd29d8af64969880f0e9026d81
thread:issue:148116,issue #148116: [inductor][triton] Explicit kernel-arg mismatch checks,stale/low urgency,2,stale/low urgency,1,0,148116,,,,0,383deed7700c98a2568d87d80f29b875992b445914323c9242b1a181591016dd
thread:issue:148144,issue #148144: Torch export does not preserve original edges between nodes,stale/low urgency,2,stale/low urgency,1,0,148144,,,,0,ab18e9955b945021e2662a95a89d953a1945e1fb0dc833d356f27d617619da89
thread:issue:148146,issue #148146: Unify OpOverload._get_dispatch and HigherOrderOperator.dispatch,stale/low urgency,2,stale/low urgency,1,0,148146,,,,0,ae08ed79100e1a6ff1284564175471e996218644ec7b3296ba9c82b5babe059d
thread:issue:148150,issue #148150: [pytorch elastic] [RHEL] multiple processes call to dist.destroy_process_group() cause an RuntimeError: CUDA error: CUDA-capable device(s) is/are busy or unavailable,stale/low urgency,2,stale/low urgency,1,0,148150,,,,0,3e8ac76c5d36a16b2bd39acadc17e4c58c98374555809e58ec5040fa96a007b7
thread:issue:148162,issue #148162: Specifying device_id in init_process_group causes tensor parallel + pipeline parallel to fail,stale/low urgency,2,stale/low urgency,1,0,148162,,,,0,3054c334611fc79f4ad6539a9e65bd2ff1cf6197a365d6ce8710de72f94fa15b
thread:issue:148193,issue #148193: [Request Help] “torch._dynamo.exc.UserError: Dynamic control flow is not supported at the moment.” “torch._dynamo.exc.UncapturedHigherOrderOpError: Cond doesn't work unless it is captured completely with torch.compile.”,stale/low urgency,2,stale/low urgency,1,0,148193,,,,0,5ba93bc43f14c6d73c4de566a3ee78372df60462632073a8d6bdaa86316351b4
thread:issue:148196,"issue #148196: [inductor][triton] Decide how to deprecate ""old triton versions""",stale/low urgency,2,stale/low urgency,1,0,148196,,,,0,67a34d17c5882dfd05b0d13e420c208608a364cd8d14d6e468179794b5535b0f
thread:issue:148219,issue #148219: MPS vs Metal vs CPU performance comparison,stale/low urgency,2,stale/low urgency,1,0,148219,,,,0,ad9691355156564a1dc156a1e7d6ed75923639ad5e3b0e5b0e514f3e0b1752f5
thread:issue:148232,issue #148232: Add Structured Knowledge Accumulation (SKA) Layer to PyTorch,stale/low urgency,2,stale/low urgency,1,0,148232,,,,0,76cd3edde0d1e26fc00eec833fa7019686ea3420269cf2824b517d87fe646346
thread:issue:148247,issue #148247: `torch.multinomial` outputs inconsistency on ARM and x86,stale/low urgency,2,stale/low urgency,1,0,148247,,,,0,e7a3db4169ff480c5f5d52b2ccb548aae168dd59862dc4603413d4728724919e
thread:issue:148257,issue #148257: [FSDP2] HSDP with globally sharded fp32 weights and optimizer states,stale/low urgency,2,stale/low urgency,1,0,148257,,,,0,35ef645e44a9a8a62e1ef5551438dfb0e62400d9ea6d8fe17f8b8ee29a7a8f82
thread:issue:148269,issue #148269: Should DTensor support `Shard()` placement without dim requirement?,stale/low urgency,2,stale/low urgency,1,0,148269,,,,0,f578d9d2a534284060e6431b760d02335defa1e7a2d99d074db6fe293fdf90c4
thread:issue:148289,issue #148289: Triton Kernel Rejects NamedTupleVariable Arguments,stale/low urgency,2,stale/low urgency,1,0,148289,,,,0,6e1dc6d53250b01df03335aa9cdf8763a918dcb0865805b0dba7d7f65019c640
thread:issue:148302,issue #148302: broadcast_object_list not release GPU,stale/low urgency,2,stale/low urgency,1,0,148302,,,,0,8fcdd8bf0b662f2ee73e720327ff660a0a0fe947a2984935780244287605d4b6
thread:issue:148306,issue #148306: torch.vmap incompatibility with DLPack functions,stale/low urgency,2,stale/low urgency,1,0,148306,,,,0,8bbe8fa5072bae94ca25cbde965f0e3dda20554bedfd47f90ac7eef07984cd2f
thread:issue:148379,issue #148379: Improve nested jagged tensor select performance on batch dim,stale/low urgency,2,stale/low urgency,1,0,148379,,,,0,8667c9638424f8c57e94533eeb7000d2d1b94ea4ddd04105b3246ce1d1ddf3c2
thread:issue:14839,issue #14839: Test OpenCV4 in CI,stale/low urgency,2,stale/low urgency,1,0,14839,,,,0,7e9f6b668591ec209f3caf846589169271cbe4309d0b9cebd477aa1b6d10f924
thread:issue:148440,issue #148440: torch.distributed hangs between 2 Mac Devices,stale/low urgency,2,stale/low urgency,1,0,148440,,,,0,ad8b4fe91af19ec2b4cba7eb17fa43f1cce887433a1f175d8de7e0d051d1c660
thread:issue:148451,issue #148451: BC-linter should ignore testing/linter/adapters/,stale/low urgency,2,stale/low urgency,1,0,148451,,,,0,ede8a8dae2df066dabc8189fe2d7017a1f76a7d6a4742891f4b2b3b51a050df4
thread:issue:148504,issue #148504: [FSDP2] improve error msg for duplicate wraps,stale/low urgency,2,stale/low urgency,1,0,148504,,,,0,8db148a756cd5f828e493ed813f2a232638bdcc785a6bab19373010662f4d392
thread:issue:148518,issue #148518: Preview (Nightly) version cuda12.8 cannot find torchaudio file,stale/low urgency,2,stale/low urgency,1,0,148518,,,,0,85c951324324e12545d61ad6a342b6a0a5c6729ed9d42c73afa21186002a573b
thread:issue:148536,issue #148536: Computing only the n first rows of a distance matrix with pdist,stale/low urgency,2,stale/low urgency,1,0,148536,,,,0,b1f1106f0365a5ac0de1ab446fe5d8b470fe0669cc9bb0a50dd105f6b7f4ef39
thread:issue:148546,issue #148546: Pytorch nightly broken Flash Attention 3 compile with sycl commit - TypeError: _write_ninja_file() got an unexpected keyword argument 'sycl_cflags',stale/low urgency,2,stale/low urgency,1,0,148546,,,,0,d99b811e13f682298258a3268bf390bb216a62db3b07eb3f7ed9c1b650a318ae
thread:issue:148548,issue #148548: Issue with Sparse Tensor Matrix Multiplication and Broadcasting,stale/low urgency,2,stale/low urgency,1,0,148548,,,,0,59044cd23d9a4c87e375846e774cd92ea0701a982920a8811cc337216043b0b0
thread:issue:148552,issue #148552: Return type annotation of `Tensor.long()` etc is not narrowed down to dtype-specific names `LongTensor` etc,stale/low urgency,2,stale/low urgency,1,0,148552,,,,0,a19fa0bfb162ef759c6b81fa1ae7ec13a39423fcd6f3b368fb19a7c5e62e8d3d
thread:issue:148564,issue #148564: [Feature Request] Dynamic shapes API requires spec for all arguments.,stale/low urgency,2,stale/low urgency,1,0,148564,,,,0,e4d7fda3d2748bbca4c15b7087dc2b976f2a7ed9eb6ceaccad939b45245ed4be
thread:issue:148568,issue #148568: Suggested fixes sometimes not enough in export,stale/low urgency,2,stale/low urgency,1,0,148568,,,,0,196b8611e1b7938aeb7e4b4a1aaeb378e8ff4d5ccb8a213d48ad6771f505d130
thread:issue:14864,issue #14864: [discussion] Recommend a different file extension for models (.PTH is a special extension for Python),stale/low urgency,2,stale/low urgency,1,0,14864,,,,0,e74a9715940c61c11f7dea3f6f81bd0fb8b21b829120dbe06de8da0653e2057d
thread:issue:148670,issue #148670: MPS Backend Error: ComplexDouble (complex128) Conversion Fails When Diffusers Transformer Creates 64‐bit Complex Tensors,stale/low urgency,2,stale/low urgency,1,0,148670,,,,0,3e2780626a649c74398129b77cc828aafd3ac6fd0615c14eb44f1a9a8f42e521
thread:issue:148681,issue #148681: foreach CUDA tests flaky on CUDA 12.6+ due to flaky profiler results,stale/low urgency,2,stale/low urgency,1,0,148681,,,,0,c23e25708cb08f1103973725899b0964f8fce301c7f43f13934595810ca64da6
thread:issue:148689,issue #148689: NUMA Binding Integration with torchrun,stale/low urgency,2,stale/low urgency,1,0,148689,,,,0,c655f2d6fc4656f30415c5b91030250a8a205985d1cd4a570a80b9d184a62634
thread:issue:1487,issue #1487: [feature request/proposal] Relax scale_factor for nearest neighbor upsampling,stale/low urgency,2,stale/low urgency,1,0,1487,,,,0,e9c6af6f9c089efce93c38174ba17051d194a17e3171be92b93fa4161adc1e46
thread:issue:148756,issue #148756: FSPD ValueError: expected to be in states [<TrainingState.FORWARD_BACKWARD: 2>] but current state is TrainingState.IDLE,stale/low urgency,2,stale/low urgency,1,0,148756,,,,0,2d4eb738ca6fcdfcab13ab01b31886f6c17acb4829f35eeefbbff1dbea04ee73
thread:issue:148762,issue #148762: magma builds should be part of the docker image builds,stale/low urgency,2,stale/low urgency,1,0,148762,,,,0,60fa4d530d8512aa8b400eaf64f38b21b825d8416aabd8e0a5b87bc6a80e667d
thread:issue:148775,issue #148775: Calling `torch.linalg.lstsq` with a wrongly-shaped `out=output` argument does not properly reshape `output`,stale/low urgency,2,stale/low urgency,1,0,148775,,,,0,63117410322a9d68c08f37fbfca19ac9d5922cf6bf0c1ba3f4f64de58a8da7c4
thread:issue:148807,issue #148807: Native Infinite Sampler for Datasets,stale/low urgency,2,stale/low urgency,1,0,148807,,,,0,b4034abadcf0f3d2c51fe78f8c4fc5ecaf092eb877389883066029da748f7758
thread:issue:148828,issue #148828: MiniCPM-o compilation issue,stale/low urgency,2,stale/low urgency,1,0,148828,,,,0,f6fc03712d31d74a0b8af6c3e434db6f32a838e7cc3e2f34242f13f247129f45
thread:issue:148831,issue #148831: FSDP2 and autocast compatibility issue,stale/low urgency,2,stale/low urgency,1,0,148831,,,,0,6a663b308f767d04ecc750c334f4c23ca273c5b0e253d47fec475873bc72febd
thread:issue:148837,issue #148837: NVLS support in Pytorch,stale/low urgency,2,stale/low urgency,1,0,148837,,,,0,b83830140b6c8992eacbd2e3d41ffedbbaf30f416c889fd5818d4a063820b2f3
thread:issue:148858,issue #148858: RuntimeError: OffsetBasedRNGTracker instantiation requires the presence of CUDA/CUDA-like device,stale/low urgency,2,stale/low urgency,1,0,148858,,,,0,da5fc60a958ccf61a2f6941c0af63a892c44a310d969734105cdaf2e10c53a78
thread:issue:148885,issue #148885: test_memory_profiler_viz failed on cudamallocasync,stale/low urgency,2,stale/low urgency,1,0,148885,,,,0,075af8be81c818635eaa45b9887ad0c21f7f1e453a69e106c37909ec357f20bd
thread:issue:1489,issue #1489: Optimizer should track parameter names and not id,stale/low urgency,2,stale/low urgency,1,0,1489,,,,0,b87a761f11019deb59bdb35bc862ee48300eada0f400a9a57f2f59f01318e48b
thread:issue:148920,issue #148920: Rewrite cpp extension tests to not be crazy,stale/low urgency,2,stale/low urgency,1,0,148920,,,,0,5fdbce1da58dd77631fde895b3248ab74abce5ff23f6e0912e076de9dd1d1094
thread:issue:148927,issue #148927: default cudagraphable policy for custom op,stale/low urgency,2,stale/low urgency,1,0,148927,,,,0,8afa6059aee450b68ca65965ef2e4170af7358e0f6621d0631a2b62c0d892b2d
thread:issue:148954,issue #148954: Aborted (core dumped) double free or corruption (out),stale/low urgency,2,stale/low urgency,1,0,148954,,,,0,390473a3ac56668dc88144b274eb7769e5f1f5e75c749bf40c108bd108c48ebe
thread:issue:148960,issue #148960: Does torch compile affect results ?,stale/low urgency,2,stale/low urgency,1,0,148960,,,,0,bdd893797e4f123375d224a22b890c85c22b58f91bb97e7eb11bc2fec8bf03ba
thread:issue:148976,issue #148976: `dist.barrier()` fails with TORCH_DISTRIBUTED_DEBUG=DETAIL and after dist.send/dist.recv calls,stale/low urgency,2,stale/low urgency,1,0,148976,,,,0,002c300bbc9fa48b2667e190f87ff7e2f2df7034e19ede51df82fcf9f023c00a
thread:issue:148983,issue #148983: inference_mode Tensors do not always need to be guarded on,stale/low urgency,2,stale/low urgency,1,0,148983,,,,0,435d7270e8d734060eb0ad8c8726c9687f9ab5f4c1c4eedb67757317303221a4
thread:issue:148984,issue #148984: [Sync file] the new file is not sync properly between pytorch/pytorch and pytorch/benchmark,stale/low urgency,2,stale/low urgency,1,0,148984,,,,0,9201e3a689f20af159faf14f92672b34e80b55897caae86ecdb9e3d0bae85788
thread:issue:149013,issue #149013: Unexpected out-of-boundary behavior in `grid_sample`,stale/low urgency,2,stale/low urgency,1,0,149013,,,,0,0526d51583a489cdb13a3b05f5ab2a0a1e09bc8061a33082567c1ed6e1db728d
thread:issue:149016,issue #149016: DISABLED test_var_mean_tile_reduction_True_dynamic_shapes_cuda (__main__.DynamicShapesGPUTests),stale/low urgency,2,stale/low urgency,1,0,149016,,,,0,f461de161c020099fe480d1b5400fb60756d7fe46d391bf02a061a2c7b07c170
thread:issue:149040,"issue #149040: Only 2D, 3D, 4D, 5D padding with non-constant padding are supported for now",stale/low urgency,2,stale/low urgency,1,0,149040,,,,0,196866825dbb9c276bdd727f033aa6b28764c2e2c30bc9fc3799a29082b9851e
thread:issue:149049,"issue #149049: `pytroch.distribute` should support ""meta"" device tensors",stale/low urgency,2,stale/low urgency,1,0,149049,,,,0,42be019e959c9964fabf910bdbe9519b1f5098c956c4694cf830189e3b30aae6
thread:issue:149062,issue #149062: Reserve customized modules in torch.compile's dynamo tracer,stale/low urgency,2,stale/low urgency,1,0,149062,,,,0,0401880f60d5500f76c367e1238c84900c6f17789f1f4c53478ce1b60e4edff1
thread:issue:14907,issue #14907: Missing dilation from several pooling modules (AvgPool),stale/low urgency,2,stale/low urgency,1,0,14907,,,,0,430627009fe4df9f1c8e454458ebf6898991700d8bb23ddb1dea653d0b31db2c
thread:issue:149090,issue #149090: Better warning once for `cuDNN/MIOpen` not enabled,stale/low urgency,2,stale/low urgency,1,0,149090,,,,0,610b63287e8326f58868fbe4db103273024c828fe43de3dbd2e92a8fdf708f6f
thread:issue:149094,issue #149094: How to skip backward specific steps in torch.compile,stale/low urgency,2,stale/low urgency,1,0,149094,,,,0,d36d444a8aebd7cad31464b48ac6774f89f32fe6a8f47aadfe6b289a46a6b732
thread:issue:149096,issue #149096: How to determine which part of torch.compile undergoes recompiling after caching,stale/low urgency,2,stale/low urgency,1,0,149096,,,,0,666199d432442badf0418df6aaf1002f476509301d666b31dc988c25ae80260a
thread:issue:149100,issue #149100: Memory leak when using get_model_state_dict with FSDP-sharded models,stale/low urgency,2,stale/low urgency,1,0,149100,,,,0,adf567069d9e2592f83c01e8d26067af959de09af4fb204f7bf4d0df8c69ee55
thread:issue:149110,issue #149110: [XPU] Enable Windows CI/CD test for XPU,stale/low urgency,2,stale/low urgency,1,0,149110,,,,0,1b8da49921578e9b7af09ae157622ad03f5814b1acfddcc316aa3c10df023830
thread:issue:149120,issue #149120: Add `keepdim` parameter for `torch.nn.functional.cosine_similarity`,stale/low urgency,2,stale/low urgency,1,0,149120,,,,0,7773a3c852b55cf4c3b3e8ee842633cdda6f23c8a7a7656ecade0787477e108a
thread:issue:149155,issue #149155: torch.multiprocessing.Queue Zeroes Out Tensors on Retrieval,stale/low urgency,2,stale/low urgency,1,0,149155,,,,0,85ee80732168e698492d68b8ead955cdf0e9e3e8c25e6191601ce47ee25df479
thread:issue:149169,issue #149169: [RelEng] wheel testing for new arch versions,stale/low urgency,2,stale/low urgency,1,0,149169,,,,0,2dc276cba57a9f3741f4e11d2b7638de3dd120baf0273d860da65e1bdaa8ef82
thread:issue:149177,issue #149177: [Dist] Async op isend and irecv bug,stale/low urgency,2,stale/low urgency,1,0,149177,,,,0,7e1fe0cef3a11638f8664033980e711863c39ca7b2a837e467b89dba4b61b091
thread:issue:149184,"issue #149184: [macOS] instantiating optimizer after torch.set_default_device(""mps"") throws ""RuntimeError: Placeholder storage has not been allocated on MPS device!""",stale/low urgency,2,stale/low urgency,1,0,149184,,,,0,64f8241e205119e3f56f8a47f1971f139b0d922a075d3f2473bb46c5294d592e
thread:issue:149187,issue #149187: Issue with Shared CUDA Tensor Reference Counting in Multi-Processing,stale/low urgency,2,stale/low urgency,1,0,149187,,,,0,a59b85594ea8c80af215920363c1dc12663abfaaa54d83a601fe5aa4416a4eed
thread:issue:149194,issue #149194: LSTM slow on PackedSequence,stale/low urgency,2,stale/low urgency,1,0,149194,,,,0,3e5db92c820774c3a24f3c317f7ea349a03701aa1f54c452d520e285e94a0949
thread:issue:149196,issue #149196: (Will PR) Multiprocessing with CUDA_VISIBLE_DEVICES seems to give the wrong device,stale/low urgency,2,stale/low urgency,1,0,149196,,,,0,9509f798e9afa0b45f9f27231f1dff675ecc846af4da571fe4f6c761ac57a70b
thread:issue:149209,issue #149209: Wrong argument in `CompileCounterWithBackend` when running `torch.utils.benchmark.utils.compile.bench_all`,stale/low urgency,2,stale/low urgency,1,0,149209,,,,0,85a372731ea3810c8e1aa2dd009bbfbade439f1951eb8de2c17959e950980282
thread:issue:149224,issue #149224: Unexpected results in using torch.save API,stale/low urgency,2,stale/low urgency,1,0,149224,,,,0,042c36bfdc4cff981151402fab8f00c0f9a5408f5752b17ddad5f389f7e9ec69
thread:issue:149227,issue #149227: Unexpected behaviour in using torch.nn.utils.rnn.pack_padded_sequence API,stale/low urgency,2,stale/low urgency,1,0,149227,,,,0,78bf1c8de136879d2220d5061751d8d08b204827faf642892860661485994ea3
thread:issue:14925,issue #14925: Am I able to run caffe2 and gloo example: resnet50_trainer.py using pytorch v1.0rc1?,stale/low urgency,2,stale/low urgency,1,0,14925,,,,0,a3f6fdd24bf47a120679bdae8519f22fc1261577335bae0995ed84da032b6a0e
thread:issue:149250,issue #149250: Implement batching for torch.isin operator,stale/low urgency,2,stale/low urgency,1,0,149250,,,,0,0b8c67ab0c2ff12c38fe6038e3cec843e316eb12929ca1c0d174c186445d892d
thread:issue:149269,issue #149269: Add Optimizer: Sharpness Aware Minimization,stale/low urgency,2,stale/low urgency,1,0,149269,,,,0,d7f694ba5cc6715792b3d187984a0d652ae8aa3c8092cbd2f20a5f53d33c80b3
thread:issue:149284,issue #149284: No examples in documentation for masked_fill and masked_fill_,stale/low urgency,2,stale/low urgency,1,0,149284,,,,0,8207fc100e26a64354467d96833939dbd477b9e30cc33b74ca91d3633ab5a7a9
thread:issue:149301,issue #149301: Unexpected results w/ LayerNorm -- suspecting possible memory issue?,stale/low urgency,2,stale/low urgency,1,0,149301,,,,0,1b66c3e8fffbf5738d1e15b8aee0fc72ceb40b7a6c83e9db253f704f7c4b3e17
thread:issue:149304,issue #149304: Machine with isolated cores is not respected by torch,stale/low urgency,2,stale/low urgency,1,0,149304,,,,0,98cc0185d7a721fb1e90c2be249e6f3c7e105da4356f750f8b46e1c006584a0c
thread:issue:149313,"issue #149313: Training error: due to torch.load(weights, map_location='cpu') - Weights only load failed.",stale/low urgency,2,stale/low urgency,1,0,149313,,,,0,9e81f40f5572f40d58185afc2c0acbefddb2dfa637c1719020df58f8fe5fd065
thread:issue:149315,issue #149315: How to Retain Computational Graph in torch.func.jvp() for Parameter Gradients?,stale/low urgency,2,stale/low urgency,1,0,149315,,,,0,9d372a0e26f4936e5186d27ca04903aea6ad729368f9f6fb608afac85e7b131c
thread:issue:149318,issue #149318: BypassFxGraphCache should have better error message for HOPs (also we should support more),stale/low urgency,2,stale/low urgency,1,0,149318,,,,0,fb4aeaacf22dfdbeeb58cb020a624f90c304ad0f43902664e53a59924edb3a3a
thread:issue:149319,issue #149319: Need some sort of pattern matcher recipes somewhere,stale/low urgency,2,stale/low urgency,1,0,149319,,,,0,3287604356a5ab110977cecace354d7dfb5e1ef8fd3fb03c8d5bf0e96f9f3846
thread:issue:149324,issue #149324: Unguarded Usage of Facebook Internal Code?,stale/low urgency,2,stale/low urgency,1,0,149324,,,,0,b5b1c6bc9a949525219c654e99229ba8e15e84ee5048ba7a34e47d89d08f7151
thread:issue:14939,issue #14939: caffe2/predictor missing from libtorch,stale/low urgency,2,stale/low urgency,1,0,14939,,,,0,ce6f83dd0321a943e598437ef288ff1213440ce2f3d994e4b28ed60bd194004d
thread:issue:149393,issue #149393: Torch RPC examples from docs say usage is deprecated.,stale/low urgency,2,stale/low urgency,1,0,149393,,,,0,aa72d96fa38a2cdd1236ad2658dda3428b0088e1790eb149246b73803057d23d
thread:issue:14940,issue #14940: caffe2::Predictor: Cannot find operator schema for GivenTensorFill. Will skip schema checking.,stale/low urgency,2,stale/low urgency,1,0,14940,,,,0,807b617cc051247750c20e52edff0fd68cc8766388891ca2312f597fd3912e2c
thread:issue:149440,issue #149440: [pt2] Support statically launching triton compiled cuda kernels,stale/low urgency,2,stale/low urgency,1,0,149440,,,,0,c42d009abaf8c509086ab837fe6ebfb88a11b02e4bd74e1d0b99fe6b7696224c
thread:issue:149452,"issue #149452: Support save_cubin (and therefore, support cpp_wrapper use cases)",stale/low urgency,2,stale/low urgency,1,0,149452,,,,0,fd57930ce013bae1cf168b7c483a566da7f3204a6dc79cecf662c13e621b1ea0
thread:issue:149453,issue #149453: Support launch_enter and launch_exit hooks,stale/low urgency,2,stale/low urgency,1,0,149453,,,,0,b6fbb05c2af9d73961d5dec50c2eb1ca8441205a8aecba96ae97fe97f09d4aad
thread:issue:149454,issue #149454: Support user defined triton kernels,stale/low urgency,2,stale/low urgency,1,0,149454,,,,0,1e3d87cfe73fe2c96220f7d1fce1fc4bb8e084a03a20b77168b35e59489676fe
thread:issue:149455,issue #149455: Support num_ctas > 1?,stale/low urgency,2,stale/low urgency,1,0,149455,,,,0,6e13461ada3494a37fc08ccaf0b48724b0256a26af1745a35f48b88c28838b16
thread:issue:149516,issue #149516: ```StateDictOptions``` in combination with ```cpu_offload=True``` and ```strict=False``` not working,stale/low urgency,2,stale/low urgency,1,0,149516,,,,0,9b42ff9b510afd758b512f9dfe09b166a7ee8d25561f08e9e5f8774e8a1b29d9
thread:issue:149545,"issue #149545: `torch.load(..., map_location=""meta"")` hangs indefinitely",stale/low urgency,2,stale/low urgency,1,0,149545,,,,0,5154e4ea1679c4569fb7fcb43b9821db5f1b7c83b35ff4819de69ee4e80b6674
thread:issue:14959,issue #14959: get/set device in c++,stale/low urgency,2,stale/low urgency,1,0,14959,,,,0,768f1a3650b49104465d3ff90e491b87465cf1c28b82f5bcf422ade3f54ff1c6
thread:issue:149603,"issue #149603: [inductor] API for user-controlled fusion ""escape-hatch"" mechanism in inductor",stale/low urgency,2,stale/low urgency,1,0,149603,,,,0,5879010a3243cac1599efd2f468af04b0a8438522bd8cd188404c7059ad23ec4
thread:issue:149609,issue #149609: Batching Implementation for aten_nested_tensor_from_mask_left_aligned,stale/low urgency,2,stale/low urgency,1,0,149609,,,,0,546bc8d9b8e13f2e9930416403af92c29f7677b81df93b316fc51093a2af7e87
thread:issue:149620,issue #149620: `vmap` not working on `torch.range`,stale/low urgency,2,stale/low urgency,1,0,149620,,,,0,c8d81e1cf3521d551d2367ea0556058688b47033bfc6cb3b24b37ce4054c3bb1
thread:issue:14963,issue #14963: PyTorch 1.0 source build fails (error in caffe2/utils/fatal_signal_asan_no_sig_test.cc),stale/low urgency,2,stale/low urgency,1,0,14963,,,,0,42ec1876bdfb1e6ad933fe583d19880660447405727b3fc5cf1bc80e3e36d8ab
thread:issue:149660,issue #149660: cd: Migrate binary builds off of Jinja,stale/low urgency,2,stale/low urgency,1,0,149660,,,,0,393d6843c2f30a368c5959e7b0bea6c602b78300dcefb9cd0d1243786712a346
thread:issue:149694,issue #149694: torch.det results in nans for torch.func.hessian,stale/low urgency,2,stale/low urgency,1,0,149694,,,,0,07acf1845a7818c901ffe533733a22e832a0ab28f5d43620baaf690fff277589
thread:issue:14970,"issue #14970: [Caffe2] ""Same"" padding",stale/low urgency,2,stale/low urgency,1,0,14970,,,,0,bc41723851f7d30474fe0d6516f8da3185d3c9ed817061ff56f455d9db184d0a
thread:issue:149718,issue #149718: Constraints for distributions with mixed support,stale/low urgency,2,stale/low urgency,1,0,149718,,,,0,d545259e7274eb8dfcedc8758a12c5f2a7b0eceb2f9a76041bdaaa642c557b3e
thread:issue:149720,issue #149720: Implement `permute` for masked tensor,stale/low urgency,2,stale/low urgency,1,0,149720,,,,0,cbf6d96fcf0ecd225ea34279356b723eee25afc19f7f4ff96a8a4cfa56d1ed85
thread:issue:149724,"issue #149724: `torch.linalg.ldl_factor_ex`, `torch.linalg.ldl_factor`, and `torch.linalg.lstsq` Raise INTERNAL ASSERT FAILED",stale/low urgency,2,stale/low urgency,1,0,149724,,,,0,4d632107afa6ad6e99f1f75ce6953c2b79bc3dd1f14a5fdb2e6f54e4985122d0
thread:issue:149743,issue #149743: dynamic annotations that still allow duck sizing,stale/low urgency,2,stale/low urgency,1,0,149743,,,,0,d170e1a2ef1f6b9efb879408e4a4bd7312ec01f4ba60d6615f234d748590dd6e
thread:issue:149762,issue #149762: terminate called after throwing an instance of 'c10::Error',stale/low urgency,2,stale/low urgency,1,0,149762,,,,0,963543523ea41b1faacdab9f12135f683f146bb6e3f91db0c5e039bc8c86ce12
thread:issue:149788,issue #149788: `torch.vmap` does not work with tensor subclasses as expected,stale/low urgency,2,stale/low urgency,1,0,149788,,,,0,74641df4eedb45a4e7788565fa80249ad0bdeecffa871d9739bebc1164660a67
thread:issue:149796,issue #149796: Cannot compile SGlang with Torch 2.7 or Torch 2.8 and CUDA 12.8 (sm_120).,stale/low urgency,2,stale/low urgency,1,0,149796,,,,0,3f98dd9af2f52e723fef4348c7fce3f9a982c7049515e8609ec82d4b66a3fc4b
thread:issue:149798,issue #149798: Build Extension Failed with setuptools==77.0.3,stale/low urgency,2,stale/low urgency,1,0,149798,,,,0,da1446c63b6e740ea9863b7c71f61b7b6dfc2ec845ad7184a03684002785f58c
thread:issue:149802,issue #149802: (With PR) `torch.cuda.MemPool()` internal assertion failure when changing devices,stale/low urgency,2,stale/low urgency,1,0,149802,,,,0,1842fe84508f383952b1525dbd761092f2bb3e0064c500d82f76c217bf70fd6a
thread:issue:149803,issue #149803: checking out NCCL when it is not used,stale/low urgency,2,stale/low urgency,1,0,149803,,,,0,98a99fb5e842fd3e117c614f91b99fe6145aa63a5d6294930bb2b07aa28086f4
thread:issue:149804,issue #149804: torch.linalg.norm RuntimeError with torch.func.grad( vmap( hessian(.) ) ),stale/low urgency,2,stale/low urgency,1,0,149804,,,,0,58e562c016395d8ac5622da5d85f134a69e7553dc686445e4e077d6a8e86e93e
thread:issue:149815,issue #149815: comparison operators only accept scalars as the 2nd argument but not as a 1st argument,stale/low urgency,2,stale/low urgency,1,0,149815,,,,0,9e315f847ce1fea1ec5192311fa45243160ae7113b36d3c7d7928f21130f395e
thread:issue:149821,issue #149821: `Aborted` error when using `torch.cuda.memory.caching_allocator_delete`,stale/low urgency,2,stale/low urgency,1,0,149821,,,,0,436faa0fc030760822e68a2b0f7d293016dbb17a33bcfef0b56a6cf7c68e8cf1
thread:issue:149837,"issue #149837: Intermittent ""AssertionError: can only test a child process"" Warning with PyTorch DataLoader on Colab",stale/low urgency,2,stale/low urgency,1,0,149837,,,,0,54b103b52a62a61d0bec572e8b93d648acc80d0e7e5ff84944f8e8f7a7007b31
thread:issue:149841,issue #149841: `weight` parameter on functional not in nn.Module,stale/low urgency,2,stale/low urgency,1,0,149841,,,,0,e6852b678543649173b865b88ae81f93ed86dc9923f0f0484b024b92caa0a2d3
thread:issue:149845,issue #149845: CUDA error: no kernel image is available for execution on the device RTX5090D,stale/low urgency,2,stale/low urgency,1,0,149845,,,,0,6767d185ebb1371d82a1c6d02baf255886f79f6bb064ef304fb44455f0b18f4d
thread:issue:149872,issue #149872: Do all lazy imports for torch.compile in one place?,stale/low urgency,2,stale/low urgency,1,0,149872,,,,0,71f8aa21249b4110b979880738b973c81afd8e7f9c4b43e93d03a00f45c4c425
thread:issue:149879,issue #149879: Inductor Pattern Matcher's register_replacement function only works with functional `search_fn`s,stale/low urgency,2,stale/low urgency,1,0,149879,,,,0,c3d560a1f0c2630b166ed0ac1c394d2bab6e54c2200134701e3d89a69becc43b
thread:issue:14989,issue #14989: [Caffe2] CNN Training on CPU is faster than GPU,stale/low urgency,2,stale/low urgency,1,0,14989,,,,0,7823c5769612b11f3739d31627adbcdd474312e524916fb53067d7ceca379e2d
thread:issue:149906,issue #149906: [ued][chatTTS][dynamo] graph break on should_compile_partial_graph=False,stale/low urgency,2,stale/low urgency,1,0,149906,,,,0,a5880a431f8c8e3547b7db963694d9a199b1473004108a01949a7821e4869101
thread:issue:149908,issue #149908: [ued][ChatTTS][guards] Too many recompilations,stale/low urgency,2,stale/low urgency,1,0,149908,,,,0,06fbe3823fcf370e6b90c4113028dd25c15235c993fa9a3f4e3dea63eb2f9347
thread:issue:149934,issue #149934: inconsistant behavior to pass tensor/tensor_list from torch.distributed module to cpp API,stale/low urgency,2,stale/low urgency,1,0,149934,,,,0,6ac21dfab36dd2eadd80dc8efd3cfd7634871e3cbf5394abc428d9d922ab434e
thread:issue:149944,issue #149944: include cudagraph skip reasons in tlparse output,stale/low urgency,2,stale/low urgency,1,0,149944,,,,0,bce5743953407051a7ebc158de1cbc2e3eb1afd439916e387037a81cc75de41f
thread:issue:149956,issue #149956: [ued][flux][dynamo] Wrong error message with dynamo.disable,stale/low urgency,2,stale/low urgency,1,0,149956,,,,0,c00a8589ab03138830bde2ff400b6bcdd700541423f81ad43aae2b2705086823
thread:issue:149957,issue #149957: [ued] Prohibitive warm start latency,stale/low urgency,2,stale/low urgency,1,0,149957,,,,0,b1c721f1f2cf8017cb55da61bcc879bfdbde04fb59a14aa436bb0c5806639cc4
thread:issue:14996,issue #14996: as_tensor does not use the device of the default tensor type,stale/low urgency,2,stale/low urgency,1,0,14996,,,,0,8f31978de5cac593ea22410926992a64cde87e73679252ace6cad3618df445bb
thread:issue:149962,issue #149962: [ued][wan] Dynamic shape issue with tolist,stale/low urgency,2,stale/low urgency,1,0,149962,,,,0,2fa92b983f0d6a1be80339a86e0ab2c93fd8aa33c741ae9d7a820f30c274cddd
thread:issue:149970,issue #149970: [ued][deepssek-vl2] Graph break on copy.deepcopy,stale/low urgency,2,stale/low urgency,1,0,149970,,,,0,4d34e81780b6bb5cec989bf06abff204cf714ce4e176725f0090911fce700747
thread:issue:149984,"issue #149984: ncclSystemError: System call (e.g. socket, malloc) or external library call failed or device error",stale/low urgency,2,stale/low urgency,1,0,149984,,,,0,ca8b5e27a4721aa9a1b070d270e5a4a8e8f197af2a1c90059dbbeace860ac49e
thread:issue:149990,issue #149990: [Async TP] Fuse all-gather-matmuls for float8 rowwise training,stale/low urgency,2,stale/low urgency,1,0,149990,,,,0,359e9467fdd73d1693776008d49b038044fe169c3528ed3d2afdb2b630d4871f
thread:issue:150017,issue #150017: `__setitem__` with bool mask and dtype mismatch fails,stale/low urgency,2,stale/low urgency,1,0,150017,,,,0,78d4cbdf2141a9f38e7dc10ee12d44494af5551e0e411a6703da3a6398249aec
thread:issue:150021,issue #150021: [ued] I want to see the full sizes/strides in TORCH_LOGS=recompiles,stale/low urgency,2,stale/low urgency,1,0,150021,,,,0,81cce284d3cce46f17601b6d042ba10bf5c8dc6de56c1ecbd2407342e3406ce9
thread:issue:150049,issue #150049: 2.7 test docker image has NCCL with older CUDA,stale/low urgency,2,stale/low urgency,1,0,150049,,,,0,f06eda7d2fa6bc2a8699d603668ad1e45073305a0d87f339a2c739e40bd4682a
thread:issue:150078,issue #150078: [RFC] Remove periodic/unstable jobs that has been continuously broken for more than 30 days,stale/low urgency,2,stale/low urgency,1,0,150078,,,,0,acaaa767a2106dfcf2bdee5557ba4e7ed23dd252209ad11f0662c272d0ee0e16
thread:issue:15009,issue #15009: [Caffe2] Caffe2Config.cmake,stale/low urgency,2,stale/low urgency,1,0,15009,,,,0,bd87704c17b1d2b9d4d945b2fb89e70796d8d780a92b7b031860fd9bd66f3e50
thread:issue:150096,issue #150096: FSDP OOM during `sync_params_and_buffers`,stale/low urgency,2,stale/low urgency,1,0,150096,,,,0,057a011f059b0b98a12e36de1bc14ea950d2ed0326d51ac0fd96f5ff430c04a1
thread:issue:150118,issue #150118: PyTorch 2.6 License Issues,stale/low urgency,2,stale/low urgency,1,0,150118,,,,0,c8b08f90b8f7bcc30287bd6d532d682bfbf0b90cd9f0ca97ed1a1b9c07a89653
thread:issue:150124,issue #150124: torch.unique undocumented behaviour,stale/low urgency,2,stale/low urgency,1,0,150124,,,,0,5f77186464df65441bf0190e6c0fa670db443704bf0e8706b74721b7304c368f
thread:issue:150140,"issue #150140: FSDP2 issue with mp_policy, checkpoint() and float input",stale/low urgency,2,stale/low urgency,1,0,150140,,,,0,eb0cbb8cd25d15dc3d26f735ee083a4da78f68eabfca371e3e6f1a07dae0dacb
thread:issue:150186,issue #150186: `MaskedTensor` 10x slower than `Tensor` compared with `nn.functional.softmax`,stale/low urgency,2,stale/low urgency,1,0,150186,,,,0,7f288c82a0ec558a058a1babb9c2729c8648311767d1ae71fd6375646c85400f
thread:issue:150190,"issue #150190: auto_functionalize HOPs are cacheable, also we should check that we allow for manual overrides for cacheability",stale/low urgency,2,stale/low urgency,1,0,150190,,,,0,cd9c4c76c2ee86142d71c166269ebaffeeb3ee3d5f02a844f39979ff51383f10
thread:issue:150242,issue #150242: Feature Request: Enhance Nested Tensor Operations for Direct RoPE Application,stale/low urgency,2,stale/low urgency,1,0,150242,,,,0,51e71a1934a6b70e2da2966ab1ad4744ed1f42e29f3dcaa76b9c3f8bfc8dc9ae
thread:issue:150252,issue #150252: RuntimeError: Subtracting Reconstructed Jagged Nested Tensor Fails with Shape Mismatch,stale/low urgency,2,stale/low urgency,1,0,150252,,,,0,5f872dd2b14bf4c81af4340ba584df1ac053d3a59dfd101c787c4736cf97ae85
thread:issue:150257,issue #150257: Type hint bug report relating to Sequential,stale/low urgency,2,stale/low urgency,1,0,150257,,,,0,1c64cc94fd8f55085114ee313a7ec7017af3e69a048b294f2b0c74706c04fd2d
thread:issue:150284,issue #150284: torch.fill bug,stale/low urgency,2,stale/low urgency,1,0,150284,,,,0,59674ec2e282317b2fe1b26a21bfc2ba046613e9afc16ec548444d088437c394
thread:issue:150328,issue #150328: ☂️ Update submodule dependencies to supported version of Cmake,stale/low urgency,2,stale/low urgency,1,0,150328,,,,0,ca8cc8bd61badc1839249bfd6c3dee0c89bfd457d2830eaf41a4a358aa792f9e
thread:issue:150336,issue #150336: [parallelize_module] uneven sharding + use_local_output breaks,stale/low urgency,2,stale/low urgency,1,0,150336,,,,0,0e925440ee58ec78074350ce62fc8129f7f86362922c3589b546cea606b4221c
thread:issue:150364,issue #150364: AOTI doesn't error if dynamic shape gets specialized during lowering,stale/low urgency,2,stale/low urgency,1,0,150364,,,,0,63eccaa34892c90a4e13bfacb8cd59e78c4d7f28f2752b6440ae35fa5df2c85b
thread:issue:150368,issue #150368: Support `copy` kwarg in `torch.reshape()` following Python array API standard,stale/low urgency,2,stale/low urgency,1,0,150368,,,,0,cbc5843bf16306fffc9664a8bbcd884fb56f926cc0b050242a500638c5d86dae
thread:issue:150399,issue #150399: torch wheels are unusable if CUDA RPMs are installed on the system (was Import error in nvidia/cuda:12.6.3-cudnn-devel-rockylinux9),stale/low urgency,2,stale/low urgency,1,0,150399,,,,0,6d86ee8aeede321c4f6950b48bd659a895df6bdbe1837c4d0413bc42132af516
thread:issue:150402,issue #150402: Feature Request: Add minLSTM and minGRU Modules,stale/low urgency,2,stale/low urgency,1,0,150402,,,,0,cb41ddde73422a4d8fb348e138e6fdaed9c6aca4df8e206e1fcfae2fdc8cff22
thread:issue:150476,issue #150476: Better test coverage on _inductor/scheduler.py,stale/low urgency,2,stale/low urgency,1,0,150476,,,,0,81ebc8a7fb8946e2d2b43e1c8b2114b06f315bc6944c8940c59c655ced027dc6
thread:issue:150482,issue #150482: API to specify static input indices for cudagraph,stale/low urgency,2,stale/low urgency,1,0,150482,,,,0,451578ff1c7275464a17a2e5e229363604bcf0d802098b1005f3b75d431eab45
thread:issue:150484,issue #150484: API to specify cudagraph sizes,stale/low urgency,2,stale/low urgency,1,0,150484,,,,0,d657f93b3cec7a955aa381895b46a9f9fa213fdd38006925f281aac8281e28ce
thread:issue:150488,issue #150488: Lightweight CUDAGraph backend,stale/low urgency,2,stale/low urgency,1,0,150488,,,,0,0316bebd5af402742c8147a125cd407544847762b76f7a1d6ae045b849fba0c3
thread:issue:1505,issue #1505: Cannot find Intel MKL,stale/low urgency,2,stale/low urgency,1,0,1505,,,,0,f18df6256ea0d5ee757e8bd6a4c10dc66bda4ac3fc937179c2ef7ca5d267a289
thread:issue:150517,issue #150517: Pinned memory doubles memory usage for tensors slightly over 128MB,stale/low urgency,2,stale/low urgency,1,0,150517,,,,0,f3481eac9834aca20412622f34d47e4b32c726b6dc8a5e14d2b3bcf6d2bcf648
thread:issue:150531,issue #150531: Intermittent SSL certificate expiry warnings for `download.pytorch.org` (load balancer?),stale/low urgency,2,stale/low urgency,1,0,150531,,,,0,c45f0b6a2970e7ada13f2fc205063850da1785a19b049182135df02db43eacef
thread:issue:150544,issue #150544: [ONNX] dynamic_axes does not rename dynamic dimension in torch.onnx.export,stale/low urgency,2,stale/low urgency,1,0,150544,,,,0,3dc64ec38cb32f4c81858be8a4c8ff73cabd27a6380cefc56c9546f6f7689e20
thread:issue:150548,issue #150548: Segmentation fault when using torch.tensor from a non python created thread,stale/low urgency,2,stale/low urgency,1,0,150548,,,,0,8d24ba1a0e548f9e93cdaf3fe7158e3679388f776aa2f7388aae45891eadcf5e
thread:issue:150571,issue #150571: Improve speed of pytorch docs build,stale/low urgency,2,stale/low urgency,1,0,150571,,,,0,7cb994f944a7d28c373ae0df1e2182768b796102be52b5c46cf4c3e34331b944
thread:issue:150574,issue #150574: [mem profiler] mem fragmentation and pynvml view,stale/low urgency,2,stale/low urgency,1,0,150574,,,,0,d26be3a625d4eb75656cfbe53daf7274d69d7570e3fd60c7bbf54b5df64eae43
thread:issue:150585,issue #150585: [distributed] Crash when trying to use default PG after creating new PG,stale/low urgency,2,stale/low urgency,1,0,150585,,,,0,b0b1490862312dadd51b42f5fe61d3259e6531dfffa8d6ba3d6335dd0cb627d4
thread:issue:150614,issue #150614: conv2d fp8 support,stale/low urgency,2,stale/low urgency,1,0,150614,,,,0,0e1fd1bdabae577cf068a0f792926862baf957479dfd15e0ebefa0d0739e30dd
thread:issue:150663,issue #150663: Proposing torch.empty_cache (device_type) as generalization of torch.cuda.empty_cache() .,stale/low urgency,2,stale/low urgency,1,0,150663,,,,0,8a7b0bb1b58322ad82d24a64772a98f81a01909466dd3b47a9abc5bc708e733a
thread:issue:150700,issue #150700: CuDNN + H100 Gives Weird Gradients,stale/low urgency,2,stale/low urgency,1,0,150700,,,,0,c56d268486f6fa15e94d88ac1d97a9d24bae58c9df807c6721e9c39d411e905a
thread:issue:150718,issue #150718: Codegen or Lint for python-api.md,stale/low urgency,2,stale/low urgency,1,0,150718,,,,0,1383f58cfe2f4ee72e359e855d9e14ae5a242d121991a8d13c19c030cef93a5b
thread:issue:150734,issue #150734: Training/Fine-tuning fails with PyTorch 2.8 + 4x 5090 GPUs using DDP/FSDP/DeepSpeed,stale/low urgency,2,stale/low urgency,1,0,150734,,,,0,2ae80edbf628c10830e0a1bbacde47017ad4290ee23535d70d02d198ba7ec784
thread:issue:150799,issue #150799: FSDP in hybrid mode throws _saved_grad_shard error when backward is called on cross-rank all-gathered loss,stale/low urgency,2,stale/low urgency,1,0,150799,,,,0,fb1f00d9a323255ff87e08da58735e4e961edef69ea20b04cd90ee4605e40d27
thread:issue:150836,issue #150836: `Aborted (core dumped)` in `torch.cuda.nccl.reduce`,stale/low urgency,2,stale/low urgency,1,0,150836,,,,0,3572bbab33a0417a7b1fa71bfc5e2263c49677c1a4cceaa19b001fd06ae55f2f
thread:issue:150861,issue #150861: Allow non-Tensor subclass in `torch.Tensor._make_wrapper_subclass`,stale/low urgency,2,stale/low urgency,1,0,150861,,,,0,66c826af632e0c6cc544b981b230c1b4907d4bfa7f85afd71b9248b9689b95f1
thread:issue:150864,issue #150864: [dynamo] `torch.compile` graph breaks on `setattr` of type objects,stale/low urgency,2,stale/low urgency,1,0,150864,,,,0,054e056fc51293e2c53b4e93819825efb8f5957c7abf37ad8d6c6db38f04d422
thread:issue:150871,issue #150871: register_replacement should also respect arg_kwarg_vals,stale/low urgency,2,stale/low urgency,1,0,150871,,,,0,2390c0d8b0dd27808a9b218c1beee487e2d9296d8801468972426c1a4cb37a90
thread:issue:150873,issue #150873: Add Created On | Last Updated On to the docs,stale/low urgency,2,stale/low urgency,1,0,150873,,,,0,965cb3a1ae4e2089e7f1e88257dcc01b40b63fc46b01f65319cf1ffebd032885
thread:issue:150922,issue #150922: StrictMinMaxConstraint issue in pytorch 2.4.0,stale/low urgency,2,stale/low urgency,1,0,150922,,,,0,58d10a2949ef1137c51f3d35b47d963df684495f52fa929b967dbf4c2f197808
thread:issue:150928,issue #150928: ShardTensor gather will encounter an error when a local tensor on certain ranks has zero elements,stale/low urgency,2,stale/low urgency,1,0,150928,,,,0,9d003996366d34f89dc78b7f632c3dfe6aa93cdabc80156833ff52ab4763db2a
thread:issue:150961,issue #150961: [ued] `torch.compile` cause more than 2x slow down with HF diffuser GGUF Auraflow,stale/low urgency,2,stale/low urgency,1,0,150961,,,,0,548e751ddef18c770536d9697773fffe218ce4e4b4090761ba5d966387c25abb
thread:issue:150963,issue #150963: [ued] `torch.compile` yields lower latency when compiling transformer blocks only for ComfyUI GGUF Flux,stale/low urgency,2,stale/low urgency,1,0,150963,,,,0,742fb20c1ad52f3d151131ded08e669d5636a0ab96790068cf72e06805a6007d
thread:issue:150972,issue #150972: Escape hatch: way to dynamically add or remove tags from custom operators,stale/low urgency,2,stale/low urgency,1,0,150972,,,,0,71c1fde37050ab89be4febd0ec2c69b13d9345f224a00bb21b631d6210b39738
thread:issue:151015,issue #151015: update user defined triton kernel table to include strict vs non-strict difference,stale/low urgency,2,stale/low urgency,1,0,151015,,,,0,17099ca5179aade047b115133b96a3b585e70bb29be1920123e3874f65b110d0
thread:issue:151030,issue #151030: Unexpected memory usage in FSDP 2 Hybrid Sharding (HSDP),stale/low urgency,2,stale/low urgency,1,0,151030,,,,0,cb3af2b551c858e3e72e3a2a30e326b81802196d16f143064404cf38b24f20ff
thread:issue:151037,issue #151037: [release] CPU perf benchmark latency increase for 2.6->2.7 on c5.24xlarge and A100 instances,stale/low urgency,2,stale/low urgency,1,0,151037,,,,0,2f30420b589049b152494053d2e860746daea920afece428150716f054eddee9
thread:issue:151064,issue #151064: [ONNX] Use dlpack to transfer tensors when onnxruntime implements proper support,stale/low urgency,2,stale/low urgency,1,0,151064,,,,0,77fb3d80846a2ed053d900acb258afc8645ca2848b346b179e07b88df67a6a7c
thread:issue:151066,issue #151066: [ROCm] QR decomposition is much slower on MI300x than A100,stale/low urgency,2,stale/low urgency,1,0,151066,,,,0,bda7b3d477b24e653c065a707e33916e429ab49fd5831c3141704f2f8943bd4f
thread:issue:151110,issue #151110: Update epsilon logic to improve numerical stability,stale/low urgency,2,stale/low urgency,1,0,151110,,,,0,f2f5a46f7bc5d6981946cacda1b30cc76912a38bf19b2f21e27825cb40544f89
thread:issue:151113,issue #151113: Failure in test_vmap_autograd_grad_nn_functional_conv2d_cpu_float32,stale/low urgency,2,stale/low urgency,1,0,151113,,,,0,76df1291a48265a62089b7ab6dcc51063adcfa12adfa738d63033e4fbf9385a0
thread:issue:151159,issue #151159: Failed to destroy or init process group after calling _abort_process_group,stale/low urgency,2,stale/low urgency,1,0,151159,,,,0,083873e2c416d3402426439f922e629c3fd73f4cee676efa5617623f2e6ab0f8
thread:issue:151182,issue #151182: A problem discovered when computing complex matrices in deep neural networks,stale/low urgency,2,stale/low urgency,1,0,151182,,,,0,a11d22eb58fdf312fcd20a940996a1621ab1cbc2da1f14c445992d14b5f1ef0a
thread:issue:15120,issue #15120: Different implementations of upsampleBilinear between pytorch and caffe2,stale/low urgency,2,stale/low urgency,1,0,15120,,,,0,06416709089a52b9c006754a4d99ff8af6fa4c52f406a1690108cd73e59902ce
thread:issue:151222,issue #151222: [FSDP] Detail information about parameters when raising errors,stale/low urgency,2,stale/low urgency,1,0,151222,,,,0,288ef762f59c7540ef21d7a2a0e463188f4140ef29c0a214a528876dbc715631
thread:issue:151240,issue #151240: [funcol] wait() breaks the chain for backwards,stale/low urgency,2,stale/low urgency,1,0,151240,,,,0,874aa37d5918df6d5e431b0d696515e3da03eb4ef9425f10d484e64cf555b80b
thread:issue:151251,issue #151251: Actually support LOAD_BUILD_CLASS,stale/low urgency,2,stale/low urgency,1,0,151251,,,,0,ded410c6f1fd215dac687a939651348749cb6004880109ab7dda5597d2bf233e
thread:issue:151269,issue #151269: [PP] Hang when num_microbatches < stages for Interleaved1F1B,stale/low urgency,2,stale/low urgency,1,0,151269,,,,0,32b4e262000e78f9e145cfccd028d939dd98881f0ff2512740bf621628c057ac
thread:issue:151286,issue #151286: vmap- out of memory,stale/low urgency,2,stale/low urgency,1,0,151286,,,,0,5f21eea50671d5e44ecf2e2d5a1e3851da61b1814aab7b060a4aee57ff182660
thread:issue:151303,issue #151303: [Feature Request] Experimental support to Moore Threads GPU MUSA,stale/low urgency,2,stale/low urgency,1,0,151303,,,,0,bc1ecd97221c680e06132b54c223540f02a997a1bbd3fb473390d8a93554a955
thread:issue:151308,issue #151308: [Intel GPU][UT failure] Depthwise conv related UT failures,stale/low urgency,2,stale/low urgency,1,0,151308,,,,0,8fdd1c6bb90febd2d1b6c38b55aa5adeb6492e153a754c26af0863921e5dda41
thread:issue:151322,issue #151322: [CI] No workflows scheduled on PRs,stale/low urgency,2,stale/low urgency,1,0,151322,,,,0,aaa9f18cbd9c163a66dd7df123057708b40eb90fa831dccdfe05b25972bad24c
thread:issue:151402,"issue #151402: RuntimeError: curPos <= (kUpperBound - kAppendInterval) INTERNAL ASSERT FAILED at ""../torch/csrc/jit/ir/ir.cpp"":698, please report a bug to PyTorch.",stale/low urgency,2,stale/low urgency,1,0,151402,,,,0,c4190a87dd55773d06fdbde772de1a7b869c1d658d4bfed55608b118c2c8bd36
thread:issue:15142,issue #15142: Update third_party/googletest - Ability to skip tests in GTEST,stale/low urgency,2,stale/low urgency,1,0,15142,,,,0,1ebf4c3736d3dd077264050cdc3f588d33f2ccfdc2f32ee35908708172e6256d
thread:issue:151433,issue #151433: The compare interface of the memory snapshot visualization tool “_memory_viz.py” has a bug,stale/low urgency,2,stale/low urgency,1,0,151433,,,,0,bae622e6b408a3d35dce65fbcda5fc02f224bc78865e308c7cd754a25d543f24
thread:issue:151457,"issue #151457: When distributed.destroy_process_group() is executed, new memory usage will be generated on device 0, which may cause OOM under extreme conditions and thus abnormal exit.",stale/low urgency,2,stale/low urgency,1,0,151457,,,,0,ad4cad977d6d84cb67cb04b1bdd6e850c4395b0e4ccfb2eb21bc66b248a9f91d
thread:issue:151463,issue #151463: Inconsistent Output from nn.Conv2d with padding_mode='circular' When Using MKLDNN on a AVX2_Processor,stale/low urgency,2,stale/low urgency,1,0,151463,,,,0,e0bdaa9a108788a4c285947c51e48673f4856e8ac582a5a02c6c53ebfb33d015
thread:issue:15149,issue #15149: Use std::variant to represent C++ side enumerations (with binding support),stale/low urgency,2,stale/low urgency,1,0,15149,,,,0,ec0986d2ba54e50eb8c52a30e294589b99541f94926d07a0b2e1adf0001d6c8d
thread:issue:151555,issue #151555: [Inductor] Convolution auto-generation through Triton backend,stale/low urgency,2,stale/low urgency,1,0,151555,,,,0,4c0c31b998e7de2811c96bd76f62693ae80acb371be361971e0eb6c028c93548
thread:issue:151556,"issue #151556: Compatibility Issues Between DCP, LoRA, and Hugging Face Models",stale/low urgency,2,stale/low urgency,1,0,151556,,,,0,c0e93b71b71d65e98d6daaa8621696686eaf84823ab12fb1fb8c33c610ca455f
thread:issue:151591,issue #151591: Device Error on vmap,stale/low urgency,2,stale/low urgency,1,0,151591,,,,0,ed1ff4f986cfd65b3908ccef502d265102ffce648ea94ef2843069df5d1144c7
thread:issue:151592,issue #151592: Xcode 16+: duplicate LC_RPATH '@loader_path',stale/low urgency,2,stale/low urgency,1,0,151592,,,,0,9478cd6f7cf4220538c106d6c15128e47b3e6c98279287b64d3048fb35c7bcb2
thread:issue:15161,issue #15161: Maybe a bug when using DataParallel,stale/low urgency,2,stale/low urgency,1,0,15161,,,,0,4219a01ff3bf5dee33be38f8290e498d7ce0269a11bb64ff3629f2352cf56083
thread:issue:15167,issue #15167: cudnn not found,stale/low urgency,2,stale/low urgency,1,0,15167,,,,0,410efe421c6fd0b62144263c0ca7c36abcd3f10bff8130793af7164db323d2c8
thread:issue:151695,issue #151695: FP8 Support for FlexAttention,stale/low urgency,2,stale/low urgency,1,0,151695,,,,0,cc2b8332d126e73f0bcd2fbcc9447dddf7ac8bfdf4de370eb59c30975e0db6f2
thread:issue:151717,issue #151717: [NJT] Dropout with p=0 still drops values,stale/low urgency,2,stale/low urgency,1,0,151717,,,,0,2f7b9e7e17cda45cf5107fc6882ecf8579742381d59e6b1a679ef78fba4d7a76
thread:issue:151759,"issue #151759: ""_get_pg_default_device"" deprecated warning in ""Getting Started with Distributed Checkpoint (DCP)""",stale/low urgency,2,stale/low urgency,1,0,151759,,,,0,f2e59ca2f9a3ac77236e924d579eed65adcac5aadc2831267ce6d08cc04619db
thread:issue:151776,issue #151776: [dynamo] Some inefficiencies around handling __torch_function__,stale/low urgency,2,stale/low urgency,1,0,151776,,,,0,38123479ce2f800f1f8fe4f2cbf8e1eff33cc7a198ed76828bd8859173a4baaa
thread:issue:151793,issue #151793: Create decomp for searchsorted,stale/low urgency,2,stale/low urgency,1,0,151793,,,,0,727193fbdaace3bce41ef1fd293b801df056b7319ab809f3cc92906a8f2ded7e
thread:issue:151858,issue #151858: construcing DTensor on a 2D device mesh SIGTERMs,stale/low urgency,2,stale/low urgency,1,0,151858,,,,0,db8a1c4c876d275b2bf6f14f8f5cb0e8ed97ddcaa85a64f512dabdc99c28eda6
thread:issue:151878,issue #151878: Potential error in calling scaled_dot_product_attention in pytorch/torch/testing/_internal/distributed/_tensor /common_dtensor.py ?,stale/low urgency,2,stale/low urgency,1,0,151878,,,,0,d3672dd5abaa54b5c50de2e67d62a38ac7cc9986a0372d33d9c107a6a4a81a7f
thread:issue:151884,"issue #151884: The ""eager"" and ""aot_eager"" backends have different behavior for the expected gradient tensor of the torch.expend_as operator",stale/low urgency,2,stale/low urgency,1,0,151884,,,,0,1900c426c52a0c8efea60910cb6e95bb6e646aed1188058fb6ed8ec8f854fc3c
thread:issue:151934,issue #151934: `torch.bmm` is slow on non-contiguous BF16 CPU tensors,stale/low urgency,2,stale/low urgency,1,0,151934,,,,0,b3ea1ae94aa93359a4336e322fb6cbda6a88d3ba4f7a25d7f6bae79b1f376917
thread:issue:151965,"issue #151965: [rfc][c10d] RDMA APIs (read/write, rkey)",stale/low urgency,2,stale/low urgency,1,0,151965,,,,0,fc83015c5b8ac520b44f39e5f087e0105586af0eb7d356396856c3fa79f3a6ec
thread:issue:151978,issue #151978: torch.gather returns incorrect output on cuda after unsqueezing and expanding with double precision,stale/low urgency,2,stale/low urgency,1,0,151978,,,,0,0f712951c6661e7739adf26ae8d4e3ba6d6f43a562b5e21de54aff2770c86c95
thread:issue:152097,"issue #152097: When using torch to convert to oxxn model, testing the inference results with actual images shows tensor mismatch",stale/low urgency,2,stale/low urgency,1,0,152097,,,,0,3b955daa38bc4ad6981020c4a8267aa6d9a4cd06dd8e3938f7e1391eaa3629ba
thread:issue:152127,issue #152127: [C10D] Make collectives backwards throw an error,stale/low urgency,2,stale/low urgency,1,0,152127,,,,0,c7ddd3c186aba75bcbee7c1e9d6b111d845b66ba8f76d25db6fdf2a627ba5a4a
thread:issue:152134,issue #152134: [RFC] Proposed Changes to Feature Tracking & Classification for PyTorch Releases starting Release 2.8,stale/low urgency,2,stale/low urgency,1,0,152134,,,,0,6de05a77f3a60216d06a9d748679cfa028c0b906b90d618adbb7b5a8e482e5a5
thread:issue:152150,"issue #152150: [Scaled MM] Update to support on B200 TN, NT, NN, TT Layouts are supported",stale/low urgency,2,stale/low urgency,1,0,152150,,,,0,68db4ffecee35363c4a519b3aa5e6ab6e841503dfbe241b0d4a2ea16d30397cc
thread:issue:152152,issue #152152: padding_mode `reflect` works different from others in Conv,stale/low urgency,2,stale/low urgency,1,0,152152,,,,0,80f9897a55b19a398b43c35fae025da4f815ad2b3dc86066e41089dcb0c0f0d7
thread:issue:152188,issue #152188: The input for layers other than the first layer should be the hidden state from the previous layer.,stale/low urgency,2,stale/low urgency,1,0,152188,,,,0,d4fc1892f45c966dadd6163d4b8efe6dc66a25a46d86888ee86ec2bec60e5466
thread:issue:152190,issue #152190: HUD Dashboard sort by perf speedup doesn't do anything,stale/low urgency,2,stale/low urgency,1,0,152190,,,,0,ac41b1eb722024b8ae0e6b04f3c8b1a78b4da12b1b35ea59e64cb383b6ef0548
thread:issue:152212,issue #152212: [Release improvements] Have cherry-pick bot always add the current release to the PR,stale/low urgency,2,stale/low urgency,1,0,152212,,,,0,bfe639bcef9a1739436fbe83943d58f808a016323d81082e8934d3c59291f409
thread:issue:152257,issue #152257: [FR] Support BSHM-layout scaled_dot_product_attention without transpose.,stale/low urgency,2,stale/low urgency,1,0,152257,,,,0,c67a021aa9812f0a4e9cdeceed11979b689a6bbfc03ee0408f09ecadd1695167
thread:issue:152261,issue #152261: Context Parallel -- unsharded output doesn't match output without CP.,stale/low urgency,2,stale/low urgency,1,0,152261,,,,0,4511049e9052e288f58c4d41515f0a5718bd15e56f2b5d72f3684ff48436a475
thread:issue:152269,issue #152269: Arbitrary Code Execution Risk in `torch.distributed.utils.overload` When Misused in Type Annotations,stale/low urgency,2,stale/low urgency,1,0,152269,,,,0,498733f403fd5ef6ad693c4e63edcabbe8507759fae35657c7fb83d6915e0067
thread:issue:152294,issue #152294: Unexpected overflow behavior when using `torch.addcmul`,stale/low urgency,2,stale/low urgency,1,0,152294,,,,0,401561b4c59fd60ffb8432096f098a5fe4490aeaff8092ae6599719925e2c33c
thread:issue:152295,"issue #152295: `vmap` not working on `torch.arange`, `torch.scalar_tensor`, and `torch.ones`",stale/low urgency,2,stale/low urgency,1,0,152295,,,,0,028548a8016889cccb5c62c45d9b3fcf0d82ccd335b7309e4e33cd70a8995525
thread:issue:152308,issue #152308: bizarre behavior with torch module's Attribute Error,stale/low urgency,2,stale/low urgency,1,0,152308,,,,0,66935c9d391389b81303b147b751a33769b699abe1c12dd4fb028cc8b84be793
thread:issue:152337,issue #152337: Unusually slow draft_export time,stale/low urgency,2,stale/low urgency,1,0,152337,,,,0,992e16d2affed4c1cfd84df75573927790b4cfa4e45be31ab45947003690bc7b
thread:issue:152381,issue #152381: Outdated contents in dynamo overview,stale/low urgency,2,stale/low urgency,1,0,152381,,,,0,9add0b94102d0d9b9b5bcd6e4ffcf08ac009755595fab0d0e4545a9d96c27a38
thread:issue:152426,"issue #152426: [Manylinux 2.28] Migrate Docker container to use gcc 13, CUDA 12.6 and gcc14 CUDA 12.8",stale/low urgency,2,stale/low urgency,1,0,152426,,,,0,5b8b73b2a51420384e3c6cf5a87b04f64b10c10faaf750345d8aedf665323beb
thread:issue:152437,issue #152437: `nn.CrossEntropyLoss` accepts negative target probabilities,stale/low urgency,2,stale/low urgency,1,0,152437,,,,0,2bcacea355b806aa601e7bb3098cce134388e44dc67e52ec13dc77aa5bc189d5
thread:issue:152489,issue #152489: UNSTABLE Lint / Lint URLs / linux-job,stale/low urgency,2,stale/low urgency,1,0,152489,,,,0,187f4e075982a797c8e3042fe332402d9ec348c9bf6b4d2a7b814ef466f38d46
thread:issue:15249,issue #15249: [caffe2] controlled forward pass,stale/low urgency,2,stale/low urgency,1,0,15249,,,,0,e6368939b5e174f3321b78f619243ee97d4fd0b0b9a2e1e831152919fc8f8fb8
thread:issue:152527,issue #152527: can't reconstruct the communication group using PyTorch.,stale/low urgency,2,stale/low urgency,1,0,152527,,,,0,883785776bbef000c36f209c728aeff049e763a47e8160b8dab28f7bd267ccfe
thread:issue:15253,issue #15253: [Feature Request] cdist: pairwise distances between two sets of tensors with batch mode,stale/low urgency,2,stale/low urgency,1,0,15253,,,,0,19f5084d2425605abfb44cb963063f0c9cdbed9f571f0f6fb84b190ce2807a43
thread:issue:152550,issue #152550: MPS varying seq len SDPA memory leak,stale/low urgency,2,stale/low urgency,1,0,152550,,,,0,5ae7761046a4fa07e80d87403f4bdb4c7779ce7745f898c4d0c1ca5d5a718192
thread:issue:152566,issue #152566: [Benchmark] High compilation time variance on benchmark dashboards,stale/low urgency,2,stale/low urgency,1,0,152566,,,,0,a903c41147fb067ae2a613e2f109f4607296454edd0bcb9844303a85ad317a29
thread:issue:15259,issue #15259: err:torch.nn.CrossEntropyLoss,stale/low urgency,2,stale/low urgency,1,0,15259,,,,0,d369aa8d034d6659e8aee77e3c425145ec22fa12f2a014bcb03edf55bf91f598
thread:issue:15260,issue #15260: MultiGPU for gru,stale/low urgency,2,stale/low urgency,1,0,15260,,,,0,38229d6f178fce6ec3bec105632ac61af9906848d0293c06a5146888fc4fbeb2
thread:issue:152607,issue #152607: Loops impacting output when utilizing hooks,stale/low urgency,2,stale/low urgency,1,0,152607,,,,0,635df5c6bd3e3d2544882a429182108bfb32b42308af1728bde7a596888b475e
thread:issue:152608,issue #152608: [triton pin update] Run Inductor CI on pin updates for Triton and the PyTorch nightly branch,stale/low urgency,2,stale/low urgency,1,0,152608,,,,0,3f61a74ff3dda6186268873867f39a033e68499aed10dd6ad2d060fbf07b9286
thread:issue:152645,issue #152645: ProcessGroupGloo.allgather_into_tensor_coalesced crashes with CUDA tensors,stale/low urgency,2,stale/low urgency,1,0,152645,,,,0,64d5a84dacd4a984f506df11a480e4551df925714026e5303fa50a70a58f049c
thread:issue:152647,issue #152647: Check for if two tensors are overall similar instead of bitwise similar?,stale/low urgency,2,stale/low urgency,1,0,152647,,,,0,b6f9275608d43f37c9fffe8a518b7530cfe6131cfa245a12d9342486bac58e10
thread:issue:152699,issue #152699: [CI] [anaconda] Triton windows build,stale/low urgency,2,stale/low urgency,1,0,152699,,,,0,ed534e1dbacb34b3c5563ecdf9101220731d1821a5c619ba063d5e6678984278
thread:issue:152717,issue #152717: Cannot mask a DTensor,stale/low urgency,2,stale/low urgency,1,0,152717,,,,0,6190fbe0a696c3830adeaea5551cbf4e83a5fa53c2c2a9901f49db358e6371af
thread:issue:152720,issue #152720: When scoped_libary is destroyed the fake impls are not cleared,stale/low urgency,2,stale/low urgency,1,0,152720,,,,0,9423ad57e290669d131474bd85c3232cdee80e00e2dd0598ef071ce7e6a0dd30
thread:issue:152731,issue #152731: Inconsistent float16 overflow behavior between CPU and CUDA devices,stale/low urgency,2,stale/low urgency,1,0,152731,,,,0,47264938f23918a521deb9437b623440d33ff2dc6b50f7f7d9c9ef9db39962ab
thread:issue:152736,issue #152736: [MPS] TensorIterator and accuracy,stale/low urgency,2,stale/low urgency,1,0,152736,,,,0,f3720729ccae502429631a34bf1c17486ef930415a406785ab894221f1dd2153
thread:issue:152746,"issue #152746: [FSDP2] fully_shard(mesh=(shard, shard)) for intra and inter node all-gathers",stale/low urgency,2,stale/low urgency,1,0,152746,,,,0,6e6e99e207b4262edbbfe2a19e95e691b41440197f2311eac087cf793e7347d4
thread:issue:152752,issue #152752: Checkpoint sequential doesn't raise clear error when segments is greater than number of functions,stale/low urgency,2,stale/low urgency,1,0,152752,,,,0,03eb00ba52d9f5ef57326b49b0796f2aef6ecb38b65fbda393820010fd965cb1
thread:issue:152756,issue #152756: Cuda-12.9 removed libnvToolsExt.so.* and is now purely header nvtx3,stale/low urgency,2,stale/low urgency,1,0,152756,,,,0,10d9782fa81bbc4f723ba900c969f6d66e6735c7798dc266856496de797a024e
thread:issue:152829,issue #152829: Don't hardcoded support for DTensor to_local/from_local/redistribute into dynamo,stale/low urgency,2,stale/low urgency,1,0,152829,,,,0,5724b2f1446cb5eea6a1b24faa99eadfe799bd23957c567b0209770ee6018cf9
thread:issue:15284,issue #15284: index_add_ with scalar values instead of tensors,stale/low urgency,2,stale/low urgency,1,0,15284,,,,0,a65c07f35d78311562e557754489b7f8a699ca4c031b67c83105d1753a13bf6a
thread:issue:152869,issue #152869: DISABLED testAssertNotRegex (__main__.CPythonTest_Assertions),stale/low urgency,2,stale/low urgency,1,0,152869,,,,0,2d23370c0e62bb88bf3d3f102b394e59a4e41b529164109c152cdcea82b9aa04
thread:issue:152870,issue #152870: Docs Update `wrap_triton`,stale/low urgency,2,stale/low urgency,1,0,152870,,,,0,b77b593991de7562308e9302823ebea37fb4aea706d0fffcb41b411a8f114123
thread:issue:15288,issue #15288: possible unsafety in torch.distributions.kl_divergence for Bernoullis,stale/low urgency,2,stale/low urgency,1,0,15288,,,,0,e0abaeb57044b2f569f6517742773323631ee4c9b34bca552c9ec9da72aa8ed9
thread:issue:152884,issue #152884: UNSTABLE Lint / Link checks / Lint URLs / linux-job,stale/low urgency,2,stale/low urgency,1,0,152884,,,,0,984a18c5786d7656f87807eb67a9cfdceeb7ebedbc3f449ec997bb20ba6f37d2
thread:issue:15290,issue #15290: [caffe2] Installation problem on OSX,stale/low urgency,2,stale/low urgency,1,0,15290,,,,0,36b14012357c538ad15a95611fd38c6ae572b095cb57f67b70f89bf06f11bdbe
thread:issue:15291,issue #15291: linked error of Pytorch 1.0 release,stale/low urgency,2,stale/low urgency,1,0,15291,,,,0,bfe8fea64474b7746228676d2301f6ac9c288d5b8a09fc925e08c7a046a99bee
thread:issue:152922,issue #152922: Enable 12.8.1,stale/low urgency,2,stale/low urgency,1,0,152922,,,,0,9db1129e46837dbd5fed19b0e799f39a04e302366be080d11b3ae65d0f4fc44f
thread:issue:15294,issue #15294: manylinux2014 compatible wheels,stale/low urgency,2,stale/low urgency,1,0,15294,,,,0,a0920dce7c95f9dc12bb5ddc47c3847ad91a3c36dd08d9a65b1a7bf975f8b7fe
thread:issue:152962,issue #152962: TestNestedTensorOpInfoCUDA.test_compile_backward_matmul_cuda_float32 Test Failure,stale/low urgency,2,stale/low urgency,1,0,152962,,,,0,63360bc0e19c1d2562916a44ce1b3ed69c74bdcfc8162d11325b509e48e1d0ac
thread:issue:152964,issue #152964: [FSDP2] need dummy forward/backward to stay SPMD,stale/low urgency,2,stale/low urgency,1,0,152964,,,,0,7574835e4165e757fbcf99a9eac421ccb22a8f4b979a814c2e23989017d9a0e7
thread:issue:15298,issue #15298: Momentum problem (1-momentum is correct?) in BatchNorm2d,stale/low urgency,2,stale/low urgency,1,0,15298,,,,0,c975a43856425e2a5c6a7f07024b01834bd2da2492e29a4b864688ae471d93c0
thread:issue:152992,issue #152992: conv2d with int8 on CUDA: GET was unable to find an engine to execute this computation,stale/low urgency,2,stale/low urgency,1,0,152992,,,,0,aa3d747c7ee47ef99effbe1d47247fd18368fd2cd7e9aaa6aee8b28d92e894fb
thread:issue:152994,issue #152994: [dynamo] Actually support functools.lru_cache,stale/low urgency,2,stale/low urgency,1,0,152994,,,,0,78fb6ff8d52b0a5863677876ad001b3d270f8d153b9b9f61e74d9440f02eabcf
thread:issue:153025,issue #153025: Multiple CUDA graphs utilizing multiple CUDA GPUs encounter illegal memory access during replay,stale/low urgency,2,stale/low urgency,1,0,153025,,,,0,4dd4b4fd95107ec245831f4e9c3a57584e417f05880addc37bde35a3555b66ed
thread:issue:15303,issue #15303: CMake Error at CMakeLists.txt:10 (find_package) in C++,stale/low urgency,2,stale/low urgency,1,0,15303,,,,0,e8c8cb6f84e0cd20e0020c2e6660deb815db7e4d5d4478fc1250f3ad7c741c03
thread:issue:153041,issue #153041: gen_alias_from_base ruins the result of view after inductor generated a copy for the results of the view operations.,stale/low urgency,2,stale/low urgency,1,0,153041,,,,0,be2503980b9b526986fbb0a924bc52e5859c5daf75021374335116da288be608
thread:issue:15309,issue #15309: libtorch without cmake,stale/low urgency,2,stale/low urgency,1,0,15309,,,,0,8a83a6832ad0dd306c8b3cba86097a17fc0c34ea47ca2cb21ea675aa50706249
thread:issue:153103,issue #153103: [Inductor] Investigate computing global amaxes via atomics instead of 2 stage reduction approach,stale/low urgency,2,stale/low urgency,1,0,153103,,,,0,0c0fac08ff6e127c3dfe751318275cc438b4c23c1ea057e7546ad0d0fc4ec0b1
thread:issue:15313,issue #15313: AttributeError: module 'caffe2.python._import_c_extension' has no attribute 'get_cudnn_version' when Caffe2 is not built with CuDNN,stale/low urgency,2,stale/low urgency,1,0,15313,,,,0,bd5651db71fe60c3bcc58c2f3f4900eede6f21169d27c03f4b85157895a19b33
thread:issue:153139,issue #153139: Floating Point exception in Convolution with disabled SMT,stale/low urgency,2,stale/low urgency,1,0,153139,,,,0,8f6a8bc31ad8a828a0451858c01ca5af28e542be96a036163f899215c9cb97b8
thread:issue:153187,issue #153187: Inconsistent overflow handling in `torch.clamp_min` between CPU and CUDA for float16 tensors,stale/low urgency,2,stale/low urgency,1,0,153187,,,,0,72c186a5a6035f29926332a6734fb5651b4f67cb3c9fc09bfcbe2270a9cbfa83
thread:issue:153210,issue #153210: Inconsistent Error Handling in `torch.addcmul` with `int16` Tensors,stale/low urgency,2,stale/low urgency,1,0,153210,,,,0,bcb6e2e47500ddfef7146ac90f7f716b2841324f9071ee24947144bdddeeb44f
thread:issue:153234,issue #153234: `torch.quantile` perform differently on cpu and cuda,stale/low urgency,2,stale/low urgency,1,0,153234,,,,0,8b12d6001c2702bb3b8ed4e97473123feaa47120cbd1ebf3115383364fe0a7bc
thread:issue:153238,issue #153238: [NJT] can only chunk if the 2nd dimension is ragged,stale/low urgency,2,stale/low urgency,1,0,153238,,,,0,45b4dc6107a261de064f3b8a2c40f2455d69703f54418523b2f71659a0f01e1d
thread:issue:153241,issue #153241: Online softmax is disabled on the fly,stale/low urgency,2,stale/low urgency,1,0,153241,,,,0,af9108f486188ccd1e93f50a91975ce13eed0ea7b97c0463aefa6e1a0ca06430
thread:issue:153242,issue #153242: [RFC] Add a SlimTensor representation to help AOTInductor generate standalone binaries,stale/low urgency,2,stale/low urgency,1,0,153242,,,,0,b4d9469bb2ea40a971b056e3647c36e00436f781eb50006d430440fe356f07fe
thread:issue:153273,issue #153273: Auto format lint not making suggestions,stale/low urgency,2,stale/low urgency,1,0,153273,,,,0,13b4c253e231e4cfd2d553ac5887c3cc7b26ad555af5e4a8918145be65ee6192
thread:issue:153327,issue #153327: FPE when calling `torch.pixel_shuffle()` with empty tensors and large upscale_factor,stale/low urgency,2,stale/low urgency,1,0,153327,,,,0,a340bd9bf5e2c7b8e32ed493f746a96902f323281b9b95b4c439b3213b3b1f6d
thread:issue:153329,issue #153329: Segmentation fault when converting sparse COO tensor with complex values to dense,stale/low urgency,2,stale/low urgency,1,0,153329,,,,0,ae1a6d5d71290a3fcdac2c388fb0fd5bb9b9ee564f7c8714218eead5f2e48111
thread:issue:153334,issue #153334: Selective Activation Checkpointing on custom autograd.Function,stale/low urgency,2,stale/low urgency,1,0,153334,,,,0,175bd3a5bdd3cd2ef9759b5df4143a1e3e0e272ce5ce3c6c14a48bb2e15f3990
thread:issue:153337,issue #153337: `torch.combinations` exhibits excessive memory usage and hangs for moderate `n` and `r` due to `n^r`,stale/low urgency,2,stale/low urgency,1,0,153337,,,,0,06a2cb9f0c9b5167e84aa5d8d9c37b243d86012a34967d289f159a6407b5c36a
thread:issue:153360,issue #153360: torch.aminmax reducing over all dims should not accept keepdim,stale/low urgency,2,stale/low urgency,1,0,153360,,,,0,f21f68766582e1a441b42da82540bc820fee1709a33570d19be5f953aa1bcefd
thread:issue:153361,issue #153361: MANIFEST.in is not being used,stale/low urgency,2,stale/low urgency,1,0,153361,,,,0,3dde5218bebedd1beca5cf008253166d902ab54d5f8a1c460086510b3b1da721
thread:issue:153463,issue #153463: Feedback about Class Tensor,stale/low urgency,2,stale/low urgency,1,0,153463,,,,0,5006400cac64d8d7754bd6794069a2c540902b010df0c17ceeb0adbf0f19793e
thread:issue:153471,issue #153471: RFC for vector length agnostic SVE Vectorized class,stale/low urgency,2,stale/low urgency,1,0,153471,,,,0,937a2568b6a63923acd530ca0c8157cc953311439a3f6190301919c9dcf9a665
thread:issue:153472,issue #153472: Multihead Attention does not work with jagged tensors due to __torch_function__,stale/low urgency,2,stale/low urgency,1,0,153472,,,,0,e457b462b949269829c89c23860779efc8aeb06d2c19cc1db4aeabfb44e2d653
thread:issue:153497,issue #153497: NotImplementedError: Could not run 'aten::log' with arguments from the 'SparseCUDA' backend.,stale/low urgency,2,stale/low urgency,1,0,153497,,,,0,6bbd79ba3e44b7744432054ea665b29dcabbaf7d36d87799d4dbd2718aeb961f
thread:issue:153502,issue #153502: Add MXFP8 Support to scaled_grouped_gemm,stale/low urgency,2,stale/low urgency,1,0,153502,,,,0,6a2de96c616e0234049dbcdfa065654aef7a0feb4b149744046268cf22708215
thread:issue:153531,issue #153531: Documentation: explaining the STFT formula,stale/low urgency,2,stale/low urgency,1,0,153531,,,,0,7d3cc33f28e21f50568f1e35c4505bd69d27b77630881f7f6fa99760a8d254d9
thread:issue:153532,issue #153532: [RFC] [Inductor] Custom pass registration interface,stale/low urgency,2,stale/low urgency,1,0,153532,,,,0,285693ff17dfb9bb0e8931f04620237fdb03a89d1fce5d6bcf2448108de280a4
thread:issue:153534,issue #153534: `pca_lowrank` on CUDA is significantly slower than on CPU,stale/low urgency,2,stale/low urgency,1,0,153534,,,,0,57e3ab7f8760b0c1f3e69fc24eaff10ae5738b213800218fd827c01f62fbe559
thread:issue:153536,issue #153536: `torch.utils.data.default_collate` raises misleading warning for read-only NumPy arrays,stale/low urgency,2,stale/low urgency,1,0,153536,,,,0,4dc64eac47ccc3e0a8555117217c63fa37ed1ed607a13b5a829c6401dcd9d9af
thread:issue:153542,issue #153542: torch.set_ on a view does not sever view relation,stale/low urgency,2,stale/low urgency,1,0,153542,,,,0,7fd6efb833c22c7c40391518e21607903653963ff0ac2f471198b41f45d9a0db
thread:issue:153594,issue #153594: Significant DataLoader worker startup delay and sequential-like initialization with num_workers > 0 on Windows (using 'spawn'),stale/low urgency,2,stale/low urgency,1,0,153594,,,,0,c86d60968827550177636497caa52f367d3586c1fcd9a91718b50f45d11d153e
thread:issue:153606,issue #153606: native/indexingUtils.h `AdvancedIndex` has internal linkage,stale/low urgency,2,stale/low urgency,1,0,153606,,,,0,a6c5e571098afb2391e8c84347e6c4ea2c0e76ba20c630db490bdb5f2b65ca6d
thread:issue:153638,issue #153638: Can't run lintrunner locally on Windows machine,stale/low urgency,2,stale/low urgency,1,0,153638,,,,0,fdc2e77391d2a5ab84c636331f6125cb28db6244bba2f6f0c5a783d7e5c070c1
thread:issue:153649,issue #153649: Use opmath_t and not double compute in fused optimizers,stale/low urgency,2,stale/low urgency,1,0,153649,,,,0,ab6ac178dec1efb70528f5ea581e0c7c87b02c65f9c624150a4c0096fc99e8fb
thread:issue:153673,issue #153673: [FSDP2] revisit NCCL group coalescing vs copy-in/copy-out,stale/low urgency,2,stale/low urgency,1,0,153673,,,,0,8bdbbaa2df54296738f1dc91cc306cbe45dcf003949411648513055e40e58e76
thread:issue:153737,issue #153737: SystemError: PY_SSIZE_T_CLEAN macro must be defined for '#' formats - PyTorch compile fails with Python 3.12,stale/low urgency,2,stale/low urgency,1,0,153737,,,,0,82c18f3dbe1939af903b2685f4a517a420c249d49ddb329c6a0e59f42ac41cb5
thread:issue:153745,issue #153745: [RFC] [Feature] Intra-Device Heterogeneous Memory Allocation Support,stale/low urgency,2,stale/low urgency,1,0,153745,,,,0,54d7ce77e25260708bc2e883f5f6f6eeac6150e0cf619fa2be5d8fb5cf5c3d74
thread:issue:15378,issue #15378: Projective Transformation grid generator,stale/low urgency,2,stale/low urgency,1,0,15378,,,,0,d8df45ccb7a34062bfb0002b34ef2998bb8fd364aefa6f2770fea1644fec49c5
thread:issue:153826,issue #153826: Unable to reproduce dynamo bechmark script results,stale/low urgency,2,stale/low urgency,1,0,153826,,,,0,b0b4cb474fa6fa1c18dfb603760743d8114ff56fda00c9604453a05afaca2094
thread:issue:153832,issue #153832: export torch.cond fails at capturing identity,stale/low urgency,2,stale/low urgency,1,0,153832,,,,0,c607a0bd0e77d879890385ecd25eec2b15215e6e588a2bf3695211b2751a23f8
thread:issue:153842,issue #153842: Duplicated `milestones` keys in `torch.optim.lr_scheduler.MultiStepLr` after loading a checkpoint,stale/low urgency,2,stale/low urgency,1,0,153842,,,,0,8488888e9a896137e4bfc2830a3d9739d6b4cb5cca81bd4a40ff2f0abb9222a8
thread:issue:153861,issue #153861: [BUG] `functionalize` silent data corruption with pre-strided tensors,stale/low urgency,2,stale/low urgency,1,0,153861,,,,0,ae52c64ea320d84b172d1b40f0e517c552e23c2967d000926c3c285fa696ef36
thread:issue:153863,"issue #153863: cmake: add USE_SYSTEM_{KLEIDI,CUDNN_FRONTEND,CUTLASS,FMT} options to USE_SYSTEM_LIBS",stale/low urgency,2,stale/low urgency,1,0,153863,,,,0,04830f99574992563ad819077ed04fe7ff699289f6a8a01e754c2df612156e5d
thread:issue:153919,issue #153919: FPE in `torch.remainder`,stale/low urgency,2,stale/low urgency,1,0,153919,,,,0,3e9b74f60c819ac575d89256b569ad0415419289c5e94f6b84f80d2998024ae7
thread:issue:153930,issue #153930: [FX] Support `torch.cond` in `torch.fx.symbolic_trace`,stale/low urgency,2,stale/low urgency,1,0,153930,,,,0,66a969959c77fc4ab490b3e5e705fc0c0d2bb5e1a7898bb83a283de0ce8119ba
thread:issue:153933,issue #153933: PyTorch build fail with GCC 15.1.0 due to third_party/gloo/gloo/types.h:66:11: error: ‘uint8_t’ does not name a type,stale/low urgency,2,stale/low urgency,1,0,153933,,,,0,4c363d59177c14cdb506ae32822165e7076842791a6042515e62154ba79816be
thread:issue:153957,issue #153957: MPS backend gradient correctness issues with large shapes,stale/low urgency,2,stale/low urgency,1,0,153957,,,,0,6d1a759aab3aa87728405448e55d4f1686cfc5c8054a65d7a4fc2d655735943c
thread:issue:153967,issue #153967: Memory Corruption in `torch.batch_norm_update_stats`,stale/low urgency,2,stale/low urgency,1,0,153967,,,,0,89817e3b13ed17ba8f33866c1a824c061bef21b021a9021982baac022561068f
thread:issue:153976,issue #153976: MPS (Metal) CTC Loss Implementation (Gist link),stale/low urgency,2,stale/low urgency,1,0,153976,,,,0,331766b561d4f736b80cd6084f86c62c4134961fbd51caa80e3f7cde0319cc62
thread:issue:154017,issue #154017: [RFC] Generalized Per-Operator Device Capability Registry for PyTorch Operator Testing,stale/low urgency,2,stale/low urgency,1,0,154017,,,,0,bb6228e65b77db58ec82d90997355c616bd07b83f11e4253f82ee2991d73db69
thread:issue:154028,issue #154028: Add clamped FP8 (E4M3) cast for overflow-safe inference,stale/low urgency,2,stale/low urgency,1,0,154028,,,,0,1cdca4c3ad9b2370cf6e09091efca18d8b66f61e8d982fd50ac4efe7eeeb3924
thread:issue:154074,issue #154074: UNSTABLE Build manywheel docker images for s390x / build-docker-cpu-s390x,stale/low urgency,2,stale/low urgency,1,0,154074,,,,0,e05d0552e5c0a8e971e20a9315ac6f718cfc012615b40b2ecbeb6a99aad48e08
thread:issue:154083,issue #154083: torch.compile joint trace materializes constant tensors,stale/low urgency,2,stale/low urgency,1,0,154083,,,,0,b342459e2b0fe3493e3af628e2e448cbda81ce340c50772a46cae75e7bc8d101
thread:issue:154086,issue #154086: [cuBLAS] Strange behavior with `out_dtype` argument in `torch.addmm`,stale/low urgency,2,stale/low urgency,1,0,154086,,,,0,9d7e297360baa3cf7328698a8468056d1a4aeee9977342822776eb96015b184f
thread:issue:154096,issue #154096: CMake Error (CMAKE_MINIMUM_REQUIRED) when build 2.7.0,stale/low urgency,2,stale/low urgency,1,0,154096,,,,0,d4482037142deed8109745691979a824a5b90c2bb65be0322dad05def488fb85
thread:issue:154128,issue #154128: matmul uses excessive memory in batch cases with more than 3 dimensions,stale/low urgency,2,stale/low urgency,1,0,154128,,,,0,74b6fce28b2ffebea0e244f66910b769651ea2e1ef43ffbc0b47512b58b88436
thread:issue:154187,issue #154187: pybind11 loading for c10::Scalar NYI,stale/low urgency,2,stale/low urgency,1,0,154187,,,,0,2498e4065af3d3c4132e02ee84327ae3560a30ba999708c5df17a5b609d52220
thread:issue:154214,issue #154214: AssertionError: Scalars are not equal! ```test_require_stride_expanded_dynamic_shapes_cuda```,stale/low urgency,2,stale/low urgency,1,0,154214,,,,0,d819677523629192a17cf4dfb66de923ec9728877aaae41bcce5a620220e334c
thread:issue:154226,issue #154226: NotImplementedError when computing JVP of Attention,stale/low urgency,2,stale/low urgency,1,0,154226,,,,0,b0ed74b3f5afaf18ee845900d393c72b00e851779426da91100994f2a90e53b4
thread:issue:154241,issue #154241: `RNNBase` modules break parameter sharing due to `flatten_parameters()`,stale/low urgency,2,stale/low urgency,1,0,154241,,,,0,73fecba7b8146ffbb5f34cee86907d25c598da28fa41e5359ddf980836238e7e
thread:issue:154248,issue #154248: [Testing][CUDA] test_cublas_addmm_reduced_precision_size_10000_backend_cublaslt_cuda_float16 failure with linux-focal-cuda12.6-py3-gcc11-slow-gradcheck,stale/low urgency,2,stale/low urgency,1,0,154248,,,,0,29e1df6f7efc367c6bb628dd2326bd7a0cdc3838134cde11d2fcc514ef52ea69
thread:issue:154272,"issue #154272: [FSDP2] for mixed precision, input casting can get blocked when cuda streams are full",stale/low urgency,2,stale/low urgency,1,0,154272,,,,0,e3a2c2a1642e6de9dba7b5a7f3748da76e40c46d832a223dd3b24df7defdaae5
thread:issue:15431,issue #15431: IDEEP error:could not initialize a memory descriptor,stale/low urgency,2,stale/low urgency,1,0,15431,,,,0,169ce5b4dd52ca50b65f71ef0eca8bb785db08ddc926b8be0c055332b3bf2217
thread:issue:154311,issue #154311: torch.inf.to(torch.int32) produces different values on CPU vs CUDA,stale/low urgency,2,stale/low urgency,1,0,154311,,,,0,62db1e42004380a776f90ce72a52aabbd02b0ea66c6746f5477b18a460b16717
thread:issue:15432,issue #15432: caffe2: multi-thread(or multi-instance) predict is much slower than the single thread or single instance,stale/low urgency,2,stale/low urgency,1,0,15432,,,,0,b6c2d731a5a523e8eab4ace52ca50b3807e81648de690bff322667c3599cd83b
thread:issue:154322,issue #154322: float16 → float32 conversion yields unexpected zero matrix for matrices > 43000 × 43000 in MPS,stale/low urgency,2,stale/low urgency,1,0,154322,,,,0,f15945bfff12cd2f3185aa370f91387a16892516cc737bc8b3e1f8478d4c1b86
thread:issue:154332,issue #154332: Support jvp for flex attention,stale/low urgency,2,stale/low urgency,1,0,154332,,,,0,86a603863d3856d28998e75ad435ed628dfc041477163f83c78811e7e0e7aa9e
thread:issue:154351,issue #154351: bfloat16 Conv2d slower than float16 on 4090,stale/low urgency,2,stale/low urgency,1,0,154351,,,,0,6a9da691e63825186e1d134078a54fc33126ab119d53108e207d484dd4313e8f
thread:issue:154419,issue #154419: Crash in `torch.sparse.softmax`,stale/low urgency,2,stale/low urgency,1,0,154419,,,,0,144875f77299166e3607e77394797e1b9ca3d49bca4f88903981ed73c08d0167
thread:issue:154428,issue #154428: Inconsistent sin(x) output between CPU and CUDA for very large arguments,stale/low urgency,2,stale/low urgency,1,0,154428,,,,0,93e4b3ac4160bd4bf0e443cfead130654649fbf114d8ed2e98c5e250e9ce972b
thread:issue:154456,issue #154456: torch.compile tracing doesn't seem to be using the cache,stale/low urgency,2,stale/low urgency,1,0,154456,,,,0,dc2c8944d96268d917b71f534eecd80867a43d314c96840f0e74a79d9e6eb591
thread:issue:154463,issue #154463: Mega Cache/ Torchcompile cache Not working,stale/low urgency,2,stale/low urgency,1,0,154463,,,,0,7d983eecbcae57e49a2af3fc4b08699465a86f5920dddf0cafe9fdc1d22d0ac6
thread:issue:154474,issue #154474: Multiple torch.fft APIs produce inconsistent results for infinite inputs between CPU and GPU,stale/low urgency,2,stale/low urgency,1,0,154474,,,,0,5ff22e73d3cbbb541b5bc3e12c0ec4e53cc289e0c141be752b39792ac5d65549
thread:issue:154489,issue #154489: [wishlist item] Assume data-dependent info based on tensor construction,stale/low urgency,2,stale/low urgency,1,0,154489,,,,0,28d36a1aa593677857241034e73fec511405e64781070a0045e511a57a72c4e2
thread:issue:154496,issue #154496: torch.fft.irfft(n) doesn’t handle non-Hermitian inputs in a consistent way,stale/low urgency,2,stale/low urgency,1,0,154496,,,,0,e8a3133856e9f14f5dd4c5579daae45e7e7d8bd42c4e06db2d749e3a2289d50a
thread:issue:154501,issue #154501: `torch.add` and `torch.sub` return incorrect results when computing complex numbers containing `inf` on both CPU and GPU,stale/low urgency,2,stale/low urgency,1,0,154501,,,,0,af7d214ba7716fd1f15fe8cb985d2823ed63f0096c34197f517fab881db9d67d
thread:issue:154557,issue #154557: grouped_mm optional zero initialization of the output,stale/low urgency,2,stale/low urgency,1,0,154557,,,,0,cf4256755adc1e2b905be0aa7b24223d7978f1cf564bc5f62f47cab3da667159
thread:issue:154560,issue #154560: Enable GB200 for dynamo/torchbench tests,stale/low urgency,2,stale/low urgency,1,0,154560,,,,0,1a19a2eeaaf7f80d2ef84af948f7147a9f387ed4971d742cdf77a9b597bed6d6
thread:issue:154602,issue #154602: Cudnn attention is very slow when sequence length changed in every step,stale/low urgency,2,stale/low urgency,1,0,154602,,,,0,8e869068a562dcad8340bd929fb5c767192cf8433be171bfebdc3a209a5b5033
thread:issue:154603,"issue #154603: RuntimeError: devptr INTERNAL ASSERT FAILED at ""/build/source/c10/cuda/CUDACachingAllocator.cpp"":3696, please report a bug to PyTorch. entry in cache has missing shared_ptr",stale/low urgency,2,stale/low urgency,1,0,154603,,,,0,6dde6159d232e0956341f3911fd8c2e7f6cb53c143ec50a921577b7bfeed6d9f
thread:issue:154606,issue #154606: `torch.cumprod` for complex128 produces inconsistent results between CPU and CUDA,stale/low urgency,2,stale/low urgency,1,0,154606,,,,0,1985f54470799fad1c1019d81d204062d131dab33ddda1e7ab2efcd8605ea983
thread:issue:15465,issue #15465: Transplant caffe2 on my DNN-accelerator,stale/low urgency,2,stale/low urgency,1,0,15465,,,,0,b4303b1df433909f30a7172b84d38c8b1bfb2a40307a0791cc97b9f4a20017fa
thread:issue:154791,issue #154791: [discussion] Specialized frontend method for computing Gram / covariance matrix,stale/low urgency,2,stale/low urgency,1,0,154791,,,,0,ab4884f0a5f031024602f311035b261719f14c4bbd05d707075e7b3dc07a082b
thread:issue:154810,issue #154810: Multi-dimensional tensors in datasets might get incorrectly flattened when fetching data from dataloader which is specified 'batch_sampler' when created,stale/low urgency,2,stale/low urgency,1,0,154810,,,,0,617ef83efa55e807b4822a44c15f99e529fc4d169ff0c600574dd632843baf8f
thread:issue:154836,issue #154836: [FSDP2] fix unit test test_all_gather_extension_outer_size_stride,stale/low urgency,2,stale/low urgency,1,0,154836,,,,0,7586bf11d92400acea790b5fb1bcc16acfb5e027f13417608fe13ea7d020f7ae
thread:issue:154842,issue #154842: `torch.linalg.solve` does not raise an error for singular matrix on CPU.,stale/low urgency,2,stale/low urgency,1,0,154842,,,,0,a081dbfa010316cd7d95d987ff736ac3a5b55681adbb493fefc9718766db3cfa
thread:issue:154847,issue #154847: don't require recompiles when switching between torch.Tensor vs AsyncCollectiveTensor graph inputs,stale/low urgency,2,stale/low urgency,1,0,154847,,,,0,846f701cc09fdcb4f79f4868148b0968c31ad1774fe12cedca79cef22e4a9b1b
thread:issue:154855,"issue #154855: test_tensorboard.py::TestTensorBoardSummary::test_histogram_auto Fails with s390x-periodic / linux-manylinux-2_28-py3-cpu-s390x / test (default, 7, 10, linux.s390x)",stale/low urgency,2,stale/low urgency,1,0,154855,,,,0,7a2c1917feb32a9265740a7cab173ccbf48284d8efc329b63f6d863250101936
thread:issue:15491,issue #15491: LibTorch: include cmake files for all distributed headers,stale/low urgency,2,stale/low urgency,1,0,15491,,,,0,a68fd1815e11eee25c9cf359103ef4e932ecd0f272654926ef57307f95603da9
thread:issue:154960,issue #154960: [FSDP2] all_gather_copy_in for cpu offload,stale/low urgency,2,stale/low urgency,1,0,154960,,,,0,7a643d31fea3b6d8573ddb59d5c69b758af92f3fa6c8921d73c46084d6388e44
thread:issue:155001,issue #155001: The profiler does not seem to be able to record cuda runtime nodes,stale/low urgency,2,stale/low urgency,1,0,155001,,,,0,359e49558141a6c2d7e2a74733c73bd6e5f539b28ec9b92393744c2ee85574ce
thread:issue:155011,issue #155011: Please support libtorch for XPU,stale/low urgency,2,stale/low urgency,1,0,155011,,,,0,0f64d10a9275acfc48ef7392100a674a0639634bfee4df10c51682de5c3217df
thread:issue:155065,issue #155065: Flex Attention and Nested Tensor: very high VRAM usage,stale/low urgency,2,stale/low urgency,1,0,155065,,,,0,fdd83f617ea231a5b7bcefec373654244be11feef0857fc659cbe75860b1df28
thread:issue:155105,"issue #155105: [dynamo && compiled autograd] compiled autograd recompiles when using custom autograd functions and compiling the forward with backend ""eager""",stale/low urgency,2,stale/low urgency,1,0,155105,,,,0,13bf369c54c38767dd96ae8a81bb626d11555fcc189db3ab92dc9676fc9a95fd
thread:issue:155157,issue #155157: Multi-worker dataloader for `IterableDataset` holds open process on macOS Python 3.12,stale/low urgency,2,stale/low urgency,1,0,155157,,,,0,227566be05310136a9f31802ada7c671da946b6178dbd3c98d6f8bbe24498da2
thread:issue:155193,issue #155193: Should be ReLU6(Module),stale/low urgency,2,stale/low urgency,1,0,155193,,,,0,94422a65016efd6cd817fc92bde513e60b0c9c67216700c3ea04bd7a53443ec3
thread:issue:155203,issue #155203: support for cuDNN 9.8+,stale/low urgency,2,stale/low urgency,1,0,155203,,,,0,97308c2daeeca1267b4ffd0cd2dd2dc8d50c80a22f723bc525e812c272b6781c
thread:issue:155230,issue #155230: [scan] scan is broken in nightly,stale/low urgency,2,stale/low urgency,1,0,155230,,,,0,6fda9761d6c17ca83145eec77a72771cfe1e2bec92d750a8c9e871c848970657
thread:issue:155266,issue #155266: torch.compile with InvokeAI: 'UserDefinedObjectVariable' object has no attribute 'proxy' / compute_ancestors KeyError op4,stale/low urgency,2,stale/low urgency,1,0,155266,,,,0,aabb5777048a2c82197044a2ca1ee9b341f7fc3bd4b48b07b4cec4d647a82eec
thread:issue:155332,issue #155332: compiler cache not work,stale/low urgency,2,stale/low urgency,1,0,155332,,,,0,c2efd60c17ac8b5dcd7c62bfc03b9c20e7799c04d8da5e97edd4531f560bcd67
thread:issue:155333,issue #155333: [A100][cusparselt] Non-determinstic correctness issue with some algorithms,stale/low urgency,2,stale/low urgency,1,0,155333,,,,0,ae36bbb508745f890aeca93fb871947b046b61c6183eae36e1e7ee14c7d5d322
thread:issue:155350,issue #155350: CUDA_HOME doesn't seem to work with setup script,stale/low urgency,2,stale/low urgency,1,0,155350,,,,0,f0fa66f0ef4c2bc3b689aaf45a70717134316a91e27c1071eea256831a1392e3
thread:issue:155378,issue #155378: RuntimeError: Could not find libnvrtc.so. Please make sure CUDA is installed.,stale/low urgency,2,stale/low urgency,1,0,155378,,,,0,6e8726b83e7624323a620f91fd5ac69e6e3ac8ef2761a12934b37ed3deee6e22
thread:issue:155386,issue #155386: OpOverloads should have annotations,stale/low urgency,2,stale/low urgency,1,0,155386,,,,0,f245c1e5bf9491297e0fb39be1af8b16f478b3caf2dcec824aea9ef993dad8ca
thread:issue:155472,issue #155472: [RFC][c10d] Make it easier to act on group,stale/low urgency,2,stale/low urgency,1,0,155472,,,,0,37508932a80e166dc9a27220f86d9a1af4aa9c59eef4efc3b3d47ed7a7236e67
thread:issue:155519,issue #155519: [Tracking] Triton 3.2 deprecation,stale/low urgency,2,stale/low urgency,1,0,155519,,,,0,1f38773d829e43ed7441b4e1cd2fab010c77b11e42caef70c0f433c081b26bbc
thread:issue:155552,issue #155552: Reproducibility of results without AVX512 by setting ATEN_CPU_CAPABILITY=avx2,stale/low urgency,2,stale/low urgency,1,0,155552,,,,0,4462a56c26f5bc02e9226346330d29694deec4719a81f1a356bf4c3a52a9b9b7
thread:issue:155616,issue #155616: Additional Documentation About Size Checking with Custom Operators,stale/low urgency,2,stale/low urgency,1,0,155616,,,,0,89ae6b3cc022a48e845de07300038de131fb216c935265340da4003e447366af
thread:issue:155667,issue #155667: [libtorch] Debug and Release should be shipped together on Windows,stale/low urgency,2,stale/low urgency,1,0,155667,,,,0,b1334df33b59b5a9bd9cedbbe94f6d69bb496150e19f1c380dc9f1a0cb58b2b1
thread:issue:155671,issue #155671: torch.clamp throws overflow error on CPU but not on CUDA,stale/low urgency,2,stale/low urgency,1,0,155671,,,,0,6ced6b1b116e16fd5422514eacfba5241cb8222687ed030d3497f5fc01ba5e99
thread:issue:155680,issue #155680: [FSDP] optimizer_state_dict fails if FSDP model is not on global rank 0,stale/low urgency,2,stale/low urgency,1,0,155680,,,,0,e5aa2953aa1c58e0e2eae93f5aa207769bd2cf1ee5f86d01c30dd80b41ca8542
thread:issue:155703,issue #155703: [pt2] [precompile] Support cudagraphs in BundledAOTAutogradCacheArtifact,stale/low urgency,2,stale/low urgency,1,0,155703,,,,0,57ec06b92ce256d7d0ba57aed285db074275a0fb33caf94b3179bb4204376422
thread:issue:155798,"issue #155798: `nn.RNN(...).to('cuda')` fails with `cuDNN error: CUDNN_STATUS_BAD_PARAM` on GPU, but works on CPU",stale/low urgency,2,stale/low urgency,1,0,155798,,,,0,73677bc94a3225f2e1a922450608bcf48390f9e68abb8b9223cca273620d0e66
thread:issue:155825,issue #155825: Running dispatch modes on compile-disabled regions of a compiled model,stale/low urgency,2,stale/low urgency,1,0,155825,,,,0,1afeb3a7b8ff1f039109e04570bcc7ce42c852bd6e89cc98790ca6f9e95d5600
thread:issue:155838,issue #155838: intmm is being compiled inconsistently with errors.,stale/low urgency,2,stale/low urgency,1,0,155838,,,,0,e9a287df54fa0e4d9cc5a3aae788b694306ad0c495d7224c2d7578e0a6c019bf
thread:issue:155997,"issue #155997: ONNX export via Dynamo sets `dft_length = 1` in `DFT`, breaking shape-inference for `torch.fft.rfft`",stale/low urgency,2,stale/low urgency,1,0,155997,,,,0,5557c93fdf0654c6531a5714c6d556cec16d73588ac47308766d6e5c714de7e9
thread:issue:156020,issue #156020: torch.fft.ifft for complex64 produces inconsistent results between CPU and CUDA,stale/low urgency,2,stale/low urgency,1,0,156020,,,,0,fd8b2f2c79645215f736e79119cdfc3513afc7c1cad0a2e4f431352db70098d8
thread:issue:156024,issue #156024: Ошибка установки torch для CUDA 12.1 на GTX 1660 Ti,stale/low urgency,2,stale/low urgency,1,0,156024,,,,0,ccec3db16212fe817e85dec4bdf5be242e4814a28d33fdf8a2eab070ee223f26
thread:issue:156062,issue #156062: torch.nn.functional.conv_transpose3d return inconsistent results when weight containing inf between CPU and GPU,stale/low urgency,2,stale/low urgency,1,0,156062,,,,0,44ce0920a3157b4c0347d5881bcb6a8f48db23e6df0e4fb0a47315535692759d
thread:issue:15607,issue #15607: documentation for adding a new type via C++ extensions,stale/low urgency,2,stale/low urgency,1,0,15607,,,,0,0170fd69698bbe0f5cfa3e0cfc133abca7e0d1ca69ec7ba8169d83fcd07a15e8
thread:issue:156072,issue #156072: [typing][docs] `torch.amin` and `torch.amax` do not document `dim=None`,stale/low urgency,2,stale/low urgency,1,0,156072,,,,0,12ebe604c46d2f95c85ddbcdfd2ec2149c0136c9851a15b653f14bc16885cb2e
thread:issue:156073,issue #156073: [codespell] fix typos in the codebase,stale/low urgency,2,stale/low urgency,1,0,156073,,,,0,4d111404f766919450f17f19d9e92a1cb4e2413fa8e97f92295bbf528fc3f9c7
thread:issue:15608,issue #15608: [Caffe2] How to switch to test phase?,stale/low urgency,2,stale/low urgency,1,0,15608,,,,0,6fbe3b2aac3b464f3e9e8f4850e0abd8e17f8051545e393b0e7e3f6dd96ce8b8
thread:issue:156085,issue #156085: Idea: Add SBOM Generation (and optional vuln scan) for better supply chain insight,stale/low urgency,2,stale/low urgency,1,0,156085,,,,0,14f32d16e551ed37e0078ddcfe1fd42e729f309b47bf00d5034f58e697a5a74b
thread:issue:156177,issue #156177: nn.Module._load_from_state_dict is always called with strict=True,stale/low urgency,2,stale/low urgency,1,0,156177,,,,0,c3051d7222c0dee1ba572ab84cfec713328c474526e65c0eff9458cf7819e8e3
thread:issue:156238,issue #156238: Upgrade torch._scaled_grouped_mm to SM100+,stale/low urgency,2,stale/low urgency,1,0,156238,,,,0,d9d995f209bcaa0348a97543518db8b0ce605dafb88e09ffe09bf0bf7e06df77
thread:issue:156258,issue #156258: [ONNX] Create a tutorial for exporting hf transformers model,stale/low urgency,2,stale/low urgency,1,0,156258,,,,0,8bea967cf223430711bf89d8d5db8bedfbf7fd9478fa89d4acff0271b355f4f8
thread:issue:156411,issue #156411: [compile] torch._dynamo.exc.TorchRuntimeError: Failed running call_function aten.lift_fresh_copy.default,stale/low urgency,2,stale/low urgency,1,0,156411,,,,0,49289503b45bc79e2b7655e7c5928531632840a924146fe766f578e596dd6663
thread:issue:156434,issue #156434: Suggest to use the torch cmake target instead of ${TORCH_LIBRARIES} in the c++ docs,stale/low urgency,2,stale/low urgency,1,0,156434,,,,0,4799e97a8ffe12c6803cda58ba0029ee851eda11cc85e64c6a69df59516e475d
thread:issue:156444,issue #156444: Tensor.is_pinned() raises error after renaming privateuseone backend.,stale/low urgency,2,stale/low urgency,1,0,156444,,,,0,124906ecde3001c066df39323c694e259618aeb4ff622a4a6912b8592a5b1608
thread:issue:156520,issue #156520: [user empathy] compile for `transformers` model,stale/low urgency,2,stale/low urgency,1,0,156520,,,,0,106c16d54aa3c4fb10a7f79d7d3b4c93db31ebe54bdde7c4f91192aa071c7223
thread:issue:156607,issue #156607: [OSS tooling] pytorchbot fail to revert a PR,stale/low urgency,2,stale/low urgency,1,0,156607,,,,0,e514458dfdb4b6a84f713f571de4bd957c4146ae025408e247f5260da17b3472
thread:issue:156618,issue #156618: THPStorage_shareCuda zeroes CUDA tensor,stale/low urgency,2,stale/low urgency,1,0,156618,,,,0,7c4d360abccae4fe15d89470a07efcd327479d83b9fc1e7bfe28f928ce51d8ec
thread:issue:156622,issue #156622: Add a test to torch.library.opcheck for if a C++ meta kernel specializes on a symint,stale/low urgency,2,stale/low urgency,1,0,156622,,,,0,f77849c8eac029a9e8027a4aa45cad06430c9a7bf04a43278977c2a4466de6ab
thread:issue:156680,issue #156680: torch compile does not support SyncBatchNorm with fullgraph=True,stale/low urgency,2,stale/low urgency,1,0,156680,,,,0,cff4f7e2e2ac5bed93ecf234d8ac9924d213e8c8c72cf4802f8bcadd8b10ef54
thread:issue:156694,"issue #156694: Vendored wheels on PyTorch pip repository are outdated (e.g., `cmake`, `certifi`)",stale/low urgency,2,stale/low urgency,1,0,156694,,,,0,25df0dda4988e167d85081b4d00b54e2a6d131e4c9c988c75c0651bba3a841e2
thread:issue:156701,issue #156701: Add generator argument to torch.rand_like,stale/low urgency,2,stale/low urgency,1,0,156701,,,,0,dd1262381dc28c1840e62d659de3fcd7dac0ebd881377c6ad181b0761b4dfadc
thread:issue:156725,issue #156725: CMake improperly configures pybind11. 3 different versions of pybind11 in use at the sametime.,stale/low urgency,2,stale/low urgency,1,0,156725,,,,0,2762cd5c0a68d40322d0f34d0db898d75a488fb95f8295a4ea894a51f24face2
thread:issue:156733,issue #156733: FX graph printing do newlines for parameters by default,stale/low urgency,2,stale/low urgency,1,0,156733,,,,0,7fa3703cfa186056fbb1a8304c71c1c1938fb443f9c1a47582dd1a9cd4b794ef
thread:issue:156745,issue #156745: [v.2.8.0] Release Tracker,stale/low urgency,2,stale/low urgency,1,0,156745,,,,0,1583eb9d39aa31d7763c0766d8c1c7f946369341c4c146e7600bc30372eb5c10
thread:issue:156792,issue #156792: backprop fails in HF SegformerForSemanticSegmentation on MPS backend with torch > 2.4,stale/low urgency,2,stale/low urgency,1,0,156792,,,,0,9ad4ef18ebf5b5c2d34da7f1db5996199cb03406e246342e5b8ebeb450230b5f
thread:issue:156810,issue #156810: Add the XPU item to pytorch.org/get-started,stale/low urgency,2,stale/low urgency,1,0,156810,,,,0,7d9d98e8e7d4bbdc4cbfefa2d11de5369a6acd068e06c5205cdc3ceea7ac6e9b
thread:issue:156811,issue #156811: Allow to set a seed for random tensors in torch.onnx.export,stale/low urgency,2,stale/low urgency,1,0,156811,,,,0,595c023bc482691e295e432acfd0f5be199a6cb5543a6b0a39979dfada032115
thread:issue:156820,issue #156820: put compile-based cudagraph re-recordings in tlparse,stale/low urgency,2,stale/low urgency,1,0,156820,,,,0,32e80dc531af1cc9dc246fad7ec7f389ab7261c9eca118a3c75f965522583186
thread:issue:156846,issue #156846: Can't alter the learning rate though loading optimizer states,stale/low urgency,2,stale/low urgency,1,0,156846,,,,0,2180e6aa4c9af6b5950604c28b848c613624c46be13dfe0ffc435ffa2d905c56
thread:issue:156873,issue #156873: [ONNX] Missing repeat_interleave for dynamo=True,stale/low urgency,2,stale/low urgency,1,0,156873,,,,0,d4a043a9d35143d0b6c7f02b0df26d1c0462b07e8ad17cfc7916f90934c3c410
thread:issue:156956,issue #156956: Allow clamp to work on absolute values and preserve sign,stale/low urgency,2,stale/low urgency,1,0,156956,,,,0,2b77df1c2bc86242faca4d8dd4c58be486d35a369690d9c3951d7212fc13bcb8
thread:issue:156959,issue #156959: Gross mismatch in PDF between CUDA and CPU for multivariate Gaussian mixture models,stale/low urgency,2,stale/low urgency,1,0,156959,,,,0,42460a451c36d2675244b357b6b006252438db1bcd39805c6d1f4a5629e1007b
thread:issue:157015,issue #157015: [WIP][RFC] Compilable flex_attention + Context Parallel,stale/low urgency,2,stale/low urgency,1,0,157015,,,,0,b5a6a75c87afe74d2b12873a3ee0bc1dbc65c9af9c1e60e745a8da835801465c
thread:issue:157054,issue #157054: NVFp4 Cublas Error,stale/low urgency,2,stale/low urgency,1,0,157054,,,,0,c3cd46ac1c0b692a744575f36790c63793c5f5cc108648830c26ae4cbf56bd76
thread:issue:157075,issue #157075: Improve debug message for metadata guard failure,stale/low urgency,2,stale/low urgency,1,0,157075,,,,0,d845a7c290cde16665976063385d0471a0ff6c27f84720779eb8feddfd8e5899
thread:issue:157101,issue #157101: Documentation update: Contribution.md,stale/low urgency,2,stale/low urgency,1,0,157101,,,,0,98b4da67886fe34d4aa14b3c59885282f2904a8121cd46ea8bd5bb89f155f7bf
thread:issue:157139,issue #157139: Distribute ARM Images,stale/low urgency,2,stale/low urgency,1,0,157139,,,,0,b37cfe3c186bc5859b7dec78801bc55eae533b8b28a025838792dba63a66240d
thread:issue:15716,"issue #15716: Nn.dataparallel with multiple output, weird gradient result None",stale/low urgency,2,stale/low urgency,1,0,15716,,,,0,ec52ffdaae880d51eecd8f38310b025f7c1f6980de10f5d8841291ae12ef43cd
thread:issue:157218,issue #157218: [Feature] Taylor expansion pruning,stale/low urgency,2,stale/low urgency,1,0,157218,,,,0,b85138a6546f8f6fc717ca0d82f13be4db656b12cb0fb0d07d1760c16b83970f
thread:issue:157220,issue #157220: [ONNX] torch.nn.functional.interpolate \w antialias=True isn't op.Resize compatible,stale/low urgency,2,stale/low urgency,1,0,157220,,,,0,8af3f8262809e367ec6be6820da8d8fa6761d88ef4083b5a34dc19b0b26a84f0
thread:issue:157266,"issue #157266: Better typechecking of `int` only-operators `|`, `^`, `&`, `<<`, `>>`, `~` and `@`",stale/low urgency,2,stale/low urgency,1,0,157266,,,,0,309dc706250ca626a59778f6e42d800f0302e1a1473c1e4c1d0b20a48b711291
thread:issue:157272,issue #157272: `torch.reciprocal` and `torch.divide` for Complex `inf` Incorrectly Returns `NaN` Only for Tensors with >= 4 Elements on CPU,stale/low urgency,2,stale/low urgency,1,0,157272,,,,0,95d43e69386b8829b7054e5e0bccc53a60efbda8e0069f3f9a55dfbb28d0b372
thread:issue:157321,issue #157321: Symmetric memory test failed with TORCH_SYMMMEM=NVSHMEM,stale/low urgency,2,stale/low urgency,1,0,157321,,,,0,0350c643cb5f92d3f5e458b6aa8eecc8fff0172e7f6a185c8374ccb764d4cb3c
thread:issue:157334,issue #157334: Inefficient 2D convolution compared to JAX,stale/low urgency,2,stale/low urgency,1,0,157334,,,,0,6c060f6af925999790b8c371e7752badb187a943461c4c02d1a7252bd27ff60b
thread:issue:157379,issue #157379: [CI] Need better way to detect OOMs especially on pet instances,stale/low urgency,2,stale/low urgency,1,0,157379,,,,0,d13e78afa860a67aae8f527d6c396b9d88c2520fb0e115884bd52f971372d9ad
thread:issue:15738,issue #15738: Beta Distribution values wrong for a=b---> 0,stale/low urgency,2,stale/low urgency,1,0,15738,,,,0,cba80c2191721970c6e5304b3d7816682716e39aabbb5ae40cffd9705eabcde5
thread:issue:157391,issue #157391: [FSDP2] document the contract for modifying DTensor model.parameters(),stale/low urgency,2,stale/low urgency,1,0,157391,,,,0,7a89a46c8a24ec425cfb3f98f8044bacc8eaa2e0039c61b5eeb5defd7fd00ed8
thread:issue:157445,issue #157445: DDP+TP composition does not work as expected,stale/low urgency,2,stale/low urgency,1,0,157445,,,,0,98d12c0b20697cf507fb52a899b6a7bc4744fb24226f9f3ec87ada4521a0f7c5
thread:issue:157465,issue #157465: Nightly NCCL builds are missing optional features from NCCL,stale/low urgency,2,stale/low urgency,1,0,157465,,,,0,ca0f3e24e0e0232db39fe3c856d8449134bd594ee772c5626d6a7d7ed64d053c
thread:issue:157502,issue #157502: Failure with cub::TransformInputIterator in 12.9 periodic CI test,stale/low urgency,2,stale/low urgency,1,0,157502,,,,0,db1cc465d10844fc51f3deba5a4ccd09898a9fd3733345c696d30dd1fd8cb9c2
thread:issue:157545,issue #157545: test_dtensor.py::test_dtensor_save_load_import conflicts with autoloader importing torch._dynamo,stale/low urgency,2,stale/low urgency,1,0,157545,,,,0,0a0c35507ac5b7e95f65f15e2faeb7744566f5cddcb341f8fcb43b471a268996
thread:issue:157548,issue #157548: Nightly cu128 aarch64 wheels haven't been built for weeks,stale/low urgency,2,stale/low urgency,1,0,157548,,,,0,8db170eee73d213a5bfc0805eb10a73863bc974cb7ed0b1d2b7d641b6db05f73
thread:issue:157591,issue #157591: PT2E Quantization Migration Tracker,stale/low urgency,2,stale/low urgency,1,0,157591,,,,0,91136f389014e0130d1669319adea7df1be967a915b0c2b07aceeb130c7635e8
thread:issue:157606,issue #157606: Both DTensor TP and SP are missing the last collective in the backward pass,stale/low urgency,2,stale/low urgency,1,0,157606,,,,0,be6f83fcabe6bf907e35e19f2262ff84f2da8a9a2bf2bf54dbef305f9954165f
thread:issue:157624,issue #157624: file_name is not correctly read in here,stale/low urgency,2,stale/low urgency,1,0,157624,,,,0,6ac8c72fd10d5e50ae755f59e2bef8566fc4c25341127c8a0c99814a0ac42037
thread:issue:157627,issue #157627: Regression: torch.distributed.gather_object segfaults,stale/low urgency,2,stale/low urgency,1,0,157627,,,,0,0f4241bf8fa19edcc49790eee3350987dd447902cac108f4a5a9ab30cd4e57cc
thread:issue:157662,issue #157662: OffsetBasedRNGTracker's run_state_sync causes deadlock due to inconsistent broadcast order across ranks,stale/low urgency,2,stale/low urgency,1,0,157662,,,,0,2caf42d1d3dd4a70b369688b91c430d61b0213ceb3cac474ca1e8bb367a1c32e
thread:issue:15771,issue #15771: Implicit conversion error in caffe2,stale/low urgency,2,stale/low urgency,1,0,15771,,,,0,cddf9fc53feadb6b773810463a62d01a8b7977ea3cdca041a73a67eb5d32bbc9
thread:issue:157714,issue #157714: AssertHandler::printMessage On Intel GPU,stale/low urgency,2,stale/low urgency,1,0,157714,,,,0,48bf979d5372416ec1018ff476a38a6a106b1a78ec794ce8a4a29c972569ac00
thread:issue:157721,issue #157721: [Precompile] Caching precompile and presence of packages in general should be safe re: guard serialization,stale/low urgency,2,stale/low urgency,1,0,157721,,,,0,2602d0fe8a19b2cbc828c9d18fa3a4eb5f723afb0f1954c2406620beefd0e893
thread:issue:157764,issue #157764: Drop SSE4 support in oneDNN,stale/low urgency,2,stale/low urgency,1,0,157764,,,,0,d382623ddf18e027afb69b7a2bca241e42cab156bed96bc179068adc3d7ce478
thread:issue:157808,issue #157808: sparse_csr bfloat16 matrix multiplication backward is 10x slower than float16,stale/low urgency,2,stale/low urgency,1,0,157808,,,,0,cfd7f131f6f2b2bb1914afa30474c52196e9151a7fb856ffe7af7d58a7c4378d
thread:issue:157918,issue #157918: [Precompile] Umbrella task,stale/low urgency,2,stale/low urgency,1,0,157918,,,,0,1a372eaaea4784a45fa9213f7e0687746b9d30b36af5f7a1d5cc1e29d08972e9
thread:issue:157919,issue #157919: tighten pt2 FX graph invariants,stale/low urgency,2,stale/low urgency,1,0,157919,,,,0,e09f1093412985c1dc1b0a9c572eacd59a692919a101be4c9a4aeca1e99cbef4
thread:issue:157950,issue #157950: [RFC]: PyTorch Low-Precision GEMMs Public API,stale/low urgency,2,stale/low urgency,1,0,157950,,,,0,60c3d839dadaa481567c09d10408965288497a8606336cbd96910d0c179ef5da
thread:issue:157976,issue #157976: support DTensor's local_map in compile,stale/low urgency,2,stale/low urgency,1,0,157976,,,,0,0beb72d42dd876104bea767458a112b200df4451226797abdd8eb3c0833d38e0
thread:issue:158041,issue #158041: pow_test fails on Aarch64,stale/low urgency,2,stale/low urgency,1,0,158041,,,,0,2f8033478d4365cf8202f214e05257c684b32066ca5aebc31e14cbbcaec7da57
thread:issue:158080,issue #158080: loaded storage won't be cached during deserialization if location is meta,stale/low urgency,2,stale/low urgency,1,0,158080,,,,0,bd29c1af5cb08286c3dbb83f20da8ca85a1b73994095e7762e891824336dec92
thread:issue:15812,issue #15812: Possible regression in incremental build,stale/low urgency,2,stale/low urgency,1,0,15812,,,,0,075a24547ed35b0a0edcca9f7a8453385af8674cb8d79b3cb518f8c97653e4bf
thread:issue:158125,issue #158125: index.Tensor doesn't properly account for boolean masks acting on multiple dimensions.,stale/low urgency,2,stale/low urgency,1,0,158125,,,,0,6078352a1e43c6d11cce83d19349cfa428cc2adeeea342c1c97c19f9405da3db
thread:issue:158127,"issue #158127: A ""Merge-when-ready"" label when ci finishes?",stale/low urgency,2,stale/low urgency,1,0,158127,,,,0,821502303e762380fc9b863e931fad57b26ec0ba4d8ceb8559b269f11e2c0065
thread:issue:158169,issue #158169: Invalid `torch.linalg.lstsq` check during vmapped implementation,stale/low urgency,2,stale/low urgency,1,0,158169,,,,0,8b008bb1f56f86d99c665d5a7a8f9de4c74a92d768759f8a052229a5cfa7eefe
thread:issue:158172,issue #158172: `torch.fmin` has inconsistent overflow behavior on CPU and GPU,stale/low urgency,2,stale/low urgency,1,0,158172,,,,0,0b6603ff7e65d61fe3d59b6c80768999adf0f21167fde6f532d574523e641805
thread:issue:158187,"issue #158187: [torch.save] Integrate FastPersist by allowing f (a.k.a., file-like object) to save storage objects",stale/low urgency,2,stale/low urgency,1,0,158187,,,,0,2a74316e60601cc34c56b40d90fe9aeffc3e73327f5bd939a5841aa80d64b273
thread:issue:158199,issue #158199: GPT-2 compiled model perf/accuracy,stale/low urgency,2,stale/low urgency,1,0,158199,,,,0,cbbe4259f3f1f9c37dcbaceb99cb434668843f380fd6bceaccaa989a16e6d205
thread:issue:15821,issue #15821: No test coverage for kwargs of AvgPool2d and AvgPool3d,stale/low urgency,2,stale/low urgency,1,0,15821,,,,0,df77002f5dafac324889de22c51f98607d62642a422a118c0b8e17bffecd1e6d
thread:issue:15822,issue #15822: AvgPool2d doesn't test if kernel is smaller than input size,stale/low urgency,2,stale/low urgency,1,0,15822,,,,0,d6eadae4703306f5d94264f45d15192ca84382eef33fa7c1470076203af09a39
thread:issue:158240,issue #158240: [Windows][inductor] Windows compilers building was slower 4x than on Linux.,stale/low urgency,2,stale/low urgency,1,0,158240,,,,0,3e05960de525b99ee4db75acd420382e6db79a3d6af5f0624eec0c20a1d14723
thread:issue:158244,issue #158244: [QST] [DTensor] MoE + EP support,stale/low urgency,2,stale/low urgency,1,0,158244,,,,0,9052610ea2437b35e55606ed9320b483a729a34ac49c8961872e9f5ff2257631
thread:issue:15829,issue #15829: computing entropy of a tensor,stale/low urgency,2,stale/low urgency,1,0,15829,,,,0,2a6b2443cdb9c33a90ffc5758007f215d0c10bd840cfdd36754808d01a5584cd
thread:issue:158326,issue #158326: Support for etcd v3 api as rendezvous backend,stale/low urgency,2,stale/low urgency,1,0,158326,,,,0,3b06b32cce807dec029d6ebda226fd928c0169c5ad62af7daeb84555ac86c9f7
thread:issue:158364,issue #158364: [feature request] [memory_viz] Flexible time/range memory event filtering controls to skip loading all memory events from pickle in the browser's memory,stale/low urgency,2,stale/low urgency,1,0,158364,,,,0,727085954558e475edbe954956f915bebbc93b17c2891d29780bf1cc2193c33f
thread:issue:158401,issue #158401: torch.compile hits internal tuple index out of range inside subclass_meta.unwrapped_idx,stale/low urgency,2,stale/low urgency,1,0,158401,,,,0,71b027890628ed6fef14455628e3bb9b0ac3a465b9436a1281d0460acf32609a
thread:issue:158412,issue #158412: Inconsistent torch.absolute results on complex128 between CPU and CUDA,stale/low urgency,2,stale/low urgency,1,0,158412,,,,0,576739823bc3663feacb67a460d1457ed51dba41641c81be659244a9396f0a2d
thread:issue:158419,issue #158419: torch.copysign for float16 produces inconsistent results between CPU and CUDA,stale/low urgency,2,stale/low urgency,1,0,158419,,,,0,e09bcbcc0592f7420dc8e075144791c38b7164caf1a03f39dd6fb35ee2b3bf98
thread:issue:158428,issue #158428: Inconsistent torch.cholesky_inverse results on complex128 between CPU and CUDA,stale/low urgency,2,stale/low urgency,1,0,158428,,,,0,a7675b723e56334a19898070538c9e2a889668ed2511e56e7487c16e9771fb74
thread:issue:158444,issue #158444: Proposal to introduce CUDA streams in BatchLinearAlgebraLib.cpp,stale/low urgency,2,stale/low urgency,1,0,158444,,,,0,12c9bfc178ea18faf18ece8607991c17baca196837392ef448bc070ab7079e70
thread:issue:158566,issue #158566: torch.compile fails with UncapturedHigherOrderOpError when using torch.ops.higher_order.scan,stale/low urgency,2,stale/low urgency,1,0,158566,,,,0,101139553a956ae1a5fca2560fe2c37ccea2a955d44067490a3876df32e59e0e
thread:issue:15864,issue #15864: Deformable Convolution,stale/low urgency,2,stale/low urgency,1,0,15864,,,,0,2119990a8e7fb3897ff4079c0724e19fa4a41590e7504eb6def1db51fe926ec3
thread:issue:158641,issue #158641: support checkPoolLiveAllocations for cudaMallocAsync,stale/low urgency,2,stale/low urgency,1,0,158641,,,,0,d1ef8143a3ae1a4b54461f3abbe7d8ecf786b9fcabcdfac240098b16671c81c5
thread:issue:158664,issue #158664: DISABLED test_sum_all_cpu_float64 (__main__.TestReductionsCPU),stale/low urgency,2,stale/low urgency,1,0,158664,,,,0,558b8411b95003b2b317b7626d48bb42b913903610efbc3721881c77ac0119b9
thread:issue:15869,issue #15869: RuntimeError: [enforce fail at pybind_state.cc:1111] success. Error running net train,stale/low urgency,2,stale/low urgency,1,0,15869,,,,0,2a7794249914b87917bef69de67bc34a55e88eb9dace32edf0f95330a2ca2b75
thread:issue:158699,issue #158699: reports of torch.compile'd rmsnorm numerics different from eager,stale/low urgency,2,stale/low urgency,1,0,158699,,,,0,e842aca8126637ebc7216b702a81b5e0a02596c51a222e37e319513508f01399
thread:issue:158738,issue #158738: searchsorted handling of nan,stale/low urgency,2,stale/low urgency,1,0,158738,,,,0,40cac66c2a1abb770d516c8e5478e4134ab42e12f255beac75ad305d8eeb22b0
thread:issue:158782,issue #158782: [BE] linter for base docker image configurations with docker_builds.yml,stale/low urgency,2,stale/low urgency,1,0,158782,,,,0,52cf12144dac6b10b5845513eeaca52472bddbecef6abc1eedfa465706d90c4e
thread:issue:158783,issue #158783: [BE] linter to detect unused docker images,stale/low urgency,2,stale/low urgency,1,0,158783,,,,0,bc15d95ded3525ca79b65ed2132066bd7c1cdd03ab8b5f11185b49cacbbc0e52
thread:issue:158785,issue #158785: [lint] Set_linter suggests wrong changes,stale/low urgency,2,stale/low urgency,1,0,158785,,,,0,62e1b7c00c043763041ecd73194852d991b86bb9293c3054465a91ed24478fb9
thread:issue:158819,issue #158819: [FSDP2] return original unsharded nn.Module from FSDPModule,stale/low urgency,2,stale/low urgency,1,0,158819,,,,0,46bee035b4c8a919cae673fe893e36da3e15b67c1c498befbcdc34783e3ee7c4
thread:issue:158831,issue #158831: Severe Performance Degradation in Batched SVD on A100-40G,stale/low urgency,2,stale/low urgency,1,0,158831,,,,0,d28389b63724d37aba3cf64a811f0ef16a04f27568e5ec6c4e2fe0a806ada1af
thread:issue:158832,issue #158832: Lazy Layer Norm,stale/low urgency,2,stale/low urgency,1,0,158832,,,,0,fe3df6734dfb1b02b237034715795b93157b6445838800ec5cb53c4b6f30f7ba
thread:issue:158871,issue #158871: torch.classes module doesn't set all module attributes,stale/low urgency,2,stale/low urgency,1,0,158871,,,,0,de9566bd3057db110accd91a12f547aa34f6e3cd3b9340299bbcd95e0ebd52bb
thread:issue:158873,issue #158873: Erroneous Recompile Warnings with Multiple Models,stale/low urgency,2,stale/low urgency,1,0,158873,,,,0,5eb24f3f2faa2c1b477c7ff216f0b6e3f2326eeb734f2c73ec2b62eddd0c0700
thread:issue:158876,issue #158876: UNSTABLE pull / linux-jammy-py3_9-clang9-xla / test (xla),stale/low urgency,2,stale/low urgency,1,0,158876,,,,0,42ec608f164649b28e7ccc5ad0a6b693c86102f129dae626b1ad2bd02d0edc88
thread:issue:158895,issue #158895: Update docs about running lint,stale/low urgency,2,stale/low urgency,1,0,158895,,,,0,5916b397ec8922921a8b99607375ce6d3aae7151a9254642b53d4ac6edc36047
thread:issue:158936,issue #158936: TorchAO FP8 training + HF transformer model + torch.compile + DDP does not work,stale/low urgency,2,stale/low urgency,1,0,158936,,,,0,379503c77812faefcc3ed87fe6c6948ccfb69ebfb533e5256c01ddd47da4f8ba
thread:issue:158969,issue #158969: Scaled dot product attention accepts head_num for attn_mask,stale/low urgency,2,stale/low urgency,1,0,158969,,,,0,69529a8bb9b57726aa3f105bacfdf51b345fed132d03db7b9c8531f325058822
thread:issue:1591,issue #1591: Pad PackedSequences to original batch length,stale/low urgency,2,stale/low urgency,1,0,1591,,,,0,d9eabe3c5bafa141dbcd5163dc081da3d4f01aff8b79a3f0fee5c3180d28768a
thread:issue:159122,issue #159122: CI lint unexpected failure,stale/low urgency,2,stale/low urgency,1,0,159122,,,,0,2f53d2d90a6553f3ef215100b019a8575180bea6447bcc6d62dbdf8133ef519f
thread:issue:159138,issue #159138: Some bugs that related to `ReflectionPadNd` on input dim and padding,stale/low urgency,2,stale/low urgency,1,0,159138,,,,0,f91f206d59dffa669de8b63f49c79c8ba3fb6b24566930bce31ee401da8fd43a
thread:issue:159149,"issue #159149: ❌ Error generating speech: NVML_SUCCESS == r INTERNAL ASSERT FAILED at ""../c10/cuda/CUDACachingAllocator.cpp"":995, please report a bug to PyTorch.",stale/low urgency,2,stale/low urgency,1,0,159149,,,,0,3f9d294b05ab0d642dbb42adf8e18d53a4b91f0c9b0db0059bf4415c54bf0a49
thread:issue:159156,issue #159156: Fully remove the protoc dependency in core,stale/low urgency,2,stale/low urgency,1,0,159156,,,,0,098f495971a5e3742b108aee56f0a0b4b6db63beac0b10f173ab8c49e84d1d5e
thread:issue:159164,issue #159164: Improvements to CUDACachingAllocator When expandable_segments Is Enabled,stale/low urgency,2,stale/low urgency,1,0,159164,,,,0,6a8a14ad89c87bce2ef157f4dc778d9424d4c69a897344467b2f78fce05ae501
thread:issue:15918,issue #15918: Broadcasting and Additional Dimensions for pairwise_distance,stale/low urgency,2,stale/low urgency,1,0,15918,,,,0,92c455b1d430fefbcd1c5eef557f394dbba5224970da71fe068ee37b2abf9e9a
thread:issue:159226,issue #159226: Add support sdpa kernels on gfx1101 (AMD RX7800XT or W7700),stale/low urgency,2,stale/low urgency,1,0,159226,,,,0,a9241a60ee3fb94d8912da7530909106960f19703c1a6ae809572151bbcb1212
thread:issue:159228,issue #159228: [Precompile] mark_static_address behavior,stale/low urgency,2,stale/low urgency,1,0,159228,,,,0,847495386367c4dcb4abbc2a232e7845196543400b6b868301d392ed525fb070
thread:issue:159232,issue #159232: Support installing Python bindings in CMake,stale/low urgency,2,stale/low urgency,1,0,159232,,,,0,3f5160f063a8d1a47a320c56449464bed0af0a599678b6d08cacdfa1a4c2d95d
thread:issue:159234,issue #159234: Lingering reference to setup.py in CONTRIBUTING.md,stale/low urgency,2,stale/low urgency,1,0,159234,,,,0,2e3ce620b752453e53d295bdd77761426c6407b599c69075bd756f9d0f795602
thread:issue:159301,issue #159301: #158878 seems to break `test_mempool_empty_cache`,stale/low urgency,2,stale/low urgency,1,0,159301,,,,0,132d6b6c39cf765dbd6c2687f8c2ee76a31278d99e0815574302c885aea6f0f2
thread:issue:159333,issue #159333: torch._foreach_copy_() copy all nan as same number,stale/low urgency,2,stale/low urgency,1,0,159333,,,,0,68a661f47a788ad04e747b0e61840af86845d84ee7a5e54ba800f4920d4e5306
thread:issue:159363,issue #159363: pad_sequence overflows when padding with larger value than int64.max - 512,stale/low urgency,2,stale/low urgency,1,0,159363,,,,0,e6dbe6b5a149dfa31447c33b887935c062a952c05ad51b1c84ae3827b7a24398
thread:issue:159384,issue #159384: Graph partition performance,stale/low urgency,2,stale/low urgency,1,0,159384,,,,0,18393eacc0ec78902cc6e0d2ba09703c5f60eaf103266b04c9b9dad57e11d24f
thread:issue:159498,issue #159498: [FSDP2] provide profiler metadata for memory bandwidth calculation,stale/low urgency,2,stale/low urgency,1,0,159498,,,,0,92e21cf4ad134fc9c6684fa87000a1c79242de40adbd7549178af1e9cf3bcfc8
thread:issue:159510,issue #159510: Add more tests to h100 ciflow,stale/low urgency,2,stale/low urgency,1,0,159510,,,,0,59d0727d85465e0d6211028e136b4fa8161902212709862c4a7f4113b2aaf84e
thread:issue:159536,issue #159536: Document NCCL Symmetric memory,stale/low urgency,2,stale/low urgency,1,0,159536,,,,0,edabe89c0f5bd86f87feacdde097bc1c80865dac680bdfa21541d94ddcb5c8d5
thread:issue:159559,issue #159559: Support NCCL register userbuffer for P2P communication in NCCL backend,stale/low urgency,2,stale/low urgency,1,0,159559,,,,0,661e4cbe5b4cf2fb16b410ee2da26150cfa0cf52eefabb40ff705f05b62179aa
thread:issue:159564,issue #159564: Re-enable inactive_split_bytes when expandable_segments:True,stale/low urgency,2,stale/low urgency,1,0,159564,,,,0,6a166ce6186734d65284ddd442e50e31b3ee08d4649bd4ee70f901aa4b7bbe04
thread:issue:159622,issue #159622: Performance bug on `inplace` of `nn.ELU`,stale/low urgency,2,stale/low urgency,1,0,159622,,,,0,6a803a492439a8baa682c24d6cf2f8a032120fd425d43cb86dd2ca333ec488f7
thread:issue:159667,"issue #159667: Non-tensor auxiliary types don't work in torch.func.{grad, value_and_grad}",stale/low urgency,2,stale/low urgency,1,0,159667,,,,0,a6f2e95bcaacc10b40b0eb7c80eae0c6875639d47ac6414f27a54c1329fac9ef
thread:issue:159674,issue #159674: OOM when using use_mem_pool due to restriction on retry,stale/low urgency,2,stale/low urgency,1,0,159674,,,,0,eaced352257e836c983fe527df500d11b8c0165bbaa4d2a41df35651be5a4008
thread:issue:159676,issue #159676: [Benchmark Dashboard] Track quarterly change,stale/low urgency,2,stale/low urgency,1,0,159676,,,,0,8a749c1881f8027d370d8d2108a945a3e477a5d4cc2a449ea6e8d9722d3d9323
thread:issue:159680,issue #159680: [Benchmark Dashboard] Kernel-level benchmark,stale/low urgency,2,stale/low urgency,1,0,159680,,,,0,4bfc1b351771d30ac312f0546f8f836181d6e445deb8800d53d3371f804458d9
thread:issue:159739,issue #159739: torch.empty should consider np.bool while parsing args,stale/low urgency,2,stale/low urgency,1,0,159739,,,,0,8c4946f10293f6878924d7c85be830aa73ee61b0cda7fc45acc19eaad8ad8a89
thread:issue:159740,issue #159740: Better error message when requesting pinned memory on cpu-only system,stale/low urgency,2,stale/low urgency,1,0,159740,,,,0,0aef7448038c5c1a2c11a4e9d23c50c4baf468390398e0f1f5fdb8bc0ad725d3
thread:issue:159769,issue #159769: Inductor codegen for float8 dynamic quantization ops for scaled_grouped_mm backward pass is slow,stale/low urgency,2,stale/low urgency,1,0,159769,,,,0,1e435e4bf97e6a576ee847aa66916bb12cc24046ddec90cd2219d42bb21b377e
thread:issue:159781,issue #159781: MacOS Python 3.13t wheel is build for macOS 14.0+ ARM64 while Py3.9-3.13 wheels are for macOS 11.0+ ARM64,stale/low urgency,2,stale/low urgency,1,0,159781,,,,0,d5bca4a39a8d310c633db0c020129d3bc35c9ad68d2ffeb129d4afa342e1a459
thread:issue:159784,issue #159784: Immutable assignment (akin to `array.at[...]` in JAX),stale/low urgency,2,stale/low urgency,1,0,159784,,,,0,e77eb00b93121845f836da4245a0636ec4b902f915dc111ee5639940465c526a
thread:issue:159838,issue #159838: C++23 forward declaration error :`invalid application of 'sizeof' to an incomplete type 'torch::_export::Graph'`,stale/low urgency,2,stale/low urgency,1,0,159838,,,,0,4261590f1fc06517e99451def6a2876535ef3c4355e58632eb1e2380f23d698d
thread:issue:159848,"issue #159848: `tree_flatten(some_pytree, lambda x: isinstance(x, Tensor))` yields `[None], _`",stale/low urgency,2,stale/low urgency,1,0,159848,,,,0,bac463f0cdf40fdbc48081a6afcee51a5f59484946475f7e75dd40de8a0c2365
thread:issue:159866,issue #159866: BE: enable auto reformat List to list and similar typing,stale/low urgency,2,stale/low urgency,1,0,159866,,,,0,0736a50cf09a9c7a7e690e93b211a0313cbf2a85299953e033176b1213860332
thread:issue:159870,issue #159870: `torch.nn.functional.sigmoid` produces inconsistent results on different device for complex inputs,stale/low urgency,2,stale/low urgency,1,0,159870,,,,0,4a5669e252d77685921b95c23b8ccb80267a5d088b7a4090b5753f4806d34191
thread:issue:159886,issue #159886: torch._dynamo documentation,stale/low urgency,2,stale/low urgency,1,0,159886,,,,0,7c022baa7f18d781b9d71730968f786684a7b00a9230895bf0936b336f2417e2
thread:issue:159914,issue #159914: `test_non_contiguous_input_mm_plus_mm` looks broken on B200,stale/low urgency,2,stale/low urgency,1,0,159914,,,,0,a4bacecdefe0cd612fe64872a30904419d154e9e6d4db8b497d59c66ccd361dd
thread:issue:159932,issue #159932: Cannot view dtype of tensor which is contiguous due to numel=1 but has non-standard striding,stale/low urgency,2,stale/low urgency,1,0,159932,,,,0,01fd0c700a76bd7f14fdf231bf124a9b142690abaf133c575c9164ee747bfa6f
thread:issue:15994,issue #15994: More data type support for gather_map,stale/low urgency,2,stale/low urgency,1,0,15994,,,,0,30632c387f79733f058bbd3e148f8f99b284ae91ff2cc17b7bf2a919ae3d53d6
thread:issue:159942,issue #159942: Is is possible for pytorch pipeline parallelism to support dynamic shapes(input/output),stale/low urgency,2,stale/low urgency,1,0,159942,,,,0,8eac400622c94af7b57475ac661e947bdb59ae014fee46fe0c3c92b0c97741a7
thread:issue:159947,issue #159947: Add official statically linked libtorch libraries,stale/low urgency,2,stale/low urgency,1,0,159947,,,,0,9d63a732e5c9030db0962e75251051b1e8d1c10cc76ec624004957e3c4735ab2
thread:issue:160071,issue #160071: [TorchDynamo] excessive stack use: stack is 3286 deep on A100 with AMD EPYC CPU,stale/low urgency,2,stale/low urgency,1,0,160071,,,,0,777f874850a6a3935cc99ad52188029f3db6509c0775649a1f4f4377ff329e86
thread:issue:160122,issue #160122: Add tensor-aware put_signal & wait_signal Triton wrappers to NVSHMEM Triton Kernels,stale/low urgency,2,stale/low urgency,1,0,160122,,,,0,2aabc1f6e46b4ba41a62557ea13819c6df23c4289ac70363a1bed7d8f455391f
thread:issue:16014,issue #16014: [docs] Missing argument description (value) in scatter_ function documentation,stale/low urgency,2,stale/low urgency,1,0,16014,,,,0,35d755482cbea53e19927002e4977593bad86f5f015defb8c67803c76cd97d28
thread:issue:160142,"issue #160142: Add a ""max-performance"" mode to torch.compile for aggressive optimization",stale/low urgency,2,stale/low urgency,1,0,160142,,,,0,e7f2d9c1c40c4bf528aee54db272f4037dafe7d12a952d0ae32c69fa2f5269f0
thread:issue:160167,"issue #160167: libtorch: CMake 4.1: Error evaluating generator expression: TARGET_PROPERTY:torch,INTERFACE_LINK_LIBRARIES",stale/low urgency,2,stale/low urgency,1,0,160167,,,,0,fa418c91f5a27b2adb9b5759c0be7774a225bfe02254aa2bee1a4492a0536d1b
thread:issue:160225,issue #160225: Building cuda kernels with debug information causes some cuda kernel launches to fail due to capacity constraints,stale/low urgency,2,stale/low urgency,1,0,160225,,,,0,fd6b3769042f63464999cb2b0522804785ddc671a91c82659018655b86216a86
thread:issue:160240,issue #160240: Migrate test.sh into python based tests,stale/low urgency,2,stale/low urgency,1,0,160240,,,,0,f23e06f54508e5d5fa9d98760302a9e29f749d244342246c1b4e12d05dc21cc6
thread:issue:160248,"issue #160248: PyTorch 2.8: gloo-cuda linking issue, undefined symbol",stale/low urgency,2,stale/low urgency,1,0,160248,,,,0,484583941ecf97ec51cf42401e1d124ef8fed6259f92c346d4f31e892d9e4f11
thread:issue:160281,issue #160281: Improve CUDAGraph Tree heuristics on starting new generation,stale/low urgency,2,stale/low urgency,1,0,160281,,,,0,4dd466792340bae11634f2e7ec093dea3f67702c1250a2c57dd26f87dde6c810
thread:issue:160308,issue #160308: Reserved memory is much bigger than allocated memory in multi-stream scenario,stale/low urgency,2,stale/low urgency,1,0,160308,,,,0,bee87d61c6ac1f20b57729a1f4a28a08c3b73209db5a0b411130b81dadde28a0
thread:issue:160337,"issue #160337: [RFC] Simplify bookkeeping of DeviceMesh slicing, (un)flattening, ...",stale/low urgency,2,stale/low urgency,1,0,160337,,,,0,3421fa666cec78040fc804d9aab4a40de413df563a4d3888bb42edcb740115b1
thread:issue:160359,issue #160359: Use `tf32x3` through Inductor,stale/low urgency,2,stale/low urgency,1,0,160359,,,,0,d84d61173ea36e109e82960d3af6f2f9b4799460e9f4105ae9feaf853ed39224
thread:issue:160387,issue #160387: DISABLED RecordDebugHandles.Basic (__main__.test_jit),stale/low urgency,2,stale/low urgency,1,0,160387,,,,0,3bdfd6784593aec699fdc2b2917967341e21c090a5a649d5600dd7dbff469d61
thread:issue:160405,issue #160405: [Expandable block] how to get the best-fit free block,stale/low urgency,2,stale/low urgency,1,0,160405,,,,0,d0943e1a75d9faf8a153f4f7f99f4ac49321e78a9a389e665ae8c498ff012a6c
thread:issue:160415,issue #160415: [MPS] Conv3d type mismatch with AMP,stale/low urgency,2,stale/low urgency,1,0,160415,,,,0,0f5184c8948034d69fb9edaca4479709f138f3a1f4469c710fe68c719a7188e0
thread:issue:160438,issue #160438: torch.autocast with no_grad() on aliased model disables gradients for subsequent forward on same instance,stale/low urgency,2,stale/low urgency,1,0,160438,,,,0,e9824a8e4955e7d8456f644d4a2bbebce2c006c72dcdb5b4dfaa37cf8a0451d7
thread:issue:160517,issue #160517: Enhance output layouts of all-sparse matrix multiplications,stale/low urgency,2,stale/low urgency,1,0,160517,,,,0,4ec291c01630720859866d7beb3673564aeacc8b7448877f176c8220d0b65299
thread:issue:160629,issue #160629: torch.arange not implemented for complex numbers,stale/low urgency,2,stale/low urgency,1,0,160629,,,,0,31555cc80d235b1666e006095044baa553618b1679b6aa492af84aba5d5cba84
thread:issue:160651,issue #160651: `dequantize_per_channel` in `quantized_decomposed` not compatible with Float8 types,stale/low urgency,2,stale/low urgency,1,0,160651,,,,0,42e2eb9cc854da919166231089487ccb45ed216df48de6371fe1fe6e802de91d
thread:issue:160787,issue #160787: [Intel Arc XPU] NaN Values in Gradients after SDPA with is_causal,stale/low urgency,2,stale/low urgency,1,0,160787,,,,0,298ea6a1baf598ba8d81564b59667392580306a6a8c7d6a7a65aa574f8669ca1
thread:issue:160816,issue #160816: torch._scaled_mm crash with cublas error when out_dtype is not specified,stale/low urgency,2,stale/low urgency,1,0,160816,,,,0,777290a452541808c0a166decdeab76888d51260a0711b56eb85cd790364ac5c
thread:issue:160838,issue #160838: backward on 5090(sm_120) much slower than on 4090,stale/low urgency,2,stale/low urgency,1,0,160838,,,,0,7bb478a81ac8037cdc095ae2b2b07ff9809ace883547fa206777b593fd909436
thread:issue:160849,issue #160849: Add new shuffle mechanism for SGD/Adam --- CorgiPile without full data shuffle,stale/low urgency,2,stale/low urgency,1,0,160849,,,,0,e4ba6d8b530d0527d32885a515a01139e019a8e241ec42d5fb62cab8b8c87058
thread:issue:160856,issue #160856: Discrepancy between PyTorch and TensorRT when casting from FP32 to BF16 causes significant accuracy mismatch,stale/low urgency,2,stale/low urgency,1,0,160856,,,,0,d17713d001308d7416c9449e548a9890739128ebdecce7e16225d3bd4520dafc
thread:issue:160891,issue #160891: Please add `torch._grouped_mm` on SM_120 (RTX 6000 Blackwell) hardware,stale/low urgency,2,stale/low urgency,1,0,160891,,,,0,9c0df4b88d1cc8455cd91dd6b3a0117ae6093622527890c30010437f692a87b5
thread:issue:16090,issue #16090: build failure with NNAPI enabled (Caffe2 path but under pytorch umbrella),stale/low urgency,2,stale/low urgency,1,0,16090,,,,0,9fc24a52abae485432b753ec1feb521c1d24ad72e7eb2449b4883865a29b678d
thread:issue:160937,issue #160937: Module produced by export.unflatten is not runnable for Qwen3 and Llama3,stale/low urgency,2,stale/low urgency,1,0,160937,,,,0,813d88c8d4894d75f8596598b501388944065869405cd1f5e307192777002d60
thread:issue:161001,issue #161001: Add some way to associate functional custom operators and their in-place out variants,stale/low urgency,2,stale/low urgency,1,0,161001,,,,0,25c767866af9f3e82aa7c5ec5288302337035f88d9bd38eca55acb6a9fbecfa6
thread:issue:161029,issue #161029: MPS does not support deterministic index operations,stale/low urgency,2,stale/low urgency,1,0,161029,,,,0,d717b87230ec5aa5e43c66ff63ca023a5096d5fd7d3424d02a36976fed5225d2
thread:issue:161068,issue #161068: [ATen] Linkage error with TensorBase::data_ptr<T> and TensorBase::mutable_data_ptr<T>,stale/low urgency,2,stale/low urgency,1,0,161068,,,,0,201a8087f49fc2f5d2ec4c6d810ed1045601d1c92cc36cc5d303c2a2d7c371f4
thread:issue:161105,issue #161105: UED guard check failure,stale/low urgency,2,stale/low urgency,1,0,161105,,,,0,985abaa3d5437982f4f5a8df96c099ff69b535380b81de83bf4719f32ea9118f
thread:issue:161108,issue #161108: AoTInductor gives segmentation fault with torchbind registration,stale/low urgency,2,stale/low urgency,1,0,161108,,,,0,10236030b316f0ff37ab325725fe3b8ab29ddea724cf504eddc05e33cf11fa3d
thread:issue:161163,issue #161163: Dropout performance degradation after version 2.6.0,stale/low urgency,2,stale/low urgency,1,0,161163,,,,0,31eb132281f39863723fd2ee615cdcac2544115fad1712ad55d680a4699ba6a2
thread:issue:161281,issue #161281: There should be an easy way in CI to disable individual failing inductor-periodic tests,stale/low urgency,2,stale/low urgency,1,0,161281,,,,0,ac01b35b4366403f59eba23b44bd533dbef047b508518d5b4aaaf978fb7c38c1
thread:issue:161283,issue #161283: Implement half-vs-float option for MPS output match tests,stale/low urgency,2,stale/low urgency,1,0,161283,,,,0,f482d0884f3d2d5d1a06c7c198f537cd2ade397b8172084d0bf43aa2e01256f8
thread:issue:161287,issue #161287: Add partial shape inference support when creating a `view()` of a Nested Jagged Tensor,stale/low urgency,2,stale/low urgency,1,0,161287,,,,0,72ce78dcf2623409c01c993327a28f0627fa4464dc0cdab0fbca77131db9e3e8
thread:issue:161296,issue #161296: Make `torch.func.functional_call` compatible with params and buffer input as `tensordict.TensorDict`,stale/low urgency,2,stale/low urgency,1,0,161296,,,,0,a6857d753bfedee6399980fce2d3a4b0359703bcd8374d4049845ff66302bc4d
thread:issue:161330,issue #161330: better tool to repro issues in 1 dynamo graph,stale/low urgency,2,stale/low urgency,1,0,161330,,,,0,92abcbffb9d4b35278e4fb5ff53c4e15ac0f60e31e86b0316446b685286b5f96
thread:issue:161357,issue #161357: `reshape` support for sparse tensors (needed for `einsum`),stale/low urgency,2,stale/low urgency,1,0,161357,,,,0,37b184b741189d5ded87c4695a0d0a67602b7c2401155a04ac81b6f844415f20
thread:issue:161365,issue #161365: orch.onnx.errors.UnsupportedOperatorError: Exporting the operator 'aten::linalg_inv' to ONNX opset version 16 is not supported.,stale/low urgency,2,stale/low urgency,1,0,161365,,,,0,9c3bb9a9d722367cb51e5aef655ff3139d16bd2a57851f2bad121557cd030994
thread:issue:16138,issue #16138: [Caffe2] resize5d stride/dim issue,stale/low urgency,2,stale/low urgency,1,0,16138,,,,0,3cc7b5c0e408fe49df2fd08c318b54d3504a09353026a7ed5f98db8d2baaac8a
thread:issue:16148,issue #16148: CategoricalCrossEntropy Loss runs with wrong tag,stale/low urgency,2,stale/low urgency,1,0,16148,,,,0,f8e47a1e7b1b76d07d97161639af8f0761750dd35d21b7a22a79fd53ef8e562a
thread:issue:161500,issue #161500: nn.TransformerEncoder NaNs with float mask and float src_key_padding_mask only when in both eval() and torch.no_grad(),stale/low urgency,2,stale/low urgency,1,0,161500,,,,0,5f92cf5977ff5549ba2875e4b2bda8cd4ddd914a6163b8d34c509a5d71a3d812
thread:issue:161509,issue #161509: Update install commands in c++ Debugging,stale/low urgency,2,stale/low urgency,1,0,161509,,,,0,3d6a99002d171e30fb347df34ef46456c145a0c0d8d4f2fae706ceccfb8c87a8
thread:issue:161524,issue #161524: boolean index_add: Discrepancy between CPU and MPS backend,stale/low urgency,2,stale/low urgency,1,0,161524,,,,0,b82d57ae47f581602eb5a095efbf574c67ea2ca6bc8252130a5cfcb44de24d4a
thread:issue:161621,issue #161621: evaluation forward() of Transformer model is slower than training forward() while autocast,stale/low urgency,2,stale/low urgency,1,0,161621,,,,0,20b328cd7cbb72f857e7bd69fd4354f92a2e9abbe723f184ea9fb561615e745a
thread:issue:161778,issue #161778: RuntimeError: ptrDesc->finalize() with conv1d,stale/low urgency,2,stale/low urgency,1,0,161778,,,,0,9b7936f1c19850d1b88a9096364d97740cfa41f2267516e384a8b81b99fdaf95
thread:issue:161853,issue #161853: torch.nn.FractionalMaxPool3d occurs bug with big kernel_size,stale/low urgency,2,stale/low urgency,1,0,161853,,,,0,294832e5d42c9074b80afda517dfcc1a08ed577d977efebcc710f83db8f36e49
thread:issue:161869,issue #161869: Add LazyGroupNorm,stale/low urgency,2,stale/low urgency,1,0,161869,,,,0,afd6016c055ac7f227f3559d23f27eb1fc0ff3ebf7c9a7e079a7c311a406f202
thread:issue:16187,issue #16187: Sparse matrix multiplication is too slow,stale/low urgency,2,stale/low urgency,1,0,16187,,,,0,05b752f77284cd0eff75ddc37244e565b215e0bad7b1749f65826c0b188ca232
thread:issue:161886,issue #161886: Directly monkey patching torch.Tensor.__getitem__ breaks tensor construction,stale/low urgency,2,stale/low urgency,1,0,161886,,,,0,094ccf6070d83522a02762f03dcedf35c1438da4f07ea5678d0629f1353c7988
thread:issue:161937,issue #161937: AOTAutograd runtime alias reconstruction applies view on non‑contiguous base under torch.compile (inductor + DDP): “Use .reshape(...) instead”. pre_grad_custom_pass and joint_custom_post_pass can’t override,stale/low urgency,2,stale/low urgency,1,0,161937,,,,0,b3f2a54b29089f3751f9be517db12a7b441430a7482d20e7ec3a377e9247db50
thread:issue:16195,issue #16195: how to implement crf predict for segmentation task in Caffe2?,stale/low urgency,2,stale/low urgency,1,0,16195,,,,0,b2c4b03a38397f00aaaa61ea56b8520cd977683742cc1faa482b337d9178aeaf
thread:issue:161954,issue #161954: [RFC] CUDA code decoupling and directory refactoring,stale/low urgency,2,stale/low urgency,1,0,161954,,,,0,f98f605c9ad57e2b4123288ec34e9f190e63ef23e891e05d90b9306e2d5560c1
thread:issue:161960,issue #161960: WeightsUnpickler error: Trying to call reduce for unrecognized function <class 'code'>,stale/low urgency,2,stale/low urgency,1,0,161960,,,,0,97a0a62b7cb957a3db20b114a1f7c8a9ffcbcb4cee58eee1d723286f38e9512f
thread:issue:161970,issue #161970: Confusion behavior of PyTorch when combining boolean indexing with integer indexing,stale/low urgency,2,stale/low urgency,1,0,161970,,,,0,d084a5f2f20bb221c745020e95b130006fdb6860884788764afe6196f493294d
thread:issue:162004,issue #162004: UBSAN appears to be active for ATen but not c10,stale/low urgency,2,stale/low urgency,1,0,162004,,,,0,90fad46becddbf709261991aac6ac551da485756d69b25a8008f5280c2017039
thread:issue:162127,issue #162127: [CUDA] Need support for Cuda 13.0,stale/low urgency,2,stale/low urgency,1,0,162127,,,,0,c8bb242083d29fedfea19acbf0d4dd3e449bc58dd784635858c94b8a3a6307e4
thread:issue:162154,issue #162154: Linear algebra operations hang indefinitely doing nothing when run in subprocesses,stale/low urgency,2,stale/low urgency,1,0,162154,,,,0,5b4116f391755293a1b1ecf47580f9b2e3c50096cc19b00fe6d3379c8fd85058
thread:issue:162182,issue #162182: CI jobs for x86_64 and x86_zen (AMD CPU) are overwriting each other,stale/low urgency,2,stale/low urgency,1,0,162182,,,,0,527c4214658a0ac270dede0246ba0d3fa7cebd74bd0ae9361fa6318a4c443ae1
thread:issue:162198,issue #162198: [Feature Request][Inductor] Support ops.store in ops.masked,stale/low urgency,2,stale/low urgency,1,0,162198,,,,0,b7752d49552dafdefabef073485529db8ae1ff8afcaeab706b2b4ad63170e22c
thread:issue:162259,issue #162259: [Triton] Support multi-GPU with different types of GPUs (4090s + 5090s on same machine),stale/low urgency,2,stale/low urgency,1,0,162259,,,,0,497283a3a283a38512a73909a2ec4b5c6334b27e728d85a5a889674767ba6f7e
thread:issue:162280,issue #162280: libcufile symbol mismatch error in CUDA 13.0 libtorch build,stale/low urgency,2,stale/low urgency,1,0,162280,,,,0,61f97355082714aff0d1107fb4027b824c6f2ddf607fe5d918b3e797bd76b0c3
thread:issue:162283,"issue #162283: Failure after building torch with ""pip install -e . -v --no-build-isolation""",stale/low urgency,2,stale/low urgency,1,0,162283,,,,0,79fc517b1bb1a070faa840ea74b52861a168c9640703d370a8825a91d4bbb7ef
thread:issue:162346,issue #162346: The batching rule of aten::affine_grid_generator is not implemented,stale/low urgency,2,stale/low urgency,1,0,162346,,,,0,bf8dd612a6fe293670e46af72055e42adde78fbf3e39d00b2f891cd67043f16b
thread:issue:162352,issue #162352: Hiding C/C++ symbols in `version_script.lds`,stale/low urgency,2,stale/low urgency,1,0,162352,,,,0,a29e927cedde5eaca115f336d07ab13a60e83f85c7b1caefaecb76abc3ab99db
thread:issue:162362,issue #162362: the result is wrong for sdpa with sliding window attention,stale/low urgency,2,stale/low urgency,1,0,162362,,,,0,90829257eebdeda01b0965a3a22c9728237f35fc1198e40394558993c1011367
thread:issue:162375,issue #162375: `torch.compile` fails with `pack_padded_sequence` + `pad_packed_sequence` (FakeTensor error),stale/low urgency,2,stale/low urgency,1,0,162375,,,,0,9098ee1ac349620b2fdb23fe9d7da1783a374479779dc75d82e886e55a54f4b1
thread:issue:162398,issue #162398: Unexpected variable type when parsing YAML in torchgen,stale/low urgency,2,stale/low urgency,1,0,162398,,,,0,85affff8ff8ac2a98e43b6b0c786ccc8986bb9e8059e259d40c83e9e35dbb21d
thread:issue:162402,issue #162402: Issue with torch.compile - sm110a(Nvidia Jetson Thor),stale/low urgency,2,stale/low urgency,1,0,162402,,,,0,ed4ec480f7692cc9f7a8acf5d3b0078717d7a40e94569c247120595287305ce8
thread:issue:162449,issue #162449: optimum-cli export onnx for model intfloat/multilingual-e5-large-instruct fails with sufficient accuracy,stale/low urgency,2,stale/low urgency,1,0,162449,,,,0,7ae04201f3182388a48123d90d545dbd9e59c733dfe1c20f14f7228fef8eb28b
thread:issue:162458,issue #162458: torch.float8_e5m2 unique() functionality support,stale/low urgency,2,stale/low urgency,1,0,162458,,,,0,6ae80c802b61cbda9ae989065967ca3f560600265c884388e5e1975046a265c8
thread:issue:162482,issue #162482: [sparse] `torch._sparse_broadcast_to` creates false-positive coalesced tensors with inputs of shape `()`,stale/low urgency,2,stale/low urgency,1,0,162482,,,,0,483d27522480f86aa0156289ce3a79881cdf7502263333d465290ec373267092
thread:issue:162489,issue #162489: BatchNorm1d exhibits NaN outputs on GPU due to extreme input values causing numerical instability.,stale/low urgency,2,stale/low urgency,1,0,162489,,,,0,a55e0eb62827da8b30ea43926d50664859dd3817c7a825877b99e799bf37e048
thread:issue:16254,issue #16254: I hope Caffe2's python interface add DataLoader Module Similar to pytorch's DataSet and DataLoader,stale/low urgency,2,stale/low urgency,1,0,16254,,,,0,1419727034d20d353e941b477900d0c36143cb5607b41f7c973c586092ca3b27
thread:issue:16257,issue #16257: why check ArgumentInfo is_pod? suffer bugs,stale/low urgency,2,stale/low urgency,1,0,16257,,,,0,a5dcd017d4ede9bb3f05a490493c719bd570d5b1ccf7224f7ede45a16a67d2b4
thread:issue:162582,issue #162582: calling _record_memory_history(enabled=None) before _dump_snapshot would clear the timestamp information,stale/low urgency,2,stale/low urgency,1,0,162582,,,,0,1c179956b67119d8a7ce982e42ec6599d1883b247e09844e278aa37f7745ee2a
thread:issue:162592,issue #162592: Multi-Batching in F.scaled_dot_product_attention does not work.,stale/low urgency,2,stale/low urgency,1,0,162592,,,,0,6c5b441e9c6877b096791bd53b7b8508acc54e563c480c0429e9b5375a9f393b
thread:issue:162606,issue #162606: Tensorpipe - ROCm support,stale/low urgency,2,stale/low urgency,1,0,162606,,,,0,abce22ce4eae73a0559485055a68f201164bdd2d26470cb5ee5a1f441a65d11a
thread:issue:16266,issue #16266: support for multiple torch.cuda.max_memory_allocated() counters,stale/low urgency,2,stale/low urgency,1,0,16266,,,,0,e595d17ea5ad3579a791a45c0d33064b5d2cecd434e14d8b7fe178cbd8cf1d7a
thread:issue:162674,"issue #162674: NotImplementedError: ""addmm_sparse_cuda"" not implemented for 'Half'",stale/low urgency,2,stale/low urgency,1,0,162674,,,,0,abd20daece70067af912b0abccde9056848a0404397328626fec035d17d64e49
thread:issue:16268,issue #16268: Complete dtype support for torch.norm,stale/low urgency,2,stale/low urgency,1,0,16268,,,,0,b2491f65e94225834769fa4af647ab1545b2f5b43a6c1aec225751437ef18409
thread:issue:162685,issue #162685: Prevent py::class_ with no pointer type from being used on structs that inherit from intrusive_ptr_target,stale/low urgency,2,stale/low urgency,1,0,162685,,,,0,efde54e288790233df81ad7a46faddfe9a6b77433a14414ddc687bea51ccbbcd
thread:issue:162711,issue #162711: `scatter` / `gather` index type unclear,stale/low urgency,2,stale/low urgency,1,0,162711,,,,0,261d669eda91311409bb12f641635bd3db2f39e93f84598e32143c30dca4733a
thread:issue:162712,issue #162712: FPE in torch.range,stale/low urgency,2,stale/low urgency,1,0,162712,,,,0,0bcd84d6698d1dd5b44aae37bcf0d3af059d82802872a4c549ab451e1bb2b72d
thread:issue:162719,issue #162719: linalg.eig does not get parallelized on CPU,stale/low urgency,2,stale/low urgency,1,0,162719,,,,0,2ddc0ecc450eae58e89fa6c6a9439e62d86b1376e7eacc6d173060ea40acde04
thread:issue:162782,issue #162782: Is `torch.nn.functional.gumbel_softmax` going to be deprecated?,stale/low urgency,2,stale/low urgency,1,0,162782,,,,0,89e7399fc221210a81676e3976d98800ddf2be7f8d1c3ebccb26718fd030b1ef
thread:issue:162786,issue #162786: [MSVC] LINK : fatal error LNK1181: cannot open input file 'sleef\src\libm\CMakeFiles\mkrename.dir\mkrename.c.obj',stale/low urgency,2,stale/low urgency,1,0,162786,,,,0,ab7d19cd03aec819a2d7e7cf276cb6bea0ccde64b7ae271b83fedaa2159a22ad
thread:issue:162790,issue #162790: torch.tensor and torch.as_tensor flatten (concatenates) list of 1d tensors instead of nesting them,stale/low urgency,2,stale/low urgency,1,0,162790,,,,0,d8a2518e35c682109318244e8ca19e29671ae6a48bbc69a1125d5a2ff1231698
thread:issue:16280,issue #16280: caffe2: cudaHostRegister() and mbind()-related test failures on ppc64le,stale/low urgency,2,stale/low urgency,1,0,16280,,,,0,353ce21ca74907501f873a62d446eca96eb32524e305ade3110e16b196f570c5
thread:issue:162837,issue #162837: XPU backend is falling back to run on the CPU.,stale/low urgency,2,stale/low urgency,1,0,162837,,,,0,784784601b3528f02a426695fb9d61c55074ac522f533f4dd5d28346467f52c8
thread:issue:162870,issue #162870: [RFC] library function with 64+ arguments,stale/low urgency,2,stale/low urgency,1,0,162870,,,,0,63a52031a0e50163893e3a872133f2648ef5180dc5325d38b9f6958cf7bdd525
thread:issue:162907,issue #162907: Poor bf16 precision in backward torch.asin,stale/low urgency,2,stale/low urgency,1,0,162907,,,,0,dc90f00d40904ce3407b452e09cf46c6d23abb85ed3b5b24128e78c83cab294a
thread:issue:16291,issue #16291: Confusing documentation with distributions.Categorical about logits,stale/low urgency,2,stale/low urgency,1,0,16291,,,,0,bab35afe3912c730a77efccde9f3bd018d36a9fd8d13e538cdb5a202ce84d189
thread:issue:162911,issue #162911: `torch._foreach_...` functions error on zero-length inputs,stale/low urgency,2,stale/low urgency,1,0,162911,,,,0,3012cdcca2ed43ba61187353873b28e84152fcdf79317c22b7b5b14f349215ce
thread:issue:162964,issue #162964: [feat] OpInfo-like concept for subclass edge conditions to handle,stale/low urgency,2,stale/low urgency,1,0,162964,,,,0,b8349a8f0c845edebe1f5611ddec97c9aed9eaecb090c1c8bfb2d14bfd0cf4d6
thread:issue:162971,issue #162971: [CD] Reasonable time constraint for binary builds,stale/low urgency,2,stale/low urgency,1,0,162971,,,,0,fbb9333d60ca01b6ebe1cbf025102023a9eab2d64ad0c01983cd639866ee7965
thread:issue:16299,issue #16299: caffe2::CudnnConvOp::RunOnDevice() fails on Squeezenet,stale/low urgency,2,stale/low urgency,1,0,16299,,,,0,7b1bd54a907b156c6c4445a6764eaafabef04312251af320d2d107832bab92da
thread:issue:163013,issue #163013: Compiled HOPS do not support compilation when run under a custom DispatchMode,stale/low urgency,2,stale/low urgency,1,0,163013,,,,0,0b9ff1c693f25defcb0ecbcdf13b3b3e27e7fd978fcc554741828f4f9a914f60
thread:issue:163048,issue #163048: [compile/aot-autograd] `torch.autograd.grad` w.r.t. an intermediate tensor works in eager but fails under `torch.compile` (“tensor not used in the graph”),stale/low urgency,2,stale/low urgency,1,0,163048,,,,0,f3fddff1cdeebf29f3071964dcf287788feef82789f4340ca8cea8ba40efbed0
thread:issue:163052,issue #163052: CPU dynamic quantization (qint8) cannot be used with multiprocessing,stale/low urgency,2,stale/low urgency,1,0,163052,,,,0,5fd99866e47d51a56b3ec2f3acaba56feb7f77022b0b99d6ee464f46aee23681
thread:issue:163062,issue #163062: GIL is not released when creating a device tensor using a shape depending on data from another device tensor,stale/low urgency,2,stale/low urgency,1,0,163062,,,,0,d59e4c0375a2808f16137a2be4ab4368ce25d329e2f78dfc860c2218ef8567b1
thread:issue:16318,issue #16318: Pytorch with CUDA aware OpenMPI for Infiniband not working with HCOLL and MXM,stale/low urgency,2,stale/low urgency,1,0,16318,,,,0,3dfce813c22a36af7e7ea411a8bda79c8cd1876ed99268c2febba7c23e419484
thread:issue:16322,issue #16322: We're binding a bunch of crap to 'torch' namespace which shouldn't be there,stale/low urgency,2,stale/low urgency,1,0,16322,,,,0,a95c8613f58703c3966a9f7a34a9090f2d2a59258011552d516edc7ae4f56213
thread:issue:16328,issue #16328: Should torch.arange take a layout parameter?,stale/low urgency,2,stale/low urgency,1,0,16328,,,,0,2c13dc6840695f4bd35117a80105820bde5a7bdeaf1f64c1a78335a1ed64bb8f
thread:issue:163290,issue #163290: Tracker for migrating xformers to torch stable ABI,stale/low urgency,2,stale/low urgency,1,0,163290,,,,0,80d605eadf5d7516ef65d7ccc7f5dc72c771abf8c8539bace9963cfed64f7c5e
thread:issue:16330,issue #16330: support `unique_indices` option for `unique`,stale/low urgency,2,stale/low urgency,1,0,16330,,,,0,e954e0f610c8881245ad379e15f2c96dabc20449ce3c2804cf97ab7b5674b557
thread:issue:163302,issue #163302: [vLLM CI] Periodically run B200 tests,stale/low urgency,2,stale/low urgency,1,0,163302,,,,0,a81aadd2832151c27d4b502c61c2ef31400848b66117c0cd94c261abf9388e1d
thread:issue:163400,"issue #163400: Add Flash Dynamic Mask Attention (FDMA) as an SDPA backend for sparse, extremely long-context attention",stale/low urgency,2,stale/low urgency,1,0,163400,,,,0,2c62586f8ffe445fe047a6987cbda2f9349b75c190c9cf4a69329f6c3e4f222f
thread:issue:163425,issue #163425: `torch._grouped_gemm` has poor performance for large batch sizes with dtype BF16,stale/low urgency,2,stale/low urgency,1,0,163425,,,,0,a68f4417f4b9c87b5f0f2502b6bd1de9a476d8cd76017f4adec88ba891a96c6c
thread:issue:163428,issue #163428: [vLLM] Crash during inferece in torch dynamo on 2xRTX2080Ti,stale/low urgency,2,stale/low urgency,1,0,163428,,,,0,f024cfba8d9d08acf3ebf5a7148135955ac53adad149941e6c7b8b31f272106a
thread:issue:163502,issue #163502: at::Tensor::to_sparse_csr() does not produce correct result in Debug configuration when tensor is big sparse CSC on CUDA device in LibTorch 2.8.0,stale/low urgency,2,stale/low urgency,1,0,163502,,,,0,34c6b08f36b14ce2e70ca7ebc8f582b1cbb7aae4b5a811100170869f9dba4cea
thread:issue:163574,issue #163574: Bfloat16 large numerical discrepancy on Blackwell,stale/low urgency,2,stale/low urgency,1,0,163574,,,,0,758bc54bb5c85a61723847604a30e605ff43bf6f8df4de26def46a00a17727f2
thread:issue:163577,issue #163577: Lintrunner failures on files that only manifest on CI,stale/low urgency,2,stale/low urgency,1,0,163577,,,,0,f98538de762612e35409bd836a98a4983cc27d673df88c0df65e559d1af414d1
thread:issue:163624,issue #163624: [aoti] [xpu] [null-pointer-deference] potential npt issue in `sycl_runtime_wrappers.h`,stale/low urgency,2,stale/low urgency,1,0,163624,,,,0,24ac18d2e4bffc3f44878c75c273943bd48b13a587ce5c4e5752b7141498e31c
thread:issue:163657,issue #163657: UNSTABLE periodic / linux-jammy-cuda12.4-py3.10-gcc11 / test (legacy_nvidia_driver),stale/low urgency,2,stale/low urgency,1,0,163657,,,,0,9558c91b06391bfbbdd73fd8c500e406092a254fe502a5ddae69445698fe8b2a
thread:issue:163659,issue #163659: Allow double in native_functions.yaml as a schema type,stale/low urgency,2,stale/low urgency,1,0,163659,,,,0,5c7d8d7536b4f1a447b24457cf0d41aa7da62a990ae97d33e1f5d3f99e50e8af
thread:issue:163709,issue #163709: Inconsistent handling of backend floating point precision settings (related to PR #125888),stale/low urgency,2,stale/low urgency,1,0,163709,,,,0,21baea22e67ae939cd5c82c54c0879804ff2e422be0418952bae47742c711479
thread:issue:16375,issue #16375: Python-bound C++ frontend modules don't handle attributes well,stale/low urgency,2,stale/low urgency,1,0,16375,,,,0,96427f54f778fe263cdf403cf5160379b39323f35aaddb0d7168b28a514a1fc8
thread:issue:163785,issue #163785: Revisit guarding on unbacked inputs !,stale/low urgency,2,stale/low urgency,1,0,163785,,,,0,3e0e90d9b440c72566cce839bca09abd5679628354bd897cc099d569e38b7f20
thread:issue:163789,issue #163789: [docs] instructions to locally build docs are underspecified,stale/low urgency,2,stale/low urgency,1,0,163789,,,,0,531b5845ef195e963ae3a51ce42d71a06951f85cfef08ac23f3cdacbea1f6a7f
thread:issue:163821,issue #163821: Support for nn.BitLinear layer,stale/low urgency,2,stale/low urgency,1,0,163821,,,,0,85fea1678015edcdf36bd22781ce327b3728ba7608400f25ef2ff9ba9a7ef6b3
thread:issue:16385,issue #16385: Adding new module to caffe2,stale/low urgency,2,stale/low urgency,1,0,16385,,,,0,f0f96994a30d413d0c24f9eeea0862818cede944ab841a37e470e83eedf7da8a
thread:issue:163902,issue #163902: Add banner for docker sevs to hud,stale/low urgency,2,stale/low urgency,1,0,163902,,,,0,1165461d1de03edd1445e4c6c84e536babfb93563f01b028f9c572ed89127c06
thread:issue:163950,issue #163950: ImportError: cannot import name 'skip_code' from 'torch._C._dynamo.eval_frame' (unknown location),stale/low urgency,2,stale/low urgency,1,0,163950,,,,0,0449e14cd4e5e86173ff40155ce34cc1efc38b841b4bc8067355f93776e00e70
thread:issue:163982,issue #163982: Need to update Magma version in Pytorch,stale/low urgency,2,stale/low urgency,1,0,163982,,,,0,7465710622916497ed3b63f6ba17d7f52149883e060955bc9c6aa92f88668180
thread:issue:163984,issue #163984: [inductor] Convert complex operations to real-block matrices,stale/low urgency,2,stale/low urgency,1,0,163984,,,,0,0a8aa28060335bb7f602c4d5db82219a0179a747ddb642fa9fcf2de518cc513a
thread:issue:16402,issue #16402: error with nccl when distributed training on caffe2,stale/low urgency,2,stale/low urgency,1,0,16402,,,,0,18b2fe9bd0e20bdb17a6f439555d773211a8203043bf317584c38c67afd2b98a
thread:issue:164041,issue #164041: [vLLM] AssertionError: out= ops aren't yet supported,stale/low urgency,2,stale/low urgency,1,0,164041,,,,0,d07992c13c0a4c57c4eb25c80aeedc1bf67fdc6ce324e8fdf4de5944d0329de7
thread:issue:16405,issue #16405: error on cmake_version from tools/build_pytorch_libs.py,stale/low urgency,2,stale/low urgency,1,0,16405,,,,0,6eafe3d90400aa7103b16d99a34fc3298cae912e02811c004580b1328965820e
thread:issue:164061,issue #164061: GPU Memory Leak due to distributions,stale/low urgency,2,stale/low urgency,1,0,164061,,,,0,936ae0bf12a8109590115b8a14ba4051d019adfe7395741f4ecfe2019680538c
thread:issue:164062,issue #164062: [strict-export][install_free_tensors] Tracker for issues,stale/low urgency,2,stale/low urgency,1,0,164062,,,,0,c2ab20f715ee3a43af93004f336d2ce1b62f5ca6265616cfa14c3640072075f5
thread:issue:164089,issue #164089: Failing Operator Benchmarks,stale/low urgency,2,stale/low urgency,1,0,164089,,,,0,0c046a35c2b24d2fc84625a9f0de4751afd396e4b97bf928a1fff9ba5f5b4dad
thread:issue:1641,issue #1641: [Feature request] In-place 'max' method for Tensor,stale/low urgency,2,stale/low urgency,1,0,1641,,,,0,42c98b00f8012ac10e44f889432cb52aff5a0b699786dbdce1039f2168097b0f
thread:issue:164116,issue #164116: qta and export onnx error,stale/low urgency,2,stale/low urgency,1,0,164116,,,,0,6517fc7546f95657adf2ad835dc2ae9823f0de7f0b88977919b380ffbe158bf7
thread:issue:164122,issue #164122: [CUDA][Sparse][Linalg] torch.sparse.spsolve is unavailable,stale/low urgency,2,stale/low urgency,1,0,164122,,,,0,fb42b958256981e4cc5a7960dfb458652379e15ac2da6d2cbfa34159921cbd5c
thread:issue:1642,issue #1642: feature request: On-the-fly Operation Batching in Dynamic Computation Graphs,stale/low urgency,2,stale/low urgency,1,0,1642,,,,0,8795c0e3a1aa06c41ff4b07ba150bbf8c158d906cfd59d2f22330cbf3f31493b
thread:issue:164249,issue #164249: Accuracy issue with sum-reduction on A100,stale/low urgency,2,stale/low urgency,1,0,164249,,,,0,185fdcb6a88fc6cfc5fadf3077eecffa5cf2f9d02fc0917210b6f7f329fc33d2
thread:issue:16432,issue #16432: A bug in parallel.data_parallel when module_kwargs is not None,stale/low urgency,2,stale/low urgency,1,0,16432,,,,0,b82150d848a7726753306b4eda5821dc8c46ac4b1d1d5b98fc76687de5a32657
thread:issue:164331,issue #164331: Anomaly mode (detect_anomaly) should populate backward stacks with torch.profiler.profile(with_stack=True),stale/low urgency,2,stale/low urgency,1,0,164331,,,,0,a16f2c0a60181396dfb8510f31a8fea8912ea6fa63f2c450920cf4fec777fdb4
thread:issue:16436,issue #16436: How to retrain the modelzoo model in Caffe2?,stale/low urgency,2,stale/low urgency,1,0,16436,,,,0,2db247039d7c16bd4ecad2b03175d2d54d9ebcb72266dc44f17629a968763cd8
thread:issue:164493,issue #164493: [CI][CUDA][SymmMem] B200 Periodical Job Failure with test_triton_sum_reduce_int64,stale/low urgency,2,stale/low urgency,1,0,164493,,,,0,ed1d0015544fef5d78e25c501c931b1e6973b453ccd149b4e36feef20b207ea5
thread:issue:16458,issue #16458: TensorRTOpTest.test_vgg19 is flaky,stale/low urgency,2,stale/low urgency,1,0,16458,,,,0,d2dd539b57f906c691f2758920f8ff7903bd5fe6bbcdfe059ae0ff89fa6be0b3
thread:issue:164707,issue #164707: Significant difference in performance between CUSPARSELT and CUTLASS when doing SparseSemiStructuredTensor matmul,stale/low urgency,2,stale/low urgency,1,0,164707,,,,0,f686a7dbb0a3fd91eb65ee92041afad449e1235cfd7822762a6c146817ff84a9
thread:issue:164732,issue #164732: Fake tensor leakage in aot_autograd by itself,stale/low urgency,2,stale/low urgency,1,0,164732,,,,0,47c8b844b87be6d98edb965145c60b7a3699e509eabee6e253db242ea62370a2
thread:issue:164756,issue #164756: [PP] Unshard and reshard not working with FSDP,stale/low urgency,2,stale/low urgency,1,0,164756,,,,0,978e0e24d85e2929ce6b8259523f15ae2dbcf9310e5954268da2c7c20e8fb365
thread:issue:164759,issue #164759: Bug : Side effect in th.func.vjp of th.func.jvp,stale/low urgency,2,stale/low urgency,1,0,164759,,,,0,4dc236342af30f3bcbef1a4029c968f7f40f2286189c59c5cbac18b4eab65faf
thread:issue:164786,issue #164786: How should we handle PyTorch build flags in torch/headeronly for custom ops?,stale/low urgency,2,stale/low urgency,1,0,164786,,,,0,9cd9c96a2480fd52810e5edb03b41a81b3a8e9f1ef91e306b4f86f7f3593451f
thread:issue:164799,issue #164799: Check behavior of multicast object release when using clang to compile PyTorch,stale/low urgency,2,stale/low urgency,1,0,164799,,,,0,068607c83e3430c68543d78929f2202dea8e90e56fd57a21701a09e18cbe6065
thread:issue:164813,issue #164813: flex_attention not functional in NVIDIA RTX 2080Ti,stale/low urgency,2,stale/low urgency,1,0,164813,,,,0,ea9d0f980e671e6c36f9669d0056f9148a052602653201f2834ee975da3d8f96
thread:issue:164843,issue #164843: `set_model_state_dict()` causes `AssertionError: found no DeviceMesh from dtensor args for c10d.broadcast_.default!`,stale/low urgency,2,stale/low urgency,1,0,164843,,,,0,f6fcfc9ce8f0675aa81780f2425cf90ede198b8ee55fd9e25a2391186ba81247
thread:issue:164860,issue #164860: Driver update on H100 and A100 instances,stale/low urgency,2,stale/low urgency,1,0,164860,,,,0,89c1583ad37985db227a1709e1af07cbf7f26da5600d37b80374e1aa201e46c4
thread:issue:164862,issue #164862: Fix incomplete exception handling in validate_input_col function,stale/low urgency,2,stale/low urgency,1,0,164862,,,,0,8983bf6217bc73f0920b49dda031eee518b91deca7d324dae297e48f0c0b11ef
thread:issue:164872,issue #164872: Tensor subclass slice during `inference_mode` fails,stale/low urgency,2,stale/low urgency,1,0,164872,,,,0,9bff6b1b4b7a7ad62e7f41bc59c9b8979b0b0408ba33523e8bbe005b026de5ec
thread:issue:164878,issue #164878: Ban and remove plain asserts with no message in our python code,stale/low urgency,2,stale/low urgency,1,0,164878,,,,0,9acd6ca266e1ffb59ee9ce3381111051e2bff2d70582ad06c8b51a04bbc0a9f3
thread:issue:164891,issue #164891: Better API or GuardRails for Distributed Optimizations with non-SPMD graphs,stale/low urgency,2,stale/low urgency,1,0,164891,,,,0,8de6880696482a73d61058dd7490b73fc0869f9964b84bd9349553c28bd9f809
thread:issue:164925,issue #164925: add support for distributed backends to specify default backend in multiple backends format,stale/low urgency,2,stale/low urgency,1,0,164925,,,,0,e5d19ad30a31cd1c8adab419696b44d63f0f9127b6fe376beb43f5ae7155e9f4
thread:issue:16494,issue #16494: caffe2 softmaxwithloss problem,stale/low urgency,2,stale/low urgency,1,0,16494,,,,0,e1855404aa8c6ded89b6470410fae8ec70e50453140df5db338360033bf30c25
thread:issue:164951,issue #164951: Docker checkouts take 30+ min on H100 runners,stale/low urgency,2,stale/low urgency,1,0,164951,,,,0,5c05c1214f65cedb06f0e3bd7205042de4ddfac4158f820ae411c7072bc3f7ee
thread:issue:16499,issue #16499: TorchConfig.cmake always sets _GLIBCXX_USE_CXX11_ABI,stale/low urgency,2,stale/low urgency,1,0,16499,,,,0,446e1fdc18e004d45dd056744723ddc33cd28645f3c19d34a90637ad73f0d32e
thread:issue:165042,issue #165042: constant ones/zeros tensor,stale/low urgency,2,stale/low urgency,1,0,165042,,,,0,f038f4238b7b03a8c186c6ee40084a9cc9a0c91753fc6419bc8e6a0ae5670b3c
thread:issue:165100,issue #165100: Header files not found during build,stale/low urgency,2,stale/low urgency,1,0,165100,,,,0,7dea1011bb5699c7109af39e91dcb72ff69ba31642b18fc4f0313aced398b8dd
thread:issue:165122,issue #165122: RPC: abort when RRef from remote() is mistakenly called with wait() due to unhandled exception in thread pool,stale/low urgency,2,stale/low urgency,1,0,165122,,,,0,cfbc6d137d78c3461382379219a5fe1f9ae80c48cc2b24eafc8c8194fac4cd47
thread:issue:165173,issue #165173: Add CUDA 13.0 in CI,stale/low urgency,2,stale/low urgency,1,0,165173,,,,0,e3f0d139f6f6812e2267e11ae60620b0d849923d59fa08b0201d0990d62f5a38
thread:issue:16523,"issue #16523: The word ""ints"" is not rendered correclty in docs.",stale/low urgency,2,stale/low urgency,1,0,16523,,,,0,8269af324ebd47e10fc7a8165e86baf42afe153a69e5fbb2a1bd882b08f6853f
thread:issue:165251,issue #165251: out_dtype for einsum,stale/low urgency,2,stale/low urgency,1,0,165251,,,,0,2bc0ba0ab9c9211b60062ca6b2ad30b067416e8a07f5a340af5a3cbbeaf2311f
thread:issue:165277,"issue #165277: autowrap_modules should be tuple[ModuleType, ...]",stale/low urgency,2,stale/low urgency,1,0,165277,,,,0,8445bfded59ebccdaa98b45ab5d417b5a179e11b63835f64e525c277a634c6e2
thread:issue:165312,issue #165312: TransformerEncoder crashes with IndexError when num_layers=0,stale/low urgency,2,stale/low urgency,1,0,165312,,,,0,6f38cfa8ec8a30bf6809052866070c65996de0285bbd2d8d490c096c28b02466
thread:issue:165319,issue #165319: Memory leak when converting from numpy array,stale/low urgency,2,stale/low urgency,1,0,165319,,,,0,b1730f983a0640c4bb72d264b6e8a78acc8537b3cf0fc8161d07c307eeb47b1b
thread:issue:165324,issue #165324: How to enable Bfloat16 when using torch.func.jvp,stale/low urgency,2,stale/low urgency,1,0,165324,,,,0,fa3a5b222fa8f40638ff7831b2cf60a3a9f0483f18b101b38ec711d080d5f7d5
thread:issue:165409,issue #165409: [torch.compile][fullgraph][functionalization] In-place op on overlapped view from unfold fails: “tensor being mutated has internal overlap” (eager OK),stale/low urgency,2,stale/low urgency,1,0,165409,,,,0,b6b3a2494dd8b443498a717dc688b25f07cee29eaecadc33f47d26cef56dc2e8
thread:issue:16542,issue #16542: Error with setting tensors to use cpu in packed_padded_sequence when CUDA tensor is set as default,stale/low urgency,2,stale/low urgency,1,0,16542,,,,0,8670873a569a5278e408311ba5ebdef5c3d7f94d03e62ce0e3414b9aed28bba9
thread:issue:165421,issue #165421: [torch.compile][fullgraph][dynamic shapes] aten.masked_select.default rejected as “Dynamic shape operator” (eager OK),stale/low urgency,2,stale/low urgency,1,0,165421,,,,0,96604bfe40c3b559e3b53fa993091c3aaacb2b2aa6ee353305ff296e7ae42473
thread:issue:165425,issue #165425: Support FFT operations in FLOP counter,stale/low urgency,2,stale/low urgency,1,0,165425,,,,0,ea1313366cd0de948f915786f138d3f73974f592bed3e5c310a20738e0c2ba33
thread:issue:165537,"issue #165537: `torch.linspace` with `inf` endpoint and `float16` dtype returns all `nan`,inconsistent with NumPy",stale/low urgency,2,stale/low urgency,1,0,165537,,,,0,50c9f69ac319b5a1550e65d6a3f24af6e1682f25f807f0c99393685f27fdaa4e
thread:issue:16556,"issue #16556: KeyError: 'No translator registered for layer: name: ""res3b_relu_norm""\ntype: ""Normalize""\nbottom: ""res3b""\ntop: ""res3b_relu_norm""\nnorm_param",stale/low urgency,2,stale/low urgency,1,0,16556,,,,0,f34d2d66d82acadf6f2df7eb0e90ce96bcd47c1771e79088cf469b60ba312c75
thread:issue:165570,issue #165570: Add support for element-wise operations to flop counter,stale/low urgency,2,stale/low urgency,1,0,165570,,,,0,976c63100de73a2576095e1a6e752c88e842e5e0ed00a89922a4f7200adc4454
thread:issue:165578,issue #165578: Out of tree backend documentation does not seem accurate,stale/low urgency,2,stale/low urgency,1,0,165578,,,,0,232c875af6c6c539ff0cfcf2547202e7dec737cf4e185e9f51e60feb22b9eef9
thread:issue:165621,issue #165621: Feedback about RNN,stale/low urgency,2,stale/low urgency,1,0,165621,,,,0,e3e2e495fb194b2af7c3d86113e98b4d9a91cd9f1b86cda4a31d10955ee6e1f1
thread:issue:165626,issue #165626: Code Style Cleanup: Replace python3 -mpip with the more readable python3 -m pip,stale/low urgency,2,stale/low urgency,1,0,165626,,,,0,8974425ac7566d00a1375309f37a27cd18f29b584079eecd817730cebea371d2
thread:issue:165634,issue #165634: make_fx pre_dispatch=True produces incorrect graph when BatchedFallback is triggered,stale/low urgency,2,stale/low urgency,1,0,165634,,,,0,313c00c83006dc7cb9e84705f35b61981b8fe47cfca8dd17707964dbaecac081
thread:issue:165684,issue #165684: Python version in test CI job doesn't align with the python version in the docker build log,stale/low urgency,2,stale/low urgency,1,0,165684,,,,0,10d97e619c43735f4e1b8726e6415a497ed5eb4e345239d69ce83063c1d6ad7a
thread:issue:165715,issue #165715: Missing spans in CPU section of profile likely caused by memory allocation with expandable segments,stale/low urgency,2,stale/low urgency,1,0,165715,,,,0,70120ff1b8dca42ca0d5dfa75148f30f7165cf249057ca64a5bf19ae868646b1
thread:issue:165730,"issue #165730: autocast('cuda', enabled=False) silently ignored on NPU without warning, causing precision issues",stale/low urgency,2,stale/low urgency,1,0,165730,,,,0,e112aa112f0378e19c50ac56dd4a9909cfd56af29394b1729675467d5722f2bc
thread:issue:165739,issue #165739: Inconsistent in `torch.fft.irfft` and `torch.fft.ifftn` for complex64 inputs containing `Inf`,stale/low urgency,2,stale/low urgency,1,0,165739,,,,0,f968f1289c1b1d3b6ea20f840a25447545438c08289a173ab01005c5cc68ef13
thread:issue:165741,issue #165741: Ctx or at least Thread level switch for fp32 IEEE precision tolerance,stale/low urgency,2,stale/low urgency,1,0,165741,,,,0,966c3411733df306a87fcfe58c7e0b4611713d8a08366d80adbdf748dfc6c19c
thread:issue:165744,issue #165744: Inconsistency in `conv_transpose2d` for float16 with inf weights between CPU and CUDA,stale/low urgency,2,stale/low urgency,1,0,165744,,,,0,b1e96e24d30e624b85e602c4917241f98c909b9263f2bd71b831abe78865d758
thread:issue:165799,issue #165799: `torch.where` does not accept scalar argument when `out=` is passed,stale/low urgency,2,stale/low urgency,1,0,165799,,,,0,56d2e516726f2d10a1a6533a2c7bc51fdbdf98b677825739cd606665261fc775
thread:issue:165811,issue #165811: [RFC] A Python backend registration API,stale/low urgency,2,stale/low urgency,1,0,165811,,,,0,e19b25320a9f5db4e57e73084969cd6d5d8273bb8aa8c1c929c8f207ae7f0133
thread:issue:165826,issue #165826: “Request: Add CUDA support for NVIDIA RTX 5070 Ti (Ada Lovelace Refresh)”,stale/low urgency,2,stale/low urgency,1,0,165826,,,,0,ff4cd0b5a2da735d2f36ce11f67baa2d42faf05a2f1a03243c29e54e67fb9b8f
thread:issue:165880,issue #165880: torch.vmap doesn't work on most of the models,stale/low urgency,2,stale/low urgency,1,0,165880,,,,0,86d0732b7a6d694ee2c172befbcd6751c4616e0e29524e7437188593fe6e73e8
thread:issue:16589,issue #16589: Make sure `data_ptr` for non-zero-size input tensors stays the same after the VariableType dispatch,stale/low urgency,2,stale/low urgency,1,0,16589,,,,0,31748cb36a35489055da29086d231a18f2f6d07247097c89deebaaafc9fa9d44
thread:issue:16590,issue #16590: Benchmarking pre-trained model in Caffe2,stale/low urgency,2,stale/low urgency,1,0,16590,,,,0,12b9bd83ec96382249493aa0fe35d3fb57e51bc117da20488fef9ee09934bb0a
thread:issue:165902,issue #165902: torchcodec in pytorch url,stale/low urgency,2,stale/low urgency,1,0,165902,,,,0,bb365c1b3a6eb29ab686a6660bba14cd74ebca241e95343f117053e22cc2ea26
thread:issue:165906,"issue #165906: `torch.bmm` returns wrong results for large tensors with torch.set_float32_matmul_precision(""medium"")",stale/low urgency,2,stale/low urgency,1,0,165906,,,,0,8853302ffc0a3877860c1bedd7b50f9509520f724ba8e430d66b170776efacd6
thread:issue:165912,issue #165912: Runtime assert on unbacked input inside torch.cond improperly gets lifted to top level,stale/low urgency,2,stale/low urgency,1,0,165912,,,,0,ed253a445027eeac13ee350b2191d309c3e0b6c41c151cb45e28d681287e56a8
thread:issue:165976,issue #165976: Internal Assert Failure with Specific Large Padding Value in torch.nn.Conv3d,stale/low urgency,2,stale/low urgency,1,0,165976,,,,0,49a469033615a9df26920b2787e8657023fdeb823434b63d9ee920c71d9f4758
thread:issue:165982,issue #165982: LELU Activation Function: Proposal for PyTorch,stale/low urgency,2,stale/low urgency,1,0,165982,,,,0,bb771be2a18c8c4a1a3b0803d8a8eff7fe9888133d85f8cc3549b086398037f6
thread:issue:16599,issue #16599: Feature Request: Earth Mover's Distance Loss,stale/low urgency,2,stale/low urgency,1,0,16599,,,,0,3a67d20916ed2bd4d264bf662f399349d2a70deb01b5f76e060a8829f2f8f14d
thread:issue:165992,issue #165992: AttributeError: module 'triton.language.math' has no attribute 'tanh',stale/low urgency,2,stale/low urgency,1,0,165992,,,,0,598b326579a70a931a911eb9613d46336444ec67984ff55c83495e0a51e81fa6
thread:issue:165999,issue #165999: [RFC] Improve the performance of `associative_scan()` with `cuda.compute`,stale/low urgency,2,stale/low urgency,1,0,165999,,,,0,b559fed926a0d83e4bde737f2f80b2b9f074ddfa1f9924b24ffc70e0c657ac60
thread:issue:166011,issue #166011: cusparse deprecation warning in CUDA 13.0 inductor build,stale/low urgency,2,stale/low urgency,1,0,166011,,,,0,b59b077229fd23556b3924bb9270f7e8b0d00ec876b4eac3a9d6563fd22f08ac
thread:issue:166054,issue #166054: Add XLA to `at::accelerator::getAccelerator` and `at::accelerator::isAccelerator`,stale/low urgency,2,stale/low urgency,1,0,166054,,,,0,de46eddf7469c3bc3e5f56cd2d6f23735ac5c33e708d5db059ebc56480f59949
thread:issue:166064,issue #166064: Add submodul for `six`,stale/low urgency,2,stale/low urgency,1,0,166064,,,,0,c6ea5a83d8f531a5340bdb208f115b00d6ed431f6c7b5682aa5edb1ceb179aff
thread:issue:166072,issue #166072: UNSTABLE trunk / linux-jammy-py3-clang12-executorch / test (executorch),stale/low urgency,2,stale/low urgency,1,0,166072,,,,0,397609aa69d2cf444a9fb923d30e56dfbfd20d6e3ebc10bc4602ac4c806e4f38
thread:issue:166080,issue #166080: Feature Request: masked convolution,stale/low urgency,2,stale/low urgency,1,0,166080,,,,0,8d46d7d5a464d3401010335d928b8f2eed38820b4468c82e737060702ca623e5
thread:issue:166101,issue #166101: Compiling libtorch requires GLIBC_2.35,stale/low urgency,2,stale/low urgency,1,0,166101,,,,0,a2121624e70ca488d3cc179148d0be73468190b2ce5323b4853e82fd24d108f4
thread:issue:166114,issue #166114: [BUG]interpolate operator triggers severe performance degradation when enabling deterministic computation,stale/low urgency,2,stale/low urgency,1,0,166114,,,,0,9fe5902cf8ccbc1f363292e691957c4cd2b19ab6bd95294526f7ff8469c7656c
thread:issue:166116,issue #166116: [CCA] CUDACachingAllocator always release physical memory handle when the expandable segment unmaps.,stale/low urgency,2,stale/low urgency,1,0,166116,,,,0,f084062eb3fa03c1a45bd5f973a41c36c10bf3d1966f102a3cd68d15351b60a6
thread:issue:166131,issue #166131: Return a view from tensor(requires_grad=False) in autograd function may cause incorrect requires_grad attribute.,stale/low urgency,2,stale/low urgency,1,0,166131,,,,0,177b97d633afc103fe3dc8c7ba11bc863406515dfe8aefe6313e4236052049da
thread:issue:166167,"issue #166167: Proposal: include 'extras packages' for supported versions of PyTorch-adjacent tools (e.g., numpy, onnx)",stale/low urgency,2,stale/low urgency,1,0,166167,,,,0,231ae6fa9a7a110251385127baa03a5561b91979fdeab7e2f1c5f9dd64e5ab43
thread:issue:166186,issue #166186: We should have an env var to turn off all dynamic shapes,stale/low urgency,2,stale/low urgency,1,0,166186,,,,0,f894787dde9a923f8af23bf71d91e584dec08b38c051ccc9c3f057623e6c402c
thread:issue:166219,issue #166219: Why are there so many warnings when building the C++ libtorch project? How to resolve it?,stale/low urgency,2,stale/low urgency,1,0,166219,,,,0,0fb3945686c4d8f93738e0a2457bb4ec4a8ec8e225917c912561e14df44989eb
thread:issue:166233,"issue #166233: license: Is it possible to stop using Conda in the Dockerfile? Due to Conda’s licensing issues, many companies have already received legal warning letters.",stale/low urgency,2,stale/low urgency,1,0,166233,,,,0,af762716edb3648a11996ffb855c3dbd46222e6f3f0ae8e42c138e33a9195bce
thread:issue:166234,"issue #166234: !handles_.at(i) INTERNAL ASSERT FAILED at ""/pytorch/c10/cuda/CUDACachingAllocator.cpp"":430",stale/low urgency,2,stale/low urgency,1,0,166234,,,,0,094f0097889ed32539a3dd5356d2ebe83d9b53f3a97c7f9ff0663fa3db102490
thread:issue:166243,issue #166243: [BugReport] Exponential distribution may generate zero values on CPU backend,stale/low urgency,2,stale/low urgency,1,0,166243,,,,0,d9fda374511901225c59d1b82baf855099d2f617b6b38f0ce941263eb984dada
thread:issue:166378,issue #166378: Multihead Attention's in_proj_weight is initialized inconsistently,stale/low urgency,2,stale/low urgency,1,0,166378,,,,0,e38c2118d8b84771c266349af646181d54d397fbc5aa8407c62deaee541f7991
thread:issue:166401,issue #166401: torch.masked.argmax returns different results on CPU vs CUDA when multiple maximum values exist,stale/low urgency,2,stale/low urgency,1,0,166401,,,,0,85c2d5fb8b601ff106b5b612e5469167d0615b375cf80c7ff202c58b8002dd40
thread:issue:166409,issue #166409: `torch.polygamma` produces inconsistent results between CPU and CUDA for moderate orders,stale/low urgency,2,stale/low urgency,1,0,166409,,,,0,6e52c9a39d2b78f2e934f0e1d4a6135fb6af7bed8b3986a1bc746ccf363555be
thread:issue:166511,issue #166511: RuntimeError when finetuning Gemma 3 using Unsloth on Databricks,stale/low urgency,2,stale/low urgency,1,0,166511,,,,0,e88d593351eeb287d02bfa94b6515f802b0effdddf0fb0bf4af615992a22969a
thread:issue:16655,"issue #16655: CUDA cache doubles on the second batch and causes OOM, `empty_cache` doesn't empty it",stale/low urgency,2,stale/low urgency,1,0,16655,,,,0,07d4ac9bb32b78b7d31b1792481a6e0fed29f7b697812b29699267c1fc4adffb
thread:issue:166557,issue #166557: Two conflicting definitions of `torch.backends.cudnn.rnn` since #125888,stale/low urgency,2,stale/low urgency,1,0,166557,,,,0,ceec14eb63681a34fac15f386087408273ad2499997c6482a3cebfdc257a6e08
thread:issue:166611,issue #166611: CuTe DSL NVFP4 GEMM/Grouped GEMM kernels,stale/low urgency,2,stale/low urgency,1,0,166611,,,,0,d6e20ac2c56d7a598c946c4b3ca91a7a0a01e7a189fb4e5d144330d8c0610d81
thread:issue:166646,issue #166646: Allow kwargs in torch.utils.checkpoint.checkpoint().,stale/low urgency,2,stale/low urgency,1,0,166646,,,,0,456cac4e08a84be1bc96ad6ec6b1a28d418e1daa2f6f4885808730db09dc17dd
thread:issue:166651,issue #166651: Make `_grouped_mm` public and add out_dtype,stale/low urgency,2,stale/low urgency,1,0,166651,,,,0,c95af10359f8fec10462bd8e3d5e30247fe1af170b439b47b5cab7636aae10d3
thread:issue:16666,issue #16666: numa_test.py fails with AssertionError,stale/low urgency,2,stale/low urgency,1,0,16666,,,,0,10a54c97c8303fa914e0be5b9f887311e094d2a9b545f2298ecdc290706d75dd
thread:issue:166686,"issue #166686: pytorch can not load libtorch trained pt pth model ,",stale/low urgency,2,stale/low urgency,1,0,166686,,,,0,8d2eace77930281aa23b7d1b09d2a301e6cd6598d196242689c8d8d67ed0a1af
thread:issue:166696,issue #166696: `torch.accelerator.is_available()` fails when privateuseone is used,stale/low urgency,2,stale/low urgency,1,0,166696,,,,0,0049f6cd4c3dd3283a231dcbbf7effbf4a46c6eff4db6011ea6d232e7d086b02
thread:issue:166733,issue #166733: benchmark and advertise torch.compile's ability to reduce triton kernel launch overhead,stale/low urgency,2,stale/low urgency,1,0,166733,,,,0,699515da698ac48dc51b46b4739682347fb72f97b6926937c7b69f980c60fffc
thread:issue:166823,issue #166823: Change tab order of docs to have main content before left sidebar,stale/low urgency,2,stale/low urgency,1,0,166823,,,,0,3204b7934db474cca7b661827f15d63c58e7d15b6c0c1dc397acb28c57bdf71b
thread:issue:166853,issue #166853: torch.randn on CPU is not accurate with dtype=bfloat16,stale/low urgency,2,stale/low urgency,1,0,166853,,,,0,cefe2b71e03a02bf727388aeefed39528682940b236be8519dcab64d6e8091f9
thread:issue:16695,issue #16695: Check whether _cudnn_rnn_flatten_weight can avoid changing the TensorImpl or Storage pointer of tensors in `weight_arr`,stale/low urgency,2,stale/low urgency,1,0,16695,,,,0,54b88eb2ba6e5cc79ab7c66c134ae90af9901910156a387f90997cdb24d7edfa
thread:issue:16700,issue #16700: [feature request] Store accumulated gradients in separate GPU or on CPU memory,stale/low urgency,2,stale/low urgency,1,0,16700,,,,0,72c696d448046f95c3fd29e631f8cfbb585c7e78b1ec9b58ae72d2e82e21b872
thread:issue:167027,issue #167027: combine compiled vectorized function without recompiling already compiled part,stale/low urgency,2,stale/low urgency,1,0,167027,,,,0,336c5191d6be9beb778f0827307ffb92bf28d528022ef96c9da55bb5b51cf38e
thread:issue:16703,issue #16703: split_with_sizes should accept a LongTensor as the split_sizes parameter,stale/low urgency,2,stale/low urgency,1,0,16703,,,,0,30c351eb464e6d9fb31883dcd914157ed1495669079eaab9cb8973a5c179a17c
thread:issue:167063,issue #167063: `torch.onnx.export` with `dynamo` breaks `FakeQuantize` exporting,stale/low urgency,2,stale/low urgency,1,0,167063,,,,0,2653b5c2e30a33414a75cc256641131a652e6ca026ed49f4685f07e17d472ef8
thread:issue:16710,issue #16710: culibos linker errors on binary_linux_conda_3.6_cu90_build,stale/low urgency,2,stale/low urgency,1,0,16710,,,,0,2230e55dfa3e328518d6ff5957d59b287aeec2677cf075b36b8566488a0b91c9
thread:issue:167119,issue #167119: Pyrefly may regress type coverage - consider reenabling mypy,stale/low urgency,2,stale/low urgency,1,0,167119,,,,0,1e53fa7ab2403f602248a71d6947f40f71dd16f339b6b7f0e487d79d5332c06e
thread:issue:167170,issue #167170: [XPU][Tracker] Enable `torch._scaled_matmul`,stale/low urgency,2,stale/low urgency,1,0,167170,,,,0,00301e7c66413c20307bd28a75cbf8de1c736a40b69b892b3adc8c704dbabc00
thread:issue:167210,issue #167210: [RFC] Introduce the Unified Memory Pool and Pluggable Allocator in torch.accelerator,stale/low urgency,2,stale/low urgency,1,0,167210,,,,0,fa8286fbb3e36533c3ab00b339612c98c2173c53782e9d0ef0855df9f50a3a73
thread:issue:167229,"issue #167229: The quality of button ""Ask AI"" is very low. I propose removing it. The button gets too much area of webpage, made it inconvenient.",stale/low urgency,2,stale/low urgency,1,0,167229,,,,0,1d982fd9c1b6fa39bac31646e3d63b201ad1b6ef8f20f5698e2866d166dcb889
thread:issue:167254,issue #167254: Fine control of fp32 precison in backward phase,stale/low urgency,2,stale/low urgency,1,0,167254,,,,0,f552e052a050ccc840a06409a00be97b776f99d9448d60bde0d5d22080ce3081
thread:issue:167257,issue #167257: Support cuda event input/output in graph with streams,stale/low urgency,2,stale/low urgency,1,0,167257,,,,0,5048858667532cebe737de4ca2d7980dfd77f9e53470c04f30a1cd88af56ac13
thread:issue:167349,issue #167349: AOTAutograd cache bypasses on DTensor.to_local(),stale/low urgency,2,stale/low urgency,1,0,167349,,,,0,8ed634f91874ff4eee5e142949ab2534f2124079027afd47d788e641121e6341
thread:issue:167351,issue #167351: move to gcc11 for CUDA 13 CI tests,stale/low urgency,2,stale/low urgency,1,0,167351,,,,0,88a39da83c4343a6bc090458d343f85ab81f841d4382117770977328ceb08154
thread:issue:16737,issue #16737: torch.save overwrite,stale/low urgency,2,stale/low urgency,1,0,16737,,,,0,9ee08757d7e3308ed640308dcb572d58f3ad69679569c768de706180b4a7bda7
thread:issue:167409,issue #167409: FFT operations hang indefinitely on RTX 5000 Ada (driver 570.86.15),stale/low urgency,2,stale/low urgency,1,0,167409,,,,0,6f1ebdc2ea369231576531d644bacd4d40f1d9381844dd8f5b52d7b67d3e7ebd
thread:issue:167420,issue #167420: conv3d with dilation and bias = severe performance degradation in bf16/fp16,stale/low urgency,2,stale/low urgency,1,0,167420,,,,0,ddd49811014719c987c0ea1b994f5c6a09f87fbc040665d21b337f5c9c9435ec
thread:issue:167447,issue #167447: Memory related APIs in torch.accelerator are not supported on MPS,stale/low urgency,2,stale/low urgency,1,0,167447,,,,0,b6fb17d74af8d08ce992f033b5a8bb6aa3f2f1a9511909dfcb3af5a3d4c77b48
thread:issue:167520,issue #167520: improve selective decomposition tracing time,stale/low urgency,2,stale/low urgency,1,0,167520,,,,0,1680885ac4cc30947b1eaf0938e8d3a09039f8f6c886cc643419d701eff76ccd
thread:issue:167526,issue #167526: Missing documentation for CUTLASS backend,stale/low urgency,2,stale/low urgency,1,0,167526,,,,0,5ca655552f8b04fd520f555f9f5f727f9294cea9562b4d7fa0d9549543055345
thread:issue:167558,issue #167558: per_page=1000 doesn't work in hud.pytorch.org,stale/low urgency,2,stale/low urgency,1,0,167558,,,,0,8987d847584f7172ca3c386f1119747f960d8fd3e4f06d88593518ea1f3e104c
thread:issue:167670,issue #167670: PyreFly results are dependent on local python version,stale/low urgency,2,stale/low urgency,1,0,167670,,,,0,fa5d7b85e86002bd4a8507c8e7adf79225a7a8962b375e86c5e648dbfe472176
thread:issue:167721,"issue #167721: Minimal, comprehensive test suite",stale/low urgency,2,stale/low urgency,1,0,167721,,,,0,f2af51c23be3fc97d8dac179f438ddf627ba613c541251d6a4a90a0b67a33a1f
thread:issue:167724,issue #167724: CUDA error: invalid configuration argument for tensor operation,stale/low urgency,2,stale/low urgency,1,0,167724,,,,0,b5459dccbd671a1ba0536ec375448251a26a46616131fbe0670421c8733d3bd7
thread:issue:167746,issue #167746: [xpu] Balance xpu ci test jobs workload,stale/low urgency,2,stale/low urgency,1,0,167746,,,,0,f2ebdeb85b2513a9056e4f183388ee91a832bd21aec42a2fd4cb07b60147a1e5
thread:issue:167754,issue #167754: Getting fallback when using scaled_dot_product_attention() despite CUDA kernels being installed and available,stale/low urgency,2,stale/low urgency,1,0,167754,,,,0,ad140e870ca1f7bdc634d5543815d2bbf294616a2be4dca674d503ff5ce9b3c3
thread:issue:167762,issue #167762: Conan Support,stale/low urgency,2,stale/low urgency,1,0,167762,,,,0,c481a3602e1c6bdabc4fb157214e73b20595fc6229191079728ce7580f390144
thread:issue:167765,issue #167765: [RFC] Extend MemPool support to SymmMem CUDA backend,stale/low urgency,2,stale/low urgency,1,0,167765,,,,0,f205854999c55e149de677f27534cb36098d6e328bfce1c4a02e5e035bfe4315
thread:issue:167825,issue #167825: extra mem copy for value to assign when dispatching `index_put` to `masked_fill` on PrivateUse1 backend.,stale/low urgency,2,stale/low urgency,1,0,167825,,,,0,cb19041fd1d80c58e815db43ddfabbe02674bcdaa698bd29c21bdf723866e808
thread:issue:167895,issue #167895: stable-diffusion 2 gone from huggingface,stale/low urgency,2,stale/low urgency,1,0,167895,,,,0,5e0f8c4e87b31cb297f9321735301331c5c9fa294f5c6039c19683ce5e288326
thread:issue:167950,issue #167950: Insufficient documentation about the batching logic of `torch.linalg.solve`,stale/low urgency,2,stale/low urgency,1,0,167950,,,,0,a4b633a6f02692d1abe5e0c75c1490461a7927baa42d141d1423a52279b4208f
thread:issue:167970,issue #167970: Thread Sanitizer support is clunky right now.,stale/low urgency,2,stale/low urgency,1,0,167970,,,,0,1b5036c5a643f0b820b624489b09ac43cc232481c3741be7f9fcbdd287f82068
thread:issue:167979,"issue #167979: Jupyter kernel restart during eval after first epoch (MNIST, CPU and CUDA, PyTorch 2.10.0.dev20251114+cu128 on Windows 11 + RTX 5080)",stale/low urgency,2,stale/low urgency,1,0,167979,,,,0,09e9017904faec54200340e9a9935d75aad3c65f76efe5873ff253e2c2621fa8
thread:issue:168010,issue #168010: Misleading error message for device mismatch in `conv2d`,stale/low urgency,2,stale/low urgency,1,0,168010,,,,0,c7f91148983aeaa425f2a4e7b139fec19f669d230855064a2fd2198148f57a84
thread:issue:16804,issue #16804: Wrong description of positive class weight in BCEWithLogitsLoss,stale/low urgency,2,stale/low urgency,1,0,16804,,,,0,859aa25b99a7f8e2b0eb594c2d1dae7fec40dc87681dd9cf5670a0bcd180fe89
thread:issue:168128,"issue #168128: [proposal] Make docs webpage layout leaner, reduce number of menu/nav bars",stale/low urgency,2,stale/low urgency,1,0,168128,,,,0,e3ccfb72ce38a923fa1cbbd1bc6df3310a7d766cae2d0db81b7b96e28567ff14
thread:issue:168158,issue #168158: A more efficient alternative to the Tensor.record_stream(...) api is needed,stale/low urgency,2,stale/low urgency,1,0,168158,,,,0,911ddce19cee0b1a0bf9a0ecf8a963a53227dd4fd8c7d541511717ee782bc062
thread:issue:168304,issue #168304: Finish pytree flattening path in create_native_op_schema,stale/low urgency,2,stale/low urgency,1,0,168304,,,,0,1944cd4d4bfc63477a2cd3a2d169be6879f8999e98d9e81370cddf2379b4e15a
thread:issue:168376,issue #168376: `TORCH_SHOW_CPP_STACKTRACES=1` often fails to display all function names,stale/low urgency,2,stale/low urgency,1,0,168376,,,,0,b75621e11c8e0fe62a8253dc70687f3e3cb776446b7d9e7be53e59f210df1f71
thread:issue:16843,issue #16843: Allow to build pytorch for a *specific* architecture,stale/low urgency,2,stale/low urgency,1,0,16843,,,,0,dc9b82c27f96e275f270d185b68e23d53673679fc0cc5c7afc2f1eef8c9fafe2
thread:issue:1686,issue #1686: Feature Request: load_state_dict should take filenames,stale/low urgency,2,stale/low urgency,1,0,1686,,,,0,0afdadab707c0d55a58294329d754fa2305802e8c84f548ce9fe7526292d4541
thread:issue:16873,issue #16873: Allow building C++ custom ops that imports another custom ops,stale/low urgency,2,stale/low urgency,1,0,16873,,,,0,1084dee1b88901686b3d3b8ff4af17a2d832e3959997e87456b168cf57bf69c5
thread:issue:168907,issue #168907: make lint fails on Mac,stale/low urgency,2,stale/low urgency,1,0,168907,,,,0,eadc537900cb10b378fd6d8052cf66eede031784820a6e89961b5b5f1019c12f
thread:issue:168935,issue #168935: [ROCm] enabling compile should not cause signficiant accuracy loss on many important models,stale/low urgency,2,stale/low urgency,1,0,168935,,,,0,eb3b4629e37de805b46fd9f5c046d0010e284dd757dec0280e65fbda7d6b5cd4
thread:issue:16896,issue #16896: `nn.Linear` allows 1d input tensors,stale/low urgency,2,stale/low urgency,1,0,16896,,,,0,0a3ff6e4ee96fc303e37201c924e5351854319a56209719a56c38b641d7f37cb
thread:issue:168967,issue #168967: implement __str__ for Schedular classes,stale/low urgency,2,stale/low urgency,1,0,168967,,,,0,5d0a03c2892b564853d6a2d4d4b60a2fcda7ff34dcad08dcdd5b2503ebf29104
thread:issue:168977,issue #168977: Easier way to correlate DebugMode with FX graph,stale/low urgency,2,stale/low urgency,1,0,168977,,,,0,3cc2fa8ca78014a7a5e2e7b7794dc5a8702a70afac656b0e9eee345006b4bbf8
thread:issue:16899,"issue #16899: Give clearer guidance about multithreading in PyTorch, and how to disable it",stale/low urgency,2,stale/low urgency,1,0,16899,,,,0,85736c4e2f5eba241929c50cd5d42eb42b02ba14829e2e4e8a815118cd13dbfd
thread:issue:16901,issue #16901: Issue with dataloader using pin_memory = True,stale/low urgency,2,stale/low urgency,1,0,16901,,,,0,4dcde8707c5fb6ec1bd8689a7ec440a833b489ab7212d4f156f31f2567f02a9b
thread:issue:169055,issue #169055: `[feature request][discussion][cuBLASLt]` fused backward support for `addmm` with epilogue fusions.,stale/low urgency,2,stale/low urgency,1,0,169055,,,,0,83c3cd560e39c0f52e1754590b6b0c5cec7ba8f3c0a2319756fcc4e66712a2d6
thread:issue:169058,issue #169058: [MPS] Inconsistent overflow behavior when casting Float to Int8 compared to CPU/CUDA (Saturation vs Wraparound),stale/low urgency,2,stale/low urgency,1,0,169058,,,,0,0e853909660bf76ed41f6bd37ddf67ade0aaa5465de34165ad02b9e7d377171d
thread:issue:169072,issue #169072: Properly fix destructive hardware interference warnings,stale/low urgency,2,stale/low urgency,1,0,169072,,,,0,fb46cb9a444e26e00923f5ae5457bb8ee9a283250b62ebfffeaefdb4945376bf
thread:issue:169073,issue #169073: Replace deprecated __args__ type access in typing libraries with proper typing.get_args,stale/low urgency,2,stale/low urgency,1,0,169073,,,,0,80cfadc770e3a003ca959b61fb32fa0478ac7e298b19449f87779c379f1bde12
thread:issue:169078,issue #169078: Custom Non-Integer Indexing,stale/low urgency,2,stale/low urgency,1,0,169078,,,,0,a4780e61f718116d522d9f18c12960df87ab8df8c6d6c026cedcc1be94bb2962
thread:issue:169080,issue #169080: Processing & Tiling Massive Images,stale/low urgency,2,stale/low urgency,1,0,169080,,,,0,1f17214a33f5980f96ff5472b74df0d5ec59aacc90e8fce39c23c9226c5ee750
thread:issue:169119,"issue #169119: `torch.compile(backend=""cudagraphs"", fullgraph=True, cuda)` fails for sparse CSR → COO (`torch.compile does not support sparse Tensors`)",stale/low urgency,2,stale/low urgency,1,0,169119,,,,0,3d83ce8517e54f8ce78cf2bd8c3497f836420646fd5f0ab4609df7802bc6372a
thread:issue:169133,issue #169133: vLLM tlparse sometimes missing compilation_metrics,stale/low urgency,2,stale/low urgency,1,0,169133,,,,0,d176efebc2612f622fc89fac7cdf235433dd0d3ac774b632e753a158448d0f93
thread:issue:169146,issue #169146: bad numerics for variable length attention with cudnn,stale/low urgency,2,stale/low urgency,1,0,169146,,,,0,09a47012eb2c1000ad49ca8e5aa6cc0dc9e0cc14365a31a6a6fb849ba941c0ff
thread:issue:169152,issue #169152: [debug mode] support pipeline parallelism,stale/low urgency,2,stale/low urgency,1,0,169152,,,,0,8987e0eb7c554106dcc4727af54c24c9187b411ed5f100d9c89bd1e6b432e619
thread:issue:169154,issue #169154: AOTIModelPackageLoader occupies too much disk space when launching multiple instances,stale/low urgency,2,stale/low urgency,1,0,169154,,,,0,64ff1379eed7c979e78c4c0822aaea8db36420e65d99e974e8b00692a6f0a288
thread:issue:169163,issue #169163: Silent failure of torch.stack,stale/low urgency,2,stale/low urgency,1,0,169163,,,,0,a69bb8b25717daeb004651c70ae0c5a0be3ceda2c33bbebb5e5669cde9d28c39
thread:issue:169170,issue #169170: Optimizer __repr__ Performance & Stability Improvement,stale/low urgency,2,stale/low urgency,1,0,169170,,,,0,4084e53545a1a2a35f8e4deafe4f68fea132b558ff34c068d0218f2d4b228717
thread:issue:169184,issue #169184: [RFC] Speed-up checkout operations in CI with treeless checkouts,stale/low urgency,2,stale/low urgency,1,0,169184,,,,0,2f3fb679e522f9c6b0751a49ceef23127962fde8b469f933df58ca150168402e
thread:issue:169191,issue #169191: dim broadcast in masked tensor between mask and data.,stale/low urgency,2,stale/low urgency,1,0,169191,,,,0,4d6b1cbd47fc9b32f833a61660496b97ac5b46cfc15430b3202fc4f30d8352e2
thread:issue:169213,issue #169213: CPU bad-free in `torch.nn.utils.rnn.pad_packed_sequence` / `unpack_sequence`,stale/low urgency,2,stale/low urgency,1,0,169213,,,,0,13fc0677161b93fcce2f82f8bf5a12f4a52444a41143b76a3b7a7b67c37ee4ba
thread:issue:169237,issue #169237: Large discrepancy between CPU and GPU norms for identical float32 tensor,stale/low urgency,2,stale/low urgency,1,0,169237,,,,0,ae53369171d1dd3bd4ff273d5d9c8823fa9c1e0c664704db5daef2a4e8d3b356
thread:issue:169266,issue #169266: Translation units with thrust::pair should include thrust/pair.h,stale/low urgency,2,stale/low urgency,1,0,169266,,,,0,1897c10ecc799e2ed61f958ee4af9588aa7923cafc6f9e1c81bbabb7b57e93e1
thread:issue:169290,issue #169290: [MPS][Inductor] LayerNorm/BatchNorm/GroupNorm fail to compile with dynamic=True due to invalid Metal codegen (VLA & Min),stale/low urgency,2,stale/low urgency,1,0,169290,,,,0,c7eba46fb8d3b8625fb5b734dfd8290708f5d19315ee739179970d5968133396
thread:issue:169294,issue #169294: Conv2d CPU performance 2-3x slower on macOS due to missing MKLDNN backend,stale/low urgency,2,stale/low urgency,1,0,169294,,,,0,8991f7182508e3479d70374f5bbd7afb159bd102a28207024bd2c9d6e3e85905
thread:issue:169318,issue #169318: [debug mode] disable debug mode temporarily,stale/low urgency,2,stale/low urgency,1,0,169318,,,,0,33b5343e58df8161fc32561b5848213214aece485ef0b3a51f02c9ea2aaf51e8
thread:issue:169371,issue #169371: C++ Generator API is platform dependent,stale/low urgency,2,stale/low urgency,1,0,169371,,,,0,4f05b6b0332c44de9545b71586a8aa33583b268bb8f7c68b19f8914b8127a034
thread:issue:169378,issue #169378: [rocm triton 3.6] rocm-periodic-dynamo-benchmarks-test : convnextv2_nano.fcmae_ft_in22k_in1k fail_accuracy,stale/low urgency,2,stale/low urgency,1,0,169378,,,,0,e60582b7bedd912d86af952f4ecf7ac6a18f54ad36584d5016b5370832416590
thread:issue:169390,issue #169390: [CI] [CUDA] SIGIOT error with test_cuda.py test_graph_capture_reclaim_4_streams,stale/low urgency,2,stale/low urgency,1,0,169390,,,,0,c18ca5eff1980818f498654ca27149fe4d2444ea2e68f8184748c351f56b23e2
thread:issue:169458,issue #169458: [download.pytorch.org] obselete dangling `xformers` index on CPU only branch,stale/low urgency,2,stale/low urgency,1,0,169458,,,,0,cdefcb0eaf1b285e82f4a57ad609b28965250dbdb0f22ccd08d4eef21f329dca
thread:issue:169469,issue #169469: AOTI infinite recursion from a scalar buffer,stale/low urgency,2,stale/low urgency,1,0,169469,,,,0,0661595555b38c3fa21f32668d7c9cafef305e41409b3eb599330cecc77650d4
thread:issue:16954,issue #16954: torch.multiprocessing.pool.Pool broken,stale/low urgency,2,stale/low urgency,1,0,16954,,,,0,711d451802bc7fbf94d4a4a30307172275f9b7c4620ea3d720c9f2d4409af500
thread:issue:169675,issue #169675: Header-hiding when TORCH_TARGET_VERSION is set is broken for editable torch installs,stale/low urgency,2,stale/low urgency,1,0,169675,,,,0,99c6788b07f86b7a06130e3e440e1cb031f4f664d539fc0581415b1c4fd46140
thread:issue:169738,"issue #169738: [MPS][Inductor] AdaptiveMaxPool{1,2}d produces incorrect results (numerical correctness)",stale/low urgency,2,stale/low urgency,1,0,169738,,,,0,0d0036c777e5ee9ad095ca8b1f44d77b042e10fb5e11e5cfcb517adbc391c557
thread:issue:169763,issue #169763: Multiple torch.load() call on the same BinaryIO,stale/low urgency,2,stale/low urgency,1,0,169763,,,,0,db8ddc4ad53425edb8afabe9e37fdad1896dd23c062c3fc61ef0741671214910
thread:issue:169797,"issue #169797: When augment_with_fx_traces=True but the user has misconfigured the FX config, raise an error",stale/low urgency,2,stale/low urgency,1,0,169797,,,,0,9b01e265c690b05a6fb9f1a45b0e818c7bc5304a22e63ed7b804008353ef0af9
thread:issue:169816,"issue #169816: Cannot load or export TorchVision quantized models on Windows - ""Quantized backend not supported""",stale/low urgency,2,stale/low urgency,1,0,169816,,,,0,f97cfbd813e0c3507ed46da121ba1b27e9b3ca48c494f9bedf30ef55b87de028
thread:issue:169870,issue #169870: Capturing ViewAndMutationMeta for training graphs for PyTorch 2.8,stale/low urgency,2,stale/low urgency,1,0,169870,,,,0,abe96ca680548fed241ca8fed5093c013130c34a9613650319ee216756c72ebe
thread:issue:169893,issue #169893: Investigate which submodules in third_party/ can be omitted from stable header hiding,stale/low urgency,2,stale/low urgency,1,0,169893,,,,0,8f8251db094a88f1f65d3bcac6cad44dc1eb7752ae56bf185e63178d7c1a351c
thread:issue:169909,issue #169909: [CI] cudaErrorInvalidValue in test_std_cuda_kernel_launch_check_error_show_cpp_stacktraces,stale/low urgency,2,stale/low urgency,1,0,169909,,,,0,9a85616cb7f792a5ae41bd38e7b9f205cbd31a217bb69676c983a468193f5880
thread:issue:169955,issue #169955: [Bug] Potential command injection risk (shell=True) in subprocess,stale/low urgency,2,stale/low urgency,1,0,169955,,,,0,a2410e12c167314357becba87d31b43d790f6c1131f4d97ad39000230e210b50
thread:issue:169956,issue #169956: `TestLinalgCUDA.test__int_mm` fails false/false case on A100,stale/low urgency,2,stale/low urgency,1,0,169956,,,,0,fd520385218ca6b52d0ca73d7ab5afdabbaa8b74923888a1fde8d1d0b41ec7dc
thread:issue:170175,"issue #170175: [MPS][Inconsistency] nonzero() incorrectly preserves non-contiguous strides from sliced inputs, causing .view() crash on CPU/CUDA,but pass on MPS",stale/low urgency,2,stale/low urgency,1,0,170175,,,,0,44a20109448b693f6dbe40d10806bfc80d4d03f7fdd8d4f94a94c94d7fe81a0f
thread:issue:170266,issue #170266: pytorchbot does not react to Copilot PRs,stale/low urgency,2,stale/low urgency,1,0,170266,,,,0,5cdd5a757b2738ee23ef8036655422a2f6e7229d810aed08945283dc76babfbe
thread:issue:170284,issue #170284: selective activation checkpoint compatibility to async activation offload,stale/low urgency,2,stale/low urgency,1,0,170284,,,,0,58abdf4b720d6ef7f6815f515cb98bd2db66a2c716b81b986f2cfd1dc7d0daa0
thread:issue:170345,issue #170345: [vllm] Pytorch failure lazy_imports.py is failing on torch._inductor.async_compile for release 2.9.0 and 2.10.0 rc1,stale/low urgency,2,stale/low urgency,1,0,170345,,,,0,c4a6db93e8fee2ba1103a2fe3a5a14e6c50222501d2c7d545980f1822dfc1fc9
thread:issue:170350,issue #170350: [feature request] Checked version of indexing/embedding ops,stale/low urgency,2,stale/low urgency,1,0,170350,,,,0,17ec5cd66aa2a41215d7167f708382fa54e92b99d571a8114ac6a65b8928f0b7
thread:issue:170365,issue #170365: [CI][CUDA][GH100] LinAlg Unit Test Failures,stale/low urgency,2,stale/low urgency,1,0,170365,,,,0,0fd32fe45b84cb4163e460083bcb21b41ac982733506d38dc28973093f948b71
thread:issue:170382,issue #170382: Feedback about torch.nn.functional.interpolate,stale/low urgency,2,stale/low urgency,1,0,170382,,,,0,25afef6ea067d25a2421cdb22d8fbe8418fcb675440288c96a62a604d8a0a8d9
thread:issue:170402,issue #170402: Write more test cases for Conv2d to check for 0 steps,stale/low urgency,2,stale/low urgency,1,0,170402,,,,0,5b45bb4af132eb2c96a0ecd41cba0024089465074781e1b199a0015a94e45b9a
thread:issue:17041,issue #17041: Errors running distributed example,stale/low urgency,2,stale/low urgency,1,0,17041,,,,0,690d91ee0b8a233dcdb0a17488b4e32ae310d06318740e032496652bfaf85adb
thread:issue:170458,issue #170458: must_in_dispatch takes a long time in torch registered ops,stale/low urgency,2,stale/low urgency,1,0,170458,,,,0,36ee6ad976974a4644fc0ece8e65be39da4e94f095218cf02804607684bbc4fb
thread:issue:170495,issue #170495: vLLM Blackwell Fusion and Compile Tests failing PyTorch 2.10,stale/low urgency,2,stale/low urgency,1,0,170495,,,,0,8088357727f63fabe862cd9b46796663981905ee3ff3b68c8ee96f87910cc371
thread:issue:170536,issue #170536: accept int padding value for F.pad,stale/low urgency,2,stale/low urgency,1,0,170536,,,,0,4db60931194780e4fdd53e5d625f843ee2e24913b536a8c94e4812a5d9720173
thread:issue:170542,issue #170542: BuildExtension(no_python_abi_suffix=True).get_ext_filename result is incorrect when ABI part is not present in the filename.,stale/low urgency,2,stale/low urgency,1,0,170542,,,,0,2e352527621c3b4ce26dee2741a9561d1dac25ea4b4bb738976de550a08b3350
thread:issue:170563,issue #170563: vLLM v1/distributed/test_eagle_dp.py Tests failing on PyTorch 2.10,stale/low urgency,2,stale/low urgency,1,0,170563,,,,0,80607b9ddd673ced478d49e0a20b6eb6ea4469821885b709a5b9fb6877c0cef0
thread:issue:170573,"issue #170573: Silent Heap Buffer Overflow in standard ops (add, pinv) via implicit view resizing",stale/low urgency,2,stale/low urgency,1,0,170573,,,,0,20f60e48c6ee30f544d383eb4626d9769afc00349f4a884751b4b6f868aea027
thread:issue:170674,issue #170674: Flex attention cannot support page size/block mask size != 128 for prefill.,stale/low urgency,2,stale/low urgency,1,0,170674,,,,0,87698c899451dbe36cb9c78507dcb36bcefc56b2f52ce7fd323a69c3e11d931e
thread:issue:170691,issue #170691: [Feature Request] Add StateSpaceModel to torch.nn,stale/low urgency,2,stale/low urgency,1,0,170691,,,,0,8717413067ed5047f68086d65d26b57a14f89586a7f26ccbd1b7ad3fab99224a
thread:issue:170694,issue #170694: Proposal to simplify `DeviceThreadHandlePool`,stale/low urgency,2,stale/low urgency,1,0,170694,,,,0,86a487f522dcd72e5b3f7b3c09437a1c6f04d62025c791183d3fc72324c8924d
thread:issue:170851,issue #170851: Add forward_by_train and forward_by_eval convenience methods to nn.Module,stale/low urgency,2,stale/low urgency,1,0,170851,,,,0,4d9a4db8b29da0e8d4933ebffa7c04d4728d18f68dd8145f97799d034350c35f
thread:issue:170936,issue #170936: NaN-checking anomaly mode should work with DTensor,stale/low urgency,2,stale/low urgency,1,0,170936,,,,0,3b1243014f79c2aaff42e2b39299b2cd41daf6c5c47a3abd47db6d8f5a016bce
thread:issue:171039,"issue #171039: When using torch.multiprocessing for CUDA IPC, memory cannot be released.",stale/low urgency,2,stale/low urgency,1,0,171039,,,,0,3d66990c2ae01a155874ab3f74c9480acedf2adc2c5db0b668f710132480ddbd
thread:issue:171100,issue #171100: Feedback about torch.bernoulli,stale/low urgency,2,stale/low urgency,1,0,171100,,,,0,9e49cdb6db5dfd57a83e34536ba81aa480af6ea459eb0623e030029bf355cbc7
thread:issue:17113,issue #17113: Distributed training jobs do not terminate properly if there is a crash,stale/low urgency,2,stale/low urgency,1,0,17113,,,,0,71e1b3e4c05a50d0795cbe991724663285780d0f4fc94b15114186279d1467ed
thread:issue:171179,"issue #171179: NCCL error: /pytorch/torch/csrc/distributed/c10d/NCCLUtils.hpp:268, NCCL version 2.22.3 ncclSystemError: System call (e.g. socket, malloc) or external library call failed or device error. Last error: socketStartConnect: Connect to IP<51423> failed",stale/low urgency,2,stale/low urgency,1,0,171179,,,,0,0e81f0674be4cb47e209f0c0bb1a152b081c045c6dc5b40eacd245533a5d7b68
thread:issue:171184,issue #171184: No hash in CachingAutotuner meta information for some kernels,stale/low urgency,2,stale/low urgency,1,0,171184,,,,0,4de216de65ec6a14ca4efc15f05c8f6c8489b20cb0bdcdc52fd152974792aa06
thread:issue:171263,"issue #171263: [CUDA Graphs] Capture failure leaves generator state corrupted, causing ""Offset increment outside graph capture encountered unexpectedly"" on subsequent RNG operations",stale/low urgency,2,stale/low urgency,1,0,171263,,,,0,4a3afa76cd6572a256e42cf9be0e665e41afca553b8ab0748a5d6439d044ae8d
thread:issue:171286,issue #171286: Memory Viz Y-Axis Uses SI Units (GB) Instead of Binary Units (GiB),stale/low urgency,2,stale/low urgency,1,0,171286,,,,0,634054bbe9a6651ae6ea1b821a350e4cfdafda83f49ae143ec553cdab2cd26ec
thread:issue:171323,issue #171323: Segmentation fault when calling `torch.tensor()` after importing xgboost,stale/low urgency,2,stale/low urgency,1,0,171323,,,,0,43d4add1ae1de9015728d7d149577386e60221709a86a3f8ebe3354661075f0f
thread:issue:171401,issue #171401: Feedback about Flatten,stale/low urgency,2,stale/low urgency,1,0,171401,,,,0,237def58e208dc04a59b40595d1891d29e491578da4ac74d5fb3942f29dd03b3
thread:issue:17150,issue #17150: Conv2d layers should accept 4-tuple for padding argument,stale/low urgency,2,stale/low urgency,1,0,17150,,,,0,047bfccf55bba41a39ff010b04e1b4aaa924b68e4d88b702ba34c5339ff0ecc2
thread:issue:171534,issue #171534: Add proper FMA support for Pallas backend,stale/low urgency,2,stale/low urgency,1,0,171534,,,,0,690e73e3eb40f679e22f09b4241b3d994b76f83b136cc4b797b375e2a5605fc2
thread:issue:171536,issue #171536: `torch.nn.PixelUnshuffle` throws RuntimeError: Stride calculation overflowed on CUDA with empty input and large downscale factor (CPU succeeds),stale/low urgency,2,stale/low urgency,1,0,171536,,,,0,d76f828c006e9412aeed64de6ace2389a50827618691b6ded7198d5fbe032467
thread:issue:17169,issue #17169: Global Second Order Pooling,stale/low urgency,2,stale/low urgency,1,0,17169,,,,0,94485ee003a39c04f3738a0740c302fd427a4f586b5e0b0856c0df1075bd767f
thread:issue:171697,"issue #171697: When the pinned_use_background_threads option is first set to True and then set to False, the background threads may not stop",stale/low urgency,2,stale/low urgency,1,0,171697,,,,0,dec2c2e452148536efc46e648018c540609c1a2af74ffb30ad5f135d38800b68
thread:issue:171771,issue #171771: [RFC] Optimize CUDA Allocator's Synchronization Behavior During OOM,stale/low urgency,2,stale/low urgency,1,0,171771,,,,0,d0baef9f78f6d5f85c3a3bd5373dc322b098830a1e2141b5c9de49189659c0d6
thread:issue:171956,issue #171956: [Inductor] Eliminate triton-related hardcode,stale/low urgency,2,stale/low urgency,1,0,171956,,,,0,c5a494c29df74234a81057fe14feb14abf595a157c90e863b2b7e26c3fd00418
thread:issue:172085,issue #172085: Many test failures when GPU does not support BF16,stale/low urgency,2,stale/low urgency,1,0,172085,,,,0,3cfef809441136b6d9966b2f622780ddc412f6eea4772eae81701f377c083244
thread:issue:17249,issue #17249: Proposal: Add __tensor_wrap__ method similar to numpy __array_wrap__,stale/low urgency,2,stale/low urgency,1,0,17249,,,,0,fb923efbbfcb84a3bc591899991ba631a51a0ce99225801c60defb79a13aee4a
thread:issue:17257,issue #17257: Generated `__init__.pyi` contains invalid default values,stale/low urgency,2,stale/low urgency,1,0,17257,,,,0,6c5a59437d36b7b00e52b0b434dc31c851ca8524ef9aea2d1f323c89f6cbb3b0
thread:issue:17268,issue #17268: Generic object to tensor dispatching,stale/low urgency,2,stale/low urgency,1,0,17268,,,,0,dc6ee9df21f4af2bd7529c61ee1d316bf1e02b922327d29bb57cc6dfd3685e6f
thread:issue:17286,issue #17286: Use standard docker image for XLA build,stale/low urgency,2,stale/low urgency,1,0,17286,,,,0,fced46957f2d67d2d2a85e82ac6cef40de5fe0545460f0de2b1bd0729e1d2f70
thread:issue:17313,issue #17313: Implement Adaptive Input Representations for Neural Language Modeling,stale/low urgency,2,stale/low urgency,1,0,17313,,,,0,cf31875c61df09e08fcec56a635414501a132d2d376a2a8422a18ace7bea2e50
thread:issue:17350,"issue #17350: torch.nn.CrossEntropyLoss with ""reduction"" sum/mean is not deterministic on segmentation outputs / labels",stale/low urgency,2,stale/low urgency,1,0,17350,,,,0,5721048d6239464189357373913f7b879bbcd5d5fd7eb96c6b96bc09300f1884
thread:issue:17354,issue #17354: testConvnetBenchmarks intermittently segfaults,stale/low urgency,2,stale/low urgency,1,0,17354,,,,0,17dd84ac5ba9a827d4189c90d1cca5a42c919d0efbbcd27115db6e17a8e5094a
thread:issue:1736,issue #1736: Expose optimizer options as attributes when there's a single param group,stale/low urgency,2,stale/low urgency,1,0,1736,,,,0,9b2b2544fbf581c62e3bc16f362bc795d27870729fdb7bc2288aa8be54f52f80
thread:issue:17437,issue #17437: Automatic aggregation of a mix of sparse and dense gradients is not supported yet,stale/low urgency,2,stale/low urgency,1,0,17437,,,,0,705174f8beca2796dc6095a8edd027ed63f2f41fce70c4c5561eb7ffe21d0217
thread:issue:17458,issue #17458: Overflow in fbgemm,stale/low urgency,2,stale/low urgency,1,0,17458,,,,0,0be0ddf4d4d8a7c8e012972e4b65ebce53de0d3a9377021c43bf71336d0e66b8
thread:issue:1747,issue #1747: In-place bernoulli_ has more functionality than torch.bernoulli with output parameter,stale/low urgency,2,stale/low urgency,1,0,1747,,,,0,bb9c26f4b1c73d6cdf7a98831745ca9ee8a15abe4ecf97f58122b317abf6c50a
thread:issue:17495,issue #17495: RuntimeError: storage_.IsType<T>() ASSERT FAILED,stale/low urgency,2,stale/low urgency,1,0,17495,,,,0,cdfd87cf0954fb284c453d7689f55eb3285885ed40aad793086b7d66a145b9df
thread:issue:174986,issue #174986: `torch.istft` returns an unhelpful internal error,stale/low urgency,2,stale/low urgency,1,0,174986,,,,0,b8b293d54d2f00fbb7530acd1a09fc75512edbafbd480c7833b2458e5cbc0a60
thread:issue:1750,issue #1750: Add support for colors (and maybe other attributes) to NVTX API,stale/low urgency,2,stale/low urgency,1,0,1750,,,,0,a126fe1127cb4f6d282994fbc3a6b5dd7cf60200533cf3a68c7014340cc0ba78
thread:issue:17525,issue #17525: cmake fails with -DBUILD_PYTHON=OFF -DUSE_NNPACK=OFF,stale/low urgency,2,stale/low urgency,1,0,17525,,,,0,237c33557a63148a1e189c7bb007b4dea9a009543062b7a817cf04163d8728b6
thread:issue:17538,issue #17538: ONNX->TensorRT parser library duplicated,stale/low urgency,2,stale/low urgency,1,0,17538,,,,0,adba4eae0282370e0ca37d44eb17aefd4fe6ef4f633e1da8ccaf41c2e1456c60
thread:issue:17554,issue #17554: Add ASSERT for calling accessor on a GPU tensor,stale/low urgency,2,stale/low urgency,1,0,17554,,,,0,a3ff251c0540859567b4363404f690c406ba25ebd4ef184c59377ce89c39e0c0
thread:issue:17612,issue #17612: The documentation to Module._version isn't visible.,stale/low urgency,2,stale/low urgency,1,0,17612,,,,0,35e1d292b44001c28c4fa28a727495851fbcbf716fe5ddda8600bf0a4750ec41
thread:issue:17614,issue #17614: Document whether it is possible to train TorchScript modules,stale/low urgency,2,stale/low urgency,1,0,17614,,,,0,2d7bb6fb50a77982ee2d09af4c243b063f2abb75d656dcc9c0654f7b28a750ed
thread:issue:17632,issue #17632: Caffe2 C++ script for classification/object_detection with CMakeLists.txt,stale/low urgency,2,stale/low urgency,1,0,17632,,,,0,8f02bf2f7c80f2414352192c7ce24d215763e551d4b664022ffe17d498c9b874
thread:issue:17634,issue #17634: [Caffe2] android app crashed while linked with libcaffe2_detectron_ops.so,stale/low urgency,2,stale/low urgency,1,0,17634,,,,0,e2b3e88e7fcf79179d8c13edc6b54364adce7396a9020a0fed50c74a7ca8b5a2
thread:issue:17644,"issue #17644: Not depend on third_party submodules, but self-built libraries?",stale/low urgency,2,stale/low urgency,1,0,17644,,,,0,4853b036cde2f99a46847b3aa5c41bb15b1e72946e5b92c095819965f06728bd
thread:issue:17655,issue #17655: TracedModule 'to' attribute doesn't work for tensors created on forward.,stale/low urgency,2,stale/low urgency,1,0,17655,,,,0,ac807d96af6317741fae57fa0670850a0803266d13c7f57acb98b76cec7b33bb
thread:issue:17658,issue #17658: C++ API: Crash in cudnnDestroy() when deconstructing,stale/low urgency,2,stale/low urgency,1,0,17658,,,,0,59196e02376f3f5c3e6d74b3d93daa797618aa92edc7e7a47dc28caf128f0b2e
thread:issue:17666,issue #17666: Building pytorch on ARM failed,stale/low urgency,2,stale/low urgency,1,0,17666,,,,0,428a22f8b73b0f00299af95d98ffb7173d54ea5398d1646f0c7dfeb140c511c6
thread:issue:17688,issue #17688: Improve save() method in torch.jit.ScriptModule,stale/low urgency,2,stale/low urgency,1,0,17688,,,,0,6cc343d92cffc752d99fa4e53f07a2c67b199e395d259913a42792041224c732
thread:issue:17703,issue #17703: Training hangs when using DistributedDataParallel in two pod on two nodes,stale/low urgency,2,stale/low urgency,1,0,17703,,,,0,9ab55dc3c4a688fd4b90bacd6de2a8acc685348635d56b4dfe02c517adafd03c
thread:issue:177082,issue #177082: Stable Linux binaries of LibTorch have broken soname and versioning in libgomp,stale/low urgency,2,stale/low urgency,1,0,177082,,,,0,43759d2ceda9e2cddcb15b28b8e29d0ad0b9c062ca24ef3cb2c75fd1df1f229a
thread:issue:17712,issue #17712: Can we to build Caffe2 custom ops independently from a Caffe2 build ?,stale/low urgency,2,stale/low urgency,1,0,17712,,,,0,b921e2d32485095dfeabfc136356aaa964be8ae21426b8f87ca745e438318195
thread:issue:17745,"issue #17745: distributed data parallel, gloo backend works, but nccl deadlock",stale/low urgency,2,stale/low urgency,1,0,17745,,,,0,7bf14451b2f94077b49d11f8fbd2e48bdf3715f04cb7e9bd5982669e4f06392c
thread:issue:17796,issue #17796: [Caffe2] install error,stale/low urgency,2,stale/low urgency,1,0,17796,,,,0,fcfef7c57fe94ca8e77de011775262dda76f1a4bd457f7acdf7d3cee7844d65c
thread:issue:17797,issue #17797: Non-coherent result for C++ with multithreading and GPU,stale/low urgency,2,stale/low urgency,1,0,17797,,,,0,4e97bdba2a21f6793c1020e27c16cfe57f70a784d98263b72b35d3e3dd4c4055
thread:issue:17798,issue #17798: Feature Request: deterministic CUDA torch.nn.CTCLoss,stale/low urgency,2,stale/low urgency,1,0,17798,,,,0,e05ccfcfa6bf1f85f3611434ed145eddf50db8951661ce9038b771810d1a7098
thread:issue:17800,issue #17800: [Caffe2] cudnn mismatch,stale/low urgency,2,stale/low urgency,1,0,17800,,,,0,28a64f84771384a753afa520cd4d49eb999abe534ec2feb9bb6710d59da9424f
thread:issue:17835,issue #17835: PyPy support,stale/low urgency,2,stale/low urgency,1,0,17835,,,,0,72752c099a191d23e9e96784b2fd817916e88dff9d5f86043e9168770daafbee
thread:issue:17849,issue #17849: Tensor::options() returns false for requires_grad when it is true,stale/low urgency,2,stale/low urgency,1,0,17849,,,,0,32b2a396c31969bf91f9583b787e4c30852a067952e6740dc8b13124d7c44853
thread:issue:17850,issue #17850: Build fails for caffe2/CMakeFiles/caffe2.dir/__/aten/src/ATen/native/mkldnn/Conv.cpp.o; possibly MKL-DNN issue,stale/low urgency,2,stale/low urgency,1,0,17850,,,,0,c747b1a11f10c640445fcd4cb057aa8e6431180469f69727ee81366fb8e1f898
thread:issue:17872,issue #17872: Linking Caffe2 with sequential MKL but still calling multi-threaded mkl,stale/low urgency,2,stale/low urgency,1,0,17872,,,,0,188e741dc3d2d59233411f3135abe4cbaec295295c238cd328f9322673728296
thread:issue:17879,issue #17879: C++ nn::Sequential push_back() copies module if the module is concrete type,stale/low urgency,2,stale/low urgency,1,0,17879,,,,0,657a8ddb3934c11938ecf0ba55bf1e14146c457cf29cc27cfd8eb4315822fa99
thread:issue:17887,issue #17887: Caffe2 building failure,stale/low urgency,2,stale/low urgency,1,0,17887,,,,0,99a36d807f223737f3108c0bf96bd669a36d6ab92fb2008b8a1cc901b23ab406
thread:issue:17907,"issue #17907: [caffe2] resnet_trainer example fails with MNIST dataset, when parameter num_channels=1 is provided",stale/low urgency,2,stale/low urgency,1,0,17907,,,,0,baf8f7db451c642d0f9ec2afe950e880ec2f19ac40835cb4a0d22bf9d53e661d
thread:issue:17920,issue #17920: Build compact libtorch from source with cmake,stale/low urgency,2,stale/low urgency,1,0,17920,,,,0,668047e74daaaf07888c87dcd8c2f1150be55f7656af47f6f27acfdc9bb2f96c
thread:issue:17936,issue #17936: [caffe2] resize_op_test.py::TestResize::test_nearest FAILED,stale/low urgency,2,stale/low urgency,1,0,17936,,,,0,ef4737e543653e3d9efaa526ee95e8981ca363bc2ee3d48f1a173205091268bb
thread:issue:1794,issue #1794: Feature request: reverse_padded_sequence,stale/low urgency,2,stale/low urgency,1,0,1794,,,,0,f439967008a09a8f0f6b7a17b72cd67c56b4762725f3219a0c49ded5e44fc598
thread:issue:17970,issue #17970: Binary not operator causes crash when Jit module is executed on different device,stale/low urgency,2,stale/low urgency,1,0,17970,,,,0,458d087b1cd8d02c91a7837d7bc96c9095880f8be61e82db4edcdb154731e166
thread:issue:17984,"issue #17984: ""unknown builtin op"" error with static library",stale/low urgency,2,stale/low urgency,1,0,17984,,,,0,81e5141029e8e944fd2d1320131565cd4a4eab95b107f9d6c07e61afb70470d2
thread:issue:18000,issue #18000: Unsupported type of tensor c10::half when running resnet50_trainer.py in Caffe2,stale/low urgency,2,stale/low urgency,1,0,18000,,,,0,15c0871b68383783e35f72d69071a69f7d2553c43d4b20af93a4f8386911e497
thread:issue:18005,issue #18005: Support 'bytes' type in torchscript,stale/low urgency,2,stale/low urgency,1,0,18005,,,,0,c4215a98bf5b48f67497c6aaa59f8c44022519cabdafed796fa88d9c05c4720f
thread:issue:18011,issue #18011: caffe2 Segmentation fault (core dumped),stale/low urgency,2,stale/low urgency,1,0,18011,,,,0,cdd6a7c2df32a58a064cf5c280568a9527fb51bca04eb907b4390b83037f3673
thread:issue:18094,issue #18094: [JIT Script] Need support on distributions.Categorical.,stale/low urgency,2,stale/low urgency,1,0,18094,,,,0,4f6e8408f187468fea7c25a7f4f0d442563b53db8dc1aa9792121c2b29b9f819
thread:issue:18101,issue #18101: Android NNAPI support of Caffe2,stale/low urgency,2,stale/low urgency,1,0,18101,,,,0,5119867f3bb332115f859343c2334d3ae1a16174e7ec680a86a55839220f3e16
thread:issue:18106,issue #18106: copy.copy not working for ScriptModule,stale/low urgency,2,stale/low urgency,1,0,18106,,,,0,686400529548d44cfe9f2bb4d99792350762f419e1015e0fd11976a5e0826f3a
thread:issue:18157,issue #18157: JIT does not batch linear layers in an ensemble,stale/low urgency,2,stale/low urgency,1,0,18157,,,,0,86e48c9d771834b069654d91e501689186c52e0c5f85df4b8b8ec73774674c82
thread:issue:18206,issue #18206: Rename ignore_index to ignore_target in CrossEntropyLoss,stale/low urgency,2,stale/low urgency,1,0,18206,,,,0,7efa98208022210f50997a1b811f60ce20d5e811485172bcc6a5df739d7fc52c
thread:issue:18212,issue #18212: [JIT] bitwise NOT does not handle tensor shapes correctly under JIT,stale/low urgency,2,stale/low urgency,1,0,18212,,,,0,d42eb4ada9ad7602819644b0513b3515d07f3688ca45e1333735a4241450a4a3
thread:issue:18220,issue #18220: Context Manager that disables training mode with in a nn.Module.,stale/low urgency,2,stale/low urgency,1,0,18220,,,,0,a156374132592ac31ba4b094330879ff8104c3d924678d41498468f0c82aa528
thread:issue:18229,issue #18229: Add support for tuple type deduction in C++ custom operators,stale/low urgency,2,stale/low urgency,1,0,18229,,,,0,2d39ef7b5846508213fd885b49dbf101ea7cabd7f48971bd896c3e09ee9f1c2b
thread:issue:18273,issue #18273: [caffe2] Broken Operators Catalog,stale/low urgency,2,stale/low urgency,1,0,18273,,,,0,e1582372e47e044e3206f78ef63f9a708440a18d8633285cc57503db4a223783
thread:issue:18283,issue #18283: Doubly freed pointer in torch::cat error handling when called via pybind11.,stale/low urgency,2,stale/low urgency,1,0,18283,,,,0,5a1d6da2e61137c984607ce141e1eb74d887c758f4f5e150cf92637b68af579d
thread:issue:18315,issue #18315: [FR] support default size (scalar) in torch.randint,stale/low urgency,2,stale/low urgency,1,0,18315,,,,0,e5ced67aa400de7711bd456cbb9a5eaf248e6030c58a0294423da58971c9825b
thread:issue:18317,issue #18317: Add a RandomBatchSampler ?,stale/low urgency,2,stale/low urgency,1,0,18317,,,,0,be90539ad034ce47fa856c113b3f40712931f3f228e52c3cb7589aff83666d19
thread:issue:18351,issue #18351: Port SpatialConvolutionMM and VolumetricConvolutionMM to ATen,stale/low urgency,2,stale/low urgency,1,0,18351,,,,0,1acd6deafd2c193ef9efd334f73f6f1156822c43955888456edb4924b473d80d
thread:issue:18414,issue #18414: Add layer-wise adaptive rate scaling (LARS) optimizer,stale/low urgency,2,stale/low urgency,1,0,18414,,,,0,70a166c664c500b60220416d4da5f6d633f35c1e6bb3be9d0fcaf02d5c185a8e
thread:issue:18434,issue #18434: improve jit error message for legacy constructor,stale/low urgency,2,stale/low urgency,1,0,18434,,,,0,9929cf03bf9c3efc3d1d4adefcbe0594d25bdb603539524f70dfba6d7a9274c2
thread:issue:18436,issue #18436: UnpicklingError when trying to load multiple objects from a file,stale/low urgency,2,stale/low urgency,1,0,18436,,,,0,4f585251eee32e96cb90b0f5d907325360c40df0e58be18a5a6caaf5d2a6173f
thread:issue:1861,issue #1861: Factorized Output Layer,stale/low urgency,2,stale/low urgency,1,0,1861,,,,0,86539f4c0c19c5f45ba6a0703cf793c4efb42c6d1b8533b41d55ec24729c423e
thread:issue:18630,issue #18630: documentation for C++ / libtorch autograd profiler,stale/low urgency,2,stale/low urgency,1,0,18630,,,,0,f814698da52da9be85a635941dea7b2e2a8a6c025adf6e01079fc1fcefc75f90
thread:issue:18643,issue #18643: Memory not being deallocated in backward(),stale/low urgency,2,stale/low urgency,1,0,18643,,,,0,754c2fa7b2b7964a79b602d0b287eacc537abb7bb11962293605fb78f865c6fb
thread:issue:18658,issue #18658: [FR] add CPU information in collect_env.py,stale/low urgency,2,stale/low urgency,1,0,18658,,,,0,26a78c927c4a727e06f1acb7afc8cfba397051bec7693ef97a8b9362cb1e5aa1
thread:issue:18677,issue #18677: How to compile/install caffe2 with cuda 9.0?,stale/low urgency,2,stale/low urgency,1,0,18677,,,,0,2123bcffbcbc9927806caa2b0ade4bce5b3b3926bef38094e33212acccdf6827
thread:issue:18692,issue #18692: Add build tests for feature environment vars,stale/low urgency,2,stale/low urgency,1,0,18692,,,,0,127d46c6d11a43afd326c0a4bb066f1b3913b75abc363e5ce24aa105bb57b037
thread:issue:18728,issue #18728: MaxPool with n-dimensional tensors,stale/low urgency,2,stale/low urgency,1,0,18728,,,,0,7118673662fdaada483ceafcb650d94a3f88a0cdd7db11008ad8f53e7e11768d
thread:issue:18730,issue #18730: Numerical instability KL divergence RelaxedOneHotCategorical,stale/low urgency,2,stale/low urgency,1,0,18730,,,,0,9b6175c367bf648f64e54a1bbc6ac41145b1fb08a6af290494cf34087329b9dc
thread:issue:18776,issue #18776: Value of torch.backends.cudnn.benchmark Baked into JIT-Traced Modules ( 150x slowdown on ConvTranspose2d() ) [jit] [libtorch] [cudnn],stale/low urgency,2,stale/low urgency,1,0,18776,,,,0,bc47c231ffe3206b12cd8a2b7d060b56e6d86335380b790546945c23bf94b698
thread:issue:18796,issue #18796: [Feature Request] Flattened indices option for max pooling,stale/low urgency,2,stale/low urgency,1,0,18796,,,,0,3fb02d9b9899599aa18d8246466c3191a2946fe40f4df2dc7ffac6b1472c3ad8
thread:issue:18805,issue #18805: Running custom operator tests manually is too difficult,stale/low urgency,2,stale/low urgency,1,0,18805,,,,0,da40a6d16aad0b9b7fd545f11bf9bad13c02cb03355ec4d0acf60f3910999676
thread:issue:18849,issue #18849: More efficient STFT on CUDA,stale/low urgency,2,stale/low urgency,1,0,18849,,,,0,f6590dc9f0869846f462e9ffc7e90682029cd09f93f00b3bc098ddd4466a80c5
thread:issue:18850,issue #18850: Allow tracing of models which output `None`,stale/low urgency,2,stale/low urgency,1,0,18850,,,,0,7b5a2a1e2974182d54195f7c3374b4a1336eff023eb4e4a42fab62229052086c
thread:issue:18920,issue #18920: [c10d] CUDA tests for C++ reducer,stale/low urgency,2,stale/low urgency,1,0,18920,,,,0,23c6d31198f61a9da3034802d58a7567b253cb61737a4933e550405b1238c3ad
thread:issue:18933,issue #18933: Completion of error handling,stale/low urgency,2,stale/low urgency,1,0,18933,,,,0,77fc3da1fa1e96fd886f3ce0070802e4e19597a761a40166802b7dab769f2a51
thread:issue:18944,issue #18944: Caffe2 on google colab:,stale/low urgency,2,stale/low urgency,1,0,18944,,,,0,3230bf1d010063ce11b8b035d522f3dcc4edbb04fc59cfd1ec025f8f3554c2a5
thread:issue:18987,issue #18987: Performance issue of unique on CPU,stale/low urgency,2,stale/low urgency,1,0,18987,,,,0,9990da702888c0e4ee62f4e91a099d1acb34c4d0f23a3746f870a9b45551618f
thread:issue:18993,issue #18993: ninja: build stopped: subcommand failed.,stale/low urgency,2,stale/low urgency,1,0,18993,,,,0,8e08ca86d150545eb909a39f483eefd6f814aff091329fe513b68d358cd1a5f3
thread:issue:18998,issue #18998: torch.from_PIL() Request ?,stale/low urgency,2,stale/low urgency,1,0,18998,,,,0,7cc32565fa18bca0943412382b970576b9f4e744d0777e94a5e8393f9f58bd2e
thread:issue:19016,issue #19016: Clean up and consolidate DDP tests,stale/low urgency,2,stale/low urgency,1,0,19016,,,,0,9f9eb61d059bbfd833c0e84f41c857b65affecc9d8c07e51acd3072e8149b231
thread:issue:19107,issue #19107: pytorch/caffe2/onnx/backend.cc:1668:57: error: invalid conversion from ‘google::protobuf::int32 {aka int}’ to ‘onnx_torch::TensorProto::DataType {aka onnx_torch::TensorProto_DataType}’ [-fpermissive],stale/low urgency,2,stale/low urgency,1,0,19107,,,,0,7a39e7f87a6ea5d2a1e33774cbde9d1574bf2a1b6f0b3ab561ddf4bdc7684784
thread:issue:19117,issue #19117: [jit] Can't `torch.jit.script` a lambda,stale/low urgency,2,stale/low urgency,1,0,19117,,,,0,27fb8dea1c827f4289922046bcafed6e5dae0b1039478efa8f78a909b4244697
thread:issue:19125,issue #19125: Multi-gpu via torch::nn::parallel::data_parallel,stale/low urgency,2,stale/low urgency,1,0,19125,,,,0,50812342f833e546c6ae2296dc308627559f2ace1b3b17383aea1d083284ebda
thread:issue:19126,issue #19126: weight_norm doesn't support eta and returns nan for zero weights,stale/low urgency,2,stale/low urgency,1,0,19126,,,,0,2a5cec0720518ecae3ed40e31afdeffd0b02602216c40f61f5b864feeb9e385a
thread:issue:19143,issue #19143: Support memoryview() method on torch.Tensor,stale/low urgency,2,stale/low urgency,1,0,19143,,,,0,2c656247847609bba03efea72a26b6eacb74ea656b0b9eaa4e5c3e816ed71d0a
thread:issue:19150,issue #19150: Improve unit test coverage of torch.unique,stale/low urgency,2,stale/low urgency,1,0,19150,,,,0,29eba1ca6785fcbdac8b050569ec3a7ccae861be92078907d0d31176741ce140
thread:issue:19160,issue #19160: Suggest model.eval() in torch.no_grad (and vice versa),stale/low urgency,2,stale/low urgency,1,0,19160,,,,0,68179ae037a306ab271a5022b9160041769fe15d28e36ec3703d2f6c4572085f
thread:issue:19163,issue #19163: LayerNorm is very slow (almost frozen) in CPU of multiprocessing,stale/low urgency,2,stale/low urgency,1,0,19163,,,,0,c14603c3286fb5caabad0e2c93701c0a5714f9e6f576fa7c484ca9a9cd008a27
thread:issue:19172,issue #19172: [Feature Request] Common constants in the torch.* namespace,stale/low urgency,2,stale/low urgency,1,0,19172,,,,0,b2c9703e2adbb6a2b61bc5d73357b10a2007a05c5e89916b6a98334c50702a29
thread:issue:19229,issue #19229: Different behavior of torch.nn.MultiMarginLoss on CPU/GPU Tensors,stale/low urgency,2,stale/low urgency,1,0,19229,,,,0,18204e3c104aca089b3a22fab434463668233391da506b1564948ec4e730a13b
thread:issue:19244,issue #19244: [Caffe2] Retraining saved model,stale/low urgency,2,stale/low urgency,1,0,19244,,,,0,a2704880af943d28b2e0bfb8483a36d3be1058f72afc20a15c2a5283c6007022
thread:issue:19263,issue #19263: GCP Base Image Wrong CUDA Version,stale/low urgency,2,stale/low urgency,1,0,19263,,,,0,b746c2020e4270461ee3931223d66d04ee6a232338788972cdb8600a52d48d1c
thread:issue:19267,issue #19267: torch.distributed.broadcast should default to current stream,stale/low urgency,2,stale/low urgency,1,0,19267,,,,0,1fb984e65e2332a1312f2231bd7e0ee53b644adec420514485120da56487aaa4
thread:issue:1927,issue #1927: [feature request] time-distributed layers for application of normal layers to sequence data,stale/low urgency,2,stale/low urgency,1,0,1927,,,,0,97d25e5dab25dbef4e17ee580ad8b98cb549cf4f455e77d3fd20471a5d21cc75
thread:issue:1932,issue #1932: Feature Request: ReLU on LSTMs and GRUs,stale/low urgency,2,stale/low urgency,1,0,1932,,,,0,e1fb21d93b226809aba1af601cfaba2bf76ee06c03ef19a49f56023147984a53
thread:issue:19330,issue #19330: Multiprocessing on distributed Multi-nodes shutdown error: ‘spawn’ on slave node leads to semaphore_tracker leaked,stale/low urgency,2,stale/low urgency,1,0,19330,,,,0,548502025401d28682f3c34797bc63e2ca994c668aff2dc211e4b9ef62793ee0
thread:issue:19335,issue #19335: CUDA optimization: using `__restrict__` whenever possible,stale/low urgency,2,stale/low urgency,1,0,19335,,,,0,f0fc1f463ae678e4156822d8955daf34032e8cf13f13cff100fa1e9e6142f8cb
thread:issue:19349,issue #19349: Different behavior when trace model.,stale/low urgency,2,stale/low urgency,1,0,19349,,,,0,f6e174035e86ca0aa5a25dd4d0909d60f257b8985918bcbf48900120c923d9c6
thread:issue:19408,issue #19408: Massive memory overhead over NumPy,stale/low urgency,2,stale/low urgency,1,0,19408,,,,0,73e4e5a625d8c0abc711d19bd9db21b15f009e4ac3d1d8dec94cd831c4b7e873
thread:issue:19416,issue #19416: [caffe2]Reloading model gives segmentation fault,stale/low urgency,2,stale/low urgency,1,0,19416,,,,0,182a4f68c8d239f6da9c156680cc09aa33403cffd2c93a33631df69d62d2b719
thread:issue:19437,issue #19437: The `unary_kernel` call should be pushed into CopyKernel.cpp and it should completely replace the current `copy_kernel`.,stale/low urgency,2,stale/low urgency,1,0,19437,,,,0,c4bb507ec04c02616b55401e9bb8510d91a3255a0d67d55653a37cf50d04515c
thread:issue:19453,issue #19453: How to load PyTorch model with LSTM using C++ api,stale/low urgency,2,stale/low urgency,1,0,19453,,,,0,96cb1e825431fa711ef9fb42e20be6be40824c38437cf2541c2bc38fe3685ec3
thread:issue:19499,issue #19499: C++ API 'nn::Sequential' has inconsistent behavior with python conterpart,stale/low urgency,2,stale/low urgency,1,0,19499,,,,0,4b2fbddb037ad487bf460e7d3570a36abbb61487ec87708874d95c93df176006
thread:issue:19504,issue #19504: [jit] Bad error when calling `ScriptModule`s with attributes/parameters,stale/low urgency,2,stale/low urgency,1,0,19504,,,,0,538c067ef40059929258186de409d4560cf7d432c9948df343db8e2e405b587b
thread:issue:19529,issue #19529: failed to load model which is saved as text format(pickle_protocol=0) instead of binary format,stale/low urgency,2,stale/low urgency,1,0,19529,,,,0,105e407e387ada14f0360cca17f8e12d8afd7f30ad1e76c172f6075090430db9
thread:issue:19581,issue #19581: [jit] Do aten::values dispatch at build time instead of runtime,stale/low urgency,2,stale/low urgency,1,0,19581,,,,0,85fc1e0e8f7e5bfdce370ca569cbbf166917b10715c89ccc82969424c2daa22d
thread:issue:19599,issue #19599: cudnn conv doesn't check batch_size > 0,stale/low urgency,2,stale/low urgency,1,0,19599,,,,0,cc38bf69f2822371d5410d940ac85442a535b5accb26f1694edd1accc2e89e11
thread:issue:19628,issue #19628: [jit] nn.LSTM errors in nn.ScriptModule,stale/low urgency,2,stale/low urgency,1,0,19628,,,,0,2696224963702d8c73269c0085032618075db54692bc7d28c91191793ffac90e
thread:issue:19652,issue #19652: [docs] torch.set_flush_denormal(...) to mention default mode,stale/low urgency,2,stale/low urgency,1,0,19652,,,,0,f8a28da8a62472b7d3f761bf0a35723937890372d28688cfb829edcd80672376
thread:issue:19672,issue #19672: torch.set_flush_denormal not working on some (old) OSX machines,stale/low urgency,2,stale/low urgency,1,0,19672,,,,0,78b1469d1dd3bd1b513756a7cdaf3a6b56697a3b5b8075f4d0cf342ea455eb06
thread:issue:19682,issue #19682: View in Sequential - reasoned case for it,stale/low urgency,2,stale/low urgency,1,0,19682,,,,0,7ecfcb5859572376f9d91b71a9ba03933fbb40d274932a6dd78d3781246a4b2b
thread:issue:19708,issue #19708: Test failure for depthwise3x3_conv,stale/low urgency,2,stale/low urgency,1,0,19708,,,,0,46e9fa69bea788436d8a49e318ece54dd27a0f956488f23e08fea8005a4bb005
thread:issue:19825,issue #19825: Performance issue master (a25b79531),stale/low urgency,2,stale/low urgency,1,0,19825,,,,0,e85470d2c39f89a5c5065667c72fe7e7931e65e0d7b7d39e208225215a4ca9d0
thread:issue:19911,"issue #19911: BatchNorm1d does not support batchsize>65535 in eval mode with 3 dimension (NxCxL), raise CUDNN_STATUS_NOT_SUPPORTED",stale/low urgency,2,stale/low urgency,1,0,19911,,,,0,2bec646c0ca42350116fe649bbfb7adb82944f4a3ad5a687f85a6dcc28b25297
thread:issue:19930,issue #19930: caffe2/resnet50 assert in fetch_blob() when base_learning_rate = 0,stale/low urgency,2,stale/low urgency,1,0,19930,,,,0,023017a5b5621a9d8200c0dc61bbb017bd921f57da86d0a7760e129f57bc89a5
thread:issue:19953,issue #19953: Performance difference between 0.4.1 and 1.1.0,stale/low urgency,2,stale/low urgency,1,0,19953,,,,0,e316b10b86108fc7597e0befd374bd5812a482e83bd48769e294686c1e298fd9
thread:issue:19960,"issue #19960: On first construction, CUDAContext changes default CPU allocator behavior",stale/low urgency,2,stale/low urgency,1,0,19960,,,,0,43f16992552a0174f9ec14b42797e59e53bb0f0751de8ececd27f19daf425c82
thread:issue:19978,issue #19978: [JIT] traced model with optimization shows no performance improvement,stale/low urgency,2,stale/low urgency,1,0,19978,,,,0,3f5789ec6b52386f7ff036d364fa89b3a72ba79b3ec25c742d1bf1f95934de30
thread:issue:20005,issue #20005: install error from source,stale/low urgency,2,stale/low urgency,1,0,20005,,,,0,d9e3a0cef1842f2328f97f77a5ee94683fd686a85c1aca4d377f74eecee69a15
thread:issue:20006,"issue #20006: RuntimeError: invalid argument 10: ldb should be at least max(1, 0), but have 0 at ../aten/src/TH/generic/THBlas.cpp:36",stale/low urgency,2,stale/low urgency,1,0,20006,,,,0,2e9dd98e553f37e190cd02d2b942eccb0da992b9c0771e13a0932a475e7ccdab
thread:issue:20022,issue #20022: Precision of sparse float embeddings differs from dense embeddings on CPU,stale/low urgency,2,stale/low urgency,1,0,20022,,,,0,db1b247d910714a29b7c80bf3a0119af5ae49b44bc09758e40bc8f1e1ab8d476
thread:issue:20056,issue #20056: Creation of too big multidimensional array returns empty tensor.,stale/low urgency,2,stale/low urgency,1,0,20056,,,,0,f1adc3c04d7e46b984fa3cb3ef94a180a76e11cd285374e26b75b94e5e519632
thread:issue:20058,issue #20058: Implement noise_shape keyword for Dropout layers,stale/low urgency,2,stale/low urgency,1,0,20058,,,,0,a1a15eb1e16ce8b9973390b076a208686cce672ffd501f0446ecbcc618b4ec31
thread:issue:20086,issue #20086: CosineAnnealingLR giving unexpected learning rates on PyTorch 1.1.,stale/low urgency,2,stale/low urgency,1,0,20086,,,,0,54cca7cdc5bd2f74835a98414831515b9f4f9879e44bd0700c9a2d16260ff760
thread:issue:20111,issue #20111: Split libtorch binary build CI job into separate variants,stale/low urgency,2,stale/low urgency,1,0,20111,,,,0,61d0a06ce9f38335d9980a5e1390d860e7fa57aa6e434561a4e19a951f460cf6
thread:issue:20119,issue #20119: [JIT] Source highlighting doesn't line up when tabs are used for indentation,stale/low urgency,2,stale/low urgency,1,0,20119,,,,0,efef7a56619d8fc05d3c67860c4002f8e9882e5368d633ed712b385246a2b8e6
thread:issue:20138,issue #20138: Inconsistant values of lr_scheduler.get_lr and lr in optimizer.param_groups,stale/low urgency,2,stale/low urgency,1,0,20138,,,,0,9c550e155d243d6d290c9b4856d0ca936a1055fcc2be5d3b9138ac8ed64ac9d9
thread:issue:20151,issue #20151: RuntimeError: Given input size: (2048x1x1). Calculated output size: (2048x-5x-5). Output size is too small at /pytorch/aten/src/THNN/generic/SpatialAveragePooling.c:48,stale/low urgency,2,stale/low urgency,1,0,20151,,,,0,3ead02e7fd1d526fbf23ce73d80352011fb1ab7e5658ba5fbc2e5c637894db08
thread:issue:20165,issue #20165: torch.nn.threshold cannot accept tensor as a threshold,stale/low urgency,2,stale/low urgency,1,0,20165,,,,0,14f3e6a66c1c409aa5c41b197f3083ace5473b85c16314ea964b47c77b227a28
thread:issue:20183,issue #20183: Latex Errors when Compiling documentation to latexpdf,stale/low urgency,2,stale/low urgency,1,0,20183,,,,0,dba2abe3df990f61f8e0dee7f65ed57b1b57ad0fdaed5d6455af1505fcd542e7
thread:issue:20189,issue #20189: Unify Caffe2 and PyTorch OpenMP initialization,stale/low urgency,2,stale/low urgency,1,0,20189,,,,0,bde4e0d4d9d15438855062e0f1d2131db9a8e1aaf2aa482b39c3f1e733b74af2
thread:issue:20204,"issue #20204: SyncBatchNorm should support 2D input (B, C)",stale/low urgency,2,stale/low urgency,1,0,20204,,,,0,7004933a59ea5f20817e774d1264e2ffb021b891b4995b2f5ebbec267ab62b51
thread:issue:20206,issue #20206: Failed to build with system protobuf,stale/low urgency,2,stale/low urgency,1,0,20206,,,,0,961173aedc511082bb2f351bc0c553b3d798a971bdc11ad7ff47aa770341c4f4
thread:issue:20245,issue #20245: ProcessGroupMPI tests in test_c10d.py,stale/low urgency,2,stale/low urgency,1,0,20245,,,,0,b3581f2a55c9a59ddbfb05512002b829e3e125c151a6eafdf70f9c382626e2a4
thread:issue:20248,issue #20248: Sparse tensors can't be used in DataLoader running many workers,stale/low urgency,2,stale/low urgency,1,0,20248,,,,0,d8bdc44c793b5f2792ddb2520aacd0ffaa4be194fc1b1b137705e2cbffed987f
thread:issue:20268,issue #20268: [Caffe2] Convert caffe to caffe2 with solverstate,stale/low urgency,2,stale/low urgency,1,0,20268,,,,0,afc52e77ef968462dad520ac470a88fe9d0fc2dcebcedc2fe64a216c3d8d0c65
thread:issue:20272,issue #20272: Redundantly saving sizes of SavedVariables in autograd Function,stale/low urgency,2,stale/low urgency,1,0,20272,,,,0,c367034424ba1db126e6912567ac08fcd586a8997ef3e6e3ad3678a65c36774c
thread:issue:20273,issue #20273: Linker errors when linking statically (avx perfkernels) [Caffe2],stale/low urgency,2,stale/low urgency,1,0,20273,,,,0,73b29b112bc2c693edb8e89985d05daf7cb13f0b3f6a472a92db43dcabce0f4f
thread:issue:20315,issue #20315: libtorch+opencv Mat result error: different from the python ones,stale/low urgency,2,stale/low urgency,1,0,20315,,,,0,bfe2ca42779286c7f3f014faba3a40f65c31b0f0a8d7c839b7192e3d292aafcc
thread:issue:20342,issue #20342: TensorIterator resizes output to a scalar if there are no inputs,stale/low urgency,2,stale/low urgency,1,0,20342,,,,0,3097d54ecaca8b11c3861f31af880f268bec0787f878bd649943985a4390bd01
thread:issue:20343,issue #20343: torch.distributions.Binomial.sample() uses a massive amount of memory,stale/low urgency,2,stale/low urgency,1,0,20343,,,,0,e9b471dbd76bac224cc8f01ecd7eefee368dc39b1737f8e6ba66b5405dcc70e6
thread:issue:20359,issue #20359: [feature request] Run examples from docs as tests,stale/low urgency,2,stale/low urgency,1,0,20359,,,,0,5b17bf55b2e6d6ff1347b839badec8fd661609dea75e9f783e878b7a6cf6b7ed
thread:issue:20363,"issue #20363: Define portable M_PI replacement, use it instead of non-standard M_PI in math.h",stale/low urgency,2,stale/low urgency,1,0,20363,,,,0,4c8f4ee8fa43cf3b2b98bbd6376a6f4c09f3da28d931e654a20450a3550d609e
thread:issue:20375,issue #20375: Better documentation / molly-guards around use of multiprocessing with spawn in Jupyter/ipython notebooks,stale/low urgency,2,stale/low urgency,1,0,20375,,,,0,e16c5d3849f42d1ef5210fd3ffc215574896619d74d3efeb785cf6db7c66cfd9
thread:issue:20405,issue #20405: CosineAnnealingLR has unexpected behavior with large step,stale/low urgency,2,stale/low urgency,1,0,20405,,,,0,c6a6a799342200bcf10874ef885671fb1b4ed7002cdab783cd311a40a2a0398a
thread:issue:20412,issue #20412: String in tensor,stale/low urgency,2,stale/low urgency,1,0,20412,,,,0,07255193a2ba1cefec48d78c96daebd749335e36102ebb425452dfcd9cc35a46
thread:issue:20416,issue #20416: Class based Sampler for Class Incremental/Continual Learning research,stale/low urgency,2,stale/low urgency,1,0,20416,,,,0,83141064017ba369d2dc133bcb6ea6aeb12067e2c2388016bed487346fd5e952
thread:issue:20424,issue #20424: pos_weight argument in torch.nn.BCELoss,stale/low urgency,2,stale/low urgency,1,0,20424,,,,0,599b0cc444ffb504b2ef87a4781b63b89373f22ba55e5b517258312ad7310ea9
thread:issue:20466,issue #20466: Lint rule to prevent direct use of #pragma omp,stale/low urgency,2,stale/low urgency,1,0,20466,,,,0,4880a97ee1ce0acf04ffe389d1aecbc6f64306e77fc74048d11b47623db820bd
thread:issue:20481,issue #20481: [docs] Automatically detect docs missing in rst,stale/low urgency,2,stale/low urgency,1,0,20481,,,,0,054c46c23f05eb115787ccb8daa66b0b27df4daa558716130a070ce9f5e0d999
thread:issue:20488,issue #20488: RoiAlignTest.CheckCPUGPUEqual is still flaky,stale/low urgency,2,stale/low urgency,1,0,20488,,,,0,8e019ef2a6ba3c9c3b4989a6d3cde0c5f4ac8f9e85315c74bf23039a5de9af15
thread:issue:20531,issue #20531: c++ torch::nn::Sequential increments count on name errors,stale/low urgency,2,stale/low urgency,1,0,20531,,,,0,ed58f5c5a009807535c1a4e06b87d996d1282e7510b21303ccf92d341412f793
thread:issue:20551,issue #20551: torch.norm produces incorrect results,stale/low urgency,2,stale/low urgency,1,0,20551,,,,0,2e445520f5f10407bc8a48fac9f7e0515bbfa0984d113beae4f925b7c04b28c6
thread:issue:20562,issue #20562: Performance issue when accessing an extremely large (10GB) longtensor,stale/low urgency,2,stale/low urgency,1,0,20562,,,,0,b7adad7c477e4c672151978e6699a51e044dc440f21db3ad637c9da426d646b0
thread:issue:20576,issue #20576: NVRTC_ERROR unknown when using self-built libtorch,stale/low urgency,2,stale/low urgency,1,0,20576,,,,0,fe5983fda630fd29b53f5bb9a4348f071858a206244101dc7ed0df5f9d9411f0
thread:issue:20591,issue #20591: @ignore annotation for user defined type,stale/low urgency,2,stale/low urgency,1,0,20591,,,,0,29cd18349cf1c828ab3f193637de654f9628666ab6a1afe0ff16696d8d01c814
thread:issue:20613,issue #20613: groupby function,stale/low urgency,2,stale/low urgency,1,0,20613,,,,0,eaf3cb453433430e8dde4432203968e7a640a35d6a17b07cbfce4f589df515c1
thread:issue:20641,issue #20641: Main page broadcasting (?) example image bug,stale/low urgency,2,stale/low urgency,1,0,20641,,,,0,e12aa1eaba913be22771b2493a8b51b628e09df40565e0c74ddf4e1be7a903d6
thread:issue:20655,issue #20655: nn.Embedding backwards slow under high row contention,stale/low urgency,2,stale/low urgency,1,0,20655,,,,0,d60b3be813ad46f50a257da6832515d662d78c33d880aaee8e2034f76ad3d302
thread:issue:20682,issue #20682: [distribution] Support for various domain for AffineTransform,stale/low urgency,2,stale/low urgency,1,0,20682,,,,0,baed588d8232fa5b5ab3624ba99ef39329ed3c294f9f467982a6b72f16e95bb5
thread:issue:20686,issue #20686: In-source build causes repeating filename annotations (Windows doesn't support out-of-source build),stale/low urgency,2,stale/low urgency,1,0,20686,,,,0,f5001505509413d1ab54973dd38440969fc614ffbb88beb06aed23be196dcac2
thread:issue:20734,issue #20734: [jit] set up views for Autodiff and autograd hooks,stale/low urgency,2,stale/low urgency,1,0,20734,,,,0,bf58090d97432f57b14ef051ce46c399773933ab5ff44821e5185270156306e1
thread:issue:20753,issue #20753: cmake for Torch unusable in archlinux,stale/low urgency,2,stale/low urgency,1,0,20753,,,,0,20ee6c56689ea5d204268bd6bacb062d7c5700aecfa846f23a414e24f6ec64b2
thread:issue:20776,issue #20776: Caffe2: executor_test.py::ExecutorGPUResNetTest fails with an AssertionError,stale/low urgency,2,stale/low urgency,1,0,20776,,,,0,8e5b54215171f56e97f66812760cf90e8afefe64430235e88de29c8b5340a498
thread:issue:20777,"issue #20777: Caffe2: operator_test/instance_norm_test.py::TestInstanceNorm::test_instance_norm_gradients test case fails with ""RuntimeError: dim() called on undefined TensorError from operator""",stale/low urgency,2,stale/low urgency,1,0,20777,,,,0,488468304fb9cc7993eadf6f5f55d3d8f2ece6fc08c844f7e665ace7d082f7b1
thread:issue:20864,issue #20864: Import warning when using the wrong version of CUDA,stale/low urgency,2,stale/low urgency,1,0,20864,,,,0,9e3458ba68ed799ecf688560410185910f88eae01d596035f23f0923af53b197
thread:issue:20935,issue #20935: torch.full_like missing documentation for out input variable,stale/low urgency,2,stale/low urgency,1,0,20935,,,,0,63a04bc5906fe04cca2eea60fa37d240881b9654814f38a722ce79b949c755a1
thread:issue:20963,issue #20963: official libtorch static build zip file error,stale/low urgency,2,stale/low urgency,1,0,20963,,,,0,f4f9a78387bfef62896db2c68595b420608fa7984f673e57e95e640203dc9de7
thread:issue:20967,issue #20967: Track running stats regardless of track_running_stats=False,stale/low urgency,2,stale/low urgency,1,0,20967,,,,0,948dc70a566ebbbb3cc7ad713de80b409069cb466dbd9b2e98886ed3444d79a2
thread:issue:20975,"issue #20975: build caffe2 operators failed, 'sorry, unimplemented: non-trivial designated initializers not supported'",stale/low urgency,2,stale/low urgency,1,0,20975,,,,0,c163830a74562ac073d87b7a4bd205d9536a875bbe8b4894fb8f0c49c598c0f1
thread:issue:21012,"issue #21012: Error when creating new caffe2::Predictor(_initNet, _predictNet)",stale/low urgency,2,stale/low urgency,1,0,21012,,,,0,5601241f1de6df11503f5883a51e5877822db56011bde1140dc545998b3b7a16
thread:issue:21015,issue #21015: How to use Infiniband for cpu-cluster with backend gloo?,stale/low urgency,2,stale/low urgency,1,0,21015,,,,0,a984de9fdc7f51e87801ae7eb5663c6398b6e1bd86c81791db1af60f88a1226a
thread:issue:21016,issue #21016: [utils.bottleneck] throws initialization error for cuda profiling,stale/low urgency,2,stale/low urgency,1,0,21016,,,,0,27f6a7590996cfbebe8d339dc607d5f094d53215133602172f73978466373df0
thread:issue:21028,issue #21028: Zero-dim Tensors (scalars) should be printed at full precision,stale/low urgency,2,stale/low urgency,1,0,21028,,,,0,8cb9be6d7cc8e1297363aad98dd879c4884e2f6664440ee081fad082bb0287b5
thread:issue:21050,issue #21050: output values not same and much slower than Python API,stale/low urgency,2,stale/low urgency,1,0,21050,,,,0,8aa74cd10eddb0053e5cc5df61879ba6e4c154054d44e1dc601c51327998e913
thread:issue:21061,issue #21061: Misleading Error when doing Large Batch Matrix Multiplication,stale/low urgency,2,stale/low urgency,1,0,21061,,,,0,e14481a851cfe28d656b405b3f42e39f7e782a6eb6375a39a912f9cfd2aabecd
thread:issue:21120,issue #21120: [RFC] Adding MKL-DNN Int8 functions to PyTorch/Aten/JIT backend,stale/low urgency,2,stale/low urgency,1,0,21120,,,,0,8b2d24cd8e1c53cadcc49a541a6a13392c335da3cbda6db96392f51b22c9c67a
thread:issue:21135,issue #21135: [cmake build] can't build pytorch with install mkl library,stale/low urgency,2,stale/low urgency,1,0,21135,,,,0,d5b52b8a945dca26acda4dabde2629360c4b8dc4ca1c13740d84d0e487daac1d
thread:issue:21144,issue #21144: Data Parallel Implementation Improvements,stale/low urgency,2,stale/low urgency,1,0,21144,,,,0,9a5776a5f14da7b1c290f6933974fc864f6c98f9caeb00acf822c6de6dea5dd7
thread:issue:21165,issue #21165: Getting Access to Blob/Tensor reference in jit::script::Module,stale/low urgency,2,stale/low urgency,1,0,21165,,,,0,e2b4b5e21000c362722ce9e7ef9e9e2be559758ebdf66c61815075c1bb1f9c30
thread:issue:21176,issue #21176: Adding a method called `T` in native_functions causes undefined behavior on Windows,stale/low urgency,2,stale/low urgency,1,0,21176,,,,0,562679b3c1e5a52f849311d077e73129d0e0646d79d0b2a369c14755daca7c1a
thread:issue:21275,issue #21275: weight_norm is not supported in TorchScript,stale/low urgency,2,stale/low urgency,1,0,21275,,,,0,84c9cbfd47f75fc571fc379df55b495856a9aca2a4774d799aed7726d959207c
thread:issue:21277,issue #21277: [caffe2] check Range Operator inputs with bug,stale/low urgency,2,stale/low urgency,1,0,21277,,,,0,3f2b028a965e3c452b96f92940b700acd011795a384d8e9958d6b865be8d55ad
thread:issue:21290,"issue #21290: Title of docs page includes ""PyTorch master documentation"" even for non-master branches.",stale/low urgency,2,stale/low urgency,1,0,21290,,,,0,a6329a2c39af1f10f49207539f08b01f07d5e234e346de1e56f6352246d4c61b
thread:issue:21318,"issue #21318: Undefined symbols for architecture x86_64: ""testing::internal::UntypedFunctionMockerBase::UntypedInvokeWith(void const*)"" on Mac OS X",stale/low urgency,2,stale/low urgency,1,0,21318,,,,0,3e388bf48910057849d2df1d8098b3794d5a9268e8e6a4a7a8fbbbb419316906
thread:issue:21399,issue #21399: collect_env ignores conda environment,stale/low urgency,2,stale/low urgency,1,0,21399,,,,0,ff42267dc183029b13f0bb76ef3823c897dad34e16122acb0636a8916b4097cd
thread:issue:21405,issue #21405: how libtorch can work with tensor data as same as pytorch,stale/low urgency,2,stale/low urgency,1,0,21405,,,,0,d74ecd672c3bb8ad02f4b3b9b99aa32093f678baae1ca6c7b3f0741c1419eaf1
thread:issue:21411,issue #21411: Implementation of Group equivariant convolutions,stale/low urgency,2,stale/low urgency,1,0,21411,,,,0,f65626e92f5ba4b68470ab0f65735a115a77bf5350104a436ec22309d4fdcd82
thread:issue:21442,issue #21442: [JIT] kwarg with default doesn't work for class instantiation,stale/low urgency,2,stale/low urgency,1,0,21442,,,,0,214229eb74dc99fd645abdf4dd940694138846bb060574a9ed128360686d24cc
thread:issue:21454,issue #21454: [JIT] Memory Leak during tracing?,stale/low urgency,2,stale/low urgency,1,0,21454,,,,0,39c9af8c9bb67bdd3d49be5110966c6fc33399ab7bb51df853532dafbb16bb31
thread:issue:21459,issue #21459: RuntimeError: cublas runtime error,stale/low urgency,2,stale/low urgency,1,0,21459,,,,0,427fbae81a81ffc5ffbd0fe367966ce4dda089940c1fdc56b080b541cdf20d5d
thread:issue:21462,"issue #21462: Slow convolution with large kernels, should be using FFT",stale/low urgency,2,stale/low urgency,1,0,21462,,,,0,e1ce5a035b87263026fd77d3abc24786fb06ff5235e098fb272f444f48380778
thread:issue:21477,issue #21477: Not obvious how to install torchvision with PyTorch source build,stale/low urgency,2,stale/low urgency,1,0,21477,,,,0,f31d21ee00c60b8f864763792ed052fe62eeaa57abd15bc5d6414180bd1c3870
thread:issue:21518,issue #21518: `attn_mask` in nn.MultiheadAttention is additive,stale/low urgency,2,stale/low urgency,1,0,21518,,,,0,3b09e6f72c6d35672ed670c19494e4a0a4012525e1d2c30100e71d4cc628af40
thread:issue:21544,"issue #21544: [jit] In pickler, don't memoize if not necessary",stale/low urgency,2,stale/low urgency,1,0,21544,,,,0,bc24c21a78f0554387dfe1c354d4cf26d78b69bc7895e8afc70add7fb981e8db
thread:issue:21553,issue #21553: Feature Request: beta cdf,stale/low urgency,2,stale/low urgency,1,0,21553,,,,0,52594fe4b99229c3cae1237f1906abff349d3e5dd8063ecbeae82c0c5620e8e4
thread:issue:21554,issue #21554: Pytorch hangs when dataloader multiprocessing workers are killed,stale/low urgency,2,stale/low urgency,1,0,21554,,,,0,97e846ab2cb56a08cce84d367f8ef0b768d149ba196c3ef2a891a353a411a521
thread:issue:21567,issue #21567: torch.bernoulli's parameter generator not documented,stale/low urgency,2,stale/low urgency,1,0,21567,,,,0,17a968d7963035b032dc07c63db292aab8bf9301e906d84e15f95c0d0a50865e
thread:issue:21587,issue #21587: RuntimeError: cublas runtime error,stale/low urgency,2,stale/low urgency,1,0,21587,,,,0,f20fc5d4754e13bf91908dc503806b5fa9821a40ac9af9fb472bf53003ca5164
thread:issue:21602,issue #21602: Strange latency overhead of F.conv2d,stale/low urgency,2,stale/low urgency,1,0,21602,,,,0,e715e1aa1a8341c007f705bf80b1e371ec512d1850596894071458b374e35136
thread:issue:21645,issue #21645: Batch Dataloader and Dataset,stale/low urgency,2,stale/low urgency,1,0,21645,,,,0,b944e7b65ab7fdedd3445c89327582f6fff425c15c62893acdc50bf2404e067e
thread:issue:21673,issue #21673: [FR] Diagonal Transform for Distributions,stale/low urgency,2,stale/low urgency,1,0,21673,,,,0,c9b00cd11b529d472379cbe22b425d5e9dfd7fa711a3ce2df506ec4e3b65fd81
thread:issue:21682,issue #21682: C++ module API footgun: assigning to parameter doesn't update `parameters()` list,stale/low urgency,2,stale/low urgency,1,0,21682,,,,0,eae28d107192ed9f1e5320b18c5d77d6535c201d4def72e71715b2cad7b5d70b
thread:issue:21684,issue #21684: The cuda problem in caffe2,stale/low urgency,2,stale/low urgency,1,0,21684,,,,0,1c0c1b17e2375abd8303d4ca8a9c0a4c54c4c788f3368e8961893c2be0d5a0e2
thread:issue:21738,issue #21738: symbol lookup error: libmkl_intel_lp64.so: undefined symbol: mkl_blas_dsyrk (binaries built with static linking -DBUILD_SHARED_LIBS=OFF fail due to dynamic linker problem),stale/low urgency,2,stale/low urgency,1,0,21738,,,,0,f2007cfa60ea9ec09ac5946480ee43d5cfc7bd9d23534d039614a83670b3b3af
thread:issue:21745,issue #21745: torch.save also saves docstrings into pickle for some reason,stale/low urgency,2,stale/low urgency,1,0,21745,,,,0,e436e4d1381794faec5c63d6338d6c45346963e43ba4ca2668ff75814024d8f7
thread:issue:21766,issue #21766: Missing header files for prebuilt libtorch,stale/low urgency,2,stale/low urgency,1,0,21766,,,,0,3334b59e703406131ca2dabe2c56e3899bb0b73ead640f397cbf6b4f851131ba
thread:issue:21777,issue #21777: typo and missing return statements,stale/low urgency,2,stale/low urgency,1,0,21777,,,,0,c9cab046809f71d13146203eda3f362deb83967ad29e3de5b74e9265d4015921
thread:issue:21779,"issue #21779: Use of word ""elements"" in `torch.utils.data` samplers",stale/low urgency,2,stale/low urgency,1,0,21779,,,,0,b9aece799a3102aebf390f5c74ea4b7ed26db09680e52a369b8687014ee7424e
thread:issue:21780,issue #21780: cumsum cuda numerical instability,stale/low urgency,2,stale/low urgency,1,0,21780,,,,0,e658ca4291ed56c74d71cc62619db6d16bd3375ff095f82214033a7c866c22b2
thread:issue:21794,issue #21794: Linker errors when building project with OpenCV,stale/low urgency,2,stale/low urgency,1,0,21794,,,,0,199531895d6ddaed67b2c6456e232ef7006477b29843da99ee158f96f1adb08c
thread:issue:2180,issue #2180: Discrepancy in BCEWithLogitsLoss and ClassNLLLoss,stale/low urgency,2,stale/low urgency,1,0,2180,,,,0,9910b6c988b7103fcd18f80d6b66ae37a998caea99dbbc89eb5c9b43d42b6b82
thread:issue:21818,issue #21818: C++ ABI - Coupling different libraries issue,stale/low urgency,2,stale/low urgency,1,0,21818,,,,0,51949b9885f3f79971651fc109b746d8f01f2f92c0cb3c6ea9f2189c2ba6d84f
thread:issue:21824,issue #21824: How about add torch::end for slicing in c++ frontend,stale/low urgency,2,stale/low urgency,1,0,21824,,,,0,8ad98da55d03815264de953c7a3171ed1414b43a6d333604a3c643132a6ccda2
thread:issue:21828,issue #21828: torch::zeros is slow for small tensors (C++),stale/low urgency,2,stale/low urgency,1,0,21828,,,,0,c9b9b46c8d22ca528b403c02bd82808494729d38693301724a8580a259d66c6c
thread:issue:21847,issue #21847: Logging mode for saying when tensor broadcast occurs,stale/low urgency,2,stale/low urgency,1,0,21847,,,,0,799244e04a39a5487c4e9f9cedc6330575d354df04cc344f9f081dfd932e1ebe
thread:issue:21856,issue #21856: Batch Normalization axis,stale/low urgency,2,stale/low urgency,1,0,21856,,,,0,3432549469efd66100d05965485fcf9012a891932ba8b4cabbaca45455423e90
thread:issue:21899,issue #21899: Crash when using tensor.set_data() function in libtorch on windows,stale/low urgency,2,stale/low urgency,1,0,21899,,,,0,0743898753f3ac25cc05bf2645d4431514ae11b9f76622590d47c0f945e3ba3d
thread:issue:21902,issue #21902: contradictory output values,stale/low urgency,2,stale/low urgency,1,0,21902,,,,0,c34621b82150b1d4dd86fe03863afc2559e31359bfed97d944d4c6e2f38581df
thread:issue:21956,issue #21956: nn.init.orthogonal_ doesn't work with multiprocessing,stale/low urgency,2,stale/low urgency,1,0,21956,,,,0,d6794e9062c430fa86571234511e6badd5dc0413ebb6854c625dd214a92b69c8
thread:issue:22022,issue #22022: nn.modules.functional.h does not support optional arguments,stale/low urgency,2,stale/low urgency,1,0,22022,,,,0,4c50ef07d00d04f98bd06cc5519d4114a2fae0abe97eb1f606e5b94b7397384c
thread:issue:22024,issue #22024: Consolidate definition of operators/gradients where possible,stale/low urgency,2,stale/low urgency,1,0,22024,,,,0,2de7a0d7771e29226ab0d90c7ed52d78e323761d8c2d268fbee887556524f539
thread:issue:22035,issue #22035: Have a different way to check if gradient was computed in the optimizer (not checking for None),stale/low urgency,2,stale/low urgency,1,0,22035,,,,0,742317c4c5a5891a6dddd8ba0dbfaa88119e3411ca85e0e239fe731510cf467b
thread:issue:22040,issue #22040: SigAbort while running the Caffe2 unit test - thread_init_test - built on Clang7 +glibc 2.23,stale/low urgency,2,stale/low urgency,1,0,22040,,,,0,d8ce2705af05dc749508ee73af910634bc2ac6b35e6a383276e865d4e1225617
thread:issue:22049,issue #22049: Cannot update part of the parameters in DistributedDataParallel.,stale/low urgency,2,stale/low urgency,1,0,22049,,,,0,780563852b7701d329e12c20a8741df1045983eef6f651aa7e41b186dd11af2e
thread:issue:22122,issue #22122: ASSERT FAILED at /pytorch/torch/csrc/autograd/generated/VariableType_0.cpp:12721,stale/low urgency,2,stale/low urgency,1,0,22122,,,,0,ca7bba0e42234e004684a09572d67195fd4296948fa01880892f69a957f486f9
thread:issue:22128,issue #22128: Automatic rank selection when using file:// initialization method,stale/low urgency,2,stale/low urgency,1,0,22128,,,,0,e98faf8de7d609e8575d2306bc3177690aa57aa010d8daf066fcc978421c22e4
thread:issue:22136,issue #22136: [dataloader] SIGCHLD handler should poll the queue for exception first,stale/low urgency,2,stale/low urgency,1,0,22136,,,,0,58a53a6cf765759341740e279c37a07a688ac22de403a4ad80c941a3d70fb81f
thread:issue:22169,issue #22169: [RFC] NestedTensor - 0.0.1,stale/low urgency,2,stale/low urgency,1,0,22169,,,,0,3700ea2b13968c0a90c78764c901431730a19bcece6703e01ec8db118fb14fc6
thread:issue:22220,issue #22220: [doc] nn.Module.forward documentation unclear,stale/low urgency,2,stale/low urgency,1,0,22220,,,,0,88996d835b4d06a6ee6f6f3103ce56f2abe517bee3d795316ef6dcab5eedf00d
thread:issue:22232,issue #22232: [FYI] Introducing Quantized Tensor,stale/low urgency,2,stale/low urgency,1,0,22232,,,,0,e7790bc3f315b83cfd687d7895024b6f955e0e73c29b9affcb52806fd6360afc
thread:issue:22252,issue #22252: [libtorch] header warning suppression,stale/low urgency,2,stale/low urgency,1,0,22252,,,,0,5611413fe62ff4b44167afd6c64ec76dc34b5f2f47dc41f173fe51abbddb578c
thread:issue:22274,issue #22274: second derivatives of unfold,stale/low urgency,2,stale/low urgency,1,0,22274,,,,0,7096689950a1c20abd190784e7047ab048809fc70ce907dcab852780c322d702
thread:issue:22277,issue #22277: Handle all IntArrayRef expansions in ATen,stale/low urgency,2,stale/low urgency,1,0,22277,,,,0,4d1da9207f8a820c02a912e85a4bb3870fb6173c5947a7e1be7f910f7ac74c5e
thread:issue:22330,issue #22330: Training CNNs with deconvolution,stale/low urgency,2,stale/low urgency,1,0,22330,,,,0,1a2973494f6fed11fd8038d7f48fb61a38e49d617d56bb5b6b218b5c77235ad9
thread:issue:22331,issue #22331: Integer division by Zero giving large number results instead of NaN/inf on Windows,stale/low urgency,2,stale/low urgency,1,0,22331,,,,0,b1cfd1e13e75357cc8dcba5401afdd77f21242cf6a992d50e51b7b0e4b25180a
thread:issue:22340,issue #22340: Reducer bucketing based on autograd profile,stale/low urgency,2,stale/low urgency,1,0,22340,,,,0,50f49e13e258d97107acfe37dc5a220fc608665b0094369da9918767dbebfe2d
thread:issue:22358,"issue #22358: When I run python setup.py install to install Caffe2, I have an error: ""No such file or directory: 'nvcc': 'nvcc'""",stale/low urgency,2,stale/low urgency,1,0,22358,,,,0,8e50dd30bb96cb25b5ffbb654b330aaa66ed20b5458f1ed1f69d1b496a041c6f
thread:issue:22370,issue #22370: returned non-zero exit status 2.,stale/low urgency,2,stale/low urgency,1,0,22370,,,,0,d8597dfa0045dda0376c297ac35c0ed0584849bb054e8735ba97016f9f276ba6
thread:issue:22375,issue #22375: Label.dim Enforcement Check in AccuracyOp,stale/low urgency,2,stale/low urgency,1,0,22375,,,,0,2e40b5f861aefe9ded282812f2f78ddc99a56c179fd86b17a65ecdd8b6187a79
thread:issue:22383,issue #22383: Storage operation failing on second GPU,stale/low urgency,2,stale/low urgency,1,0,22383,,,,0,0275d2f61979cd5d5c4724a1fd642ba152f7babf7c817eab54c1317ffa554fdc
thread:issue:22400,issue #22400: Sparse allreduce for ProcessGroupNCCL,stale/low urgency,2,stale/low urgency,1,0,22400,,,,0,ecda92cd3125a9f4e630c271cb6f4592b84bfd2dda12cfc26de87915b57ebf48
thread:issue:22406,issue #22406: Build failure with setup.py,stale/low urgency,2,stale/low urgency,1,0,22406,,,,0,cd7898caeb3511162911104e78241bb4200960f643115bdfc677eeba27ae6af4
thread:issue:22415,issue #22415: Add support for serializing Mkldnn Tensor,stale/low urgency,2,stale/low urgency,1,0,22415,,,,0,f8038d4fc318ac4cda32ec99613a912ea7a2405b36954e84f8c884cafe599226
thread:issue:22447,issue #22447: Weak Symbols Resolution Causes Segmentation Fault in External Libraries,stale/low urgency,2,stale/low urgency,1,0,22447,,,,0,46fb7e6fff28612e9a2d4d319597feb9bd62ba56f9b94e21d21aa788826051e3
thread:issue:22454,issue #22454: Tracing an RNN does not support torch.nn.utils.rnn.PackedSequence as input,stale/low urgency,2,stale/low urgency,1,0,22454,,,,0,10e07c46fc02781c5014e09a64ec74adc96e080496c50093f9b3e179bf8a7e7a
thread:issue:22455,"issue #22455: Move csrc/distributed/c10d/{comm,reducer} to libtorch.so",stale/low urgency,2,stale/low urgency,1,0,22455,,,,0,847544d3978c484175056869d410145586c1f5546a4554ee5389bad05a00401a
thread:issue:22487,"issue #22487: ""Floating point exception"" after trying the method from the issue #22382",stale/low urgency,2,stale/low urgency,1,0,22487,,,,0,bf7cddba985d6a24ea18b83831902c933e03bd8effa84146cd920a62341c2d6a
thread:issue:22494,issue #22494: Error in equation,stale/low urgency,2,stale/low urgency,1,0,22494,,,,0,a3316262e45818663744d65f05399b80ac503618f8b607e3c7a907ca65a7b93c
thread:issue:22495,issue #22495: `binary_linux_libtorch_2.7m_cu100_devtoolset3_build` times out after running for 5 hours,stale/low urgency,2,stale/low urgency,1,0,22495,,,,0,066e845d77aa74c39550a5b33d2743a61aac1575bf22abd197e0808e86a11bbc
thread:issue:22513,issue #22513: No assertion when using scatter_ on a non-contiguous tensor,stale/low urgency,2,stale/low urgency,1,0,22513,,,,0,2be9ebd4a808c6c9c05736fb63a2e9dac7663499510813064510445e32465c45
thread:issue:22532,issue #22532: Know which function is used by conv and force to use a function,stale/low urgency,2,stale/low urgency,1,0,22532,,,,0,8007188532e944c9e388ae3e0f6c46da481c2e17399baf92fcbf16af866a687d
thread:issue:22536,issue #22536: libtorch new op,stale/low urgency,2,stale/low urgency,1,0,22536,,,,0,66840d3a8b3594f3fbc0a9ed158b896e2f06597ce6d9b76b0c71e46ba4e8c33f
thread:issue:22541,issue #22541: make[2]: *** No rule to make target 'libtorch/lib/libc10.so',stale/low urgency,2,stale/low urgency,1,0,22541,,,,0,b355d3f9ad99b88a03da20ae5df142e2ce9a329aa85d06f80e12f1202472a84a
thread:issue:22550,issue #22550: Eigen Tensor library for convolutions on CPU,stale/low urgency,2,stale/low urgency,1,0,22550,,,,0,7dc537760b51d177823961a206e40629f8ad39b084712be8d39cfa6f3d1e8e54
thread:issue:22557,issue #22557: CPU torch.exponential_ function may generate 0 which can cause downstream NaN,stale/low urgency,2,stale/low urgency,1,0,22557,,,,0,fbbecf3b37800c590c7039afe00f6fe493b3bc52ed83f6341ad0dcadd6aec520
thread:issue:22573,issue #22573: Batched symeig and qr are very slow on GPU,stale/low urgency,2,stale/low urgency,1,0,22573,,,,0,6709c4619bcf9a691bb5420438d7def4489582cd73a2a4dc0e24eb39bd874c41
thread:issue:22577,issue #22577: Double backward 3 times slower for conv2d with padding = 1,stale/low urgency,2,stale/low urgency,1,0,22577,,,,0,7db19ac2a57dd60b37efb48a6c8e91719f09f650549384399807da49b052b153
thread:issue:22578,issue #22578: torch.gels runs 100 time slower on gpu than on cpu,stale/low urgency,2,stale/low urgency,1,0,22578,,,,0,1eae4b804f15d0dc65d14e8cb6efa15f05f0e43097bb83c2e6259691f0068ccc
thread:issue:22580,issue #22580: Pytorch compilation error on Mac OS,stale/low urgency,2,stale/low urgency,1,0,22580,,,,0,65e0cd19c825331f8d87a5c8e3943a81ff547d5f0259ff1c621ec82b2ab584c6
thread:issue:22585,issue #22585: [dataloader] Mysterious error when using spawn start_method,stale/low urgency,2,stale/low urgency,1,0,22585,,,,0,64c48f02fbd1b988caeeb5b2dff8491e6e1b2745d31921d8ac7f31d39262b31f
thread:issue:22604,"issue #22604: ""CrossEntropyLoss"" should mention in its name that it takes softmax for target",stale/low urgency,2,stale/low urgency,1,0,22604,,,,0,27a2d6678a9098e92db17fad8044f1d03176b9fefa7905eebaf2059d71051542
thread:issue:22609,issue #22609: caffe_translator TranslateCrop fails when more than one dimensions is cropped,stale/low urgency,2,stale/low urgency,1,0,22609,,,,0,b18093e8ce4c51450745ec7dd4ec0b330e2cc6e56a1f29f64004e1bb3f6c0136
thread:issue:22669,issue #22669: BFloat16 numeric limits should contain more info,stale/low urgency,2,stale/low urgency,1,0,22669,,,,0,7fd4a8761cd0009f759caea834c33301609ee6813ddf79ce0da057f27c01eab6
thread:issue:22682,issue #22682: CMAKE_PARSE_IMPLICIT_LINK_INFO Function invoked with incorrect arguments,stale/low urgency,2,stale/low urgency,1,0,22682,,,,0,89589abefd7480f2100fb45ed99f39f37eddb71293e94bb886b32cbf96fee2a9
thread:issue:22687,issue #22687: DispatchStub should report what operator it failed to find kernel for,stale/low urgency,2,stale/low urgency,1,0,22687,,,,0,9c855e0df67d2ab531075e768bbc67a4e8976550374969aa9ac46a8f34552de6
thread:issue:22707,issue #22707: torch.fill_() exists and modifies the input tensor: Expected or bug?,stale/low urgency,2,stale/low urgency,1,0,22707,,,,0,8da7828e11e6859d93e9e1e0ee74685dac4a224fb8d6895fe416b6110aa79921
thread:issue:22755,issue #22755: [RFC] InstanceNorm default affine value,stale/low urgency,2,stale/low urgency,1,0,22755,,,,0,7d08c74a73dffbeb40ad19c39b81bcff35b7cf57a713c1dd3c601f4a5cddfc02
thread:issue:22766,issue #22766: assert_no_internal_overlap should pass const char*,stale/low urgency,2,stale/low urgency,1,0,22766,,,,0,a84981a000cd0dc367e57eedb9117777c0b9e51a08abd13b402ea945e6886238
thread:issue:22778,issue #22778: In-place updating the original value tensor should also update version counter of sparse tensor's values_ tensor,stale/low urgency,2,stale/low urgency,1,0,22778,,,,0,6b5571b35bfb8e8bf91e946329a486667846a54ed93ffb561611a1dfde9e36ed
thread:issue:22785,issue #22785: Getting cuda runtime error (48) with Jetson TX2 when running simple program,stale/low urgency,2,stale/low urgency,1,0,22785,,,,0,169db44592395b3059f71fc246626fe185dcf28c0244fe424c1a805a74811e82
thread:issue:22788,issue #22788: Pytorch deadlock from distributed multiprocessing,stale/low urgency,2,stale/low urgency,1,0,22788,,,,0,fd01e9bc89fe8716e7afb7a483c92abb61dfe6c1174c638be5b9f1c6b5d1eac6
thread:issue:22791,issue #22791: How to use mpi backend without CUDA_aware,stale/low urgency,2,stale/low urgency,1,0,22791,,,,0,23251ea50f7487edfd0878fa25768bfc533956d6f05fc852d56a361086f5e18f
thread:issue:22803,issue #22803: Port `fmod` operator from the TH code to Aten,stale/low urgency,2,stale/low urgency,1,0,22803,,,,0,0e748532959789d1d183d4d5f7b65bcf6e17d132321da09dfe554138a2ed00b2
thread:issue:22815,issue #22815: TensorImpl de-virtualization,stale/low urgency,2,stale/low urgency,1,0,22815,,,,0,8578d2a5520fef7141ba348571bae57cee07e67e9f2138246a65569eb6c6efc0
thread:issue:22848,issue #22848: [feature request] Log-determinant for symmetric positive definite matrices,stale/low urgency,2,stale/low urgency,1,0,22848,,,,0,6faf88e6a2e44b8f9a5b7922a4800dd9ee72310ea7a0b671fd8b7999e2784231
thread:issue:22859,issue #22859: CRITICAL:root:Cannot load caffe2.python. Error: DLL load failed: The specified module could not be found.,stale/low urgency,2,stale/low urgency,1,0,22859,,,,0,f62e979cd72dd453960f86e965eabd624765ee213b6a0098576c87df32789128
thread:issue:22868,issue #22868: Make it easier to bisect on PyTorch,stale/low urgency,2,stale/low urgency,1,0,22868,,,,0,b7e007f53757577b8303a6a006e14124541081947f30d0983b3d8fff74e72951
thread:issue:22879,issue #22879: Difference between dropout2d and dropout3d,stale/low urgency,2,stale/low urgency,1,0,22879,,,,0,30b2a8eddd64fc3099770460c0169a53293342b5a27bff897e2395cf3188283b
thread:issue:22911,issue #22911: JIT trace parameter sharing error if Module attributes happen to be the same,stale/low urgency,2,stale/low urgency,1,0,22911,,,,0,0e9f59c0e0c0973a5662a98e83639cf8bc0fc88a98749504893846522010fffb
thread:issue:22919,issue #22919: Unify tensor shape formatting in shape checks,stale/low urgency,2,stale/low urgency,1,0,22919,,,,0,2eda2caf17b6d260a0f4723d2720ce862fdff979f5de40a342606ecbfaca42af
thread:issue:22954,issue #22954: Support serializing IValue to bytes (and deserialize from bytes),stale/low urgency,2,stale/low urgency,1,0,22954,,,,0,1c93bdc7109f26e61845fcd51d63d4ad07cfc85d60f2d019e1204b6494c81f6e
thread:issue:22961,issue #22961: performance much worse on 2080ti than 1080ti,stale/low urgency,2,stale/low urgency,1,0,22961,,,,0,4af24c37823a2a76b70a6892ac2bbff816d4e1872e4a3aab5c079808a091294c
thread:issue:23026,issue #23026: BatchNorm1d fails on first run through GPU,stale/low urgency,2,stale/low urgency,1,0,23026,,,,0,a47bcbeec20d2ffb707c0432db558723ae100c17d21257b455f85978830e4777
thread:issue:23032,issue #23032: Proposal: Optional AutogradMeta for Variable,stale/low urgency,2,stale/low urgency,1,0,23032,,,,0,ae750a9af66819b3b6fe68daa0548429decc9dceffd1bbf876a43ee7aee3f6ce
thread:issue:23048,issue #23048: Add automatic tuning flags to utils.data.dataloader,stale/low urgency,2,stale/low urgency,1,0,23048,,,,0,6ddda684ac3f7b3ea219fbf6b317cb396944ac53346f69ea9414e1da7aaf1090
thread:issue:23054,issue #23054: ConcatDataset returns different error messages setting out of range plus index and minus index.,stale/low urgency,2,stale/low urgency,1,0,23054,,,,0,0ffa891438786cabf7eb2797a9a38bd901a1cc4781fa634fe7eb216e5ab1f379
thread:issue:23061,issue #23061: The speed of `torch.einsum` and `torch.matmul` when using `fp16` is slow,stale/low urgency,2,stale/low urgency,1,0,23061,,,,0,156f6adc8b594f5e50d7463d80f4ce737e3d9d5b1dc5dbd34eca9de663b4ead1
thread:issue:23067,issue #23067: creation of a tensor from a numba.cuda array,stale/low urgency,2,stale/low urgency,1,0,23067,,,,0,6615438579f31b9da37d53a4321699a11cfc2ec8bbe2f674008292a407cc1e93
thread:issue:23068,issue #23068: [c++] torch::conv2d() expected output_padding to be a single integer value or a list of 3 values,stale/low urgency,2,stale/low urgency,1,0,23068,,,,0,c9da9fce1d26447a9bb95ee8b5424cf91b354a342c22d46625359a1a8bea1ec9
thread:issue:23072,issue #23072: Inconsistent axis argument names in torch.diagonal and torch.transpose,stale/low urgency,2,stale/low urgency,1,0,23072,,,,0,41dc90cd13906f10213689acfc29ec44a21ccb228cfbc5b56e8493471a6f805f
thread:issue:23079,issue #23079: update docs that sorting is not needed in,stale/low urgency,2,stale/low urgency,1,0,23079,,,,0,4d86826f1fe4f815519704624e3efbd10058c9f3b1fccc64f7564298826c728d
thread:issue:23094,issue #23094: Versioning for libtorch nightlies,stale/low urgency,2,stale/low urgency,1,0,23094,,,,0,ef597f1c5193cc4eb345148c9c84f2b7e79ff357f8567c6136c6a3e64d8072e7
thread:issue:23103,issue #23103: Test utility for non-contiguous tensors,stale/low urgency,2,stale/low urgency,1,0,23103,,,,0,7e5dac0759a8a039e1b694cf3ccfbb1c22cc43f910570972814ab4466de2afc7
thread:issue:2312,issue #2312: DataParallel is not compatible with pack_padded_sequence,stale/low urgency,2,stale/low urgency,1,0,2312,,,,0,97e36c0f8904866a425dacb5574de86ee299c277e0d089be54c3f8338e13d3ba
thread:issue:23151,issue #23151: Deterministic mode for scatter_add operation,stale/low urgency,2,stale/low urgency,1,0,23151,,,,0,071a03de64bee679a8910f5e479819152ac135163e44e36476bc0c685ea67d0c
thread:issue:23230,issue #23230: Consolidate multiprocessing helpers in distributed tests,stale/low urgency,2,stale/low urgency,1,0,23230,,,,0,2621927130ce49841aa98635611f57b507545d75bb61312c96f3aa6e6534d41c
thread:issue:23233,"issue #23233: Unofficial ARMv6, ARMv7, ARMv8 builds",stale/low urgency,2,stale/low urgency,1,0,23233,,,,0,c752246bca0eaedb6ca22a8fa0c2527a541a912f2863597f217c884f4710c899
thread:issue:23299,issue #23299: Add gatherv/allgatherv primitives to support non-equal contribution,stale/low urgency,2,stale/low urgency,1,0,23299,,,,0,3fe3355668407e355d42e22eb172301abbe7c30a3cf207a1699ef04567ef554e
thread:issue:23301,issue #23301: Error while using Libtorch + OpenCV + Qt Creator,stale/low urgency,2,stale/low urgency,1,0,23301,,,,0,efffb4828dac57fa6579894fd562e4500982b3d19def52c29487e14b97c81be8
thread:issue:23305,issue #23305: make torch.utils._download_url_from_file public and add a docstring,stale/low urgency,2,stale/low urgency,1,0,23305,,,,0,f1cd331b3a6153cf175f2362a41a51ed2ed0523997235e880f6294e24462277a
thread:issue:23355,issue #23355: [jit] add named tuple as output type to the tracer,stale/low urgency,2,stale/low urgency,1,0,23355,,,,0,14043900af0787ad694ee0c38201b1c01b8aa0460bab26428c7ed9234d92c3b1
thread:issue:23364,issue #23364: dyndep function not working,stale/low urgency,2,stale/low urgency,1,0,23364,,,,0,76ac5153774dcc385eb0ef012e7ce887f2233d784972fca15ee58c65c398fafa
thread:issue:23382,issue #23382: Tensor from mmaped storage loads the entire file into memory,stale/low urgency,2,stale/low urgency,1,0,23382,,,,0,366e7fddcb9a1f26700c1513939579d8ef0f70dc53afaec0511f123fa4bf18b5
thread:issue:23429,issue #23429: Unreachable code in tanh,stale/low urgency,2,stale/low urgency,1,0,23429,,,,0,c9f9e0e64fd717a1748d72d4332d150a82cd8dc098307727fafd2e500056b258
thread:issue:23434,issue #23434: einsum equation with conditional mask works in numpy but not in PyTorch,stale/low urgency,2,stale/low urgency,1,0,23434,,,,0,0ae2bf465f49daef0befd41c5a280cc4263565ce9371f0e44444c082d287efc2
thread:issue:23482,issue #23482: Build error due to unintended include path /usr/include,stale/low urgency,2,stale/low urgency,1,0,23482,,,,0,2138415788c686ec743f8c4727802210d123723e7fdbb1ec47a148e7f444cfa7
thread:issue:23509,issue #23509: Enable PyTorch Bfloat16 for CPU and add MKL-DNN bfloat16 optimization for Cooper Lake,stale/low urgency,2,stale/low urgency,1,0,23509,,,,0,8541e201ee7b810031d8da165b92f44f70980daa38809505ae5bef0d39bd5d06
thread:issue:23512,issue #23512: Build reconfiguration should consistently honor env variables,stale/low urgency,2,stale/low urgency,1,0,23512,,,,0,b0eac4fbe2a84ac731ce46a840c089b4b6d8ae39185f58574f22b6c8bd10e889
thread:issue:23540,issue #23540: Error from PyTorch when finalizing Python embedded in C++,stale/low urgency,2,stale/low urgency,1,0,23540,,,,0,4bed6df9e489e319c514ce3e9eaecc53c5b29c89e5a41b02a7cc437f43c88401
thread:issue:23547,issue #23547: model use dilated conv backward in v1.1.0 is ~3x slower than in v0.4.1 on 1080Ti,stale/low urgency,2,stale/low urgency,1,0,23547,,,,0,c436010a1ac035ad95ae17e3f71761323fb505d95a60e39612caa040cbd964f3
thread:issue:23655,issue #23655: TorchScript GPU Fuser Doesn't Handle In-Place Operations,stale/low urgency,2,stale/low urgency,1,0,23655,,,,0,58510128f677399892018b6a9ab0b2028d1356b260ac9860cac8ab5e3cd9e974
thread:issue:23657,issue #23657: Accelerate PyTorch just-in-time compilation using MKL-DNN,stale/low urgency,2,stale/low urgency,1,0,23657,,,,0,c86d83e7ccd374440ed4e0cb23ddd8d6dc131db9768342117d7aa032779cdbf7
thread:issue:23676,"issue #23676: Bogus ""Your compiler (clang++) is not compatible"" message",stale/low urgency,2,stale/low urgency,1,0,23676,,,,0,e88ea4908b9ea0c88854b732b32ff48b6fa1fb45b2beebedadcbaadbf1a7fc46
thread:issue:23720,issue #23720: segmentation faults when using multiprocessing_context='spawn' with large number of processes,stale/low urgency,2,stale/low urgency,1,0,23720,,,,0,d2d04907116b27a036f70288cb5abedcf95b7ce893a5cf05c9599bd4046235c9
thread:issue:23721,issue #23721: Using PyTorch on AWS EFA network,stale/low urgency,2,stale/low urgency,1,0,23721,,,,0,9cc74aa36e1868af33ebee3ccae5c87c32d1569c18572b032688607cd53eee15
thread:issue:23730,issue #23730: Failed to build pytorch ...,stale/low urgency,2,stale/low urgency,1,0,23730,,,,0,832f86ee919cd6713286028e5918e595214461f832fb77a7abb18f5930832a68
thread:issue:23732,issue #23732: nn.Module.forward signature with **kwargs,stale/low urgency,2,stale/low urgency,1,0,23732,,,,0,5a593854af24ce4368bcfa852769097bbb13962844b4659e70510b62de09511a
thread:issue:23749,issue #23749: Wrong device in graph - Tensorboard SummaryWriter,stale/low urgency,2,stale/low urgency,1,0,23749,,,,0,a3f084f85ca6140fc8a921a5dee606cd664818c9b9dbdf7419be318873aa201b
thread:issue:23768,issue #23768: Better documentation about PyTorch's dependencies,stale/low urgency,2,stale/low urgency,1,0,23768,,,,0,7782e57d42c94c446c4e642a8741e1fe8ee9b91f3576483e87f822c661129894
thread:issue:23771,issue #23771: SummaryWriter doesn't read comment if log_dir precised,stale/low urgency,2,stale/low urgency,1,0,23771,,,,0,28f27062930ce803fb44941280cb91b327dea9bdde2587c3b589ce04beac8263
thread:issue:23780,issue #23780: Construction of MultivariateNormal much slower on GPU than CPU,stale/low urgency,2,stale/low urgency,1,0,23780,,,,0,6e58a29fa595ebfb0b707f20653aa643d2ce0bed488e57e22b4312ef4b1d4e52
thread:issue:23787,issue #23787: Better version of chrome://tracing,stale/low urgency,2,stale/low urgency,1,0,23787,,,,0,9a7fb18640db7bdc07151aceb6a88a2784b11c92fe63ac44f4ad3d9f1bea6ddf
thread:issue:23849,issue #23849: Make MultiProcessTestCase pickable,stale/low urgency,2,stale/low urgency,1,0,23849,,,,0,b3f64a4cba11e5712c3c2f4897f7edf066a7443461be807e72da70918cfa6601
thread:issue:23850,issue #23850: [quantization] jit::class_ for packed weights,stale/low urgency,2,stale/low urgency,1,0,23850,,,,0,b3b6071a1043a54dad9cce41e0da90f2725277c3fd2326d8c1dee5669789e37f
thread:issue:23854,issue #23854: Fan out calculation broken for group (depthwise) convolution,stale/low urgency,2,stale/low urgency,1,0,23854,,,,0,5d245bcb7ce30c7a8e6c7b2422ae8edab5cd0e3a33c3d1af87d099c28c0f7499
thread:issue:23859,issue #23859: CUDA: THTensor code complains about devices not matching when creating tensor from blob,stale/low urgency,2,stale/low urgency,1,0,23859,,,,0,05dd9892a1fa28da031da9bd7d3cc5b7c5c49d2d8008df647a8e160813896148
thread:issue:23865,issue #23865: model->to(device) costs over a millisecond when doing nothing,stale/low urgency,2,stale/low urgency,1,0,23865,,,,0,62e10c6fb1dc6901d5c0b36910063fbfb2ce87b93d05826cc7e317234d6cfe53
thread:issue:23890,issue #23890: [JIT] Can't use ndim in script,stale/low urgency,2,stale/low urgency,1,0,23890,,,,0,e475d0dc98de0dbfadac31fa3b74b3c2fcb4d44950cda61c2287118feaf7b62e
thread:issue:23896,issue #23896: Remove USE_C10D flag,stale/low urgency,2,stale/low urgency,1,0,23896,,,,0,c4197e838b97a53d2d4c90b3eee6186b02704365d1e1661aa63eab4ad03ac8b7
thread:issue:23938,issue #23938: torch.nn.DataParallel causes incorrect gradients,stale/low urgency,2,stale/low urgency,1,0,23938,,,,0,7ee0cc3cdd093ba0ad24d50937f6331cca9fa8459f02fae6b1b5e761bf787f42
thread:issue:23958,issue #23958: [jit] Python @property's not supported in TorchScript,stale/low urgency,2,stale/low urgency,1,0,23958,,,,0,8ce0f336c2868ebd297483578fee7a7c8432789938e6e30a796167b0f0cc7a0f
thread:issue:2400,issue #2400: Add CRF Layer,stale/low urgency,2,stale/low urgency,1,0,2400,,,,0,8dc6bcaed6ad74251c7067abeb201dcfabb93ef2c7d15be488f38451f92512ba
thread:issue:24005,issue #24005: Using `torch.utils.checkpoint.checkpoint_sequential` and `torch.autograd.grad` breaks when used in combination with `DistributedDataParallel`,stale/low urgency,2,stale/low urgency,1,0,24005,,,,0,6571af263bdf10d92e3cba55fd5841b21ff5ee80a173c09b46056548c1c3ce26
thread:issue:24016,issue #24016: Multiplying a very large CUDA tensor with another tensor yields unexpected result,stale/low urgency,2,stale/low urgency,1,0,24016,,,,0,565d38642989562903175649e2365da7eda66c987a26ea98222853f1a507270a
thread:issue:24031,issue #24031: fractional_max_pool2d_with_indices silently ignores output_ratio if output_size is provided,stale/low urgency,2,stale/low urgency,1,0,24031,,,,0,9f08585ceebace98de6059f83f63423d6003fcfa8b5d2e4493dbec44bb7eca31
thread:issue:24037,issue #24037: Tests do not pass with the latest protobuf,stale/low urgency,2,stale/low urgency,1,0,24037,,,,0,764142b1f8b7f7a85b2cdc7291dc4a4d9fefb930a1c556cd69f857350acacadf
thread:issue:2407,issue #2407: Hard-negative mining using __getitem__ directive in Dataset class,stale/low urgency,2,stale/low urgency,1,0,2407,,,,0,8a346fe193921ba7c9acdbc49bebfee955ab96df523d2c95dca7d8cf8b8c454c
thread:issue:24079,issue #24079: [Caffe2] build android in v1.1.0 with headfile error,stale/low urgency,2,stale/low urgency,1,0,24079,,,,0,623cb2cde322b56bf2ee591e5d19afb2086c3f1b99dc87eabe7ce24be57c89c7
thread:issue:24081,issue #24081: Multi-gpu example freeze and is not killable,stale/low urgency,2,stale/low urgency,1,0,24081,,,,0,b9c29433e943e71e164e829e92d8813ae09eb6408c172de521e41c438f4fde8a
thread:issue:24089,issue #24089: torch.nn.functional.grid_sample with 'circular' border conditions,stale/low urgency,2,stale/low urgency,1,0,24089,,,,0,baccd2e3b13113cf8de1b48952bf7541ebe0e07aff27ab969f373bf37c510cb8
thread:issue:24102,issue #24102: [JIT] script doesn't convert dtypes back to torch.dtype from long,stale/low urgency,2,stale/low urgency,1,0,24102,,,,0,2dbf89c1beb9b26df021b246011421b31d8c0e69805893c1d516e1cdca3412b8
thread:issue:24130,issue #24130: [RPC] Add type annotations for RPC-related Python files,stale/low urgency,2,stale/low urgency,1,0,24130,,,,0,2d70086507f0a06f547c67ff5f85280403949bc42a5d2642b3eb04a49d6c3923
thread:issue:24139,issue #24139: Allow incompatible shapes in load_state_dict(strict=False),stale/low urgency,2,stale/low urgency,1,0,24139,,,,0,521b74f34e7a6efc1dd8ad843cf93071085ee6bf7801b54435d4dbf5eb8cd809
thread:issue:24155,"issue #24155: ""To compact weights again call flatten_parameters()"" is printed every step for every GPU",stale/low urgency,2,stale/low urgency,1,0,24155,,,,0,196f4c73410f037419ed38783a355d110f800ed06e21b9b74d818dbc9bfc4b1e
thread:issue:24167,issue #24167: Failed to build pytorch with NanoPi M4,stale/low urgency,2,stale/low urgency,1,0,24167,,,,0,e4e4e620859592a6e24fd51b4ef5f7c9c5319b7f3a4164b454962a11349ce017
thread:issue:24168,issue #24168: torch.unique is inconsistent with NumPy's unique,stale/low urgency,2,stale/low urgency,1,0,24168,,,,0,040588a21c46354787da9472548da18a050d6d3e149f3236a99a2421a2c6ea59
thread:issue:24176,issue #24176: torch.fft crash when used with nn.DataParallel,stale/low urgency,2,stale/low urgency,1,0,24176,,,,0,cc1cf6e4c84e022ef9867257246ec6664fc594ed3c37b40560e97fcf1041ea0b
thread:issue:24193,issue #24193: TensorIterator stubs are designed for merge conflicts.,stale/low urgency,2,stale/low urgency,1,0,24193,,,,0,05dd5c1866425d7e2f30c9696cba221dc48d4903b2ab419011425767d713eaad
thread:issue:24205,issue #24205: deprecate cuda arch 3.5/3.7 in nightlies,stale/low urgency,2,stale/low urgency,1,0,24205,,,,0,aadfcb6324d2d21098473707e1e830cad8169360a70aa906bf04b70bd9edcaee
thread:issue:24225,issue #24225: Allow forward method to be defined with .define() in new TorchScript API,stale/low urgency,2,stale/low urgency,1,0,24225,,,,0,a81117cf0d7cd2c2939ffa52865d505219699fae95026883cb530127b65e6a1e
thread:issue:24229,issue #24229: Confusing error message for Custom Class type mismatch,stale/low urgency,2,stale/low urgency,1,0,24229,,,,0,47f914870c26a25856fb25ec82142826fd168091cb55129a0f5b25ead16f3b5e
thread:issue:24236,issue #24236: Tensorboard: Add disable flag for debugging,stale/low urgency,2,stale/low urgency,1,0,24236,,,,0,2317bc8ae5887015301801404bfafa2d7af0ac7ea986972b58077235af19aef7
thread:issue:24248,issue #24248: Loading custom Torchscript C++ operators in python segfaults due to ABI compatibility issue between pytorch and libtorch,stale/low urgency,2,stale/low urgency,1,0,24248,,,,0,0dc3f8525c636c3b837f00db057e6907d0ef0fa68bbb5e2e49a4968cb36cfb65
thread:issue:24261,"issue #24261: ""PyTorch core"" thread local flag",stale/low urgency,2,stale/low urgency,1,0,24261,,,,0,898e10e1dee5b18ac23192297facb3b45428f25eaba0e946e78e5165c1f26256
thread:issue:24264,"issue #24264: TensorIterator ""builder"" options should be documented.",stale/low urgency,2,stale/low urgency,1,0,24264,,,,0,4075d4a7132e9c402e7fde3311d83f48922b5adc5ebdd81d32ebbac76c3d13ef
thread:issue:24273,issue #24273: Check PyTorch version when initializing process groups,stale/low urgency,2,stale/low urgency,1,0,24273,,,,0,ae525a1ce28a69c32589b0280416b402c74a9fe367fb942c7333ede197615969
thread:issue:24303,issue #24303: Port `masked_fill` operator from the TH code to Aten,stale/low urgency,2,stale/low urgency,1,0,24303,,,,0,074d34aab30e21724113673b0df115622b707caad2f1214382ef703b260ba276
thread:issue:24313,issue #24313: Value_select to perform region-wise selection,stale/low urgency,2,stale/low urgency,1,0,24313,,,,0,9df6fc1f9cae31f0bf2279494138ce25977d9ecc996b7aaad79d079c1cd58352
thread:issue:24336,issue #24336: Auto tuner takes too much time in serialized model,stale/low urgency,2,stale/low urgency,1,0,24336,,,,0,eb70d3b8e761cd3faa364e41efe65e3866f1c1efb9ab06485f038303aec0c824
thread:issue:24343,issue #24343: Unified representation for enum types,stale/low urgency,2,stale/low urgency,1,0,24343,,,,0,5be0fcac7907ec0fa415e4340c8e4696dce09fcb152bc37492bdab03bc703175
thread:issue:24353,issue #24353: Visual Studio Code not providing autosuggestions for submodules,stale/low urgency,2,stale/low urgency,1,0,24353,,,,0,2ea64530ca2018ca64d329a39dd9b8828052f7828b196731f32a8a71821015a2
thread:issue:24354,issue #24354: Default warning handler in C++ doesn't seem to unique warnings,stale/low urgency,2,stale/low urgency,1,0,24354,,,,0,5c817da4bef5f1237f4597103ebfcbaa4a56a15cbc72c19528f32f75365dbea1
thread:issue:24401,"issue #24401: When running model forward with large batch size, it reports the error: THCudaTensor sizes too large for THCDeviceTensor conversion",stale/low urgency,2,stale/low urgency,1,0,24401,,,,0,b0f6a556fbb041e5ae7cb6f82b51eb7013607051f506fad26c407ef22c62c49b
thread:issue:24417,issue #24417: 1.0rc0-6216 installs empty directories under include and duplicates under /,stale/low urgency,2,stale/low urgency,1,0,24417,,,,0,30dc1773fb4dd043431621715f13beec6c52caa616288cc0f080eab8fa7976d2
thread:issue:24419,issue #24419: Building Python bits separate from C++ bits and making one play well with the other,stale/low urgency,2,stale/low urgency,1,0,24419,,,,0,f11e14c864911d27dbef7b24a42f6fbd7f1800d7fd827adc5764ca873b60c696
thread:issue:24432,issue #24432: Use a ScriptModule on GPU that was saved from CPU,stale/low urgency,2,stale/low urgency,1,0,24432,,,,0,efe2e0cb220463952e0a1502eb1fbd0e1128f93a8ae4e3926bad0f14df68fb37
thread:issue:24456,issue #24456: Build PyTorch 1.2.0 occur `recipe for target bin/test_parallel' failed,stale/low urgency,2,stale/low urgency,1,0,24456,,,,0,116c333ef16f87732672e3126b8b378f7f9c1083f907b5b536670873025b4769
thread:issue:24463,issue #24463: Crashes on torch.cuda.memory_allocated(device),stale/low urgency,2,stale/low urgency,1,0,24463,,,,0,ab77fe3cc1293c97d9c78471e41571b7bc6e4f7f15f3b29a2244e85a833b181c
thread:issue:24468,issue #24468: [Tensorboard] Write summaries to S3 or GCS bucket,stale/low urgency,2,stale/low urgency,1,0,24468,,,,0,2e8f8d19af11b591145a79a2bdd3ea9bbd6c7d3a736cb0898e7227fd69edf10e
thread:issue:24500,issue #24500: Failed to compile PyTorch on IBM Power 9 architecture with CUDA 10,stale/low urgency,2,stale/low urgency,1,0,24500,,,,0,6db5a57d304bd47246ad1da142fc69bcc0d6096a2c3c7a307d23d8f88d37b6fd
thread:issue:2466,issue #2466: Autograd test failure on ppc64le,stale/low urgency,2,stale/low urgency,1,0,2466,,,,0,8c45b1971da7ee3e8371c118f25e8123ab116fe4a8cf2cbad56c8082590e2ef2
thread:issue:24786,issue #24786: [jit] Dict iterator invalidation doesn't match Python,stale/low urgency,2,stale/low urgency,1,0,24786,,,,0,8ab73425ffe8bb741629a85aeea6f53d66813d0db75e2c1824d9cd790ecbbda6
thread:issue:24798,issue #24798: Installs empty directories under Python's sitelibdir,stale/low urgency,2,stale/low urgency,1,0,24798,,,,0,ee432d8a9bfbff453a5ffdf7a16bee63900a89e1fe2cda1d2a8f9e12fdaaea00
thread:issue:24810,"issue #24810: Doesn't install the python module ""torch""",stale/low urgency,2,stale/low urgency,1,0,24810,,,,0,10d1a0e23eea3ae30a0b703f96401f33df8c6f6ac92556a445eb39f03093afb6
thread:issue:24831,issue #24831: subprocess.CalledProcessError: Compile source in NVIDIA TX2,stale/low urgency,2,stale/low urgency,1,0,24831,,,,0,c034867b6b8e94484e0691eab9000bc79a1c75fdf594f982d9141bc903b56139
thread:issue:24835,issue #24835: NetworkX's Version,stale/low urgency,2,stale/low urgency,1,0,24835,,,,0,cfa47a2cf0079cc443192fb2829c38fafcf036f1eebddfaed6e1c3a37b6ecd0c
thread:issue:24836,issue #24836: Gloo scatter gives wrong result for stride != 1,stale/low urgency,2,stale/low urgency,1,0,24836,,,,0,699bb2c9221a12e53840258b801fe8fc2c689ef0bd67101729238c37b12075d2
thread:issue:24946,issue #24946: [RPC] Make ProcessGroupAgent send task non-blocking,stale/low urgency,2,stale/low urgency,1,0,24946,,,,0,4ed0fe7e66777b0afb58d313fd2c4db5a038745027de446c5268c3a24612ca69
thread:issue:24956,issue #24956: [RPC] Fix logging initialization warning in ProcessGroupAgent,stale/low urgency,2,stale/low urgency,1,0,24956,,,,0,d38ae1c4ca406da898a8aa2f93ef871267fb4e87a232cc903ef0b15b6cc9b184
thread:issue:24963,issue #24963: torch.cuda.synchronize blocks CUDA execution on other threads using other devices.,stale/low urgency,2,stale/low urgency,1,0,24963,,,,0,0842f1de7b9a103d22c9aff39dbcdacaccf7a360c88dea479e681d1352607ec4
thread:issue:24984,issue #24984: DataLoader slow down when `pin_memory=False`,stale/low urgency,2,stale/low urgency,1,0,24984,,,,0,cfab7f057c1ea46ccee9466c1432494114d98ea5d8718ad8643435d72962662e
thread:issue:25004,issue #25004: [C++] Call find_package(Torch REQUIRED) more than one time in downstream project causes CMake configuration error,stale/low urgency,2,stale/low urgency,1,0,25004,,,,0,8bab68473d0d6ceff6439d7a43507505d616c1108912d4f330eead84d04989cd
thread:issue:25034,"issue #25034: make add_module accept tuples as well or change containers(ModuleList, Sequential, etc) to allow this",stale/low urgency,2,stale/low urgency,1,0,25034,,,,0,4576960dd25fb754816e925c77c8478b29be232c976a46c3e50f186d86e4d300
thread:issue:25045,issue #25045: [Distance functions] F.pdist backward CUDA invalid configuration,stale/low urgency,2,stale/low urgency,1,0,25045,,,,0,7cc6e51441d14bec74c0ff4626fe0474837ec7193e404f9244165db2e9de48e9
thread:issue:25047,issue #25047: ScriptModule and nn.Module parameter ordering difference,stale/low urgency,2,stale/low urgency,1,0,25047,,,,0,9bd8a2f4e5e681f2ccc6555aa50adf79767d09ac7dad9c4bc41307989f47ff79
thread:issue:25057,issue #25057: ctc_loss computes different losses and gradients on batched utterances vs. individual utterances,stale/low urgency,2,stale/low urgency,1,0,25057,,,,0,1b6bf22d66a1f6a43a14d0926c70b76d8040daf747a1b1ad8f6af25058799ad4
thread:issue:25070,issue #25070: [FR] torch.(Generator|random).seed allows specifying the seed value,stale/low urgency,2,stale/low urgency,1,0,25070,,,,0,0883a6639574db8c747c23eb09db8938beeff71ab22cc65269d85656ca1bdba7
thread:issue:25071,issue #25071: Generator objects should not always use the same seed,stale/low urgency,2,stale/low urgency,1,0,25071,,,,0,556fd9f7cb36c6a102a0869c14443cc68a8e1f9e4c3a4de8fa0aaf1d74dea336
thread:issue:25091,issue #25091: JavaScript (Web Assembly) target for trained models,stale/low urgency,2,stale/low urgency,1,0,25091,,,,0,11e9859679d77b0b60d6cd17257a288adf24e0dcad7a5e7622201e78d326ca43
thread:issue:25092,issue #25092: No way to correctly reset weights of a model with spectral norm,stale/low urgency,2,stale/low urgency,1,0,25092,,,,0,63e9d29639a7edc5aaad95b919f7cc9b6a99aa4ef443b21cef8a75bbb3abc98f
thread:issue:25112,issue #25112: [RFC] NestedTensor - 0.0.2,stale/low urgency,2,stale/low urgency,1,0,25112,,,,0,ce1d917e59b96b3c9610342f90fdf57938b59b91b8c6bb7ecaff312265b4ba7f
thread:issue:2512,issue #2512: detach_() variant that affects all past uses too,stale/low urgency,2,stale/low urgency,1,0,2512,,,,0,a920e18f20c9b1c92f51ef3ebe44ade19e236a6ab9e191305d0ad04c3fa3bbb8
thread:issue:25137,issue #25137: Serialization does not work for quantized modules,stale/low urgency,2,stale/low urgency,1,0,25137,,,,0,32ea46aa7835547114104384580abe86f4adfc4e0d4c3005fb2a172da4edaea5
thread:issue:25156,issue #25156: euclidean distance between two tensors,stale/low urgency,2,stale/low urgency,1,0,25156,,,,0,8266c60920709cd3eedf0197fd542006e0fb793545561a6939c3741d1c9b1f65
thread:issue:25297,issue #25297: torch.as_tensor(bytearray(...)) seems to leak memory,stale/low urgency,2,stale/low urgency,1,0,25297,,,,0,abe7f1bf194006d006b0c265c1f9ec8d1dbb41d61e4d5e0b07a0380c4a1a42bb
thread:issue:25314,issue #25314: ConstQuantizerPtr is misnamed,stale/low urgency,2,stale/low urgency,1,0,25314,,,,0,27e0ce814787506097cfc2ecd526b6736c26e30acae85719dc18af4572edcbec
thread:issue:25329,issue #25329: [jit] Bad error for incorrect container type,stale/low urgency,2,stale/low urgency,1,0,25329,,,,0,59bf483053693af9fe5887f31476114da477384dcfb2d3d8a70910ab3c87ecde
thread:issue:25381,issue #25381: regarding builtin_function_or_method,stale/low urgency,2,stale/low urgency,1,0,25381,,,,0,aeffe609308e5e9cc1065935b49e75106c7a428b7ab9f17500dd8100be3cfd82
thread:issue:25387,issue #25387: [DRAFT] Auto-casting in JIT - Automatic mixed precision,stale/low urgency,2,stale/low urgency,1,0,25387,,,,0,275fc67e08f466a2f9884e72bc142878126d21400bf1c9a160d4f2878b0c4189
thread:issue:25410,issue #25410: [jit] NamedTuples don't respect `__new__`,stale/low urgency,2,stale/low urgency,1,0,25410,,,,0,b9ebec90770b47f2e84224f9d6ea35200d3604fba08c32e99d67b13732b4a970
thread:issue:25417,issue #25417: Add a mode to check input tensor sizes in allreduce_coalesced,stale/low urgency,2,stale/low urgency,1,0,25417,,,,0,4151bb4b61c0acfe8981d1d302277a9893b4fa40cb0c4a99ac4ba600578a7f23
thread:issue:2545,issue #2545: Counter-intuitive Patience & Cooldown of ReduceLROnPlateau,stale/low urgency,2,stale/low urgency,1,0,2545,,,,0,d77c574d5cb0dd286492590c2444ac15615b72d72fa4c168ba833ddc0ea6af08
thread:issue:25460,issue #25460: Build link not right,stale/low urgency,2,stale/low urgency,1,0,25460,,,,0,1455279b6f54cc95988d699898cdbd203d9f602ea0490451b6b6e95e01b41554
thread:issue:25480,issue #25480: Vectorize bool operations,stale/low urgency,2,stale/low urgency,1,0,25480,,,,0,0e1052009ccae241ae5977e3ef4133832f7897e298a4c0ab3e4207f54d246473
thread:issue:25514,issue #25514: Multithreaded backpropagation with custom autograd.Functions,stale/low urgency,2,stale/low urgency,1,0,25514,,,,0,67bfe3c28e98146be7a0c8b5bd83fd18f25a23ed99d50e84a795e4183e6d5ab2
thread:issue:25520,issue #25520: pytorch c++ api cannot call operator() on torch::nn::Sequential,stale/low urgency,2,stale/low urgency,1,0,25520,,,,0,c7d79e98f334dd18815cc33d6c4bef3e0614de92a11cbdec11f8783d12b56493
thread:issue:25548,issue #25548: The inference speed of the torch c++ dynamic library compiled manually is slower than the torch library officially provided,stale/low urgency,2,stale/low urgency,1,0,25548,,,,0,ac5cdb2bb24ebc2037a945f36bf0c68783d8c5218486e8b19bfb3729bb5747bc
thread:issue:25552,issue #25552: Usage of DDP on a module that doesn't require gradients,stale/low urgency,2,stale/low urgency,1,0,25552,,,,0,de4e0ade86f59ccb10f633c3b88d73fa0681613f475ef3bb391f63bcefdae803
thread:issue:25570,issue #25570: Remote memory access similar to MPI one-sided in pytorch,stale/low urgency,2,stale/low urgency,1,0,25570,,,,0,924c494e51ce7d186e79087266e547c01fa4f67a188c4a682ebbd7c07904caee
thread:issue:25601,issue #25601: Avoid non-POD data in thread_local,stale/low urgency,2,stale/low urgency,1,0,25601,,,,0,1aaae5e459832dd648c26a8d8e1ed56c1bd6281b60ef0ec9e10993d70528ce83
thread:issue:25635,issue #25635: [C++] Support negative index in `torch::TensorAccessor::size()`,stale/low urgency,2,stale/low urgency,1,0,25635,,,,0,ee6b961cf2b2a508c24e607f028df8770b504805218e98aafa5153befa3bf04a
thread:issue:25641,issue #25641: [distributed] all_gather on a List of Tensors directly,stale/low urgency,2,stale/low urgency,1,0,25641,,,,0,c067c913779b3ba2ba05ddc693096c707d53d5822cc256dd862d36fa53a4cd6d
thread:issue:25646,issue #25646: libtorch forward memory leak,stale/low urgency,2,stale/low urgency,1,0,25646,,,,0,49d4694da6922e560bed0495356f6804fb28fef5dbb68b6d8da347eaa2b4a06e
thread:issue:25648,"issue #25648: Torch.jit.trace unexpected error with `torch.cat(…, dim=-1)`",stale/low urgency,2,stale/low urgency,1,0,25648,,,,0,f73ee573d23b3ddb927261db04418bfcef96f8ed95ceec4875d9874c1fbb861c
thread:issue:25661,issue #25661: finfo operator not bound into JIT,stale/low urgency,2,stale/low urgency,1,0,25661,,,,0,d02d2b72edb59e96e79d86f81b5486faef4c86b4ea66ebaf262722c9ecd49b10
thread:issue:2569,issue #2569: Add a [strong] hash function for tensor data,stale/low urgency,2,stale/low urgency,1,0,2569,,,,0,285a35a4eaca77664ad7f0efd2950601ffcc3c02ef6d2235dce384f19980eecc
thread:issue:25714,issue #25714: Backtrace prints many <unknown function>,stale/low urgency,2,stale/low urgency,1,0,25714,,,,0,0074d8050b68d9d85b948975aa6b485663b2d729e64fc338757065ff599fa715
thread:issue:25747,issue #25747: Int32 overflow in bincount indexing,stale/low urgency,2,stale/low urgency,1,0,25747,,,,0,e7a31a83456c104c89a43cee88a1ca43c21c1de6185ea8ef771b9b03cb3c9911
thread:issue:25752,issue #25752: torch.cuda.empty_cache() write data to gpu0,stale/low urgency,2,stale/low urgency,1,0,25752,,,,0,94e53582770c8a1380b6b35c0c17837769030ca114888c1c6807a06a62819134
thread:issue:25767,"issue #25767: Model parallel with DDP get `Socket Timeout` error when using NCCL, while GLOO works fine",stale/low urgency,2,stale/low urgency,1,0,25767,,,,0,72feb0076af70b340faf405bfba9e53cf5793507c064955faa818a89072ed1dc
thread:issue:25783,issue #25783: Detaching a distribution's `log_prob` to block gradients only w.r.t its parameters,stale/low urgency,2,stale/low urgency,1,0,25783,,,,0,2e9de035ff3c9d3beca96abacab1258083815cc6698062230cd1698f3c6665f1
thread:issue:25785,issue #25785: [Feature request] modified Cholesky decomposition,stale/low urgency,2,stale/low urgency,1,0,25785,,,,0,b4a4641dbe59da32159e72af9cbe5f8f31b6cabf50e19e5623a01bf0417d29c5
thread:issue:25803,issue #25803: [jit] `random` module support,stale/low urgency,2,stale/low urgency,1,0,25803,,,,0,584ad1eb71248b43fafa1342ec55d299c4d1434622ce4cb1b0345f47a13ade4f
thread:issue:25814,issue #25814: (PyTorch1.1 and 1.2) RuntimeError: Can't detach views in-place. Use detach() instead,stale/low urgency,2,stale/low urgency,1,0,25814,,,,0,daee8090ca3ecff5871baf30cf98762ecbd2ae79e2811f9fa409bff8bd502065
thread:issue:25820,issue #25820: Pytorch master can not build with computer capability 3.0 under Mac OS X 10.13.16 with Nvidia GT 750m,stale/low urgency,2,stale/low urgency,1,0,25820,,,,0,aba095320172312a78f381684aac4e74b913dcbfe04f24ef752bee2ebe844ec8
thread:issue:25821,issue #25821: Conv2D 2x~20x slower than Tensorflow when channel count is small,stale/low urgency,2,stale/low urgency,1,0,25821,,,,0,a470db96bf9619ae8463dd3380f8cf685192c8cea72a4b05e44452e91754702c
thread:issue:25880,issue #25880: Incorrect lable read with ImageInput Op of Caffe2,stale/low urgency,2,stale/low urgency,1,0,25880,,,,0,387889664af8565c4b6944b96789b45ac6196fa06ae75b185838c960a9ad7a0e
thread:issue:25892,issue #25892: load_state_dict on CPU first,stale/low urgency,2,stale/low urgency,1,0,25892,,,,0,fce7c3b2aed70e976ffaf572339da84a5f2183c06ffe4e6b8a4c8efeeb2e89b7
thread:issue:25941,issue #25941: TestAutograd.test_deep_reentrant fails with SIGBUS on macOS,stale/low urgency,2,stale/low urgency,1,0,25941,,,,0,58f0e558fcf42a264f30d2abc74692681b5285d8de05d582692f8acbff079040
thread:issue:25978,issue #25978: Provide a way to select SVD algorithm in PyTorch?,stale/low urgency,2,stale/low urgency,1,0,25978,,,,0,925efdb4312945896a07fb0d4c10d64ff851aa7503869c7b4edc1f9cf4cf4ac7
thread:issue:25979,issue #25979: [C++] `Module::pretty_print` is broken,stale/low urgency,2,stale/low urgency,1,0,25979,,,,0,8077315a1c2f0114a0db85e9ada2e67f837824fb8663f95744b2c57ef9e925df
thread:issue:25991,issue #25991: Why doc building isn't failing us for referring to a non-existent method?,stale/low urgency,2,stale/low urgency,1,0,25991,,,,0,0b7ab1802bab1664a6b998c9cd6d46856a43c6b60bf63d25c9a80a9caf4b0de6
thread:issue:26058,issue #26058: sccache stats can cause whole build to fail,stale/low urgency,2,stale/low urgency,1,0,26058,,,,0,036099a78a6e6f71a76e3086551000e44d129e818a7e484a911b80843b701f1c
thread:issue:26086,issue #26086: [Feature Request] Trace / Script C++ models,stale/low urgency,2,stale/low urgency,1,0,26086,,,,0,e6566a00938134f34505cb52007e7254f4b0869abb555692ec4827ad85b86d55
thread:issue:26097,issue #26097: [RFC] TensorBoard extensions and improvements for PyTorch,stale/low urgency,2,stale/low urgency,1,0,26097,,,,0,2762d1093c9623ee4ed393c5554047a6145390822a1421c2621274bdd0b8c6f0
thread:issue:26109,issue #26109: Support the AVX512 runtime dispatch,stale/low urgency,2,stale/low urgency,1,0,26109,,,,0,4c13c077866120f652366d40358292042e80d44e8dfea3773ed298694deb64ef
thread:issue:26110,issue #26110: Access data_ptr in RNN.cpp,stale/low urgency,2,stale/low urgency,1,0,26110,,,,0,4d1f5e182abf7eec022906555834a8c78bac0ac4c89ab46e64ec360f515421f1
thread:issue:26119,issue #26119: Make GloballyUniqueId a common type for both rpc and dist autograd,stale/low urgency,2,stale/low urgency,1,0,26119,,,,0,b3abb6f1ea09b64c8e96d7168bee84cd11374e0846789862e3c694255065199e
thread:issue:26136,issue #26136: There should be gating around BFloat16,stale/low urgency,2,stale/low urgency,1,0,26136,,,,0,b8b6de34f1050f53260b86f543d229e96361f55a7137ea302ef0218e971ff055
thread:issue:26139,issue #26139: c10 List API hard to use,stale/low urgency,2,stale/low urgency,1,0,26139,,,,0,c2f6ef9f4e3d3ab07ae767176b39bd2ee22f5186c2a559710a7e6ea2a3b46df2
thread:issue:26218,issue #26218: Default adam epsilon to 1e-7 when on fp16,stale/low urgency,2,stale/low urgency,1,0,26218,,,,0,d6ee440659166949dc384b43f9a09ee5484ba6c920a7fc1f7af0fcc90b5f7ec2
thread:issue:26228,issue #26228: Higher dimension support for `MultiLableSoftMarginLoss`,stale/low urgency,2,stale/low urgency,1,0,26228,,,,0,1e635835a882a0aa85d554b657d7a29a426d51d04aa141b74c05fe377f6fe8ad
thread:issue:26258,"issue #26258: ""git describe"" shows incorrect version 1.0 instead of 1.2",stale/low urgency,2,stale/low urgency,1,0,26258,,,,0,9c19e658b344f9fec1f25ec66b090e5d1035bd23af56622dc664ae82bfc06af7
thread:issue:2629,issue #2629: [feature request] add pairwise ranking loss,stale/low urgency,2,stale/low urgency,1,0,2629,,,,0,e175512ba7c66fe85271ff25e14d66007ecbe14e59881ccc4d19f0d178303408
thread:issue:26334,"issue #26334: Supporting ""cdf"" for Student-T distribution",stale/low urgency,2,stale/low urgency,1,0,26334,,,,0,cff6f9e4828c5c5e2aa18a3226ac0bb2e0cf8ecdfc2ac2ade535893255367108
thread:issue:26338,issue #26338: Behavior of F.dropout in eval mode,stale/low urgency,2,stale/low urgency,1,0,26338,,,,0,339df2b0778ca6d9341a723988a43d03b3144b3e0a0bea31b25b63f0aef7960e
thread:issue:26340,issue #26340: CPU version of PyTorch on PyPI,stale/low urgency,2,stale/low urgency,1,0,26340,,,,0,96c8609c9fc6328cbdf45cab39d7437222958c00f2b870e66f5c830216aae654
thread:issue:26344,issue #26344: How to get rid of zombie processes using torch.multiprocessing.Pool?,stale/low urgency,2,stale/low urgency,1,0,26344,,,,0,608e6dcf2b3bd490e918d44c1cad6a8a662ee6815e3bfb10dd625a677e896cf1
thread:issue:26354,issue #26354: torch.tensor / torch.as_tensor not working with list of tensors,stale/low urgency,2,stale/low urgency,1,0,26354,,,,0,8b7e75d17c2bc1bd695536f0a43aeb77f5bae47ceb77c22c2bd4e0525269f8bf
thread:issue:26379,issue #26379: DistAutogradContext should be cleaned up in case of node failures.,stale/low urgency,2,stale/low urgency,1,0,26379,,,,0,d399b02f4339ae295bbdeaea4c40b2b0e684aa78789e10f0ca728e3e7376112d
thread:issue:26396,issue #26396: NNPACK condition should be changed (ARM processors),stale/low urgency,2,stale/low urgency,1,0,26396,,,,0,2611d46f317616b01febca4353ced60674724a75b9cbd1b206b9f13778c71ded
thread:issue:26409,issue #26409: Generated file not getting cleaned up by clean,stale/low urgency,2,stale/low urgency,1,0,26409,,,,0,d070632cc04cdecc2c6fd400ee6f9075464c1163cc315cf5085ce5d39175e261
thread:issue:26428,issue #26428: Remove TensorOptions logic from generated code,stale/low urgency,2,stale/low urgency,1,0,26428,,,,0,de6d7f11b5a5109842996921f28cca8c69bb2127eb875e9c6206b08b1f9ed485
thread:issue:26446,issue #26446: Avoid sending zero grads over the wire in distributed autograd backward pass,stale/low urgency,2,stale/low urgency,1,0,26446,,,,0,9fa750d30a4782d8b8f09fc46820ef1844881f91fe93ba1720faad1c9d7ce593
thread:issue:26447,issue #26447: parallel_for may hang when called in main process and then on daemon process,stale/low urgency,2,stale/low urgency,1,0,26447,,,,0,f9c18165a31e1da3fac7e117f00926766721fd45f0aa75110c55742bdd9120dc
thread:issue:26456,issue #26456: [libtorch]Same model in CUDA and CPU got different result?,stale/low urgency,2,stale/low urgency,1,0,26456,,,,0,1f5d6fd52d997d1e4fb274667e002d09e084a3fd66ee615acd53896df4aed19f
thread:issue:26527,issue #26527: Multilinear map,stale/low urgency,2,stale/low urgency,1,0,26527,,,,0,f16e386162a5237d9f6edce1560adf5d3c66cf5bacbdfdd54411e8a2f722ca58
thread:issue:26534,issue #26534: MKLDNN+AMD BLIS path for PyTorch,stale/low urgency,2,stale/low urgency,1,0,26534,,,,0,937bc720c5107f142d1dcbef21b264c39193296bab1ad1da5066de40b7fcc0d4
thread:issue:26551,issue #26551: No way to disable mse_loss broadcasting warning,stale/low urgency,2,stale/low urgency,1,0,26551,,,,0,6a5f3a0f046c03355a83642de42055ceb453836d662ddb0a857e5c5b30a24305
thread:issue:26567,issue #26567: Make `torch.save` serialize a zip file,stale/low urgency,2,stale/low urgency,1,0,26567,,,,0,c3c8f1e841e4d55a970186c06db94d488ae7bb2bad85ee79203e13470aa91252
thread:issue:26659,issue #26659: torch::nn::Sequential not compatible with torch::nn::RNN,stale/low urgency,2,stale/low urgency,1,0,26659,,,,0,8e0d67d8d834547a68f5384d7b9ab9f86eef0d59f90b8c3b7524f4a540e2934e
thread:issue:26691,issue #26691: Support FPGA Xilinx,stale/low urgency,2,stale/low urgency,1,0,26691,,,,0,f555f035f6563e36a9bad3076166b9047413bb193fad84740a3851b1aa332573
thread:issue:26730,issue #26730: [jit] Bad error when instantiating TorchScript class with incorrect types,stale/low urgency,2,stale/low urgency,1,0,26730,,,,0,026b3d1be7162b446c9130b5f9d510a8bd85156788522f014badffa3af727f71
thread:issue:26744,issue #26744: [jit] Default args don't work with TorchScript classes,stale/low urgency,2,stale/low urgency,1,0,26744,,,,0,667922da6a9227b1138e74155fa7f561c8a400b654c16efaef99c7e87abdf5bd
thread:issue:26781,issue #26781: Matrix corresponding to convolution by a 2D kernel (convmtx2),stale/low urgency,2,stale/low urgency,1,0,26781,,,,0,759ae8747ed93d36d4eb49ba93402b602b94730af9c91da58acb5f39b1f32cd3
thread:issue:26790,issue #26790: Importing tensorboard jams CUDA device selection,stale/low urgency,2,stale/low urgency,1,0,26790,,,,0,4d7a58194a261d36da6593823871cac5c711069c1e311c388244c0e730c3a579
thread:issue:26796,issue #26796: setup.py install error,stale/low urgency,2,stale/low urgency,1,0,26796,,,,0,0be1bbfebd7b8cc7d41955a72f1a97179ac8df697103d2e767319bcecab60245
thread:issue:26802,issue #26802: Google Summer of Code,stale/low urgency,2,stale/low urgency,1,0,26802,,,,0,16e2fb4fb006b01c3988c394d8a86f2ff98da85c912b3b6c95fa4afe0fba7f04
thread:issue:26863,issue #26863: [BUG Report]Integrate libtorch to ffmpeg but memory leak happened!,stale/low urgency,2,stale/low urgency,1,0,26863,,,,0,5ee2a00c9b871d8cb6f3a9ecc251cf903118f7248ef3aaf9a8e8135e02f472be
thread:issue:2691,issue #2691: Feature Request: Support grad of grad in fused RNNs,stale/low urgency,2,stale/low urgency,1,0,2691,,,,0,13f6001df9183b84281ed3b29cb84519a1722935a9b6efacd4efadc802846df6
thread:issue:26964,issue #26964: AnyValueTest.CorrectlyAccessesIntWhenCorrectType UBSAN failure: owncast of address 0x60300105d750 which does not point to an object of type 'Holder<const int>' Sep 27 00:01:03 0x60300105d750: note: object is of type 'torch::nn::AnyModule::Value::Holder<int>',stale/low urgency,2,stale/low urgency,1,0,26964,,,,0,9741b8a9820091f6f6994dff3683e1387d5dd1efcd1502727cc49232d25204af
thread:issue:26967,issue #26967: Lint rule to test for creation of tensor in native/ without options(),stale/low urgency,2,stale/low urgency,1,0,26967,,,,0,85589102c96f010dee58186f7b220f6db721d48a0c05d46f25ad237d1aeb1ad8
thread:issue:26990,issue #26990: torch.quantized.modules.floatFunctional -- add a little color commentary to the doc,stale/low urgency,2,stale/low urgency,1,0,26990,,,,0,6f7e1afff9742dbc64ec3843d64823432721b2d9a7ba9b619f7e3151247db8cf
thread:issue:27003,issue #27003: [JIT] Script doesn't preserve builtin torch named tuples upon python return,stale/low urgency,2,stale/low urgency,1,0,27003,,,,0,e37ff6feacd39bf6dec8d31637b613198e06278757bece4cae082aa096ad47ca
thread:issue:27015,issue #27015: Missing bin and include when building with torchvision on CentOS,stale/low urgency,2,stale/low urgency,1,0,27015,,,,0,c38e2f12f42ff09027243313d694505c6c00c48fc95bb5fc6fa88605c51e6275
thread:issue:27021,issue #27021: Cmake warnings during build,stale/low urgency,2,stale/low urgency,1,0,27021,,,,0,e6bdea0fbf3b4d018db4e460c5136ae143ef870e81cdbc8ed57b70361c1b936c
thread:issue:27024,issue #27024: Is it an incompleted dst tensor synchronization in CUDA device to device copy ?,stale/low urgency,2,stale/low urgency,1,0,27024,,,,0,b6a51eb01409ba8b705d113cfcda18da419427ce1e60c330c44311e76ad8b972
thread:issue:2703,issue #2703: Have ppc64le docker images?,stale/low urgency,2,stale/low urgency,1,0,2703,,,,0,334991d4e7b9c9cdea2194503056baf62d5742160f9bf3d8a803d1cee1159187
thread:issue:27030,issue #27030: Tensorboard logging image WITH LABEL,stale/low urgency,2,stale/low urgency,1,0,27030,,,,0,f7834afc6867e9fd1ca30438948232d4ec1eaa4e708e2f6c4e69116591be7a95
thread:issue:27034,issue #27034: RuntimeError:[enforce fail at context.h:48] option.device_type() ==PROTO_CPU. 1vs0,stale/low urgency,2,stale/low urgency,1,0,27034,,,,0,83f61eaf14712a03b9b0939d34c333cd4581b5af143b5921bd9d50e35ce4be5d
thread:issue:27046,issue #27046: The Gather problem in DataParallel: dimension are not matched.,stale/low urgency,2,stale/low urgency,1,0,27046,,,,0,84ec6280c4dea637a8ebd6dc62517d811c66410be2187cf712abf499680be06b
thread:issue:27055,issue #27055: torch::NoGradGuard no_grad get wrong when I use batchsize!=1,stale/low urgency,2,stale/low urgency,1,0,27055,,,,0,6b507370182a008d015cfe9920d433bc1acc9f2116dfb8e3c00c2ec91aa83f50
thread:issue:27094,issue #27094: [jit] Document what types can be traced,stale/low urgency,2,stale/low urgency,1,0,27094,,,,0,b2fc9ec41e6e77bcb0f1a6ab4672058fa4bb66ba188aff159acaaa16bb2c3c6e
thread:issue:27138,issue #27138: CUDAPytorchToCaffe2.MutualResizes is flaky,stale/low urgency,2,stale/low urgency,1,0,27138,,,,0,418a89e57e2f394146d23f83207b6070ea9533884c91c066a7183034ab1dde6a
thread:issue:27147,issue #27147: Stop binding in-place methods to `torch.*`,stale/low urgency,2,stale/low urgency,1,0,27147,,,,0,666ec4681f93b00dd733c3c5566b8744616564d9b116e4f95fbabe64de94666e
thread:issue:27175,issue #27175: Easier way to create tensors with names,stale/low urgency,2,stale/low urgency,1,0,27175,,,,0,e4fc0423ad53b5d2edae200dc4dd8cad6e5db5df8dc5ddd1555185aeb3539a19
thread:issue:27218,issue #27218: pytorch data_parallel oom on gpu:0,stale/low urgency,2,stale/low urgency,1,0,27218,,,,0,4791f17ba71d640782da2711143b9c776200ba5c7956b47d0ddcf28cb23ae61d
thread:issue:2724,issue #2724: DataLoader gets stuck after model initialization,stale/low urgency,2,stale/low urgency,1,0,2724,,,,0,fb1cbb0f3755e9c843e403a9c3d1af8fb6b12ca617a732aca516d04fcfdabf67
thread:issue:27276,issue #27276: Add in-place view (view_),stale/low urgency,2,stale/low urgency,1,0,27276,,,,0,3e62d7ab248c5b37ac49c299269d810599eccba11cbd3c25e63dfe01700ec36a
thread:issue:27297,issue #27297: CUDA error: device-side assert triggered(insert_events at /pytorch/c10/cuda/CUDACachingAllocator.cpp:569),stale/low urgency,2,stale/low urgency,1,0,27297,,,,0,7e49ee638d4a2d399ab0ce09041a1c74c9c3257628443a47ded125204b337684
thread:issue:27300,"issue #27300: Tensorboard add image with boxes, labels, and confidence scores.",stale/low urgency,2,stale/low urgency,1,0,27300,,,,0,e7a6ff9372eb5852684fbdde9ed893bb1321b31ac635bc0b031a67379ee90350
thread:issue:27336,issue #27336: Allow one inferred axis in torch.split,stale/low urgency,2,stale/low urgency,1,0,27336,,,,0,3d96695163a62e933f0388a873854e789f292eab30c9c903903ee638dbd78831
thread:issue:27343,issue #27343: [jit] `ScriptFunction`s are loaded as `ScriptModule`s,stale/low urgency,2,stale/low urgency,1,0,27343,,,,0,336d57e08a3d35dbd18aaeebe3a2adfd6696e5f32dca4dac60b7179921a25372
thread:issue:2739,issue #2739: Proposal: simplify overloaded Tensor function signatures,stale/low urgency,2,stale/low urgency,1,0,2739,,,,0,fe949c3562783cb4a024d6997cfdbaa763656200fa0fa55d155d2e5852675720
thread:issue:27406,issue #27406: [jit] String frontend doesn't support default arg values,stale/low urgency,2,stale/low urgency,1,0,27406,,,,0,90a984f3bb096a2135f78d810ce1dd36af9e3fdd58ddd934fa200725d374c9a2
thread:issue:27409,issue #27409: Installer not setting rpath for MAGMA (OS X w/ GPU),stale/low urgency,2,stale/low urgency,1,0,27409,,,,0,916006e8b4d496680176444b81f2501c454b02eef9d96c53cc9beda5801cd02e
thread:issue:27421,issue #27421: [dataloader] Sampler abstract constructor API minor proposal,stale/low urgency,2,stale/low urgency,1,0,27421,,,,0,f2f5a5411945da800c5ef8193599f1019844b4b781131dfd0345daccf8a9cbff
thread:issue:27497,issue #27497: TSAN failure related to mkldnn,stale/low urgency,2,stale/low urgency,1,0,27497,,,,0,7688b771a57f66322df47273b6662118bf0c4832149c31392b7770cdc1862796
thread:issue:27505,issue #27505: [JIT] ndimension not supported,stale/low urgency,2,stale/low urgency,1,0,27505,,,,0,00b2f2a7e1692482f5ac93016bf52bbd420af4447f10217b206705af20671d0b
thread:issue:27517,issue #27517: Allow explicit gradients in torch.distributed.autograd.backward() API,stale/low urgency,2,stale/low urgency,1,0,27517,,,,0,8c2f0ecc27ee0eb50abd8b1e87bbe610364ab8f4d9d42696e30d77c3851d4871
thread:issue:27540,issue #27540: CTCLoss cuda backend large batch handling takes up to 1.8x more memory,stale/low urgency,2,stale/low urgency,1,0,27540,,,,0,2a20f4975e50e6d98651b14a845e6a009b202d0948e6f336341e981548c458d3
thread:issue:27542,issue #27542: Make topk sort stable,stale/low urgency,2,stale/low urgency,1,0,27542,,,,0,1963152ff71c164fee745c5746bca4cf7c3365de04dca414ae19da9bda2764f1
thread:issue:27579,issue #27579: Support Anomaly detection for distributed autograd.,stale/low urgency,2,stale/low urgency,1,0,27579,,,,0,826badff247817d2cd1aeec439748e5783bcb4b35c263388cc924d6781503c2d
thread:issue:27588,issue #27588: RuntimeError: cuDNN error: CUDNN_STATUS_MAPPING_ERROR,stale/low urgency,2,stale/low urgency,1,0,27588,,,,0,73a5b164028f4f4342919cc57cab3ca7723ca666c11bb0d133ef76b047e8e8c9
thread:issue:27608,issue #27608: Unify warning logging mechanism,stale/low urgency,2,stale/low urgency,1,0,27608,,,,0,6f5631c905169d3610cdefe6bae052c8811a488167f4113fb1795fd62af51380
thread:issue:27641,issue #27641: torch.distributed.autograd.backward() should populate .grad field on Tensors by default.,stale/low urgency,2,stale/low urgency,1,0,27641,,,,0,bb1b1c985073d1a51518ab8d45fe336a39fab11f58724f01349299b0090a7483
thread:issue:27642,issue #27642: Test re-entrant backward works with torch.distributed.autograd.backward(),stale/low urgency,2,stale/low urgency,1,0,27642,,,,0,2ae2f119f41c48add8b71f08d415221a084eb7087e480e3b6410d2eab9fc2f0e
thread:issue:27671,issue #27671: Avoid RTTI in DistEngine,stale/low urgency,2,stale/low urgency,1,0,27671,,,,0,fcf8fbba1005860dffbce5016c6c95f7ec69b5c888323835bdda2c28fbe3cad5
thread:issue:27687,issue #27687: `nn.Sequential.__setattr__` appends to the execution list,stale/low urgency,2,stale/low urgency,1,0,27687,,,,0,de988259fb13feea764fe208ea225e72d5890dbb80fafce882af2c2d651deb57
thread:issue:27708,issue #27708: Documentation makefile should include torchvision,stale/low urgency,2,stale/low urgency,1,0,27708,,,,0,8fcf9d2e6a89ae3320db4dc29d2443f781e6fa696a6026af5e9dfedabc2eb6f1
thread:issue:27740,issue #27740: Scripting torchvision.models.detection.maskrcnn_resnet50_fpn,stale/low urgency,2,stale/low urgency,1,0,27740,,,,0,c00768d4014de2a05a8a7d32aab782bd1692fa21b1dc2f4fda98b5016111dc1f
thread:issue:27752,issue #27752: Version number is still duplicated in a bunch of places,stale/low urgency,2,stale/low urgency,1,0,27752,,,,0,ccf60e84926f0eac3bebeee3d297465f333e2a3b8612815e1a434806bc744d5b
thread:issue:27780,issue #27780: torch.cuda.default_generators documentation referenced but don't exist.,stale/low urgency,2,stale/low urgency,1,0,27780,,,,0,ea88aa3218a262f29c722abd46418a55c07ca2ac89e02c13e748d0dc34045bdf
thread:issue:27815,issue #27815: Improve the error message when trying to install in a 32-bit Python environment,stale/low urgency,2,stale/low urgency,1,0,27815,,,,0,3317fc1ce52c7675b983c219a108f34ae07802d8180ae27e587fe58e8aa103ec
thread:issue:27839,issue #27839: RuntimeError: invalid argument 0: Sizes of tensors must match except in dimension 1. Got 24 and 195 in dimension 0 at /opt/conda/conda-bld/pytorch_1565272271120/work/aten/src/TH/generic/THTensor.cpp:689,stale/low urgency,2,stale/low urgency,1,0,27839,,,,0,3819eb94858416b6407b303978e4ff04ebdae08f02ca2049f78076fc8e27daa8
thread:issue:27840,issue #27840: caffe2 install VS2019 CUDA 10.1 lib\torch.lib : fatal error LNK1248: image size (10028FA9F) exceeds maximum allowable size (FFFFFFFF),stale/low urgency,2,stale/low urgency,1,0,27840,,,,0,ea47dc832eaed7e85454a8757b37ef16a49d2e1b9e59136cb06348516703ff3f
thread:issue:27843,issue #27843: Be able to build torch.distributed documentation easier,stale/low urgency,2,stale/low urgency,1,0,27843,,,,0,cf436ac739f14fcc3d8426336533cafd45d0c289871a6875d6f004cd7c5c050e
thread:issue:27926,issue #27926: `num_batches_tracked` update in `_BatchNorm` forward should be a single scalar update on host regardless of the residence of the layer,stale/low urgency,2,stale/low urgency,1,0,27926,,,,0,338480f143ea461f9768f12e322102316fcb362e5df9015f91f181fde7ecc030
thread:issue:27936,issue #27936: Specifying `pos_weight` in F.binary_cross_entropy_with_logits leads to RuntimeError: class size not match,stale/low urgency,2,stale/low urgency,1,0,27936,,,,0,932380dc1fd0b2df48cdc3f07071d1b414d357348e675df6ab364abd20154730
thread:issue:27955,issue #27955: Expand Pytorch C10D backend to dynamic load third party communication library,stale/low urgency,2,stale/low urgency,1,0,27955,,,,0,063c25c7b9b37110d7236731bba145e7efcc53b43823da188014d17b3d23b181
thread:issue:27958,issue #27958: how to use libtorch library in cuda file with nvcc compiler(c++)?,stale/low urgency,2,stale/low urgency,1,0,27958,,,,0,121a14f592ef73d92fb6e388652d92dec66f6d673f88dfce9a3b75985b49519a
thread:issue:27970,issue #27970: can't load model on cuda after call cudaDeviceReset functions,stale/low urgency,2,stale/low urgency,1,0,27970,,,,0,dd787cce155cadf4b99bfc9feda7c4a0fbd855313959079cd135a0ba878fbe69
thread:issue:27975,"issue #27975: Sending CUDA tensors via queue between processes, memory of Consumer process grows infinitely",stale/low urgency,2,stale/low urgency,1,0,27975,,,,0,f2592290465c94aac161288264e91cc599a1c623e4f9d57a955377d495345d42
thread:issue:28055,issue #28055: Traced resnet101 leaks memory during `forward`,stale/low urgency,2,stale/low urgency,1,0,28055,,,,0,ce3ac421ae8fdbc25d520aa768bfed2decf56c94a56550ba6923456fce7c0f01
thread:issue:28093,issue #28093: Move options logic in _like functions from parsing layer to implementations,stale/low urgency,2,stale/low urgency,1,0,28093,,,,0,010fb120d0a1d0cf3d6f3ae84dc0844b52ef8ea0ee29452dc81055a0ebd7c721
thread:issue:28110,issue #28110: Stable docs show dead __config__ section link,stale/low urgency,2,stale/low urgency,1,0,28110,,,,0,b22b02bc65e835b6d3453f14f888250ee2ac6c86c4582e1ade3ca9325cb1d7fe
thread:issue:28142,issue #28142: Named Tensor: Support -1 in `unflatten`.,stale/low urgency,2,stale/low urgency,1,0,28142,,,,0,0432f83c751e3c92b6cca729159dcdc25aaa1c792e95ef54b3121ff083fa0494
thread:issue:28146,issue #28146: Named tensor: align_to align_as error messages,stale/low urgency,2,stale/low urgency,1,0,28146,,,,0,c2072cd9ede7224efedf90a640cb88c7ed840c42b2aab925b9878d7ab54c181a
thread:issue:28194,issue #28194: New Stochastic Optimization Algorithms in Pytorch,stale/low urgency,2,stale/low urgency,1,0,28194,,,,0,b40915c23d7af38f0a427af53c0207ea5f67217d94b3cf72906ec70777341abc
thread:issue:28218,"issue #28218: index_sub, index_mul and index_div",stale/low urgency,2,stale/low urgency,1,0,28218,,,,0,425c237ab2695f9fa1859a28d18e79587ffa34627aa7d132c8b7fc337fe9864d
thread:issue:28220,issue #28220: TorchScript custom ops like `cuda` `byte` etc. doesn't support memory_format argument,stale/low urgency,2,stale/low urgency,1,0,28220,,,,0,9404e8b07efd95104b94dd69acb77d612bf27a9306ae0532819091335910af57
thread:issue:28221,issue #28221: TorchScript doesn't support torch.channels_last or any other memory format constants,stale/low urgency,2,stale/low urgency,1,0,28221,,,,0,fd07bd0f684b23441331e617580f8bc83ed2653f1ad8b8eee36ba656e1d53886
thread:issue:28223,issue #28223: Convert manually bound `cuda` `cpu` `byte` `float` operators to native_functions,stale/low urgency,2,stale/low urgency,1,0,28223,,,,0,c7cd72be268e6943152a9f655d46a02fa406275f667dc3648f8f2b073d328f59
thread:issue:28233,issue #28233: torch native functions cannot be used with inspect.signature,stale/low urgency,2,stale/low urgency,1,0,28233,,,,0,d39883046e5c460f1a20fe807fcf877b45f24981d030bc782af9166543b31a1d
thread:issue:28239,issue #28239: support class annotations in __init__,stale/low urgency,2,stale/low urgency,1,0,28239,,,,0,e256570294ba03deff1c813abfa333294e3d29bfe1f4fa25bc105b5264822b63
thread:issue:28244,issue #28244: ProcessGroupMPI reports incorrect world size,stale/low urgency,2,stale/low urgency,1,0,28244,,,,0,be2316df71b78173fb588fcf6869de90ab1bea06909fa0d9ccfa94f9af7899d6
thread:issue:28249,issue #28249: Add scopes to autograd profiler,stale/low urgency,2,stale/low urgency,1,0,28249,,,,0,9200d2c5974be5879b60912286bce9d57ed2f4a622af8ed6d54cf01c538b594e
thread:issue:28258,issue #28258: [jit] TorchScript classes don't work in notebooks,stale/low urgency,2,stale/low urgency,1,0,28258,,,,0,93d04a7fafbe92bf603f4882a15d8be500727581b0fa368c70906a8aeddfa74f
thread:issue:28275,issue #28275: nn.parallel.replicate in v1.1+ is much slower than v1.0,stale/low urgency,2,stale/low urgency,1,0,28275,,,,0,cecfdf1baf9fb8bd76645d90e19f7d8d88682967ebae7b608b7d73c1bd3f50fe
thread:issue:28390,issue #28390: torch.multinominal ignores elements from cumulative distribution,stale/low urgency,2,stale/low urgency,1,0,28390,,,,0,1aa2f4778bbf2017f2db644603ef5d572f1f60b57be19727efc94bf495de0f26
thread:issue:28441,issue #28441: gradient of Dirichlet.log_prob gives nan,stale/low urgency,2,stale/low urgency,1,0,28441,,,,0,9deb85b50d61174e76dc2b0b74395df315c0165ee5d175d4d0e8997fe8f877de
thread:issue:2849,issue #2849: Learning rate scheduler have different APIs,stale/low urgency,2,stale/low urgency,1,0,2849,,,,0,1b64a0d04c23d769adb6c1fff2c676ab469ed674d6396d6e61c121c700291c9f
thread:issue:285,issue #285: Keyword arguments passed to module's __call__ aren't forwarded to the hooks,stale/low urgency,2,stale/low urgency,1,0,285,,,,0,a6d4dfa895a000f754a1de8e0af295cbf6e3ac6e8b0a30e4a40486733cc5436f
thread:issue:28515,issue #28515: Get rid of libc10.so,stale/low urgency,2,stale/low urgency,1,0,28515,,,,0,fa7e7d5076d72ed977898aeb6e774d959a52bb22c934c058c239da23ef5395f0
thread:issue:28517,issue #28517: [JIT] __repr__ support for ScriptModules,stale/low urgency,2,stale/low urgency,1,0,28517,,,,0,68884878448db97eb8c738412d35aa634461070f6805184951fef107c59b2343
thread:issue:28519,issue #28519: [jit] C++ Documentation,stale/low urgency,2,stale/low urgency,1,0,28519,,,,0,ea7ef30678c8f392787ae8cfad97ec440336dad7b03d6324914dfe2308219e0a
thread:issue:28532,issue #28532: TestTorch.test_c10_layer_norm fails if you run it on a build of PyTorch with BUILD_CAFFE2_OPS=0,stale/low urgency,2,stale/low urgency,1,0,28532,,,,0,a2d3d6c77647e07e4ae93be938fe56bc65c7ae39c3cc022684bd4260c6e4cb9b
thread:issue:2854,issue #2854: Add the new lr_scheduler which called poly,stale/low urgency,2,stale/low urgency,1,0,2854,,,,0,7a1ce746b56f94d15b269e16654e1902f68b85711262d4f0d76e7c4f9d373f63
thread:issue:28549,issue #28549: Expose DifferentiableGraphBackward to python,stale/low urgency,2,stale/low urgency,1,0,28549,,,,0,bcd4a0dcf54ec6db457af8e5bee7fe6a69b130f86d15dc8c6a16a248e28ba9b3
thread:issue:28552,issue #28552: Doing rendezvous twice can cause hangs,stale/low urgency,2,stale/low urgency,1,0,28552,,,,0,cd334c22835b619c12d4c3c3373d6f79e1236de1b34008a1fc96b107c6a68e9d
thread:issue:28558,issue #28558: We should suggest using as_strided_ instead of set_,stale/low urgency,2,stale/low urgency,1,0,28558,,,,0,07a871a921569fd07e7ffdc9a6d52c689530163f758dae8c2819179fc2b152ac
thread:issue:28577,issue #28577: `Tensor.__reversed__` breaks protocol for reversible objects,stale/low urgency,2,stale/low urgency,1,0,28577,,,,0,c0121b3663dc6f45195a2ebe6667a12036630e27e78c4e65e02bf35f5b09d41e
thread:issue:28655,issue #28655: Cmake fails due to bad python call,stale/low urgency,2,stale/low urgency,1,0,28655,,,,0,a1655375659f37634245a50cc3161ef8018b21488e2fbc584a85efd8256a9485
thread:issue:28721,issue #28721: [docs] Improve docs of nn.MultiheadAttention,stale/low urgency,2,stale/low urgency,1,0,28721,,,,0,5308a0bfd1522c8c53a6c52c5a19f663aada55278d90d77d272a1a1802867073
thread:issue:28742,issue #28742: torch.as_tensor fails to create named tensors,stale/low urgency,2,stale/low urgency,1,0,28742,,,,0,8e19f5b5edf2935774cb183b08fad0ab2b042bca264ee348d93b3d6d21050d14
thread:issue:28745,issue #28745: Problem when installing Pytorch from source on CentOS 7.4,stale/low urgency,2,stale/low urgency,1,0,28745,,,,0,09f5605e507ad5fa4667db5caecf4d4c1c061149361b2ac93647fa61be3bd740
thread:issue:28746,issue #28746: super().__init__() not called in torch.nn.Module.__init__,stale/low urgency,2,stale/low urgency,1,0,28746,,,,0,c2aee6874ca2b6b88b0ceaa00ce0682cbe27fedc1a8e4e9675ef45e3cced50a4
thread:issue:28754,issue #28754: ABI backwards compatibility,stale/low urgency,2,stale/low urgency,1,0,28754,,,,0,eb0ecd90a8ce89a205ccd15d246f70c204fb90a463100ed2cb32d65af14866a8
thread:issue:28756,issue #28756: cmake allows both MKL and MKLDNN to be OFF; aten/src/ATen/CMakeLists.txt then ignores c++ sources,stale/low urgency,2,stale/low urgency,1,0,28756,,,,0,82adf5131077d6f58cc2537905cc50b01cee928989339e60cb3ba0912be99490
thread:issue:28757,issue #28757: [feature request] Docs for fuse_conv_bn_eval,stale/low urgency,2,stale/low urgency,1,0,28757,,,,0,14dd108e0d403844fd7cf19bde1b81a81cd0f653a3fa158d567b1e257fd24c68
thread:issue:28761,issue #28761: torch.tensor() is very slow when it is passed an h5py Dataset.,stale/low urgency,2,stale/low urgency,1,0,28761,,,,0,c089cfc276681674ff315f5b8b0bbc984d8372f9b77382cddb4ab4e6fe2d9573
thread:issue:28777,issue #28777: cudnn.determinstic=True causes dilated convolution to be >10x slower,stale/low urgency,2,stale/low urgency,1,0,28777,,,,0,0fe8a7c1640050f8ce54d36b1cd716baa414721869f7be505b9f1787dfe129dd
thread:issue:28780,issue #28780: 'import torch` fails with Illegal instruction,stale/low urgency,2,stale/low urgency,1,0,28780,,,,0,a539c0648f54916b8b2df270337d45b75c6e9f901276d9a604a1a3450831ddf9
thread:issue:28791,issue #28791: [CUDA-MEMCHECK] TestTorchDeviceTypeCUDA.test_pin_memory_from_constructor_cuda fails,stale/low urgency,2,stale/low urgency,1,0,28791,,,,0,b7f02a4ffc212e24173a4913ac60bd48b1296758009e2f38dca6fc649ca1cf58
thread:issue:28794,issue #28794: torch.from_file not documented,stale/low urgency,2,stale/low urgency,1,0,28794,,,,0,f2e104c9b6aea94f5744376bb5dc3817342e29fbee97e5e26fa9fda147e07659
thread:issue:28829,issue #28829: torch.cuda.close request to be able to reset communications,stale/low urgency,2,stale/low urgency,1,0,28829,,,,0,5a5005a85c6bc7e874a2aad59c8c14cf2653c87ef78a0b5bbb41a1169ee5bbba
thread:issue:28845,issue #28845: Implement operator<< for bfloat16,stale/low urgency,2,stale/low urgency,1,0,28845,,,,0,cdd0642aabe9e0aaddca474e55caeed50c42ddd5cc93d5e6ac75dbe990119e09
thread:issue:28859,issue #28859: Remove bce_with_logits in derivatives.yaml,stale/low urgency,2,stale/low urgency,1,0,28859,,,,0,8f822ef098d0104e55c68a010bdb08a099393db7692807efee86097f5e8fef32
thread:issue:28863,issue #28863: Command for downloading torch 1.12.0 CUDA 10 linked to cu92 version,stale/low urgency,2,stale/low urgency,1,0,28863,,,,0,cc225fcbb58523bdd3a261db2d6fc14b6553d59d78ec9e05b8c14cdb1182a950
thread:issue:28882,issue #28882: Support RRef[T].__call__(*args) which invokes T.__call__(*args) on owner,stale/low urgency,2,stale/low urgency,1,0,28882,,,,0,ff2ac5b3c7ef634ee427f595ae0552141568a42cbd4ada7e469694d5eaea0e61
thread:issue:28884,issue #28884: Why attn_mask is not 3D tensor in nn.MultiheadAttention?,stale/low urgency,2,stale/low urgency,1,0,28884,,,,0,1d7e667849abcd4672cd41e6d43ac19a779ee7095a1b16a7bd2bd9674b83207b
thread:issue:28906,issue #28906: Font used in documentation is not always sharp,stale/low urgency,2,stale/low urgency,1,0,28906,,,,0,9f928b04735633785fd72ede5a92cdf35e1caebc3139a36e66ac4aa4c6feae90
thread:issue:28925,"issue #28925: Provide rpc, remote and dist autograd C++ APIs and register them as Prim::ops",stale/low urgency,2,stale/low urgency,1,0,28925,,,,0,cfc6ced3f3f99efb1e1b468aed6c017936136c8be0b0a8ccce2df0fbf646238c
thread:issue:28929,issue #28929: torch::tensor(scalar) behaves differently from at::tensor(scalar),stale/low urgency,2,stale/low urgency,1,0,28929,,,,0,5da93cb6209aeb7faa62a485f4086723ebf328f9c6eea3d2aaafa6410399e2d5
thread:issue:28930,issue #28930: Add support for multidimensional input to `at::tensor`,stale/low urgency,2,stale/low urgency,1,0,28930,,,,0,2ab3cb5d16157a32909d7087750fe7073352d74886b4f4f145aa664a30247347
thread:issue:28938,issue #28938: Cannot select version in the tutorials page,stale/low urgency,2,stale/low urgency,1,0,28938,,,,0,846f9379c319c17c8be8b5fb434566da5d46e03fc714e60d2aa6f5ce8a235837
thread:issue:28962,issue #28962: [feature request] Some activations modules missing inplace argument,stale/low urgency,2,stale/low urgency,1,0,28962,,,,0,83a7bb7c53ee7071f4a5dc91b7f8310b5593f78e13b7d75cb0b2c6ce00c8fcdc
thread:issue:29010,issue #29010: torch.stack does not yet support named tensors,stale/low urgency,2,stale/low urgency,1,0,29010,,,,0,20150d6ef8a4dc839d33839732d7ea7befd04b27c776940f23a3f2b4398f82be
thread:issue:29021,issue #29021: Named Tensors: size of a named dimension,stale/low urgency,2,stale/low urgency,1,0,29021,,,,0,451e665fa9396349ca44a13c9b468eb80a590e6acb1bbdfb3c912245f728d253
thread:issue:29026,issue #29026: RuntimeError: !t.is_cuda() INTERNAL ASSERT FAILED at /pytorch/aten/src/ATen/native/sparse/SparseTensorMath.cpp:591,stale/low urgency,2,stale/low urgency,1,0,29026,,,,0,bcf052e35d8c9173fab4cee0ab272cdba3750d65007fd9ed8a387390497cf448
thread:issue:29033,issue #29033: Move allgather_coalesced functionality to comm.cpp.,stale/low urgency,2,stale/low urgency,1,0,29033,,,,0,32e8898a3022791e95368e7f31db02e97ed4da60a13cb808c64c4c77054cbbe4
thread:issue:29044,issue #29044: Add Wishart and inverse-Wishart distributions,stale/low urgency,2,stale/low urgency,1,0,29044,,,,0,b43190cc6c446239ad03f3a005f88be68ad2a4a486b6f4edb04aeb3401e6f7b7
thread:issue:29057,issue #29057: Docker issue for Pytorch 1.3,stale/low urgency,2,stale/low urgency,1,0,29057,,,,0,44ba796b49e2721a125e66928a28669c1016c0b3bb4b379966cb99d4c229e1e8
thread:issue:29063,issue #29063: Trace of torch.tensor is impressively convoluted,stale/low urgency,2,stale/low urgency,1,0,29063,,,,0,d57a7fe26f2efba19c3d427f7567fd27a8442d2b9bbe99adbada17f3af216056
thread:issue:29108,issue #29108: a retrained and saved jit module could not be reload.,stale/low urgency,2,stale/low urgency,1,0,29108,,,,0,2bba29c4b9b2fc9990ee3edd74a65a7410d2b4f579cb21b41630f78dc07e6d39
thread:issue:29115,"issue #29115: crash when call dist.new_group(ranks=local_ranks, backend='gloo')",stale/low urgency,2,stale/low urgency,1,0,29115,,,,0,1042d3f99b1deee171e2c93b277dee11a3ad47cb34fb0f991b7410126446a875
thread:issue:29172,issue #29172: Modules without copying in multiprocess,stale/low urgency,2,stale/low urgency,1,0,29172,,,,0,c1c7af416a2696f5a08e522b759f5b969ab03d2ddf3fa5e121ce7642ab8826d6
thread:issue:29190,issue #29190: How to run two different jit models in two GPUs respectively in one scrip?,stale/low urgency,2,stale/low urgency,1,0,29190,,,,0,0f8887dda8843a13c98125db3fe1656192aa5ac0fad87b2048db862fe1049e11
thread:issue:29198,issue #29198: RuntimeError: CUDA out of memory with available GPU memory,stale/low urgency,2,stale/low urgency,1,0,29198,,,,0,f84c1b251feefcec69af3bff6730affcae8d008b3e8367d0ca41498899f5df1a
thread:issue:29202,issue #29202: Add instructions for building torch.distributed on macOS,stale/low urgency,2,stale/low urgency,1,0,29202,,,,0,b660f677cc0afc1540eebccdd2a08ca72b0d390a990d3e45a0a01dfb6715a491
thread:issue:29235,issue #29235: Redo our library structure,stale/low urgency,2,stale/low urgency,1,0,29235,,,,0,ff694c0e4c6fff71b2d5e8f9d1962fc884093049f9bb66fc8c4b8dd31ca36adf
thread:issue:29254,issue #29254: Add support for ivalue float scalars,stale/low urgency,2,stale/low urgency,1,0,29254,,,,0,075a92983e5871019c08609dd224e6e899ead66b94351cfc918ae5369839d649
thread:issue:29369,issue #29369: nn.Conv(n)d constructor doesn't check for the number of kernel dimensions,stale/low urgency,2,stale/low urgency,1,0,29369,,,,0,243b5952cc421670c1c910b3de580e2e7822c24c83c1c98cb16dbf17770f88bc
thread:issue:29372,issue #29372: torch.std() returns nan for single item tensors.,stale/low urgency,2,stale/low urgency,1,0,29372,,,,0,0e4b08a15a8fd11115adf0d4ee72e9db7b63aefab1200573d7fb7212a483055e
thread:issue:29377,issue #29377: [jit] Saving a `ScriptFunction` to a buffer doesn't work,stale/low urgency,2,stale/low urgency,1,0,29377,,,,0,80faaa7816e5ceff56b9ca4c1940efd4d901172720b3e283cd20bdae63dbb9e7
thread:issue:29378,issue #29378: Support out= parameters with autograd,stale/low urgency,2,stale/low urgency,1,0,29378,,,,0,28ba2d87da9ec970e1eb68d96a68bb4bc79ad30ac32a8a131e703bd33622c168
thread:issue:29398,issue #29398: [feature request] torch.kthvalue to support a new argument largest,stale/low urgency,2,stale/low urgency,1,0,29398,,,,0,7787019d0a4257130e37462e48caef1acb2d86372743bd4beedee32cd45636db
thread:issue:29402,issue #29402: [RFC] RPC timeout,stale/low urgency,2,stale/low urgency,1,0,29402,,,,0,46ba22d4fe64f545faca13dca7f252121dcaed57d76e939246717a5f68c0b917
thread:issue:29419,"issue #29419: IValue can't be constructed from one of `int`, `long`, or `long long`.",stale/low urgency,2,stale/low urgency,1,0,29419,,,,0,1fbaeff943ba140d2069912700b720f4851b8bcbd3ac1415db9c1b7a29cb50d6
thread:issue:29429,issue #29429: NLLLoss reduce=True returning nan in float16,stale/low urgency,2,stale/low urgency,1,0,29429,,,,0,7294be5e6589151b94921b508f9961e79d04b22de343d2adc52b2614ba3e07d5
thread:issue:29510,issue #29510: torch.stack: bad shape error message,stale/low urgency,2,stale/low urgency,1,0,29510,,,,0,ca9bf21c3cebf375c865689ba985982bab8dcc00a05e587ffd410dc8b2fb8fa1
thread:issue:29516,issue #29516: RuntimeError: CUDA error: invalid device ordinal,stale/low urgency,2,stale/low urgency,1,0,29516,,,,0,c6f786960496f965af69c53defa1ff23af5d55c48154aa69643f04de5e0ba87c
thread:issue:29528,issue #29528: Half precision cdist,stale/low urgency,2,stale/low urgency,1,0,29528,,,,0,d0cb4aee34ac62b320317d6d2b4b52afa81f53cdf8249ade23fe6ba016f3a879
thread:issue:29545,issue #29545: Tensorboard GPU Problems,stale/low urgency,2,stale/low urgency,1,0,29545,,,,0,85fab2889cc0bde3ade4d626009793cba896ceae4fd42734985279b9707da1fd
thread:issue:29553,issue #29553: `print` uses lots of GPU memory,stale/low urgency,2,stale/low urgency,1,0,29553,,,,0,f047a46b8282b2584f8fadeba49ec5f5c40fac2683640b013af3b2c6dbc2fc11
thread:issue:29597,issue #29597: [jit] Script class attributes aren't automatically added,stale/low urgency,2,stale/low urgency,1,0,29597,,,,0,815680f0aff902d7a67150b59999136f75254e1788dfd1d29dedbd6acfec7cd4
thread:issue:29637,issue #29637: Compiled functions can't take variable number of arguments,stale/low urgency,2,stale/low urgency,1,0,29637,,,,0,0e090b4d6f0aa3b52a58647331c15d94781b288ca70eeb31ffb2c4432f6f75d1
thread:issue:29642,issue #29642: error: ‘struct torch::jit::RegisterOperators’ has no member named ‘op’,stale/low urgency,2,stale/low urgency,1,0,29642,,,,0,080766976755739b5f03b5df3114458b8feca1f21a44bdcf6cf35fb8cbe7c2fa
thread:issue:29657,issue #29657: Improved detection of repeated observers,stale/low urgency,2,stale/low urgency,1,0,29657,,,,0,4e50a2d4317eac97b25ca9a9eb94761a1a7d6c23f2a3562f137d2a851a940c2a
thread:issue:29662,issue #29662: [jit] Dict construction fails at runtime,stale/low urgency,2,stale/low urgency,1,0,29662,,,,0,72b7908d32ca10920db3d52bb0817436e13f89136be6c7f48b57c1604baa86ab
thread:issue:29692,issue #29692: [FR] add generator= kwarg support for torch.randn and torch.rand,stale/low urgency,2,stale/low urgency,1,0,29692,,,,0,817afd45a228cca417087ee0b4037d9948a9717fe96cd9c4331904f6101f96b8
thread:issue:29710,issue #29710: Reliable way to identify RuntimeErrors (CUDA),stale/low urgency,2,stale/low urgency,1,0,29710,,,,0,c9a43e1bce5e9796b94d0a3b6db8faaface3ba25fd3bd89d1d3e63a3495e8ce0
thread:issue:29713,issue #29713: [FR] F.pad support syntax sugars for specifying the padding amount,stale/low urgency,2,stale/low urgency,1,0,29713,,,,0,5cca71310bf5a8077f4a58dcf40605bb3606d59478cb8d1c0c59f14d8f9e1769
thread:issue:29717,issue #29717: CPU and CUDA error messages are divergent in type promotion,stale/low urgency,2,stale/low urgency,1,0,29717,,,,0,0044874b0bb93b1da4c9b0e1dc841ee1d1e0d1c173e7e2467921ae1ebea329d9
thread:issue:29720,issue #29720: Provide a mechanism to set global state per test in thread-safe manner,stale/low urgency,2,stale/low urgency,1,0,29720,,,,0,5132fcf267767151a463e107f4428a6340d3555a8dad92ef9516a60e42c6f46c
thread:issue:29722,issue #29722: Slow (20-50x) RNN tutorial/example when torch is installed using pip comp. to conda installation,stale/low urgency,2,stale/low urgency,1,0,29722,,,,0,83d2f2e5a38ed68f5a1397ac3a83d309a6535c4c765011911eaa5f0a2bff5a0c
thread:issue:29734,issue #29734: Tensor.nbytes() returns itemsize * numel for sparse tensors,stale/low urgency,2,stale/low urgency,1,0,29734,,,,0,e299e2e84799a896d4534043ce9e8303d77e7a2a7d3eaa302d8baedbad1b1d46
thread:issue:29750,issue #29750: [jit] Printing the graph doesn't include function calls,stale/low urgency,2,stale/low urgency,1,0,29750,,,,0,424a882204549743816af62c5638dfa8e1d00ab7506f309bbefb7fe910f64bbd
thread:issue:29754,issue #29754: [FR] general nll_loss and cross_entropy along arbitrary dimension,stale/low urgency,2,stale/low urgency,1,0,29754,,,,0,f5c9dd06e10e1c8d488340ec2d13f6aa17908e7273724a60222a051c3ab3e052
thread:issue:29800,issue #29800: unsqueeze has 'out=' option documented but not implemented(?),stale/low urgency,2,stale/low urgency,1,0,29800,,,,0,2e9855d0cd7198327e9bf2401768e62dd478d1a485b94ce97612f4b9a75f090e
thread:issue:29804,issue #29804: Parallel data loader performance degradation for IterableDataset with num_workers > 1 (but not for Dataset).,stale/low urgency,2,stale/low urgency,1,0,29804,,,,0,6e594db20d9837f7153729b3865be54c65233d8e6c726356d107d4de164bc2ae
thread:issue:29822,issue #29822: [jit] Traced `cat` on GPU doesn't support negative indexing,stale/low urgency,2,stale/low urgency,1,0,29822,,,,0,45b30fac702f120561d302efc97518aee5f8d669cd18db785b4582f132baaaf8
thread:issue:29844,issue #29844: Fold DispatchStub into c10 dispatcher,stale/low urgency,2,stale/low urgency,1,0,29844,,,,0,cf1e02a27249ecce88323174f8d5f6a7106df2f998281913c3ff34b48c734ad9
thread:issue:29860,issue #29860: torch.save/load shows raw path on the pickle_module arg,stale/low urgency,2,stale/low urgency,1,0,29860,,,,0,a9c3a90d7053720328625802f3ddc0f4da8afacfc3f09dd02ce9513422cc4b87
thread:issue:29961,issue #29961: Add support for integer matrix multiplication (particularly for dtype = torch.int8 ),stale/low urgency,2,stale/low urgency,1,0,29961,,,,0,ec5c6efa135a75892484ab8bab52005b970506e8bde8ef7714e778760f37fe15
thread:issue:30150,issue #30150: I have reconstructed LSTM model and tested by mnist data but the loss is not changed (loss=2.3),stale/low urgency,2,stale/low urgency,1,0,30150,,,,0,3bbf5f259b91f9417dd341ad2f18c3fcdc6ec8dec4bfc387b56ef3b21c0d1a97
thread:issue:30155,issue #30155: Transform caffe2 to trt failed,stale/low urgency,2,stale/low urgency,1,0,30155,,,,0,ec8bc2cdb7fb4f3c7d32b272a04d3ea46aa5140bf0138538afcafd612ed04254
thread:issue:30161,issue #30161: Support struct that is initializable / mutable in CPP,stale/low urgency,2,stale/low urgency,1,0,30161,,,,0,d5b17932fc52cefc500e79732765f115c709b8254228f6e81757269f516a9025
thread:issue:30162,issue #30162: [docs] Newline / whitespace / comma missing in formula for PoissonNLLLoss,stale/low urgency,2,stale/low urgency,1,0,30162,,,,0,e6f9b806bde6c493f41a71a7e58adb2f75a7c73fdbb81a6907ddfdfa97cd11bc
thread:issue:30182,issue #30182: [quantization][graph mode] SubgraphRewriter discards SourceRanges,stale/low urgency,2,stale/low urgency,1,0,30182,,,,0,cdc5446368d5afb70314c75397258ea91bfbf2ebf2a92abcd08f301115f9ac1b
thread:issue:30225,issue #30225: c10:Error: could not unlink the shared memory file,stale/low urgency,2,stale/low urgency,1,0,30225,,,,0,e15e6838e1aeb41c6a87322debe6ebb9479319034b9a45496c8393f68ccbc184
thread:issue:30226,issue #30226: Wrong substitution of aten::to,stale/low urgency,2,stale/low urgency,1,0,30226,,,,0,5c3e1b3b2a60d6fdccfe44a399d307ba0402828d92dc13a12d99fedb426cb8df
thread:issue:30246,issue #30246: CUDA masked_select uses way too much memory,stale/low urgency,2,stale/low urgency,1,0,30246,,,,0,427e2506fc4a02393485994facbcc3fdafde34a3ce123e6f76ae28ae9419d1fe
thread:issue:30248,issue #30248: Add scripts for comprehensive benchmark TensorIterator,stale/low urgency,2,stale/low urgency,1,0,30248,,,,0,79ee21c6e1a5d6ec12d474818963fb1a3ec1ca4adc3cf70b165bdbd8706a77a6
thread:issue:30263,issue #30263: pairwise_dist eps argument is confusing,stale/low urgency,2,stale/low urgency,1,0,30263,,,,0,21cc260cfb8597a66e3c1e530ee4f8ad3aca3d53cb3e1b24c0652b0ff3e8013b
thread:issue:30380,issue #30380: Add Google Colab as an option on the 'Get Started' page.,stale/low urgency,2,stale/low urgency,1,0,30380,,,,0,ba1e37de3b10325903ca9514b225e88997cd5e2d0dae2bd33efda8ef2e88ab68
thread:issue:30384,issue #30384: Docs missing link to torch.__config__,stale/low urgency,2,stale/low urgency,1,0,30384,,,,0,87f852e9cbb437c5a06cd408c835048415138a9631e2a794ca589c88062e30ea
thread:issue:30388,issue #30388: Increasing memory usage on CPU,stale/low urgency,2,stale/low urgency,1,0,30388,,,,0,03dcfb980d19288278574c0be3acde63d1fd8dcbf4811d6cef6c84e73df65154
thread:issue:30399,issue #30399: Conv2d: Inconsistent results on Raspberry Pi 3B,stale/low urgency,2,stale/low urgency,1,0,30399,,,,0,5e2f3a6408b3743994a24aa6e67091c888fc07c72461beabbb0bd61917b093d1
thread:issue:30401,issue #30401: Unexpected difference torch.multiprocessing.manager.queue and torch.multiprocessing.queue,stale/low urgency,2,stale/low urgency,1,0,30401,,,,0,f5a572a3d1f36dfb8e6e4a210ee9d9101279761b43a6fa5d6aae520ffea227ff
thread:issue:30402,issue #30402: There is no support for `weight_decay`/`momentum` in SGD for sparse tensors.,stale/low urgency,2,stale/low urgency,1,0,30402,,,,0,bae459ae0f6b2835e255d533bbf9468810f2849bdaf5b78c1ef430013229b43c
thread:issue:30403,issue #30403: last updated timestamp,stale/low urgency,2,stale/low urgency,1,0,30403,,,,0,c6665afc2a7717b76391f0e56d96babed78b354388be91e9c3205fb3f7096f1d
thread:issue:30404,issue #30404: affine_grid CUDA / cuDNN support for Half removed in 1.3.x,stale/low urgency,2,stale/low urgency,1,0,30404,,,,0,f676c983d3b07879060223cbf881994ce8817f51940b1cdbae5bd1e293f01b92
thread:issue:30405,issue #30405: Codegen refactoring master task,stale/low urgency,2,stale/low urgency,1,0,30405,,,,0,41b347883e643d71a16f3465d7cee827b65b4dfb84896e5bef09c9dd1aaf0b6d
thread:issue:30413,issue #30413: nn.functional should maintain API parity with nn where possible,stale/low urgency,2,stale/low urgency,1,0,30413,,,,0,3832902bca7fd8c13203a3d33e72ac2cb9bbe70df662b911fdc9d8a870defcbd
thread:issue:30418,issue #30418: More detailed information about TensorType in error messages,stale/low urgency,2,stale/low urgency,1,0,30418,,,,0,55213a5983d63d283f955ca74bd4ff2505fd6e78404c5034b07e572b661f9a9c
thread:issue:30440,issue #30440: Redundant counter in batchnorm impl,stale/low urgency,2,stale/low urgency,1,0,30440,,,,0,d01cacdc8eb75d749e0c91caf18cc645933d564c8f5f814a0b0006ec6a9990f9
thread:issue:30461,issue #30461: `torch.multiprocessing.spawn` fails when `join=False`,stale/low urgency,2,stale/low urgency,1,0,30461,,,,0,b82492b1ee8899d6c22e1ac23074c4fc6bd4ab28b0a8c7188086e2e58c553ce4
thread:issue:30463,issue #30463: cuda support,stale/low urgency,2,stale/low urgency,1,0,30463,,,,0,99f7a10d10168f9328dad045f8e98cfb84e4fc252ccc20450fa040dd820e96db
thread:issue:30469,issue #30469: Add single input/output tensor scatter/gather to ProcessGroup base class,stale/low urgency,2,stale/low urgency,1,0,30469,,,,0,4eb3baf5075b96e4ee302bdc228e9c0b096eaa87903f2ccbd90d93c24190527b
thread:issue:30489,issue #30489: Torchscript Precision Issue with PyTorch/vision pretrained model inception v3,stale/low urgency,2,stale/low urgency,1,0,30489,,,,0,ebf544285fce6972dbd928f13f95946f1c6960e61f8ee71c471f5ffe300c485a
thread:issue:30505,issue #30505: Adding a function to change the process_group in DistributedDataParallel,stale/low urgency,2,stale/low urgency,1,0,30505,,,,0,f41f14077ffeb9403ac563a5bb17e75f35078bd22940ee17d5a4a51b065ea688
thread:issue:30528,issue #30528: Warn about driver CUDA mismatch in torch.cuda.is_available(),stale/low urgency,2,stale/low urgency,1,0,30528,,,,0,c00f90b8b8d51bdcfe3d2c51488f1b541f3495899ea0cb31a35dcb8648c138e3
thread:issue:30533,issue #30533: Document memory characteristics of in-place ops,stale/low urgency,2,stale/low urgency,1,0,30533,,,,0,3e8ca58204f4e076b6c1d1608d3a35d1f8884626fbd2b4001695e6b1201061af
thread:issue:30535,issue #30535: [FR] nn.init.* accept None as input tensor,stale/low urgency,2,stale/low urgency,1,0,30535,,,,0,573d01e567efcf3f18142b8101fb56285e9c650be989b23bb38870f5abf827c1
thread:issue:30569,"issue #30569: Assertion `index >= -sizes[i] && index < sizes[i] && ""index out of bounds""` failed.",stale/low urgency,2,stale/low urgency,1,0,30569,,,,0,aa657ddc291c802bb72b45d63b3c9f79cf1593c935f817ef8235885cfcd000c0
thread:issue:30589,issue #30589: weight norm missing p=,stale/low urgency,2,stale/low urgency,1,0,30589,,,,0,822d5975af831b55c272042e9b9212f05be01e2d200ec8f6c0c906162a0b4192
thread:issue:30596,issue #30596: Overlapping strides not supported by cublas,stale/low urgency,2,stale/low urgency,1,0,30596,,,,0,16dbe9db5ac6e229a13f00345d68c1a2c0d39d0dc64d1ea787bc1eb9a08aaa5b
thread:issue:30611,issue #30611: Refactor/consolidate code for generating test tensors,stale/low urgency,2,stale/low urgency,1,0,30611,,,,0,fc31ca08e709b7baaa34972dfe6e9a75e5ab01c054f1ea702c7dcf68360b05be
thread:issue:30658,"issue #30658: Dedicated inverse AdaptiveMaxPool1d operation (e.g., AdaptiveMaxUnpool1d)",stale/low urgency,2,stale/low urgency,1,0,30658,,,,0,c35b09c9cd020996768ff0a34d3f9170f558f7cdaa9a9f0a3e6ea0b79ac4f095
thread:issue:30711,issue #30711: Sorting in embedding_dense_backward_cuda takes very long time,stale/low urgency,2,stale/low urgency,1,0,30711,,,,0,8a5c96556c60bbf83a5b8f0ed1a21bf28a07b77f706781956f0a746d98045491
thread:issue:30718,issue #30718: Method to broadcast parameters/buffers of DDP model,stale/low urgency,2,stale/low urgency,1,0,30718,,,,0,ca3eb147ac22375746e33d0805a0b66c217622263922718411aaa7d6f01521b3
thread:issue:30723,issue #30723: Distributed Package asynchronous send/receive not working as expected (Gloo),stale/low urgency,2,stale/low urgency,1,0,30723,,,,0,d5040ce78a1c8ef2fabf6f60d72267bccef0456b826cee75af9ef14b180c9676
thread:issue:30744,issue #30744: Issues in linking libtorch c++,stale/low urgency,2,stale/low urgency,1,0,30744,,,,0,cca1b22ae59171b5afc8915097d969ca7da9fac2248ff5404e6c9700ed04ce18
thread:issue:30754,"issue #30754: JIT, nn.utils.weight_norm and {save,load}_state_dict produce wrong results",stale/low urgency,2,stale/low urgency,1,0,30754,,,,0,50aa6e103ab521803affa07bd8899f77c37e7ac7eac9d47ff8f012b8a915b26c
thread:issue:3076,issue #3076: BN slows down double-backprop enormously,stale/low urgency,2,stale/low urgency,1,0,3076,,,,0,1eae080395ded4b1ecacd54a743171118bf1048ea682020fb590740c2f94135d
thread:issue:30762,issue #30762: Page for `torch.__config__` 404s,stale/low urgency,2,stale/low urgency,1,0,30762,,,,0,8b4f78f3867d242ce8722bc37b8a33139f4e29fe403076e994271610f0a1283c
thread:issue:30780,issue #30780: Remove Ops bound in Python Layer for Legacy Reasons,stale/low urgency,2,stale/low urgency,1,0,30780,,,,0,219a323b2614620e4f774a2babc15ad08705091aa6d30353cb42deeaf761a9a5
thread:issue:30788,issue #30788: Master Task: JIT/C++ Parity With Pytorch Python API,stale/low urgency,2,stale/low urgency,1,0,30788,,,,0,36dc3470e3f11e857ed4200c6c75e2385a131552307e083083e655eecf1062ef
thread:issue:30807,issue #30807: Sending sparse tensors over RPC not yet supported,stale/low urgency,2,stale/low urgency,1,0,30807,,,,0,c718f17e1288f7a3dcdb9b74df6f352085e5a6bea179e80dfec17ec0fc6dd455
thread:issue:30867,issue #30867: save model definition file,stale/low urgency,2,stale/low urgency,1,0,30867,,,,0,1f88ecc0089765a6cd13f071962bd2407d94f5735a9238ff1dcee630049f863e
thread:issue:30899,issue #30899: Adding max_norm constraint to an Embedding layer leads to an error,stale/low urgency,2,stale/low urgency,1,0,30899,,,,0,73f31685696d2355e84d631250a395636fb11fb2b1be1b11f4ac1c0b7ee2fba3
thread:issue:30934,issue #30934: Spurious negative output in convolution of positive tensors,stale/low urgency,2,stale/low urgency,1,0,30934,,,,0,87b86c3c24687bcc60562b0d57d3d46a1c8d959fb68b2fde7ce1aa357674169d
thread:issue:30947,issue #30947: Wrong initialization with kaiming_uniform_,stale/low urgency,2,stale/low urgency,1,0,30947,,,,0,d52a763ae94d5ddb845a23a5d8ccf6b3e6a25c9a936c429f2fe1c7e30c4e0ac7
thread:issue:30953,issue #30953: false CHECK FAILED at ../aten/src/ATen/core/function_schema_inl.h,stale/low urgency,2,stale/low urgency,1,0,30953,,,,0,7d359f4024252650a62a04d4e291b86ea5bad803f1384f843b6487b658354f55
thread:issue:30962,issue #30962: How can I add masks to parameters,stale/low urgency,2,stale/low urgency,1,0,30962,,,,0,6d83d5cb07b21145e9075bef04cdf1657bea027a05e33e7039aff63880fb5b9c
thread:issue:30987,issue #30987: Remove `.data`,stale/low urgency,2,stale/low urgency,1,0,30987,,,,0,a720cea52ee3c33a86a1b9a8c9cb0176592d8cabc8d2740c634c4921234decac
thread:issue:31038,issue #31038: Pytorch 1.3.0 on RTX cards: CUDA error: an illegal memory access was encountered,stale/low urgency,2,stale/low urgency,1,0,31038,,,,0,15eb47f9b6646850ce0bd4b90152ca18af10c20c16f19cb47a62f49f019d8a25
thread:issue:31048,issue #31048: MathJax too small in Firefox,stale/low urgency,2,stale/low urgency,1,0,31048,,,,0,0aa5ee7103ad043d51eb34ad4f18de806a467367ad2a06f6dfc899e18b7ff1c1
thread:issue:31085,issue #31085: Common lookup of generic types across full script and mobile parsers,stale/low urgency,2,stale/low urgency,1,0,31085,,,,0,76062096d8002761b613e2ce0e93ced307cb08b01ae9c315f845e7b0a36930c0
thread:issue:31107,issue #31107: Add torch.version.nccl,stale/low urgency,2,stale/low urgency,1,0,31107,,,,0,a70afd503b3de389586eecd82f59a1940ca84c4cff3f3b8cbc03a599d848a61f
thread:issue:31110,issue #31110: torch.nn.Softplus threshold argument bug?,stale/low urgency,2,stale/low urgency,1,0,31110,,,,0,84703275922e2e5d586ae113f6d26971401e943c3bd5da7e0480d436f06a7f1b
thread:issue:31141,issue #31141: [JIT] tensor(device=...) and tensor.to(device = ...) does not work properly in traced functions and modules,stale/low urgency,2,stale/low urgency,1,0,31141,,,,0,9dbfba0ca5230d45f37b2c5dbc7682e3f0aac50ca7a3400ae06ed5362472d6ed
thread:issue:31178,issue #31178: torch::nn::functional::interpolate crash,stale/low urgency,2,stale/low urgency,1,0,31178,,,,0,cbd61451b9d98d9ffd462e461b75d679c30c7ff077345cbcbc0736a4a3d762ac
thread:issue:31185,issue #31185: Retain Subgraph or Save Intermediate Grad support?,stale/low urgency,2,stale/low urgency,1,0,31185,,,,0,7cd3f7c526c68c0ae616e9a53c622ac9b62dc8f972f66e5aa1c7303e10583070
thread:issue:31191,issue #31191: [JIT] Slice with optional not supported,stale/low urgency,2,stale/low urgency,1,0,31191,,,,0,0c309f1260056eb7aca345fcbad205789bbca87cedd7e9ae9beb955b0a2dcbed
thread:issue:31202,issue #31202: Request for Lint Pass to Detect Modification on Parameters/Attributes during TorchScript Inference,stale/low urgency,2,stale/low urgency,1,0,31202,,,,0,90bcb573865415efc9a0fdefba65195a52f3162c2a83c0bdba4c636f20ebe85c
thread:issue:31232,issue #31232: Default shuffle behavior of DistributedSampler,stale/low urgency,2,stale/low urgency,1,0,31232,,,,0,ffb6686e9476868377fdac213b4b3a9948771a79576aebbbcdfbf340e472f8a2
thread:issue:31259,issue #31259: Script method can't call a scripted function when it is decorated with `@torch.no_grad`,stale/low urgency,2,stale/low urgency,1,0,31259,,,,0,ca797cc688a659a7d9f2ef73b20000c7ae7ffde3d046a4e5dad4705ba46f909c
thread:issue:31261,issue #31261: pinned memory requires DeviceGuard in multi-process envs,stale/low urgency,2,stale/low urgency,1,0,31261,,,,0,e5f1886213a58af25361e9cf8c3f9dd297dae4a97c66dec055cf9240a02f087c
thread:issue:31264,issue #31264: Failed to config caffe2_rocksdb in cmake,stale/low urgency,2,stale/low urgency,1,0,31264,,,,0,736eef5b04cd796ae1614ea7797fdda8c63f2a8e90647bdda58085cc708b8ac8
thread:issue:31277,issue #31277: nn.MultiHeadAttention with different similarity measures,stale/low urgency,2,stale/low urgency,1,0,31277,,,,0,ae02ecdbd024981dc938e930b43ede08c037c9feb33586b27ff20c9e4468bf37
thread:issue:31289,issue #31289: Intel OMP multiprocessing assertion failure: Assertion failure at z_Linux_util.cpp(2338),stale/low urgency,2,stale/low urgency,1,0,31289,,,,0,40593e5571f90545a2dd244dfa0354361293e83c588fcbd79712b2cdff6cb4fe
thread:issue:31300,issue #31300: torch runtime error when manual link libmkldnn.so,stale/low urgency,2,stale/low urgency,1,0,31300,,,,0,0b3a562f9a897f990bf05560b6fd9746f0854bd4e5ce494b347b1db05644b023
thread:issue:31308,issue #31308: [mac] Failure to import torch,stale/low urgency,2,stale/low urgency,1,0,31308,,,,0,b739fd930d95e5f11e33a2a98f063f2a17bdcdb90471b7dc3c80ac013389293d
thread:issue:31316,issue #31316: quantization - Missing operations needed for object detection,stale/low urgency,2,stale/low urgency,1,0,31316,,,,0,d4d3be2e7bd182a19a7ed1349c39706811f45ba2fe8fc767a90922e553bd5ec9
thread:issue:31321,issue #31321: Ability to download docs HTML for offline use,stale/low urgency,2,stale/low urgency,1,0,31321,,,,0,ec666e570ebada75d4c791cc2a519745d8afeef660e06047863f116819102249
thread:issue:31360,issue #31360: DDP/MP not yielding nontrivial speedup,stale/low urgency,2,stale/low urgency,1,0,31360,,,,0,90a7fafb24b0f1cf09e5cbf46c0c88c9e262253834841d33c4364ad94edecf62
thread:issue:31367,issue #31367: `clip_grad_norm` allows negative `max_norm` values,stale/low urgency,2,stale/low urgency,1,0,31367,,,,0,5cab43b503b7c60f11899d1e958866c09fa13b0bf363b14a832422fdf0a624e3
thread:issue:31386,issue #31386: [jit] Python type hints in TorchScript classes don't work,stale/low urgency,2,stale/low urgency,1,0,31386,,,,0,740140bb614f4ae48ede3ddca1d9f55c85f9d344b131ca9129147feae6a96062
thread:issue:31391,"issue #31391: Error ""builtin cannot be used as a value"" when add Python snippets in C++",stale/low urgency,2,stale/low urgency,1,0,31391,,,,0,0039edb92336458c22a3a539edfa461e0691486b9b2ee1dd3557e9617e3a9959
thread:issue:31422,issue #31422: Support DataParallel with PackedSequence,stale/low urgency,2,stale/low urgency,1,0,31422,,,,0,11c300e25d785148a5862b932e057005ad4468a50337b20eb3a36ea4109f6ccc
thread:issue:31423,issue #31423: Add option in LSTM layer to access all cell states of all time steps,stale/low urgency,2,stale/low urgency,1,0,31423,,,,0,4e5fae9d6e7224684b739579d878e6bf12fef44af26e369e0bd1d1b3c897de7f
thread:issue:31425,issue #31425: cuDNN convolution does not handle empty input tensor,stale/low urgency,2,stale/low urgency,1,0,31425,,,,0,bd96c3ad393fa043a96d756c0de2b4b3d0af12a859ca88132f984aac54329685
thread:issue:31430,issue #31430: [jit] `del` with slices doesn't work,stale/low urgency,2,stale/low urgency,1,0,31430,,,,0,f2ad34326711e60bd8f3ddcabd2be04845dfc46427a271c3ee7df1635e15eb58
thread:issue:31460,issue #31460: DataParallel doesn't properly handle kwargs,stale/low urgency,2,stale/low urgency,1,0,31460,,,,0,9b2862ccf5106fc29d7dda13e132e4b9f8b60854db73f75e9ff476addeb81cfb
thread:issue:31467,issue #31467: Distributed hangs on process termination with world_size=1,stale/low urgency,2,stale/low urgency,1,0,31467,,,,0,81e97370247e4e0a0170bc64b38193d8928a9cdde6a11382177cd40456c15ee8
thread:issue:31471,issue #31471: Mnasnet0_5 first layer shape incorrect,stale/low urgency,2,stale/low urgency,1,0,31471,,,,0,7f8caec7da7938eeaa6b29514c6bb97e3d5c97aaf5059acf47889e5d6716c1f6
thread:issue:31478,issue #31478: DataParallel has different tensor copy behavior if batch size = 1,stale/low urgency,2,stale/low urgency,1,0,31478,,,,0,c9c1e27f32411b11ecb83e446f5ad6ed6e06b2260e145923b6b9351af29020ee
thread:issue:31487,issue #31487: Disable PSIMD? Why Pytorch is trying to download PSimd when PSIMD_SOURCE_DIR is defined?,stale/low urgency,2,stale/low urgency,1,0,31487,,,,0,da64db0ac42b689db9790e32a42dba738284d31925ac48da7851c9e49484260b
thread:issue:3152,issue #3152: Sparse tensor .new(size) can be confusing,stale/low urgency,2,stale/low urgency,1,0,3152,,,,0,f34a1267a10eaded934dae053dcee3d074bbec0836c1fbe4dd72ccb9ddc501df
thread:issue:31545,issue #31545: IndexExpressions (or slice) for jit.script functions,stale/low urgency,2,stale/low urgency,1,0,31545,,,,0,9155ec795494dce84dca324ccb73dfd378dcf248547242e9cded2aff8b784b68
thread:issue:31546,issue #31546: [Feature Request] Make torch.solve output NaN for singular matrix,stale/low urgency,2,stale/low urgency,1,0,31546,,,,0,7a560bf01b7993aa7cbd0b769ca5c98f605b2be347dfaa82e836876557f38768
thread:issue:31548,"issue #31548: c++ PReLUFuncOptions declared, not used or valid",stale/low urgency,2,stale/low urgency,1,0,31548,,,,0,b972d9b7c6bc7b7d85177b05b8330eb20e35cf4e60d54285311e669eae26710f
thread:issue:31551,issue #31551: TorchBind broken on rocm,stale/low urgency,2,stale/low urgency,1,0,31551,,,,0,cebaaf34c680c9c7e98349e02d3566d52e8e80cb6546881649f36627376d0ca8
thread:issue:31553,issue #31553: allow setting different batch size splits for data_parallel.py and distributed.py,stale/low urgency,2,stale/low urgency,1,0,31553,,,,0,146f807692769bc68f97c471c618160ec03c41e14b12a91cd3988cd4104dc218
thread:issue:31565,issue #31565: Force libtorch to use CUDA context,stale/low urgency,2,stale/low urgency,1,0,31565,,,,0,fe469bccd1975142fa7a4222750c4c4f5cf352ff587a892148c94e8e797b5996
thread:issue:31570,issue #31570: Can't pin storage memory,stale/low urgency,2,stale/low urgency,1,0,31570,,,,0,ccca50ef2df1168c0f8abcf4ce886f909e890fd614da3ebaf1fbe9b1321673ab
thread:issue:31571,issue #31571: Bug in ForkingPickler for multiprocessing spawn context for shared storages on Linux,stale/low urgency,2,stale/low urgency,1,0,31571,,,,0,fd95320dba0613e7ecb79b6e3c5c97643f83d728d664313505cf15c71fc515c0
thread:issue:31574,issue #31574: attribute and register_buffer are not the same on gpu,stale/low urgency,2,stale/low urgency,1,0,31574,,,,0,bfa9189f0e4793e0ad6a26eb90d2535ff803d29d5986122d41e7ba52e7120b97
thread:issue:31579,issue #31579: Option to apply weights to gradients when using DistributedDataParallel,stale/low urgency,2,stale/low urgency,1,0,31579,,,,0,be88c1444826b6efaa55dd29413f7a8cf6c693fa6a43f9b898f952e97bea2ace
thread:issue:31596,issue #31596: pytorch forward hangs in multiprocess environment,stale/low urgency,2,stale/low urgency,1,0,31596,,,,0,f7380444db31b35f4bba484ee089dfa3430a255e7289f43eb29522f4b81b1121
thread:issue:31598,issue #31598: ATen not compiled with MKL support,stale/low urgency,2,stale/low urgency,1,0,31598,,,,0,c179b9bd888ab114cfd6bd056f781f65d4d24cf028532b5e9ed5d9028b56c61b
thread:issue:31606,issue #31606: No auto-suggest capacity for Transformer,stale/low urgency,2,stale/low urgency,1,0,31606,,,,0,2cbb8059afa5e6c0fc1c205f3fb703b1c4af8555e7a12132480942820677e8fe
thread:issue:31615,issue #31615: 'torch.load' report 'bad pickle data',stale/low urgency,2,stale/low urgency,1,0,31615,,,,0,8dd8d36f5acf70677448b375f6371d9f02e13d041fbe94492f6927e80bdb84bb
thread:issue:31651,"issue #31651: update embedding at indices, other than those passed as input, in the case of sparse tensors",stale/low urgency,2,stale/low urgency,1,0,31651,,,,0,4e33f58cc3a76af81dd4aebcd0244b59bb4543c6fc5a482dfbea2188dfe22da9
thread:issue:31655,issue #31655: pytorch out of memory when calculation squared difference of unfold,stale/low urgency,2,stale/low urgency,1,0,31655,,,,0,882568538bd4c71a043a89093e9ba55971b0e80c4ee5cf98d84a6681aaaa122b
thread:issue:31657,issue #31657: Use of Sequence collections for abstract classes in Dataset,stale/low urgency,2,stale/low urgency,1,0,31657,,,,0,7514e84ce8058c775ef196eb7f8898ea78fed58c3932c552306d8ff7d38ab9ac
thread:issue:31679,issue #31679: broadcast randperm with dim specification,stale/low urgency,2,stale/low urgency,1,0,31679,,,,0,74d6db24069c21f308b1570c0a894e18584177a700e6de1fb9fc42708aa3229b
thread:issue:31685,issue #31685: What is the significance of torchvision._is_tracing()?,stale/low urgency,2,stale/low urgency,1,0,31685,,,,0,a5039c2e5dcae0e6caae92532d58d06d8da01d1e284dd51ef2d77eab19508565
thread:issue:31688,issue #31688: Parallelization: more balanced work distribution among workers,stale/low urgency,2,stale/low urgency,1,0,31688,,,,0,5ab454f0357fac1356bb4efde45cb4e1b569c295d491e5c280eaf3cf7970cdec
thread:issue:31689,issue #31689: DataLoader does not consider default floating point type,stale/low urgency,2,stale/low urgency,1,0,31689,,,,0,052f0bf04312c1258cba2c95ae1f0673daa5c2b0895b414d0ef4200a2d45fddc
thread:issue:31697,issue #31697: the cmake problem with build from source ?,stale/low urgency,2,stale/low urgency,1,0,31697,,,,0,b78bfd09931d87c2ffdcbe0ed5d9c43f810f6e6c531d8bd74fa5b626ec15996d
thread:issue:31710,issue #31710: AttributeError: module 'torch.distributed' has no attribute 'init_process_group' on torch 1.3 aarch64,stale/low urgency,2,stale/low urgency,1,0,31710,,,,0,2c0e844f9005e7fcbacb79abb91e416ac22689d97e29a48152a8dc3c086f4561
thread:issue:31742,issue #31742: sparse tensor eliminate_zeros,stale/low urgency,2,stale/low urgency,1,0,31742,,,,0,ab0ef26f790b65020c5bc67bc6f2ff7cef8102a915bd134c92034d378c9b4c57
thread:issue:31752,issue #31752: Need a launch utility for Distributed RPC framework.,stale/low urgency,2,stale/low urgency,1,0,31752,,,,0,451bd9e66a3d37601078d9508950b36a1627def961ce75bab505930522873e55
thread:issue:31759,issue #31759: Allow using digits in names of named tensors,stale/low urgency,2,stale/low urgency,1,0,31759,,,,0,12931b9dc21ac89d1b740f5b462cd2ff6407712aa7679b5e16063c7bec1e91a0
thread:issue:3176,issue #3176: Add safety checks in `index_add`/`scatter_add`,stale/low urgency,2,stale/low urgency,1,0,3176,,,,0,60067f3afae4ab49d821f4e14424459f79273a4011bd483e8e191f3bb2f5454f
thread:issue:31776,issue #31776: Documentation for `scatter` incorrectly states that index values must be unique,stale/low urgency,2,stale/low urgency,1,0,31776,,,,0,740ee6532ab3c0bc1a5504880d971f274de89b8216693d80d0f274f2c94ce5d7
thread:issue:31815,issue #31815: ImportError: /home/xx/anaconda3/lib/python3.7/site-packages/torch/lib/libtorch_python.so: undefined symbol: _ZNK5torch3jit5Graph8toStringE,stale/low urgency,2,stale/low urgency,1,0,31815,,,,0,00eefd03be3a2d3cc3498b807f5f2e9fb27edce2215bf5c1c6fd6112fea2b928
thread:issue:31818,issue #31818: How to distinguish different layers in hook？,stale/low urgency,2,stale/low urgency,1,0,31818,,,,0,8012ca6b7aa000723d8d06a5a2469b8309d192feb17430dfa72f9f75ea37b54e
thread:issue:31842,issue #31842: Support rpc/remote torch script call with script class/module name and class/module method name,stale/low urgency,2,stale/low urgency,1,0,31842,,,,0,28598537d12b45de941bfb920bfb284b7d9c8909d712575356ec3e87585de69d
thread:issue:31856,issue #31856: Fix softplus clampling issues using logsigmoid,stale/low urgency,2,stale/low urgency,1,0,31856,,,,0,b487b9d4d24929c70579ef3d7daa7f10e7a893c7847f18be4a3cbbea3765d349
thread:issue:31866,issue #31866: Negative indices in chunk could cause Out of Range access on loss.backward in JIT,stale/low urgency,2,stale/low urgency,1,0,31866,,,,0,b62a5631390aa563239ec5feaa11e6c6bd6c6f5b47867034328055fc0d3790c7
thread:issue:31867,issue #31867: The model training time is increasing between runs if the same DataLoader reused to train multiple models.,stale/low urgency,2,stale/low urgency,1,0,31867,,,,0,826a3cc9e800cd045f28b59e7ff7945fe51ebcc25c1f9f3bf40d669351a40cb9
thread:issue:31882,issue #31882: Example cmakelists for custom cuda operator?,stale/low urgency,2,stale/low urgency,1,0,31882,,,,0,ea9e63fb5b853aba26fa43aece42dff15d0c649e3acf6b05613d0e743f67ec14
thread:issue:31883,"issue #31883: The dependency target ""nccl_external"" of target ""gloo_cuda"" does not exist.",stale/low urgency,2,stale/low urgency,1,0,31883,,,,0,dea053c99e0395378f642c9064cc9f8c623be25a3c62db47f3e63996b70be074
thread:issue:31895,"issue #31895: torch.masked_select out argument can easily be misused, because output shape is dynamically computed",stale/low urgency,2,stale/low urgency,1,0,31895,,,,0,bf077f7d57dec95c3535db6b45e4eebfcea2f123126d9ce4d49b4b2d6c8f86dd
thread:issue:31916,issue #31916: I'm not able to build pytorch with tensorrt (current master),stale/low urgency,2,stale/low urgency,1,0,31916,,,,0,22b6a5998e6a73f221cfa213e99c447bcce6cce41a1d53f697a422ff34c99b12
thread:issue:31937,issue #31937: RPC and dist_autograd should respect no_grad mode,stale/low urgency,2,stale/low urgency,1,0,31937,,,,0,1e094f7c0a995c13f8e96bcac4fe9eea0ea12739474262d2ff2144eb1292f45a
thread:issue:31942,issue #31942: Support sparse inputs for torch.block_diag,stale/low urgency,2,stale/low urgency,1,0,31942,,,,0,34cc3c9b9e92837b45b6abcfde13c325b914449e06384daf604c1bb41d3631f6
thread:issue:31943,issue #31943: error when building pytorch 1.1.0 from source,stale/low urgency,2,stale/low urgency,1,0,31943,,,,0,57ff78efb58c4a2808bb1916a1c0301586e157902ae05ad6b4e9df94c22001d4
thread:issue:31953,issue #31953: JIT tests are linked directly into libtorch and register operators even when unused,stale/low urgency,2,stale/low urgency,1,0,31953,,,,0,83d0b9313a6a56a174ef04aa55aa828d80d0bac28a0d54c40678262d80d44f17
thread:issue:31979,issue #31979: Reuse spawned subprocesses in RPC tests,stale/low urgency,2,stale/low urgency,1,0,31979,,,,0,7f0045aa79fab4679968d876651bb984c811cc34a273a928fdb4dc486ac9fe23
thread:issue:31994,issue #31994: Trying to disable cuda to run torch on OSX run it on CPU,stale/low urgency,2,stale/low urgency,1,0,31994,,,,0,fe843f1f2e5df1298e2abd678340028911457e18f915a50020d363b4fbc80365
thread:issue:32018,issue #32018: DistributedDataParallel non-floating point dtype parameter with requires_grad=False,stale/low urgency,2,stale/low urgency,1,0,32018,,,,0,c70e3caea31a1aaa9d016fd9952564f948876b522de5d850d8aa895595a1ea82
thread:issue:32023,issue #32023: Inaccurate batched GRU results on CPU,stale/low urgency,2,stale/low urgency,1,0,32023,,,,0,7e468b65066d1b371e371ba2b0815ddba924d6c43f2edbf7800a99e64c178a1d
thread:issue:32030,issue #32030: Dockerfile for people to quick-start contributing,stale/low urgency,2,stale/low urgency,1,0,32030,,,,0,8771e94165abb156ca477c05a4debf6410a88e7e86861c69326bc8f8b2d45bb4
thread:issue:32033,issue #32033: run detectron test test_spatial_narrow_as_op.py fails,stale/low urgency,2,stale/low urgency,1,0,32033,,,,0,6e6ba2ad75b72e1bbadf59e01a2406ce175fb075b3f6be18c3485d969c72d73e
thread:issue:32034,issue #32034: Extend `ConcatDataset` to return dataset index,stale/low urgency,2,stale/low urgency,1,0,32034,,,,0,e206458928d2d2cd03cabe385aec731dbed042fcd323490c04d620d7074a8a07
thread:issue:32036,issue #32036: Internal assert failed using multiple GPUs with DataParallel,stale/low urgency,2,stale/low urgency,1,0,32036,,,,0,7b992dbfade825c96ad54a2693e5b5595bcba71fa77a3ad6c934b0a4834f5cee
thread:issue:32039,issue #32039: Internal assert failure for user IValue unwrapping,stale/low urgency,2,stale/low urgency,1,0,32039,,,,0,28ae91bcd56f9275e3f7791131dbe7d75ce09d3b44f2b724e8ca97864abe5a3c
thread:issue:32047,issue #32047: Sobol point implementation,stale/low urgency,2,stale/low urgency,1,0,32047,,,,0,9067bb6784a22afbceefbd044d031ee0ccd62727e8fe7e87e8aef9ef4ca2c49a
thread:issue:32083,issue #32083: build pytorch from source fauled: undefined reference to `cusparseGetErrorString(cusparseStatus_t)',stale/low urgency,2,stale/low urgency,1,0,32083,,,,0,3fd65f6feb9a1e56a883f7174a5f09d250bf1568afb20835bab80084467ba399
thread:issue:32087,issue #32087: Out-of-date link to pytorch ci dockerfiles in jenkins readme,stale/low urgency,2,stale/low urgency,1,0,32087,,,,0,42dbc9899fc8730c54b174db023456d3aed30c1a6ac7a0953366d6e32b11e56f
thread:issue:32088,issue #32088: Batch convolutional layer with 5d weight tensor that is not contiguous,stale/low urgency,2,stale/low urgency,1,0,32088,,,,0,c24a3ef9ae726bd24e2aeb821dbe7c9edc5056ac26d5736a940c4cdef8b03811
thread:issue:32098,issue #32098: torch.gt ge lt le,stale/low urgency,2,stale/low urgency,1,0,32098,,,,0,9f6372a817ffb0b6d9daf9e85ec78a6e8c264a65824bd192188e9b5e643fd018
thread:issue:32105,issue #32105: TensorRT: CheckDims() need adjustment for EXPLICIT_BATCH?,stale/low urgency,2,stale/low urgency,1,0,32105,,,,0,cda64a60b65473ea85005df48034ec5ec76fc88ecc6df8e6712633dfe0a47482
thread:issue:32113,issue #32113: Inconsistent linking flags result in error when building lib/libc10_cuda.so,stale/low urgency,2,stale/low urgency,1,0,32113,,,,0,5250b6af7f8ddeb5ead3e49f5bdfb03e2c44be90f0ebeb2fa5d5813594471959
thread:issue:32130,issue #32130: Idempotency Keys for RPC Retry,stale/low urgency,2,stale/low urgency,1,0,32130,,,,0,5c073630874872ccc8b9a4233b98b655c0d5ac9831f3901b0e48e9251b8b4596
thread:issue:32132,issue #32132: Slighty out of tolerance for `test_mv` and `test_cholesky_solve_batched_cuda_float64`,stale/low urgency,2,stale/low urgency,1,0,32132,,,,0,9d36c7434cbeb0b5e42c44b6c994cb31adf58a3da5a0c259b9834637218f1d30
thread:issue:32137,issue #32137: torch.nn.functional.normalize epsilon too small for half precision,stale/low urgency,2,stale/low urgency,1,0,32137,,,,0,2e2b253ad98b8aaf3507053c796c768639b69e2b0eac4e880c4af79d2f34810c
thread:issue:32141,issue #32141: We should change DeprecationWarnings to UserWarnings in 27361,stale/low urgency,2,stale/low urgency,1,0,32141,,,,0,81424f433cc01de90cc7073405515d4b53294c942ba02fd45fdc6f03f6a5aa29
thread:issue:32150,issue #32150: pin_memory may change the type of instance returned by collate_fn.,stale/low urgency,2,stale/low urgency,1,0,32150,,,,0,1ec015d7fafdbf30f491696aab5ecb193760b17523150c59cc2d0b16948b3f8e
thread:issue:32162,issue #32162: Using Tensor.to(device) after distributed all_reduce intermittently causes deadlock with NCCL,stale/low urgency,2,stale/low urgency,1,0,32162,,,,0,586c55c3f2636e82d1988d5c5daf370828b3d6f16ef5d3036334fa8cebe41f23
thread:issue:32165,issue #32165: Serialization inconsistency with pickling tensors breaks caching,stale/low urgency,2,stale/low urgency,1,0,32165,,,,0,84c1af882e19dddcee7843c591c67e8822ae0f048b71c8d7294093564632cc1a
thread:issue:32166,"issue #32166: C++ randint returns float32, python returns int64",stale/low urgency,2,stale/low urgency,1,0,32166,,,,0,68555969ae58b87c036dc720a2354ab60162b71759d09f240644390266890b81
thread:issue:32167,issue #32167: Support in-place pinning of memory,stale/low urgency,2,stale/low urgency,1,0,32167,,,,0,4340445b755e30cbf2981c621d213bb9631756a998e6cd348e24bdeefabac130
thread:issue:32210,"issue #32210: dist.send/recv ""IndexError: map::at"" error when bool tensors are used with mpi backend",stale/low urgency,2,stale/low urgency,1,0,32210,,,,0,0e332e691e6b3fa66e15b9f228e8d88db8e2fb4e5d173f953c0ca88aa7992934
thread:issue:32214,issue #32214: Pytorch compilation error,stale/low urgency,2,stale/low urgency,1,0,32214,,,,0,ffb691918c353440dde54673393a197f84528133917c8e895ab39cbfca9bf158
thread:issue:32219,issue #32219: [Feature proposal] improved algorithms for checkpointing,stale/low urgency,2,stale/low urgency,1,0,32219,,,,0,f9ab630eca3446c5fc0a522b5ddcc8dd88a5b28ebf6e59736251ac9c2f207d5e
thread:issue:32227,"issue #32227: 0 INTERNAL ASSERT FAILED at /pytorch/c10/util/intrusive_ptr.h:348, please report a bug to PyTorch.",stale/low urgency,2,stale/low urgency,1,0,32227,,,,0,13058b0380dc4cee5deba464c8c3926fb9f2afbfb371bc90349b7cc803dcec5a
thread:issue:32237,issue #32237: PyTorch C++ API docs only tracks master branch,stale/low urgency,2,stale/low urgency,1,0,32237,,,,0,78b0fcdfe0f36caa77e7dfb19dbfff5328147f52c7db860a65d9fce69e0c2b95
thread:issue:32253,issue #32253: [TorchScript] Device comparison implemented in eq but not ne,stale/low urgency,2,stale/low urgency,1,0,32253,,,,0,4fdb7c2393a1533f2acc47ef03f976e636af9eefce3f9d7e954a7e3a2e148590
thread:issue:32262,issue #32262: models after prepare/prepare_qat contains `qconfig` which prevents pickling,stale/low urgency,2,stale/low urgency,1,0,32262,,,,0,b1df2af72821cae12bd00d3d6609bfaecb4e6991cf1d9615a2234c575a390349
thread:issue:32264,issue #32264: torch.nn.functional.threshold not work with LongTensor,stale/low urgency,2,stale/low urgency,1,0,32264,,,,0,78874008c7856c8f3a6dbebb22462b99f4fed62524aef1556b6c583a99b9f911
thread:issue:32268,issue #32268: Numpy array functionality in torchscript.,stale/low urgency,2,stale/low urgency,1,0,32268,,,,0,01c0734195d3a5602501d2a50dd3f1bc03cafd054bf96c95b2677f145fc6e398
thread:issue:32269,issue #32269: Support strategy to train large model that exceeds GPU mem and DRAM mem,stale/low urgency,2,stale/low urgency,1,0,32269,,,,0,d9a89e9d82ae177aeb53d0a19d5e0d96c63402e910b663d1fc9604d8668a3ed1
thread:issue:32287,issue #32287: Python 3.8 Windows JIT test failure,stale/low urgency,2,stale/low urgency,1,0,32287,,,,0,d2abd9d466b9d0301f1b1f1436ece9c7fe7369fc0aae294de7662fa994d54f44
thread:issue:32305,issue #32305: [FR] histc should (optionally) return a long tensor,stale/low urgency,2,stale/low urgency,1,0,32305,,,,0,8d615142c268247d26700c3e81894a0c2c9a774b64201fd8e8e00207c4ad8ed5
thread:issue:32322,issue #32322: deadlock when using mp.spawn multiprocessing,stale/low urgency,2,stale/low urgency,1,0,32322,,,,0,0dcacd633935dd3c8100ebc7a500672ecf6bade8326599c5262540c9a55b732e
thread:issue:32337,issue #32337: Save LocalResponceNorm as a single aten node when convert a pytorch model to jit model,stale/low urgency,2,stale/low urgency,1,0,32337,,,,0,65dce62c9824363d6f26493648203c3c090d793704874546a1845518d092da6c
thread:issue:32357,issue #32357: Pytorch JIT Compilation Does Not Finish (Infinite Loop?) for Deeper Models,stale/low urgency,2,stale/low urgency,1,0,32357,,,,0,bce95df99cc3d3099cc488734a629609adba9db24db986b414028709a0f38d2c
thread:issue:32358,issue #32358: torch.bartlett_window not jitable,stale/low urgency,2,stale/low urgency,1,0,32358,,,,0,d6bddddba60bf84bf2b68f8724ab885dcc908642b9ab4dc236296388de5f0872
thread:issue:32363,issue #32363: Possible ProcessGroup::Work::abort correctness issue,stale/low urgency,2,stale/low urgency,1,0,32363,,,,0,3d0a34ef4f6321224ab7f6feec5bd4b068ad571614f309507232ccd091aebc11
thread:issue:32370,issue #32370: Backward `Functional.conv3d` is slow when cuDNN is enabled,stale/low urgency,2,stale/low urgency,1,0,32370,,,,0,563a2dc67c7405c9073a41709cb8fc834b70028f14ca42bd0d837219e575e84a
thread:issue:32372,issue #32372: Migrate processGroup::async_work to use Future,stale/low urgency,2,stale/low urgency,1,0,32372,,,,0,0fcdfc6c1b1bdf42381a1817c848c5a6516cfab9b7e767925b619a4beca5c7df
thread:issue:32375,issue #32375: Make _rpc_sync_torchscript and _rpc_async_torchscript work with autograd profiler,stale/low urgency,2,stale/low urgency,1,0,32375,,,,0,ef56370dcbd45946481fb06de41d35860bcd5a710eae4cbb7ef66fb66854e069
thread:issue:32390,issue #32390: [JIT] Compile group of functions/modules,stale/low urgency,2,stale/low urgency,1,0,32390,,,,0,b25408bd8d38bd0ae6d2287f59cfd3cf7a249c40cac6092b2cdcf3c5b36e871d
thread:issue:32402,issue #32402: torchvision,stale/low urgency,2,stale/low urgency,1,0,32402,,,,0,5e8aa95a301fe2e73a00569b331d1085ac815c81d6f16078fb89be5e333169e3
thread:issue:32403,issue #32403: How to accelerate the compiling of pytorch,stale/low urgency,2,stale/low urgency,1,0,32403,,,,0,5a867ef95925b44f4c4f2839cc5d0d87f690bbb4c1f88c29fd13896606edc639
thread:issue:32406,issue #32406: KL divergence for diagonal Gaussian distributions,stale/low urgency,2,stale/low urgency,1,0,32406,,,,0,bba9f281aa03ac6f7dd40c931e4ca1e86c439f559e596f0fd4a90409575bcf3d
thread:issue:32415,issue #32415: DistributedStreamSampler: support stream sampler in distributed setting,stale/low urgency,2,stale/low urgency,1,0,32415,,,,0,beb45eb61bac5523400645c1472779c99b7cf073847bc2b8caa7136524ca9c7d
thread:issue:32416,issue #32416: Is there a way to use SYSTEM_INSTALLED 3rdparty libraries?,stale/low urgency,2,stale/low urgency,1,0,32416,,,,0,0f736243e2b43d971a4adee9c2262509cde3900aadd029ba66ce2c08524496da
thread:issue:32419,issue #32419: MAGMA libraries,stale/low urgency,2,stale/low urgency,1,0,32419,,,,0,55895858dd73004f1b03a31fdaeffd4a6efccbff0caf353a64c1647edf9eacff
thread:issue:32421,issue #32421: Segmentation fault in C++ API torch::from_blob(...).clone(),stale/low urgency,2,stale/low urgency,1,0,32421,,,,0,cea6b788e57c25326e23fefaba3388b7dddad849697172a7e0cde61a4b62867a
thread:issue:32427,issue #32427: [C++] Don't use DeprecatedTypeProperties in torch::utils::reorder_tensors_like,stale/low urgency,2,stale/low urgency,1,0,32427,,,,0,a7bd7e015bb9012774bc320fc7f92624a3d86bfa7a8b08bb800a6a6270500935
thread:issue:32444,issue #32444: DataParallel does not work with sparse parameters,stale/low urgency,2,stale/low urgency,1,0,32444,,,,0,df17fab3d27d537d1bdaffee7d8ed2acc70e7032f5c6c6030471623ce14f15d6
thread:issue:32452,issue #32452: Numba Enhancement Proposal (NBEP) 7: External Memory Management Plugins,stale/low urgency,2,stale/low urgency,1,0,32452,,,,0,7e39007974a4c0b30617c5d617631a9cc5eaddddca05f8232eb68baccc3ce9d4
thread:issue:32488,issue #32488: TracerWarning When Using Tensor Size in Torchscript Trace,stale/low urgency,2,stale/low urgency,1,0,32488,,,,0,b4aaa00759ceedb6c91d6b3f54909606929eebe77e881d9c6a0bc7e235bf19ee
thread:issue:32490,"issue #32490: RuntimeError: has_marked_unused_parameters_ INTERNAL ASSERT FAILED at /pytorch/torch/csrc/distributed/c10d/reducer.cpp:290, please report a bug to PyTorch.",stale/low urgency,2,stale/low urgency,1,0,32490,,,,0,58c861f8600cff2d3a5e38fbbadb54f841bbcb8903e868d1a1922334b5e8ea7e
thread:issue:32497,issue #32497: Don't unnecesarily send cleanup dist autograd context RPCs to other nodes,stale/low urgency,2,stale/low urgency,1,0,32497,,,,0,9563cc3c91e9761ead6f49c5938044a5b58c179cb645c43c00825c55fc2e641b
thread:issue:32504,issue #32504: LibTorch operates very slowly on data blobs from GPU,stale/low urgency,2,stale/low urgency,1,0,32504,,,,0,8d97d7dbb6764c7432117206ce65bb03a92cc85a3228c687481aee2b66a06ad6
thread:issue:32510,issue #32510: Tensor.random_ is not implemented for bool on CUDA(but implemented on CPU),stale/low urgency,2,stale/low urgency,1,0,32510,,,,0,e6d5e9f89ccc63590d2ef239af53569aaed36b8483ce695b467ffcb3eb229160
thread:issue:32511,issue #32511: TorchScript C++ API Tracking Issue,stale/low urgency,2,stale/low urgency,1,0,32511,,,,0,4b237f24dee273268dbdf6f62529f189642c2126dc6ef7815d0a01254a0af5cd
thread:issue:32524,issue #32524: Pytorch 1.4 does not detect gpu,stale/low urgency,2,stale/low urgency,1,0,32524,,,,0,64e95cdaa388e677908100627d7061976ddc3f64eca77c5b30e95bc734c34349
thread:issue:32531,issue #32531: Batched torch.eig() and gradient of torch.eig() for real eigenvalues,stale/low urgency,2,stale/low urgency,1,0,32531,,,,0,03df681bdc8835eb546e230b7ce72129f78bc54373074babaac2b899fff53b25
thread:issue:32549,"issue #32549: ""Tried to register multiple operators with the same name and the same overload name"" error is confusing",stale/low urgency,2,stale/low urgency,1,0,32549,,,,0,850eeb7a8de796326af8b1a0ea189f92004e155c7afefb98d26eeff1b95d74d5
thread:issue:32552,issue #32552: MultivariateNormal.rsample: use eigen-decomposition when Cholesky fails,stale/low urgency,2,stale/low urgency,1,0,32552,,,,0,ee68f0f9460f46c82f4839ac0a9c1a9c0a3d313ce0d77b448b97a7f09d052f17
thread:issue:32565,issue #32565: Reduce RPC branches for Python/Built-inOp/TorchScript,stale/low urgency,2,stale/low urgency,1,0,32565,,,,0,9c9ef8ee6ed12b06984d1bb2b1e11540cb3bf997d858a4d584b776291825d267
thread:issue:32575,issue #32575: Interrupting a DDP worker while using the CUDA MPS server causes CUDA to hang until reboot,stale/low urgency,2,stale/low urgency,1,0,32575,,,,0,07b2ad41db197da4a8d5205be5c7e85e3329331c48f23595406cada1f6f85171
thread:issue:32576,issue #32576: Document different gradient wrt to jax when nesting,stale/low urgency,2,stale/low urgency,1,0,32576,,,,0,19d89a048de5803b14ef85690c5a2d47c83a10b1843d5f2a324d3e565198404c
thread:issue:32580,issue #32580: SGD optimizer with deprecation warning,stale/low urgency,2,stale/low urgency,1,0,32580,,,,0,8bccbdf0275e22d2baba4937dbbcffb1329820b51b6d987b9e5ec8eb36daf55c
thread:issue:32589,issue #32589: User-friendly handling of types and devices,stale/low urgency,2,stale/low urgency,1,0,32589,,,,0,654fc7f394efe817caeff5cc249b65d397fa7623924cf9fc3e826f389db22096
thread:issue:32614,issue #32614: Logical AND and OR for Tensors in C++ API.,stale/low urgency,2,stale/low urgency,1,0,32614,,,,0,2898183d19969536b822ac8d29b81d21bd62179873ee691724311f994dc92dfb
thread:issue:32641,issue #32641: changed format of trace graph in torch 1.4.0,stale/low urgency,2,stale/low urgency,1,0,32641,,,,0,20621e97a3972cde946e8e4b7d0d74da03f239368cbde730912cb47fc462b596
thread:issue:32687,issue #32687: Building PyTorch ignores my current Python version,stale/low urgency,2,stale/low urgency,1,0,32687,,,,0,6231624018423e808403b4f07da9631efe6da687848f8695f52f2b2b901fa1f1
thread:issue:32690,issue #32690: How to customize build torchscript model to be used in end devices codebase,stale/low urgency,2,stale/low urgency,1,0,32690,,,,0,ef55562f7e159a6c5742045223628d745c4a1c40adc391dc64f5608c0c98e36f
thread:issue:32694,issue #32694: ninja: build stopped: subcommand failed.,stale/low urgency,2,stale/low urgency,1,0,32694,,,,0,6e1975c5aceb6e36fd7b47b6419d8aaec7b08019f9fd48e3ea83b1eae7586b59
thread:issue:32695,issue #32695: torch.tensordot has inconsistent signature with torch script,stale/low urgency,2,stale/low urgency,1,0,32695,,,,0,f485cd7283e00cdb9059d28ac9f10e1a033d2e177239fa6117b38d80194ffc46
thread:issue:32698,issue #32698: Add Julia Bindings to Torch backend,stale/low urgency,2,stale/low urgency,1,0,32698,,,,0,36244cf186b637e1a337b50eaa7fbaa13f8803ae6aca0166e472685179f191cd
thread:issue:32706,issue #32706: Torch serialization does not restore tensors properly when custom __reduce__ is defined,stale/low urgency,2,stale/low urgency,1,0,32706,,,,0,100b181549b61869080ce33f4829a9e34dd44d4b80759e76ccc54e5755ccb0ed
thread:issue:32751,issue #32751: PyTorch freezes on second call to scripted densenet model from torchvision,stale/low urgency,2,stale/low urgency,1,0,32751,,,,0,bce970792fa8796b523158a30702567c6df0e0fed45601081209876bb61bf03d
thread:issue:32752,issue #32752: RPC mock mode for unit tests.,stale/low urgency,2,stale/low urgency,1,0,32752,,,,0,d19272ff2d847c00b210508339232f1acbb1482bb3f522dc73fd500c21a1c61e
thread:issue:32754,issue #32754: JIT performance discrepancies,stale/low urgency,2,stale/low urgency,1,0,32754,,,,0,cc1cfce9571e41ed9f23305919975cdaf8aa7607f9e080ef03b852f4abeff5dc
thread:issue:32767,issue #32767: Documentation is not loaded by IDEs,stale/low urgency,2,stale/low urgency,1,0,32767,,,,0,ca79c2007bfe8c637f871dd64d70df49d2292e45c8661588028b05ecfefe6462
thread:issue:32783,issue #32783: Comments Separating Class Methods from Different Classes in C++ Files,stale/low urgency,2,stale/low urgency,1,0,32783,,,,0,f0c4bb6f0887179f78bfe894e76e0e31117514de9d4f6094ffdb8baaf4effd31
thread:issue:32784,issue #32784: Implement Backend-Agnostic RPC functionality in RpcAgent,stale/low urgency,2,stale/low urgency,1,0,32784,,,,0,0a1f2f1269193274d7c9ac6915e050f4c0b557d6bf8aad84ef8ebbc66f1d968a
thread:issue:3281,"issue #3281: DataLoader ""casting"" non statndard objects to lists",stale/low urgency,2,stale/low urgency,1,0,3281,,,,0,750745e3227dfc10cf4cabe8076ac7fec5394a4ee5114dcab11b8af2e80e4002
thread:issue:32822,issue #32822: Error tracing custom autograd.Function,stale/low urgency,2,stale/low urgency,1,0,32822,,,,0,29692fd9479d37f08059c7bcd9717866eff9b3a986adc4a617f02b221bd5c2fd
thread:issue:32826,issue #32826: MKLDNN doesnt work and is slower than normal cpu mode,stale/low urgency,2,stale/low urgency,1,0,32826,,,,0,21cfa878c5f6fa0dc8bf3bcb194055f818987ea25366a31a1c412475c0065a4d
thread:issue:32840,issue #32840: Can we add support for Enum in scripted models?,stale/low urgency,2,stale/low urgency,1,0,32840,,,,0,03a10bf792f637eb43b145b01505e962e36ad28c1122a0efa7db3bf500a2e392
thread:issue:32868,issue #32868: __cuda_array_interface__ conversion does not support readonly arrays,stale/low urgency,2,stale/low urgency,1,0,32868,,,,0,7d0d238336e691db039637a12a694a2830639658be80b4428f2e316e7c37d06b
thread:issue:32872,issue #32872: Confusing error messages of tensor.scatter_ on both CPU and CUDA,stale/low urgency,2,stale/low urgency,1,0,32872,,,,0,e8c4d23ad8fe8365ff497851d0ad48689251f76326c279bd95813f27f5361d1c
thread:issue:32881,issue #32881: CUDNN_STATUS_EXECUTION_FAILED,stale/low urgency,2,stale/low urgency,1,0,32881,,,,0,dab234afa4f16f1df1ffb88b6f745834d2dae4a4a072841fff8689c170750d64
thread:issue:32884,issue #32884: [RFC] Nested scopes in autograd profiler should support RPC calls properly.,stale/low urgency,2,stale/low urgency,1,0,32884,,,,0,88541eaaa5d32b4388dd6d6dfb904eb2e5dd00123a13576913e8c9f828c42afa
thread:issue:32909,issue #32909: Some questions Concerning Intel's DNNL(MKLDNN) support in Pytorch (adding support for Intel Processors GPUs) by transitioning to DNNL,stale/low urgency,2,stale/low urgency,1,0,32909,,,,0,bc27f82cd7b485b5929b9f3f3a56eabf1528ac1f7330435a6b21a654e4d95ef6
thread:issue:32912,issue #32912: Optional seq_len argument to torch.nn.utils.rnn.pad_sequence,stale/low urgency,2,stale/low urgency,1,0,32912,,,,0,181befe05631b4403410dc7b1c2d7cf19d78ebc3ad7c38c9916a2238a85d4bdd
thread:issue:32916,issue #32916: from torch._C import * (ImportError: DLL load failed),stale/low urgency,2,stale/low urgency,1,0,32916,,,,0,cc786499797621fa6ab2f999b2ba4b3a6d2a12e59aa7460cad4d26d68e07e28f
thread:issue:32919,"issue #32919: Torchscript used to work, but now it fails with VariableTensorId error",stale/low urgency,2,stale/low urgency,1,0,32919,,,,0,271085688d7a486eb06138801c4e1294f411f3dbf2add8ab3ddb2c91cae14e20
thread:issue:32924,issue #32924: Connect timeout feature do not work in DDP with TCPStore,stale/low urgency,2,stale/low urgency,1,0,32924,,,,0,ad712255779dd290d9566f0f4511355c30946084419c9b8be7be46a87f5a6938
thread:issue:32937,issue #32937: Operation Registration Error,stale/low urgency,2,stale/low urgency,1,0,32937,,,,0,1c17c793379bd515e23b9af4e1640c55e6b0714e2e79314d535f83a01b3a7896
thread:issue:32972,"issue #32972: which pytorch do i install for running this project in my windows.link for this is :: https://github.com/xiaojunxu/SQLNet .IN this they have used python 2.7 ,but i am unable to install pytorch on python 2.7 environment .help me with this",stale/low urgency,2,stale/low urgency,1,0,32972,,,,0,429021a8ebbcee10acf23aa73d4524152530fa8188250dccad85b3e572bdbffe
thread:issue:32988,issue #32988: [jit] Dict set item type mismatch error doesn't say the type was inferred,stale/low urgency,2,stale/low urgency,1,0,32988,,,,0,c068a550957d514875b44a18e0734e94e53a4f19f3722073ad0624bcf58b216c
thread:issue:33006,issue #33006: matmul: no warning when contracting differently named dimensions,stale/low urgency,2,stale/low urgency,1,0,33006,,,,0,c896452eec6c5233c7219ff00afeea9d8f71735c562d9c19dbce8a412f99e5e2
thread:issue:33007,issue #33007: Upper/Lower attributes for named dimensions for proper Ricci notation and to generalize matrix operations,stale/low urgency,2,stale/low urgency,1,0,33007,,,,0,b5125a58459ada5f2ecc6a355b8a87df8ba06cc45cc10b0fda05bc338234bb2f
thread:issue:33035,issue #33035: Support batch linear transformation,stale/low urgency,2,stale/low urgency,1,0,33035,,,,0,75db3acc60e48369f7365497095a6e8b0c973f4c5432bbd355b629f55146e860
thread:issue:33040,issue #33040: [debatable] Better infer dtype in torch.as_tensor,stale/low urgency,2,stale/low urgency,1,0,33040,,,,0,2414f36b17f78d37927ab2798295dda3027ae9afe616470332b621199969adb1
thread:issue:33049,issue #33049: Warning when link libtorch and opencv4.2.0 together,stale/low urgency,2,stale/low urgency,1,0,33049,,,,0,48babfb4c005cc4f000fc6165b541d2e7e1701b6d7ba6f83ed5f0a5a01306524
thread:issue:33051,issue #33051: Recover from CUDA runtime error,stale/low urgency,2,stale/low urgency,1,0,33051,,,,0,0abff39738f8a8706f587d4901691c725b9e689d2ebbaf99bfba8e7333ad5f6f
thread:issue:33055,issue #33055: [docs] Strange order of items in docs contents in left pane,stale/low urgency,2,stale/low urgency,1,0,33055,,,,0,ef48cdd19e3ab53ab400cd1cb2a07659f0e4c977bfcab3395a7847ce7aa645c4
thread:issue:33081,issue #33081: DataParallel gives different gradients when using LSTMs,stale/low urgency,2,stale/low urgency,1,0,33081,,,,0,ec4a33ced0bdfc9794728c53efaa680e9c72bb90b995cdfc5cba8aa7e3e88413
thread:issue:33085,issue #33085: not able to import * from fastai.vision in Google collab,stale/low urgency,2,stale/low urgency,1,0,33085,,,,0,083dd14d7e2217752259829f4bf0aa46500e4e16f4af7743905efd6981281b0b
thread:issue:33086,issue #33086: F.max_pool*d/F.min_pool*d should support integer dtypes and bool tensors,stale/low urgency,2,stale/low urgency,1,0,33086,,,,0,1094e24c2950c21550870fb239966925eb73411d3442dc86ea86cdfe4a402827
thread:issue:33089,issue #33089: test_baddbmm_cpu_float32 fails locally for me when built with DEBUG=1,stale/low urgency,2,stale/low urgency,1,0,33089,,,,0,f7c3a2d6dc000d7b10497b4b810f66154767fec8fa88109a705f08e21e0f2660
thread:issue:33114,issue #33114: Move the custom pass execution back to the beginning of runNondiffOptimization,stale/low urgency,2,stale/low urgency,1,0,33114,,,,0,22850534adcdd750ae65d6f6b72f6bc9d1083581863b9bfa3b91db036edccf36
thread:issue:33115,issue #33115: [RFC] Add ability to get all remote parameters when constructing DistributedOptimizer.,stale/low urgency,2,stale/low urgency,1,0,33115,,,,0,df4c207f8962e90d92295aa02940c6f7a65eadc402332dba7f155603eeba96bf
thread:issue:33122,issue #33122: [feature request] make torch.multinomial behaviour compliant with rnn output dimension,stale/low urgency,2,stale/low urgency,1,0,33122,,,,0,3a8f38efd0233979b5f933aaef07af27b350451b1733a18f629696c6dd09a018
thread:issue:33129,issue #33129: Python package using CMake,stale/low urgency,2,stale/low urgency,1,0,33129,,,,0,8d485d4b7ba5171df4bab7ee2a13bd6c1cdc8274c84eb72275cf11b36e43f379
thread:issue:33166,issue #33166: TensorIterator does not work with different input/output types,stale/low urgency,2,stale/low urgency,1,0,33166,,,,0,8b7790315b6cf30542fc55449acc9899388ffa71c571cbba1a6b7d5bd391c4de
thread:issue:33180,issue #33180: Loading pretrained model,stale/low urgency,2,stale/low urgency,1,0,33180,,,,0,6138dca4d2093ae4bb0ae4d943edf7c7afe5c3b888d3ae9909a1cca8d697046a
thread:issue:33241,"issue #33241: [feature request] Add ""groups"" argument to nn.Fold and nn.Unfold",stale/low urgency,2,stale/low urgency,1,0,33241,,,,0,e2a060833fd3d4670dd7063fee74da9fb72ef0a729d9de983dfd7b7cbfd8bd1d
thread:issue:33249,issue #33249: Core dumps being created when running test_c10d.py and test_multiprocessing_spawn.py,stale/low urgency,2,stale/low urgency,1,0,33249,,,,0,50ec955d7fd03e071dfe6fbeee2cb4d8364f36a6c41264d7efd15e5a694dddd7
thread:issue:33298,issue #33298: [feature request] torch.expand to not require unsqueeze and match -1 to existing dimensions if tensor.shape.count(-1) == tensor.ndim,stale/low urgency,2,stale/low urgency,1,0,33298,,,,0,c76ef49addede505f5a39d01a88fcabddbdf0e1c2e73acd679b9d556156e83d3
thread:issue:33301,issue #33301: Don't take TensorOptions by reference,stale/low urgency,2,stale/low urgency,1,0,33301,,,,0,165c116a8d46f6972df92c1e871dccdd8facbe6120e56217d9e2c49c9048d6b8
thread:issue:33303,issue #33303: torch::var_out and dimnames,stale/low urgency,2,stale/low urgency,1,0,33303,,,,0,3e24d9f39bd2b3df23778219fa641ec37d3b91cd65ad22b22c1c97437d274257
thread:issue:33327,issue #33327: Creating Torch tensors slows OpenCV video reading a lot,stale/low urgency,2,stale/low urgency,1,0,33327,,,,0,f569bb1cd5a122ca567f10b85e1af1d56b04b3b9818f24fb233b2a54f6c1dc55
thread:issue:33341,issue #33341: how-to-adjust-learning-rate using libtorch,stale/low urgency,2,stale/low urgency,1,0,33341,,,,0,678c9eb525e6b646a1aaa823f55d1c51f68f1b346e50f965db7c626822d116cd
thread:issue:33342,issue #33342: malloc(): memory corruption (fast),stale/low urgency,2,stale/low urgency,1,0,33342,,,,0,b015651c88f501e8b4cfe4cacd68b995412bc63d7dc90661933c691e549a1190
thread:issue:33353,issue #33353: [jit] `Node` callstack is incorrect,stale/low urgency,2,stale/low urgency,1,0,33353,,,,0,732ae109895023a83bbe4f15fb2d82a12789e3d1ed5f0b65b4a5a3d596e693fd
thread:issue:33360,issue #33360: Dropout of attention weights in function F.multi_head_attention_forward() breaks sum-to-1 constraint,stale/low urgency,2,stale/low urgency,1,0,33360,,,,0,0b5bd6b1012694c5ba2980fc4dc57f52bd019962d0defa63a636f08e67f3ba91
thread:issue:33363,issue #33363: Make it easier to add new messages in RPC layer,stale/low urgency,2,stale/low urgency,1,0,33363,,,,0,4aa2ef46e8dd504febd4cc6279517724fbc5c035df578e9b620159315e8e0ae6
thread:issue:33367,issue #33367: Expose `internal::GRAIN_SIZE` through Python API.,stale/low urgency,2,stale/low urgency,1,0,33367,,,,0,a720a18c44b5ac2a58649c3fe00c0864528e3edc51a98a8012e9c98d909b4693
thread:issue:33373,issue #33373: torch.clamp_ not inplace during backward,stale/low urgency,2,stale/low urgency,1,0,33373,,,,0,4f66fbfd108f4bc00f4cc672fce373c20ea825ae3caff4da544e7960c8cea7f5
thread:issue:33385,issue #33385: Illegal instruction: 4 - OSX 10.13.6 install from source,stale/low urgency,2,stale/low urgency,1,0,33385,,,,0,05dd3f41c66d2a4f81f6dc12d0f4fe0fbf8958dc6f80da786fc48137b48f9ab6
thread:issue:33388,issue #33388: Sobol engine generates out-of-bounds samples after drawing too many samples,stale/low urgency,2,stale/low urgency,1,0,33388,,,,0,71f10655ab95028b0d47823a4d18c54bef470d7da2f9eec2124f04ebd63a3442
thread:issue:33390,issue #33390: It is not good to separate the steps of modules making and forward computation,stale/low urgency,2,stale/low urgency,1,0,33390,,,,0,98cc8c55b788dd61e607597ca428f099e5b7700d476d84bf88f02b8eac052fc9
thread:issue:33396,issue #33396: PyTorch 1.4.0 does not support using `Module` or `ModuleList` in attribute annotations in ScriptModule,stale/low urgency,2,stale/low urgency,1,0,33396,,,,0,a24f88ead8279baeb76fcb242b518abda4e292786f168d76f2bd7568a8603f8d
thread:issue:33413,issue #33413: IterableDataset with num_workers > 0 and drop_last=True drops more instances than expected,stale/low urgency,2,stale/low urgency,1,0,33413,,,,0,658b4e3b4b48097243268774fc157986820e7927798bd26975d2228a76336e5b
thread:issue:33436,issue #33436: It seems nn.Sequential.add_module() could take duplicate names,stale/low urgency,2,stale/low urgency,1,0,33436,,,,0,ccea40c8752ab5ab178cf5932893257b55e3db9c2682882de6f8bfcae84d3c91
thread:issue:33463,issue #33463: torch._C.Node.scopeName() missing in pytorch 1.4,stale/low urgency,2,stale/low urgency,1,0,33463,,,,0,eff17879d544ed1a3c4cba1bdb263070446a060d62c36015e39017323e6c6d2e
thread:issue:33476,issue #33476: Make ArrayRef::size() return int64_t rather than size_t,stale/low urgency,2,stale/low urgency,1,0,33476,,,,0,8d9ebffa20d8539312814bd2119a1b55f47b2863f70464bc34cd24e5708c98f4
thread:issue:33477,"issue #33477: Add 32-bit CI (e.g., Raspberry PI CI)",stale/low urgency,2,stale/low urgency,1,0,33477,,,,0,ebbbd6ba395418c48a4f965aa0d1b6219264ceb2f9c991a16daf6e55f3f03e12
thread:issue:33482,issue #33482: Distributed Data Parallel for computation graphs that make RPCs in forward(),stale/low urgency,2,stale/low urgency,1,0,33482,,,,0,9db008774286c5ebcf2129ad3c8e3e827eefee2cb1bed640d76445681e8e4fbf
thread:issue:33501,"issue #33501: When a Node fails to resolve to an Operator, print out the types of arguments, and all ""close matches"" in known operators",stale/low urgency,2,stale/low urgency,1,0,33501,,,,0,0104995ff633c4c419718261c93a86040ebdf30f0fb7a4747f178b17167b3f66
thread:issue:33514,"issue #33514: Make setter non-optional, e.g., TensorOptions::device(optional<Device>) -> device(Device), and add a device_opt setter",stale/low urgency,2,stale/low urgency,1,0,33514,,,,0,7cedc471129f91695edb0f8cd77beaa9ea025724a08c4d88305678fa52365a6e
thread:issue:33520,issue #33520: Reverse Cumulative Sum,stale/low urgency,2,stale/low urgency,1,0,33520,,,,0,adb29a9d6e840c2f606112cdfa872c6285ed38f4290d708cc22fc45e4536ce7b
thread:issue:3356,issue #3356: Data sampling seems to be more complicated than necessary,stale/low urgency,2,stale/low urgency,1,0,3356,,,,0,a161d5acfabc6bd3cbe3c521303bd28dc500007cc1f54aa5c5975336dc1dec2b
thread:issue:33571,issue #33571: empty_sparse shouldn't be called with memory layout but is,stale/low urgency,2,stale/low urgency,1,0,33571,,,,0,d35dc78e39f820d70e465953bfb0e60af5aa8f12972df35f4f9573fbc7844f50
thread:issue:33611,issue #33611: Resnet 32x32 in caffe2 generates ResNet with wrong number of layers,stale/low urgency,2,stale/low urgency,1,0,33611,,,,0,2b27c6deafcea5e354d86b6d84f42c408e69ea80b8becdd5dd9e54434be1d1d0
thread:issue:33628,issue #33628: Remove .data subset1 for fixathon,stale/low urgency,2,stale/low urgency,1,0,33628,,,,0,de15ed6cab87707c83134f3c5fbc2f81dbbb4fb64fa2165fe34295ea0d245d92
thread:issue:3364,issue #3364: type of torch.bernoulli and torch.multinomial inconsistent,stale/low urgency,2,stale/low urgency,1,0,3364,,,,0,e5bf1b9d35b208df39a418d1f7030eee479f5126b02e179a7041b0bea5eb48a0
thread:issue:33653,issue #33653: master build error,stale/low urgency,2,stale/low urgency,1,0,33653,,,,0,2db420104e0885b551da360703f65065cce941c217b69724fd91e8ad4135c495
thread:issue:33664,issue #33664: torch.rand() not having same values on using torch.manual_seed(0),stale/low urgency,2,stale/low urgency,1,0,33664,,,,0,66d77d9fe6e9b847dc9e21a960bd381b43d6df48058f0650e0e6ee5bc8e74fb9
thread:issue:33668,"issue #33668: /usr/bin/x86_64-linux-gnu-ld: warning: libcusparse.so.10.0, needed by /pytorch_master/build/lib/libtorch_cuda.so, not found (try using -rpath or -rpath-link)",stale/low urgency,2,stale/low urgency,1,0,33668,,,,0,109fb8e66f0c1710c241d3be9b4161c57eae39142d660a8eea233201672a5bd8
thread:issue:33672,issue #33672: [dev] `RecursiveScriptModule` does not expose `jit.ignore`d methods,stale/low urgency,2,stale/low urgency,1,0,33672,,,,0,17b6e1a3decebd79afa77e07c989245bd8957df35572496f3404daa1291a89bf
thread:issue:33684,issue #33684: Decouple Lifetime of Local RRef and RPC,stale/low urgency,2,stale/low urgency,1,0,33684,,,,0,74b3bbb1b79767b202f794f6c6f36501e8168305b88f9e7433906e17a895f37b
thread:issue:33725,issue #33725: Tensor.random_ is not implemented for bfloat16 on CPU(but implemented on CUDA),stale/low urgency,2,stale/low urgency,1,0,33725,,,,0,4d0982e75fc72aa8d37617bcac237a0895db1f6c7abae79f9509e69d4b16211d
thread:issue:33738,issue #33738: Support pipelining the backward pass and optimizer.step() for distributed autograd.,stale/low urgency,2,stale/low urgency,1,0,33738,,,,0,a1180a63ec6fa101347f22f5130f4253b7d0137b05ae055f202b0272dacce865
thread:issue:33782,issue #33782: [JIT] BroadcastingList annotations don't work with ignore'd functions,stale/low urgency,2,stale/low urgency,1,0,33782,,,,0,47bfad652b31c040641c1c6c012de129be19554d83e056ed70817b8e9be1b6ef
thread:issue:33785,issue #33785: Deepcopy fails with nn.parallel.replicate,stale/low urgency,2,stale/low urgency,1,0,33785,,,,0,cdf762831acd01dcc32451ae7a653f8e7c352d20ed36497609ffbc67047347ed
thread:issue:33787,issue #33787: MSE Loss Implementation,stale/low urgency,2,stale/low urgency,1,0,33787,,,,0,5bf3d1f2797350631b0556df794f3edb65cc895dc0f80c94ce0089d216da6301
thread:issue:33799,issue #33799: Add `start_process_in_context` to `torch.multiprocessing`,stale/low urgency,2,stale/low urgency,1,0,33799,,,,0,c8925a7280b93a81b6a28a3f69bc923c923976d3798021273db882c64f38a8e1
thread:issue:33800,issue #33800: Can't get module gradient in autograd.Function's custom backward when DataParallel is used,stale/low urgency,2,stale/low urgency,1,0,33800,,,,0,345962b0df4e0a7db8daf8fd2b48c204ebac7b05ce740089fe2e38ee3f1f2b3e
thread:issue:33811,"issue #33811: Models saved in C++ LibTorch with torch::save, cannot be loaded in python using torch.load",stale/low urgency,2,stale/low urgency,1,0,33811,,,,0,13ad60e1995ceea1db2315f26a443be5130e9df79b1df320dac16d6328264aba
thread:issue:33844,issue #33844: [jit] Returning different types with `Any` segfaults,stale/low urgency,2,stale/low urgency,1,0,33844,,,,0,751911f86ade85c4127b97f1b2af0ad973e335d2d6eea837a57e5c3a10b2a165
thread:issue:33855,issue #33855: Unclear output for Pytorch Profiler,stale/low urgency,2,stale/low urgency,1,0,33855,,,,0,9b268f6eb67065ab2254f6d430a1176150b5cb94442972ac382978fccc13e427
thread:issue:33859,issue #33859: Possibility to support int4 data type,stale/low urgency,2,stale/low urgency,1,0,33859,,,,0,8bf0d3b4184412e7ec9ec38c1e7d5354ed8d9d437b38d9a3388d6b81dfbddceb
thread:issue:33864,issue #33864: Memory leak in embedding layer and LSTM,stale/low urgency,2,stale/low urgency,1,0,33864,,,,0,e3ed5c74a1c46a745639dca879a5b90baf8adff51e994e53c7c72b254b210b80
thread:issue:33876,issue #33876: Torchscript incompatible with torch.cat for tensor lists,stale/low urgency,2,stale/low urgency,1,0,33876,,,,0,266b2c798863ca8e2c41ea569619b61b25fdc4a2ef0b770295fe3ac9c20429c0
thread:issue:33887,issue #33887: [jit] Python arg parser / TorchScript incompatibilities,stale/low urgency,2,stale/low urgency,1,0,33887,,,,0,504b3652c76bd9163a3f68ee735e7eb6ee964a1564edd76d2788055155274a98
thread:issue:33890,issue #33890: Libtorch segfault when used with libqpOASES,stale/low urgency,2,stale/low urgency,1,0,33890,,,,0,52e592499e05ad2c3e4c8cceab65cd7daea1b9fcb173f1a5eb682c46c0f31f99
thread:issue:33898,issue #33898: Docs for uniform_ don't make any sense,stale/low urgency,2,stale/low urgency,1,0,33898,,,,0,0a6a17d73a037e64835b1e061f282ed7d5108c85394cbfa1627e6a2ebbafe171
thread:issue:3390,issue #3390: High CPU use by clock_gettime syscall,stale/low urgency,2,stale/low urgency,1,0,3390,,,,0,5b8f6ca71350a290b634d9460a19bbbdbd7bdf6f5ed911f1a9a77d16683c72b2
thread:issue:3396,issue #3396: CUDA topk is slow for some input sizes,stale/low urgency,2,stale/low urgency,1,0,3396,,,,0,7c33bd2414673ffaa098af0b64eb4dd77a4fd880dff4b811481bc5082cfc3bdb
thread:issue:33979,issue #33979: TensorPipes RPC Agent CUDA Support,stale/low urgency,2,stale/low urgency,1,0,33979,,,,0,23404ac055eaceca251c92a6db54569709f728592fd3107c451a2b01bc69a39a
thread:issue:33982,issue #33982: Initialize TensorPipe RPC Agent Transport,stale/low urgency,2,stale/low urgency,1,0,33982,,,,0,2640aee8c238e0566cb7671e98a907e0ca92f0553be69dac6b5af8b30bacf37b
thread:issue:33983,issue #33983: TensorPipes RPC Agent Listener shortcut,stale/low urgency,2,stale/low urgency,1,0,33983,,,,0,24ca5701c8d7c0bce2081b037d52fb5074dfd2c1e56a68ffee1f571dacbeab6f
thread:issue:33989,issue #33989: TensorPipes RPC Agent Message Acknowledgements,stale/low urgency,2,stale/low urgency,1,0,33989,,,,0,6a5a702b6583fe844f5764dce7e0c7b6057e278ab3e208d7fac18e51473359a3
thread:issue:33990,issue #33990: TensorPipes RPC Agent Multiple Placement Retries,stale/low urgency,2,stale/low urgency,1,0,33990,,,,0,b01dd939116155a2da886e61242a0b89d828b13ce25528b0749de8bb62850a91
thread:issue:33991,issue #33991: TensorPipes RPC Agent Default Args/Result Device Mapping,stale/low urgency,2,stale/low urgency,1,0,33991,,,,0,5810fa31affc0109d9d4a285d5f2024610f044e0201c76501dc18967e071ef37
thread:issue:34001,issue #34001: _broadcast_coalesced_reshape doesn't respect zero-size tensor,stale/low urgency,2,stale/low urgency,1,0,34001,,,,0,725ad4b068c232270a4efb86428b0618484012f01e691ca4a32a3514fa8acfa4
thread:issue:34052,issue #34052: Model loaded in C++ runtime is not thread safe,stale/low urgency,2,stale/low urgency,1,0,34052,,,,0,5397d9a16e5ffd6253065aa5e0e4636f3d2633f89194ebf3474c63176ff441db
thread:issue:34063,issue #34063: [docs] Unclear description of indices arg in torch.index_put_,stale/low urgency,2,stale/low urgency,1,0,34063,,,,0,13aa5ba0dce0804c8284cfb9e3a7b0d5070fff790e8e26bdfc0da88c7664d256
thread:issue:34064,issue #34064: pip --requirement installs incorrect CUDA version,stale/low urgency,2,stale/low urgency,1,0,34064,,,,0,79312c54fc09a3af61753edf08e7131d74a96309d1ae2001e126ce1a49480e5f
thread:issue:34068,issue #34068: [C++ API Parity] Incorrect documentation for optim initialization in serialization docs,stale/low urgency,2,stale/low urgency,1,0,34068,,,,0,050fe02bf69971a91d84759a52548e1b70864015f0d8937a39344fc806ad48c3
thread:issue:34080,issue #34080: RPC API Changes for TensorPipes,stale/low urgency,2,stale/low urgency,1,0,34080,,,,0,d362d09ac8f14e669bc3afad487cbddd2b0331db3fe462155b21e860786bde86
thread:issue:34082,issue #34082: Autograd deep copy avoidance optimization unsound in the presence of views,stale/low urgency,2,stale/low urgency,1,0,34082,,,,0,c52a9c4bc185a19fe78dccd07e2de786f07d6a3b9a4a43850a6a9cccf88e5a51
thread:issue:34086,issue #34086: Enable profiler tracing tests on windows,stale/low urgency,2,stale/low urgency,1,0,34086,,,,0,4ceca9bb1e22c2282dd707207de0dd4714b4ecc44acdf8dd582257ded9230039
thread:issue:34126,issue #34126: torch::jit::script::Object::attr should throw AttributeError instead of RuntimeError,stale/low urgency,2,stale/low urgency,1,0,34126,,,,0,87d734b6de2665816e763d4df3e74d642a71d9475cd8d7bf8eabe135dc1b5084
thread:issue:34172,issue #34172: END_HANDLE_TH_ERRORS_PYBIND prevents pybind11 Exception translation,stale/low urgency,2,stale/low urgency,1,0,34172,,,,0,eb14eaae2a5f1f9b991ba5d894b38387c6af3d88a64432347b2baedceef3aa43
thread:issue:34207,issue #34207: Support vars/__dict__ on torch.return_types tuples e.g. topk,stale/low urgency,2,stale/low urgency,1,0,34207,,,,0,4e29a96e3d200617a68981d36a335ac37dab2cf4804b1249a579823fbdfd1d55
thread:issue:34210,issue #34210: [docs] Unclear arg spec for torch.full,stale/low urgency,2,stale/low urgency,1,0,34210,,,,0,b1b15827115465e7ec7df8afd55c6b93512d0d86d48f0194e929c038bd660bad
thread:issue:34300,issue #34300: Inconsistent semantics of converting inf/-inf to long,stale/low urgency,2,stale/low urgency,1,0,34300,,,,0,1f6112d732c54aa24a9dc2f459e7c419ad366a3b9b5823ddb957248bc1bfcc29
thread:issue:34323,issue #34323: PyTorch GPU memory allocation,stale/low urgency,2,stale/low urgency,1,0,34323,,,,0,5bc16d3533f6c79a008dafd7a73e5c844b18367cbb9175880381246bc6a5b82a
thread:issue:34367,issue #34367: Better testing of the autograd engine,stale/low urgency,2,stale/low urgency,1,0,34367,,,,0,a2d4caf6e6110b0871188cfd6c17e126c716d2e8f6829c0034bf8f3a3593ec7e
thread:issue:34375,issue #34375: .detach() behaves differently for dense tensors vs sparse tensors,stale/low urgency,2,stale/low urgency,1,0,34375,,,,0,eb0172d60112e1143fed1c0842609f71014a0e365ebb49e8be1e90a8f8ef13d1
thread:issue:34409,issue #34409: PyTorch build with Ideep support?,stale/low urgency,2,stale/low urgency,1,0,34409,,,,0,fad135166936e09bca62d96f5e013cb1c189b69ced4ca2c95d6feed76944000e
thread:issue:34412,issue #34412: error in pytorch Docs: https://pytorch.org/docs/stable/distributions.html,stale/low urgency,2,stale/low urgency,1,0,34412,,,,0,a2120ef70a69285e79d998f1531ab7a064556d6788c2d6849dbe523c9fb2d609
thread:issue:34451,issue #34451: Mac build from source error,stale/low urgency,2,stale/low urgency,1,0,34451,,,,0,1b8a3d15e93374bf322fc3a1d856dd2c689a78a3349ea35ae05aa19c42740d66
thread:issue:34453,issue #34453: [Feature request] Query padding mask for nn.MultiheadAttention,stale/low urgency,2,stale/low urgency,1,0,34453,,,,0,c052a078bbcda9dd3c41927f4a12a02aba955088fbac30b8bd11d9c44a76a6cc
thread:issue:34483,issue #34483: JIT: Tracing faster than scripting,stale/low urgency,2,stale/low urgency,1,0,34483,,,,0,986200b89cdeb07dcc4a4bb9049cbccb20a79adf10b5ce1da08c84a6dae69a1c
thread:issue:34516,issue #34516: Expose chunk_sizes for DataParallel,stale/low urgency,2,stale/low urgency,1,0,34516,,,,0,fae626f376ed6fcc84b6ba57194bd8ec82616e5c030be49bc0efcd5296e675c5
thread:issue:34576,issue #34576: Significant speed difference between P100 and V100,stale/low urgency,2,stale/low urgency,1,0,34576,,,,0,26aa8b6e9554dfd7dca2defecbf081977eb0dae02be3cf7b165cc493a77e2973
thread:issue:34611,issue #34611: Eigen version for PyTorch ?,stale/low urgency,2,stale/low urgency,1,0,34611,,,,0,0b7bd8c769f2c8f62324ee56d837c421122c77c717b6bf3c62f50d498277ef24
thread:issue:34621,issue #34621: A better way to show users all build options,stale/low urgency,2,stale/low urgency,1,0,34621,,,,0,d23e233bd271ee01d9c2f0ccd3120fad8caab00422c8ba62a20bd044ea45c2f2
thread:issue:34660,"issue #34660: Add ""strict"" flag to ignore missing parameters in Optimizer.load_state_dict",stale/low urgency,2,stale/low urgency,1,0,34660,,,,0,4bfe37f45a4c5ffa589087e8ce9ad5d01e426d8a9a4a181e79cc103fab311188
thread:issue:3468,issue #3468: improve performance of common CPU clone / contiguous calls with HPTT,stale/low urgency,2,stale/low urgency,1,0,3468,,,,0,3d1b35f296a84b87e947d34f7666433395685b84c78a70d92731d8eda3f1cee8
thread:issue:34715,issue #34715: Maybe gpu_kernel shouldn't ASSERT_HOST_DEVICE_LAMBDA,stale/low urgency,2,stale/low urgency,1,0,34715,,,,0,bbed0a3462d0b59100ede6602c66cff7e5b142c70bcef83097aea86a93305cf6
thread:issue:34728,issue #34728: CPU softmax performance poor when dim is not the last dimension,stale/low urgency,2,stale/low urgency,1,0,34728,,,,0,1abf120ec637fdbbc922827ed98c08924da739ad19e6c19bc5700732835e5be0
thread:issue:3473,issue #3473: Feature Request: Distributed send arbitrary objects,stale/low urgency,2,stale/low urgency,1,0,3473,,,,0,88e66c92d8f489216a2957461fa1f7b5acabef0bdbec71c80bf7df249c666ffa
thread:issue:34754,issue #34754: General Purpose Faulty RPC Agent,stale/low urgency,2,stale/low urgency,1,0,34754,,,,0,dc9d5be403c59635f44705ba2fa15f9e7808c0d59f02525dab4aa1c2096b161d
thread:issue:34760,issue #34760: incomplete implementation doubt,stale/low urgency,2,stale/low urgency,1,0,34760,,,,0,682c1966dc52a752d021da067d9afcc886709b7f6bd0f0f928dd3ef8fffe2824
thread:issue:34765,issue #34765: Add ability to return a copy to Module's state_dict member function.,stale/low urgency,2,stale/low urgency,1,0,34765,,,,0,503ce16e4f71657282f028244fec2c1e4f1b2cc8c560119686212fb31c7ac373
thread:issue:34772,"issue #34772: Unsupported ONNX op (Upsample 3D, bicubic) contrary to documentation",stale/low urgency,2,stale/low urgency,1,0,34772,,,,0,b1e3240e79f435263f57cf0d2e4a844e05cf1a5d3080780c4837ad6dc82c70a7
thread:issue:34799,issue #34799: Is it possible to run an object detection android app by using Pytorch Mobile?,stale/low urgency,2,stale/low urgency,1,0,34799,,,,0,36ba002e0f8cdf8bfd7d1a7ee8a6ebbfb0f8f5c87423801c3609cb8260801c6a
thread:issue:34812,"issue #34812: Semantic differences between forward and backward of upsampling2d/3d in channels last (NHWC, NDHWC) format.",stale/low urgency,2,stale/low urgency,1,0,34812,,,,0,3259ee82dfeccc7404691f68fe5086944333eb427c36233664f74f439d076620
thread:issue:34814,issue #34814: [JIT] Tensor.add_ incorrect schema matching on overloads,stale/low urgency,2,stale/low urgency,1,0,34814,,,,0,a60abf94982bde6e95326b0b5ff95e2d39fc6326c9240f1e9372bddace28e9a0
thread:issue:34833,issue #34833: NO_CUDA and NO_DISTRIBUTED referenced in docs,stale/low urgency,2,stale/low urgency,1,0,34833,,,,0,3791a38ee041e7737584627801834f9972a31ce25386799526a807bd43af344f
thread:issue:34837,issue #34837: [torch.jit.script] Support tensor indexing after an Ellipsis,stale/low urgency,2,stale/low urgency,1,0,34837,,,,0,064bd3b19db9bb3916fbe837532c05db0369eeb544225c62af1a62c06a89e881
thread:issue:34839,issue #34839: [torch.jit.script] Allow `range` to index into Tensor,stale/low urgency,2,stale/low urgency,1,0,34839,,,,0,15e244409041717dcf9d0613b1a526fa05cfee1585c169d8157680e488016f17
thread:issue:34874,issue #34874: Extend nn.functional.softmax for arbitrary dimensions,stale/low urgency,2,stale/low urgency,1,0,34874,,,,0,5555e56f2587269ed908288156a7f6073914c182e0ad0ca62e2c8b204a329966
thread:issue:34878,issue #34878: Set the number of CUDAStream,stale/low urgency,2,stale/low urgency,1,0,34878,,,,0,353117b60af7a28d822de58b60809cfede6618f216413fac7838f155619a3a4e
thread:issue:34879,issue #34879: Addition of Siamese loss or contrastive loss function,stale/low urgency,2,stale/low urgency,1,0,34879,,,,0,dda9e6defdedd14f3e8bc61a94b86246922aea1740a6bec7aba489505a89a8e4
thread:issue:34880,issue #34880: Expanded tensors don't work as registered buffers,stale/low urgency,2,stale/low urgency,1,0,34880,,,,0,d1afb3144899a61f1d1234ac3ec859261f41c1d889cbcaa717eafb1f398bd07f
thread:issue:34881,issue #34881: libtorch memory leak,stale/low urgency,2,stale/low urgency,1,0,34881,,,,0,5863523215753ab6025d9b6dc64bd57d151daf50e2be1ddd8f3569796ae2bfc7
thread:issue:34939,issue #34939: Enable OpaqueTensor to possess Storage then allow it to view from CPUTensor,stale/low urgency,2,stale/low urgency,1,0,34939,,,,0,00f09f02de92dfd369661be94fb9f6dde57d9bc6aca292540dd646cff40d0ae7
thread:issue:34941,issue #34941: Issues with DataParallel on Multiple GPUs,stale/low urgency,2,stale/low urgency,1,0,34941,,,,0,4f0d7da60787c9d4d10df08b170cd3acd5e994d219b3bd970e012fcb29737200
thread:issue:34949,issue #34949: Equality operator for torch.distribution.*,stale/low urgency,2,stale/low urgency,1,0,34949,,,,0,fa5da511d4236c822ff57d1ba6338702ac6e8d193a7d763e5ab5fcef8a4059a4
thread:issue:34951,"issue #34951: [feature request] Sparse (hybrid sparse-dense) output option for topk, min, max",stale/low urgency,2,stale/low urgency,1,0,34951,,,,0,8dbd63469d6641c59f7c7b9c0de91104a56f384bf1c54dc7090b842b9dc88600
thread:issue:34953,issue #34953: Graphic tool to view the backward(Gradient Graph) and forward graph in Pytorch,stale/low urgency,2,stale/low urgency,1,0,34953,,,,0,123a883ad0ccff2df2518458896f38b088e1c7e187ae9a8ca51bdd15e926e852
thread:issue:34977,issue #34977: `conv2d` is slow with specific shapes of channels_last tensors,stale/low urgency,2,stale/low urgency,1,0,34977,,,,0,ffe9c72689fe10f4a39aed39c2a7a057e5d1e0555bb8555007f15a7bdcd2b2c8
thread:issue:34998,issue #34998: Better examples in functional autograd functions,stale/low urgency,2,stale/low urgency,1,0,34998,,,,0,617f0cb6480f2d925e4dc7c28ed8f0a3f307b2eebc62fcc627a5b5222c4f1ec5
thread:issue:35035,issue #35035: How install old version pytorch 1.2.0 from source?,stale/low urgency,2,stale/low urgency,1,0,35035,,,,0,c3c143650ef9ba51723a4671127c5741a523055d2fe1ce6ba4d048fd20e0ed8b
thread:issue:35038,issue #35038: Rename Dispatcher::findSchema to Dispatcher::findOperator,stale/low urgency,2,stale/low urgency,1,0,35038,,,,0,593b691da58de6abd1e6b48dce716abf6f30615993fbda7819f4a9ad2e0f68de
thread:issue:35040,issue #35040: Deprecate and remove RegisterOperators,stale/low urgency,2,stale/low urgency,1,0,35040,,,,0,5329af460a614cc12aad63dd67f97abf9296107ed4a1b79e3a973e44feee73f7
thread:issue:35071,issue #35071: Options for printing the shape with print(tensor),stale/low urgency,2,stale/low urgency,1,0,35071,,,,0,6774de9a82116e085a85b3b02691c39299870a8e263ab71485fc2847cd1317e5
thread:issue:35074,"issue #35074: [JIT] If a python function type comment is referring to a wrong type, JIT frontend gives a not helpful error message",stale/low urgency,2,stale/low urgency,1,0,35074,,,,0,2e4641577715617f04f78c70f826d12617b93a98c4bc71d965554a082b70c524
thread:issue:35078,issue #35078: Generator C++ API should match Python API,stale/low urgency,2,stale/low urgency,1,0,35078,,,,0,5df1001ca052b67fbb0b43861bed4888cc126a0d16554a295b0cc4333fb86b2a
thread:issue:35097,issue #35097: undefined symbols when using libtorch and ITK,stale/low urgency,2,stale/low urgency,1,0,35097,,,,0,b058f6c4c0ad81af3a4375dc5ec23b2ad144abadfc0ee76baed3981e9aaad496
thread:issue:35108,"issue #35108: zip(list, tuple) throws an non-actionable error message",stale/low urgency,2,stale/low urgency,1,0,35108,,,,0,a161790511b99f9b87bfdc8639acac63119af32acf43a13fca94c29704fa3251
thread:issue:35116,issue #35116: JIT does not support class instance attribute type annotation,stale/low urgency,2,stale/low urgency,1,0,35116,,,,0,0185f4aa437a734a5cf4f18b6871b2aac4f26114268478429251eddd2233a200
thread:issue:35118,issue #35118: Customize batch size based on gpu id,stale/low urgency,2,stale/low urgency,1,0,35118,,,,0,00133fc5e8b12f93bf01538324ce1b87f3e85ea8ce2916dd6521a8fdc520ba29
thread:issue:35128,issue #35128: Autograd Engine leaks reentrant threadpool threads on deletion,stale/low urgency,2,stale/low urgency,1,0,35128,,,,0,02a0a2e61bd5b923cf6e84e96dd49e325b207fdd722cf7eac70a56a493a03de0
thread:issue:35160,issue #35160: Documentation doesn't cover MWE using launch.py script,stale/low urgency,2,stale/low urgency,1,0,35160,,,,0,5a05dda22cbe58c8743264c5762ecdd27b6e8e5b939586fd62dc8a261f3df2b0
thread:issue:35180,issue #35180: hasSpecialCase INTERNAL ASSERT FAILED: We don't have an op for aten::to but it isn't a special case.,stale/low urgency,2,stale/low urgency,1,0,35180,,,,0,a91fb8a69347d2c9e96a715c2f7e7f998a090ff2b08c998de716df0bf914dd5e
thread:issue:35208,issue #35208: The BatchNorm error in `DataParallel`,stale/low urgency,2,stale/low urgency,1,0,35208,,,,0,b4aef47ea53b45e1e58f75a4385c3faf7fa1b8d5996f5f2aedc8de9abad3fb48
thread:issue:35234,issue #35234: Raspberry Pi Zero build fails,stale/low urgency,2,stale/low urgency,1,0,35234,,,,0,28b58abf09603d27d5d47c3dd27e9a4e460d7694d5d295fd4a697cf51dfdcb66
thread:issue:35290,issue #35290: Why C++ version libtorch so slow,stale/low urgency,2,stale/low urgency,1,0,35290,,,,0,9a6c8611dc1a1a825b18bd566c7b48404450ce17fab2673ffbc937c0b5f36405
thread:issue:35312,issue #35312: Method/constructor which takes as input angle and magnitude and returns a complex tensor,stale/low urgency,2,stale/low urgency,1,0,35312,,,,0,1ebd44e6cae2c092f15f86af1dd060078264724beefb9a8788af3ecc78cc7a1d
thread:issue:35408,issue #35408: torch.jit.script works for x*sigmoid(x) but not for x*sin(x),stale/low urgency,2,stale/low urgency,1,0,35408,,,,0,e8fdf88cb3ef29b3642c1fbb1cf642758e69bee1461f1e91a21a6a9b40839190
thread:issue:35414,"issue #35414: MacOS Error: subprocess.CalledProcessError: Command '['cmake', '--build', '.', '--target', 'install', '--config', 'Release', '--', '-j', '12']' returned non-zero exit status 1.",stale/low urgency,2,stale/low urgency,1,0,35414,,,,0,d66d0e6b20bbdcdced3b24073989b84fab239ed080fa7fb540376d225aea3c20
thread:issue:35441,"issue #35441: torch::normal only supports (double, double), but at::normal supports (double, double) / (double, Tensor) / (Tensor, double) / (Tensor, Tensor)",stale/low urgency,2,stale/low urgency,1,0,35441,,,,0,d2eac54d3b9856580be60348e095e278a75a69a474b433acb2c0188c63892195
thread:issue:35472,issue #35472: Multiprocessing map gets stuck if doing inference on loaded model,stale/low urgency,2,stale/low urgency,1,0,35472,,,,0,b50eab9708274a49cef19fccb9eb44a8a4cd38ee3223ad13e174c7a9eb5d07be
thread:issue:35499,issue #35499: Support tensor.cumsum() for 1-dim tensors,stale/low urgency,2,stale/low urgency,1,0,35499,,,,0,d753405cb9755fa6c3179a61b04602d0d945d0de586fd1683e4b420fa6868775
thread:issue:35520,issue #35520: TorchScript can't use lists in conditionals,stale/low urgency,2,stale/low urgency,1,0,35520,,,,0,72deea6845af6dd1fbde9f6fcedb0b10f6f9b3804189a299468ae667d725165a
thread:issue:35534,issue #35534: Fail to load torch script model,stale/low urgency,2,stale/low urgency,1,0,35534,,,,0,f2b99f8b20a4544d957287e97ac09543fc0357caae6916607d13de7b8dd264f8
thread:issue:35535,issue #35535: Apple Review Rejected. ITMS-90338: Non-public API usage In Pytorch For iOS,stale/low urgency,2,stale/low urgency,1,0,35535,,,,0,34533ae55d520b127bdd35329fa1a0b18ca5c274d69ad73868c411b4d076e403
thread:issue:35553,issue #35553: caffe2 `DEPTHWISE3x3.Conv` test is broken,stale/low urgency,2,stale/low urgency,1,0,35553,,,,0,1e3172dcba0fe5fa930ab523f5e6e87cd8779b03fbd9cbff36e69f15fe5854d7
thread:issue:35603,issue #35603: Performance bug with convolutions with weights and inputs of similar spatial size,stale/low urgency,2,stale/low urgency,1,0,35603,,,,0,39bf9cccbd4ab4eae2a25defc2ee9958e034d1e7f65f3ffe50880dfc6840ad4a
thread:issue:35609,issue #35609: Randomly error reports,stale/low urgency,2,stale/low urgency,1,0,35609,,,,0,f0024685e971f1842b1bee90f8cff5cc7ea6a1db6400d70f8e91ca117b75c9da
thread:issue:35636,issue #35636: Caffe2 generate_proposals_op_gpu_test crashes on Windows,stale/low urgency,2,stale/low urgency,1,0,35636,,,,0,0b28f545c3a666b6a26ac0a209e25e9695a57c0d6dfa671c968010552cbe75b4
thread:issue:35640,issue #35640: How do you change Adam learning rate since the latest commits ?,stale/low urgency,2,stale/low urgency,1,0,35640,,,,0,cdbf491e511c07e88bfeeb2ed6cb84fb0f8a757d4a23b5f4ca51519d048e7b5d
thread:issue:35648,issue #35648: Caffe2 ReshapeOpGPUTest crashes on Windows,stale/low urgency,2,stale/low urgency,1,0,35648,,,,0,2d412b9ed0d2924679b35d185d84151274b1a508b56160c9831cb8ba17ead0d2
thread:issue:35651,issue #35651: Caffe2 utility_ops_gpu_test fails on Windows,stale/low urgency,2,stale/low urgency,1,0,35651,,,,0,52fca38359890f40059672c0c1b3485229432d2ad081bb52fe1545d5085cb745
thread:issue:35655,issue #35655: Wrong conv2d output on GPU when kernel has many zeros,stale/low urgency,2,stale/low urgency,1,0,35655,,,,0,e00c2bf6f8d40d415bc4de179ba2f0d8b810576b81c316e371ca06ac8c9a6928
thread:issue:35671,issue #35671: allgather_coalesced for tensors of different types seems to be broken,stale/low urgency,2,stale/low urgency,1,0,35671,,,,0,36485691eb7068b56770e35de5643158d3dcaa940417a43e887a5eabafe1fa3c
thread:issue:35681,issue #35681: TensorBoard add_scalars throws error when dict has keys of type int,stale/low urgency,2,stale/low urgency,1,0,35681,,,,0,b60bdd6b387b69e9336ac544a171938b573137ba0e3e2f39b0500f738af588fc
thread:issue:35713,issue #35713: libtorch for Windows. MNIST example does no work.,stale/low urgency,2,stale/low urgency,1,0,35713,,,,0,3642ab438cc34b1eca22d220ded4737b3e0549f208f647756f76380fd2e3516f
thread:issue:35723,issue #35723: LibTorch API on Mobile,stale/low urgency,2,stale/low urgency,1,0,35723,,,,0,453a15813f3666295c414e7ea426cd68eae1a101e7acaf451367d32825eac3d1
thread:issue:35736,"issue #35736: Simple C++ custom autograd function code throws error ""CUDA error: driver shutting down""",stale/low urgency,2,stale/low urgency,1,0,35736,,,,0,1376f79c960d3a5b80d8597def6410a80880a7bc664ae35f75982f5d7c9ed3f0
thread:issue:35755,issue #35755: Expected object of scalar type Double but got scalar type Float for argument #2 'mat2' in call to _th_mm,stale/low urgency,2,stale/low urgency,1,0,35755,,,,0,d2e57f3eb96134ec01c0e9ef705613164c6121373af7b72ad99e157f45f32b3c
thread:issue:35758,issue #35758: UserRRef should store error if it sees any and prevent subsequent usage,stale/low urgency,2,stale/low urgency,1,0,35758,,,,0,11ad25c4f519cfcea360f9add16de3a8869f6c282a69e8282665bd4b13c905e2
thread:issue:35766,issue #35766: Clarify tensor storage communication behavior in RPC,stale/low urgency,2,stale/low urgency,1,0,35766,,,,0,83f4ff30d891ced006295476e5c8307230b401581f099382077436f72fb99c6f
thread:issue:35770,issue #35770: BatchNormFuncOptions object cant be printed in C++,stale/low urgency,2,stale/low urgency,1,0,35770,,,,0,4f0f3dfd6d7696491fe2e0348179dfc30673d3e08232b7e671d5ff620b02a93b
thread:issue:35775,issue #35775: Unexpected behaviour for affine_grid and grid_sample with 3D inputs,stale/low urgency,2,stale/low urgency,1,0,35775,,,,0,c74632e68fb4fa737779d66d7a0014b46542ea6ca8e672ba7de3cfc03987cfef
thread:issue:35788,issue #35788: [jit] Structural typing shouldn't work,stale/low urgency,2,stale/low urgency,1,0,35788,,,,0,e6a40cfe4d4eccf2b658c93d4e4297fdb12c404b483fd08afefa47bdf63d4154
thread:issue:35797,issue #35797: Backward function causes device error in C++ when changing module's device repeatly.,stale/low urgency,2,stale/low urgency,1,0,35797,,,,0,664ac22422c5afa1573e3fb98ef21e99e880f170dca7d74866ab4e4eb77f896d
thread:issue:35798,issue #35798: Dropout on sparse tensors,stale/low urgency,2,stale/low urgency,1,0,35798,,,,0,4901ca0eda97f8c261f79482cab189a18c73340ccf9c1810a746dedebed2dea4
thread:issue:35820,issue #35820: [JIT] Tracer bakes List[Tensor] attribute as constant,stale/low urgency,2,stale/low urgency,1,0,35820,,,,0,4fd561410310f8275f1f787e60016061b604483c5a57689eca635b2a8873b734
thread:issue:35832,issue #35832: Linking error after marking an op `manual_kernel_registration: True`,stale/low urgency,2,stale/low urgency,1,0,35832,,,,0,45fcd6b07aab16a6bf2048f576633eb0879c17340c1c0df1aaa943d7ccb45ba8
thread:issue:35872,issue #35872: torch.autograd.set_detect_anomaly(True) does not exist in C++?,stale/low urgency,2,stale/low urgency,1,0,35872,,,,0,1894f6c9ddfc9691aef9ac29ef6f08bf519d2d18593b160ec5573b0448d6c7f5
thread:issue:35881,issue #35881: Error when loading jit traced FasterRCNN model in C++,stale/low urgency,2,stale/low urgency,1,0,35881,,,,0,a01ce1949816088a10cc7baa0a0992bc5c55aa671ca84b81fadd65a8fcbd381c
thread:issue:35937,issue #35937: MKLDNN_conv2d 2X slower than the native TH implementation,stale/low urgency,2,stale/low urgency,1,0,35937,,,,0,552a4b5e14ed051bfbc74ffaa41835f1c0e5a5d22102b08fbf411972ac4900e9
thread:issue:35942,issue #35942: Build PyTorch-1.4.0 from source failed,stale/low urgency,2,stale/low urgency,1,0,35942,,,,0,23ffc0f4af2d39583be88ec551724f32f5b737ed88d5d010826301ecaf100aca
thread:issue:35944,"issue #35944: Copy activations from one parts to another part in tensor, but report error",stale/low urgency,2,stale/low urgency,1,0,35944,,,,0,a57aff45bdde4a319c588a33fe68dfcf2a5e5050af1d2005199b705ea9e98aad
thread:issue:35964,issue #35964: TensorOptions shouldn't provide default values,stale/low urgency,2,stale/low urgency,1,0,35964,,,,0,6e063ebead87d3ee209ed34c823ed4e3d058aa58874eb2fdb868aff91d517745
thread:issue:3600,issue #3600: making .cuda() falls back to an identity function when gpu is not available,stale/low urgency,2,stale/low urgency,1,0,3600,,,,0,b9cb41a08f912fbf3ca7736feaf9e0dbc1f4d951700ba0de2407ef513dbaedb2
thread:issue:36004,issue #36004: C++ tensor print doesn't show requires_grad and grad_fn like Python tensor print,stale/low urgency,2,stale/low urgency,1,0,36004,,,,0,9055c096366bd3360b4ed6584fc4f229924aae90eae6e6fe3190385bc0660124
thread:issue:36028,issue #36028: About tensorboard in pytorch record graph in different state,stale/low urgency,2,stale/low urgency,1,0,36028,,,,0,e3e17f51f799d886a1e05fb93ab5e50140637f2dcf5cb4b3213ff36c53b1ba72
thread:issue:36036,issue #36036: torch.multinomial is misnamed.,stale/low urgency,2,stale/low urgency,1,0,36036,,,,0,01ee15b8d59bd540945115169b66c81b4d5d4ada38a001b26e43fee9f687bae1
thread:issue:36041,issue #36041: [DISCUSSION] Better user experience for debugging on Windows,stale/low urgency,2,stale/low urgency,1,0,36041,,,,0,3c0f9dd280593e3abbd5cfba2feb597428e90ccfba91de3b7b373ec6acde4cc8
thread:issue:36058,issue #36058: Half type promotion with Numpy arrays is incorrect,stale/low urgency,2,stale/low urgency,1,0,36058,,,,0,58d1bd7ca258a8b55ed7face08570fdd133059b3b138dcf677566351d9b771bd
thread:issue:36059,issue #36059: Some tutorials cannot be found from both side panel and tutorial welcome page,stale/low urgency,2,stale/low urgency,1,0,36059,,,,0,533435365fe0bad9176ec33772223d945080ca6ddab780847a2f5645fbc18465
thread:issue:36062,issue #36062: RuntimeError CUDA error despite CUDA available and GPU supported,stale/low urgency,2,stale/low urgency,1,0,36062,,,,0,cb03a6f67c40cd56feec970bf277deeaf60486471af23c64c173d7a492e10ea0
thread:issue:36066,issue #36066: Add Specific Warning/Error For Unsupported GPU or Systems,stale/low urgency,2,stale/low urgency,1,0,36066,,,,0,d14a12abea2f5c906b19bd51d2999f3446ccfd82bd5482e6d22f570dc74992d5
thread:issue:36091,issue #36091: [quantization] torch.quantized_lstm and torch.quantized_gru not documented,stale/low urgency,2,stale/low urgency,1,0,36091,,,,0,8763919fa6b7b71e18e492adc8e6b378eade2f52f2af535ada5e815e566adf06
thread:issue:36099,"issue #36099: In AutogradContext, get_saved_variables() should be renamed to get_saved_tensors()",stale/low urgency,2,stale/low urgency,1,0,36099,,,,0,c05f71339cb0336a5cc86ba18c4f49853775dae888e6cd801f4e2bc0b37fcffd
thread:issue:36125,issue #36125: build Pytorch-1.5.0-rc1 from source fail,stale/low urgency,2,stale/low urgency,1,0,36125,,,,0,308b9b61acf79f51c09663d1a6c7b252eb6c7a68d343233938f42535cddd188b
thread:issue:36140,issue #36140: Inconsistent ProcessGroupMPI work data structure for send/recv and collectives,stale/low urgency,2,stale/low urgency,1,0,36140,,,,0,ca9fac9d9274b090cd7f707dd67e8123be4060902a05650f7f7d9164e08ea85a
thread:issue:36160,issue #36160: Decouple DDP from CUDA,stale/low urgency,2,stale/low urgency,1,0,36160,,,,0,f1daf9f565162e409e2ec514ee1af70619a9a0dfa5b088e84ee177729f6c6f2b
thread:issue:36175,issue #36175: New dtype ComplexPolarFloat (phasor),stale/low urgency,2,stale/low urgency,1,0,36175,,,,0,2183bb0628a05bcfe085df86031b8b5fc5d330d5c490e5708941eec94af6624f
thread:issue:36217,issue #36217: Rectify docs for MultiLabelSoftMarginLoss,stale/low urgency,2,stale/low urgency,1,0,36217,,,,0,e42c28b089cb17a80c1643eb9b28658415abf4d6178d0e0b36cf553f017d380d
thread:issue:3625,issue #3625: Proposal: combine requires_grad and retain_grad(),stale/low urgency,2,stale/low urgency,1,0,3625,,,,0,986746de455c93488c96fe66173d13be32e1b76c5314f4ae8b07257230c69eec
thread:issue:36307,issue #36307: [docs] Matrix transpose missing in formula in nn.functional.bilinear docs,stale/low urgency,2,stale/low urgency,1,0,36307,,,,0,211fb7e2e1aae21b0afc2c700f772b50c777536494bf9900def7e7b107d5344c
thread:issue:36309,"issue #36309: torch.ceil, torch.floor should accept a dtype argument",stale/low urgency,2,stale/low urgency,1,0,36309,,,,0,bb860bf05bda5e0b89f55b2557caf5d98f3304319dc23676f7fe5c4adf10d2a4
thread:issue:36314,issue #36314: [discussion] Refactor spectral_norm to use the newly merged lowrank solvers and proposal for Linear Algebra Cookbook page,stale/low urgency,2,stale/low urgency,1,0,36314,,,,0,d274551b7592174784dce96754ab599a0e6d8fd4dfbaafcf2050f1033dc8a95b
thread:issue:36317,issue #36317: Updating a ModelDict instance with another ModelDict instance generates error.,stale/low urgency,2,stale/low urgency,1,0,36317,,,,0,2cc6b76a30165dd9b7072524f4580df4a101456f631eb785a0c71fb78baf2dd6
thread:issue:36356,issue #36356: Memory leak issue still exists in CI,stale/low urgency,2,stale/low urgency,1,0,36356,,,,0,cca05232a4cf005e6ebe1a057cd06fb655cbf8e1ad632f456fb35ecf8eb66f92
thread:issue:36370,issue #36370: Jit doesn't match schema like eager mode does,stale/low urgency,2,stale/low urgency,1,0,36370,,,,0,42770f3c766d490018450c017de343d26038ea65037ee643068d3c458c43920f
thread:issue:36378,issue #36378: CMake targets wrongly forward unknown options to NVCC (v1.5+),stale/low urgency,2,stale/low urgency,1,0,36378,,,,0,c9a487a6353a85142eafbc28905f5ea5d75d4926c170e58c6730ee7cee9fe5da
thread:issue:36380,issue #36380: Population Count Op,stale/low urgency,2,stale/low urgency,1,0,36380,,,,0,70c08fbae438348e78a48faeaf6cfa83f58d25432e60f148524ab3f8f9a3a50d
thread:issue:36386,issue #36386: Docs of distributed module do not include the full documentation for torch.distributed.launch,stale/low urgency,2,stale/low urgency,1,0,36386,,,,0,cf91fe22638c267a0766eabc3771382a35caa1580af736960310005b10d05dae
thread:issue:36407,issue #36407: Unable to build wheel from RC3,stale/low urgency,2,stale/low urgency,1,0,36407,,,,0,35674048efce5ae015e93aacf63a143ff72f02e1272dd355e9c575f0edaf3198
thread:issue:36426,issue #36426: Any reference to LPPool2d,stale/low urgency,2,stale/low urgency,1,0,36426,,,,0,871aecc4ea058c975218d1459c2479ec796569e667f5c1612e031862d2a8c36f
thread:issue:36436,"issue #36436: After `create_graph=True`, calculating `backward()` on sparse Tensor fails",stale/low urgency,2,stale/low urgency,1,0,36436,,,,0,b861109281f2d540be48ac58a49e4a7a111141d08658502cf51f714530fd2133
thread:issue:36437,issue #36437: Issue when linking C++ code with libtorch_cpu: cuda not detected,stale/low urgency,2,stale/low urgency,1,0,36437,,,,0,075daca45bc757e1197c4a2b57e1724c339fc5220fd51cb7bd7bb6c4c94b1788
thread:issue:36469,issue #36469: Wrong results for multiplication of non-finite complex numbers with real numbers,stale/low urgency,2,stale/low urgency,1,0,36469,,,,0,2d9e4d3e4be1978814882d269aa7c80caa59cc0809ec84503e8a1435e5e1ecc4
thread:issue:36508,issue #36508: Quantized _out functions don't follow same conventions as other out functions in the codebase,stale/low urgency,2,stale/low urgency,1,0,36508,,,,0,44a0b700ab9f762d732b489b06114c80668b45c1ac0330677432c2324ccf1d00
thread:issue:36516,issue #36516: DDP should divide bucket contents by the number of global replicas instead of world size,stale/low urgency,2,stale/low urgency,1,0,36516,,,,0,510a47b3f64e6c4598f30345fcd31ba3729ef3c57b3260b482e2738bae608a9d
thread:issue:36517,issue #36517: XNNPACK operators are not actually registered under xnnpack namespace,stale/low urgency,2,stale/low urgency,1,0,36517,,,,0,674c37d1957c50645126e934d24810823f2e8984929b0d17ecd08e42ef0b5a4c
thread:issue:36519,issue #36519: Restructure test_c10d.py and test_distributed.py,stale/low urgency,2,stale/low urgency,1,0,36519,,,,0,3064791dd25edfacd5c41262ce831939261c399cbf4ba1ffb375792f1d68c36b
thread:issue:36553,issue #36553: Typecasting issue in MSELoss,stale/low urgency,2,stale/low urgency,1,0,36553,,,,0,5306778b52f33405b4e7fa302a9ec284a172604a0bb83f77a4747006f297177c
thread:issue:36560,issue #36560: Allow `__array__` to automatically detach and move to CPU,stale/low urgency,2,stale/low urgency,1,0,36560,,,,0,4bfc4860d89dc4becc7afb16024d82e786df376a8290d8822607575387a592b6
thread:issue:36570,issue #36570: Libtorch build error when setting both `USE_GLOO` and `USE_SYSTEM_NCCL` to `ON`,stale/low urgency,2,stale/low urgency,1,0,36570,,,,0,5d9f9a1f4bd0c88d5407162995c252758909207ecbab7ec5952fbf6afc675c29
thread:issue:36572,issue #36572: [1.4.1] cmake3 not found,stale/low urgency,2,stale/low urgency,1,0,36572,,,,0,65572d1b376344bf60d9f955c07ed9c1120b7d25259fa722de4046ef939601b7
thread:issue:36575,issue #36575: One confusion about the CompilationUnit destructuring process in torch/jit/__init__.py,stale/low urgency,2,stale/low urgency,1,0,36575,,,,0,d763fa9275215900674635a922cb8d58d971c30751156bb9e4944850f3035802
thread:issue:36638,issue #36638: Poor elmenetwise_kernel performance becomes critical on small mini-batch sizes,stale/low urgency,2,stale/low urgency,1,0,36638,,,,0,70769df3b42c86a8d0475c0849904756a990f1575dc22d0d9161e25c31ec0ab2
thread:issue:36706,issue #36706: Reduction for `torch.int8` is super slow on CUDA,stale/low urgency,2,stale/low urgency,1,0,36706,,,,0,115c744b167f8ce8f2c09c868ef5d4480096b5e29bb0ea8adc1d044bec748d18
thread:issue:36713,issue #36713: variable name N_ conflicts with an internationalization macro in glib,stale/low urgency,2,stale/low urgency,1,0,36713,,,,0,7e645a6e8c2766696903a33c5cd05b8521050a0e2466404ec7efe20dde03d177
thread:issue:36733,issue #36733: Custom Generators don't work in JIT,stale/low urgency,2,stale/low urgency,1,0,36733,,,,0,75576c9ded528a198da59a7ccaaa3131866843e3438942448272ea42cc46a76d
thread:issue:36739,issue #36739: Custom class type name is very wordy,stale/low urgency,2,stale/low urgency,1,0,36739,,,,0,a3a5e15a9ffb7efd52929c5d33c87a5f8ec9fab9f66231a0a24e789bdd573f34
thread:issue:36748,issue #36748: return_index option for torch.unique,stale/low urgency,2,stale/low urgency,1,0,36748,,,,0,2b2af04d81fd7fe2e735f6b653d0e70e2ba38e9e2a5cc861d1eabd3f06e88c0c
thread:issue:36753,issue #36753: When you try to register a kernel that make boxed from unboxed functor doesn't support you get a horrible error message,stale/low urgency,2,stale/low urgency,1,0,36753,,,,0,750b4db658d0bf63548a40c9a22090955193c8ecaa0cf0bf7a77832b65008645
thread:issue:36754,issue #36754: Clearer guidance on when to define an operator as a method on torchbind'ed class versus standalone function,stale/low urgency,2,stale/low urgency,1,0,36754,,,,0,b06abd9db87741e4da168505a6b222386d37e9cd3d92fa48ff53c679374d8508
thread:issue:36792,issue #36792: [1.4.1] Intel MPI libs not found.,stale/low urgency,2,stale/low urgency,1,0,36792,,,,0,d91e09e5b6ffb0660dcfe86adf1dd2ffb32212f5c98d67e348c4dffab0a68a5d
thread:issue:36804,issue #36804: c++: error: unrecognized command line option '-Wthread-safety',stale/low urgency,2,stale/low urgency,1,0,36804,,,,0,13d76f88f4009c5af6be5247e171350d732fdc43522260986c54f640bed978d7
thread:issue:36874,issue #36874: Memory leak upon exception for interactive consoles,stale/low urgency,2,stale/low urgency,1,0,36874,,,,0,3edee963a91f80432878b96bf12e39d2be5ade4ade0c02ff6b89916c656f3bc5
thread:issue:36891,issue #36891: pytorch and c++ inference disagree,stale/low urgency,2,stale/low urgency,1,0,36891,,,,0,8a0b2f178ee1c979633f8691bb344949b662c16bc9b0e6e1212a26e0796b5810
thread:issue:36914,issue #36914: Problem with c10/utils/variant.h,stale/low urgency,2,stale/low urgency,1,0,36914,,,,0,2435130c217bb7321110f2c26ec8a8d5509afdd427ebf11df84aad3f5851ba70
thread:issue:36917,"issue #36917: jit trace failed due to ""failed to differentiate `prim::ListConstruct`""",stale/low urgency,2,stale/low urgency,1,0,36917,,,,0,4b810d953abfc0fc5fc4ba2eac1424ad9d59af92db07121ba6cf2b1423a6bf0b
thread:issue:36922,issue #36922: Multiprocess DataLoader with DLPack conversion sometimes corrupts memory,stale/low urgency,2,stale/low urgency,1,0,36922,,,,0,ff28aa55ecb47920b0577df99cedb5766df1654eb9966345366d2aeee0f0a525
thread:issue:36931,issue #36931: Pull hacked twins out of prim ops,stale/low urgency,2,stale/low urgency,1,0,36931,,,,0,d14550e0b7267bd9c219f68f3fe17e1165fdc82eadfc960791ac3dd471b9f1cc
thread:issue:3697,issue #3697: Deprecate inplace argument in torch.nn.functional,stale/low urgency,2,stale/low urgency,1,0,3697,,,,0,39ae5a9dcc245fed2d318ef25193836bab0f556d0e1241f25ca56bc3acdf0f85
thread:issue:36989,issue #36989: pytorch latest update(1.4) broke CosineAnnealingWarmRestarts: T_cur is not define,stale/low urgency,2,stale/low urgency,1,0,36989,,,,0,63b23e7fe38ddcd3585d0e92e52fec61185afd63131c1677a36c6969a80cb63c
thread:issue:37004,"issue #37004: RuntimeError: NCCL error in ProcessGroupNCCL.cpp:290, unhandled system error",stale/low urgency,2,stale/low urgency,1,0,37004,,,,0,14b6990a946412369446fa50955d3523ceb515d50a6c2372c64ca4c8ff2cbbe1
thread:issue:37010,issue #37010: Make it an error to def() an operator multiple times,stale/low urgency,2,stale/low urgency,1,0,37010,,,,0,3e5b60168b598b5da4bd7fc9ba59badc3cd49ab56010cfbb7a187be042172d6a
thread:issue:37048,issue #37048: copy_ slowness,stale/low urgency,2,stale/low urgency,1,0,37048,,,,0,eb9cdb8df25461f68852876c08af970958335b1aa00df17013961baee67ac6e9
thread:issue:37063,issue #37063: Misleading documentation in torch.nn.functional.fold,stale/low urgency,2,stale/low urgency,1,0,37063,,,,0,042d467961316be3cebc7c1066bb7b730103502cfdf2741c8f732c67e0dc631a
thread:issue:37065,issue #37065: Deprecated mask fill mask type can causes pages and pages of repeated messages,stale/low urgency,2,stale/low urgency,1,0,37065,,,,0,12b5b9012d1a65ab73b0bd7acb8cf98188b7ec9b7f7aee532001122ba7fa0162
thread:issue:37079,issue #37079: Result parameters from Single-Process Multi-GPU DDP training on RNN do not match local training,stale/low urgency,2,stale/low urgency,1,0,37079,,,,0,3b70a4f4e88b2e7bcaa774795bdca6bed390977a15caf3e335e2b328fa60adfa
thread:issue:37092,issue #37092: nn.Bilinear cannot be used inside nn.Sequential,stale/low urgency,2,stale/low urgency,1,0,37092,,,,0,afbed4539318a67800b373cd16235d70c1955c172a461af5f114d546b0309400
thread:issue:37103,issue #37103: Legacy fuser doesn't do remainder consistently with `aten::remainder`,stale/low urgency,2,stale/low urgency,1,0,37103,,,,0,296a47aa36f59b6dd0d70b3d9535cce717d0aa3d79a5179e305a35732110d117
thread:issue:37104,issue #37104: [JIT] Use of global value creates confusing error message,stale/low urgency,2,stale/low urgency,1,0,37104,,,,0,9e8b13bfee93eb40ced8714dee33962c6555a249640f95658705d1aaf4fbb20e
thread:issue:37113,issue #37113: pip install torch==1.4.0 is broken when using CUDA 10.1,stale/low urgency,2,stale/low urgency,1,0,37113,,,,0,0fabe34a75624424a8f3cf20f83e80c400d3a8feb61cde5c5aabbe546cff5cb7
thread:issue:37136,issue #37136: [Design][RFC] RemoteModule API Design,stale/low urgency,2,stale/low urgency,1,0,37136,,,,0,4b18252847c7471079c2c7aa80e280ae7e7233b4942e417472656f228960e3cb
thread:issue:37142,issue #37142: Propagation of channels-last layout leads to massive slowdowns in 1.5 compared to 1.4,stale/low urgency,2,stale/low urgency,1,0,37142,,,,0,d24f61855d1acb963a624f63b762939504d7bee4baaf9082a3335df967727669
thread:issue:37149,issue #37149: Cuda profiler + DataParallel + manual profiling start = strange profiling overhead pattern,stale/low urgency,2,stale/low urgency,1,0,37149,,,,0,85dea543ba333ec9b3bf97f9053e06d0550597ddd7764a48b26c776beb43f3ae
thread:issue:37153,issue #37153: CPU out of bound memory access in CUDA reduction kernel config,stale/low urgency,2,stale/low urgency,1,0,37153,,,,0,ce1fa961e58038df4a13723861994254397a1830ebb549db4eb4e36ecbc97b63
thread:issue:37162,issue #37162: torch.distributions bug in RelaxedOneHotCategorical.log_prob,stale/low urgency,2,stale/low urgency,1,0,37162,,,,0,a1fc04bc0521f215044632e6e827e4053b77c29d7dff4c8c988613d9e1b02beb
thread:issue:37204,issue #37204: Make it harder to make SIOF bugs for torchbind classes referenced by schemas,stale/low urgency,2,stale/low urgency,1,0,37204,,,,0,bcabba4e6dd199c88ada6607d601909248d8eaa13d9cdb8ba3c5dc844ad2ab00
thread:issue:37246,issue #37246: `max_norm` parameter on nn.Embedding will fail inside nn.DataParallel,stale/low urgency,2,stale/low urgency,1,0,37246,,,,0,f15a45a0bc55206e366105bfa01ff415ee11369ccb8789c93517e767d67ba05c
thread:issue:37282,issue #37282: python setup.py install error,stale/low urgency,2,stale/low urgency,1,0,37282,,,,0,3898651df0ede43dd533417ca975d5a84a8121ef8dce6a383524916da1992b33
thread:issue:37300,issue #37300: Data caching module a la `Sampler`,stale/low urgency,2,stale/low urgency,1,0,37300,,,,0,576fe518e6b14142f1c68dbac7cd725db77aec45ec41cd20e71c32fca4f563de
thread:issue:37314,issue #37314: equal_nan keyword not implemented for complex torch.isclose,stale/low urgency,2,stale/low urgency,1,0,37314,,,,0,951fb604a8c0e036093b344e94256aa1126ad2941638cba7af546afcd2bb6d73
thread:issue:37324,issue #37324: Multiprocessing: model shared between processes hangs during copy.deepcopy,stale/low urgency,2,stale/low urgency,1,0,37324,,,,0,8d7fa016dad40abf394fa9b1b82017cb95e1a14edda4f782f396ca18cdbd9ec8
thread:issue:37332,issue #37332: Concerning default confiugration for distribution packages,stale/low urgency,2,stale/low urgency,1,0,37332,,,,0,e2f680891a2804720abd2f427de5f896acce49c71d1100320ffeb6ce68fd09af
thread:issue:37343,issue #37343: qnnpack's quantized-add gives wrong result,stale/low urgency,2,stale/low urgency,1,0,37343,,,,0,98fe3750010a6b7d694f5e54b2a9ebcc3b362983f8a27303bcf2f692c7c0859a
thread:issue:37354,issue #37354: Efficient handling special gradient values in the autograd,stale/low urgency,2,stale/low urgency,1,0,37354,,,,0,fe9b02aa288edca66c8e979811ed2a5436553438cee8bf052f83f1337f017f35
thread:issue:37363,"issue #37363: cdist docs imply only one batch dimension is possible, but actually arbitrarily many are allowed",stale/low urgency,2,stale/low urgency,1,0,37363,,,,0,90136868255d656d09e4bd7a5588d11b04b8a2130272225590e932917fe8b039
thread:issue:37373,issue #37373: Move bernoulli_() to DistributionTemplates,stale/low urgency,2,stale/low urgency,1,0,37373,,,,0,a1364418f6b549138dbb646f2f4097d791fc1fa4032f1943e05ef23ff3581902
thread:issue:37384,issue #37384: Error running trace on Pytorch Crowd Counting model,stale/low urgency,2,stale/low urgency,1,0,37384,,,,0,c878d039ed11a6296e69c40d449f2d7167440ece7c58b6983a70436cf886a04c
thread:issue:37386,issue #37386: Add BufferDict container,stale/low urgency,2,stale/low urgency,1,0,37386,,,,0,42d650261b1a61346a8baef9446dfc284968e3a717fb525dbbe2272a62e7019d
thread:issue:37406,issue #37406: 3D grouped & depthwise convolution very slow on backward pass,stale/low urgency,2,stale/low urgency,1,0,37406,,,,0,3830a46296c096cfc1da66b7910162d6d5d99a19676cf1a4cb810dbda9e584cd
thread:issue:37408,issue #37408: torch.cdist() implementation without using contiguous() calls,stale/low urgency,2,stale/low urgency,1,0,37408,,,,0,20393eba6149a688e1d6d82d79a35e424037659b7b194ce5389c37ba741c252a
thread:issue:37409,issue #37409: Compatibility of subset dataset with disabled batch sampling,stale/low urgency,2,stale/low urgency,1,0,37409,,,,0,82969b7a9e21bd0be9286f1ae7e1af0f22107c1e251cc68c96c7ed16b41041ce
thread:issue:37416,issue #37416: [docs] Unclear return type of torch.randint and extra comma in arg spec,stale/low urgency,2,stale/low urgency,1,0,37416,,,,0,c6fef8a2bdb61221d6200940f9e566ea43f9ea9648ef2a0ab3c5d5ee20fbf0ae
thread:issue:37434,issue #37434: rsub incorrectly exposed in torch,stale/low urgency,2,stale/low urgency,1,0,37434,,,,0,7a88147aa0bf6826291ae6bfcd1e3c3aef532996958f9bff63ab0491fc49ae05
thread:issue:37440,issue #37440: Tensor.is_same_size not documented,stale/low urgency,2,stale/low urgency,1,0,37440,,,,0,a67ea8c9fd7f8df6f51987e9efaa800a8f944ec384e100cb69647e94c208448f
thread:issue:37441,issue #37441: Tensor.as_strided_ is not documented,stale/low urgency,2,stale/low urgency,1,0,37441,,,,0,ef200ffa913ad73c6d9da8a3166ecef6286ac65dd198e9e7287b0057423c23f5
thread:issue:37443,issue #37443: torch.clamp_max clamp_min shouldn't be there,stale/low urgency,2,stale/low urgency,1,0,37443,,,,0,4f597cbf01bee2734fd54903181075ff8c3b7bb68fcc15955377c32eaa89f80a
thread:issue:37444,issue #37444: Multi-Process Single-GPU is bad,stale/low urgency,2,stale/low urgency,1,0,37444,,,,0,69c0aa9637278ed1edbdc45509a57286f04a60477dcb9408f63bb5d85474a6f7
thread:issue:37457,issue #37457: [RuntimeError] Tensor creation using storage fails,stale/low urgency,2,stale/low urgency,1,0,37457,,,,0,ca571f59fea06accaa0612987dd800bbd5acacd46d909976cce327cd182c1b92
thread:issue:37484,issue #37484: No speedup from channels_last with DataParallel,stale/low urgency,2,stale/low urgency,1,0,37484,,,,0,dd9967ac46ebbe151337a0716d32e62de2eb1e7db4aa670630af96884016a9ac
thread:issue:37487,issue #37487: Log-linear version of cumsum and cumprod,stale/low urgency,2,stale/low urgency,1,0,37487,,,,0,13820aa3aed0bc9103c1866356a9330f6160c4eb44b5f1ad813b3db436f99f88
thread:issue:37510,"issue #37510: Cannot build pytorch with linker arguments in C{,XX}FLAGS",stale/low urgency,2,stale/low urgency,1,0,37510,,,,0,a9295c484ebab0524eb4cfebcf439e7540405ddfafab7b22fc7ba26f058af3c7
thread:issue:37512,issue #37512: Improve visibility in test suite timings,stale/low urgency,2,stale/low urgency,1,0,37512,,,,0,5bbd771a31e2e5be886673957fdc6ab0c8f9d995e8abd1a451f0e3b7edecd758
thread:issue:37549,issue #37549: Are there any differences in kernel memory between RX2080's and Quadro RTX4000?,stale/low urgency,2,stale/low urgency,1,0,37549,,,,0,d05887d4d2162d75f345fe679a47a04a71e8dec1bf96e22753bb1820e9578a19
thread:issue:37554,issue #37554: Add name to Class Parameter(),stale/low urgency,2,stale/low urgency,1,0,37554,,,,0,6848a610d2fc4d6ec2804653b7db3596e4a40fdb5782b5730392910ce89288f2
thread:issue:37556,"issue #37556: torch.cartesian_prod(*tensors) error when you have tensors with [x,y]",stale/low urgency,2,stale/low urgency,1,0,37556,,,,0,8de0364c11d9c4f7d01ae265db3fa59dfd8ab6ba4d3b9f155c2bbedada42ce2b
thread:issue:37585,issue #37585: Unable to use torch.det() inside nn.DataParallel with multiple gpus,stale/low urgency,2,stale/low urgency,1,0,37585,,,,0,db280e9631c30248339c97a99f04ec7c78120f374236806781062041e9ff868f
thread:issue:37588,"issue #37588: test_qnnpack_sigmoid failing on old CPU, built pytorch 1.5.0",stale/low urgency,2,stale/low urgency,1,0,37588,,,,0,3a61e8615e83c29eaa8948247a3d39c2a35403fccab2b0ebd9ff7de950e30e9d
thread:issue:37598,issue #37598: NCCL fails to find cuda include dir,stale/low urgency,2,stale/low urgency,1,0,37598,,,,0,c2a68f94dd877ef7b53990f5205b7c995f9c17c92f449dbf7f434165647cad4e
thread:issue:37629,issue #37629: logging_is_not_google_glog.h:24:11: error: 'const int ERROR' redeclared as different kind of symbol (v1.3.0),stale/low urgency,2,stale/low urgency,1,0,37629,,,,0,4b57caa2a32df8adfade25c6637e5500f342ebd7e4fe9eb11e3db237e0b658b4
thread:issue:37664,issue #37664: Remove everything from a GPU (including drivers),stale/low urgency,2,stale/low urgency,1,0,37664,,,,0,f2fe3daec82919001e56d34f56888563b134c9a96c9504079c7c474c4e5c58b3
thread:issue:37682,issue #37682: LSTMCell consumes x1.5 more memory on CUDA on pytorch >=1.3 comparing to pytorch 1.2,stale/low urgency,2,stale/low urgency,1,0,37682,,,,0,fef71958c52fd791172a750712872a3784c9ec4b9501caeb532d56816590b700
thread:issue:37684,"issue #37684: Add Compound key, make custom ops default to it (but keep internal users using CatchAll)",stale/low urgency,2,stale/low urgency,1,0,37684,,,,0,6b283a6c96ccf6a8dd2f51dd083fdc77afd168719e8c647e32893b819547e9de
thread:issue:37713,issue #37713: Builtin FusedLayerNorm is slower than apex one,stale/low urgency,2,stale/low urgency,1,0,37713,,,,0,ecd5b1467b8f82202009e2a19d63581a3663b3969e367286509e6ccc6874f943
thread:issue:37731,issue #37731: Variational Dropout In RNN,stale/low urgency,2,stale/low urgency,1,0,37731,,,,0,cc72b40fc385f97af77439f3323221677bbaf7e8c89578e9ae438af55820e7ca
thread:issue:37734,issue #37734: torch.cdist returns inconsistent result,stale/low urgency,2,stale/low urgency,1,0,37734,,,,0,96c331165e2516d0e0947e7f080838c0cf272206caa73b1ea0484068666c99f8
thread:issue:37751,issue #37751: Documentation of _CtxMethodMixin: must be tensors?,stale/low urgency,2,stale/low urgency,1,0,37751,,,,0,555aa227928d6f66535bbb68465b6dc75c69bd351c3921ccb9f4453a350f322a
thread:issue:37782,issue #37782: Guard Gloo and TensorPipe related code in RPC with #ifdef,stale/low urgency,2,stale/low urgency,1,0,37782,,,,0,1d99fcff9a60a12f3e3e99c8304e668d37cd1500b2857a9eba70a02bd4df4560
thread:issue:37809,"issue #37809: Add Numpy-like ""order"" argument to reshape",stale/low urgency,2,stale/low urgency,1,0,37809,,,,0,411e3b8a16cdebb125f66f6f9efb92337bcc3f7dcb27ff05ec391ecf4da1d54b
thread:issue:37814,issue #37814: DistributedDataParallel does not support Modules that take no inputs.,stale/low urgency,2,stale/low urgency,1,0,37814,,,,0,6f43cc3fbc45806ba843e453c99ce597815773df6a583d7a7cc4a9b0d63ee49b
thread:issue:37847,issue #37847: llvmlite version issue when upgrading CI docker image to python 3.8,stale/low urgency,2,stale/low urgency,1,0,37847,,,,0,5345a3aef9fcd671510c8191686c6dc0112f8089d1018e1dd05c3d747216f846
thread:issue:37860,issue #37860: autograd engine callbacks don't respect non-default cuda streams,stale/low urgency,2,stale/low urgency,1,0,37860,,,,0,866fc7fe5436d9c7debaf15a97db9d342358b11bedb95cac9ded47a4c2f4b864
thread:issue:37863,issue #37863: Deprecate type() and type_as() call,stale/low urgency,2,stale/low urgency,1,0,37863,,,,0,6b57e6a9a79077de55f8f183ca3b92404fd3ca501cf04eec90c2efbf1ab5d8c7
thread:issue:37880,"issue #37880: [RFC] Don't install CI dependencies in build scripts, install them in underlying docker images",stale/low urgency,2,stale/low urgency,1,0,37880,,,,0,61d3df41c3206fead22a897d2df4605c11d001e9f7c76f4d41b5c2c28ffa193f
thread:issue:37895,issue #37895: torch.addmv can't take as input tensors with different dtypes,stale/low urgency,2,stale/low urgency,1,0,37895,,,,0,668a4c376e740c9180c4f602ca67a47a2ef15df608f8334316db0a6bf9d97dee
thread:issue:3791,issue #3791: Make pytest stop printing docstrings in its default diagnostic output,stale/low urgency,2,stale/low urgency,1,0,3791,,,,0,b0d99aa24f8662f24b6eaaefb2d30fa9250914b219eb6542282d58fa5dbe08cf
thread:issue:37985,issue #37985: torch.save incompatible with lzma file,stale/low urgency,2,stale/low urgency,1,0,37985,,,,0,c98ae7b63c348ded9af39af7fb43e26a833de948a188a85b1aef48a6e334aae0
thread:issue:37987,issue #37987: Special methods on torchscript custom class,stale/low urgency,2,stale/low urgency,1,0,37987,,,,0,32e6272de93337a0e89a3fac8863c102afafd36f698fc7f785be6e584e83c9da
thread:issue:37988,issue #37988: JIT compilation of NamedTuple Containing a NamedTuple fails,stale/low urgency,2,stale/low urgency,1,0,37988,,,,0,bae106338614f92b3062b5913cac834ffdc1d59557661b84267e3b3cb96c8b0c
thread:issue:37995,issue #37995: unsqueeze support for named tensors,stale/low urgency,2,stale/low urgency,1,0,37995,,,,0,07668ebdde658004b2bd6b5c49e955715285ef4c488bca70743516459cc9e47d
thread:issue:38000,"issue #38000: New padding size format for F.pad, allowing named tensors and more clear syntax overall",stale/low urgency,2,stale/low urgency,1,0,38000,,,,0,9d29f1f8ffb663d35c962cc0070edb5ca8acc47bc18bf5d8d10429935de97241
thread:issue:38009,issue #38009: Resnet Model always predicting same label,stale/low urgency,2,stale/low urgency,1,0,38009,,,,0,5c793b138d7a48ebc48583333051f78cdf93dc13d751ad52bed63a13ab9c18ba
thread:issue:38012,issue #38012: Formatting of topic and sub-topic pages,stale/low urgency,2,stale/low urgency,1,0,38012,,,,0,87a89de93b670a17f9406ccb36adcb9071170da7a7dc07d0ac9afc4558b61889
thread:issue:38051,issue #38051: Div by zero error not triggered and inf not returned when dividing by 0 for some dtypes,stale/low urgency,2,stale/low urgency,1,0,38051,,,,0,6f0b38dd60be8ff3308697bb5786f2a0d562982fc05f34e505ff642f8cfbb0ed
thread:issue:38057,issue #38057: DNNL's backward pass much slower when using nn.grad.conv2d_input and nn.grad.conv2d_weight,stale/low urgency,2,stale/low urgency,1,0,38057,,,,0,e3523006be39e3420be7f1de26cda1e4947f2e4169d0b227bf4b198703ef8f41
thread:issue:38090,issue #38090: ImportError: libtorch_cpu.so: cannot open shared object file: No such file or directory,stale/low urgency,2,stale/low urgency,1,0,38090,,,,0,eb56c30288d9d2e8db83e388f5d3adbb901528170b03f691cd1073f7a65adcf7
thread:issue:38118,issue #38118: Add helpers to save/load RPC-based models,stale/low urgency,2,stale/low urgency,1,0,38118,,,,0,a3854dde51178811759047008bf3936e6b63d7246b2f007b11cbb4434d46bfc1
thread:issue:38138,"issue #38138: TorchScript to support == None, != None",stale/low urgency,2,stale/low urgency,1,0,38138,,,,0,7eebae4a3b8b383efab163b9a191ddc658489197b408b0b3eb9cb68161e7af15
thread:issue:38142,issue #38142: [JIT] Aliased Python references to script::Object escape IValue reference counting,stale/low urgency,2,stale/low urgency,1,0,38142,,,,0,df803699b999b7cab9ef61babe4f1cd4199a22a0bf7d6f6552a8b1819bfd04a2
thread:issue:38170,"issue #38170: OneCycleLR raises ""Tried to step { step_num + 1 } times"" after the value is more than expected.",stale/low urgency,2,stale/low urgency,1,0,38170,,,,0,03ecf087d622a1a434d8c3e115b73cf4f08d8d9ec549bf737456897baed6020b
thread:issue:38177,issue #38177: valgrind says libtorch has memory leak,stale/low urgency,2,stale/low urgency,1,0,38177,,,,0,c52c4e84bf5b3748df5f93b2c050d0edac7355c07e735d8d1f2e536fa15b8796
thread:issue:3818,issue #3818: Memory leak when doing backward with grad as yourself,stale/low urgency,2,stale/low urgency,1,0,3818,,,,0,4a20dbe93d9471b02b3f0c836725fe2f6955dcf2e283609178fc47e5f660ab1b
thread:issue:38184,issue #38184: Feature Request: Log uniform candidate sampler,stale/low urgency,2,stale/low urgency,1,0,38184,,,,0,cf1afc9c0af560fff55793e102abe078ea22b850b5c6a2d5d06a0b9338609d73
thread:issue:38188,issue #38188: test_interpolate_nearest_scale_3d in test_nn takes too long to run,stale/low urgency,2,stale/low urgency,1,0,38188,,,,0,6dff040281fdaa1f6cfc147065d45e2aac96672aae989aa1dc22658892dbe89e
thread:issue:38189,issue #38189: test_LocalResponseNorm_3d_custom_params (test_nn.TestNN) takes too long to run,stale/low urgency,2,stale/low urgency,1,0,38189,,,,0,a1bdd6470d7f2f33cc6efb97b6b45210bf95e65a0c460a05e1445c2f8bd30bb7
thread:issue:38190,issue #38190: test_ReplicationPad3d (test_nn.TestNN) takes too long to run,stale/low urgency,2,stale/low urgency,1,0,38190,,,,0,8e882c35e56e91999d0781eaeab9775f3bb415420b4ee19b187e7eb5fca18dff
thread:issue:38193,issue #38193: Can't Build Pytorch0.4.1 From Source,stale/low urgency,2,stale/low urgency,1,0,38193,,,,0,0bfccc4c17f1f4e7b75c00a6e24ec0026e260219c9662dc5189ea8b4435bdf4b
thread:issue:38197,issue #38197: Why no `torch.randperm_like`?,stale/low urgency,2,stale/low urgency,1,0,38197,,,,0,0a8b6d4ed69479248096ec18612b3e89f2b26a95cb71c38bb06ea84bcd487f6d
thread:issue:38202,issue #38202: IndexError reports the wrong dimension when fancy indexing,stale/low urgency,2,stale/low urgency,1,0,38202,,,,0,f5dbaa4bc9370e5b5a67ffb90b438e81e0630377863d56bfb92e39ff899e144d
thread:issue:38204,issue #38204: missing dependency `protobuf` for caffe2,stale/low urgency,2,stale/low urgency,1,0,38204,,,,0,e9588162b0bb1616208a99e7680f3109d71953f2977cf7dc63cf64361521fc7f
thread:issue:38205,issue #38205: RESOLVED: Disable zero-dim CUDA tensors interacting with CUDA tensors on other devices,stale/low urgency,2,stale/low urgency,1,0,38205,,,,0,5c94d45755d9463183b649b5d3b48a744fe3452fe9d0787f156262c3cd4ede17
thread:issue:38212,"issue #38212: Please help, building failing with MAGMA support",stale/low urgency,2,stale/low urgency,1,0,38212,,,,0,f2e3d3b84419dea171380e2e17f310aecd143af936b99956d264371049e84ac0
thread:issue:3823,issue #3823: Fuse bias to CuDNN convolution,stale/low urgency,2,stale/low urgency,1,0,3823,,,,0,62b4c9be374c31ba5c6620c3134daeac6859e7f3fe93ee5c0ed0ad74ed7b8223
thread:issue:38233,issue #38233: When TorchScripted module has bad type annotation you get bad error message,stale/low urgency,2,stale/low urgency,1,0,38233,,,,0,92b4615b13a262ce45e60be7790d76ebd8d4ac497704d42249ce5cf65e7508fc
thread:issue:38239,issue #38239: Failed to link torch_library using cmake,stale/low urgency,2,stale/low urgency,1,0,38239,,,,0,5ec9c32843a11954fd39386ab20b79496fd3629fb8931a09e43eaf83e5c5b622
thread:issue:38240,issue #38240: RuntimeError: arg_types.size() == param_names.size() - (moduleSelf_ ? 1 : 0) INTERNAL ASSERT FAILED,stale/low urgency,2,stale/low urgency,1,0,38240,,,,0,017142ac32b10fc702670a8c481c53cd55a5e7bfbef2dc4b68fe5387815370e1
thread:issue:38241,issue #38241: SIGXCPU at test_cholesky_solve with AMD EPYC 7742 64-Core Processor,stale/low urgency,2,stale/low urgency,1,0,38241,,,,0,65a9bad382acd189ea51dc7a068ce8814ae51cc51caa4fefd6d11b7ed39d3c04
thread:issue:38245,issue #38245: Make torch.rpc accept store as optional parameter,stale/low urgency,2,stale/low urgency,1,0,38245,,,,0,ce1ede95900fe4cb96152cdcf1de730fc7a64086aff1a75b52d4748ede9a69bd
thread:issue:38272,"issue #38272: PyTorch's rot90 returns a new tensor, inconsistent with NumPy's returning a view",stale/low urgency,2,stale/low urgency,1,0,38272,,,,0,2d00ae69a7e79d929c6c7c057338be5dabaf3cbcf845ee0813d4e3eef326b68c
thread:issue:38319,issue #38319: c10/macros/cmake_macros.h not exists,stale/low urgency,2,stale/low urgency,1,0,38319,,,,0,764e5b63af97c2072968ac9bbc4926c71a1d3879ad85cf0f0b7b3d3962abb1b6
thread:issue:38323,issue #38323: Support alternate casting rules,stale/low urgency,2,stale/low urgency,1,0,38323,,,,0,0109d86d2329b3c12b9fb1ff5c3593aeac3e0093a382605ed873478f64d4022a
thread:issue:38328,issue #38328: Provide an issubdtype API,stale/low urgency,2,stale/low urgency,1,0,38328,,,,0,ba99fd842d3ac18861e251ee13cdc7270fad5caff048801c77847f641bf067fb
thread:issue:38354,"issue #38354: Can I add more ""Project Documentation"" on the PYPI webpage?",stale/low urgency,2,stale/low urgency,1,0,38354,,,,0,593ea74ab461704fa35ed3d740f83e9354e7379ce9a21a8843697d3b143f7033
thread:issue:38394,issue #38394: 4D tensor support for torch.nn.functionnal.fold() (col2im),stale/low urgency,2,stale/low urgency,1,0,38394,,,,0,24f4535d0321537ade7d972ba60d24194f58cd878cecbaad2ecc978fd4795924
thread:issue:38412,issue #38412: No MKL Compatible Conda installation for PyTorch 1.5,stale/low urgency,2,stale/low urgency,1,0,38412,,,,0,7bfba4d14557837ac632700d086a24e4ec54ff006a689692e8d482849bb8ae75
thread:issue:38437,issue #38437: [JIT] torch.tensor needs a Tensor overload,stale/low urgency,2,stale/low urgency,1,0,38437,,,,0,8be75136b88b6db68b2776aeef97291be551266eacda4f91fd6902f68f8a98cb
thread:issue:38457,issue #38457: Named Tensor and Indexing,stale/low urgency,2,stale/low urgency,1,0,38457,,,,0,404f90a1d558b33c23da579ddea13ae5498ae4e2afda356d61452cc4b7b2b115
thread:issue:38470,issue #38470: [Feature] Option to have zeros/ones/full output tensor with zero strides,stale/low urgency,2,stale/low urgency,1,0,38470,,,,0,3d56c9b6c67e33087ea7aed55dec3c2b5f80589eeeaa0a6e541ad1e56ac3bea3
thread:issue:38471,issue #38471: Immediate mode API (with functional flavor) for optimizers,stale/low urgency,2,stale/low urgency,1,0,38471,,,,0,2d674fa2cca933dbed731cef66114f6f79064e42d372e3cb4293e7a3e31d8482
thread:issue:38472,issue #38472: [torch.jit.trace] torch.jit.trace fixed batch size CNN,stale/low urgency,2,stale/low urgency,1,0,38472,,,,0,95ae1de0f26e4ed7c000a05583e682ed3d1533c2189451a112083817c7f8ec5f
thread:issue:38473,issue #38473: Python builtin function next() is currently not supported in Torchscript,stale/low urgency,2,stale/low urgency,1,0,38473,,,,0,a662bee2e3fa0a7e873ccbd959994eaabffe1a7828a101d0ec501a2bd15749d8
thread:issue:38474,issue #38474: Scale parameter downcasted and rounded down in pytorch.distributions.Normal,stale/low urgency,2,stale/low urgency,1,0,38474,,,,0,4e1b027a6fe3e7c42c40dd70b8f22d15b9199fb720916f47106785ba3aa75496
thread:issue:38487,issue #38487: expected scalar type Half but found Float with torch.cuda.amp and torch.nn.DataParallel,stale/low urgency,2,stale/low urgency,1,0,38487,,,,0,a1e302f4e536bc7da48c12c87055bf2b19486425deef10300c645ed3e6ed4545
thread:issue:38510,issue #38510: [DISCUSSION] RPC server-side ThreadLocalState,stale/low urgency,2,stale/low urgency,1,0,38510,,,,0,744890e55f536ad8c95f12e7a33e13f81b67b17e0ccd78b5994f17b8859d9f83
thread:issue:38536,"issue #38536: build libtorch problem: Configuring incomplete, errors occurred! _mm256_abs_epi16",stale/low urgency,2,stale/low urgency,1,0,38536,,,,0,dabb1e8869c2e5f72b99be1437a7be11b95ad122c05c103d1854d6f1c9b197e8
thread:issue:38541,issue #38541: Doc update regarding predictability of experiments using Seeds and Workers,stale/low urgency,2,stale/low urgency,1,0,38541,,,,0,1717b66dc012377ab879e53337886ad985f8b991b8b605a8e68d6746876124d1
thread:issue:38543,issue #38543: Make rebuildBucket() to be async in c10d reducer,stale/low urgency,2,stale/low urgency,1,0,38543,,,,0,012dcffb089c1438e2039898b57f1a4d04c2ab03234b4661ecb3ad184a4ddeaf
thread:issue:38548,issue #38548: Add Distributed LR Scheduler to RPC,stale/low urgency,2,stale/low urgency,1,0,38548,,,,0,0f89237a377ddef414880b1c97cd84eda34f9a03195a1b2b7e62cd83a9c3c805
thread:issue:38550,issue #38550: Returning a tensor instead of a list in split and chunk,stale/low urgency,2,stale/low urgency,1,0,38550,,,,0,bfd82226c57ebec628876c1879f5e9053804f13f906f0a1dc0ca556d894884c3
thread:issue:38553,"issue #38553: When I use cuda(), wg = th.matmul(extra_obs, extra_obs.transpose(-2, -1)) take a mistake",stale/low urgency,2,stale/low urgency,1,0,38553,,,,0,7d8d10f1bd691794d65368f4e7790271d96bc222b1742f55f349423ea03d26af
thread:issue:38564,issue #38564: tensorboard projector mode with custom metadata_header with only one label name,stale/low urgency,2,stale/low urgency,1,0,38564,,,,0,e054c9a086dba7c99ed06512475991a0c446076ea4dc0f85f32f0c55d79d9171
thread:issue:38569,issue #38569: [quantization] Version support for quantization BC tests,stale/low urgency,2,stale/low urgency,1,0,38569,,,,0,36ba895d00beb42f7a457f959fad407e39d59031dc5b02174e15f41ca589deaf
thread:issue:38595,issue #38595: torch.norm p/ord parameter documentation is wrong,stale/low urgency,2,stale/low urgency,1,0,38595,,,,0,a9a6e790c07d6e2db7be4aada67eb93c7257cfbb6bef917deddd13ed0463ceda
thread:issue:38611,issue #38611: Investigate using -cospi(u) / sinpi(u) instead of tan(pi * (u - 0.5)) in transformation::cauchy,stale/low urgency,2,stale/low urgency,1,0,38611,,,,0,61638ae9d3ac86b1a38bb9bcadfbb72e65dc2ea60d900b837ec9e90ff90c31ec
thread:issue:38612,issue #38612: Investigate exponential distribution improvements,stale/low urgency,2,stale/low urgency,1,0,38612,,,,0,646358efe45e5231aa82a3963627fe5b2a9c6e50529e6d9ee3fb28a7175742e0
thread:issue:38613,issue #38613: Investigate using log1p instead of log in transformation functions(TransformationHelper.h),stale/low urgency,2,stale/low urgency,1,0,38613,,,,0,70c7faa22edcb293eca7c52ab24410e57257b0dab15ca21bd9716eb7b0efbeca
thread:issue:38619,"issue #38619: Can you add NMS,RoIAlign,RoIPool for libtorch?",stale/low urgency,2,stale/low urgency,1,0,38619,,,,0,5b92ac0316d1c4dec400f1f7280252da0b4799e6df33220ddb87698aeaf23459
thread:issue:38624,issue #38624: Wheels not manylinux1 compliant,stale/low urgency,2,stale/low urgency,1,0,38624,,,,0,ea67189e3055c2d89ca0ee13aca27deb207ecbd08aebb9729fa6e0243a4f05dd
thread:issue:38630,issue #38630: Torchvision error TypeError: _resolve_type_from_object(),stale/low urgency,2,stale/low urgency,1,0,38630,,,,0,7f77383620db50a6e7bf0d1741974743edfa7e8e90e33b3e8eb2464640809d0b
thread:issue:38639,issue #38639: Difference between using a python list and nn.ModuleList,stale/low urgency,2,stale/low urgency,1,0,38639,,,,0,95b58851e47c3d4668281b86529ee69c3e9b5c93c161d0f7472585b13ba0af65
thread:issue:38640,issue #38640: Maxunpool seems to give a weird error message,stale/low urgency,2,stale/low urgency,1,0,38640,,,,0,e4afb3a4b9a1f263ebbf7fb9f2032a2611012471c0cfb84a806abe72f5283469
thread:issue:38643,issue #38643: torch while loading weighs found runtime error on storage has wrong size: expected 4254413747647032608 got 1024,stale/low urgency,2,stale/low urgency,1,0,38643,,,,0,974de46e7f13842cebdc57df82fa1388e1e7667472c3ba0f999491e658412d3a
thread:issue:38650,issue #38650: Implement torch.erf for complex dtypes,stale/low urgency,2,stale/low urgency,1,0,38650,,,,0,5f1ed3d6367081268abbb61728772bdac37bb7c95b31ef93787f7ec0452010d1
thread:issue:38662,issue #38662: Unify CPU/CUDA exponential transformation formula,stale/low urgency,2,stale/low urgency,1,0,38662,,,,0,d0268fa101dfef0f40d43c7594806d1c37dea7dd235a2e0f067ef8e5563d0e4a
thread:issue:38671,issue #38671: Add CUDA callback to Python API,stale/low urgency,2,stale/low urgency,1,0,38671,,,,0,cae82ba557d3130b723315380bd41571c484365e055af78175502b4f55e8c377
thread:issue:38677,issue #38677: TestCase.assertEqual does not distinguish Python builtin types and single-element Tensor,stale/low urgency,2,stale/low urgency,1,0,38677,,,,0,726a3b0a9ca29aea7385bf80b8f43d245741693aa90d186f8df08f43a52ee241
thread:issue:38684,issue #38684: Add Plackett-Luce distribution,stale/low urgency,2,stale/low urgency,1,0,38684,,,,0,0a1741651e19335f0fb263024e9bebfa837ebf25429be15f8a2d7c5c2fd10d38
thread:issue:38709,issue #38709: NaN Loss for FasterRCNN on Multiclass Object Detection on Custom Dataset COCO,stale/low urgency,2,stale/low urgency,1,0,38709,,,,0,d0a28b73d287ffadf59a2964191abdfdfa830173bd17293ce888441e41ab7076
thread:issue:38719,issue #38719: Building NVCC (Device) failed when building from source,stale/low urgency,2,stale/low urgency,1,0,38719,,,,0,5369077484567ea76c1740c5aa9aed5ed2fb749898073ed8bc86d163f55054f8
thread:issue:38720,issue #38720: DOC: add documentation for undocumented classes and methods,stale/low urgency,2,stale/low urgency,1,0,38720,,,,0,28106a5ee72a0bc1694e55a3d1492c2f58cad81ccd577212a891fdc3a1af02cb
thread:issue:38737,issue #38737: equivalent of tensorflow's embedding_lookup_sparse,stale/low urgency,2,stale/low urgency,1,0,38737,,,,0,d5355841c7a48d6adee530c44a2e8803989a2d35d451a8eba57b2d4c34b97799
thread:issue:38765,issue #38765: LibTorch 1.5.0 not supporting GLIBC < 2.23,stale/low urgency,2,stale/low urgency,1,0,38765,,,,0,3a45f3815e1dadba13d84aa364b4161608b5c9e25bfb41485a04591237b70546
thread:issue:38782,issue #38782: SPMG in DDP does not have gradient computation and communication overlap?,stale/low urgency,2,stale/low urgency,1,0,38782,,,,0,0f7d3d56bb83fb2a3ed07ddb27ce0c66f39bf2b567e864c36273460a7addc979
thread:issue:38788,issue #38788: Convergence issues when using pytorch's native AMP,stale/low urgency,2,stale/low urgency,1,0,38788,,,,0,7aec94aed236af97a72074b331a4ed7c48b3c8f040734d591173c45e30c57356
thread:issue:38853,issue #38853: Weight decay in AdamW,stale/low urgency,2,stale/low urgency,1,0,38853,,,,0,ddca5ac958bbe3a02d8713ed46dbc2044e024651f89fc4788962b5214ea1a37e
thread:issue:38854,issue #38854: Can't compile QuantizedOpKernels.cpp using gcc-9.3,stale/low urgency,2,stale/low urgency,1,0,38854,,,,0,527bbb9037904b7e83b4fd9c1ab5540dfbbfd24902b4771502b39e339ebd6b82
thread:issue:38855,issue #38855: Can not compile GridSamplerKernel.cpp with gcc-9.3,stale/low urgency,2,stale/low urgency,1,0,38855,,,,0,370939f701873181c8ebc77fbf2fc84216c012e275455d655114abaa064e571f
thread:issue:38863,issue #38863: [JIT] named_(parameters | buffers) do not support recursive iteration.,stale/low urgency,2,stale/low urgency,1,0,38863,,,,0,3f6c2053b79ac2827aaa1fa57950086d0e4b536dee4f815e34d3515226aa7c24
thread:issue:38876,issue #38876: Override `__call__` instead of `forward`,stale/low urgency,2,stale/low urgency,1,0,38876,,,,0,f2c1107ab999b044c2cb569aae1ec66197f3eb4943a72753a68a6e7f48883f04
thread:issue:38911,issue #38911: [FR] cuda.comm.broadcast/reduce_add support `out=`,stale/low urgency,2,stale/low urgency,1,0,38911,,,,0,df4bd4346c7dab982dfb428ecbff61ce8a8f9e5af28ef8e4eef151793c45535c
thread:issue:38936,issue #38936: [FR] DataParallel arg rename device_ids->devices,stale/low urgency,2,stale/low urgency,1,0,38936,,,,0,bba519d74c2d88ddf88f5105bcfba56a262fec3d56ff7a8382678ef9cd273964
thread:issue:38940,issue #38940: [FR] Support SyncBatchNorm in DataParallel,stale/low urgency,2,stale/low urgency,1,0,38940,,,,0,865027c2e00c5f11b19688e80524d378421944d81ce4d2fe182bb50708ebe41f
thread:issue:38955,issue #38955: fold variation,stale/low urgency,2,stale/low urgency,1,0,38955,,,,0,09e9e89d0f57727868c8db1b7ffb9373fd5909a25711c803e830162d37ad5829
thread:issue:38964,issue #38964: [JIT] jit can not recognize the imported function,stale/low urgency,2,stale/low urgency,1,0,38964,,,,0,4d84ccf71fb3ea212777522585fe45b1fcc64d467371d375d0ec7cfe784cdf63
thread:issue:38973,issue #38973: ATen operator API versioning,stale/low urgency,2,stale/low urgency,1,0,38973,,,,0,fc5ef4641750c24dcfa74e9e0bedc54850610465896251ecfcfcecac7149738f
thread:issue:38974,issue #38974: ATen registrable operator list,stale/low urgency,2,stale/low urgency,1,0,38974,,,,0,9d80d1a2d21e26bd615b533b4d81935a1e999adb590aafa1b8100d97a3b1e7a1
thread:issue:3898,issue #3898: Sparse matrices in dataloader error,stale/low urgency,2,stale/low urgency,1,0,3898,,,,0,5c9cddbfce24dba08880b5cab6f85fa06cd91b47d2bfcb5f4bee5f1125720737
thread:issue:38980,issue #38980: runtime error while using default arguments in add_graph() function,stale/low urgency,2,stale/low urgency,1,0,38980,,,,0,a376ac21a268cc83abd5fffa5bdafd966ea0ef7d4f2b49515ed3e0c64bd77092
thread:issue:38996,issue #38996: [FR] API consistency for cuda.comm and distributed,stale/low urgency,2,stale/low urgency,1,0,38996,,,,0,3fc15e1427b39c8193204f2ad3a087dee918d4b083a08081c637a23026c6248f
thread:issue:38997,issue #38997: [doc] document cuda.nccl,stale/low urgency,2,stale/low urgency,1,0,38997,,,,0,effaa0daf8e144740fd98e535280bad22ed48c6ceae81368f9f2bf487d22b907
thread:issue:39009,issue #39009: Missing OneCycleLR and MultiplicativeLR in lr_scheduler.pyi,stale/low urgency,2,stale/low urgency,1,0,39009,,,,0,6cb063ab3b220858e49f79ed40de5211c8e222a652e7415bcadb9319de032f29
thread:issue:39022,issue #39022: Cap DDP total number of buckets,stale/low urgency,2,stale/low urgency,1,0,39022,,,,0,d2c3045c94ad8bafbb58d9a34c86e45cf18d004786411381851734a56fa06915
thread:issue:39058,issue #39058: Failure when loading quantized pre-trained weights partially,stale/low urgency,2,stale/low urgency,1,0,39058,,,,0,e96e5b3f03de457745f74fb4345d5c1b7c6ffa0db3b624e52939ffc5e956526c
thread:issue:39064,issue #39064: Add cpack support to CMakeLists.txt,stale/low urgency,2,stale/low urgency,1,0,39064,,,,0,94a9a8cd81d2fd9465fe0ce339a4f2c0a6822ed7dfaae336f9ae0b3a248cfebc
thread:issue:39082,issue #39082: [TensorPipe] Avoid wrapping the future message in order to do atomic test-and-set,stale/low urgency,2,stale/low urgency,1,0,39082,,,,0,7d0d161d586f4e9fee37d4b6a7029e94d95ed7c458faf2369e980335fc870482
thread:issue:39101,issue #39101: Build issue when installing pytorch with USE_FFMPEG=1 USE_OPENCV=1,stale/low urgency,2,stale/low urgency,1,0,39101,,,,0,72368c0cf6c23d09cbf021c02ebbcdda2b090f2b495243b67347bec71eb71432
thread:issue:39123,issue #39123: The MacOS compiler is generating illegal instruction for the division of c10::complex,stale/low urgency,2,stale/low urgency,1,0,39123,,,,0,7b78fa1883f1cc571048cf1c71d78c47093fea19b57db2e51f09a2727f623ed9
thread:issue:39124,issue #39124: [TensorPipe] Errors in pipeWrite should clear out the future in pendingResponseMessage,stale/low urgency,2,stale/low urgency,1,0,39124,,,,0,b1dc9efc4e8eb9b6b6d3f4f5cdbc0636c43d80b09e01b2d220577de81a202cc8
thread:issue:39128,"issue #39128: Problems implementing complex support for acosh, asinh, tanh",stale/low urgency,2,stale/low urgency,1,0,39128,,,,0,728001f99dbc189b1cc7a7d9b9c4f8f7dbbb1ab18679745019bf63b693d3d9d0
thread:issue:39130,issue #39130: `torch.backends` undocumented,stale/low urgency,2,stale/low urgency,1,0,39130,,,,0,ec977728a31ad9635edc383641182d422c8a9dac03a5d8925b960c4bebf3183a
thread:issue:39134,issue #39134: Crossentropy inconsistent results depending on tensor order,stale/low urgency,2,stale/low urgency,1,0,39134,,,,0,0784ce5eab1cb31e0a8b6817702f613379fc9a96b6b8bb4a634c7bc97e61d266
thread:issue:39178,issue #39178: Add support for rsample to MixtureSameFamily Distribution,stale/low urgency,2,stale/low urgency,1,0,39178,,,,0,d9c81af296db182620b4d4c594841db698845d0b57bc1dee2658928f6077fdb3
thread:issue:3920,issue #3920: [docs] Tensor.new is not documented,stale/low urgency,2,stale/low urgency,1,0,3920,,,,0,3a45499787bf0995638ae6d90c43eb1b10342cea5002e858133c5a1b833b9ff2
thread:issue:39226,issue #39226: Try Address Sanitizer in MSVC builds,stale/low urgency,2,stale/low urgency,1,0,39226,,,,0,ac8dd281907df2e9031aa58d8ec1cfb63e0c8f9fc00ae54212cf04cf4c7ae4e1
thread:issue:39245,issue #39245: pytorch-crf model to onnx conversion,stale/low urgency,2,stale/low urgency,1,0,39245,,,,0,6e2258b6b18434da62358af7c028c025315359ada6610f198588b6885a72f0b4
thread:issue:39260,issue #39260: Naming inconsistency: padding_mode vs pad_mode + F.conv* docs ops miss padding_mode arg at all,stale/low urgency,2,stale/low urgency,1,0,39260,,,,0,ca8ce89822551da5a75b9e4881ee3e7a5d929a2866f9d21bdf938f948fdbe19c
thread:issue:39270,issue #39270: Current implementation of c10::complex does not support being used in shared memory,stale/low urgency,2,stale/low urgency,1,0,39270,,,,0,7a664122faa0ebcf8aab542a0d170ba8c90985968507811d110cbe98d4a7eb53
thread:issue:39280,issue #39280: Let future.wait() take in an optional timeout,stale/low urgency,2,stale/low urgency,1,0,39280,,,,0,41035bb6ffd0f48399571e3be8d27fbf4f51f7550adfba8b0e30d97b6866951b
thread:issue:39298,issue #39298: Caffe 2 android app not working with proguard and minifyEnabled,stale/low urgency,2,stale/low urgency,1,0,39298,,,,0,a1bbba1de57c93dcc5c12b75fff84558fedef71361c2a61c9891ee940fcbc10d
thread:issue:39305,issue #39305: Compilation errors,stale/low urgency,2,stale/low urgency,1,0,39305,,,,0,7505df0f8cd66c8b3ceadcad8da9a0acbafc54f222fa9853977d7677eaf94ac7
thread:issue:3931,issue #3931: Suppress hidden state output of RNNs?,stale/low urgency,2,stale/low urgency,1,0,3931,,,,0,53aebac78cd04dc1882ceba07e382ea89b3ff65e773dafe2ed9fa422343fb9fc
thread:issue:39310,issue #39310: Make torch.cross `dim` parameter work intuitively,stale/low urgency,2,stale/low urgency,1,0,39310,,,,0,585336149d3bcfb68f293fcc7de511b48b19249917a02deca31591d9c977c664
thread:issue:39317,issue #39317: send a Tensor to Cuda very slow,stale/low urgency,2,stale/low urgency,1,0,39317,,,,0,5898dee2867c32d94e77d34d9b5762ec62ede5d72bf8109f01cf1e4d89d97dd4
thread:issue:39355,issue #39355: Clean up GIL that used to guard deleted RRef destructions,stale/low urgency,2,stale/low urgency,1,0,39355,,,,0,706c067f9e85e94b2a34ee2f3feae2f4f4ec3ab8d9a71f3ffb006dc06d9baf88
thread:issue:39356,issue #39356: Torch script to support dictionary with keys of type tuple,stale/low urgency,2,stale/low urgency,1,0,39356,,,,0,ff410ed6cf4c210e5c87da855713aa269a90fd9c3ac7c352cc65ac733be06cac
thread:issue:39357,issue #39357: Enable scripting modules with recursive functions,stale/low urgency,2,stale/low urgency,1,0,39357,,,,0,030cc68676675f40e89920902db3d49fa2f37d60a205aa129e26b26ec48acb35
thread:issue:39358,issue #39358: SyncBatchNorm doesn't scale well across multiple nodes on large data sizes,stale/low urgency,2,stale/low urgency,1,0,39358,,,,0,d906aff9fc1b79fc73cb844c35beec0b64413637abdd234789f5531a923ced9f
thread:issue:39362,"issue #39362: The conversion of Inplace aten::append to outplace is incomplete, creating incorrect subgraphs",stale/low urgency,2,stale/low urgency,1,0,39362,,,,0,9dca75b2cdc7669441bd2e5beea4d9c2998147f4b6a69aa98bdbd23d88fcd2e8
thread:issue:39365,issue #39365: Attributes not removed by freeze module,stale/low urgency,2,stale/low urgency,1,0,39365,,,,0,a58b22d7e8b05f2400ac3bd8ca4d7ccc92137cadf4bc134880f29b2f8f80e7ea
thread:issue:39370,issue #39370: Draw quantized tensors in tensorboard,stale/low urgency,2,stale/low urgency,1,0,39370,,,,0,7244d68661a23984888a6c6b8a11a67716677f5007024f8e03a09ad6e9524a1c
thread:issue:39395,issue #39395: Morphological operations,stale/low urgency,2,stale/low urgency,1,0,39395,,,,0,4bceadab7b5f0079d856d3cf90952aeb103e828fb82bd970110d9f80bab07522
thread:issue:39399,issue #39399: Update RPC doc to recommend async user functions for non-blocking server execution,stale/low urgency,2,stale/low urgency,1,0,39399,,,,0,8928a7a072e6a2f24e4e981fde66b234282524a464accea3e593db6d0db44e1d
thread:issue:39411,issue #39411: Allow to_here to interrupt blocking wait in the case of rpc.remote timeout,stale/low urgency,2,stale/low urgency,1,0,39411,,,,0,d8964aac4de77ac7e93d9726e831207966b955b61965184aba2e03cb6ac39ebb
thread:issue:39421,issue #39421: torch::jit::script::Module::to(torch::kDouble) also casts buffers,stale/low urgency,2,stale/low urgency,1,0,39421,,,,0,25eda13422ec874789db375809caae9842e8a86fe5d54dd06757f39d8e907635
thread:issue:39426,issue #39426: Format issue in `torch.quantization.add_quant_dequant` documentation parameter section,stale/low urgency,2,stale/low urgency,1,0,39426,,,,0,df34a43b13426ad558283e0c4e6a2a1a6310bf00fcbcd6c6418a7cad3b38175f
thread:issue:39435,issue #39435: BCEWithLogitsLoss() not equal to BCELoss() with sigmoid(),stale/low urgency,2,stale/low urgency,1,0,39435,,,,0,5194b9443a3b61a65f8915d41dfe4ec01793c9690a02efaef2d3a9cbaa0a4405
thread:issue:39437,issue #39437: error when specifying sparse=True in embedding,stale/low urgency,2,stale/low urgency,1,0,39437,,,,0,61cce2452fd0fa91ed0819598bba91cf7880e52ccee39f1fa82619e227be4c6f
thread:issue:39444,issue #39444: Torch hub: object has no attribute nms,stale/low urgency,2,stale/low urgency,1,0,39444,,,,0,fa6740c716e6877d55f39b8df4751614d44ca222319a87b85ea351ae067f3a57
thread:issue:39463,issue #39463: JIT test suite has dependencies across tests,stale/low urgency,2,stale/low urgency,1,0,39463,,,,0,97fc4ad877a3d5346e7a0bfa4df434f2070ef3adf4399a2ee95fdda064d4ec2b
thread:issue:39495,issue #39495: Misannotation of layer_norm parameters causes internal assert failure,stale/low urgency,2,stale/low urgency,1,0,39495,,,,0,8d11f8201450d649e3bb7aa93465f8b956c9c10c62f2d9aabae27e4f4cb7e961
thread:issue:39503,issue #39503: pytest suppresses stderr from Python startup by default,stale/low urgency,2,stale/low urgency,1,0,39503,,,,0,7305058d84882aaf10900db28822e0eee1dc06d7763961f60bfc9d7890cbc726
thread:issue:39522,issue #39522: Valgrind leak checking flags losses in libtorch,stale/low urgency,2,stale/low urgency,1,0,39522,,,,0,61b619c16ebb55ee2482c77bcb4623b22ce0c14eace3ce1ed34b023162d9dfc9
thread:issue:39524,issue #39524: test_nn_module_tests should run less tests,stale/low urgency,2,stale/low urgency,1,0,39524,,,,0,d8b617d92c2ae157ca72954457bcfb1203130738dc8d440bc5d40055655f2932
thread:issue:39537,issue #39537: Segfault = docker + tensorboard + pytorch,stale/low urgency,2,stale/low urgency,1,0,39537,,,,0,97959104cca9928fd62b4fd3203971a0e8d285ca9a225b1ddb42549dbf4aedf0
thread:issue:39603,issue #39603: friend constexpr in templated struct loses constexpr-ness in nvcc,stale/low urgency,2,stale/low urgency,1,0,39603,,,,0,947ffe2bd28bec9de88ad174b35a04c212ae74c252399307166d639bf0c1b158
thread:issue:39619,issue #39619: Switching from CPU build to CUDA build in conda environments is a bit tricky,stale/low urgency,2,stale/low urgency,1,0,39619,,,,0,477e267f67d483b661affb1728a9559d5ee0577f92bd338f070f0a9fc3ecf5ec
thread:issue:39625,issue #39625: All keys matched successfully missing when loading state dict on optimizers,stale/low urgency,2,stale/low urgency,1,0,39625,,,,0,26328679ab0127d966d1ba14d8f9fa22ddff953430c13fa9db1ceef4a4017a3c
thread:issue:39632,issue #39632: F.affine_grid dispatch async issue,stale/low urgency,2,stale/low urgency,1,0,39632,,,,0,9c8beb57a118954ba4bd017f91fabfff113d0768dce078cc5cd61d28a7adcd35
thread:issue:39634,issue #39634: Installation from source fails on macOS (No CUDA),stale/low urgency,2,stale/low urgency,1,0,39634,,,,0,1d6d147d103a4eb3ce1c25df149139103583a53f5b171e440b42245ed32ee652
thread:issue:39656,issue #39656: Broadcasting for torch.cross,stale/low urgency,2,stale/low urgency,1,0,39656,,,,0,d300d967f2cbfe0b458b07b92794f73680d62814ae30fa7c5b0b1e3cff4e4b83
thread:issue:39662,issue #39662: [RFC][distributed] RFC: c10d ProcessGroup extension and C++ API change,stale/low urgency,2,stale/low urgency,1,0,39662,,,,0,202f7a718944b4ffa2d1e2077c0a11eac2b51989fdba42519fe777f84f5a21ef
thread:issue:39665,issue #39665: Better err msg for tensor ctor from sequence,stale/low urgency,2,stale/low urgency,1,0,39665,,,,0,15901431173a71c7a1ba9004300b479cc1050e3272f6cfdf9f2ec3edac012acc
thread:issue:39670,issue #39670: Force JIT to do type inference even when mypy annotated,stale/low urgency,2,stale/low urgency,1,0,39670,,,,0,92a5e9ba1bc0bc2e79ff9f5542738b5cbdd4ba372cd062c6c9ff36283102975b
thread:issue:39671,issue #39671: TestTorchDeviceTypeCPU.test_float_to_int_conversion_finite_cpu_uint8 is broken on PowerPC,stale/low urgency,2,stale/low urgency,1,0,39671,,,,0,f88a2a07e5b4fac1d956ac8c172d9904c5c6680376352d554685fded5bda65b9
thread:issue:39674,issue #39674: [BatchNorm] Add boolean flags to choose the stats for normalization,stale/low urgency,2,stale/low urgency,1,0,39674,,,,0,a095126f25fbd33ad8442c4c3adf98f739866122ad7db92c839a2c4f91d4cf6e
thread:issue:39680,issue #39680: CUDA cannot be found,stale/low urgency,2,stale/low urgency,1,0,39680,,,,0,20abe090d12f0abaf0a85a509a24b4a6cd99ef64bc1b6efdfb4400b084f9ce18
thread:issue:39682,issue #39682: Run `clang-tidy` on the `aten` folder?,stale/low urgency,2,stale/low urgency,1,0,39682,,,,0,677e01f87ef24178a7a969688c17e9c6c37dc5c3c37ffa18247495c9443fb65d
thread:issue:39705,issue #39705: [doc] [distributed] example of specifying url etc in url,stale/low urgency,2,stale/low urgency,1,0,39705,,,,0,fad8ece1276e6e6793e0959fa2e868f6b5af607310f5d8c907098b3bd72d5ec3
thread:issue:39706,issue #39706: [distributed] calling nccl reduce with inconsistent dst hangs,stale/low urgency,2,stale/low urgency,1,0,39706,,,,0,15968ed393c1a372bca23c518e457e7a20b3cd4f6331407d285b6476107b463f
thread:issue:39716,issue #39716: Do not modify global random state,stale/low urgency,2,stale/low urgency,1,0,39716,,,,0,008351a7e3f84fc77ae37e85b04a8269230e8e21d0b0de183271b621ec7d162f
thread:issue:39719,issue #39719: User Objects aren't recursively scripted as nn.Module attributes,stale/low urgency,2,stale/low urgency,1,0,39719,,,,0,dbc94911006f0bf68ddd833cb346182fc69176ce257c0340019de560c9210695
thread:issue:39745,issue #39745: [JIT] exported dunder methods are ignored,stale/low urgency,2,stale/low urgency,1,0,39745,,,,0,f19a82d1c78099d6086879b5bbc86896668b76a057398770ec8babe16ff6dcae
thread:issue:39757,issue #39757: max_unpool2d and max_unpool3d cpp signature should be similar,stale/low urgency,2,stale/low urgency,1,0,39757,,,,0,16edd7063062bbf4b05141909c59a996354ddef18062ff81d41aa5f8d2fb60ee
thread:issue:39799,issue #39799: Tensor.size() API not correctly documented?,stale/low urgency,2,stale/low urgency,1,0,39799,,,,0,f79458d59bfc4364e5184cb0332f77a95f327667cbac5dac8de45bdafdb4a47e
thread:issue:39806,issue #39806: Simplify layers of optionals in `VaryingShape` and `Stride`,stale/low urgency,2,stale/low urgency,1,0,39806,,,,0,c58ff4d556a294c99e57b61edf0633093d1ae947951d85c0ca9e3d889fd64874
thread:issue:39820,issue #39820: I would like to install pytorch-nightly from requirements.txt,stale/low urgency,2,stale/low urgency,1,0,39820,,,,0,2e8a3cf5d9ce9b5fd41c0fc1af7a5237cbf5c17eb22fde956bd2d11cd8d00265
thread:issue:39852,issue #39852: Shared file-system initialization in pytorch distributed is slow,stale/low urgency,2,stale/low urgency,1,0,39852,,,,0,da0c0aa325e37d8b2d14062642832501e620b774a1cde5be3b6f50864d6245e9
thread:issue:39864,issue #39864: Parallelize arguments serde for RPC with TorchScript functions.,stale/low urgency,2,stale/low urgency,1,0,39864,,,,0,6759bca23337d7b81e15aba113cb676cde9738d0cbb2dc7bb10b0d5b560404c0
thread:issue:3990,issue #3990: Raise an error when using magma built against wrong version of cuda,stale/low urgency,2,stale/low urgency,1,0,3990,,,,0,61c0516a52268fd4da24f01bef48db6fea6ec1656aee469b23f235b4a180d13c
thread:issue:39917,issue #39917: build QT program use libtorch-cxx11-abi-shared-with-deps-1.5.0+cu101 ok with CPU but error with cuda GPU,stale/low urgency,2,stale/low urgency,1,0,39917,,,,0,e3b2e9bfff41a1804a43e6c54005b9d9dd325c35536553df7008caf2ab419cac
thread:issue:39959,issue #39959: Different max_pool2d cpp signatures due to indices.,stale/low urgency,2,stale/low urgency,1,0,39959,,,,0,5ad72016dcb6c104a8301e71009b06d836407cd19719075294c0ae42e4f21554
thread:issue:39972,issue #39972: Port old registration API to new one,stale/low urgency,2,stale/low urgency,1,0,39972,,,,0,6b73b8a2c0856cbc6f9ecd2dfd2c89f655bb2d9d69439216159ca16ad8cfc6bb
thread:issue:39990,issue #39990: libtorch 1.5 macos crash when loading on some mac,stale/low urgency,2,stale/low urgency,1,0,39990,,,,0,41ee65cbc7a1fb4cfd0af6ac1001aabaff9ab15c56539bdcf6c9d7ec513a587c
thread:issue:39993,issue #39993: How do I derive weights for CrossEntropy Loss on my custom dataset?,stale/low urgency,2,stale/low urgency,1,0,39993,,,,0,123e32b04ed16f907f779f32adb9d81a47dbbd4bd4a88a427fc3520ccba990eb
thread:issue:40020,issue #40020: Caffe2 operation switches current CUDA stream,stale/low urgency,2,stale/low urgency,1,0,40020,,,,0,2a22b60d9b194db16dca4d5631be978b91de8f0ab8b132921faa4ada25e349bd
thread:issue:40024,issue #40024: torch::jit::load -> Unhandled exception,stale/low urgency,2,stale/low urgency,1,0,40024,,,,0,4b8636aa7a0a780ad0047133eb63ad66f3240231104c2af433742d0542257e89
thread:issue:40027,issue #40027: [Android Pytorch] TorchScript traced model returns inconsistent output tensors on each run,stale/low urgency,2,stale/low urgency,1,0,40027,,,,0,ffcbc033717ffb18f8aed9cba15bf048a78dab9abea2cb2477b0c05a1021d10f
thread:issue:40030,issue #40030: matchTensor tests. Add tests to make sure that `isSubtypeOf` and `matchTensor` return equivalent results,stale/low urgency,2,stale/low urgency,1,0,40030,,,,0,8092b777e70120d1ee413f5948d44daf9f20b5aee2a2d0507d7b39e0490747e6
thread:issue:40034,issue #40034: Doing optimizing compilations in a separate thread,stale/low urgency,2,stale/low urgency,1,0,40034,,,,0,ec93dc009e6283272a7789d3e3dcd002bdf01ed7cb34e5818f453e78588ace80
thread:issue:40070,issue #40070: [JIT][to-backend] Support type refinement for container types in the generated code,stale/low urgency,2,stale/low urgency,1,0,40070,,,,0,f577557c0d86abb90b34692e7974e0627a2510a576dd9dffb8f8b606fe0e9759
thread:issue:40071,issue #40071: [JIT][to-backend] Finalize the PyTorchBackendInterface API/naming for external sharing,stale/low urgency,2,stale/low urgency,1,0,40071,,,,0,e35c259df2b42b694cdde03c6aceedbdbda221c51f13e215c438bec98c0fc982
thread:issue:40086,issue #40086: TestListwiseL2rOps::test_lambda_rank_loss fails,stale/low urgency,2,stale/low urgency,1,0,40086,,,,0,3804e6c7a5484c6c572283f6ef1e28cdf943b3cb9416b70f57f898acb032746b
thread:issue:40087,issue #40087: Make Scaling in BatchNorm optional,stale/low urgency,2,stale/low urgency,1,0,40087,,,,0,af92ecf026bba2d056cd90dd3c68be1e58e9ec141413efdc46307c5da3ac687e
thread:issue:40095,issue #40095: Bump up NCCL to 2.7.3,stale/low urgency,2,stale/low urgency,1,0,40095,,,,0,620faadbb6e4d06196b52afe8ff5b8c272b3a50f29e11abee811d2205633b570
thread:issue:40098,issue #40098: Updating learning rate with Libtorch 1.5 and optimiser options,stale/low urgency,2,stale/low urgency,1,0,40098,,,,0,b4fbfbcae79ab9fcef6292c7be477696ab07ce6f47d727de19372d34843d2bb8
thread:issue:40114,issue #40114: ppc64le: test cpp_extensions/rng_extension.cpp failure (without altivec override),stale/low urgency,2,stale/low urgency,1,0,40114,,,,0,416b9d138708e0fd749b3de6bb34ad319046085e7267c4c9faec6014ab32de13
thread:issue:40134,issue #40134: About torch.backends.cudnn.deterministic issue,stale/low urgency,2,stale/low urgency,1,0,40134,,,,0,756f958d3c5cf8daa61129c6137764c8ad4a6a875bfb8782f84b8d6fcdf2690e
thread:issue:40147,issue #40147: SyncBatchNorm doesn't work when I set track_running_stats False,stale/low urgency,2,stale/low urgency,1,0,40147,,,,0,26aa96bd1602ebf87940a7c33263af8e778e5067d5077e79ffbec70df33b1b1c
thread:issue:40215,issue #40215: Move all torch.Tensor methods to codegen,stale/low urgency,2,stale/low urgency,1,0,40215,,,,0,cde0f1cc92ce75dae6c290ffc0288930c1920853d49f74b7c44035df7cc676b3
thread:issue:40237,issue #40237: Move torch cpp Errors to c10::Error,stale/low urgency,2,stale/low urgency,1,0,40237,,,,0,e1cf66d6081e5d47582027a5fa7972d127aced5158b572b1a85957d443f0ee6a
thread:issue:40250,issue #40250: Include expanded TensorOptions version of op in at:: namespace,stale/low urgency,2,stale/low urgency,1,0,40250,,,,0,6e54ce9fac8d63f815572c4b9b026de73beefde8ca1b51ff07e26eec95602b87
thread:issue:40266,issue #40266: CUDA error: out of memory when running tensorpipe test_cuda,stale/low urgency,2,stale/low urgency,1,0,40266,,,,0,b624da0a0962aa5e2795c9b1e3f86eca5a1027a335a76f65e270020fe6578eb3
thread:issue:40269,issue #40269: caffe2 error during compilation of PyTorch with ROCm on Archlinux,stale/low urgency,2,stale/low urgency,1,0,40269,,,,0,929e4f209bc10dadbd19560eb289e3d017bdf6d2522e1cc5613d1722090b1a2a
thread:issue:40275,issue #40275: Conv3d with specific kernel size outputs inconsistent results between FP16 and FP32 in V100 GPU,stale/low urgency,2,stale/low urgency,1,0,40275,,,,0,985875d1f8518faeeb747f92339a81714e77ee112ec292d779be5a49a2fb08b5
thread:issue:40319,issue #40319: Too many labels in the repo,stale/low urgency,2,stale/low urgency,1,0,40319,,,,0,14c831ae07bef302f931115b0564163a734732c9978618ba68466eca14be5317
thread:issue:40355,issue #40355: Tracking output dimensions of the convolutional layers,stale/low urgency,2,stale/low urgency,1,0,40355,,,,0,212206a989cd48c906363be0f0b69b5e65b0034302f082587baa7058081c353f
thread:issue:40375,issue #40375: Add batched torch.combinations,stale/low urgency,2,stale/low urgency,1,0,40375,,,,0,5d1a0d983c3355d7222fd2e13951d3af0f9d25aaa930c283eeb7dc8d477c7cd4
thread:issue:40400,issue #40400: New ProcessGroups created with dist.new_group may leak memory,stale/low urgency,2,stale/low urgency,1,0,40400,,,,0,e6fcb36eb75239444d6100772ebc4d8aa9abcd9fcd4678d56f057481e0b649f4
thread:issue:40417,issue #40417: Missing explanation in torch.utils.tensorboard.add_histogram(),stale/low urgency,2,stale/low urgency,1,0,40417,,,,0,8d2ea5f27ea08e648806b7211bad7ac9aaf860d648287fad459648b9c487e8b4
thread:issue:40419,issue #40419: Add a `like` argument to creation ops,stale/low urgency,2,stale/low urgency,1,0,40419,,,,0,4d2bcdd13c51781cc2e4402f96fc05222c4e6554127ab236ab1bb77dd18f1182
thread:issue:40427,"issue #40427: Can we make torch.inverse FP16? - RuntimeError: ""inverse_cuda"" not implemented for 'Half'",stale/low urgency,2,stale/low urgency,1,0,40427,,,,0,0b9a7865199a84e5cff59861ab24c24419a956e8df8d2c54d1e1681b850c7f24
thread:issue:40441,"issue #40441: [feature request] batch_apply, a general-purpose device-agnostic batch iterator",stale/low urgency,2,stale/low urgency,1,0,40441,,,,0,0bcb1f04c1e24c9a72b7a296329ecc087b8871936725e4a0d37ccd77bdb20419
thread:issue:40456,issue #40456: Print values (but not strings) when STRIP_ERROR_MESSAGES is defined,stale/low urgency,2,stale/low urgency,1,0,40456,,,,0,b7cb8255aa1e9c57e2ad5769d6dff9e505edc86c98b85fb5a918481f7aefa4ae
thread:issue:40470,issue #40470: jit's default dtype is different in sandcastle and test_jit.py,stale/low urgency,2,stale/low urgency,1,0,40470,,,,0,76ece1ea94160c21bdabd501e99f529f68b69af984096efe175435461fa82759
thread:issue:40480,issue #40480: torch.autograd.functional.* for models,stale/low urgency,2,stale/low urgency,1,0,40480,,,,0,9e5daad693cbe8fb808b21a4e74773caeb83f5287f695e08f88f3cf9f64d5fe4
thread:issue:40492,issue #40492: Learning rate change is not applied at designated iteration with a scheduler,stale/low urgency,2,stale/low urgency,1,0,40492,,,,0,f12d60cfda9fcdbceb2f69067c95ed2369d92148381f0ad0016a9cb7c1f07479
thread:issue:40497,issue #40497: Mixed precision causes NaN loss,stale/low urgency,2,stale/low urgency,1,0,40497,,,,0,ad38f197275dd85de8016af1b739c57af1b051f8da3c2ed21a611a54a32d1f04
thread:issue:40507,issue #40507: SyncBatchNorm for JIT and a list of not supported operations,stale/low urgency,2,stale/low urgency,1,0,40507,,,,0,faa33a74a0a6bad64e51d6900ca63d9f3a03f67365e9d79a34fd62ac78f9b8bc
thread:issue:40548,"issue #40548: Embedding with DataParallel can return ""incomplete"" results",stale/low urgency,2,stale/low urgency,1,0,40548,,,,0,74ff0f4655b4b0dfb53c3ccad50e5b34cfcdc260d732ee6899449142589cbe92
thread:issue:40550,issue #40550: RemoteModule enhancements,stale/low urgency,2,stale/low urgency,1,0,40550,,,,0,ddd2b07a008d446206502d68c65613d2f48837c308cc38c546e1d962c6d64211
thread:issue:40563,"issue #40563: Compiling PyTorch with 11.0, V11.0.167",stale/low urgency,2,stale/low urgency,1,0,40563,,,,0,b688a07955f65cfe84309c81ee165d3cd69caa8a9e875e2a7e01472cff81ca74
thread:issue:40564,issue #40564: Stop registering kernels that use DispatchStub as catch all,stale/low urgency,2,stale/low urgency,1,0,40564,,,,0,660fa56d4b0e1843e7540d8d69a59e92223fc395a3c865df17722acd79147987
thread:issue:40566,"issue #40566: If all parameters are unused by forward pass in a process, backward will not work with DDP.",stale/low urgency,2,stale/low urgency,1,0,40566,,,,0,dbaf8beb004102f50593b5b7bda0ef7ac9147bcabedc8530b188d64a9a5bfda7
thread:issue:40590,issue #40590: Inconsistent behavior between numpy.exp and torch.exp on CPU for complex numbers,stale/low urgency,2,stale/low urgency,1,0,40590,,,,0,09cad319c10ffcb3a1f4247f70fba28704028a8d72083900c53e4e94e52d2ca9
thread:issue:40633,issue #40633: RuntimeError: broken pipe from NCCL,stale/low urgency,2,stale/low urgency,1,0,40633,,,,0,881818f8bd075939e0eda4b00ba61d73ec3ba641381558261fb9d7e24a62855b
thread:issue:40635,issue #40635: [JIT] Recursive compilation doesn't apply for types used in type expressions but not value expressions,stale/low urgency,2,stale/low urgency,1,0,40635,,,,0,d08161604425ed6ba9f0fd0e6328bcf95189643630641cc570be1f6744abc660
thread:issue:40659,issue #40659: [typing] overly restrictive List[int],stale/low urgency,2,stale/low urgency,1,0,40659,,,,0,798d8f01cab3d14e5da07dac46a5abe2f1932335db544f71f41a106e1173b6ec
thread:issue:40667,issue #40667: Segmentation fault in forward pass using DataParallel and multiple GPUs,stale/low urgency,2,stale/low urgency,1,0,40667,,,,0,34388606f1f25b101499817fbad5ea10ff18999319e71753fe1af88a72956cdf
thread:issue:40679,issue #40679: [JIT] Tracing BCE loss throws error when using weight,stale/low urgency,2,stale/low urgency,1,0,40679,,,,0,5dceb00f218e51c6aa86cd429b2daf82b9df6e9bf2ee8a675c0af02653cf10cb
thread:issue:40687,issue #40687: libtorch: macros in logging_is_not_google_glog.h have very common names like CHECK or LOG,stale/low urgency,2,stale/low urgency,1,0,40687,,,,0,ea11c615877f2347cfa1227a08f1fefbe40b628baafac6f4cc7b9d2123ea0966
thread:issue:40729,issue #40729: Custom c++ extension build process doesn't preserve color from compiler,stale/low urgency,2,stale/low urgency,1,0,40729,,,,0,40d891ef3e3c3a6d610ec8f9088e081794a7b06479c9b9bc9fe542da8ecebe53
thread:issue:4073,issue #4073: Feature request: Correlation module,stale/low urgency,2,stale/low urgency,1,0,4073,,,,0,c7fab22db64dc9b576aac75ea5dd221c2fb1ca786dc955a247a2294326d105dd
thread:issue:40736,issue #40736: torch::autograd::Function should set AutoNonVariableTypeMode when running forward,stale/low urgency,2,stale/low urgency,1,0,40736,,,,0,b357dc60ae688253c09b87613ebc425a38c6114603e53c823ecbb10072954372
thread:issue:40756,issue #40756: RuntimeError: rois.device().is_cpu() ASSERT FAILED at /vision/torchvision/csrc/cpu/ROIAlign_cpu.cpp:386,stale/low urgency,2,stale/low urgency,1,0,40756,,,,0,658204ecc2760f2130ae407efd447654cff37a24c2d5ab562d971767cb4f1520
thread:issue:40759,issue #40759: Quantization support for F.Softmax,stale/low urgency,2,stale/low urgency,1,0,40759,,,,0,e0c366b42550dda469a53f7bc187017eda880d0609fed60bb03f676fcd8a469a
thread:issue:40763,issue #40763: Enhance supported types of functional.pad,stale/low urgency,2,stale/low urgency,1,0,40763,,,,0,53920c96c262cf226eb6f7b6727ac2dbe38f2807bd725af1a8f81cdb5cde65d9
thread:issue:40790,issue #40790: Encoding dimension argument for F.one_hot,stale/low urgency,2,stale/low urgency,1,0,40790,,,,0,5c34c867c4acda8abd4d5ec7f3dd940ecaa0eb766b503819d1a51b7b512d9c22
thread:issue:40854,issue #40854: Build pytorch with cuda11,stale/low urgency,2,stale/low urgency,1,0,40854,,,,0,de1a9c34a517c52c57066e4027c902197347be276cdaa3e682c56e99d02c5bb7
thread:issue:40859,issue #40859: Option to allow loading state dict with mismatching shapes.,stale/low urgency,2,stale/low urgency,1,0,40859,,,,0,482ee5e85fe7450c4f3c302c44beb0c125430b87739694641ab7d4d89ff7d4cd
thread:issue:40880,issue #40880: Allow TLS to keep distributed autograd context alive,stale/low urgency,2,stale/low urgency,1,0,40880,,,,0,983f0b6602dfb6122ff4a56be58f57477dcd880012075541bdc7a87b714ccdac
thread:issue:40882,issue #40882: JIT fuser result of dropout doesn't fully match eager mode,stale/low urgency,2,stale/low urgency,1,0,40882,,,,0,626bbace9969201d930fc7555182535c2c0eb83361ea5a05c417b14d90fee769
thread:issue:40886,issue #40886: Improve error messaging when using dictionaries as inputs to a trace,stale/low urgency,2,stale/low urgency,1,0,40886,,,,0,27a1373929b8d745aca5ae44fee75d70c09ec95dbfd5061920e45cd7284f0d74
thread:issue:40891,issue #40891: [FR] torch.floor/ceil should support output int dtype,stale/low urgency,2,stale/low urgency,1,0,40891,,,,0,e56eac7b27d6676c3626d7986fa5e58a33fee75574655066c0aa73b88f23a80f
thread:issue:40916,issue #40916: Vectorized torch.eig(),stale/low urgency,2,stale/low urgency,1,0,40916,,,,0,d4f538b8e42a73d7de264393f754bc0d5c7e73eafec510a330a53a518adcc4ff
thread:issue:40936,issue #40936: [RFC] [RPC] Automatic retries of all requests in TensorPipe agent,stale/low urgency,2,stale/low urgency,1,0,40936,,,,0,994a8283534e302161a693e231eedac18e1c9a27a26c52d5722d76c09c6f14c8
thread:issue:40959,issue #40959: Will the model run slower when deployed using libtorch ?,stale/low urgency,2,stale/low urgency,1,0,40959,,,,0,a7ea9bae07ac17bac737e7ef21bc8d2569636a6fc46674520191ee7fb88d1502
thread:issue:40967,issue #40967: Inconsistent behaviour when parameter appears multiple times in parameter list,stale/low urgency,2,stale/low urgency,1,0,40967,,,,0,eb39acbc9e908f9372eaf8b451b6ab2f6b8ec8691a0b70f14cbf3221cf27c7db
thread:issue:40968,issue #40968: Pytorch 1.4 compilation hangs on AMD Epyc,stale/low urgency,2,stale/low urgency,1,0,40968,,,,0,a7914e8054b5c769aad980caee43f0d1fe1b612632dc4ecbc38d3aa8b47376d1
thread:issue:40969,issue #40969: Regarding graphs page on site,stale/low urgency,2,stale/low urgency,1,0,40969,,,,0,5b83f9cd4f9aac79820b7c54148d26436f64c41db6fc3b6e5cb49ba2697a6ade
thread:issue:40971,issue #40971: The values calculated according to the document isn't equal to the values calculated by framework,stale/low urgency,2,stale/low urgency,1,0,40971,,,,0,f726a74febcbe6c538f3a889fe679a47fe78179dabb1722c3167311bbe927d58
thread:issue:40972,issue #40972: len of dataloader when using iterable dataset does not reflect batch size,stale/low urgency,2,stale/low urgency,1,0,40972,,,,0,6bda4f80068885ffbcba817e4d6ba9acefdd1951054862c88124246e715ebeb5
thread:issue:40974,issue #40974: Add a launching script for RPC,stale/low urgency,2,stale/low urgency,1,0,40974,,,,0,5848d0f563e83b7a45d08f375f256ea923cd62469f820692ef8ccc542176a642
thread:issue:40978,issue #40978: [FR] NCCL and bool type,stale/low urgency,2,stale/low urgency,1,0,40978,,,,0,d2d3d2cd904af4cd1024ed0e0b4bf097d347ec159d6ce0657cb85b660234c338
thread:issue:40997,issue #40997: Is this a bug? The values calculated according to the document isn't equal to the values calculated by framework,stale/low urgency,2,stale/low urgency,1,0,40997,,,,0,acf39d946eda6ab8d7690e0d3c72639a14c491bf3eea492d93198f0759defd33
thread:issue:41006,issue #41006: Simplify Adam Optimizer,stale/low urgency,2,stale/low urgency,1,0,41006,,,,0,8e0c5effc22afa2300ca375cffe007e31422cb139e1ba9b552475e285bc3f083
thread:issue:4102,issue #4102: Make the generator tools data model more explicit,stale/low urgency,2,stale/low urgency,1,0,4102,,,,0,fb3fee314d231525c788bb48f57a8843bd0a92b4f6a479be111075d234d169e5
thread:issue:41027,issue #41027: out= resizing (and restriding) behavior is confusing,stale/low urgency,2,stale/low urgency,1,0,41027,,,,0,cdef03e462a19b1e7307d5fe14ec70d00f8ea123855149d78a3b0a6104267e24
thread:issue:41070,issue #41070: Cannot define v_dim in MultiheadAttention,stale/low urgency,2,stale/low urgency,1,0,41070,,,,0,77ff12acb62d76c5814dc22ed738b614e2f2f55a121503bfd8aa81390c021f2d
thread:issue:41099,issue #41099: max_pool2d always compute indices even when it's not required,stale/low urgency,2,stale/low urgency,1,0,41099,,,,0,90821c6585ab0cc7922aca422236e52da87b178570b3f6dcd66481a45ddcb92a
thread:issue:41101,issue #41101: DDP should provide an option for not touching grad of locally unused params,stale/low urgency,2,stale/low urgency,1,0,41101,,,,0,22780a0004c76682853021945e20c067ffa7ce01b23b59969f3fbc959bbce703
thread:issue:41106,issue #41106: Missing Code for Audio Classification Tutorial,stale/low urgency,2,stale/low urgency,1,0,41106,,,,0,a373fe263a137cc72acaf22ad5973dfe245c773b060d1f12167e838679cdc354
thread:issue:41119,issue #41119: TORCH_CUDA_ARCH_LIST deprecation warning,stale/low urgency,2,stale/low urgency,1,0,41119,,,,0,892ce6dcb7a33467b8d261abf12d1bef1bb7d3b405ea663c771c0cd68b3cecec
thread:issue:41135,issue #41135: libtorch cmake should PUBLIC ally advertise it as requiring C++14,stale/low urgency,2,stale/low urgency,1,0,41135,,,,0,43be98d2ac9d6841aebf22ecc6b4e4c3c407e63f4cc9e4b273401110cbefa0c9
thread:issue:41148,issue #41148: Source code for some /_modules/.../.../ not displaying on https://pytorch.org/docs/master/_modules/,stale/low urgency,2,stale/low urgency,1,0,41148,,,,0,9279b0608c63891a35ec0155ddbdf2a740da662b643f9a64084cda800c16b5f9
thread:issue:41149,issue #41149: eager mode quantization should remove qconfig in the (non-leaf module of) quantized model,stale/low urgency,2,stale/low urgency,1,0,41149,,,,0,8a8cd6086b3b71a75ad3f724fa2ec1c19c8012ba31234d2b2c2bd47606dd3228
thread:issue:41162,issue #41162: Advanced indexing gradient is extremely slow when there are many duplicate indices,stale/low urgency,2,stale/low urgency,1,0,41162,,,,0,2ee917e5848f5b0a5ad6f0c94c060849264a19750cbd39a45d3ac98f96924026
thread:issue:41164,issue #41164: Pytorch.js would be nice!,stale/low urgency,2,stale/low urgency,1,0,41164,,,,0,3b44e82d7ccd5e2d81e3c9ed994555be62c50de791611b44947fd15fe2ecb509
thread:issue:41177,issue #41177: The new version of the libtorch become slow,stale/low urgency,2,stale/low urgency,1,0,41177,,,,0,e6cb85d8335fbe04df8c4d12783566424d240924e54cf1776fcce25198de7f67
thread:issue:41178,issue #41178: Failure to build on Power9 due to FXDIV_SOURCE_DIR not being set,stale/low urgency,2,stale/low urgency,1,0,41178,,,,0,b25a7c6184081cef7e26c362f32f11d15a62555d9cd8477e9cbfb6f6f1a5fb47
thread:issue:41187,issue #41187: test_upsampling_not_recompute_scale_factor fails with Eigen/OpenBLAS,stale/low urgency,2,stale/low urgency,1,0,41187,,,,0,8fcad8f8c47af77685a21dd4c6e8840f464980aa8b480cd08e8c7baa82fc31ed
thread:issue:41213,issue #41213: libTorch cpp docs missing for Tensor::item(),stale/low urgency,2,stale/low urgency,1,0,41213,,,,0,d6db16616a8fc4bebc75e5085aa1d4527b149049e311d67225c95f73b5af4121
thread:issue:4123,issue #4123: Use the int64 version of MKL calls,stale/low urgency,2,stale/low urgency,1,0,4123,,,,0,276b9390fdbe47bc213c62bca20026e264a518089736d0ef0a7bda3b0e43d0af
thread:issue:41232,issue #41232: Cleanup git branches,stale/low urgency,2,stale/low urgency,1,0,41232,,,,0,9c7beed4e3af5ed6745b709bd2fdaed3b30bfb55c2745be6baaddb474f5c3f85
thread:issue:41236,"issue #41236: ""out of scope"" variable `cmd` used",stale/low urgency,2,stale/low urgency,1,0,41236,,,,0,c82d986d572ba46ed5bc7173071752f65b75476f02f18f129425e96e9a0ce41c
thread:issue:41244,issue #41244: torch.tan(complex) on CUDA doesn't handle nonfinite values properly,stale/low urgency,2,stale/low urgency,1,0,41244,,,,0,148d386886d697812491248f20ad9eb67b4b2081539ca07aa4acd3d59b7be6b5
thread:issue:41246,issue #41246: torch.abs(complex) is divergent from NumPy on vectorized NaN values,stale/low urgency,2,stale/low urgency,1,0,41246,,,,0,83d575cb215c7c1e7fda1885c203c61e6ff407a6976b820d950ed8b7960f2098
thread:issue:41256,issue #41256: Add nn.functional.swish alias for nn.functional.silu,stale/low urgency,2,stale/low urgency,1,0,41256,,,,0,eabea8d56866380492e4ea5244900a4a8dc099521560640049c9438f04c1bd3d
thread:issue:41282,issue #41282: [docs] Strange arg names of torch.bmm/mm and tensor.bmm/mm,stale/low urgency,2,stale/low urgency,1,0,41282,,,,0,5994ec5bccd5af1446f98633b99d239ff99e57823a772e670b52a73ca4f18db2
thread:issue:41284,issue #41284: add a reparameterized version of inverse Gaussian distribution,stale/low urgency,2,stale/low urgency,1,0,41284,,,,0,ae44e4bdcb6660e4758dc07bd12a9c6e1ad259588b50529a0bdb369732ac944f
thread:issue:41311,issue #41311: ConvTranspose1d layer behaviour under different channel numbers,stale/low urgency,2,stale/low urgency,1,0,41311,,,,0,51669a2f370371670cc090955bec981429ffa9515edaa5295328a4ce025417a8
thread:issue:41322,issue #41322: SGD documentatiuon detail on g_{t+1},stale/low urgency,2,stale/low urgency,1,0,41322,,,,0,65d3ef16a68c41cf22ffc8bb38b0180feefe001d3101e46d8d4bd577f7ee9370
thread:issue:41323,issue #41323: [discussion] Comparison operator chaining,stale/low urgency,2,stale/low urgency,1,0,41323,,,,0,4fa74fdc0e5c5e762fcdb1072240d76bc571fb2aab58197bcd63f8a6a1c7d627
thread:issue:41327,issue #41327: torch.nn.parallel.scatter_gather.gather can't gather outputs that are dataclasses,stale/low urgency,2,stale/low urgency,1,0,41327,,,,0,2403acc171edd1f1545fe6c8773b3e6b85101d5289c066197678d459d3f5d17e
thread:issue:41329,issue #41329: Reference the randomness issue in DataLoader & Dataset documentation.,stale/low urgency,2,stale/low urgency,1,0,41329,,,,0,893638a84f3961e4152e713147cd915a914d7ec8a4c095b6c6a46f14e2a91e2c
thread:issue:41369,issue #41369: [RFC] Device Placement API for RPC,stale/low urgency,2,stale/low urgency,1,0,41369,,,,0,51cdd8d5ddda86a63475bd46ff5ceb34d67e5fdff1ed3842fe5b4ae9864dbbc5
thread:issue:41383,issue #41383: Large overhead (7 microseconds) for PyTorch operation,stale/low urgency,2,stale/low urgency,1,0,41383,,,,0,ac541b697f95ed63d73a67a3c3643f7f7e8e170feb299bbfb035ae063eb7938b
thread:issue:41392,"issue #41392: About the description of the mathematical formula of nn.RNN, I think it is wrong",stale/low urgency,2,stale/low urgency,1,0,41392,,,,0,eaf62e8999780a145a2f0adaa259bc8c66d0dbff8935943531d98ae4d836b493
thread:issue:41399,issue #41399: Bottleneck when publishing the model using flask about 3 times slower.,stale/low urgency,2,stale/low urgency,1,0,41399,,,,0,6a2d1a36ec7c3fde93bb1afe3177d6665850c30698a1cd0c9d6120dfaeee03ac
thread:issue:41428,issue #41428: Add a done() API to torch.futures.Future and ProcessGroup::Work,stale/low urgency,2,stale/low urgency,1,0,41428,,,,0,90bbadc7e04349cec9cb63272bb2ea5f3a95cca7dcee66837ad0b5ce92dc76bb
thread:issue:41478,issue #41478: Implement backend fallback for Tracer,stale/low urgency,2,stale/low urgency,1,0,41478,,,,0,f9dbc3540d72c664680febebdc754e10b30289c4a962bd6c8efdee9ea6ba09b6
thread:issue:4148,issue #4148: ATen explicitly differentiated native function resolution hazard (call is ambiguous),stale/low urgency,2,stale/low urgency,1,0,4148,,,,0,3e4d6a8438724f7043e42e6c28327ebd99133b6928301cd4f23822a2348f26d6
thread:issue:41485,issue #41485: GLOO infiniband with PyTorch,stale/low urgency,2,stale/low urgency,1,0,41485,,,,0,75fcfbd51c24733dc7b30ca392b72d51af9df4d95d21e4cc5f72b3770165d640
thread:issue:41525,issue #41525: c++ indexing vs python,stale/low urgency,2,stale/low urgency,1,0,41525,,,,0,7391905eecb8cddcd7940b7437db2cfc6336ac1f910d999de3ad1fc320691b5c
thread:issue:41528,"issue #41528: torch.nn.functional.grid_sample()is doing bilinear interpolation when the input is 5D, i think the mode should add 'trilinear'",stale/low urgency,2,stale/low urgency,1,0,41528,,,,0,6660274269f9e7790a31b84a8dfa5bf5fcfd3a6a8b143f0fe98404b787bb3dd0
thread:issue:41531,issue #41531: Test failure in test_shared_allgather_nccl: NCCL error in: ../torch/lib/c10d/ProcessGroupNCCL.cpp:537,stale/low urgency,2,stale/low urgency,1,0,41531,,,,0,915731913a175872d6b7c11afc6a1515cfe174dd8f6a8706774e568a719529c5
thread:issue:41561,issue #41561: [RPC] Should we support users _not_ calling rpc.shutdown()?,stale/low urgency,2,stale/low urgency,1,0,41561,,,,0,c3b46d3c8e15fb2159cff0b6c5d1ceda2a631c4f90c915d24d8d2a4fb12cd28a
thread:issue:41571,"issue #41571: torch.cuda.BoolTensor uses 8 bits per element, not 1 bit as reported by element_size()",stale/low urgency,2,stale/low urgency,1,0,41571,,,,0,f7638711d600447ec27e50f02afc7e45f5c9a804b0e4497a7e8b184d18b5c826
thread:issue:4159,issue #4159: Variable outputs of stochastic functions should never require grad,stale/low urgency,2,stale/low urgency,1,0,4159,,,,0,9b613949687e153fd398f68de763f7c78ee6643f729659fa8062523db15efa45
thread:issue:41622,issue #41622: New Feature : A very fast algorithm for computing matrix rank,stale/low urgency,2,stale/low urgency,1,0,41622,,,,0,5d13d2ea880076b6722debbb7a4e10917c7056ecf77b792fa0c29fcb5509389d
thread:issue:41623,issue #41623: Can we have a way to reset a scheduler back to epoch -1,stale/low urgency,2,stale/low urgency,1,0,41623,,,,0,eaf5f014013e142085596ca5b02df7690bb80842f8571a1cd2c41f50249fda96
thread:issue:41639,issue #41639: Error with minimal hogwild test (multiprocessing shared memory),stale/low urgency,2,stale/low urgency,1,0,41639,,,,0,594a31cfa9f6a18d1a75086eedd4aed4bc18768fc2ba1aa26252d72bd986032e
thread:issue:41660,issue #41660: Potential bug in net_printer,stale/low urgency,2,stale/low urgency,1,0,41660,,,,0,b381664bd6f9a90643b2f8a1c8d0a2aafd4b84af6728b722403d507cf953e1d4
thread:issue:41667,issue #41667: [Feature Request] named tensor support for `tensordot`(tensor contract),stale/low urgency,2,stale/low urgency,1,0,41667,,,,0,c25066a1b9215c745ddf74fd26bcb1ef7a7cf60c2f7c5a459de9e2f0bcf88f1e
thread:issue:41673,"issue #41673: ../caffe2/perfkernels/common_avx2.cc:17:2: error: #error ( ""You found a build system error: __AVX2__ is defined (via e.g. -mavx2) "" ""but CAFFE2_PERF_WITH_AVX2 is not defined."");",stale/low urgency,2,stale/low urgency,1,0,41673,,,,0,b05804cb83739dc2cc6b8d7bea566ff339711b78a4c5d8d03087997a3b257f31
thread:issue:41681,issue #41681: TypeError: cannot create 'generator' instances,stale/low urgency,2,stale/low urgency,1,0,41681,,,,0,a34fd415faa1486ed0825b8bce85813d9aeb87dbb7d0bf3f9d326bf23053ac4e
thread:issue:41694,issue #41694: Torch.multiprocessing.spawn can deadlock,stale/low urgency,2,stale/low urgency,1,0,41694,,,,0,1ba9f4ee39e5b7ad5cf850c6a394f5feb0e07959ef79dae5cf196b83f93f189b
thread:issue:41797,issue #41797: [C++] Libtorch error in Release build only,stale/low urgency,2,stale/low urgency,1,0,41797,,,,0,d2a849c17aa641a1c8c3fe1cf1b0e57b46ecce0555c3aebb568d5dc4c0524128
thread:issue:41801,issue #41801: Caffe2 GPU Inference not working on pytorch nightly version,stale/low urgency,2,stale/low urgency,1,0,41801,,,,0,11ed6fe9da51bd933d716228574f962e8dc4c027e8dc2e2770a1fa3121441dcc
thread:issue:41816,issue #41816: install path can set prefix install dir,stale/low urgency,2,stale/low urgency,1,0,41816,,,,0,d33cd6c6ec63f9dccc099ae0d168cb8952f5600224403228057df97222978496
thread:issue:41823,issue #41823: bitwise_or / bitwise_and /... reductions across dim (+ multidim / keepdim for consistency),stale/low urgency,2,stale/low urgency,1,0,41823,,,,0,6b13b93c43a604d0c5a0bd4af1aee5d8b4371c18fcfcdf3274b4a11c47328b44
thread:issue:41839,"issue #41839: [Feature Request] Add `to`, `cpu`, and `cuda` method to optimizer",stale/low urgency,2,stale/low urgency,1,0,41839,,,,0,3af126b1baf4f4e70ba258caee2b1f6ecbd2195b8478d54d86fb4919d6d21995
thread:issue:41844,issue #41844: test_bottleneck_cuda fails on Power,stale/low urgency,2,stale/low urgency,1,0,41844,,,,0,56899ba8f993022ec982bc8db2471540bd5012d82d2e5dcdbbdef8c931314d85
thread:issue:41852,issue #41852: Difference in inference time between CUDA 10.0 & 10.2,stale/low urgency,2,stale/low urgency,1,0,41852,,,,0,89f351d41ff40b13f0153bdd037ba4ae03add2e0f1e00fbac24f52c5a0b440d6
thread:issue:4186,issue #4186: Feature Request: CPU performance optimization with MKL-DNN,stale/low urgency,2,stale/low urgency,1,0,4186,,,,0,d5414cf75801b47ea2a3f454c81aa72df907c3c457852e7c707330672eb9a79a
thread:issue:41903,issue #41903: [C++] adding type checking or type casting to torch::PackedTensorAccessor indexing,stale/low urgency,2,stale/low urgency,1,0,41903,,,,0,32da24a3c5328af1a1415eaa1335bbb834aff98a75118ca6d823c6a98a746d82
thread:issue:41904,issue #41904: Inconsistencies on ModelLayer,stale/low urgency,2,stale/low urgency,1,0,41904,,,,0,ba7246f577ebb9cf561c3ae5a349377de657cb9e6d08bfb5b74136253ad8f78c
thread:issue:41915,issue #41915: Automating the various Installation processes for Linux,stale/low urgency,2,stale/low urgency,1,0,41915,,,,0,3d8374c2b4503d8808f9b27a1ec7c207c53f53283242de692d3f63d653fca759
thread:issue:41916,issue #41916: ASAN build broken when using USE_ASAN=1,stale/low urgency,2,stale/low urgency,1,0,41916,,,,0,5155cbd95bb56372741e6c49a46433be0c0f7b6a0a97f6f5d7f93ac25035cc70
thread:issue:41918,issue #41918: ASAN build breaks when using third_party/protobuf,stale/low urgency,2,stale/low urgency,1,0,41918,,,,0,1f47e7b4994aafddfa4b2d7abb61dabbd0754d4aa27e742222760890e30aeb7f
thread:issue:41920,issue #41920: Compilation on Power fails with clang due to vec_xl,stale/low urgency,2,stale/low urgency,1,0,41920,,,,0,f902bf9a81a30a31aa9749f34420ec3e6c260318bdc1b2fc00dc33f444cd4ad7
thread:issue:41921,issue #41921: Failure to compile Eigen on Power with clang,stale/low urgency,2,stale/low urgency,1,0,41921,,,,0,86520cacc0ab90f4d9b1c4ef818a5b534f3ec4863d8d3d44df152b6bd83fc00c
thread:issue:41926,issue #41926: torch.exp() cannot be modified by an inplace operation,stale/low urgency,2,stale/low urgency,1,0,41926,,,,0,2420d6e34419b2e0dab217f38904639626f9ddede510a08f771d52cfd7e655b5
thread:issue:41938,issue #41938: Installing pytorch from source on Power9 (PPC64LE) + CUDA 10.2 + RHLE7,stale/low urgency,2,stale/low urgency,1,0,41938,,,,0,e3099d194bf05e174d906285c14a1da0ecef5ebb9b6e3221f7a37c2b9a23417e
thread:issue:41955,issue #41955: can not use nn::Functional(torch::softmax(-1)) in Sequential,stale/low urgency,2,stale/low urgency,1,0,41955,,,,0,3875714dcc0c29f5fa70ed894ac366aef8f2194d4992ab52b29b39d964e858aa
thread:issue:41970,issue #41970: torch.random.fork tries to initialize cuda even when no cuda devices are available,stale/low urgency,2,stale/low urgency,1,0,41970,,,,0,fa4867eb55eebf723f10c07eecbbc5d41e871df2804f06dbcb4453955d25a1dc
thread:issue:41999,issue #41999: PyBind11 submodule required even when pybind11_PREFER_third_party=OFF,stale/low urgency,2,stale/low urgency,1,0,41999,,,,0,dd69b5dbdbbdedccaf18c7e146b2eb10ec07a0271c5eb0b9592f4d63e6829e5e
thread:issue:42012,issue #42012: EMNIST looks different to MNIST,stale/low urgency,2,stale/low urgency,1,0,42012,,,,0,5e105ab58f7ff01dc8c5ccc15727b11bd58c46f83c6407085af4c96d820e834e
thread:issue:42049,"issue #42049: Code hangs when using `set_start_method('spawn', force=True)` in `torch.multiprocessing.pool`",stale/low urgency,2,stale/low urgency,1,0,42049,,,,0,67176127be045a3de088f91706ef3545191992b4907474e9e2b1ceaf1eecc36a
thread:issue:42053,issue #42053: Function request: np.copy (alias of clone?),stale/low urgency,2,stale/low urgency,1,0,42053,,,,0,fe030a5c7dfe78aeba07c5a0607915cedbf154ce18159d37082ab38d05635510
thread:issue:42074,issue #42074: Memory usage of torch.nn.functional.interpolate increased with v1.5.0 when run on numpy input,stale/low urgency,2,stale/low urgency,1,0,42074,,,,0,f2f73b38b8680606341f40ea05552674d94846657284f5d5f6a7b113134cf8c6
thread:issue:42080,issue #42080: To have single cuda context across multiple processes,stale/low urgency,2,stale/low urgency,1,0,42080,,,,0,c17254cf5240b0ac68d0442776c3d673da794d1aeea8af38669120d100615c44
thread:issue:42094,issue #42094: Setting threads number to the number of default by torch.set_num_threads is faster than not setting it,stale/low urgency,2,stale/low urgency,1,0,42094,,,,0,b284604befedf6390ada48e8c0c46b6d99fb1fbc3a2bc476eebfb9deeabe816b
thread:issue:42095,issue #42095: [RFC] Add Windows support to torch.distributed package,stale/low urgency,2,stale/low urgency,1,0,42095,,,,0,4837644a99c435e8a24c0988bd6edf043399ba527d3f64be15a41f20ecaa2b97
thread:issue:42112,issue #42112: Function names in title should not be capitalized in documentation,stale/low urgency,2,stale/low urgency,1,0,42112,,,,0,cf76eae1efb6f39820705343f702efd29280e3ab09c44d0f6818026cbf98d968
thread:issue:42155,issue #42155: BusError memory-mapped tensor,stale/low urgency,2,stale/low urgency,1,0,42155,,,,0,d877171f2a4c6f22fd8207fd40ddd1d494b3c7f3d5f7a75282e3c991df593744
thread:issue:42156,"issue #42156: RuntimeError: test_autograd failed! Build success, test failed on IBM POWER9",stale/low urgency,2,stale/low urgency,1,0,42156,,,,0,2fef04b7cba773a6232cfc84f4ba229c7369529018a292281d93fe26622fdb73
thread:issue:42159,issue #42159: [docs] Add example of interop with DLPack created in plain C,stale/low urgency,2,stale/low urgency,1,0,42159,,,,0,33aa416f39699cb6575172de99f0692d3ca20fc4497dacf901b43dd5e09e0f5f
thread:issue:42173,issue #42173: Exclusive flag in torch.cumsum,stale/low urgency,2,stale/low urgency,1,0,42173,,,,0,fe8b3a470d97afddf1a458e20f136b914e545ea387b572e8a994bb72c31b1f59
thread:issue:42177,issue #42177: record_function no longer appears in the docs,stale/low urgency,2,stale/low urgency,1,0,42177,,,,0,506377168e9c228bc9e9e7a641242da91bb80d0acc1ce7f391f0b835b4984c22
thread:issue:42183,issue #42183: Skip all reduce globally unused parameters in DDP,stale/low urgency,2,stale/low urgency,1,0,42183,,,,0,a365242930e0e6bbdeca4c9a4eb2c6cc790603e01be56261bd5695c2effb27b8
thread:issue:42188,issue #42188: Eliminate warning when cloning a tensor using `torch.tensor(x)`,stale/low urgency,2,stale/low urgency,1,0,42188,,,,0,50aa1dde709b77ac0f9b5dfb69f4353be66f840b9a33eee9224573c9563cc548
thread:issue:42223,issue #42223: Implementing packbits,stale/low urgency,2,stale/low urgency,1,0,42223,,,,0,aaedc4c8080ec7b961e490475de2e197c2d2b9021c13902f2004b1fe58e3d32d
thread:issue:42228,issue #42228: [docs] Bitwise ops miss arguments,stale/low urgency,2,stale/low urgency,1,0,42228,,,,0,e89ccb6e31c1c16189a6577ef493f3af376f82470238c9b91e45e63a02cadb69
thread:issue:42301,issue #42301: Is it possible that one-version pytorch supports two gpu (GTX TITAN X and TITAN) using at the same time?,stale/low urgency,2,stale/low urgency,1,0,42301,,,,0,8091392461e011e9ddcf1ad12da2e2f54fa2a42c75414944f0aee8cb57f25f7f
thread:issue:42303,issue #42303: frame_length reference in pytorch.stft documentation,stale/low urgency,2,stale/low urgency,1,0,42303,,,,0,fd208a6caf10a055c2a684213b1d6b0608d99bc2e4c2133cc4ebc825194532b8
thread:issue:42311,issue #42311: Inconsistent results when trying to enable Tensor Cores on NVIDIA T4,stale/low urgency,2,stale/low urgency,1,0,42311,,,,0,820efbe187f75205f1a103fef298a3a6875f3fb9bc6a8e0db09fcef7e1f68673
thread:issue:42332,issue #42332: RuntimeError: ProcessGroupNCCL does not support recv (torch-nightly),stale/low urgency,2,stale/low urgency,1,0,42332,,,,0,1fd65c03b55524c53c1af511783be4f576533122c61c59e4dad3ad2aa4f4d472
thread:issue:42336,issue #42336: Audit quantization functions to ensure proper argument sizes,stale/low urgency,2,stale/low urgency,1,0,42336,,,,0,2e1b8ca01d39cbae7f0e873cd9122e3f7ee9dd7341ac1af1069dc6b3f4b86c6a
thread:issue:42345,issue #42345: Generate unique `worker_id` for each node in the RPC framework for the store.,stale/low urgency,2,stale/low urgency,1,0,42345,,,,0,521cecf2df3d44b34690a58db5126c2813498a05c97f90bf97a60bffd4abb854
thread:issue:42353,issue #42353: Error while trying to install pytorch on pycharm,stale/low urgency,2,stale/low urgency,1,0,42353,,,,0,3f5e62180210b693163d3b8809549126308b3e9368356017a0d489ba389eaa92
thread:issue:42362,issue #42362: libpytorch_jni is not provided in conda builds,stale/low urgency,2,stale/low urgency,1,0,42362,,,,0,5f3fa68dd087ed3f17ca007ffd1f486e3e33aafb6dd54da39c83a9e94760be74
thread:issue:42366,issue #42366: Stop implementing JIT stack as a std::vector,stale/low urgency,2,stale/low urgency,1,0,42366,,,,0,6338d3abb00b18f9672344accae6c64b0bdd4461340c6b7de755babc32057d81
thread:issue:42402,issue #42402: Reading data speed slower than tensorflow,stale/low urgency,2,stale/low urgency,1,0,42402,,,,0,667889f28df4de29b0b1e9d91ade0efdd3ece217811d99b79361467bf4bcc663
thread:issue:42407,issue #42407: torch.distributions.multinomial.Multinomial cannot be used in batch,stale/low urgency,2,stale/low urgency,1,0,42407,,,,0,c8fafe2d6684dc539564629c3df4f316c0a5455968d33f594f680a72a0915c06
thread:issue:4241,issue #4241: single-gpu works but multi-gpu hangs,stale/low urgency,2,stale/low urgency,1,0,4241,,,,0,be541a53702624d35b0e90f184e6e6e40ec65706174603fa17fba7c2b5bf625a
thread:issue:42428,issue #42428: KeyError: xxxxxxxxxx when calling optimizer.state_dict(),stale/low urgency,2,stale/low urgency,1,0,42428,,,,0,fb44de01324313f64c64e720ce4f8284568374173808d49eb58f390ac0c86a8f
thread:issue:42439,issue #42439: How to use libtorch on Jetson TX2,stale/low urgency,2,stale/low urgency,1,0,42439,,,,0,e1c2d16d51ff262eca632a0acd6bafe26cef3b9ee9e96579bd85271c577b9579
thread:issue:42440,issue #42440: [feature request].Time Series Algorithms,stale/low urgency,2,stale/low urgency,1,0,42440,,,,0,c74b9954f05aec6e7be4640573c667f452fb1ce9796145adb081990ccfa17bbd
thread:issue:42444,issue #42444: torch.random.randperm stuck in multiprocess,stale/low urgency,2,stale/low urgency,1,0,42444,,,,0,1ee0705970950a24ef164909b6db62253bbb7b370f7614d040e26d398140cbed
thread:issue:42447,issue #42447: OneCycleLR argument `pct_start` used as a proportion not percentage,stale/low urgency,2,stale/low urgency,1,0,42447,,,,0,2e5ddd09f4b564ce1898bc669d8a41d49953292e55c207e137e78672d37d3eaa
thread:issue:42459,issue #42459: [JIT] Runtime error when backpropagating through input gradient,stale/low urgency,2,stale/low urgency,1,0,42459,,,,0,b33faa33f67c5597ca37ec0b4b8ee046138aa9e27a416425c57ebd841c2df53d
thread:issue:42460,issue #42460: Training with AMP gives overflow warning,stale/low urgency,2,stale/low urgency,1,0,42460,,,,0,641596b74acfc507df801019e17231137d86bf7997bd4749a4440e31c3899f83
thread:issue:4247,issue #4247: Feature request: sparse matrix max(axis),stale/low urgency,2,stale/low urgency,1,0,4247,,,,0,4ca243e3aa9c6b6d9005fa967c7282ec2a35e07a2edda436412706dd65bcf445
thread:issue:42478,issue #42478: Using hooks with `DataParallel` gets `autograd` error,stale/low urgency,2,stale/low urgency,1,0,42478,,,,0,298aeebb535ce6a615146ac9a10c5386d36a39ae2eab1536c0bb5f2e1cd9c5cd
thread:issue:42498,issue #42498: pytorch tests failed,stale/low urgency,2,stale/low urgency,1,0,42498,,,,0,6817f68f4c8b33da8df47d32ad46b98c9ab6ec3d8e6bfbe0752887473446e4a4
thread:issue:42526,issue #42526: Python3 segfaults or doesn't load torch after installation,stale/low urgency,2,stale/low urgency,1,0,42526,,,,0,a91651fd2dcc7832247580ae7fcc4452de81ae190e4bd0f435f10f9244746f6c
thread:issue:42529,issue #42529: renorm dim argument is extremely confusing,stale/low urgency,2,stale/low urgency,1,0,42529,,,,0,93a952e373c7c51a1c4dd6e7c1e3ece371ec63fea030c0d8c42ec8f45d9e5934
thread:issue:42539,issue #42539: Docs for Tensor.copy_ non_blocking should mention pinned memory,stale/low urgency,2,stale/low urgency,1,0,42539,,,,0,0876f7fc2ce2c454028a047f2cafc90563c4939b72f9f9be3d3f7fba949c5ceb
thread:issue:42544,issue #42544: torch rpc lose device information between rpc calls,stale/low urgency,2,stale/low urgency,1,0,42544,,,,0,608a3170894536a1255bf9300f4f1da0fd291974d2c615663931304f894ca536
thread:issue:42545,issue #42545: CuDNN RNN bindings are basically all deprecated in cudnn 8,stale/low urgency,2,stale/low urgency,1,0,42545,,,,0,e5831d4dbd0ea4976e22624731e37571747fbd049d4b28d4b640a5fd2e960a2f
thread:issue:42594,issue #42594: support LSTM for quantization aware training,stale/low urgency,2,stale/low urgency,1,0,42594,,,,0,3128fc4f8adf49eb50fddb066a3702b6c4802bcec8f939c63458b8d89530c41e
thread:issue:42597,issue #42597: Logsigmoid as chaining operation,stale/low urgency,2,stale/low urgency,1,0,42597,,,,0,31abfa50a3caeaa389ec03619185883ff0dd7f7aefe21caa18b2b833a39497e5
thread:issue:42625,issue #42625: F.grid_sample produces weird results on single-pixel images,stale/low urgency,2,stale/low urgency,1,0,42625,,,,0,42b0bb71dfbeb7fd2628e90bce01adde95f3d55e3f892cebaf0ee1b9dc48d2d7
thread:issue:42654,issue #42654: RuntimeError: each element in list of batch should be of equal size,stale/low urgency,2,stale/low urgency,1,0,42654,,,,0,178f19037bcfae7e2567f6f9e7e019141d572a139693ff3f9fcef269e360b112
thread:issue:42655,"issue #42655: When one distributed test fails in CI, the next one can fail spuriously",stale/low urgency,2,stale/low urgency,1,0,42655,,,,0,7467be5a4d8df136d51b96808f37ead7da7e482ab1cdad8befbd8b274b863071
thread:issue:42663,issue #42663: Why Conv3D is slower than Conv2D when its flops is smaller than Conv2D,stale/low urgency,2,stale/low urgency,1,0,42663,,,,0,7adfc1b99cc3bbd64ad55c0adac5278df23d002fb8e43927bd4774f2b34a9ec9
thread:issue:42665,issue #42665: Optimizer support via Libtorch C++ on Android,stale/low urgency,2,stale/low urgency,1,0,42665,,,,0,3678ff3ccf87b1c2ca604bfdf526f7489b1b6e8043493c293a0f72747bc1c385
thread:issue:42671,issue #42671: [discussion] Support __round__ magic,stale/low urgency,2,stale/low urgency,1,0,42671,,,,0,03fbd221d9656a25745bca78991bc362f62b21b380f6a1c3764a4c1b208456ca
thread:issue:42689,issue #42689: How can I specify NumPy while building PyTorch?,stale/low urgency,2,stale/low urgency,1,0,42689,,,,0,6b69a308cc36ece20f4840e6352ab521bb01bb85b21c6513ed34f8eefc8ef4ad
thread:issue:42705,issue #42705: torch.distributed.rpc package not work well with generator and lambda,stale/low urgency,2,stale/low urgency,1,0,42705,,,,0,e76fc057b427439616788d726a3fc229840cdc13ff08d5dae937eb19ef38adb1
thread:issue:42729,issue #42729: Mismatch in docs and behavior of align_corners for nn.functional.interpolate,stale/low urgency,2,stale/low urgency,1,0,42729,,,,0,7a4c0e7069ecf05ba65dec9ccb533f0651e0f023cef78142b3740ff6996caea8
thread:issue:42774,issue #42774: Missing API reference for SWALR and AveragedModel,stale/low urgency,2,stale/low urgency,1,0,42774,,,,0,3de1d2bf3f14a0f2d3f906681be7967357aa76577ce80cd5db7b977d1e2108f4
thread:issue:42791,issue #42791: Build fail on Ubuntu 18.04: caffe2/CMakeFiles/init_test.dir/build.make:106: recipe for target 'bin/init_test' failed,stale/low urgency,2,stale/low urgency,1,0,42791,,,,0,ce8acfd72b5728118f57a964f0b2e8fa037fcd8d211da191206249587392406e
thread:issue:42800,issue #42800: RuntimeError: cublas runtime error : the GPU program failed to execute at /pytorch/aten/src/THC/THCBlas.cu:27 for spectral norm,stale/low urgency,2,stale/low urgency,1,0,42800,,,,0,20ad7e99584a005ac280041b2d2fdf4f16774d84af04e6f63e7521156c28fd08
thread:issue:42805,issue #42805: Unbuffered operation,stale/low urgency,2,stale/low urgency,1,0,42805,,,,0,5efcc32f16d1bd7e4f2e42c8f1663cb0de1b3538014a1abe93cb27027957bc96
thread:issue:42807,issue #42807: build error redefinition of ‘struct c10::complex<T>’ with release v1.6.0,stale/low urgency,2,stale/low urgency,1,0,42807,,,,0,b9313cac2bd4bcf306ee7e685841d3696e2543816288fe7df00ec398f92252a4
thread:issue:42818,issue #42818: Can't registered boxed kernel for operator that doesn't support boxing,stale/low urgency,2,stale/low urgency,1,0,42818,,,,0,12e15af8789b823e23122bc5734f13596a72208f351ec6c111ddff4771231e1b
thread:issue:42829,issue #42829: torch/library.h doc rendering issue,stale/low urgency,2,stale/low urgency,1,0,42829,,,,0,f3dd52772b49d0b540db53c9c657d9fb58dd6caab07947ec8a69b9648f4df9bd
thread:issue:42830,issue #42830: Undocumented parameters in the Variables section of the recurrent bidirectional layers regarding the backward pass,stale/low urgency,2,stale/low urgency,1,0,42830,,,,0,523aac49163132ab6a6fcbd219a6a6837d970e3bfc8e7d663665b6737ec5a44f
thread:issue:42838,"issue #42838: RuntimeError: ""triangular_solve_cuda"" not implemented for 'Half'",stale/low urgency,2,stale/low urgency,1,0,42838,,,,0,890e727b53ebf41aef8dab65b3894267d034654e7e150c910c67846192e1fba4
thread:issue:42848,issue #42848: Add support for user defined types in serialization in libtorch,stale/low urgency,2,stale/low urgency,1,0,42848,,,,0,eed48edde8a494f24f1dbf62355d7d078005b7b5cb07aee7742f33434701a676
thread:issue:42858,"issue #42858: prelu_backward, hardshrink_backward shouldn't be a method",stale/low urgency,2,stale/low urgency,1,0,42858,,,,0,1f938becaaa066c1a934c221cd63ebce8be3adb225564ac7de0f2d581f93b4a8
thread:issue:42861,issue #42861: Can not get pytorch working with tensorboard,stale/low urgency,2,stale/low urgency,1,0,42861,,,,0,332e61c83ecfafeeb980835c4cec974009308bbb4aecf3110d0bd31b735248eb
thread:issue:42863,issue #42863: TensorOptions extensibility has rusted shut,stale/low urgency,2,stale/low urgency,1,0,42863,,,,0,99b6121a3d644b7eddc21e54e796acb43a7f9511b3cd21d7a0447b8d2c399bde
thread:issue:42879,issue #42879: Tensorpipe requires setting both GLOO_SOCKET_IFNAME and TP_SOCKET_IFNAME,stale/low urgency,2,stale/low urgency,1,0,42879,,,,0,72289b6c52c85cd6713bfaa0a3df7f3b0ecf09ef61aae761993aa90287fd004a
thread:issue:42882,issue #42882: Support for defining/saving custom operators with dynamic schema,stale/low urgency,2,stale/low urgency,1,0,42882,,,,0,cb0406d8c2682a4916e09c2ef17303893baaeed6789184c3cd6dfa67515ae822
thread:issue:42902,issue #42902: Avoid dynamic isCustomClassRegistered() checks in kernel call paths,stale/low urgency,2,stale/low urgency,1,0,42902,,,,0,b7189c31272c60f1c578b00ed298afa37cd7e70d8503f131afa71b6a5654646a
thread:issue:42912,"issue #42912: Add typing information into variable declarations for JIT script "".code"" output",stale/low urgency,2,stale/low urgency,1,0,42912,,,,0,f606ec7b6bf2b7617e138407b225bcc5ea4b7cbbf1d02cb50e2b2cfe768fe78f
thread:issue:42916,issue #42916: Build problems caffe2 -- pytorch from source & CUDA 11.0,stale/low urgency,2,stale/low urgency,1,0,42916,,,,0,257f46eaf8151e697063447c395389d9a74189da4be5c349d688561121ed8dc9
thread:issue:42919,issue #42919: Legacy Python2 and early Python3 leftovers,stale/low urgency,2,stale/low urgency,1,0,42919,,,,0,37fdd98e1e1e618410e59340293872d8c2033d25c0c967431a549167dd493b23
thread:issue:42949,issue #42949: Verify TorchBind works with nested class type,stale/low urgency,2,stale/low urgency,1,0,42949,,,,0,103484a3fc79fa0326618907c54ec4fd34b0f441b8efaa704eee43843f9874a3
thread:issue:42950,issue #42950: TorchBind C++ Enum Class,stale/low urgency,2,stale/low urgency,1,0,42950,,,,0,3e9ee7e07c03037c2b1e4d33a58db980b2dd5dc3b82b2b8e20249f7c683599c0
thread:issue:42969,issue #42969: Enable torch.testing._internal typechecks during CI,stale/low urgency,2,stale/low urgency,1,0,42969,,,,0,e294e64c83e3dcccd23745229284abe0d706c9cf0c1fe0daec978f91edbac00b
thread:issue:42971,issue #42971: Enable torch.quantization.fuse_modules typechecks during CI,stale/low urgency,2,stale/low urgency,1,0,42971,,,,0,7df1adc43724aee80e237c0fdc7b11b1093f9a8af546957d74c0597bf3b43045
thread:issue:43030,issue #43030: BUILD_PYTHON switched to OFF at second run of CMake,stale/low urgency,2,stale/low urgency,1,0,43030,,,,0,56e2dfb37a4af70450c9ff3b49ce116c79112e96a708c98952daa1b75a09686a
thread:issue:43033,issue #43033: No vectorization for int8 and uint8,stale/low urgency,2,stale/low urgency,1,0,43033,,,,0,e82566d3f78037254d38bccb28c22bbd368ed2d90685642c8d89b40ff86e982b
thread:issue:43040,issue #43040: undefined symbol: FLAGS_caffe2_keep_on_shrink,stale/low urgency,2,stale/low urgency,1,0,43040,,,,0,1c7ea8c4c9f465cb058b2bbb07fb40ecff3c4db55e5f7dab7da733627ff6ae79
thread:issue:43064,issue #43064: torch.dot throws an error for input tensors of different dtypes,stale/low urgency,2,stale/low urgency,1,0,43064,,,,0,1e5c539692bba7f1599f7739f0a45c76d3976453b040bbdfdb6b7fd02f51a607
thread:issue:43072,issue #43072: LSTM::permute_hidden breaks Liskov substitution principle,stale/low urgency,2,stale/low urgency,1,0,43072,,,,0,a78d25c1082c9db39c24206c0d1e1b8b596490e30fcde669d881aaaa9a25e393
thread:issue:43091,issue #43091: Enable torch.optim typechecks during CI,stale/low urgency,2,stale/low urgency,1,0,43091,,,,0,59af826d93339b5cbf6400555234dc1eabcf4464027abf846faed9268d7280b1
thread:issue:43109,issue #43109: [docs] LR schedulers' get_last_lr missing in docs online,stale/low urgency,2,stale/low urgency,1,0,43109,,,,0,197c03a01d23a8e1d3fff94dc8995dc91965eed3ccec6d473b01570580f0521c
thread:issue:43113,issue #43113: Missing caffe2_pybind11_state in pip install after cmake.,stale/low urgency,2,stale/low urgency,1,0,43113,,,,0,ccea30caae4ec96ed1f7d968aae6e1bd078e36c35b3a1e80903db36d2a340b60
thread:issue:43114,"issue #43114: out kwargs are sometimes inconsistent with returned named tuple field name (and in some cases, cannot be made consistent)",stale/low urgency,2,stale/low urgency,1,0,43114,,,,0,65705a735925c6fc37e40012cc5e60c6c02c2f0fd3a4a5a97c02f4d0e7967755
thread:issue:43120,"issue #43120: When modified the model python file, the pytorch will raise the KeyError of this file",stale/low urgency,2,stale/low urgency,1,0,43120,,,,0,bec310c107374b127f0f97b6bab268ee507b78a8a10e4b2b5042b13a72d13fdd
thread:issue:43123,issue #43123: Add CutMix transform to torchvision.transforms,stale/low urgency,2,stale/low urgency,1,0,43123,,,,0,d65a166d17872894cdc21e7a196808d6d7390dcc226f5ad2c9964e175453ce99
thread:issue:43140,"issue #43140: Pytorch ""Import torch"" lead to ""core dump""",stale/low urgency,2,stale/low urgency,1,0,43140,,,,0,3eed6288e53f340c09239e0921b4d5ae89ff50b95763144b6a9a23c70b8540f9
thread:issue:43195,issue #43195: one_hot tensors are channels_last but marked as contiguous,stale/low urgency,2,stale/low urgency,1,0,43195,,,,0,de98bfdd315049d50179fb89a48e8f7740ec5ca64a32e50af503cce148346f97
thread:issue:43238,issue #43238: Expose counters for torch.distributed to report communication overhead within a block.,stale/low urgency,2,stale/low urgency,1,0,43238,,,,0,2fe0e7fae2d983283fd1d76a29b8ba4448e42b5a49fa4544719c91feb07ad449
thread:issue:43245,issue #43245: mp.spawn 'args' are not clear,stale/low urgency,2,stale/low urgency,1,0,43245,,,,0,d28078a0fcaed23b3809abcba2571839ba3146edaae80b1b27a2eb9e16e2b9ba
thread:issue:43250,issue #43250: Support for Multi-Categorical in torch.distributions,stale/low urgency,2,stale/low urgency,1,0,43250,,,,0,7f360172bd9b02d1f2c5fa2f6178dd55608cedbed4e28ec4b0fba2459274abed
thread:issue:43253,issue #43253: Reduce number of stack frames used up by dispatcher,stale/low urgency,2,stale/low urgency,1,0,43253,,,,0,004ef8b2978395d1f62ac8c7ee0190087899079934d9792ac90dab41c5ac86b0
thread:issue:43255,issue #43255: Adding entropy function analogous to SciPy,stale/low urgency,2,stale/low urgency,1,0,43255,,,,0,32201f770bb951ea1caabc582ce4e6a94f90cdab8da6dc3947e99b0812b7779a
thread:issue:43259,issue #43259: RuntimeError: Expected to have finished reduction in the prior iteration before starting a new one.,stale/low urgency,2,stale/low urgency,1,0,43259,,,,0,5cf815659049d11fb513bbe6ecda798d3861cc2f4733c5435948a817e5bdb021
thread:issue:43281,issue #43281: Cannot find CUDA devices when the machine stays idle for a while,stale/low urgency,2,stale/low urgency,1,0,43281,,,,0,c6a1a97fe86da8d06df00931704899142d63d6cd25d36c5b4277460cb3a1ed55
thread:issue:43325,issue #43325: CUDA error: an illegal memory access was encountered (laugh_kernel at ...../cuda/CUDALoops.cuh:112),stale/low urgency,2,stale/low urgency,1,0,43325,,,,0,30fa9715592b856c878a6e3d13b106f85fc48a27af02beaa327f483befa9d181
thread:issue:43347,issue #43347: backward for dense+sparse does not work,stale/low urgency,2,stale/low urgency,1,0,43347,,,,0,beb01b1207470599011900e5ac05c2c818815bfe0baac8b20ea2a3d030af76ee
thread:issue:43352,"issue #43352: Remove warning, and update documentation.",stale/low urgency,2,stale/low urgency,1,0,43352,,,,0,7146b3805700b5607e593f4e6dcf67f743b78d2e4bd94b3636fcbc7dc0644f39
thread:issue:43385,issue #43385: OffsetCalculator.cuh(and THCIntegerDivider.cuh) should be available with PyTorch cpu-only binaries,stale/low urgency,2,stale/low urgency,1,0,43385,,,,0,caacec4139571b72201897849554860cc8d427949f21976f29c1cd4548ce30a6
thread:issue:43402,issue #43402: ModuleDict does not preserve order of initializing dictionary,stale/low urgency,2,stale/low urgency,1,0,43402,,,,0,0c5815258d00f74bdd591ca72953a59880ea2f8db33ba5168e887f29c385c365
thread:issue:43411,issue #43411: Add support of random state generator objects to nn.init module,stale/low urgency,2,stale/low urgency,1,0,43411,,,,0,ceaf8d61ba9d7ac3f5eae336c1ea640c9aec791cda354521c6895ca9e335b266
thread:issue:43429,issue #43429: Inference performance regression caused by hacky_wrapper_for_legacy_signatures,stale/low urgency,2,stale/low urgency,1,0,43429,,,,0,a3e8173d828c96dcf7b0f8fcb7024211e2661ca5eac15cdf8d678f3533272936
thread:issue:43467,issue #43467: Data loader struct pack issue(overflow)?,stale/low urgency,2,stale/low urgency,1,0,43467,,,,0,ae5ba4becbe8a3acb73ccec12529a994e0bd4bd1aac19b377e4fcbea049203f0
thread:issue:43486,issue #43486: scatter not place inputs into every device,stale/low urgency,2,stale/low urgency,1,0,43486,,,,0,f03885563f60c5731d342a65e7f3a01d59066861e36ec36130ab9b63ba190741
thread:issue:43490,issue #43490: Implement a set_printoptions method in libtorch,stale/low urgency,2,stale/low urgency,1,0,43490,,,,0,5f2effa5a2237ab42da24d894a92a03c65c31eb8f58f31b071b482943abe880f
thread:issue:43503,issue #43503: torch.quantization.quantize_dynamic document refers `module` as a parameter,stale/low urgency,2,stale/low urgency,1,0,43503,,,,0,246e1d626df13c889f8c784e4dd948f42930122bccaf88ce90ac3ac274d70d28
thread:issue:43510,issue #43510: Make busy waiting time in DDP (kSynchronizeBusyWaitMillis) a configurable environment variable,stale/low urgency,2,stale/low urgency,1,0,43510,,,,0,febfea1d02649f8b11b90e45291f5c0656de407dfa6a748e541915113ccd6b75
thread:issue:43536,"issue #43536: How do I debug ""RuntimeError: trying to initialize the default process group twice!""",stale/low urgency,2,stale/low urgency,1,0,43536,,,,0,51b1ff19030aff63b78bf4ea7fcfcdf95ca295aba112c6ce114b2254c272053e
thread:issue:43544,issue #43544: AMP much worse performance with groupped Conv2d than fp32,stale/low urgency,2,stale/low urgency,1,0,43544,,,,0,2e4e2f99dd482585c5190037b7c6d267e28a8deb4595105dd4716c73d327d2b7
thread:issue:43546,"issue #43546: NCCL WARN Your program may be hanging, this may be caused by a collective mismatch around rank 2. Please check your collective calls at and around this rank. You can use NCCL_DEBUG=INFO and NCCL_DEBUG_SUBSYS=COLL to see the collective logs",stale/low urgency,2,stale/low urgency,1,0,43546,,,,0,d155d56903843264368a38bb59dbec75fcb3db28eb91a17bfcaa89c5ff98da4c
thread:issue:43556,issue #43556: FP16 gives NaN loss when using pre-trained model,stale/low urgency,2,stale/low urgency,1,0,43556,,,,0,f6f35261c10713c280825e9ba472db4757d815d5fd882f70c7a8c3eaadd09c01
thread:issue:43561,issue #43561: [RFC] RPC BENCHMARK,stale/low urgency,2,stale/low urgency,1,0,43561,,,,0,4c6b121d038674bf3e9a23f680ce30ee26f2394c61e7ec223611c37fa0d2b05a
thread:issue:43579,"issue #43579: For the same complex dtype and same value, comparing a PyTorch tensor with a NumPy array results in False",stale/low urgency,2,stale/low urgency,1,0,43579,,,,0,b39713cafcd3e116b8acd4290824503dc5c0abd58ae017a245bc1ab786ecf12b
thread:issue:43604,issue #43604: ImportError: libtorch_cpu.so: cannot open shared object file: No such file or directory,stale/low urgency,2,stale/low urgency,1,0,43604,,,,0,c17c2bb7ae5117b50c09bb1e43d39c3e87dcaa336321f582e392f5d2bcafbc80
thread:issue:43617,issue #43617: Type hints from _VariableFunctions and elsewhere clash,stale/low urgency,2,stale/low urgency,1,0,43617,,,,0,6f81ead0fc282c14d265baa42f5e54b23d21c28ac19e94ba1e7f34765b5facc0
thread:issue:43623,issue #43623: [FR] Raise an exception when constructing non-empty index and empty values sparse tensor,stale/low urgency,2,stale/low urgency,1,0,43623,,,,0,8f3940196245121181fec852d1a8a4b1044104741026ee0cd12487d7b5ce7e90
thread:issue:43638,issue #43638: [FR] hub uses default github branch,stale/low urgency,2,stale/low urgency,1,0,43638,,,,0,8afd8ede7c0372a311661018406b19a78952a1b1ea29e5f8d7ed168aea663b02
thread:issue:43646,issue #43646: c10::string_view pybind11 custom type caster.,stale/low urgency,2,stale/low urgency,1,0,43646,,,,0,f0a273106d5436f162e72fbdbb7fb72dfbe0e68749f6d57c11d31ce7aa7e6d21
thread:issue:43659,issue #43659: Exploring supporting some cases of `__bool__` on Proxy,stale/low urgency,2,stale/low urgency,1,0,43659,,,,0,bc9b3b76878cffc9d36f91244eb57b2eeecb11c79919894fc6bc082405c375cf
thread:issue:43663,issue #43663: test_doc_template is not working correctly,stale/low urgency,2,stale/low urgency,1,0,43663,,,,0,19b2dc04d90639c8e665c7f5a267010e7e2e9d2c047b2b2dd2e9632367253947
thread:issue:43672,issue #43672: Make `torch.Generator` picklable,stale/low urgency,2,stale/low urgency,1,0,43672,,,,0,963d7d90a02c9f48728239e1fca5d1e180876555f54b716ca781bec625bd33f9
thread:issue:43677,"issue #43677: Some largeCUDATensorTest fails with OOM when running with the entire test suit, but not when running standalone",stale/low urgency,2,stale/low urgency,1,0,43677,,,,0,4858b55beec6a5b2f35efe0712f5a2c99da9c83653a618eb24735e7d7e7541a9
thread:issue:43679,issue #43679: [FX] List unpacking is broken,stale/low urgency,2,stale/low urgency,1,0,43679,,,,0,e0d1189c68bf1f9fd861c41c481e2b66c2b15472bf30f35005a670b8b677091e
thread:issue:43685,issue #43685: Syncbatchnorm and DDP,stale/low urgency,2,stale/low urgency,1,0,43685,,,,0,92a19e78b26284d33038aee296e3ce67dfa19a9284ec2b8097356c3a5675f50f
thread:issue:43692,issue #43692: named tensor INTERNAL ASSERT FAILED when indexing with a list.,stale/low urgency,2,stale/low urgency,1,0,43692,,,,0,0430f01a417aaf5a89026a99dafeff36526fd7745b5edae68a8bb321268b9170
thread:issue:43706,issue #43706: checkpoint restore of optimizers changes dtype of Floating-point state,stale/low urgency,2,stale/low urgency,1,0,43706,,,,0,be1c5a7e951f85769fa0626875ba29e9c2a6322c8eafdb9f9d5328f5f48b3c2e
thread:issue:43754,"issue #43754: Using Dataparallel with multi input error. Input and parameter tensors are not at the same device, found input tensor at cpu and parameter tensor at cuda:0",stale/low urgency,2,stale/low urgency,1,0,43754,,,,0,39cbfbae3e3d790934a6a1d12d2a55a1ad616c105ec18fd25056d336627d7cfc
thread:issue:43758,issue #43758: OSError: /lib64/libc.so.6: version `GLIBC_2.14' not found,stale/low urgency,2,stale/low urgency,1,0,43758,,,,0,53b7d0a9a6c9229f6d034347ba6e92ce35db21d582103ce3d0dc123cfbab8a1c
thread:issue:43760,issue #43760: Error : lib/libstdc++.so.6: version `CXXABI_1.3.11 not found,stale/low urgency,2,stale/low urgency,1,0,43760,,,,0,9fb8060662c642223174d167aef10822f3dcaf5ffc4b77a640ef8d7f2f0853df
thread:issue:43815,issue #43815: Buffers are moved to another device by copy and not in-place,stale/low urgency,2,stale/low urgency,1,0,43815,,,,0,cb09487b72d135db720a309ec06900b87e1fa168f759bf387a81d210f2662030
thread:issue:43816,issue #43816: Move randperm() to DistributionTemplates,stale/low urgency,2,stale/low urgency,1,0,43816,,,,0,355419c96bc7560e86ef047b352c3a08ca0d370c4991f0d2ee11ff78f515bc14
thread:issue:43865,"issue #43865: Allow custom kwargs for forward method of nn.TransformerEncoderLayer, nn.TransformerDecoderLayer.",stale/low urgency,2,stale/low urgency,1,0,43865,,,,0,85c962bae30c327497815c6f1bb4aaf7647626471d8ad5eb116a89f07c88ca8f
thread:issue:43876,issue #43876: Insert at specific index/key in nn.Sequential,stale/low urgency,2,stale/low urgency,1,0,43876,,,,0,6e5c27ff47bba12f17af663a9f0bae0ec5a0522f2e4a029cf7af54543915e103
thread:issue:43904,issue #43904: Documentation and `torch.sparse` alias for `torch.bmm` sparse-dense,stale/low urgency,2,stale/low urgency,1,0,43904,,,,0,c70e0e2b1eb5c3d8ab33177e961f36262fb315c7d34a79f6efca60b93b30bca0
thread:issue:43964,issue #43964: Make it so that leading underscore operators are truly private and can be changed without worry for BC,stale/low urgency,2,stale/low urgency,1,0,43964,,,,0,39d046fbc3e0f5f30d55e13c08e4a5095e5367d8dda5d0725919d232e9fd02c8
thread:issue:4400,issue #4400: Installation Optimise For Chinese Users Who Behind the Wall,stale/low urgency,2,stale/low urgency,1,0,4400,,,,0,47a4c01ad8e9641dc010a15a0b2ae9919a72e5b6dc27324c159b944f258d313b
thread:issue:44049,issue #44049: [FX] Can't symbolically trace callsites to ScriptFunction,stale/low urgency,2,stale/low urgency,1,0,44049,,,,0,b90093437015cfc108f297392e1a8d5bfe1593a1c84c6c2d2ece0a8fd7a8355a
thread:issue:4406,issue #4406: DistributedDataParallel doesn't converge well when using MPI,stale/low urgency,2,stale/low urgency,1,0,4406,,,,0,9cdba9f36638a53031fcf476a0cb26d3503990f9e8bc7fcb8c0ca65b3f739900
thread:issue:44084,issue #44084: [RFC] Manage CUDA Stream in TensorPipe RPC Agent,stale/low urgency,2,stale/low urgency,1,0,44084,,,,0,46167ee996abe70dc48f7f5f1dd8cbf243b28a759ac84b0db08ba7707e1f3efd
thread:issue:44156,issue #44156: CUDA memory leak in multi-processing,stale/low urgency,2,stale/low urgency,1,0,44156,,,,0,d483f66613dd9fc79d969ee6152e8ad4fae45a02c5167cf0c8eec03e32a189e5
thread:issue:44169,issue #44169: Deadlock with RPC and dist.barrier() for TensorPipeAgent and NCCL.,stale/low urgency,2,stale/low urgency,1,0,44169,,,,0,a739f8db1f08ef7ba4f33e2ab2698d4175963a02d37fbe415b9cb8a320056641
thread:issue:44172,issue #44172: build from scratch failed because of the expansion of macro,stale/low urgency,2,stale/low urgency,1,0,44172,,,,0,0b0eb0235e919f6e6761fea4935eadf2c04667fc70b295ece1d71b0d824ccb14
thread:issue:44179,issue #44179: REINFORCEMENT LEARNING (DQN) TUTORIAL is not working anymore due to matplotlib.pyplot.imshow(),stale/low urgency,2,stale/low urgency,1,0,44179,,,,0,e21ae07820bc0823c36046f549b14997412aa75aac4f5fb5c6ff812816c66499
thread:issue:44182,issue #44182: Support keep stride for neg with requires_grad=False,stale/low urgency,2,stale/low urgency,1,0,44182,,,,0,cbb2dfecc75fdf954ab443f1ffd60fa9307a6a354cd9beed555020e396838ef0
thread:issue:44185,issue #44185: caffe2: Python 3 deprecation warnings about inspect.getargspec,stale/low urgency,2,stale/low urgency,1,0,44185,,,,0,b8a42aae07c34de9ea32262ea3c550055d542b36377bc7388a68a6aaa0299fa3
thread:issue:44248,issue #44248: Additional clutter memory allocated on GPU 0 when training on GPU 1+,stale/low urgency,2,stale/low urgency,1,0,44248,,,,0,43d59cf198f8a3c07c00e8506f149559f880423fa3cb5b218de1cd5c5506b34e
thread:issue:44255,issue #44255: Allow to use system zstd,stale/low urgency,2,stale/low urgency,1,0,44255,,,,0,0068073970558f2f6235dd643a84c05a7bbd102532e41990d7df27b87d3a1d31
thread:issue:44274,issue #44274: Slower speeds when using half().,stale/low urgency,2,stale/low urgency,1,0,44274,,,,0,0e9adcc20ea55c67529344fe3adac83ca77676d0a5a0de5ee92b48560aea9e8b
thread:issue:44295,issue #44295: Request: Add PyTorch version to state dicts,stale/low urgency,2,stale/low urgency,1,0,44295,,,,0,841eeb4fd7424bc42c414c2d7a6d4d1ea57b61cdbc4f41af7c19d00fce5d78d4
thread:issue:44313,issue #44313: `Undefined symbol: torch::kNearest` when building App iOS,stale/low urgency,2,stale/low urgency,1,0,44313,,,,0,360c1997926607f46e46271688f3a331aad658d1e405c714204df69f1525738f
thread:issue:44341,issue #44341: Consider cache effects in Timer,stale/low urgency,2,stale/low urgency,1,0,44341,,,,0,8f9687645d902aa707c831e428e2e370d6d09389ef851cbcce186973bd46fad5
thread:issue:44343,issue #44343: Asynchronous Execution on CPU,stale/low urgency,2,stale/low urgency,1,0,44343,,,,0,82897f62df08e3ea722f1357578e70ba87a390a57363ddcc74514eda9d73f106
thread:issue:44362,issue #44362: Libtorch CMakeList.txt config help,stale/low urgency,2,stale/low urgency,1,0,44362,,,,0,b034901ae4f15f43b79c42567197a7f616011aed6031ce84c9dd7faecb9720c4
thread:issue:44403,issue #44403: Investigate SyncBatchNorm cleanup with NCCL Async Error Handling,stale/low urgency,2,stale/low urgency,1,0,44403,,,,0,9c6e4a5413b7e582d2ed23b08754914f566d74dbaaabf0a51c8a8b8419019fb3
thread:issue:44415,issue #44415: Batchnorm2d in PT1.6 couldn't turn off track_running_stats,stale/low urgency,2,stale/low urgency,1,0,44415,,,,0,ac2eae8b578932c275ba601c5f2c526cca37097afe7e7e13dee01890a1831d3c
thread:issue:44421,issue #44421: [tools.codegen] Rename api.legacy_dispatcher to api.native,stale/low urgency,2,stale/low urgency,1,0,44421,,,,0,76aaefc1efeb4d2a8bd570d47323c4ab4b9b303f1851cd72c6ba24e110a53ec1
thread:issue:44489,issue #44489: for_each doesn't support integer input to float output type promotion,stale/low urgency,2,stale/low urgency,1,0,44489,,,,0,2624f6f1dcc0bbe0793a49cca1034b1d391aa6a77c21b7264a17c01135ac52ab
thread:issue:44504,issue #44504: FP16 inference latency after sleeping,stale/low urgency,2,stale/low urgency,1,0,44504,,,,0,7c7082a4f06c188f7dac97c2fedbbc4317615e43001b127348fe58e2d056337e
thread:issue:44528,issue #44528: DataLoader consumes extremely large shared memory (shm) in its initialization.,stale/low urgency,2,stale/low urgency,1,0,44528,,,,0,bb6896729f7dc9a0eefa26e51968e881dd27b533c034d8d9cd3473fb24c61cc4
thread:issue:44544,issue #44544: torch.distributed server back-connects to clients on random ports leading to firewall problems,stale/low urgency,2,stale/low urgency,1,0,44544,,,,0,0696f58c2dcb6f3284bd4c54e130ebf169a677a4c63273df0531eb044d9cd302
thread:issue:44545,issue #44545: too large data in Queue cause dead lock in Multiprocessing,stale/low urgency,2,stale/low urgency,1,0,44545,,,,0,b5f72aaab34ecb026d676ab6e81c1fdb69110d55e7af8a570cdce1c14be97921
thread:issue:44547,issue #44547: floating point exception (core dumped) in training 10^4 steps,stale/low urgency,2,stale/low urgency,1,0,44547,,,,0,f3f86845ee82fb0b64cb22a646cd2b81cbfbd5c359eae65a3ee430d8542c3492
thread:issue:44551,issue #44551: PyTorch 1.6 DataParallel causes CUDNN_STATUS_BAD_PARAM in backward pass,stale/low urgency,2,stale/low urgency,1,0,44551,,,,0,bc4cfabfb4434d22bed0cdf15c92405c94c21718a22aa9c99d2b229a9995b476
thread:issue:44552,issue #44552: torch.norm does not broadcast for dim > 2,stale/low urgency,2,stale/low urgency,1,0,44552,,,,0,76e931457cdcae8394650e8331b962ecf001b302e180e15a75e7399e56621778
thread:issue:44593,issue #44593: Docs include the verbose arg of `CosineAnnealingWarmRestarts` but the actual docstring in code doesn't,stale/low urgency,2,stale/low urgency,1,0,44593,,,,0,67b4ec40b1533e3efc260ef02bed4bd002ad06a74824cfa4acd7547ebcb3b84f
thread:issue:44611,issue #44611: MacOS CPU torch.tan and torch.tanh do not compute some values properly,stale/low urgency,2,stale/low urgency,1,0,44611,,,,0,736b9b85d7da97b19523929495f4424c941e098af2f19ec90af1e5562e1fbf0c
thread:issue:44624,issue #44624: CUDA RuntimeError unrecoverably bricks session,stale/low urgency,2,stale/low urgency,1,0,44624,,,,0,1bbf718fc57f7038cfb0b466c14fb9dc07473592bd6ec64d52a2858c700837ca
thread:issue:44631,issue #44631: Multi-process Dataloader and multi-parameter exceptions,stale/low urgency,2,stale/low urgency,1,0,44631,,,,0,6661163ffaf68a76ebe89b0411bb7c4dba854d5d6018aa16c60518e7b8fec6e9
thread:issue:44657,issue #44657: Use macro to define `DispatchKey::toString` method,stale/low urgency,2,stale/low urgency,1,0,44657,,,,0,a6ec3f3a331bd4e2cef3519a3690c25a843ab160802c956079c25e059137759b
thread:issue:44676,issue #44676: Make quantized::prepack_fp16 op just do prepacking,stale/low urgency,2,stale/low urgency,1,0,44676,,,,0,b934da4fe7001fa0f4c3967fd92008d012983627ff3e2f90bd8a097feb21458f
thread:issue:44704,issue #44704: Error message concerning dtype GRU not identical between devices,stale/low urgency,2,stale/low urgency,1,0,44704,,,,0,997d7dd8f5d3e382821f836df37d5236f2e935191808764b3359bab424d7bc19
thread:issue:44719,issue #44719: 404 Page not found - inference_api.md & management_api.md,stale/low urgency,2,stale/low urgency,1,0,44719,,,,0,48792a64f6083c67548617de8f80ccd184b3577a837191074e4db7a8d2c4c734
thread:issue:44721,issue #44721: Get rid of copy_from,stale/low urgency,2,stale/low urgency,1,0,44721,,,,0,aa10befad2bf6c3ebe3fc9b250ff6d0f4c6b55a5c592afeadd319766ae832c8e
thread:issue:44724,issue #44724: 16-bit input + AMP breaks AdaptiveLogSoftmaxWithLoss,stale/low urgency,2,stale/low urgency,1,0,44724,,,,0,453fd6fe5cb22b8b6ef529bcc93c189808b5e48a5e42dcecc080872825b938f3
thread:issue:44726,issue #44726: Pytorch Installation from source fails (GCC-8.4/CUDA-10.2/RHEL7) and (GCC-7.5/CUDA-10.2/RHEL7),stale/low urgency,2,stale/low urgency,1,0,44726,,,,0,7a3687b4352ee89d2538edd73be8e0d309f4126e1b583a4099737075adfb7b82
thread:issue:44738,issue #44738: [tools.codegen] Remove byte-for-byte compatibility code,stale/low urgency,2,stale/low urgency,1,0,44738,,,,0,97d820612371066ff34bfa8606d994cefa04fefa79131adcb8dcb06e387405f0
thread:issue:4474,issue #4474: [docs] Return value type hint to be specified in docs for torch.is_grad_enabled (i.e. -> bool),stale/low urgency,2,stale/low urgency,1,0,4474,,,,0,a4cf47c6fab2ae877152e590127be3425f363654e2a9ce36f101dec07f3d25d4
thread:issue:44741,issue #44741: cblas_gemv is not being used for gemv on complex on CPU,stale/low urgency,2,stale/low urgency,1,0,44741,,,,0,7529642d8901589a94b6bcfc6abf9ee4e8e8a49ff04aa36ebe3cab1cc37a1a8f
thread:issue:44771,issue #44771: Timed out RRef can still be used in subsequent RPCs,stale/low urgency,2,stale/low urgency,1,0,44771,,,,0,e41dc46688a7a47ef2be372833c97626cad4e76dc33481e5042e1dae87eb450f
thread:issue:44809,issue #44809: Simple functions shouldn't go through dispatcher,stale/low urgency,2,stale/low urgency,1,0,44809,,,,0,83f08bc9a1b18e73f26be63edc386777238667381b6b8b205d911c67de6a1c17
thread:issue:44814,issue #44814: LNK2019 error after static compilation of Libtorch,stale/low urgency,2,stale/low urgency,1,0,44814,,,,0,a1e791117f9e3c140982ec6d3319bc1bb8840f54442731098ee418874aca78e4
thread:issue:44817,issue #44817: ComplexHelper.h contains non-inline functions,stale/low urgency,2,stale/low urgency,1,0,44817,,,,0,b8382697174b6bca275fd06d5d38a34950e46b122d47c35f39d58eff48be014e
thread:issue:44851,issue #44851: fx: Python range function used on tensor shape is not symbolically traceable,stale/low urgency,2,stale/low urgency,1,0,44851,,,,0,ce35e0d61b0586f2c6068cbbb53bbed5624538a4a6e2bd15b62a53d85ee657b1
thread:issue:44878,"issue #44878: [ONNX] Support Tuple(List(Tensors),...) in input/ouput",stale/low urgency,2,stale/low urgency,1,0,44878,,,,0,327e0055235a762684278b470c37ad234479455995be7c945d6a21c6947660a7
thread:issue:44898,issue #44898: Don't query current device on stream construction,stale/low urgency,2,stale/low urgency,1,0,44898,,,,0,c346492c9a36567e39b07911a91b7a6790ebd4e97bbb67f2661472c3d519febc
thread:issue:44917,issue #44917: Incorrect documentation of SGD momentum,stale/low urgency,2,stale/low urgency,1,0,44917,,,,0,9b900734abe8f07ee21cca791691257cc21d309fb5080b13b94c868af79f08f1
thread:issue:44931,issue #44931: Issue while writing scalars o tensorboard using writer.add_scalars(....),stale/low urgency,2,stale/low urgency,1,0,44931,,,,0,589ccf05ed11d677d7c4869e1a2e93eff76df70d660b8d8d283246b6047298c8
thread:issue:44938,issue #44938: DataParallel on CPU,stale/low urgency,2,stale/low urgency,1,0,44938,,,,0,689e8f1927347adf756d28a57ac7a7ff137c00af83e728dee0792048d6b8c187
thread:issue:44943,issue #44943: In pytorch 1.6。Run model with input no contiguous tensor will become very slow.,stale/low urgency,2,stale/low urgency,1,0,44943,,,,0,d65577b30b343db95be4dc1fa8ae7d7e4dfbaebad4732dc090141645243a1c34
thread:issue:44948,issue #44948: Clarify use of GLog/GFlags,stale/low urgency,2,stale/low urgency,1,0,44948,,,,0,335b02d15fae4464e7e613b0bdd94e7bca6adbafc4bf6b1166e1830de01c4aae
thread:issue:44968,issue #44968: [feature request] dtype argument for torch.sign,stale/low urgency,2,stale/low urgency,1,0,44968,,,,0,95b09dc38eb9738f59c799a861c1907b80b5faf4d5fd90c25bc60d2f44b19e6b
thread:issue:44969,issue #44969: Tensordot does not support Bool,stale/low urgency,2,stale/low urgency,1,0,44969,,,,0,09c622a36e5ff8a5ed551df195048a000e572b0915ea417e30c89b81a24dc3f2
thread:issue:4501,issue #4501: Bind in Python _backward ATen functions,stale/low urgency,2,stale/low urgency,1,0,4501,,,,0,cfc118721add24f14fdb3f3d30ada6c0cf4223d32849b2e6d3b825880e22fa4c
thread:issue:45042,"issue #45042: torch.distributed launch.py is hanged. (pid, sts) = os.waitpid(self.pid, wait_flags)",stale/low urgency,2,stale/low urgency,1,0,45042,,,,0,5b51ac04586ed11076d6604ec4b44bda7fa83ea03c14b4ddcc56b10ca7570a24
thread:issue:45044,issue #45044: torch rpc cannot handle UnsupportedNodeError exception,stale/low urgency,2,stale/low urgency,1,0,45044,,,,0,6f5103fe14b565cc8adaf7f0e255eb72bf9c71ba93691bc4e566a1571f09cefe
thread:issue:45053,issue #45053: about benchmark issue,stale/low urgency,2,stale/low urgency,1,0,45053,,,,0,88d9c86983f6ff7bcc0ab524037f1e41ede50ef260be11bb6288557b47e12512
thread:issue:45073,issue #45073: Compilation errors on power-pc,stale/low urgency,2,stale/low urgency,1,0,45073,,,,0,30f7b988da694ab3b01786a2dabd9668ce087bf468c2c2703610e97274c9d24e
thread:issue:45115,issue #45115: Standardized Distributions,stale/low urgency,2,stale/low urgency,1,0,45115,,,,0,10246db339425ddefe7f23944aa90d0e55c1ef9c1d64307a17f8e180f6d90f24
thread:issue:45125,issue #45125: Boolean indexing of an ndarray with a torch.tensor mask breaks for size=1,stale/low urgency,2,stale/low urgency,1,0,45125,,,,0,1314f84f0c3b472b79ac60f380f993f82c85b0ed59d77bbb80419b9a4484b573
thread:issue:45129,issue #45129: Inconsistent wheel name in https://download.pytorch.org/whl/torch_stable.html,stale/low urgency,2,stale/low urgency,1,0,45129,,,,0,4a80ff63be540920b2536fb3bd40e77d1d6acbad274c3107c2a81aff5e3899be
thread:issue:45139,issue #45139: Mention accessor/data_ptr for raw memory access in Libtorch index API document and discuss performance implications,stale/low urgency,2,stale/low urgency,1,0,45139,,,,0,af12da3d2c62e2eb276f8b71e01420a59d0b7550767d160cc5c7a9a9882741ab
thread:issue:45203,issue #45203: Staged backend boxed fallback (per-operator precomputation / precompute),stale/low urgency,2,stale/low urgency,1,0,45203,,,,0,df20aca65896a306c713c247c7b3d7b844c1e19d2a7586a633e640cf89dc93b9
thread:issue:45220,issue #45220: FX should preserve type annotations and not break TorchScript-ability,stale/low urgency,2,stale/low urgency,1,0,45220,,,,0,5d9fc03e63906a606b06a3fed11f5625f37a1c2fc8cdfc9638dde8aafb39e035
thread:issue:4530,issue #4530: descriptor 'add' of 'torch._C._VariableBase' object needs an argument,stale/low urgency,2,stale/low urgency,1,0,4530,,,,0,1d0fe2cf112c5b9a9bef7adb2c827660df1e8766cf1759fa1356390e6af04fbe
thread:issue:45308,issue #45308: Blocked version of Cholesky backward,stale/low urgency,2,stale/low urgency,1,0,45308,,,,0,8a2333085d17bcce06bd6aec81d388ba7748af391f716d6615e8b3ca93c7597e
thread:issue:45321,issue #45321: Memory leak when using share_memory_ on cuda device,stale/low urgency,2,stale/low urgency,1,0,45321,,,,0,75f92a1df56ddf353deff59d444b94f0614c09dab1f68b5e9149f60872062872
thread:issue:45352,issue #45352: torch.nn.functional.one_hot should gracefully skip negative and out-of-range indices,stale/low urgency,2,stale/low urgency,1,0,45352,,,,0,5cebd7daf350c08483d7ce58782c0e245bc56a77ab8daf4f69d0e9b3c51ae550
thread:issue:45399,issue #45399: Load ExtraFilesMap with Java library,stale/low urgency,2,stale/low urgency,1,0,45399,,,,0,9b491243427fb1091f41d047d093535932b9a4114a10c792b9715435c1664949
thread:issue:45412,issue #45412: Autocast but there are no kernels registered for this dispatch key. The operator is p�8ߎU Aborted (core dumped),stale/low urgency,2,stale/low urgency,1,0,45412,,,,0,8de59ef5cd490dc33b2403c56b71ad460cd0b7d15653ba462e272a1816b2959d
thread:issue:45414,issue #45414: Module __call__ typing,stale/low urgency,2,stale/low urgency,1,0,45414,,,,0,c6efcfb01af648093b496218460e55b60b1063f7549cb1f0929a2697461034f7
thread:issue:4542,issue #4542: Carefully audit contiguity requirements of code,stale/low urgency,2,stale/low urgency,1,0,4542,,,,0,386280f1b13a43e9746cb68dbc7ce7ab3858c670392fe649aeb3034a814a4cf9
thread:issue:45428,issue #45428: Use libtorch for online inference: an illegal memory access was encountered,stale/low urgency,2,stale/low urgency,1,0,45428,,,,0,7e3977e189798e8d03be2581a1ab100753ddbe277a5a7d8a4af32416f3d7b413
thread:issue:45444,issue #45444: Using operator[] in GenericPackedTensorAccessor impossible in cpu code in .cu files (tensor on cpu),stale/low urgency,2,stale/low urgency,1,0,45444,,,,0,d7fd8115b48bc5249cb905aecad70d4ccbeef22d631d02ddfc9f7e10b5b8bb79
thread:issue:45447,issue #45447: Retire usages of CUDA_tensor_apply helpers in ATen,stale/low urgency,2,stale/low urgency,1,0,45447,,,,0,cfd39d58304076edc0cff3fdc9dc0fd9cca1fcd02f266607fc230f9d51526861
thread:issue:45448,issue #45448: CuDNN version not found,stale/low urgency,2,stale/low urgency,1,0,45448,,,,0,328f62008c119e394cb4797d902f4ff6f0cc39a00fda4d1ba67da9e49fe9b26e
thread:issue:45459,issue #45459: Backward for sparse tensor item select does not work,stale/low urgency,2,stale/low urgency,1,0,45459,,,,0,ad9142eac48a65ddea7c35a8c90529950ec4edc6069f705e7417cdcf9db69559
thread:issue:45481,issue #45481: Combining add_scalar with add_hparams with different frequencey,stale/low urgency,2,stale/low urgency,1,0,45481,,,,0,1e366d45f30ac2b067f257fec2a4efced4b3f58879be72d8ef09b2e7f1256708
thread:issue:45483,issue #45483: Extra arguments included in the doc where they are not actually presented in the source code,stale/low urgency,2,stale/low urgency,1,0,45483,,,,0,d35ce94e64ea3b080141bfe31c9959c17c5cfec062e87b9d61073e30e3143e3d
thread:issue:45493,issue #45493: Build LibTorch for cpu using OpenBLAS: Had to manually remove a path in Caffe2Targets.cmake,stale/low urgency,2,stale/low urgency,1,0,45493,,,,0,8ec4c9c165521afb500cae8d6e1a5fed3c1017cef773f3b514e57e5cc6f27050
thread:issue:4551,issue #4551: Met 'cudnnDestroyDropoutDescriptor' while run multiply gpu-based models in multiply processes,stale/low urgency,2,stale/low urgency,1,0,4551,,,,0,d0146bfd469b7429ab94f70a6b0976e4a06dc87df77e1599dbdf4df39b811181
thread:issue:45534,"issue #45534: [FX] Make tracer return a Graph, not a GraphModule",stale/low urgency,2,stale/low urgency,1,0,45534,,,,0,f8b9a0dc0945f3eb492012eb6cc4a953afbc79f5456826f901942fe59127312a
thread:issue:45540,issue #45540: Why does NLLLoss and CrossEntropyLoss require type long?,stale/low urgency,2,stale/low urgency,1,0,45540,,,,0,68535a9dced18f63acc55a161f2deee68b8a61d41fe3922e38ca70998fc57de4
thread:issue:45560,issue #45560: Android: allow preallocation of output buffers,stale/low urgency,2,stale/low urgency,1,0,45560,,,,0,d1748d59bc87454b3dc7cb3d56f73220090a9a556c9a76cd2413a02ab0a2654a
thread:issue:45569,issue #45569: Distutils Error in torch.hub Load(),stale/low urgency,2,stale/low urgency,1,0,45569,,,,0,a87405a9ca8af058428a60fe88bd15f9af257913bea61e3cf17f4d29dc26d194
thread:issue:45573,issue #45573: Build Failed Ubuntu Cuda. CUDAHooks.cpp:97:15: error: ‘struct cudaPointerAttributes’ has no member named ‘type’ return attr.type == cudaMemoryTypeHost;,stale/low urgency,2,stale/low urgency,1,0,45573,,,,0,0fcf374b5e6378e9b6508f0e523810a1fa091e71c7ecf7c5a317d36068fa39c8
thread:issue:45578,issue #45578: Why is RTX3080 slower than RTX2020-Ti?,stale/low urgency,2,stale/low urgency,1,0,45578,,,,0,a04b7d9c091ee52b14e323cd6f1da398e09f09c3a8c32504d6ed78b1197e4b4b
thread:issue:45635,issue #45635: fx graph mode quantization tutorials,stale/low urgency,2,stale/low urgency,1,0,45635,,,,0,ad035385b36f02f476c444172f467281cf9458e34bfc4ee2ab047bc5993f595b
thread:issue:4564,issue #4564: Mixed Tensor/TensorList arguments in ATen functions with explicit derivatives,stale/low urgency,2,stale/low urgency,1,0,4564,,,,0,4557bc0e629948c28c9b14121d3466765bdc376c09d6133c59701a797956cc65
thread:issue:45683,issue #45683: [FX] Configurability of nested module representation,stale/low urgency,2,stale/low urgency,1,0,45683,,,,0,8fd7374a1d3571d545303034c77bbe4b6ef3f5d56f781512c28b1270df16ea65
thread:issue:45690,issue #45690: Complex and real results do not agree when computing reciprocal or pow(-1) of 0,stale/low urgency,2,stale/low urgency,1,0,45690,,,,0,3e896a52c53e82f45a3e3e4d6baa519c454e57bc4fdf479ca8c84051174d0b72
thread:issue:45717,issue #45717: Better documentation of vec256 API,stale/low urgency,2,stale/low urgency,1,0,45717,,,,0,7f8d9cf06d33bff3fb65f91dd00e3499067792f1de698517526f89d13bfc0d39
thread:issue:4574,issue #4574: `from` keyword in `random_` gives error,stale/low urgency,2,stale/low urgency,1,0,4574,,,,0,e528de9cea06539febd1c7568ca16235edac0e168127a625dac730de59916987
thread:issue:45760,issue #45760: Complex Number support for distributed,stale/low urgency,2,stale/low urgency,1,0,45760,,,,0,ab4838c17da992a3b24a3e21ee5bd24fff200f5a2a34434074e6423c9442ed6d
thread:issue:45764,issue #45764: resize_(0) is very expensive,stale/low urgency,2,stale/low urgency,1,0,45764,,,,0,a887a7b266a747519eef470b3e1fad9362f81e51606036a4d26fea6a93d437ac
thread:issue:45767,issue #45767: `defineMethodsInModule` no longer exists,stale/low urgency,2,stale/low urgency,1,0,45767,,,,0,3eed69d4586b6daeb81981b282d4c63334571c904e3d37dd1d30ca2a856a00ac
thread:issue:45823,"issue #45823: Traceback for ""Warning: Mixed memory format inputs detected while calling the operator.""",stale/low urgency,2,stale/low urgency,1,0,45823,,,,0,2dda2d43a59791b11ea65edadb7f5a50ee6e6bb4613588441993a6f0d38eb696
thread:issue:45840,issue #45840: Error with DataParallel and dataclass,stale/low urgency,2,stale/low urgency,1,0,45840,,,,0,ab36d852c32a0237d759c2aa066b14cea1d56c05001696ef458c14cd4f56abfe
thread:issue:45855,issue #45855: Do we have plan to offer C++ binding for prune related features.,stale/low urgency,2,stale/low urgency,1,0,45855,,,,0,3fdc09e3be551fe2aa74feec621d8d5bb7a6127461c629855946139b00d080e1
thread:issue:45910,issue #45910: with torch.cuda.amp.autocast() get out of memory error when using with torch.no_grad() during validation,stale/low urgency,2,stale/low urgency,1,0,45910,,,,0,68df53d033faba2155bb9b79b8db383f45eb1f68e7e0fb5baead0c80afd77ba9
thread:issue:45943,issue #45943: Improve distributed documentation for NCCL_BLOCKING_WAIT,stale/low urgency,2,stale/low urgency,1,0,45943,,,,0,57f54fee91c3d95b8ae18ae927794277ee8622cdcc01e606f4f4fb948410b159
thread:issue:46024,issue #46024: Reproducibility breaks down with weighted Cross Entropy loss,stale/low urgency,2,stale/low urgency,1,0,46024,,,,0,8d52e45df48174353dfbc26bf162c2dfec09de74929ab434ef06222002799344
thread:issue:46027,issue #46027: Padding mode for ConvTransposeNd,stale/low urgency,2,stale/low urgency,1,0,46027,,,,0,7be5bfea8118b76bac997ac842ef5d45ac2d66697106e8245f5734e5190142e7
thread:issue:46031,issue #46031: Custom attention,stale/low urgency,2,stale/low urgency,1,0,46031,,,,0,dfbe068fdf302c4b74a2160f86b209b8035ff600e2bdeb81f5f78eae71b96691
thread:issue:46035,"issue #46035: Error installing on source: ""src.cxx:1:10: fatal error: glog/stl_logging.h: No such file or directory""",stale/low urgency,2,stale/low urgency,1,0,46035,,,,0,02015231463f723cfa2749be5c8f7ba6c8512441792c07e8bd0f530f6b1b8414
thread:issue:46061,issue #46061: Support undispatched ops in codegen,stale/low urgency,2,stale/low urgency,1,0,46061,,,,0,d281f84402cd21f5dace6ae47893dbe424c33064b587b9bbda92940bce08028d
thread:issue:46065,issue #46065: TestDataLoader.test_proper_exit takes 2.5min to finish,stale/low urgency,2,stale/low urgency,1,0,46065,,,,0,3786a9d54c8d1f68fa1079f1a6454a33c24353d8324e427ddeaba99569cd4013
thread:issue:46066,issue #46066: TestXNNPACKConv1dTransformPass.test_conv1d_with_relu_fc takes 2+ min to finsh,stale/low urgency,2,stale/low urgency,1,0,46066,,,,0,4494d6afc9a13f45176119c51ab6c55abcbc7fdc3dd44b1a481f79ac2bb76482
thread:issue:46076,issue #46076: NCCL watchdog thread should log warnings about long-running GPU operations instead of silently hanging,stale/low urgency,2,stale/low urgency,1,0,46076,,,,0,9ab495f5dc0004fa3004fd843340be01dc775ee41d53f063ee32ad5a1e83030c
thread:issue:46084,"issue #46084: In PyTorch Tutorials, RuntimeError: CUDA error: out of memory happen",stale/low urgency,2,stale/low urgency,1,0,46084,,,,0,070cebc71ba43c73c8758e31381aace74528a08c509b8acfa34f24d9ab324eaf
thread:issue:46093,"issue #46093: Spawn wrapper should catch BaseException, not Exception",stale/low urgency,2,stale/low urgency,1,0,46093,,,,0,c6b69654afe77616eaa74b137cce175829d56aa33deee36a83f151a7c46487df
thread:issue:46102,issue #46102: test_cat_cuda (__main__.TestTensorExprFuser) fails,stale/low urgency,2,stale/low urgency,1,0,46102,,,,0,919c407d5bc2dbf4a3885e874c2b6ac5c10c35785d757768ec09dce5919687c5
thread:issue:46139,"issue #46139: Policy CMP0012 is not set, CMake, Building For C++ With PyTorch CUDA",stale/low urgency,2,stale/low urgency,1,0,46139,,,,0,d8960f2d39d6d8847c843041aaecc3717739e5d645cfc3ac5f9bd01e5d86d241
thread:issue:46149,issue #46149: seeding does not work when I initialize a linear model even if I do not use it in the code,stale/low urgency,2,stale/low urgency,1,0,46149,,,,0,34830c7db2f7f93ed4387270964cf3a0fe88b98b97222f27b8cffb30e7c7662c
thread:issue:46157,issue #46157: Ability to disable cusolver,stale/low urgency,2,stale/low urgency,1,0,46157,,,,0,46525d6c6ad1d31fc5eb1b0076deb398ba4a16384ae366ae764678e4c62c01cc
thread:issue:46178,"issue #46178: torch.cuda.amp! when I use @autocast() on DCN(DeformConv), the error ""RuntimeError:expect scalar type Float but Half """,stale/low urgency,2,stale/low urgency,1,0,46178,,,,0,071ea4cf8e46cd9fcc4231a4dba9589f99f6cb01386d28b1fab8bba21a704cb7
thread:issue:46187,issue #46187: Better support for operators that return (named) tuples of tensors,stale/low urgency,2,stale/low urgency,1,0,46187,,,,0,00f27caf454959749e602dd73073df2c67f61d21a4fd7cc16a885ba439582371
thread:issue:46202,issue #46202: pytorch not uploading data to tensorboard page,stale/low urgency,2,stale/low urgency,1,0,46202,,,,0,baa0e8a5a1bf03aab26a686dacb4de521f34b52cce5395598c2984c668c1e093
thread:issue:46210,issue #46210: [FX]Add support for len() on Proxy,stale/low urgency,2,stale/low urgency,1,0,46210,,,,0,bd2271d0eecd3c92ec332afdb12fbe368a5dc98ade77cfcd8963f04ef4a77a9e
thread:issue:4622,issue #4622: Protect user from No module named _C import error,stale/low urgency,2,stale/low urgency,1,0,4622,,,,0,19bbc4ff3564d6c8629dd8337d0c993898ac407fa830523c4f4ab0dacbb12d7b
thread:issue:46222,issue #46222: Use better tempfile creation mechanism to avoid skip windows test,stale/low urgency,2,stale/low urgency,1,0,46222,,,,0,a270687caf6a3f2c28669f92dbef741aa0e3560cdc4abef0558adeb17a2eaa99
thread:issue:46241,issue #46241: ubuntu16.04+vscode compile libtorch error,stale/low urgency,2,stale/low urgency,1,0,46241,,,,0,1c615906d2cb7ed70a4aabe47e7e8fee70441de759bcaab45c68a19bf518c58a
thread:issue:46248,"issue #46248: ""distributed"" NCCL tests fail when having more than 3 GPUs",stale/low urgency,2,stale/low urgency,1,0,46248,,,,0,a3e8780907bf49ab64b5630734195a0b5e3b185f5399a23d1ce77b12ff46159a
thread:issue:46253,issue #46253: Add short guide for updating `torch.norm` calls to `torch.linalg.norm`,stale/low urgency,2,stale/low urgency,1,0,46253,,,,0,29fbfe8d33096d6169a3f7805dbadb34367a53a76afee8aa5f3f5c36d2473ab1
thread:issue:46254,issue #46254: Enable named tensor inputs to `torch.linalg.norm`,stale/low urgency,2,stale/low urgency,1,0,46254,,,,0,6ce253ca3b19f1c6814a78643ab3ef139486f34498262ed230159540b7db091f
thread:issue:46272,issue #46272: test_distributed_* does not work with run_test.py -i option,stale/low urgency,2,stale/low urgency,1,0,46272,,,,0,7fa5ac572ff28f746d188d23c7cd844e8c2e5f3ec5d2bd4a0325ef80f2b41573
thread:issue:46276,issue #46276: test_distributed_* does not show error details from the subprocess,stale/low urgency,2,stale/low urgency,1,0,46276,,,,0,d988e308582fb6805e5011851b0b09e0aaf08a4ed49d62bd70ce53ddee33df8d
thread:issue:46291,issue #46291: ProcessGroup::Work API changes,stale/low urgency,2,stale/low urgency,1,0,46291,,,,0,f968839425c2f42eb3fa91c2ba3add3eeea4dae4816c32ad0e7e1a6e4357d9a0
thread:issue:46311,"issue #46311: Compile error when use c10::RegisterOperators::Options::kernel with the return type ""std::tuple<Tensor&, Tensor&>""",stale/low urgency,2,stale/low urgency,1,0,46311,,,,0,b1d4145ac2115d3d86d774e205c39af3489b74b04f691f5a701a8edcfcdb9bf8
thread:issue:46319,issue #46319: Remove header declarations for CPUType/TypeDefault,stale/low urgency,2,stale/low urgency,1,0,46319,,,,0,e801f23a3c985d1b9a282bd809b04fa1001a61b6ea4f252559c4d7b754ba2132
thread:issue:46321,issue #46321: Add distributed examples into PyTorch CI tests,stale/low urgency,2,stale/low urgency,1,0,46321,,,,0,63dec30cfc061d8b9e4010c21fd68d1f99ebdfe53adadf3332c86e5073b116a4
thread:issue:46350,issue #46350: torch.distributions.half_normal.HalfNormal.cdf returns negative values,stale/low urgency,2,stale/low urgency,1,0,46350,,,,0,13285d130283a07ce18be39bba6f1ab5c6cd20d072656a4cb7de6ec2dfdd534c
thread:issue:4636,issue #4636: Better header hygiene in ATen,stale/low urgency,2,stale/low urgency,1,0,4636,,,,0,52dae8be7062dd717ac3689c41b5dc52fb5ea014c16a6f0b3f5a7ee68e313f34
thread:issue:46367,issue #46367: [chore] Test pybind11 2.6.0 (RC 2?),stale/low urgency,2,stale/low urgency,1,0,46367,,,,0,089f46fd0cc691eafd394638973b2797f304039d757fc446167a013dda70ef45
thread:issue:46379,"issue #46379: Reduce ""reserved"" memory by PyTorch.",stale/low urgency,2,stale/low urgency,1,0,46379,,,,0,392db3ca76fd4b1e25d87c62b1218e39a934f74f15973a2fcab9cf6418136354
thread:issue:46381,issue #46381: [Feature] Imbalanced batch scattering in th.nn.DataParallel,stale/low urgency,2,stale/low urgency,1,0,46381,,,,0,399dcace5582f36d8fc532eb0b16f7ae5036e045c6a11c00183c15ca456fd92e
thread:issue:46382,"issue #46382: Code asserts when register new aten operations' implementation thru ""c10::RegisterOperators::op"" API.",stale/low urgency,2,stale/low urgency,1,0,46382,,,,0,a8496526cfad514c39ebd15bfaa5adbde0eef65d7da23c32dc8b004b05b28ee1
thread:issue:46386,"issue #46386: GPU memory leak when registering a forward hook with ""self"" access",stale/low urgency,2,stale/low urgency,1,0,46386,,,,0,fcf3f5e9c3ad58c86fa3ab99401b67c18d5e338526c546b577ab4c0bd3fe17d2
thread:issue:46400,issue #46400: Make Undefined reprensentable in DispatchKeySet.,stale/low urgency,2,stale/low urgency,1,0,46400,,,,0,1a15a6661280dde17e7dc57b58eaabc1c307d93ef3859bb9bcd4c41a19f0f340
thread:issue:46409,issue #46409: Custom ops get stuck in multiprocess data loader under certain environments,stale/low urgency,2,stale/low urgency,1,0,46409,,,,0,3a021fd10b7efedf2f62afa51de02117f916a3a9bf0e6f3d181fef02b537f60a
thread:issue:46428,issue #46428: send_object/recv_object APIs for c10d,stale/low urgency,2,stale/low urgency,1,0,46428,,,,0,43d40095f34aaef02f43a4193969a0fc2bbb89fd2096d8d97edabfa4dcdc2e4a
thread:issue:46464,issue #46464: [META] Assert that expected libraries appear in libtorch,stale/low urgency,2,stale/low urgency,1,0,46464,,,,0,66c8f1212e1df091a1196ca79df4c8a9e5782f99513841b8c6f9b82ed6c7c39c
thread:issue:46522,issue #46522: ModuleNotFoundError when installing PyTorch via pip on aarch64 environment,stale/low urgency,2,stale/low urgency,1,0,46522,,,,0,ba09de24110b522264044b4aa10f6c99b82d828d8d6f6d0aaa58533a4f0750bb
thread:issue:46530,issue #46530: [Request] Batched Dataset->DataLoader interface,stale/low urgency,2,stale/low urgency,1,0,46530,,,,0,66daf5cd1a9b098f4a456a1d668d7ba158a67d2a1250dd0a9971c310bdc77960
thread:issue:46597,issue #46597: Suggestion: link individual APIs in docs with tutorials that use them,stale/low urgency,2,stale/low urgency,1,0,46597,,,,0,279e99f77bf46bbb4f3fdc367c5f12dc361d04775ac61183772d81097a1135f2
thread:issue:46604,issue #46604: torch.concat doesn't raise an error in a quantized model,stale/low urgency,2,stale/low urgency,1,0,46604,,,,0,fb82b59818eb72edb867d9dc3e0b050728be8eb6ae301974c82aea6ccded01e9
thread:issue:46631,issue #46631: error in bazel build //...,stale/low urgency,2,stale/low urgency,1,0,46631,,,,0,934180587a74288463452facb43ad15f751993b02c97f59d0b320b585a5db92d
thread:issue:46643,issue #46643: Autograd support for the tensor multiplication of sparse tensors,stale/low urgency,2,stale/low urgency,1,0,46643,,,,0,04391b2f626949ab54683d210f822bdab4dcce3337a65440bdcd58ac4c6abfc6
thread:issue:46644,issue #46644: Trying to compile PyTorch for sm_30 fails with `error: identifier “__ldg” is undefined`,stale/low urgency,2,stale/low urgency,1,0,46644,,,,0,1c4641968d2a0198fb0ea1ba8df6db14aac0291efd175321056ec9bca622099f
thread:issue:46659,issue #46659: Excluding image processing operators due to no opencv,stale/low urgency,2,stale/low urgency,1,0,46659,,,,0,a136dcaa4a8f0c5073573726b6dc227ba86a0b5ccff242bbef2ee978a7dbe514
thread:issue:46705,issue #46705: torch.eye not supported for sparse layout,stale/low urgency,2,stale/low urgency,1,0,46705,,,,0,a795549713c5bedf9bfa845a74b29bf96ae1a19bdb579f5d2d88ff54b876663b
thread:issue:46711,issue #46711: A random split function that return the datasets following specific target (label) distribution.,stale/low urgency,2,stale/low urgency,1,0,46711,,,,0,69a79922bcfab786dc77196de6f97f37f6e3326ed2bf17aeb384c56c544140ee
thread:issue:46712,issue #46712: The documentation for c10::Dict is completely empty.,stale/low urgency,2,stale/low urgency,1,0,46712,,,,0,0a156265478824231d110c2a9516af1b23335aad7c52a5f87ddb01c507d21c06
thread:issue:46739,issue #46739: torch.utils.tensorboard.SummaryWriter.add_embedding fails for some label_img sizes,stale/low urgency,2,stale/low urgency,1,0,46739,,,,0,10959cf46ae3d24d0ce678f042a72c5e51930a3db251468df0ec0bc5a93c8f6a
thread:issue:46753,issue #46753: Memory leak when creating new tensors inside nn.DataParallel on multiple GPUs,stale/low urgency,2,stale/low urgency,1,0,46753,,,,0,7e3d758bca005a9cf1d284300799fba0cb1d457383a708f60c1613759d24700e
thread:issue:46760,issue #46760: How to define a new data type in native_functions.yaml?,stale/low urgency,2,stale/low urgency,1,0,46760,,,,0,367df04aaa58167f1b2d0f89ce5687c3cccacbbdf3376fdb4298859d2d679026
thread:issue:46766,issue #46766: Unable to compile from code v1.6.0,stale/low urgency,2,stale/low urgency,1,0,46766,,,,0,969305e9620ab4316a243970573fc239c1a3632c0a350e0028f65add40d7ecd3
thread:issue:46787,issue #46787: Clearer error messages for 'different devices' error in the RNN module,stale/low urgency,2,stale/low urgency,1,0,46787,,,,0,09e6b1e7d1931aceb1d799b57514da3ce8af50f920f10c9676a7074a473bf982
thread:issue:46802,"issue #46802: DataLoader gives ""Broken pipe"" error on Linux platform",stale/low urgency,2,stale/low urgency,1,0,46802,,,,0,ca38777db659b303be07dde457688beb87c7722f635167b9d634697e91df3ec3
thread:issue:46803,issue #46803: About 2 minor bug fixes on CUDA macOSX 10.13.6,stale/low urgency,2,stale/low urgency,1,0,46803,,,,0,6da56d966136f1a1b9573d9eec4b27d86f2046d36f87134e027bf259948fa35a
thread:issue:46807,issue #46807: [cudatoolkit 11.0] segfaults,stale/low urgency,2,stale/low urgency,1,0,46807,,,,0,b7bff5fac799bfa9adcd6f51cc6377277b24a5b496bd80fb99c9d9c0be3c3840
thread:issue:46809,issue #46809: Including AdaBound in the list of Optimizers.,stale/low urgency,2,stale/low urgency,1,0,46809,,,,0,1e0793682b4d64676df82f5d99bc8acfba01cae33b88376b0cf51f2c778a33cb
thread:issue:46819,issue #46819: Specify input dimensions where they are not obvious,stale/low urgency,2,stale/low urgency,1,0,46819,,,,0,06625039cd42115c4655bfc673f6f8acc4a73db6e7dcabb41b24b8a4f340722b
thread:issue:46820,issue #46820: simple v *= v_scale error,stale/low urgency,2,stale/low urgency,1,0,46820,,,,0,844aa535fb95422e0653ae200c7ad340f6085af468296574e53308878c984fd2
thread:issue:46822,issue #46822: C++ Optimizer: remove warning on Optimizer::size method,stale/low urgency,2,stale/low urgency,1,0,46822,,,,0,a974bb42cd77609421888f3177a7cfba5aebfc7fc38cc15baaefc8ba70cfa87e
thread:issue:46869,issue #46869: [META] DO NOT DELETE THIS LABEL,stale/low urgency,2,stale/low urgency,1,0,46869,,,,0,2a6be823dfdfa59ed2ed235c55cad9e9674a5f5a0d7cf06268c5cdb110be952c
thread:issue:46899,issue #46899: CUDA out of memory when using torch.load,stale/low urgency,2,stale/low urgency,1,0,46899,,,,0,bc9e7f464b2c059c8ee2c3ee96104ff8232598cc8a956ea0567fa3b8ca04501d
thread:issue:46904,issue #46904: Authentication for RPC,stale/low urgency,2,stale/low urgency,1,0,46904,,,,0,638433b531ad31bb0574368847e9375aec24ca1cec8eb91201dfc3d496f7caf7
thread:issue:46918,issue #46918: Jacobians computed by autograd.functional.jacobian with compute_graph sometimes set requires_grad True,stale/low urgency,2,stale/low urgency,1,0,46918,,,,0,860ee01ddbf6747d283279f38669e569d112d0096e4320b4f570002ebfaf2f05
thread:issue:46929,issue #46929: Pooling code does not allow sliding window starting in right padded region,stale/low urgency,2,stale/low urgency,1,0,46929,,,,0,2a789a00771a79691d1fc6a5221d69e6b09127ea0a82e87e5cd8555fbfdbb89e
thread:issue:46947,issue #46947: Add Optimistic Mirror Descent (OMD),stale/low urgency,2,stale/low urgency,1,0,46947,,,,0,94be902b2b8573aff09b13d02f95eb48cbf0bc1d1837969c58e00987b97b6a50
thread:issue:46971,issue #46971: Yolov5 detect.py(ingerence) Error:Torch.nn.modules.module.ModuleAttributeError: 'Hardswish' object has no attribute 'inplace',stale/low urgency,2,stale/low urgency,1,0,46971,,,,0,737d146493158ba2fe42e7584a21b5a4a7c954be14b3a238bb459aefa4590cf8
thread:issue:46980,issue #46980: torch.acos not supported for sparse layout,stale/low urgency,2,stale/low urgency,1,0,46980,,,,0,dc8c7aaafe73f2b375e986ceff7de6089f231bee6660c0ca89cfd9ca95420c12
thread:issue:47027,issue #47027: Fix the way imports are done to be more correct for static type checkers,stale/low urgency,2,stale/low urgency,1,0,47027,,,,0,091e50f9514ada6855b18a9fb8fa580d23e65990f6af34c5df1a6550a5a21c37
thread:issue:47028,issue #47028: torch.autograd.backward() fails to sync with other stream,stale/low urgency,2,stale/low urgency,1,0,47028,,,,0,b7658f31714c1240fbf289d0d89008349f304546611a5767ec84777f2444422a
thread:issue:4703,issue #4703: Very slow on CPU,stale/low urgency,2,stale/low urgency,1,0,4703,,,,0,b06cd4d2dd8ccc1ace1e6e8d9175b773f192684412beb3c11023c0d734edaecb
thread:issue:47109,issue #47109: Pytorch 1.7/Cuda 11.1 binaries,stale/low urgency,2,stale/low urgency,1,0,47109,,,,0,a7eea1cb2d6f977822b78c5b73820ffd2ecd99ab367d96a6311cc96ffa2ac7ca
thread:issue:47111,issue #47111: THCudaShutdown should be called before THCState_free,stale/low urgency,2,stale/low urgency,1,0,47111,,,,0,a35b4061debf152bdc221beacba3485014bd12abe57a93e8054e3cbcca46f579
thread:issue:47175,issue #47175: [feature request] Show warning if optimizer.zero_grad() was not called before optimizer.step(),stale/low urgency,2,stale/low urgency,1,0,47175,,,,0,e7d6a5616c430d2f27a17982bc93c2e4c79f93623cd702968d0b2f67725e874f
thread:issue:47201,issue #47201: torch.save has a maximum size regardless of RAM,stale/low urgency,2,stale/low urgency,1,0,47201,,,,0,1600fcb13eda888b1b962208ffb1403c14deedfcdca8a784da0c08632a392982
thread:issue:47208,issue #47208: [RFC] Add test execution time analysis CI workflow,stale/low urgency,2,stale/low urgency,1,0,47208,,,,0,a5db79417494191600eee1dd815901778505989db00c12f963ece59fcf3140e0
thread:issue:47281,issue #47281: Error: invalid use of register,stale/low urgency,2,stale/low urgency,1,0,47281,,,,0,ca577f5f3ecf952812762dd94bd66cd62398baeeecf3dfce9f7b762fd33492f0
thread:issue:47291,issue #47291: ProcessGroupNCCL NCCL lib version mismatch,stale/low urgency,2,stale/low urgency,1,0,47291,,,,0,cada319f6f85c2330fe724515dc086c7c9f6fbfe75321f3945c191e5ecee22a6
thread:issue:47307,issue #47307: [Feature request] Stochastic Frank-Wolfe optimizer,stale/low urgency,2,stale/low urgency,1,0,47307,,,,0,5cf6d159a194bd974cbbd334b3fae0c5bf089d8d85fd699bfeda8e2bd15047b6
thread:issue:4731,issue #4731: Consider disallowing Variables that require grad in NCCL/comm functions,stale/low urgency,2,stale/low urgency,1,0,4731,,,,0,3d898fe82921dfa25f705bd6f1191a1427587bf598fefd9e38fa64e1dfce42d9
thread:issue:47311,issue #47311: Quantized modules should properly implement __getstate__ and __setstate__ (copy.deepcopy doesn't work on quantized model),stale/low urgency,2,stale/low urgency,1,0,47311,,,,0,c3aeb4a83165a14f6c5c98a45c7197b374d51312c7049fff97101345829afc68
thread:issue:47341,issue #47341: ConvTranspose1d groups=channels is very slow!!!,stale/low urgency,2,stale/low urgency,1,0,47341,,,,0,3e87717192d9702dbd044466b2d7dbfdcde00188c548730c61f6d070d201e0c5
thread:issue:47351,issue #47351: no member named 'beta1' in 'torch::optim::AdamOptions',stale/low urgency,2,stale/low urgency,1,0,47351,,,,0,10baa968ee0c0067675629b9cfedaabcd8750d744e65db86187fd0b23cb66e66
thread:issue:47357,issue #47357: Conv1D formula in docs is wrong,stale/low urgency,2,stale/low urgency,1,0,47357,,,,0,145e736289c22c3226f99efef987264f79edfa65e969c1d549b11af9754ce5dc
thread:issue:47363,issue #47363: FX quantization: we should preserve original model class name through the quantization passes,stale/low urgency,2,stale/low urgency,1,0,47363,,,,0,db9a119c3016889de0ea886457730eaeb4e8d532b07cef54e096abf0003bf261
thread:issue:47378,issue #47378: torch.vmap giving INTERNAL ASSERT FAILED error,stale/low urgency,2,stale/low urgency,1,0,47378,,,,0,2d62fd43e4b98254bfa39a20f5a2e6cdf54248aa21bed421029ccecf161a3c7a
thread:issue:47433,issue #47433: unbalanced gpu memory when using DistributedDataParallel,stale/low urgency,2,stale/low urgency,1,0,47433,,,,0,703e068950cf6d4e7c23a4dbed8aec80418d294cbd8e726628a941b6c60774b3
thread:issue:47439,issue #47439: how to use torch.utils.checkpoint + gru with variable length sequence?,stale/low urgency,2,stale/low urgency,1,0,47439,,,,0,3d04658b4f2c7f140519a2fda0c393aafc4aeebaeb6c901ec5e18a3c1e3bee05
thread:issue:47447,"issue #47447: thread blocked when moving a tensor from GPU to CPU, by calling the function .cpu() in pytorch. This kind of block can be stop by any window event like mouse moving/clicking or keyboard pressing.",stale/low urgency,2,stale/low urgency,1,0,47447,,,,0,f27664f3815d6c610e8afa47b3e3993699bd9d8b77401780bc70d078d3e7f73c
thread:issue:47476,issue #47476: Reciprocal and reciprocal square root instructions are too inaccurate on ARM64,stale/low urgency,2,stale/low urgency,1,0,47476,,,,0,504eeddb6be5482d3ddae6c832d679f949a04c65d8deb332048d7f0f62a26133
thread:issue:47481,issue #47481: [fx] scripting a model with tensor list as input fails,stale/low urgency,2,stale/low urgency,1,0,47481,,,,0,4d29fa7df8f70a308c87055a7923b613f34cd418bf4d4b4686b4cd67e745380d
thread:issue:47502,issue #47502: Hackable python-only autograd engine,stale/low urgency,2,stale/low urgency,1,0,47502,,,,0,76ed50a1aabbdc80c1d1a07de5c6ee5358593aaafdc4b9d9dcc66fe4f2bae398
thread:issue:47535,issue #47535: per channel observer to work for weights of groupwise conv transpose,stale/low urgency,2,stale/low urgency,1,0,47535,,,,0,5daa55ea29d7a77a5b3eba5e3e038575486d11a1d86b44879e848bd4061bff94
thread:issue:47546,issue #47546: Whats Pytorch's policy on adding support for a wider range of hardwares for training and inference?,stale/low urgency,2,stale/low urgency,1,0,47546,,,,0,59e720db18490b13b187def06820f31708c9178f81aeb0db55dee2a1bdddcfcd
thread:issue:47557,issue #47557: Ninja-built CUDAExtension build process ignores changed #include dependencies,stale/low urgency,2,stale/low urgency,1,0,47557,,,,0,ef349a93560e72235093c237315f906b81ba491cd18d1b697cd1246d76a142d4
thread:issue:47560,issue #47560: Minumul LR is never reached in,stale/low urgency,2,stale/low urgency,1,0,47560,,,,0,a7bc1db098fd1a7f3d098aebd032479892900b7d4935b388117cf35679bb47ef
thread:issue:47563,issue #47563: Handling multiple large-scale datasets efficiently,stale/low urgency,2,stale/low urgency,1,0,47563,,,,0,8a7e83d4175b6989d367070a0602c016b74d9bddcf32cf432a81e552e6c6d6a8
thread:issue:47568,issue #47568: Failed to get generated_cpp list,stale/low urgency,2,stale/low urgency,1,0,47568,,,,0,26000d6f2c96eda4e74e694427453b7ae3a437538a7471fe9c5e53baf6d6c1bb
thread:issue:4758,issue #4758: Clang color diagnostics don't work with ninja,stale/low urgency,2,stale/low urgency,1,0,4758,,,,0,d6a2b9daf96723150f300421e8e7bf504a7d9c0fc755579622d0bb421916bc60
thread:issue:47588,issue #47588: Can we apply Weight normalization hook to a method other than `forward`?,stale/low urgency,2,stale/low urgency,1,0,47588,,,,0,d307683be3e4f19bc61184a1b99e9297eb202b09c3e39d869f9a7e2185331d43
thread:issue:47591,issue #47591: PyTorch1.3.1 Can not using namespace torch::indexing,stale/low urgency,2,stale/low urgency,1,0,47591,,,,0,8deb712a52102859026a3a91602e24acddafeff14dd29dcdd982ec86242b7c33
thread:issue:47609,issue #47609: Error while trying to load a pretrained ResNet34 VTN (Video Transformer Network) model in Android,stale/low urgency,2,stale/low urgency,1,0,47609,,,,0,deda7652a6bda1b72382d9fdec7e423506057696d0f5cd72d1c617a399cdd390
thread:issue:47610,issue #47610: Input dimension check for `torch.gather`,stale/low urgency,2,stale/low urgency,1,0,47610,,,,0,34c876286d09610d6cce449a013de4279dfca46fcd260fd09ee8db864eeea4df
thread:issue:47627,issue #47627: Allow add_embedding to have dict for metadata,stale/low urgency,2,stale/low urgency,1,0,47627,,,,0,6fd6791fb90d5778f5bad9a6361a4855f38c2786b366b34945a431b88a208335
thread:issue:47650,issue #47650: DataParallel support for scripted modules in C++,stale/low urgency,2,stale/low urgency,1,0,47650,,,,0,0e9ac526c0d7c8f0f547aab8037bf9f5547c4b6ab8a3bb5ba1ced97c922998ca
thread:issue:47681,issue #47681: How to install Pytorch on AIX7.2 without internet access?,stale/low urgency,2,stale/low urgency,1,0,47681,,,,0,99fc4faa7211c1c5f6fcb8cc55708f3529bb89b3432c2f692b2511e86d47b4ee
thread:issue:47699,issue #47699: DCNv2 Layers,stale/low urgency,2,stale/low urgency,1,0,47699,,,,0,79981852f5adfc9a5f52b5328aa0b9f742990683f584e248fdcde40592053b91
thread:issue:47739,issue #47739: Conda package should install libtorch in the standard location,stale/low urgency,2,stale/low urgency,1,0,47739,,,,0,71f055171a637e89fc3f187888064e0b06f1d8609e847885aa388102a246d5d3
thread:issue:47751,issue #47751: Make libtorch modular,stale/low urgency,2,stale/low urgency,1,0,47751,,,,0,e929e99a66f40d706d7b0a0544d69fa9ab9414758c52f71ddd34a49a007ebe35
thread:issue:47753,"issue #47753: Attempted to read a PyTorch file with version 3, but the maximum supported version for reading is 2",stale/low urgency,2,stale/low urgency,1,0,47753,,,,0,ff48b4785c325d22acd7c78988934d82db4070cbf030df83808e1042a921a576
thread:issue:47824,"issue #47824: DDP error, when I use ""if"" branch in forward function.",stale/low urgency,2,stale/low urgency,1,0,47824,,,,0,34a7baddf79be50926278fe4201ed6d0e854e36f8174a32bbd2a98b44005a826
thread:issue:47841,issue #47841: Factorial & Binomial Coefficient,stale/low urgency,2,stale/low urgency,1,0,47841,,,,0,73485ae0ee72edd1614d21b34a08ad662ed5efe067cd0226215ea21c1fc3e5bc
thread:issue:47857,issue #47857: QAT with DDP should have documentation,stale/low urgency,2,stale/low urgency,1,0,47857,,,,0,d4f8504639c26b1b98c467ff8749f70ad2abf86afdc915b9b1b7e2c6b708a6a1
thread:issue:47885,issue #47885: NCCL 2.7.8 errors on PyTorch distributed process group creation,stale/low urgency,2,stale/low urgency,1,0,47885,,,,0,9dc2be7c5e115a7dc221858defefd8f0f53505e22d47dd55677ec118695e5b95
thread:issue:47901,"issue #47901: Get forward output, grad, hessian all at once",stale/low urgency,2,stale/low urgency,1,0,47901,,,,0,338941427557d5d6e0f9a53639ff5c07f773bee10766fe566818df7e297b985f
thread:issue:47904,issue #47904: Memory leak in nn.MaxPool2d layer when run on iOS,stale/low urgency,2,stale/low urgency,1,0,47904,,,,0,c539a5cb7f308be7bc754fed80f69c50e91ea7088b37653dd1352b6385ba3140
thread:issue:47906,issue #47906: amp for custom op,stale/low urgency,2,stale/low urgency,1,0,47906,,,,0,f488673d06e70ac194cc0de9d555115f796ef062e14ebd41b3fbf3f4fd75c7ba
thread:issue:47912,issue #47912: Code becomes more than x20 slower after upgrading torch version,stale/low urgency,2,stale/low urgency,1,0,47912,,,,0,ebd0ea922fa447ee2887bce17408857bc50b4aafdc91f01f0f2fa1c96118d6c5
thread:issue:47916,issue #47916: Hash mismatch for METADATA file,stale/low urgency,2,stale/low urgency,1,0,47916,,,,0,e4f8f3fa2d9b27ce11bc99ae26879ed1ec2a673e0734113484514063c23e1992
thread:issue:47917,issue #47917: LibTorch cannot load PyTorch exported model,stale/low urgency,2,stale/low urgency,1,0,47917,,,,0,8a455be682ff67e2ae9f8ae0e6f6e83b55ed2724da9e17b2e3d8a94d788c20d0
thread:issue:47918,issue #47918: Function request: Sparse matrix inverse,stale/low urgency,2,stale/low urgency,1,0,47918,,,,0,f9bef633c651a9463590aa013be59ac49cf0d8a62d087e5794bac6043379fa6a
thread:issue:47954,issue #47954: [Feature] torch.bitwise_where,stale/low urgency,2,stale/low urgency,1,0,47954,,,,0,4185d5df03fae778452d0548cf548a519de2ab69b0a7b3fadd2f28d34414ddbf
thread:issue:47996,issue #47996: GPU Vendor-Agnosticism via Vulkan,stale/low urgency,2,stale/low urgency,1,0,47996,,,,0,702173e9a37fedf3e5276de6a2dcf153ef5046c2189401ae5f0567424dc51104
thread:issue:48000,issue #48000: Inconsistent complex results with NumPy when computing non-positive power of 0,stale/low urgency,2,stale/low urgency,1,0,48000,,,,0,d4d20c534d10c5d7bf35af983a4587460ef92bd34c04835ad5d09cd0b3d5c627
thread:issue:48033,issue #48033: Missing -fopenmp when used in torchvision,stale/low urgency,2,stale/low urgency,1,0,48033,,,,0,39784c285b8afcaf9054f2e53e8f53e773e7ca40852159cc3c0fdd6e6523e1e1
thread:issue:48064,issue #48064: FAILED: test_api/CMakeFiles/test_api.dir/dataloader.cpp.o,stale/low urgency,2,stale/low urgency,1,0,48064,,,,0,9f0cc43bd33f9ad29d3da53a0d9d8d002d8ac3525984878b8f71a8b3e55f4ec6
thread:issue:48083,issue #48083: CPU Tensor with Python MP Freezing in Docker Container,stale/low urgency,2,stale/low urgency,1,0,48083,,,,0,773bc1ed1c921f99cff6a294534b2f9176faa1235f5b0c9ea289de6f076d9fc0
thread:issue:48106,issue #48106: quantization - document get_default_qconfig,stale/low urgency,2,stale/low urgency,1,0,48106,,,,0,d6d6b713c5ed592dd8f40d5e94f0c817ba0bc99fbc87c9b6970ed642a92ff0f5
thread:issue:48153,issue #48153: More robust list comprehension,stale/low urgency,2,stale/low urgency,1,0,48153,,,,0,3596240786d50aa546b41691f659a61bf3aab4f749cf863108ea74d8343c7886
thread:issue:48163,issue #48163: nn.functional.interpolate backward in fp16 is extremely slow,stale/low urgency,2,stale/low urgency,1,0,48163,,,,0,7906b7bfb8c87f4ba849975a67139ec71d71fe5705806620636d9f3533854609
thread:issue:48167,issue #48167: [libtorch] Build shared library with libtorch and compile a static library,stale/low urgency,2,stale/low urgency,1,0,48167,,,,0,373f9912705f31d0644a023d3457deea303644186531979b9deb83a7602e6058
thread:issue:48202,"issue #48202: ""Found no NVIDIA driver"" produces too much log",stale/low urgency,2,stale/low urgency,1,0,48202,,,,0,dfc5640c8e050da98a64d88b783224f05313e641b7ba3c509c9361c3f0359131
thread:issue:48209,issue #48209: [FX] Write an inference optimization pass,stale/low urgency,2,stale/low urgency,1,0,48209,,,,0,c025f0b390fc0f0e0e6fd775a94eb266c011ef8ed01eb00ac23ddf2568a15df0
thread:issue:48227,issue #48227: Modifying values() tensor of COO tensor requiring grad throws an odd error message,stale/low urgency,2,stale/low urgency,1,0,48227,,,,0,dd53ee591dbde6a88e35d7b47c81f905e096ab784735ea1de56a1fa1451efe9b
thread:issue:48235,issue #48235: Softplus differs depending on number of elements in tensor,stale/low urgency,2,stale/low urgency,1,0,48235,,,,0,3d989a1264a5e962f11baacdc31b0f0928cfd0f21d5299c93cdc651289457b27
thread:issue:48245,"issue #48245: RuntimeError: ""threshold_cpu"" not implemented for 'Half'",stale/low urgency,2,stale/low urgency,1,0,48245,,,,0,b7a85bad1914dbc33ab27bde707d549983d81f2f28515e92fe7d6c8ad4f829d0
thread:issue:4825,issue #4825: [feature request]Support AVX512F intrinstics to vectorize operations,stale/low urgency,2,stale/low urgency,1,0,4825,,,,0,17cd361862c15a86c2d3159081e3bef5dafe59ad2e199e9eb04a00f29d635b67
thread:issue:48273,issue #48273: torch.multinomial example is incorrect,stale/low urgency,2,stale/low urgency,1,0,48273,,,,0,ac1e383c538560b15fe70e4e9191c35fc8dac742cc4ebba76335110219de04c9
thread:issue:48281,issue #48281: Batched inplace mm changes stride when out size is correct,stale/low urgency,2,stale/low urgency,1,0,48281,,,,0,ffeee3e56adf050e74a60f24d16a96fe130bf02c70c5c52ce5b4eb4643ef2a00
thread:issue:48287,"issue #48287: [doc] Error ""You've reached a dead end"" when opening torch.__config__ in the docs",stale/low urgency,2,stale/low urgency,1,0,48287,,,,0,6531da490f7f1897f2794c6e31bd1824d1acc3fe3179776288ca278965d47d6e
thread:issue:4829,issue #4829: [Feature Request] clip_grad_norm for sparse gradients,stale/low urgency,2,stale/low urgency,1,0,4829,,,,0,4bfbcdfaff249cdc9d6d84041ac5f37b1daeabe7d4d2b670d5bf5e1e2f02929c
thread:issue:48291,issue #48291: pytorch is not linked with support for cuda devices,stale/low urgency,2,stale/low urgency,1,0,48291,,,,0,38b548c941ebf94afd3def84f4fef17e76a2ceee76860c2cf8ecb12dfdd10f11
thread:issue:48305,issue #48305: [RFC] CUDA-aware future for distributed,stale/low urgency,2,stale/low urgency,1,0,48305,,,,0,c429339820c48208d41266f5866f747c4b4de6fb3c183f74afbf16c454629a34
thread:issue:48306,issue #48306: [doc] example for pairwise distance matrix,stale/low urgency,2,stale/low urgency,1,0,48306,,,,0,00c8f7232acef4801a8f5f512780f9d47c671d3941c5e12ca4c77fd6039a9e7f
thread:issue:48338,issue #48338: torch.float128 datatype,stale/low urgency,2,stale/low urgency,1,0,48338,,,,0,e638c31a2e42265791b156b5aa2175d5b0bb6fcd68e383007acf53aeb3967dcf
thread:issue:48366,issue #48366: Add Go to high level interface,stale/low urgency,2,stale/low urgency,1,0,48366,,,,0,0d89844b79d5da18c68c4a9c3a4a0be13b2f64090421ce9e81f3919443726dba
thread:issue:48377,issue #48377: Management of Gpu memory to avoid memory errors,stale/low urgency,2,stale/low urgency,1,0,48377,,,,0,0ff44db3908d4b23565a003b207fa71094b4c7591f33a0b65c48ece9436fe5f2
thread:issue:48439,issue #48439: cudnn convolution modifies the input Tensor metadata inplace when it tries to `.resize_()` it,stale/low urgency,2,stale/low urgency,1,0,48439,,,,0,d195ebfcfc7f37b931d2801485926284287c2cc02853df457011ef75d10e4c74
thread:issue:48450,"issue #48450: Qt 5.14.2, libtorch1.7.0 cuda10.2，Error after using header file",stale/low urgency,2,stale/low urgency,1,0,48450,,,,0,533ac59f34d034ae4ba1c67074aa3e972b3b39138b2d1d533e7967d44b93f96a
thread:issue:48457,issue #48457: compile error immintrin.h,stale/low urgency,2,stale/low urgency,1,0,48457,,,,0,bda5155f2a38cb2cc23e5fa477baa689b7868e88fe9ef94ec9c2254a69deac29
thread:issue:48465,issue #48465: anchored direct links to functions disappeared in 1.7.0 docs,stale/low urgency,2,stale/low urgency,1,0,48465,,,,0,093abea6085bf5f5dc35987e2e2370b60bd101d48ed3bad2aa98568643394391
thread:issue:48466,issue #48466: [master] new default signature: op=<ReduceOp.SUM: 0>from `op=<ReduceOp.SUM>,stale/low urgency,2,stale/low urgency,1,0,48466,,,,0,fdfabea1928ebe72e53580108ba6497631023375219f4f24fec2ce7510fa1116
thread:issue:48483,issue #48483: Vulkan Api Backend Build Error with GCC,stale/low urgency,2,stale/low urgency,1,0,48483,,,,0,0ceba8dce369119a3bd5b0764e53e58dfbc6f4a6ac462b15771086412515212b
thread:issue:48485,issue #48485: Add wheels for all cuda versions on pypi,stale/low urgency,2,stale/low urgency,1,0,48485,,,,0,fea04959f1492211fef343597fa4ccd3bde2b5ac9dff3e85b6ed946a85a4d933
thread:issue:48486,issue #48486: [complex] torch.abs: does not match numpy,stale/low urgency,2,stale/low urgency,1,0,48486,,,,0,7c17eb7afec89b2a59a0c8df05d0efaec7f8b7dd0b69a792f6d7508fc1497ec9
thread:issue:48511,issue #48511: Ambiguous RuntimeError raised when device of input PackedSequence does not match with nn.LSTM's device,stale/low urgency,2,stale/low urgency,1,0,48511,,,,0,586f56532adf25b1827c99f983762fff0ca403adeb42536a623af357404f5abd
thread:issue:48518,issue #48518: C++ optimizer check for duplicate parameters,stale/low urgency,2,stale/low urgency,1,0,48518,,,,0,5370a99173afe23480094b38a0991574ded49cdec80a9a4eea10812d836ca482
thread:issue:48526,issue #48526: CARU: A Content-Adaptive Recurrent Unit for the Transition of Hidden State in NLP,stale/low urgency,2,stale/low urgency,1,0,48526,,,,0,fb0de30bcea7111ae54d71e48431e3842ed8d9b0e2059c7b8df06dbce063fa51
thread:issue:48587,issue #48587: [feature request] Give `torch.cholesky` an optional fallback to test whether the tensor is positive definite,stale/low urgency,2,stale/low urgency,1,0,48587,,,,0,1f8a7cae07fb6bbf5cb397188dc5df12d15604965ecdee372e8f7718e5b8529a
thread:issue:48605,issue #48605: [NNC] Bugs Exposed in Binary Op Testing,stale/low urgency,2,stale/low urgency,1,0,48605,,,,0,b490a1ec79d99b471249c84d6bd0c7339527fa8112edc1bd3894daee15c335ff
thread:issue:48621,issue #48621: [FR] loss reduction argument accepts None,stale/low urgency,2,stale/low urgency,1,0,48621,,,,0,0ebc85c93b059dc388655e4a97d1372bfd536bf02386f29b9cb47171e40ca750
thread:issue:48626,issue #48626: Missing super().__init__() call in nn.Module,stale/low urgency,2,stale/low urgency,1,0,48626,,,,0,18d0d8f4dcb21e87d1e73075ee6f963de26f01725875a404f82f414a720a64d8
thread:issue:48641,"issue #48641: [bug] `torch.{sinh, cosh}`: Incorrect values for vectorized path",stale/low urgency,2,stale/low urgency,1,0,48641,,,,0,d71cd5f9f2b9642ed20a66ff02a979ae0d5073d584d524de0af5ca11fa4441d4
thread:issue:48645,issue #48645: Can sublist a nn.Sequential subclass,stale/low urgency,2,stale/low urgency,1,0,48645,,,,0,de24e719ecb0f065ca24cd7c3c8afeed13c93b9bbce3ed7e7db75ac86147d791
thread:issue:48648,issue #48648: Confused on project website pointer.,stale/low urgency,2,stale/low urgency,1,0,48648,,,,0,46ee39c39ca3a8c894c327f115fe3a2edfedf22011658616b84b0628a6e9487e
thread:issue:48649,issue #48649: [docs] A single additional summary page for diag* methods,stale/low urgency,2,stale/low urgency,1,0,48649,,,,0,938566326b5442b15e418b6803baa0df5ea1fad852c467764dbf998a18affcb6
thread:issue:48650,issue #48650: [feature request] Keyword-only device argument (and maybe dtype) for torch.meshgrid,stale/low urgency,2,stale/low urgency,1,0,48650,,,,0,8c181fff6a10ff5a2b820759b0d6f8bdd7d86fa0ff5d1f9464c763c628d210ae
thread:issue:48667,issue #48667: Operator registration doesn't work with noexcept functions on some compilers (c10::guts::is_function_type rejects noexcept),stale/low urgency,2,stale/low urgency,1,0,48667,,,,0,d98b0b9b07bfb7893ca0993e60cf6e94b0b64898d36fc7b420bc5d19fe935a92
thread:issue:48684,issue #48684: Remove remaining native:: references from non-native ATen,stale/low urgency,2,stale/low urgency,1,0,48684,,,,0,e4ac49fbc2fc709d7e3b5e3b2626a31f325ab23d9dd1508f96184c96d5fc1ad8
thread:issue:48702,issue #48702: provide example for distributed training with iterative dataloaders,stale/low urgency,2,stale/low urgency,1,0,48702,,,,0,bdb36d5ec2758975ad9e93977bb3160046e61e23c4db655e1730aa92fcb8dcfa
thread:issue:48706,issue #48706: test_inverse setup is flaky using MKL>=2020.1 on certain CPUs and fails on CUDA,stale/low urgency,2,stale/low urgency,1,0,48706,,,,0,7a9c6a55959b60f06402acb969b463f0a207a0caa99fb631df5c0dd942dec6f1
thread:issue:48714,issue #48714: RTX 3090 setup vs 2x RTX 2080TI setup slower? Help..,stale/low urgency,2,stale/low urgency,1,0,48714,,,,0,7b835074ec9151eed9b98bb7459c4f88d7ece3915810c0ac0837c538786ee0e0
thread:issue:48813,issue #48813: inception_v3 is not symbolically traceable,stale/low urgency,2,stale/low urgency,1,0,48813,,,,0,97cae57bc7bedbef1d566f2588f9a1e623ada5ff8a37b2eb4268b5b425fafc62
thread:issue:48832,issue #48832: cudnn cannot be pickled by cloudpickle,stale/low urgency,2,stale/low urgency,1,0,48832,,,,0,ed3db658959fa6825082d53450d9a76fa4991f7878f4c20c414200aea1d3b6a1
thread:issue:48845,issue #48845: memory_format kwarg doesn't work on most factory functions,stale/low urgency,2,stale/low urgency,1,0,48845,,,,0,6a839e37804c07db71f131116786405d9c3f5edf1847cda44ea8c7355999d67b
thread:issue:48892,issue #48892: is_non_overlapping_and_dense() does not error out for sparse tensors,stale/low urgency,2,stale/low urgency,1,0,48892,,,,0,cd6d31e199ad35494c303e1a8455b45301c135acd469aadc3132cd478039ee9d
thread:issue:48900,issue #48900: addmm with out= argument returns incorrect result,stale/low urgency,2,stale/low urgency,1,0,48900,,,,0,01cec12e2f96570226ac637e003e4ca1a6f010c1ec828ab55fb81ab6266c9e9f
thread:issue:48929,issue #48929: Domain Specific Batch Normalization,stale/low urgency,2,stale/low urgency,1,0,48929,,,,0,5de19dec5d55b7ac0fc6d858c699bc60b282f9c40274504e054cb832fa77dea3
thread:issue:48998,issue #48998: Torch1.2 can't be downloaded!,stale/low urgency,2,stale/low urgency,1,0,48998,,,,0,4c6e40da5b3fe24de2b4d465a7793f5e27ca6f2cc2559f25604eaf2a36280975
thread:issue:49023,issue #49023: Automatically rerun tests with CUDA_LAUNCH_BLOCKING=1 when they fail with CUDA errors in CI,stale/low urgency,2,stale/low urgency,1,0,49023,,,,0,4c3521c4cff1e5f96d8012b553726762a62b40dc8ea67a3df4350d4c59cd0ed4
thread:issue:49038,issue #49038: Make the forward function of a nn.Module and/or a certain function to work with only float32 when used with autocast,stale/low urgency,2,stale/low urgency,1,0,49038,,,,0,947b41de549923b2d9a7291298520e93da7bec2e806d58c36cf1b555691ec13e
thread:issue:4906,"issue #4906: Rebuild from no-CUDA to CUDA leads to: error: #error ""Expected GLOO_USE_CUDA to be defined""",stale/low urgency,2,stale/low urgency,1,0,4906,,,,0,1c8883dc5162c4a03c41c5a472472bfe19f1a7747e9fffe3515b146709c04c1d
thread:issue:49076,issue #49076: Add List Type Data through add_hparam method in torch.utils.tensorboard.SummaryWriter,stale/low urgency,2,stale/low urgency,1,0,49076,,,,0,22557ed129d21b765224c14efabe9ee2ff6c0a8afce49fb23caaa7d4ee540471
thread:issue:49082,issue #49082: swa_utils.bn_update is too opinionated in how it calls the model,stale/low urgency,2,stale/low urgency,1,0,49082,,,,0,1e5a7a92d79b37ba3b77217efebaf008934dd5a244ea27cf9eab48576520abf2
thread:issue:49095,issue #49095: NCCL Error when training with 2x 3090s.,stale/low urgency,2,stale/low urgency,1,0,49095,,,,0,a62acae0b33852f43320d303de5ba88e5683a32ea9d8352956c7519b27f48734
thread:issue:49101,issue #49101: [RFC] RemoteTensor: Use Remote Devices as Local Devices,stale/low urgency,2,stale/low urgency,1,0,49101,,,,0,c372c390c9ca9ffc6cd8f8748ee231ac44e33f20af7d9ce7408081241dd5deda
thread:issue:49110,issue #49110: [RFC] Integrate DeepSpeed CUDA kernels,stale/low urgency,2,stale/low urgency,1,0,49110,,,,0,45141b330b389f510b76f67d14b0ea7ae74b0dda144e46033480778c3bd6382c
thread:issue:49133,issue #49133: Unknown failure for test_reference_numerics_sinc_cpu_complex64,stale/low urgency,2,stale/low urgency,1,0,49133,,,,0,a5229955235d077cb3b5c095ac7600ac72e31801e0c1763909d6ea0a1fafe2a0
thread:issue:49139,"issue #49139: Hello everyone, why does libtorch half precision float 16 leak memory?大家好，为什么libtorch半精度float16会内存泄漏？",stale/low urgency,2,stale/low urgency,1,0,49139,,,,0,7ebf083ccdadbf9fe6c35e6677286d7e7768969ae12269bd658f77481712245a
thread:issue:49166,issue #49166: Implement fill_out on complex,stale/low urgency,2,stale/low urgency,1,0,49166,,,,0,bafe2cc22e03e11fbbc8f87941ff1af7f24f57221164a039fb1a9d7a40a61aae
thread:issue:49177,issue #49177: Investigate `torch.linalg.norm` performance,stale/low urgency,2,stale/low urgency,1,0,49177,,,,0,3b7fcc735a22c5f60c6ee7360c1062ff6b3c94914503ea4befc2a71dadd87c0d
thread:issue:49178,issue #49178: [NNC] Some ops have type promotion logic which adds extra casts & does compute in different dtype than eager,stale/low urgency,2,stale/low urgency,1,0,49178,,,,0,2e934eba8e138d4e61d0f074933681a9b260327b91f6b57cc3a360b88bb62057
thread:issue:49180,issue #49180: torch.utils.data.DistributedSampler allow uneven inputs,stale/low urgency,2,stale/low urgency,1,0,49180,,,,0,6b413237d5200138712aa8e303aee0a0b46c1f01c7dc809dc72691dc8960b507
thread:issue:49207,issue #49207: Provide a mechanism to limit the workspace size of cudnn convolution,stale/low urgency,2,stale/low urgency,1,0,49207,,,,0,115ede8e323d7c5120b2efe7dac58b3c33a6c7c10cf4ad4b81001bcfd0e33fb4
thread:issue:49217,issue #49217: Bug in profiler on CI machines when profiling NCCL distributed calls,stale/low urgency,2,stale/low urgency,1,0,49217,,,,0,23fbb1372da341f4c3b7e04490c1a0a6eab08654874ce43017ac975b00af3d10
thread:issue:49224,issue #49224: Torch abs op crashes on -128 int8 tensor with ASAN,stale/low urgency,2,stale/low urgency,1,0,49224,,,,0,66a729804e7fce26d4148ad9390df322f4b74c1dcbf69fb0a0532501f617bca1
thread:issue:49265,issue #49265: [FX] Update `placeholder` docs to better reflect the role of `node` and `target`,stale/low urgency,2,stale/low urgency,1,0,49265,,,,0,275f1f3785de37c55a6e8b60ea2a67bf02489db8a934a1c7028aab66b12b747f
thread:issue:4927,issue #4927: [Feature Request]would PackedSequence support unsorted sequences?,stale/low urgency,2,stale/low urgency,1,0,4927,,,,0,5debbf1073eb66c021fb657e10394837a387a3dbcc0957f207cf9afa802d4a15
thread:issue:49282,issue #49282: Bug when combining AMP with channels_last memory format,stale/low urgency,2,stale/low urgency,1,0,49282,,,,0,04672b2184d568db588a374f2280d09ceaef415d85eab6c2d050a11cc49cfc95
thread:issue:49284,issue #49284: TensorBoard SummaryWriter to remote storage unreasonably slow,stale/low urgency,2,stale/low urgency,1,0,49284,,,,0,7ec945f7786a8aee6f93eecda8d8b82b1fd15af78813b5ec2f91d3061628649e
thread:issue:49291,issue #49291: add inverse cdf for Chi-square Distribution,stale/low urgency,2,stale/low urgency,1,0,49291,,,,0,89b3d49242372ac9b618fb1649c1e4a8b2bb61dd629d270833cdf31db2b2a035
thread:issue:49298,issue #49298: [question] How hard would it be to implement 4-bit precision training?,stale/low urgency,2,stale/low urgency,1,0,49298,,,,0,bc02a6cf49184bf3cc7fb07948f4465ff64f3ce0187ec9a384930b0725210f3d
thread:issue:4930,issue #4930: [feature request] Type-1 Multi-layer bidirectional RNN,stale/low urgency,2,stale/low urgency,1,0,4930,,,,0,e92ae6d4f1847df9d88a8b3e820ae6f12cee50c2f4bba0dacb8fac70d0916d5f
thread:issue:49300,issue #49300: is this a typo in optimizer.pyi ? it says `statue` instead of `state`,stale/low urgency,2,stale/low urgency,1,0,49300,,,,0,a0aa56a18a4c2dd4349d7ee45e9c043e4e590eb9ebd4a8751ffcbc7a0034f727
thread:issue:49320,issue #49320: how could I print the log in source code,stale/low urgency,2,stale/low urgency,1,0,49320,,,,0,4f2c202368b85c89bca9d7bba4e9b4b7990020e42f46b5eb86b1dc7fcfa90ca5
thread:issue:49332,issue #49332: global_pruning costs memory after model is trained,stale/low urgency,2,stale/low urgency,1,0,49332,,,,0,26b4c45b486a2e10a8f8c494d4a48b00104daba90cdd9f3c30f2b4a7d04312d2
thread:issue:49347,issue #49347: [pre-commit hook] fails on `BLK100 Black would make changes.`,stale/low urgency,2,stale/low urgency,1,0,49347,,,,0,cec071ac8fe2cdc33fa89b7b104e460680d97dd7b603650cd46835ec76f97f75
thread:issue:49370,"issue #49370: PyEval_SaveThread: the function must be called with the GIL held, but the GIL is released",stale/low urgency,2,stale/low urgency,1,0,49370,,,,0,e70d8b9a37845f41cc27fd880832302ce8f02e7cb78f346dbc572c05c6838a8b
thread:issue:49394,issue #49394: Apparent Memory Leak with torch.as_tensor,stale/low urgency,2,stale/low urgency,1,0,49394,,,,0,63465c28799a86f990ed3890c1ae768e06c82e31bdd7c1abbca010163c9b207d
thread:issue:49398,issue #49398: Can we add try-except for list/slice indices even when auto_collation is True?,stale/low urgency,2,stale/low urgency,1,0,49398,,,,0,ab26d0c75f67e29b18faba694fd270b6dfe78fd98688de41bdad85480bf7fde2
thread:issue:49405,issue #49405: PyTorch 1.7.1 on (macOS/python 3.9/conda) links libtorch_global_deps.dylib with libomp.dylib instead of libiomp5.dylib,stale/low urgency,2,stale/low urgency,1,0,49405,,,,0,81a4b2a0d19d26a1989026ebc5fa1765c43ae469b463c6451c6f397e99b2010c
thread:issue:49406,issue #49406: Bitcode enable for iOS,stale/low urgency,2,stale/low urgency,1,0,49406,,,,0,722ab729d3901503b8bfcfe7b57d414372b2dc9e26f168be8983374cd0d24652
thread:issue:49409,issue #49409: Add global gradcheck setting,stale/low urgency,2,stale/low urgency,1,0,49409,,,,0,8c8c2a7daad586b9e4deb6344767ba39410920d914e7e2e9d74cb3278a67e053
thread:issue:49416,issue #49416: A few years after #701 and PyTorch is still using implicit __all__ imports.,stale/low urgency,2,stale/low urgency,1,0,49416,,,,0,96ac0b8e6388cb070cf397e8d7caf0c67802b9a24fee061b18f2096ebfbb19b1
thread:issue:49426,issue #49426: [doc] missing torch.pdist entry,stale/low urgency,2,stale/low urgency,1,0,49426,,,,0,8c21ed7af2b70c99a77416df25a34d1b3193d46b2424d65cb4a7f3a7c7eabb08
thread:issue:49427,issue #49427: Inserting named tensor into other fails,stale/low urgency,2,stale/low urgency,1,0,49427,,,,0,66e5de0af0d3432bae1d2db9aec5478ddf4f79bb4bddf4f41fc369a1c876cae2
thread:issue:49431,issue #49431: clip_grad_norm_ performance regression,stale/low urgency,2,stale/low urgency,1,0,49431,,,,0,4b789c36c238922c3a1ee219530c49e5cc92a079ec43151500bd299b1f5946c2
thread:issue:49445,issue #49445: [doc] how to prevent pytorch-nightly from being replaced by a released version on pip install,stale/low urgency,2,stale/low urgency,1,0,49445,,,,0,1b5d9ef9c56248fd221af7614cab6d64eb9ea921aa6ac949b7b663b802db54c5
thread:issue:49456,issue #49456: Impossible to run tests target with LibTorch as dependency inside a cocoapods.,stale/low urgency,2,stale/low urgency,1,0,49456,,,,0,8ae8f622e1c5a3c0ff5835190d4cfbedcc4089569bde9474b41313b8fe20ffec
thread:issue:49475,issue #49475: Write link in README.md,stale/low urgency,2,stale/low urgency,1,0,49475,,,,0,36b6153d7760bfefafc111b4ee8f646ca07648838fa3c0fda28a6fec2e3e65f9
thread:issue:49476,issue #49476: Use METH_FASTCALL protocol in Python bindings,stale/low urgency,2,stale/low urgency,1,0,49476,,,,0,14739472b40c61ed6beaf12a73c59d2652348005bc4b9bb4db445e5628710b98
thread:issue:49481,issue #49481: rank-0 gpu consume too high memory,stale/low urgency,2,stale/low urgency,1,0,49481,,,,0,9db24ad6837f03a960d0a3714c137384aa4040e341f1f16646c394baf1df59d3
thread:issue:4952,issue #4952: [Feature request] Optimize autograd/ATen when a gradient is clearly zero,stale/low urgency,2,stale/low urgency,1,0,4952,,,,0,4ba37d9ec0af6cec0658a37b1efdbf14b6737854e33b837d97693f70ee0fe02c
thread:issue:49539,issue #49539: Improve AssertionError for torch.nn.functional.pad 'replicate',stale/low urgency,2,stale/low urgency,1,0,49539,,,,0,d7a7a360649fa76adf44d55b564e1256f26c1abe5db5109180f634e36fa0cd43
thread:issue:4954,issue #4954: [docs] Docs website search finds duplicates and produces bad snippets,stale/low urgency,2,stale/low urgency,1,0,4954,,,,0,d403913965f21dfaeace8f2127e9fe973c99f5318aec1e260bf610233d68dc6d
thread:issue:49555,issue #49555: Gunicorn preload flag not working with PyTorch library,stale/low urgency,2,stale/low urgency,1,0,49555,,,,0,90b8d038d5001352414d71b45d69115d7d25f35f76bb0c76808378022f08b23f
thread:issue:4958,issue #4958: [Feature Request] Extract glimpses from a batch of images (as in tf.image.extract_glimpse),stale/low urgency,2,stale/low urgency,1,0,4958,,,,0,e47cd6d3de86e923d1350a86dc32c6b5ccd55e635168f5e523208f2badd1c3d1
thread:issue:49618,issue #49618: Support scope in fx Node,stale/low urgency,2,stale/low urgency,1,0,49618,,,,0,7852d7e4aa6f2ebc2c8dc7e810508765c9960d94768447b9877fc6de34806d3c
thread:issue:49630,issue #49630: Bug in CosineAnnealingWarmRestarts,stale/low urgency,2,stale/low urgency,1,0,49630,,,,0,e832e220e793d9d82b2c44b1fd368660b12aa1c574d6a1a30b8b1edad0d2bc0e
thread:issue:49653,"issue #49653: torch.cuda.amp, example with 20% memory increase compared to apex/amp",stale/low urgency,2,stale/low urgency,1,0,49653,,,,0,5b150b8d9e9effcc43010968b2aef0eabec18b5c960d243f4a10f41a50c54637
thread:issue:49672,issue #49672: Model passed through queue hangs while loading in sub process,stale/low urgency,2,stale/low urgency,1,0,49672,,,,0,1b6ceb3308a2457f8b18c99da3bbca2ec5e2dc1cb8c7fe920813dced1d7c4f06
thread:issue:49673,issue #49673: Is there any upper-level API to disable the detection of the weight’s version,stale/low urgency,2,stale/low urgency,1,0,49673,,,,0,2a11fd55034ba3a7ba1a2b7531c0da70610bed4123093211b6120a7a55eca2a4
thread:issue:49677,"issue #49677: File ""setup.py"", line 773:subprocess.CalledProcessError: Command '['cmake', '--build', '.', '--target', 'install', '--config', 'Release', '--', '-j', '8']' returned non-zero exit status 2",stale/low urgency,2,stale/low urgency,1,0,49677,,,,0,a0953fc89fbed738410aa46d7b232fdcb5072699cfebcf216b6c0767eedb402b
thread:issue:49723,issue #49723: Flattening nn.Parameters while maintaining gradients from neural network forward pass,stale/low urgency,2,stale/low urgency,1,0,49723,,,,0,b1883bae2a06277abafdc78a865a8ebdafb305bcd3f9df7b007b6b90c6647495
thread:issue:49724,issue #49724: cure unnecessary NaN values that arise from ±∞ arguments,stale/low urgency,2,stale/low urgency,1,0,49724,,,,0,ab2e3477d75312b49b378732d085e1ac6b8731688400331111851fac9985f06a
thread:issue:49727,issue #49727: RandomSampler generator created in every iteration,stale/low urgency,2,stale/low urgency,1,0,49727,,,,0,46616504b5971da1bff51c79d5a2250d7c838339757e063568d2830ee1b22ec9
thread:issue:49752,issue #49752: IndexError: Input _features.0.weight is undefined!,stale/low urgency,2,stale/low urgency,1,0,49752,,,,0,ecf5e72afb7ec0b5c66b9fb7577e97df64e0b867985c069701183f6ac5a92dba
thread:issue:49784,issue #49784: Compatibility with Raspberry Pi,stale/low urgency,2,stale/low urgency,1,0,49784,,,,0,f79466b1e51cc2b0a29dc1d8f50670890011b629ddadec0751b62d247b10f882
thread:issue:49791,issue #49791: Caching for autoregressive decoding of Transformer,stale/low urgency,2,stale/low urgency,1,0,49791,,,,0,1ea3f449d371bf0377989dc0c98106d73a44965a25cd1011b10bed6566c94663
thread:issue:49795,issue #49795: ninja: build stopped: subcommand failed.,stale/low urgency,2,stale/low urgency,1,0,49795,,,,0,d28f804ade5860d736459f6c374a84b393890b0342eda88dbf9941401e5e9417
thread:issue:49810,issue #49810: Undefined reference to `cv::imread`,stale/low urgency,2,stale/low urgency,1,0,49810,,,,0,c61548156428a66e8adc12f415363cbc2731f45184bcf866d4f310fb602d9c26
thread:issue:49825,issue #49825: Ensure inplace views are properly handled,stale/low urgency,2,stale/low urgency,1,0,49825,,,,0,ea0d145dde4e5e28b822154bfdce3bbf4d728d5f57c03e29734ed9d3af712df6
thread:issue:49844,issue #49844: BCEWithLogitsLoss gives out nan with -inf logits,stale/low urgency,2,stale/low urgency,1,0,49844,,,,0,a1ae5bb1a736dcadba0414243af4024f3d5d07f193209e5b48645cc56132446f
thread:issue:49846,issue #49846: torch init_process_group always hangs,stale/low urgency,2,stale/low urgency,1,0,49846,,,,0,e186b8ff18578b8019da8a19ce072c89c144a493724e52938eccec4fe4c4510a
thread:issue:49849,issue #49849: [dataloader] RuntimeError: Too many open files when yielding integers,stale/low urgency,2,stale/low urgency,1,0,49849,,,,0,56c2bd1f9fbb0d7e89001dfb1492351fdefa30fb91780357ead7e4201eb2ba75
thread:issue:49851,issue #49851: [Feature Request] SuperLoss (NeurIPS 2020),stale/low urgency,2,stale/low urgency,1,0,49851,,,,0,7f75a5b40b778168382427ab3f7f014fca5f5d9369157b8ca701ce8584dc16d6
thread:issue:49865,issue #49865: torch.solve on Jetson is slower than humans,stale/low urgency,2,stale/low urgency,1,0,49865,,,,0,f3f23788915a38e8fec1fa1ba40c172a5095962060094640214f9028f6153a61
thread:issue:4987,issue #4987: MultiGPU hangs Titan Xp in multiprocessing/queue.py,stale/low urgency,2,stale/low urgency,1,0,4987,,,,0,db8aa457f16dd5cca48afbbb77ed11fa1c08689553bdd0de59b918df241a26cc
thread:issue:49873,issue #49873: Accessibility of Website for Low Vision Users,stale/low urgency,2,stale/low urgency,1,0,49873,,,,0,eef96698aa6bbb66dc4e65d17c72646312cb1bf892e5466719bd6df4696bef61
thread:issue:49874,issue #49874: Accessibility Coding Needed,stale/low urgency,2,stale/low urgency,1,0,49874,,,,0,4326a1f3924f858306aec36f4a7e5afd07b103644aae2e123d3e8e0eba072972
thread:issue:49885,issue #49885: LR scheduler `get_lr()` bug,stale/low urgency,2,stale/low urgency,1,0,49885,,,,0,4a3d7a237657625ef9ba0250d734db2dee49c1fc86b7f0ec0708b9f59bc6dc63
thread:issue:49891,issue #49891: Support DDP find_unused_parameters=True mode when combined with Pipe,stale/low urgency,2,stale/low urgency,1,0,49891,,,,0,02b4cc08252da555a36312914d54f3b471da2a61780de71680b04e56a1e00526
thread:issue:49911,issue #49911: [RFC] Speed up python function and arg serialization in RPC APIs,stale/low urgency,2,stale/low urgency,1,0,49911,,,,0,433600ece87bc386290de76469cb0587711201b1efec139636f6422e1f502a0a
thread:issue:50031,issue #50031: Inconsistent function name between stub and implementation in `torch.optim.swa_utils`,stale/low urgency,2,stale/low urgency,1,0,50031,,,,0,52f9507f84d06ccd9e39ff13e420307b3cab4ae6d2729154d10a42e251ee0857
thread:issue:50036,issue #50036: channels_last format convolution is slower than normal NCHW,stale/low urgency,2,stale/low urgency,1,0,50036,,,,0,c01afa54e4c3fc899b7e837d99169fdf7c7306557d11ce227663f63af43b32c7
thread:issue:50068,issue #50068: Magma functions that don't have queue argument create cublas handles for each call,stale/low urgency,2,stale/low urgency,1,0,50068,,,,0,6d8041120d26fc34a6a40681aab77ce1ff9d5d59eb30d15c0e388f133f6cfdef
thread:issue:50076,issue #50076: resize_ documentation does not match implementation when memory_format is given,stale/low urgency,2,stale/low urgency,1,0,50076,,,,0,8c1a37aaaf8bed52b448bc5b5b1f39fe0538da02301b0d8eb83a0b0c7ea2fb7d
thread:issue:50090,issue #50090: Torch native_layer_norm OP out-of-bounds access,stale/low urgency,2,stale/low urgency,1,0,50090,,,,0,be2b6b016ed151a28deb949eafb221d3d102939a50ccab2c9ec222bdcea8a8a4
thread:issue:50094,issue #50094: Broken LAPACK links in the documentation,stale/low urgency,2,stale/low urgency,1,0,50094,,,,0,a425b6f447633c0e597b45c76cc1593b15b03e9ef22aee3384195e31ed880355
thread:issue:50098,issue #50098: Errors when coercing complex numbers of various sizes,stale/low urgency,2,stale/low urgency,1,0,50098,,,,0,d011d9f3791159f8748ff48daf45ee5680573864277affcae2a04c5cfe4e8b5b
thread:issue:50118,issue #50118: torch.where scalar/tensor documentation is unclear and not formatted,stale/low urgency,2,stale/low urgency,1,0,50118,,,,0,72d46fdda0c56509e99856a82c4bb48787e3e626a1de66710cc1646ba8139ad5
thread:issue:50136,issue #50136: Inconsistent Variable Naming in FindTorch.cmake,stale/low urgency,2,stale/low urgency,1,0,50136,,,,0,cfbccaf702dff377c67a58973b77ba1a84596952f639c0aab66aff9b439e2c5f
thread:issue:50143,issue #50143: FX call_function does not work if target is a class member function,stale/low urgency,2,stale/low urgency,1,0,50143,,,,0,4d2fe92c3759f666a331aae3cb93ca4769623feb4b9014f955f4ee90b020c974
thread:issue:50170,issue #50170: Automated Mixed Precision not documented to work with nn.DataParallel,stale/low urgency,2,stale/low urgency,1,0,50170,,,,0,2e3ee1efa9656b2f336c62261eda40376708a00146680401f14ef17d8387112d
thread:issue:50177,issue #50177: [FX] Static specialization of callsites (e.g. chunk),stale/low urgency,2,stale/low urgency,1,0,50177,,,,0,9a0d4af056a780b0db64e554cf3852098095d988318eed905cce2e6c65758a9d
thread:issue:50185,issue #50185: CUDACachingAllocator is not GC-aware,stale/low urgency,2,stale/low urgency,1,0,50185,,,,0,c8f46ff310789cc36a48fbf6bdbd88f5bfa277f49f4aa6ba71d94b30820527b8
thread:issue:50190,issue #50190: Calling emptyCache in Conv_v7.cpp causes performance degradation,stale/low urgency,2,stale/low urgency,1,0,50190,,,,0,e07be89bf08aa4eae7845edc72ad2cc1c9e3bc10704fe15fc889d7e5d9ec81d6
thread:issue:50198,issue #50198: deterministic implementation for adaptive_avg_pool2d_backward_cuda,stale/low urgency,2,stale/low urgency,1,0,50198,,,,0,8810ee5474ed98e324c0b2982fc9e3a4d7257aa647f4ce074008ccd311eabd0a
thread:issue:50201,issue #50201: Excessive memory usage caused by Samplers storing lists of indices,stale/low urgency,2,stale/low urgency,1,0,50201,,,,0,84e7a5898b246659488f5da8015f75da74ebb41b8e8a4cf396436fad5c130f3a
thread:issue:50206,issue #50206: Build Fail with vulkan on ARM64 and wayland drivers,stale/low urgency,2,stale/low urgency,1,0,50206,,,,0,7cc22e21803ebb505647eb0b327d130558fb328f69a875f5d99e1f1220ebecbc
thread:issue:50210,"issue #50210: Pytorch 1.7.1 hangs with multi-gpu, while Pytorch 1.6.0 works correctly",stale/low urgency,2,stale/low urgency,1,0,50210,,,,0,c9c21849da52d07549e34ddd6879cb42401696d73c267d356e5eba154c8e7c34
thread:issue:50213,issue #50213: torch.float_power out= and inplace variant errors on non-matching output dtype instead of casting,stale/low urgency,2,stale/low urgency,1,0,50213,,,,0,a616d8c296c267cc4e93987bd407ac08174f79a01845409cd2a75806a2287b71
thread:issue:50219,issue #50219: `requires_grad_` in `no_grad` context returns incorrect value with tensor subclasses,stale/low urgency,2,stale/low urgency,1,0,50219,,,,0,3121b3766388ce1881ce11bc8e9c0a05c4641f2bd6c7288716383a195c8662c2
thread:issue:50258,issue #50258: More general MultiHeadAttention and Transformer modules,stale/low urgency,2,stale/low urgency,1,0,50258,,,,0,e4b8bf0c38438098449891c54dbb3ca35002712e01d277bf8c4c6f809e14bbfb
thread:issue:50268,issue #50268: CUDA error when using `binary_cross_entropy_with_logits`,stale/low urgency,2,stale/low urgency,1,0,50268,,,,0,a8496255296828e5cf662960204d435272785fd32be879515ef996faba5bd152
thread:issue:50273,issue #50273: torch.cross is divergent from np.cross,stale/low urgency,2,stale/low urgency,1,0,50273,,,,0,75a9a329f16e5701d2057042745b1254b863ef6ef05a486dc1390f91edae1744
thread:issue:50274,issue #50274: torch.equal is divergent from np.equal,stale/low urgency,2,stale/low urgency,1,0,50274,,,,0,5eb7fab4a621204e2e16592b6e71b77a4c5710ce440534687d072a0a92ba6a8c
thread:issue:50275,issue #50275: torch.transpose is divergent from np.transpose,stale/low urgency,2,stale/low urgency,1,0,50275,,,,0,a4510d73e48216f4c40d36d452ff1b6f342c15006b0ec75f3b4dbb4c2d773d54
thread:issue:50276,issue #50276: torch.meshgrid is divergent from np.meshgrid,stale/low urgency,2,stale/low urgency,1,0,50276,,,,0,09778a460ef2067648ff5901acbcef92cd7d87f110861501926483a9ed2873ab
thread:issue:50303,issue #50303: Way to solve GPU Host Thread Contention?,stale/low urgency,2,stale/low urgency,1,0,50303,,,,0,a29a9d935aeddcff791bb2d18a8e1ae6da5eeb70531922cc949ac6123d8352be
thread:issue:50323,issue #50323: running a job on multiple gpus with qsub,stale/low urgency,2,stale/low urgency,1,0,50323,,,,0,d2bd3ffb0bdf3a429e9823eb2fdf4fb476e697e2a65dfcca8a72bebbe0bd607a
thread:issue:50328,issue #50328: RTX3090 performs no better than 1080ti,stale/low urgency,2,stale/low urgency,1,0,50328,,,,0,89aaf364ae032bbb676d79017806d7026f2767e0119a4cd3f593771dace60841
thread:issue:50332,issue #50332: torch.Tensor.random_ is divergent from NumPy's np.random.random,stale/low urgency,2,stale/low urgency,1,0,50332,,,,0,a8167e80a5ee1012fbb34e638f371c08bce7c7d359f3bb832f952224e23a09e4
thread:issue:50333,issue #50333: Function Request: scipy.stats.pearsonr,stale/low urgency,2,stale/low urgency,1,0,50333,,,,0,9fc8e5cf01086d2982664d175ab3f505460787896c3db801a72c34b4e9b35f6b
thread:issue:50346,issue #50346: how to save weights when using RPC framework,stale/low urgency,2,stale/low urgency,1,0,50346,,,,0,78885125c5f6fe1df30dc47d597a9efc772def1f1067726c463232912898e0a3
thread:issue:50352,issue #50352: torch.tril_indices is incompatible with np.tril_indices,stale/low urgency,2,stale/low urgency,1,0,50352,,,,0,8359f9cd4ac4da6a741eee3c716358b2766e2e54514a08d9d780e11b98d70e62
thread:issue:50354,issue #50354: for CNN in fp16 execution time depends on input scale,stale/low urgency,2,stale/low urgency,1,0,50354,,,,0,cbc405e67a6501e5bc562559bdd4cb4c8dbb4769a8c292e8e91bc8c83e0e636f
thread:issue:50360,issue #50360: error: conversion from ‘std::vector<at::Tensor>’ to non-scalar type ‘at::Tensor’ requested,stale/low urgency,2,stale/low urgency,1,0,50360,,,,0,02371e69b4b413afc1575f348e6fcaa170202eddc47bb2ab97a9a12c56b68dbf
thread:issue:50366,issue #50366: Optimizer Closure: Enable skip by returning None loss.,stale/low urgency,2,stale/low urgency,1,0,50366,,,,0,ac6774dcc41d421cc57c63baba147a9010ef11e1e947b8db49115d9606febadd
thread:issue:50375,issue #50375: Provide a set of C++ foreach APIs that will take tensor pointers as an input,stale/low urgency,2,stale/low urgency,1,0,50375,,,,0,5cc8b60e68e8d614973278a1d1dea6e1ef949f02b6a3832cba33ed747faa34f3
thread:issue:50402,issue #50402: ValueError: optimizer got an empty parameter list,stale/low urgency,2,stale/low urgency,1,0,50402,,,,0,9ddc9b67d4dff694a6441e4bce344babc49a5f92cd387b16e3d1401f6ae381c6
thread:issue:50462,issue #50462: Standalone OSS RPC benchmark,stale/low urgency,2,stale/low urgency,1,0,50462,,,,0,5bc953f92ae200779856f7778a910259ee124653082f473401ddab2fb13ee752
thread:issue:50468,issue #50468: [RFC] Make RRef proxy APIs non-blocking,stale/low urgency,2,stale/low urgency,1,0,50468,,,,0,008ad99c5b14b82d956354194d092c5ae319a24cd103c5a24d86728d05985cc1
thread:issue:50471,issue #50471: linker error when trying to use Metal backend in PyTorch Mobile,stale/low urgency,2,stale/low urgency,1,0,50471,,,,0,8c389e46ed75a9aa7acc5b7a5d51e1672c52bd8c421412ea980d9ba03650d930
thread:issue:50473,issue #50473: Make use of eps for numerical stability more consistent,stale/low urgency,2,stale/low urgency,1,0,50473,,,,0,1047c1d0dc12e0716d3620795a343838d4372e27fb5ac3d1de4d4959956dd45c
thread:issue:50530,issue #50530: Unable to compile CUDAExtension with Pytorch 1.5 or above,stale/low urgency,2,stale/low urgency,1,0,50530,,,,0,a92176919872dc3b154331d59067249695312e7215b72d18ddd4868c619cb22e
thread:issue:50549,issue #50549: Channels last doesn't improve speed when using SyncBatchNorm,stale/low urgency,2,stale/low urgency,1,0,50549,,,,0,fc860288a6143859621fbd7879cd870f4138ac82d0fe9ebd5acceff9c8191907
thread:issue:50552,issue #50552: test_fn_grad_fft_fftn_cpu_complex128 and test_fn_grad_fft_rfftn_cpu_float64 are failing under TSAN,stale/low urgency,2,stale/low urgency,1,0,50552,,,,0,27858fa4de31e58775c1aac3ee008972855ab23b55a8f890d94edaa15eaee3f1
thread:issue:50613,issue #50613: Discourage slow gradchecks,stale/low urgency,2,stale/low urgency,1,0,50613,,,,0,3658fb926f94ad77be7f0ab08a80a85de405c9be98b3da390bb23f59822e01e4
thread:issue:50616,issue #50616: Use unified type for distributions.constraint API,stale/low urgency,2,stale/low urgency,1,0,50616,,,,0,6a49c83cdafd93d54590025bc2a7db4ccd3ffd3cc9f7be87e937b96eafde9882
thread:issue:50650,issue #50650: build from source fail in jeston tx2 Python3.8,stale/low urgency,2,stale/low urgency,1,0,50650,,,,0,41d7285b5115751a824a3312706b2684cc96d8788ddf15e156406b636337d6fa
thread:issue:50694,issue #50694: CosineAnnealingWarmRestarts lacks verbose functionality,stale/low urgency,2,stale/low urgency,1,0,50694,,,,0,a5b794e77e5ad59f934583e69a549d6379daa86b76f942fe4c46d62f0652b95a
thread:issue:5070,issue #5070: NVIDIA_DRIVER_CAPABILITIES env variable is missing in pytorch docker images,stale/low urgency,2,stale/low urgency,1,0,5070,,,,0,a66d14277aa2e0737fad009d5a1ae69df74ae7e48e17da8ae0b0d07f49e254a0
thread:issue:50718,issue #50718: dependency_links is deprecated,stale/low urgency,2,stale/low urgency,1,0,50718,,,,0,26104791a85e87b1a6e311d240178d62a1dfd26916852385f6139cb93f27784a
thread:issue:50723,issue #50723: [PyTorch Mobile] Can't use Vulkan backend,stale/low urgency,2,stale/low urgency,1,0,50723,,,,0,202805a4faa871cc2750866f5f52671f5b29e5c1048a005f6780db0b80d12f62
thread:issue:50737,issue #50737: Adding visualizations for indexing and other functions,stale/low urgency,2,stale/low urgency,1,0,50737,,,,0,a268b7f3646b102cd5d0688eedfa993e17d226585b73bf26809faed7476b5bb5
thread:issue:50820,issue #50820: NCCL_BLOCKING_WAIT=1 makes training extremely slow (but if not set then OOM on one device will hang training),stale/low urgency,2,stale/low urgency,1,0,50820,,,,0,806a7914849f441657d9894b861c88742982e108485d850b20543bd5325d31fe
thread:issue:50829,issue #50829: pin_memory leads to GPU memory being used but it doesn't initialize CUDA,stale/low urgency,2,stale/low urgency,1,0,50829,,,,0,3d20cf6e65e0f54c44d6332aac1aea7a0cc8a161122e9c3d1652350b098a9748
thread:issue:50830,issue #50830: Pull upstream cmake's change in FindCUDA/select_compute_arch,stale/low urgency,2,stale/low urgency,1,0,50830,,,,0,ae4cce61914756feea7766a81947014c354056d3a4b25766f594e11d25e12403
thread:issue:50831,issue #50831: Dataloader Prefetch data to GPU by cudaMemPrefetchAsync,stale/low urgency,2,stale/low urgency,1,0,50831,,,,0,430f7f4118a17064da702f5009d1a3c28fb07072e8dc365457768e91e7562039
thread:issue:50835,issue #50835: Insufficient shared memory (shm) while training,stale/low urgency,2,stale/low urgency,1,0,50835,,,,0,7c9a17e06ab84e50bd7ddbd9874ffe40fc35decd9b85896c554d79cdd570c9e0
thread:issue:50852,issue #50852: [TorchScript Performance Deep-dive] problems discovered in TS performance deep-dive,stale/low urgency,2,stale/low urgency,1,0,50852,,,,0,a1434e772d7ed10bda640d53b0a5ea4058e82dd765256ae9569706717ba89a7e
thread:issue:50888,issue #50888: Does pytorch support cuda 11.1,stale/low urgency,2,stale/low urgency,1,0,50888,,,,0,759384eefc04fb05682997beb461d55a2359a42427bdbc7f1accd1957565a795
thread:issue:50890,"issue #50890: torch.unique(x, dim=1, return_inverse=True) returns inverse for only the last sub-tensor along dim",stale/low urgency,2,stale/low urgency,1,0,50890,,,,0,70d4cad668960234b8949c8148411d33168d621e82d78535f14c9bd3363e5bf8
thread:issue:50925,issue #50925: THCCachingAllocator::cuda_free_mutex has no effect,stale/low urgency,2,stale/low urgency,1,0,50925,,,,0,f984c4d206976f913e46d4c514b8960b1b643c340b4a6de2d1216559043c6f3c
thread:issue:50939,issue #50939: dataloader bug need help,stale/low urgency,2,stale/low urgency,1,0,50939,,,,0,4db916c494d7ccdf780172f40daf8731d250bf5d428faa2599082d79f8fbf253
thread:issue:50944,issue #50944: libpytorch macos build: static library eigen_blas_LIBRARY-NOTFOUND not found,stale/low urgency,2,stale/low urgency,1,0,50944,,,,0,e2748073a0855512f9bab2e86dd36398a65d96a86a02f5937baafa1cc7b3e8b2
thread:issue:50953,issue #50953: Dispatch-less structured wrapper / composite / alias kernels,stale/low urgency,2,stale/low urgency,1,0,50953,,,,0,41c0ae1f4a4d45e316c1ac7bf477f1ebc0a9eab7e6cd67d8403d6c004f25fa18
thread:issue:5096,issue #5096: [feature request]Add an env variable to cover different pathes when testing code with openmp,stale/low urgency,2,stale/low urgency,1,0,5096,,,,0,a6f76b8ad91c0fbfecfa055aff51e4b8c686e79619a34a4d85d32d4aeed4e174
thread:issue:50987,issue #50987: Scalar/Tensor arg type for op schemas,stale/low urgency,2,stale/low urgency,1,0,50987,,,,0,82a0490577ea6c66290fcc97f5d426f718208acd4b33b46cf581160ed5297ee0
thread:issue:51003,issue #51003: Proposing new features: Developing Echo State Network layers,stale/low urgency,2,stale/low urgency,1,0,51003,,,,0,d6364a144a14903a036e441cecad9e9e72c524e797a8460edb2a147f2f3a0602
thread:issue:51008,issue #51008: Suggestion: Assimilate activation images to PyTorch brand color,stale/low urgency,2,stale/low urgency,1,0,51008,,,,0,21769ec2f73e2e28b5b8b323ee26672c877f8c860a0b041e8322c444b2890e5f
thread:issue:51011,issue #51011: DataLoader is slow in spawned processes,stale/low urgency,2,stale/low urgency,1,0,51011,,,,0,39f0f5a112445e3843d02f8cf4d98f6855783462be9ff8c77ab5b4dbf7a33b49
thread:issue:51014,issue #51014: [pipeline] gpu util + peak mem reporting to tune partitions and chunks,stale/low urgency,2,stale/low urgency,1,0,51014,,,,0,38d5bb70e03430455707c82d4e3bf86e00003a16e17c6622a86333baf524059b
thread:issue:51015,issue #51015: /usr/local/include/google/protobuf/stubs/pbconfig.h:3:37: note: expanded from macro 'GOOGLE_PROTOBUF_HASH_MAP_H',stale/low urgency,2,stale/low urgency,1,0,51015,,,,0,7b4554606d2584494b6add4f0f2ffdc31ddb5e1a29413e98989075d051627ca8
thread:issue:51020,issue #51020: arm64-v8a not compiling due to libpytorch_jni.so,stale/low urgency,2,stale/low urgency,1,0,51020,,,,0,39487038142ceac9b94cf9d532b65ec0354811a5752975ccaf327f950212d931
thread:issue:51038,issue #51038: The pooling code is ignoring sliding windows that start within the right padding,stale/low urgency,2,stale/low urgency,1,0,51038,,,,0,f34e96ef5d055ea0e3c72c124b3acc5ee7690dd605e383cc31c6111e55577153
thread:issue:51044,issue #51044: CuDNN 8 with benchmark=True takes minutes to execute for certain configurations,stale/low urgency,2,stale/low urgency,1,0,51044,,,,0,5fc67c407b079d2ce104fb17360a0ed6ebcacb9c777b3ad931eddc0da268d25d
thread:issue:51054,issue #51054: Class-based structured kernels instruction count regression,stale/low urgency,2,stale/low urgency,1,0,51054,,,,0,dbaf7577ec786c9d2441a8dd7544c1de16046f82a4d9ab3786a60b94bc00a06a
thread:issue:5106,issue #5106: Saving model with runtime code changes,stale/low urgency,2,stale/low urgency,1,0,5106,,,,0,b561a53e169adac00f9ed4c4036ba86e98b4d60ebe412182a6c6c9ec50d22693
thread:issue:51074,issue #51074: error: ‘PyFrameObject {aka struct _frame}’ has no member named ‘f_lasti’,stale/low urgency,2,stale/low urgency,1,0,51074,,,,0,afd124171c712391324d6e51b06563ca633cde4a742dd5420dff3b5220d6ae49
thread:issue:51075,issue #51075: RFC: Private CUDA memory pools,stale/low urgency,2,stale/low urgency,1,0,51075,,,,0,9b02616436ebce2e1c7824694da7fadc21b716f9f6b6fc4216defaae30f63d72
thread:issue:51084,issue #51084: dyld: Symbol not found: _PyBaseObject_Type Referenced from,stale/low urgency,2,stale/low urgency,1,0,51084,,,,0,5ebfbd58715daf9bde67d69300c78fb40fcdcd4d188ecd4e192ffac4b13736c4
thread:issue:51112,"issue #51112: torch.tensor(x) fails inconsistently, assumes 0 index exists in 'Series'",stale/low urgency,2,stale/low urgency,1,0,51112,,,,0,1879a49f039fc86e9c074ec0776fd7545c577a628ca21252ffa5883b5dfe8745
thread:issue:51117,issue #51117: TestLinalgCPU.test_cholesky_solve_autograd_cpu_float64 fails on different random seed,stale/low urgency,2,stale/low urgency,1,0,51117,,,,0,f9f23a13884a4bd183976aa884cf3e80e3350033af4ac112420edce298c12809
thread:issue:51151,issue #51151: -fno-omit-frame-pointer by default in our builds,stale/low urgency,2,stale/low urgency,1,0,51151,,,,0,5083175ffe2047509db1a0500b2c7cb75f7db1e0088fbe3e88f575a34096b9a3
thread:issue:51152,issue #51152: Make it clearer when a tensor was quantized symmetrically,stale/low urgency,2,stale/low urgency,1,0,51152,,,,0,a1acb066e145cc27aae349aeb641f079fb4f263963ac909c42bd88463070e4d5
thread:issue:51197,"issue #51197: [docs] use `versionadded`, `versionchanged` and `deprecated` directive",stale/low urgency,2,stale/low urgency,1,0,51197,,,,0,c36dce7986ea9bfcffb6a6ffbaa261c9fac3b2a4cf1d70a3914754ae21df2d4f
thread:issue:51199,issue #51199: [numpy] torch.ceil and torch.floor don't support integer inputs while Numpy does,stale/low urgency,2,stale/low urgency,1,0,51199,,,,0,7a47de77677a91d66dcf157f519d94362c302340faf633780a5172e0540abbee
thread:issue:51207,issue #51207: Initialize NCCL backend with MPI,stale/low urgency,2,stale/low urgency,1,0,51207,,,,0,1d232ba9a96f6302d82d86553a4d98e2101a3e0c7a07e725e39ac27df458dc95
thread:issue:51210,"issue #51210: Complex->Real cast is a warning, calling `real` or `imag` on non-complex tensors is an Error.",stale/low urgency,2,stale/low urgency,1,0,51210,,,,0,101cbd68ce917047ff3b54a22a632dd19f6e62b551cb4fa200f4697086c263cd
thread:issue:51224,issue #51224: `torch.cuda.device` should look like it is inherited from `torch.device`,stale/low urgency,2,stale/low urgency,1,0,51224,,,,0,deb40ceea9af0cae5afe55f85a6422bee74fd82223168d07e1c6e5e44985154f
thread:issue:51229,issue #51229: std::type_index doesn't provide reliable equality for getCustomClassTypeMap,stale/low urgency,2,stale/low urgency,1,0,51229,,,,0,76c14e62e40a7f0909ae6fc24fdf5ebe856f19677488c45089a07718011b1b2b
thread:issue:51231,issue #51231: macOS error building 1.7.1 from source on Catalina 10.15.7,stale/low urgency,2,stale/low urgency,1,0,51231,,,,0,8f9001ac252528f97b7aaf7120b1f5379e1a82c38554fba690e66fab3b5778e3
thread:issue:51256,issue #51256: [search] duplicated entries in results,stale/low urgency,2,stale/low urgency,1,0,51256,,,,0,032b83a33f6b206d986aeea59f573da66658adc9511ccd8b251bd5604b995273
thread:issue:51280,"issue #51280: [ux] Proposal to have t() === transpose(-1, -2), since batches are very frequent",stale/low urgency,2,stale/low urgency,1,0,51280,,,,0,723ac19bac05a25021bf83069de385ed645ced0c0f49e040e21787238b15989c
thread:issue:51282,issue #51282: Bind tensor.unflatten to torch.unflatten namespace (torch.flatten exists but not torch.unflatten),stale/low urgency,2,stale/low urgency,1,0,51282,,,,0,169d288f56324fa42bde1898bc8ab912721d9452119f7595c8a90ebbb283381f
thread:issue:51284,issue #51284: result_type doesn't take dtypes and doesn't match numpy,stale/low urgency,2,stale/low urgency,1,0,51284,,,,0,fd43bfaba088d9a68d79b619373b3c4f522bb1223cdc23f35250677f74a9e71d
thread:issue:51302,"issue #51302: ""make install"" skips some required libraries for static build",stale/low urgency,2,stale/low urgency,1,0,51302,,,,0,dc66b3b9f1cd758a5951dd09aa48b4a5c3663d49c4fdabf3bca54946438fc7dc
thread:issue:51320,"issue #51320: Pytorch not working properly (I don't know how to summarize it, see below)",stale/low urgency,2,stale/low urgency,1,0,51320,,,,0,c152f7896f63634adb8d0cc247e6892c92c370a4327861d83c7cdd376d7e1efd
thread:issue:51351,issue #51351: Run time error,stale/low urgency,2,stale/low urgency,1,0,51351,,,,0,518f11f64483339ce17125e7b6063ad3f01720fd312c3412a313bd526d5953be
thread:issue:51385,issue #51385: DataParallel copies the model onto GPUs sequentially,stale/low urgency,2,stale/low urgency,1,0,51385,,,,0,9067f74a6344bbb3fbaa939e0495caeb66d6fda2e4536f4fe01125505b74820f
thread:issue:51393,issue #51393: max_pool2d CPU forward performance is poor,stale/low urgency,2,stale/low urgency,1,0,51393,,,,0,2b7ad92a4514a13b7620d27275fa1523b545acfd94f50a35022f0d18791ab805
thread:issue:51406,issue #51406: [numpy] `round` and `trunc` not supported for Integral Type while Python and NumPy supports them,stale/low urgency,2,stale/low urgency,1,0,51406,,,,0,5cffae4cd14d3efc1eb9c2d4a595a0531a96c02ca934e1f585681fb904ccd423
thread:issue:51407,issue #51407: Additon of levenberg-marquardt optimizer in TORCH.OPTIM,stale/low urgency,2,stale/low urgency,1,0,51407,,,,0,773cdfd435975883ff743a2985a88c4671f0575ec363447351784ec017766e6e
thread:issue:51410,issue #51410: Caffe2 RNN Cell Names not rectified,stale/low urgency,2,stale/low urgency,1,0,51410,,,,0,4b8174bef56dfa4bfd783c34ccf37a196274feadbae5046b7f5b1274e23ca65e
thread:issue:51418,issue #51418: Pytorch num_worker>0 code worked first time and then it never worked with same setting again,stale/low urgency,2,stale/low urgency,1,0,51418,,,,0,2f12400128b758744e256ce3783efad72e445449eb430d1a4b44abb02ea1a2eb
thread:issue:51426,issue #51426: Different results on RTX 2060s vs GTX 1060 with gpt model,stale/low urgency,2,stale/low urgency,1,0,51426,,,,0,296b070a2f04f48a02a979019265ab66f0f7b167e913d7ffdc38136de02de4a7
thread:issue:51429,"issue #51429: dist.init_process_group('nccl', ...) hangs in some combinations of pytorch+python+cuda version",stale/low urgency,2,stale/low urgency,1,0,51429,,,,0,d8d5f9325074da846bf0f45e6de9f27cc370f34b844548e093e8b7c6461effc3
thread:issue:51440,issue #51440: Support unity 3d,stale/low urgency,2,stale/low urgency,1,0,51440,,,,0,c566add4f1911a8f73e538af705a78849ade0911476b51d105f5b28aab672ce4
thread:issue:51451,issue #51451: Add support for conan package manager,stale/low urgency,2,stale/low urgency,1,0,51451,,,,0,f32f111bbb6c73d6f3d8d0a82c6237126a7f668c4db00d2f8afa53f97bc49809
thread:issue:51456,issue #51456: Optional tensor type,stale/low urgency,2,stale/low urgency,1,0,51456,,,,0,25bfc568bef599523843081ea76672a03bbea96782cac3995f1c0341ed0e0ccf
thread:issue:51493,issue #51493: Add branch predictor hints to prefer `Context::deterministicAlgorithms() == false`,stale/low urgency,2,stale/low urgency,1,0,51493,,,,0,c5af791dfbb9e8b89ab867ed1d00aa2d084ce2f1edb2d6802b057a15a408c609
thread:issue:51511,issue #51511: Python-3.9: Importing numba and torch triggers internal error,stale/low urgency,2,stale/low urgency,1,0,51511,,,,0,e1710cd2bb8d8558456d833a7e4673fcd7b7a1440539d796d2e80b922cd2c5ab
thread:issue:51535,issue #51535: Why isn't there a pip-installable PyTorch for Raspberry Pi?,stale/low urgency,2,stale/low urgency,1,0,51535,,,,0,3b1179e149b6a2dd0d15a16e1dc36bc055f7b52d5a10a6a209521c6cc7580d61
thread:issue:51558,issue #51558: torch.reshape fails to keep the memory format,stale/low urgency,2,stale/low urgency,1,0,51558,,,,0,429a8beeeae9f577dc5a2cda02ef69296dd837228d02677aaa94b3d45fa79f67
thread:issue:5157,issue #5157: BCELoss - weight parameter shape incorrect,stale/low urgency,2,stale/low urgency,1,0,5157,,,,0,4eaca17a5de855bcbc97be05fcb402f27ce925439f8997cc9c35c63f1f55a2d2
thread:issue:51575,issue #51575: [doc] being able to switch version while keeping the same doc,stale/low urgency,2,stale/low urgency,1,0,51575,,,,0,05495cf3b0e0baa3ccc9029cd13db486c7167af8bfff62a94a3f2b4f2e7573f4
thread:issue:5159,issue #5159: TakeBackward taking a significant portion of backward time,stale/low urgency,2,stale/low urgency,1,0,5159,,,,0,05f1845410e815c31b8ecc16ba58a2e8ab097391b8a263e021896265e713d6c0
thread:issue:5161,issue #5161: ASAN detected leaks on python -c 'import torch',stale/low urgency,2,stale/low urgency,1,0,5161,,,,0,910c49020c036a45d377660a23a71b792ffdbb9aaa1e67a235fb18599b5e1c9e
thread:issue:51622,"issue #51622: `fractional_max_pool{2, 3}d` inconsistent between CUDA and CPU",stale/low urgency,2,stale/low urgency,1,0,51622,,,,0,fe4ff5ececf940995304260e7132fd120df4d73895a9506cf96b9d756548fbfc
thread:issue:51623,issue #51623: [docs] FractionalMaxPool2d : _random_samples arg is present in signature but undocumented.,stale/low urgency,2,stale/low urgency,1,0,51623,,,,0,3a711d8af7be8b1a813e2a8e50e9c17dcc00f5d50df83c0b95b6b59d705f8e31
thread:issue:51644,"issue #51644: Segmentation Fault: Garbage collector, cuda memory",stale/low urgency,2,stale/low urgency,1,0,51644,,,,0,f8efba4ef05bfc3072477925c487706c558db1c31934eb0e999e0218bf4942f9
thread:issue:51663,issue #51663: Torch DataLoader fails to reraise error from sqlalchemy.,stale/low urgency,2,stale/low urgency,1,0,51663,,,,0,628c0acc95255f0d0cd2dc5d0e04d9927e8781a7745d26b523259c48ab5f92c3
thread:issue:51711,issue #51711: Weird behavior in Conv2d padding when changed to 'reflect',stale/low urgency,2,stale/low urgency,1,0,51711,,,,0,4b66e498376c796ed0a7139a1fd3a4d3e00fa9abb749a120e0edce118dc3583f
thread:issue:51714,issue #51714: Inconsistent result between NumPy and PyTorch for 0^(z) when Re(z) < 0,stale/low urgency,2,stale/low urgency,1,0,51714,,,,0,6234a3ccc87cc490bf72dcb154c65f281f60a1931e1bb74bdef56841f587b950
thread:issue:51776,issue #51776: BatchNorm3d error : CUDNN_STATUS_NOT_SUPPORTED,stale/low urgency,2,stale/low urgency,1,0,51776,,,,0,596522b7e99f1baaf5baf8a8ce400f5d44a9829a1e810dbcd81ea637bbf679c5
thread:issue:51777,issue #51777: Analytical metric package,stale/low urgency,2,stale/low urgency,1,0,51777,,,,0,fd8a1ed9250ae1bca140114d899e758cd4309ccd1247de47b67fbc0bbd90252d
thread:issue:51778,issue #51778: Assertion on the existence of RRef user-facing methods' docstrings,stale/low urgency,2,stale/low urgency,1,0,51778,,,,0,8a534f52a04f931a838f46c017f04ac1c33df56187882b5d8a3e45853fe4661e
thread:issue:51781,issue #51781: Libtorch lock thread,stale/low urgency,2,stale/low urgency,1,0,51781,,,,0,a4b2185e74b6e08edbf74ac9a48904234a977b6cfc91b0d833b68d8bf6d6cfcc
thread:issue:51794,"issue #51794: optim.Optimizer should copy ""params"" before modifying them",stale/low urgency,2,stale/low urgency,1,0,51794,,,,0,cec340b7a52d11a1e5e562c97947ba81ab7f869af98c689e88ec1d3a00e959e3
thread:issue:51836,issue #51836: TypeError: can't pickle _thread.lock objects when using spawn / forkserver,stale/low urgency,2,stale/low urgency,1,0,51836,,,,0,0d91df63300b128f3af615c1c43a2736ed31c9c70d2c4b080b2effe28d148813
thread:issue:51849,issue #51849: .numpy() array failes to keep original storage around,stale/low urgency,2,stale/low urgency,1,0,51849,,,,0,9dc5c9d67450fcef8d03183c1fa52ef7414cd10c6ffd5c22b6c9bff0039c55b8
thread:issue:51855,"issue #51855: Attempted to read a PyTorch file with version 3, but the maximum supported version for reading is 2. Your PyTorch installation may be too old. (init at /pytorch/caffe2/serialize/inline_container.cc:132)",stale/low urgency,2,stale/low urgency,1,0,51855,,,,0,146309d2ff7cb288683d3fafc59a36c4bd205b0146a8e2023a5d0405619c6235
thread:issue:51868,issue #51868: discrepancies between TorchConfigVersion.cmake to git branch,stale/low urgency,2,stale/low urgency,1,0,51868,,,,0,02f1431db42fc30a654e930524a908dd473621a67fda1496a4e722c0804c23d1
thread:issue:51906,issue #51906: torch.nn.functional.binary_cross_entropy(_with_logits) outputs NaN,stale/low urgency,2,stale/low urgency,1,0,51906,,,,0,1ac5bd210bc0f269bebe6a48c8e5896ab412243b4b3723b72189eed136882dd1
thread:issue:51911,issue #51911: torch.nn.functional.grid_sample outputs NaN,stale/low urgency,2,stale/low urgency,1,0,51911,,,,0,daba15c686c353e2328b681ee6cca9230c8dfac4a32682e821f2133de1202afe
thread:issue:51933,issue #51933: Performance bugs of transpose2d on A100,stale/low urgency,2,stale/low urgency,1,0,51933,,,,0,dd5fb95df8e77bc4400a25f7d75110bd82c9eddac8af64d69caa2e98bee06d4a
thread:issue:51965,issue #51965: Feature Request:Fast Cosine Transform,stale/low urgency,2,stale/low urgency,1,0,51965,,,,0,73b98b2642db2d0dd218ac91741efa168d2e4e45f321a1f7ce0ea342fb6f8db8
thread:issue:52005,issue #52005: Implementation of Normalized Euclidean Distance (NED) in pytorch,stale/low urgency,2,stale/low urgency,1,0,52005,,,,0,37fa332b1526b99dbcb2ad8bf2f4ed83165221a37312d4bfa3f6c143268e2f93
thread:issue:52011,issue #52011: CPU eval BatchNorm2d is not threaded,stale/low urgency,2,stale/low urgency,1,0,52011,,,,0,c8c64224bad6f5235634dee9ef241050316cf07e5a5678e45b4ee85aa67ca3a9
thread:issue:52022,issue #52022: [FX] Interpreter doesn't GC values,stale/low urgency,2,stale/low urgency,1,0,52022,,,,0,ca854af7d9a9d82ccaf10a0ad539e37d165943d7017c1468185615137469a748
thread:issue:52040,issue #52040: torch.distributed.TCPStore doesn't work with dual IPv4/IPv6 network interface,stale/low urgency,2,stale/low urgency,1,0,52040,,,,0,39f8fc884946090c87fc8df0fa41c6819aa048068a8bf5d95dc98bd62ac32cac
thread:issue:52041,issue #52041: device (+dtype) arguments for torch.stack / torch.cat,stale/low urgency,2,stale/low urgency,1,0,52041,,,,0,ecc84c93eff8d5c4c0954568bb8def6e82025312c95dd82205da34aebbedbb74
thread:issue:52052,issue #52052: common_jit.check_against_reference() should check gradients w.r.t. weights,stale/low urgency,2,stale/low urgency,1,0,52052,,,,0,2932468c8f7145c0138380689c135f0666608ad9f079bee36ed613c8e8cda972
thread:issue:52096,"issue #52096: Unable to build: subprocess.CalledProcessError: Command '['cmake', '--build', '.', '--target', 'install', '--config', 'Release', '--', '-j', '4']' returned non-zero exit status 1.",stale/low urgency,2,stale/low urgency,1,0,52096,,,,0,f3a545cab9407ce9d814f7918c3a60e1a601f07b9ae3b6fa5b12ccb32fd50d43
thread:issue:52134,issue #52134: Build using Py_LIMITED_API and then build wheels with the stable ABI abi3 tag,stale/low urgency,2,stale/low urgency,1,0,52134,,,,0,538042e9c0caa7f13735acc85ef5b6c5823b061df35d6348ff2f5fd591bd89a8
thread:issue:52145,issue #52145: Trying to initialise CUDA twice in a process with no visible devices hangs the process and terminal permanently,stale/low urgency,2,stale/low urgency,1,0,52145,,,,0,397b61915f0d7b5869dd54f889babf08c51e25cbaf57fb579862306f74d22852
thread:issue:52168,issue #52168: CMake errors building project that uses PyTorch,stale/low urgency,2,stale/low urgency,1,0,52168,,,,0,bec8d5fc96f4140650a89e1e055e49d1800e69c32ad77a34ce0eb6f2e4224bdf
thread:issue:52206,issue #52206: from_numpy function rounds float values on Jetson NX,stale/low urgency,2,stale/low urgency,1,0,52206,,,,0,0d4fecc533579179776d811def06cabdea2a946befe476c2fdf1783fc2920339
thread:issue:52208,issue #52208: C++ standard library confusion,stale/low urgency,2,stale/low urgency,1,0,52208,,,,0,7f4dae57669f50ac50ae81ea086afafacd3b70ebce45c2a24fa01f094abde0f7
thread:issue:52230,issue #52230: Removing mutation on block input should print any meaning message.,stale/low urgency,2,stale/low urgency,1,0,52230,,,,0,159b007de28afb73e133d04faefc792a62341a7ccc7d077c89061a07097bb190
thread:issue:52236,issue #52236: CosineAnnealingWarmRestarts LR scheduler fails when lash_epoch != -1,stale/low urgency,2,stale/low urgency,1,0,52236,,,,0,b589fa120aeb9644aeff1811db708cbe50cc3a5b8b42b94a9d0a639ea96c3e1b
thread:issue:52256,issue #52256: [docs] Bad docs SEO,stale/low urgency,2,stale/low urgency,1,0,52256,,,,0,c63cfe3878f37f5f61a7f71fba451267ea26b3a9bf00cba3da47e578c3c60c75
thread:issue:52262,issue #52262: tolist called on torch scalar does not return a list => proposal to support new arg force = True to provide a work around,stale/low urgency,2,stale/low urgency,1,0,52262,,,,0,632574dace2fbd8a537d3e6fa0f161b7b5a2048b8511ca4b6267016c6a8593cd
thread:issue:52266,issue #52266: GPU (+CPU/RAM?) self-check script or instructions in core,stale/low urgency,2,stale/low urgency,1,0,52266,,,,0,d272b28d525f590a8f99b5c2cf652f1695479c3c18dd249dccbad5e552a2a5a7
thread:issue:52274,issue #52274: Port unary elementwise ops to structured kernels,stale/low urgency,2,stale/low urgency,1,0,52274,,,,0,cfd333712da8f70a737882bae73a2fd391c889b3a4533f5b7c62771189dcd342
thread:issue:52279,issue #52279: [docs] Official apex -> torch.cuda.amp migration guide,stale/low urgency,2,stale/low urgency,1,0,52279,,,,0,3b204c010c11fe6be4959e8dbfce1935a4d9c92823ec8aa72b03f67ccbfd8c3b
thread:issue:52291,"issue #52291: ""LayerNormKernelImpl"" not implemented for 'Half' - CPU",stale/low urgency,2,stale/low urgency,1,0,52291,,,,0,f11447b73d6b66cc63f0890775a2a27efad67d3250534689352ed618725e4a59
thread:issue:52295,issue #52295: Printing a Tensor that is being vmap'ed over in C++ raises an error,stale/low urgency,2,stale/low urgency,1,0,52295,,,,0,dd852eefcc0e3c141aacf49b97e0f731243150187ee6b13ac57231cb98b800d7
thread:issue:52298,issue #52298: [FR} Implement Skellam distribution,stale/low urgency,2,stale/low urgency,1,0,52298,,,,0,5fcb83225565fc312cb928944309c0a6f396f59135e524202e54dc408a2f19ed
thread:issue:5231,issue #5231: [feature request] Stratified splits in random_split function,stale/low urgency,2,stale/low urgency,1,0,5231,,,,0,fa2696dc77df4f1dad44256f3a748eeb5d34fd3461266da942dad719c5f0abf9
thread:issue:52318,"issue #52318: TorchConfig.cmake entries of if(ON) are not accepted by CMake as ""true"", to be replaced by if(1)",stale/low urgency,2,stale/low urgency,1,0,52318,,,,0,a136fc8d7d5caf624866b5a519961a63a325b51e0c20778fe428ede0154fc53f
thread:issue:52334,issue #52334: Normal_like operator,stale/low urgency,2,stale/low urgency,1,0,52334,,,,0,05a0a41ca36ea5c644555f1541601c3dc19a6b2c27209411343f7a792471a904
thread:issue:52390,issue #52390: autograd.functional.vjp with function that mutates inputs in-place can cause confusion,stale/low urgency,2,stale/low urgency,1,0,52390,,,,0,52c341321057a3dde3999aa909809b76094d0331c5faa0cc01aefc734a5fa802
thread:issue:52457,issue #52457: Add autograd tests to verify correctness for R -> C cases,stale/low urgency,2,stale/low urgency,1,0,52457,,,,0,59b3da6480dfc6d4a533348bcb40dd5fb6a98b0f17e8e73e9145f21cee0cacdf
thread:issue:52465,issue #52465: [nnc][perf] Performance decrease with CPU fusion on `freeze(script(pytorch_mobilenet_v3))`,stale/low urgency,2,stale/low urgency,1,0,52465,,,,0,2671ba55390a913873a2a7e7557bf1f03766fd5cd5f26ea70743851d6af3d8ac
thread:issue:52478,"issue #52478: Could you please increase the wheel compiled Linux (aarch64) GPU, thank you very much!",stale/low urgency,2,stale/low urgency,1,0,52478,,,,0,0741c888e96f945f543174d4ef266cc53b5768ad2a467abed71233b8e6773381
thread:issue:52495,issue #52495: link failed when using custom build pytorch on Android,stale/low urgency,2,stale/low urgency,1,0,52495,,,,0,02a367e9a02b59c1b5d96beba8cd18bdbc19dc81759782107f1a5759f9ba8ff9
thread:issue:52538,issue #52538: Helper methods not preserved during AST rewrite,stale/low urgency,2,stale/low urgency,1,0,52538,,,,0,f99933e1f63773acbf39e096d86e3e6ae5c098c1b0cf1105cc1d1a0a5e4de13e
thread:issue:52552,issue #52552: Cumulative integration?,stale/low urgency,2,stale/low urgency,1,0,52552,,,,0,5179eb4df883290ddc61c7c55585223eeac3a59dfe315e4baa6b1b03db5a1e33
thread:issue:52553,issue #52553: Does pytorch have the inverse activation function of nn.LeakyReLU?,stale/low urgency,2,stale/low urgency,1,0,52553,,,,0,c48639a1c98b691f8a18d547eff932f988190a5c5b7cbd07af9359390f9eac8c
thread:issue:52564,issue #52564: Submodules not rewritten during AST rewrite,stale/low urgency,2,stale/low urgency,1,0,52564,,,,0,5ac8e89d4a18f68532ef01df5dd2a83e7b3c63bfae92cfa0d4e430248e22eb3f
thread:issue:52569,issue #52569: Omitting mutability annotations in op schemas can lead to subtle errors,stale/low urgency,2,stale/low urgency,1,0,52569,,,,0,7ddd32deaaaec9930330987a27d124a5bc631c7744fc3fc6c419c416c03b888b
thread:issue:52570,"issue #52570: RTX 2080s performs better than RTX 3080 in Semantic Segmentation inference process(Libtorch,win10),why?",stale/low urgency,2,stale/low urgency,1,0,52570,,,,0,dbca2ccd9a3cb51171d73863492d3bfd489037675278635c962b78a7af45d4c6
thread:issue:52575,issue #52575: Error: ‘str’ is not a member of ‘c10’; did you mean ‘c10::aten::str’? while using libtorch,stale/low urgency,2,stale/low urgency,1,0,52575,,,,0,c381d8813e61db09ec3f6a858d948e81b57cb4f5859b27c2262bb2c3864cf853
thread:issue:52578,issue #52578: version1.7.0 is ~1.3x slower than 1.4.0 on ResNet18,stale/low urgency,2,stale/low urgency,1,0,52578,,,,0,c56b28f7f69212e35dcfa68a033d18ebda4e6a1f1ed40e5ad9f5e35f9c832d3b
thread:issue:52583,issue #52583: [TensorExpr] Add IR Verifier,stale/low urgency,2,stale/low urgency,1,0,52583,,,,0,dc4e4e5e92b160bd9503f7e71453f2cc7afac993f2a34e0db8af060ccbacd592
thread:issue:52596,issue #52596: pickle is a security issue,stale/low urgency,2,stale/low urgency,1,0,52596,,,,0,030d3927a96bc649d9808236f050ae970ab868aa4873a8e2aca2a02db997ca71
thread:issue:52597,issue #52597: Custom scatter function for DataParallel,stale/low urgency,2,stale/low urgency,1,0,52597,,,,0,8c305d7881ebf6f5f521d5e1e47730a18b983bf848ae91c9a70040522de9c04c
thread:issue:52598,issue #52598: CUDA error: no kernel image is available for execution on the device When convert cpu tensor to gpu tensor,stale/low urgency,2,stale/low urgency,1,0,52598,,,,0,3a9217339a33f2863ce82237d5c0705eadddbe03c0dae6b0ace9d5b91c7c23e9
thread:issue:52605,issue #52605: Torchaudio 0.8.0.dev nightly requires torch 1.9.0.dev which is missing,stale/low urgency,2,stale/low urgency,1,0,52605,,,,0,220a8ab7c863ef6c8f8ce879eacaf939c7bdf4a462b730fdaa7f4a181a5561fa
thread:issue:52625,issue #52625: Add slicing to distributions,stale/low urgency,2,stale/low urgency,1,0,52625,,,,0,868d550b87407a97f99e282925d53188194898b0e813dccf52b632b079eb1de7
thread:issue:52644,issue #52644: [FR] torch.load's map_location supports int,stale/low urgency,2,stale/low urgency,1,0,52644,,,,0,3d463d5fbe6f7fe70555cd15562e44aec9f13c9fd3fddfee77d3f62b1022c4ad
thread:issue:5266,issue #5266: [feature request] warnings for functions with unspecified dim arguments,stale/low urgency,2,stale/low urgency,1,0,5266,,,,0,6533b2178463c00629772aca3f64a4ab6804b5c23587c15c6d324b506668f580
thread:issue:52668,issue #52668: Independent axis support for torch.trace,stale/low urgency,2,stale/low urgency,1,0,52668,,,,0,c4a9cc5ce0b205295d9cdd5fc1d80c9b73fc1711e18e540fe968ce41c09a6dc9
thread:issue:52673,issue #52673: Add custom types to PyTorch,stale/low urgency,2,stale/low urgency,1,0,52673,,,,0,bdda674d2c04b2afbc3963475e571033404d08894851bc4e31a5b297af5db050
thread:issue:52680,issue #52680: Add Quantized{CPU|CUDA} support to Structured Kernels,stale/low urgency,2,stale/low urgency,1,0,52680,,,,0,f8431ed313c6f0c935e71bcefbef1aee70ff90ecbcdc344aadadc235bdb403a8
thread:issue:52688,issue #52688: [BE] Consolidating scripts/ and tools/ folder,stale/low urgency,2,stale/low urgency,1,0,52688,,,,0,e34c47c2fbf92b362a6e5f1542c8f52ca0ee274aaf6ad3fb6200a5fdeb23e253
thread:issue:52694,issue #52694: [TensorExpr] Loop Transformations: `compute_inline` should use `Let` for inlined expressions,stale/low urgency,2,stale/low urgency,1,0,52694,,,,0,85c8c6e43c5c7919f2f375939d18e270bed349db3dc2ea1877c878b9f6e6e077
thread:issue:52712,issue #52712: Refactor `torch.linalg.norm`,stale/low urgency,2,stale/low urgency,1,0,52712,,,,0,af6f35caa85a9baf945d354a2fc0c8763049c60d837fadadb7166f4e7f81a4a2
thread:issue:52718,issue #52718: Discrepancy between CPU->GPU and GPU->CPU data transfer speeds,stale/low urgency,2,stale/low urgency,1,0,52718,,,,0,a143ecdc94a619c3be5f013edc2fbdbab4726fc968d9fc106f1686ecca064236
thread:issue:5272,issue #5272: Unsafe out= keyword argument with tensors sharing storage,stale/low urgency,2,stale/low urgency,1,0,5272,,,,0,14b4fdb87683c7019ff7204609dc9daaac5829669ce18c65c947e33a0e4911e5
thread:issue:52722,issue #52722: Creating torch tensor as a function of index value,stale/low urgency,2,stale/low urgency,1,0,52722,,,,0,f7de4c9655378a53b67efa08858f9354d41237217dcd3c16981dfd449d8a75e3
thread:issue:52738,issue #52738: [OpInfo] Improvements for sparse ops tests,stale/low urgency,2,stale/low urgency,1,0,52738,,,,0,a9ff31768dbe4d76d1ee590d1288cfb2cc22c8f80031d68cf722bb442edbee0d
thread:issue:52744,issue #52744: Gradient checkpointing support in C++ API,stale/low urgency,2,stale/low urgency,1,0,52744,,,,0,918e1b6ebb566b9d65a925186b3f7c3c68af720548044072c850a2ec44f31355
thread:issue:52745,issue #52745: Sort torch.topk() output following input index order,stale/low urgency,2,stale/low urgency,1,0,52745,,,,0,4f7b98b3b41bb322809b928135ae5910b14d6ae0be5f25e44dce376f44258851
thread:issue:52747,issue #52747: [proposal] Make torchvision and other libraries' docs searchable via main docs search,stale/low urgency,2,stale/low urgency,1,0,52747,,,,0,e9f8d74e3deee7750a910dc25b8cadbc2120edfbf52576874a6ec0ab945393c4
thread:issue:52762,issue #52762: Basic feature missing,stale/low urgency,2,stale/low urgency,1,0,52762,,,,0,56a9d5a700a483e464b4c2b460f75c5fa8be8406ab79d8ce24ccf32b72b69982
thread:issue:52767,issue #52767: [nnc] Grouped convolutions,stale/low urgency,2,stale/low urgency,1,0,52767,,,,0,dae32ff221d0a2824dcb1cacef0079a39957106397024abad81c64cdf18ebaea
thread:issue:5280,issue #5280: BatchNorm1d raises RuntimeError (CUDNN_STATUS_BAD_PARAM) on 3D input.,stale/low urgency,2,stale/low urgency,1,0,5280,,,,0,907df7cc90d2d1504c8100266f1ecfe19fa94dc1a0cd68e4c931b17225135b3d
thread:issue:52845,issue #52845: [feature request] F.interpolate_as,stale/low urgency,2,stale/low urgency,1,0,52845,,,,0,abd8e713006d0227747ef56cb0d160400db9fa258696e46429f12dc48ffcc6d6
thread:issue:52850,issue #52850: How to skip the images in a custom dataset and deal with None values?,stale/low urgency,2,stale/low urgency,1,0,52850,,,,0,6a7dd83abf48c469368f09fa245d2f87a2b05186a824c3fb153d1936ca951808
thread:issue:52851,"issue #52851: `embedding_bag(..., include_last_offset=True)` should always error if `offsets[-1] != input.size()`",stale/low urgency,2,stale/low urgency,1,0,52851,,,,0,5a7c8c61d10e5ed3a2d95dc68aa296e870b056746bb983b9b36eb2c673f58773
thread:issue:52911,issue #52911: LR Scheduler load_state_dict does not properly update scaling,stale/low urgency,2,stale/low urgency,1,0,52911,,,,0,2f4557396a1295077bddb6b9c2313212dc1c734b28c3b9a330588716cb9fe271
thread:issue:52913,issue #52913: Create a standalone python execution viewer like livepython,stale/low urgency,2,stale/low urgency,1,0,52913,,,,0,374dbe7bba7a09fa1b91eb8bc2d84a6bc0c472d01ffa2c3f0bea9bdb7724aa40
thread:issue:52940,issue #52940: [JIT] Provide a way to access a flattened optimized graph from Profiling Executor,stale/low urgency,2,stale/low urgency,1,0,52940,,,,0,11ea3b37b43b9c501e24e75fd1b472e14552e26a7ccd53ad438f6a929c64c4fb
thread:issue:53023,issue #53023: [codegen] Make it easier to codegen call to API,stale/low urgency,2,stale/low urgency,1,0,53023,,,,0,7454b3316570d44a8725c1552a29e6a1968ef34692d9f3862b028a30147c3e83
thread:issue:53026,issue #53026: Ensure that nn.MHA and functional multi-head attention is supported by FX graph mode quantization,stale/low urgency,2,stale/low urgency,1,0,53026,,,,0,a8d6e9f772f5782b35345437992e4978c1866036cfd8b871bd68c7094c0bbc48
thread:issue:53072,issue #53072: [NNC] Simplify after each transformation,stale/low urgency,2,stale/low urgency,1,0,53072,,,,0,01ce80cffc41e113d8610fd051b4d4e2a565973b875517cf0b63bb4bb5a335ed
thread:issue:53090,issue #53090: 'Tensor' object has no attribute 'astype',stale/low urgency,2,stale/low urgency,1,0,53090,,,,0,83f8d8d1c695745835e7a6ad24dc6880aafbdc2de7dc2e56952d945a22374565
thread:issue:53094,issue #53094: group conv in amp too slower,stale/low urgency,2,stale/low urgency,1,0,53094,,,,0,72e940bd6a0568cd2f2c18174b67d2765db8b815a64c55c312536dac245d2fe9
thread:issue:53095,issue #53095: Reset mask for torch.cumsum?,stale/low urgency,2,stale/low urgency,1,0,53095,,,,0,4fe0f2f052239818092c4b909469a5cfe2b8e3103daed00bfac9ef68b0528edd
thread:issue:53103,issue #53103: ModuleList not recognised as reversible by mypy,stale/low urgency,2,stale/low urgency,1,0,53103,,,,0,f0fc21021ce5e568920da32de6e563a7c5db93a1ba3cb79614e7d842a6a68081
thread:issue:53121,issue #53121: [fx] Handle models with Optional inputs,stale/low urgency,2,stale/low urgency,1,0,53121,,,,0,d54ebf45cdda0ad16f9037c6e902af104fa2f9593fa7657ece0a404b70a11817
thread:issue:53124,issue #53124: Deprecate and remove torch.set_default_tensor_type,stale/low urgency,2,stale/low urgency,1,0,53124,,,,0,f1ec5ec3e248642e3944f11c1e03d2ee745aeb55383d032a76eb2be2e803ea9b
thread:issue:53130,issue #53130: Lint rule to forbid bare assert() in cuda,stale/low urgency,2,stale/low urgency,1,0,53130,,,,0,acec1d8b590989ced8edd2b39067daa4731f8d914581adf8ca08c137624b832b
thread:issue:53141,issue #53141: Structured kernels for operators that don't have out= variants,stale/low urgency,2,stale/low urgency,1,0,53141,,,,0,f8e673410bbe63a3e8cb2d8ff4a89c759eb9df5a0310c2bf9a8000ebc05d2727
thread:issue:53157,issue #53157: doxygen and pytorch documentation,stale/low urgency,2,stale/low urgency,1,0,53157,,,,0,4dadd7b4749c0c2b09794b58fd2781638c6c5450c798ef477fef16e4d602b124
thread:issue:53178,issue #53178: Exceedingly different F.conv2d outputs on cuda/cpu,stale/low urgency,2,stale/low urgency,1,0,53178,,,,0,8f2fbee6d24c7c00a8d14aec140704cf64d82aae2d3fab772b449b8cae4d0c6b
thread:issue:53185,issue #53185: [NIT] Unpickler persistent_load monkey patch should fall back to original persistent_load,stale/low urgency,2,stale/low urgency,1,0,53185,,,,0,c5ca60509be35ad19da652fe4d1480a189761b8ec525d8e07da208829405373c
thread:issue:53191,issue #53191: RuntimeError: cuDNN error: CUDNN_STATUS_INTERNAL_ERROR,stale/low urgency,2,stale/low urgency,1,0,53191,,,,0,2bfcab1d2b8b34515ff054b1da731a5cb3020c0cfccb371e61ea3ff0587c6f4a
thread:issue:53193,issue #53193: [nnc] Prevent user from scheduling after prepareForCodegen,stale/low urgency,2,stale/low urgency,1,0,53193,,,,0,ab52c1956b2a41dfdddbc1431a6549758f54ad3e3afabbc2d1fdd2cfb60d48db
thread:issue:53194,issue #53194: Relative performance of histc vs bincount,stale/low urgency,2,stale/low urgency,1,0,53194,,,,0,eebb2b829df408a168755c56624717b18662330e929f8fcd6d5ab724b8118c87
thread:issue:53256,issue #53256: `index_copy_` test fails on PyTorch/XLA,stale/low urgency,2,stale/low urgency,1,0,53256,,,,0,96a90009d99dd718149c814befcc1916f41bc1c5d953eb7a187231afc2170d5b
thread:issue:53257,issue #53257: Inconsistent `out=` behaviour on advanced indexing operations.,stale/low urgency,2,stale/low urgency,1,0,53257,,,,0,acfd5318e259fc117fa18044c9ef2edacefc1b10ef5f467bdbe3d8e8486cfe80
thread:issue:5326,issue #5326: TestMultiprocessing.test_fd_sharing hangs with ASAN,stale/low urgency,2,stale/low urgency,1,0,5326,,,,0,32b84db33c53905a2a2a99a92b20cab176a7dd75f5604f799fe7477a6c68e3cd
thread:issue:53261,issue #53261: Dimension argument names in torch.diag_embed/diagonal vs. transpose/transpose_,stale/low urgency,2,stale/low urgency,1,0,53261,,,,0,770f1c871192d1626fff92863d9e63e4908e3b5c01ffbac8c5822eb11c12ed39
thread:issue:53263,issue #53263: Different gradients in torch.matmul depending on input shape,stale/low urgency,2,stale/low urgency,1,0,53263,,,,0,321c0bb5e0729792c699b1b534f363608122045b4a1208513038add2feb82e23
thread:issue:53284,issue #53284: Find a good namespace home for torch._assert_async,stale/low urgency,2,stale/low urgency,1,0,53284,,,,0,db731771228b54b0d2780e9c7a3dd4ef9264cda17e0473fda77d4aedd4dd6a36
thread:issue:53337,issue #53337: Should we remind users not to use the dataset on GPU when it's the argument of DataLoader?,stale/low urgency,2,stale/low urgency,1,0,53337,,,,0,fe83f8ecdf71c2ce336c233ec41ebd7733d3d1109fd61fd485cefa7286731435
thread:issue:53345,issue #53345: CUDA error: an illegal memory access was encountered when updating model weights using GradScaler,stale/low urgency,2,stale/low urgency,1,0,53345,,,,0,89f0c025341348c8240c9f2bef455365f310f6faa75cea09f6060d93a05eaab3
thread:issue:53352,issue #53352: OpInfo mechanism to test for nondeterminism,stale/low urgency,2,stale/low urgency,1,0,53352,,,,0,f5368b9a117c19ee7cca62404ffbbef7976f4f1616578001c2e7e4cf8095b524
thread:issue:53360,issue #53360: [bug] torch.cumsum: functional and method variant promotes all ints to Long but inplace don't,stale/low urgency,2,stale/low urgency,1,0,53360,,,,0,829fdbc84dcbbf1db76271789656c7b77ab111c8a60fa61b2191c0d6b5877446
thread:issue:53365,issue #53365: Pytorch mobile Vulkan API usage,stale/low urgency,2,stale/low urgency,1,0,53365,,,,0,fb08256895b5b5d5ab489c6fcbb600946046a5e06633ff3c0bba384ea32bac4e
thread:issue:53386,issue #53386: Default pytest is truncating stacktrace (somehow!!!),stale/low urgency,2,stale/low urgency,1,0,53386,,,,0,a1cf1cd2ed7092d0cd77154dc2c25ae1d390078f2380f78c0d9ed0e5f14e0724
thread:issue:53405,issue #53405: When I ran the 'Get Started with Data Data Parallel' tutorial the code got 'RuntimeError '!There are no hints,stale/low urgency,2,stale/low urgency,1,0,53405,,,,0,7110a416df7f4958ce492f774f73d2179025764e3d9cafbd6198cc38a679c7a0
thread:issue:53407,"issue #53407: torch.matmul doesn't handle zero-sized inputs in some cases, leading to batched grad failures",stale/low urgency,2,stale/low urgency,1,0,53407,,,,0,b06aea349bfd967ff5d287e38ed465e86ebce698aaac4ab0daeeddbb83b369dd
thread:issue:53427,issue #53427: Backward compute time in DDP logging may not be accurate,stale/low urgency,2,stale/low urgency,1,0,53427,,,,0,18fa8475b70c8e8742cd7bf8f139c5b1190e96dd61413334bb0460d434fde0f3
thread:issue:53438,issue #53438: Pipe should automatically build appropriate partitions.,stale/low urgency,2,stale/low urgency,1,0,53438,,,,0,e05cfb52a36ed046045f5098e5c4488de261f52a81a6c88a9f8b5bb5e2d6014b
thread:issue:53440,issue #53440: Add complex support for torch.unique,stale/low urgency,2,stale/low urgency,1,0,53440,,,,0,7ffbed3f905fa0da4b148137ca2f4ccdd49eca89c6edebfadb21ba95c4e7726a
thread:issue:53442,issue #53442: IRSimplifier performs unsafe transformation on mod: m*n%m -> 0,stale/low urgency,2,stale/low urgency,1,0,53442,,,,0,e489f1266464dcf8bdd720703f2033b73b7e4f7f571c550b58b9be3491d3b03a
thread:issue:53462,issue #53462: Second order derivatives for F.grid_sample(),stale/low urgency,2,stale/low urgency,1,0,53462,,,,0,526a3e32ff8d1b4c6685f316ee4c21e03ed1d3f3fda5641f93cbcced3d6f8b18
thread:issue:53463,issue #53463: [Feature Request] return type support for `torch.norm` and tensor iteration in for loop,stale/low urgency,2,stale/low urgency,1,0,53463,,,,0,c477dd169a7aa0becae2a4cce746328c8c74b68d808d24e5d60d09fcb0a1d6b5
thread:issue:53473,issue #53473: RuntimeError: CUDA error: CUBLAS_STATUS_INTERNAL_ERROR when calling `cublasCreate(handle)`,stale/low urgency,2,stale/low urgency,1,0,53473,,,,0,8b93b12ebe27fc29ea1729fb1cceb9e848b9ba9f81a7cc773e6ce41437859131
thread:issue:53477,issue #53477: Show methods of the class in the right sidebar,stale/low urgency,2,stale/low urgency,1,0,53477,,,,0,591b0e34168f5352ec56d2c76619d16db8cf2600d44ac4b01ff704f4bcf30941
thread:issue:53491,issue #53491: Very slow backward speed when using gather with small-range indices,stale/low urgency,2,stale/low urgency,1,0,53491,,,,0,879cf0b1592fe91e089cd4d14b6cd79ed5117276a547b447cb0a6afb1790900a
thread:issue:53510,issue #53510: Improve memory format testing,stale/low urgency,2,stale/low urgency,1,0,53510,,,,0,7bf0831630dec560ddd8449c2cec88d58d38045820a1029bb573d78496f6035f
thread:issue:53584,issue #53584: How to delete Module from GPU? (libtorch C++),stale/low urgency,2,stale/low urgency,1,0,53584,,,,0,c15630aa4f996f922cae2cb1fa18a47498c50515bdb845f99e7e430be7d45d37
thread:issue:53622,issue #53622: Properly design manual_cpp_binding (make it less error prone),stale/low urgency,2,stale/low urgency,1,0,53622,,,,0,6262599304f162eba32d41c43a95970d252a6f44715c00350075e2c5b1afde12
thread:issue:53627,issue #53627: test_metal.py must skip when not compiled with metal support,stale/low urgency,2,stale/low urgency,1,0,53627,,,,0,d1f67d756a0acd48e938b4d7e382caca87b9e29cb2694e562ac6ddd4e6ba9e67
thread:issue:53642,issue #53642: torch.unsafe_chunk and torch.unsafe_split 's documentation does not render on the website.,stale/low urgency,2,stale/low urgency,1,0,53642,,,,0,7f046c5045f184c8e0b401a5501b5455d55935877351befb1b54b6db8cf43093
thread:issue:53649,issue #53649: vmap gradgradcheck test fails for unfold operation,stale/low urgency,2,stale/low urgency,1,0,53649,,,,0,255cd168039d57d0c545c64da08e3aa79d0541901573a64c09d588ce97010fee
thread:issue:53658,issue #53658: torch.distributed with NCCL can hang if the first operation is a barrier,stale/low urgency,2,stale/low urgency,1,0,53658,,,,0,5639f157c9738f5780714d9b5098047569459b34c2d6d5af2b696cc480138545
thread:issue:53666,issue #53666: torch.autograd.functional.jacobian/hessian allow tensor in addition to function,stale/low urgency,2,stale/low urgency,1,0,53666,,,,0,93d720a3d8438ae1fb7268c1c26b6893f88a4a4de04f131bacc495b9f414a5a2
thread:issue:53696,issue #53696: Support GPU/CPU communication in RPC,stale/low urgency,2,stale/low urgency,1,0,53696,,,,0,372f6c8ddb58b4a01b55b7c4c7be7408862e3f635f42948cb3c2c1444972449b
thread:issue:53708,issue #53708: torch.load with Exception,stale/low urgency,2,stale/low urgency,1,0,53708,,,,0,8aa92170c00a2d765e9b42a65c2fb76008d4e1793e5aafdced418581b0bcfcea
thread:issue:53710,issue #53710: Ninja recompiles certain unchanged source files in cpp extension since torch 1.8.0,stale/low urgency,2,stale/low urgency,1,0,53710,,,,0,769ae6bfda1141965fc02f1ec727dffb43b0aa6d8baad3dfb4f7196e8f004dfc
thread:issue:53732,issue #53732: [FX] Allow customizing code generation for certain Nodes,stale/low urgency,2,stale/low urgency,1,0,53732,,,,0,86bc4996716d3be9a2aa741e377517348c5cdefe074382372f5e7c1d35c0029b
thread:issue:53758,issue #53758: ZeroDivisionError: float division by zero in Adam (bias_correction1 is zero),stale/low urgency,2,stale/low urgency,1,0,53758,,,,0,546c8d4991f55112a99641ac9cf2896e9f962b1f6d64bb735e9283bd5673ebf2
thread:issue:53785,issue #53785: Support for one-hot of dtypes besides torch.int64,stale/low urgency,2,stale/low urgency,1,0,53785,,,,0,566edac7b1684f4e46c110a7ff17890988c89e32213648672beb23d63962fbab
thread:issue:53794,issue #53794: [fx] GraphModule __init__ when provided a Module w/ inconsistent buffer handling,stale/low urgency,2,stale/low urgency,1,0,53794,,,,0,25e3a835a93ceba5cf16c5e44ba8b8c6a183e3f3960cfae9376bef654491525c
thread:issue:5385,issue #5385: Handle python_arg_parser dtype constants better,stale/low urgency,2,stale/low urgency,1,0,5385,,,,0,85581fe845bfbed557f1217170ebaec06dcc26c0ce40c4aedec8fed5c7b50f14
thread:issue:53866,issue #53866: Add comprehensive subgroup testing to torch.distributed,stale/low urgency,2,stale/low urgency,1,0,53866,,,,0,541b474b2d268fce247db05c175736aaf9dec4acf7434c04df9f77ad0142b653
thread:issue:53890,issue #53890: GoogleNet pretrained/scratch bug?,stale/low urgency,2,stale/low urgency,1,0,53890,,,,0,73435529d04afcff6af5140a57cae30bda227abe53523e2daabf3592b7d27e2f
thread:issue:53891,issue #53891: Check Archive.org for Failed Dataset Downloads,stale/low urgency,2,stale/low urgency,1,0,53891,,,,0,2eeb70508636ca065eea7ee5334a8da5afd97ba774ff44a887c25487eff50ab3
thread:issue:53910,issue #53910: There should be a CI build with OpenBLAS in addition to MKL,stale/low urgency,2,stale/low urgency,1,0,53910,,,,0,37b155c934297031593a17764f1eb2041d28abfbe231e3b4511fc1a3ddfe7373
thread:issue:53935,issue #53935: SubgraphRewriter doesn't handle tensor constants in replacement graph,stale/low urgency,2,stale/low urgency,1,0,53935,,,,0,9e1d3fe27b41fae1382f17ec3be6064c22b096bc2ea4db2d0dead4869fed7f6f
thread:issue:53952,issue #53952: [Pipe] supporting None and non-Tensors in forward's input/output,stale/low urgency,2,stale/low urgency,1,0,53952,,,,0,786e7836279520d265ec6dc0f5cc22266466ea84457a2a9f5b54e20baaa1aa8d
thread:issue:53957,issue #53957: Is pytorch 1.8.0 incompatible with cuda 11.2 or what is the reason for this error?,stale/low urgency,2,stale/low urgency,1,0,53957,,,,0,82ee50e3418043ede0f7995eabcb0f3d8ab63d5e9cfc60a0eee638caac0ad968
thread:issue:53983,"issue #53983: hope to support something like ""torch::manual_seed_for_mulit_thread""",stale/low urgency,2,stale/low urgency,1,0,53983,,,,0,d0860558df9b278ef10876748e247b2ea40cc6b20a400c273217b15559347bd6
thread:issue:54001,issue #54001: [codegen] Change public C++ to accept out/mut arguments by const&,stale/low urgency,2,stale/low urgency,1,0,54001,,,,0,9e1f5176eb22dada773a42c4064ddb85f27eeb33124c9beaeae1c8480be50eb7
thread:issue:54021,issue #54021: Support scatter/gather for NCCL backend,stale/low urgency,2,stale/low urgency,1,0,54021,,,,0,bea530646525afcc505b88a9fe9e4ed7c6d24b5e06f5bc9bc3b30b560709e870
thread:issue:54030,issue #54030: [FR] Configurable gain value for kaiming_normal_ / kaiming_uniform_ in torch.nn.init,stale/low urgency,2,stale/low urgency,1,0,54030,,,,0,ced5dfbc3f79015d30e086496ed8f1157e4414abd62eb1eb76320354e99041c3
thread:issue:54035,issue #54035: Web support?,stale/low urgency,2,stale/low urgency,1,0,54035,,,,0,e28ace990583f439831c25614ae60b38e6ca7efabd05802a8f2c63b14c952c44
thread:issue:54047,issue #54047: RuntimeError: cuDNN error: CUDNN_STATUS_NOT_INITIALIZED in pytorch lightning,stale/low urgency,2,stale/low urgency,1,0,54047,,,,0,be265dd87549dbdbde2b26087e750855e06754673f1495730f9e7680350ba406
thread:issue:54055,issue #54055: Wrong initialisation gain for SNNs/SELU,stale/low urgency,2,stale/low urgency,1,0,54055,,,,0,de3488ca11484bf4cf0c38011c9c182d615eb20f1714cb81aa4548b1627760bb
thread:issue:54058,issue #54058: c++ convert from std::vector<Tensor> to c10::List<optional<Tensor>>,stale/low urgency,2,stale/low urgency,1,0,54058,,,,0,b3148203d4e8bc96694d5e44f8f8ce159a5faca83f4633a9e8349d56db5c5022
thread:issue:54062,issue #54062: Many advanced indexing operations have untested large tensor branches,stale/low urgency,2,stale/low urgency,1,0,54062,,,,0,2d706cd95b83817f961eab94a453a5f8e3cef7f518b809bbbffd91f5de82d542
thread:issue:54071,issue #54071: [FX] Add guards in files in `test/fx` to make sure they're not run directly,stale/low urgency,2,stale/low urgency,1,0,54071,,,,0,885925d0a0279cb1dbab46fc1135e2cf09a249b4afcdddd8c38b1f637dd8d184
thread:issue:54077,issue #54077: [cuda extension] tutorial should mention device guard,stale/low urgency,2,stale/low urgency,1,0,54077,,,,0,39dc7df5b8edef5340d2409838b5073f18841a4adc3391d6ff853a25f7c784d6
thread:issue:54082,issue #54082: Dispatch table for linalg_norm is fishy,stale/low urgency,2,stale/low urgency,1,0,54082,,,,0,80597c962d82d78cbf89e844f97a9d3b9cd2f825eea3d4e9fa65fbb4d372f419
thread:issue:54096,issue #54096: cdist skip for manual backward is kind of janky,stale/low urgency,2,stale/low urgency,1,0,54096,,,,0,4f3abd361e741c8ddabd045dd3c74f76b58588e7f9e8efdace360be54fc7937f
thread:issue:54104,issue #54104: TorchVision and TorchAudio wheels for AArch64 absent from https://download.pytorch.org/whl/torch_stable.html,stale/low urgency,2,stale/low urgency,1,0,54104,,,,0,d15c3007750aa1b8aba29c4e113dda6c57340741c3f09b79db499e3f0cd07b64
thread:issue:54116,issue #54116: torch.put is divergent from np.put,stale/low urgency,2,stale/low urgency,1,0,54116,,,,0,920353bb3269272d94483d643ea1c5d84b5daaa6e34bdac0fdf1f0b4010c6ea9
thread:issue:54135,issue #54135: torch.kron of a transposed input error,stale/low urgency,2,stale/low urgency,1,0,54135,,,,0,d2cf17ef5b70eb74bcb5484dcdf4e89bef01b74c61d15cb9cfcf613fd2a988f8
thread:issue:54137,issue #54137: Add additive angular margin loss,stale/low urgency,2,stale/low urgency,1,0,54137,,,,0,f463c6233698914bc73888ad66b0d848920d4a261d0c0bded716a21689469922
thread:issue:54147,issue #54147: Implement Truly Parallel Ensemble Layers,stale/low urgency,2,stale/low urgency,1,0,54147,,,,0,117bd243b65a58a71e66fcb70d37ab59d9046b3c40527d5318150158a435d2a9
thread:issue:54149,issue #54149: OSS build doesn't hard fail on nodiscard,stale/low urgency,2,stale/low urgency,1,0,54149,,,,0,ad518da8e4b3e8ba915a06e1b53291969b402ecddf184b50acbd80ef2d70d935
thread:issue:54192,issue #54192: OSS Mobile Errors Point to fburls,stale/low urgency,2,stale/low urgency,1,0,54192,,,,0,075311933393ed5c9351c19160a453e47fbb64c0bff6662f12d4d990d246f638
thread:issue:54201,issue #54201: Provide half-away-from-zero rounding mode on Tensor::round,stale/low urgency,2,stale/low urgency,1,0,54201,,,,0,deef8be107aa9b8dc187759c2b35527036b3903f0db217bba933bcfef248de6a
thread:issue:54209,issue #54209: [docs] Incomplete examples for torch.Generator,stale/low urgency,2,stale/low urgency,1,0,54209,,,,0,cfd0958b9111cee8e991971aba62b3520567c5df5a1d738954678bde6b292dbf
thread:issue:54213,"issue #54213: [docs] torch.clamp / torch.clip do support None as placeholder of min/max, but this is not documented",stale/low urgency,2,stale/low urgency,1,0,54213,,,,0,ae29a91b6802a66cfb2d1c46abec14aded00999f63b5c88d64791d248ea5c207
thread:issue:54216,issue #54216: Add inplace variant for torch.minimum / torch.maximum,stale/low urgency,2,stale/low urgency,1,0,54216,,,,0,bb0aa470476d1f25411e9174d8b5b7b8eacce088e26c79024550ee2fc6a4c32f
thread:issue:54232,issue #54232: DistributedDataParallel: `DDP(model)` hangs on non-master node,stale/low urgency,2,stale/low urgency,1,0,54232,,,,0,4e96d46f2f1fcdd46f3a5fbbe5ef7ad82813d6331298863084a12365d7356d38
thread:issue:54269,issue #54269: SummaryWriter add_image() docs do not state that it is expecting images in a certain format,stale/low urgency,2,stale/low urgency,1,0,54269,,,,0,e4bc77b27acc0babe43ef698fb8a60264e7526d0b0c43a114fb7d739f29ba075
thread:issue:54278,issue #54278: [Static Runtime] StaticRuntime.EmbeddingBag test case broken,stale/low urgency,2,stale/low urgency,1,0,54278,,,,0,8258c4bbd56b4ec4bd2ca2c028412b249c502f5cd62c9094b1d8df10fe310560
thread:issue:54279,issue #54279: [Static Runtime] static_runtime_benchmark compile-time error,stale/low urgency,2,stale/low urgency,1,0,54279,,,,0,b104b97ef678c61a65c93c699fcb0b594cab03d18a9a52266e17806a8bed0808
thread:issue:54282,issue #54282: test_randperm is failing on CPU-only build,stale/low urgency,2,stale/low urgency,1,0,54282,,,,0,fdce3fcffd9f8bda452fcdea2cc8d8c6948abdfb702762cf7e6e23ff39d99210
thread:issue:54283,issue #54283: Split VariableTypeManual.cpp,stale/low urgency,2,stale/low urgency,1,0,54283,,,,0,6ce60a23091a68a982179c78f90ce39c69abc65076eb47bf12c2602e9b761eac
thread:issue:54307,issue #54307: cuDNN error upon the backward method with a sliced tensor output from nn.GRU.,stale/low urgency,2,stale/low urgency,1,0,54307,,,,0,352acb2d60dcd13dbc90fc34618626c897852bcb5648fcba4e72ebe43ec40d35
thread:issue:54309,issue #54309: [discussion] Support torch.matmul: strided x sparse in addition to sparse x strided,stale/low urgency,2,stale/low urgency,1,0,54309,,,,0,cd1ce6e52c048afab9bc0adf9643d6aa4ae0655911afd3fba484ce64a12ef962
thread:issue:54332,issue #54332: Unit testing failures when porting PyTorch wheel,stale/low urgency,2,stale/low urgency,1,0,54332,,,,0,384456c67261f0e500438ab9af0c49e606804a1893772400e634aaf06c8e3d58
thread:issue:5434,issue #5434: RuntimeError: $ Torch: not enough memory: you tried to allocate 72GB. Buy new RAM!,stale/low urgency,2,stale/low urgency,1,0,5434,,,,0,09b0c7fca723e2b3a371f1cafc86bb9c181f603b7a7fedc3669321d5d886af73
thread:issue:54340,issue #54340: RuntimeError: polar does not support automatic differentiation for outputs with complex dtype.,stale/low urgency,2,stale/low urgency,1,0,54340,,,,0,61d6c700a3327b7229cbd1b291240c45bffcb1c3ce62d777e59b111003a1404d
thread:issue:54346,issue #54346: Missing tests for gradcheck,stale/low urgency,2,stale/low urgency,1,0,54346,,,,0,fa7e7aef73492c9e9ed479bee7226baa0119df38cd63acaccb59ccbc65e00a75
thread:issue:5435,issue #5435: Gaussian Sampling,stale/low urgency,2,stale/low urgency,1,0,5435,,,,0,ae4b78120b566ab0a2e1626d359cb2bae3a5ae95f50e1b5011664e1b844b7eb5
thread:issue:54351,issue #54351: Numpy.float64 vs native python float breaks DDP,stale/low urgency,2,stale/low urgency,1,0,54351,,,,0,ffb73bc05f1746037e9082fcd9bb334c8fa4d3c40cfbf9e59c88a2cb3d78f390
thread:issue:54356,issue #54356: Generate delegates for all non-structured kernels,stale/low urgency,2,stale/low urgency,1,0,54356,,,,0,e400b8fb93b9ea7b543517d1b43cc5395d76ec7831bf89ede5ae431ec9cf94da
thread:issue:54362,issue #54362: [package] Buck integration for dependency analysis,stale/low urgency,2,stale/low urgency,1,0,54362,,,,0,021352fd880f3e582b774fc6b24baf967c4ee96ff8d932b578e72b954a660e80
thread:issue:54364,issue #54364: [package] Clear and minimal interface for what a package provides,stale/low urgency,2,stale/low urgency,1,0,54364,,,,0,0a87bae762107cf7aac206da832bd9dca8ea7d913e271ff9e8f9aff365ff8aad
thread:issue:54366,issue #54366: [package] Heuristics for extern-ing common dependencies,stale/low urgency,2,stale/low urgency,1,0,54366,,,,0,13fbc5b35cca34ba8644677db697c3c7e6592b3b4310677b6479af9ba5246f15
thread:issue:54368,issue #54368: RMSProp documentation is confusing,stale/low urgency,2,stale/low urgency,1,0,54368,,,,0,1d3f1895498b3cb67a36f451bfb28e4ebc5642cf28e60326b7a25bf81e1fa48c
thread:issue:54375,issue #54375: [proposal] Integrate autograd graph visualizer for displaying crashed locations during detect_anomaly,stale/low urgency,2,stale/low urgency,1,0,54375,,,,0,2d6f3ff36846b931e9e8d4584c2a379e3e93800ca9057604b66332dd5db8c4d4
thread:issue:54387,issue #54387: [docs] torch.lerp has unmatching signature and explained parameter names,stale/low urgency,2,stale/low urgency,1,0,54387,,,,0,ac1daf3c9f604604a9da5ca964551d69a6616618571cb39e863e3843f31cd5e8
thread:issue:54393,issue #54393: [docs] torch.optim.lr_scheduler,stale/low urgency,2,stale/low urgency,1,0,54393,,,,0,fab4f4429241e4c33c80ddf115bd8df2b19f5df8045953959a0893b7c71d6c5c
thread:issue:54394,issue #54394: Half Normal Log_Prob not defined for 0,stale/low urgency,2,stale/low urgency,1,0,54394,,,,0,fd0fec69f8e7aab0dc6ba5ec2ea29b367f9e1ef83f854e5b284d9673cde6c05a
thread:issue:54407,issue #54407: test_clamp fails on ppc64le,stale/low urgency,2,stale/low urgency,1,0,54407,,,,0,ac14b0aa922d712b6c9bbb1970fa7ccd9beafe2a51282b88404ed8c6867be98e
thread:issue:54445,issue #54445: add tests to cover DDP on mixed dense-sparse models,stale/low urgency,2,stale/low urgency,1,0,54445,,,,0,c8cb3eb0b788877ad59d1c075a50289280e30d36d2e4c7a715576084e3bae612
thread:issue:54499,issue #54499: Caught an integer overflow or wraparound in torch.nn.MaxPool1d,stale/low urgency,2,stale/low urgency,1,0,54499,,,,0,f855ac7611db345ccd7467a42515a2f0f7bb166408ca15f96fc6d7eeabbed2b3
thread:issue:54503,issue #54503: torch.fx may have problem for annotations,stale/low urgency,2,stale/low urgency,1,0,54503,,,,0,37a5160c75aa5ca3e0767e70b11ac8913402ef66adb83c15988c000d87a41fdd
thread:issue:54546,issue #54546: Let backends specify a schema version when registering kernels,stale/low urgency,2,stale/low urgency,1,0,54546,,,,0,f04dd16f970d4a9621cffba1abff4fc9f2d1d4f5f0cf4229cdf3a6ebd0d2b2b7
thread:issue:54564,issue #54564: why single pytorch process is displayed on each GPUs with memory 0MB,stale/low urgency,2,stale/low urgency,1,0,54564,,,,0,d63e2548aaf72d0df6f5c252f02a2814b1d1939cf705f0c651c5f047919d55c7
thread:issue:54566,issue #54566: Support docstring definition in torch/library.h operator definition,stale/low urgency,2,stale/low urgency,1,0,54566,,,,0,8c7a378d1ac0bf76b10ada85ef347dda11d83b6bf39282eb88d514e959400c02
thread:issue:54574,"issue #54574: Add OpInfo metadata for ""is_torch_functional"" and a skip for these ops in TestOperatorSignatures.test_get_torch_func_signature_exhaustive",stale/low urgency,2,stale/low urgency,1,0,54574,,,,0,f51c210572d751e863b341ea546286a0c8d18581308b3939279e6093ae665d88
thread:issue:54588,issue #54588: Make searchsorted and bucketize API consistent,stale/low urgency,2,stale/low urgency,1,0,54588,,,,0,600c11bd247901b743fa22f227608ef21e966682b4b09609027bd7eee4fe81b5
thread:issue:54591,issue #54591: Various issues in derivatives.yaml,stale/low urgency,2,stale/low urgency,1,0,54591,,,,0,78de49601b0b1c98a1798e4349b5cf01fb2de02508e8447a8ffc0fd6c9caf7ec
thread:issue:54614,issue #54614: Dispatcher TLS to bypass loads of dispatch key from tensor arguments,stale/low urgency,2,stale/low urgency,1,0,54614,,,,0,4a34ce1866da44dbb86992d325ef93e2b727af92ddf336292e4ce4669c0ec6f8
thread:issue:54620,"issue #54620: torch.pow(tensor, tensor) throws RuntimeError for dtype bool",stale/low urgency,2,stale/low urgency,1,0,54620,,,,0,ae9fa62612ecaa217b7b1e8638ecd8cf83d05205e7e65582e47cba59779c020c
thread:issue:54629,"issue #54629: ""Skipped!"" tests should have more descriptive skipped reason",stale/low urgency,2,stale/low urgency,1,0,54629,,,,0,96540e33781a552cf8c69d44fb8c95a60dd75bc62df187bfed33bdaf0d3bb539
thread:issue:54630,issue #54630: torch-tensor-repr gdb macro doesn't seem to work in opt mode,stale/low urgency,2,stale/low urgency,1,0,54630,,,,0,cd35c338f8538569ace27b35ca9edee5b29e29267c86d7186c54593d262f8069
thread:issue:54646,issue #54646: Optional enhanced logging for operator calls for backend implementor debugging,stale/low urgency,2,stale/low urgency,1,0,54646,,,,0,59a868f34026f5fb80e0ff3a021e33d553a90eea13cea6b96b05c9a517a9fc56
thread:issue:54655,issue #54655: qsize not implemented error,stale/low urgency,2,stale/low urgency,1,0,54655,,,,0,c280d4bc5000616d2a75de34028d5999f4071d46c0cedb48bb0910d0fe68f0d5
thread:issue:54677,issue #54677: [docs] use sphinx link/reference checking features,stale/low urgency,2,stale/low urgency,1,0,54677,,,,0,e05a8dc142f12ff54891c175d66bbb2e7df48f5fdc789d28609654a73d6f8bab
thread:issue:54713,issue #54713: Test test_large_cumprod_cuda_float16 gets killed due to (probably) OOM,stale/low urgency,2,stale/low urgency,1,0,54713,,,,0,063bb102a62350304532fb32ef5478d85d6e4c606dc42bd04c336fbeadcfda37
thread:issue:54743,issue #54743: [Feature request] Add batched matrix support for torch.diag,stale/low urgency,2,stale/low urgency,1,0,54743,,,,0,2ed50a9b058f641dd60b5d6ba9d5bcfefa04619880de4c91d3f6e1e26addbadd
thread:issue:54753,issue #54753: torch.sqrt for negative values should either return complex tensors or clearly throw a domain error/warning,stale/low urgency,2,stale/low urgency,1,0,54753,,,,0,0b1659c30caf99795bb0d18a162703ee96f970f29cfb3b7d0916f5f953e7da87
thread:issue:54757,issue #54757: [nnc] computeAt and index flatten broken in presence of if statements,stale/low urgency,2,stale/low urgency,1,0,54757,,,,0,f447d415e4791a0b13d95443c0e0f9b6a546d3cce020a657345966516edff6cd
thread:issue:54774,"issue #54774: RuntimeError: ""log2"" ""_vml_cpu"" not implemented for 'Half'",stale/low urgency,2,stale/low urgency,1,0,54774,,,,0,1fc93ff3a3025eda5fdd6e9088c6e556564f295ed94af17b208132124d7529fc
thread:issue:54785,issue #54785: [FX] Cannot symbolically trace class method,stale/low urgency,2,stale/low urgency,1,0,54785,,,,0,3ec3a768ece9f8cc5d8109521ea89fc1bb7b45c01091ea58ef65a6facd702aa9
thread:issue:54800,issue #54800: [package] Detection of dependencies introduced by torch.ops.load_library,stale/low urgency,2,stale/low urgency,1,0,54800,,,,0,81d954b4038da6ed8c2a63a17c021d9b76a4a40df1a92e5b9d023d7f83983f78
thread:issue:54802,issue #54802: [nnc] simplify is sometimes broken after computeAt,stale/low urgency,2,stale/low urgency,1,0,54802,,,,0,208e95a4c76e8a2e38dbb23a41fe2d32aaffbb2230dae58531511a9597b74986
thread:issue:54815,issue #54815: TensorExpr `LoopNest.get_loops_for` misbehaved after loop distribution transformation,stale/low urgency,2,stale/low urgency,1,0,54815,,,,0,d2d1bee36d9915cbe73f1b065b9d129140d888cd7529736d278227ca09e260d1
thread:issue:54816,issue #54816: Per channel weight observer for ConvTranspose,stale/low urgency,2,stale/low urgency,1,0,54816,,,,0,66355a78f49a5abfcb466bdca7312ec692afe73d45d1541c608ce0629d6028d8
thread:issue:54823,"issue #54823: Fixing pytorch distributed training (DDP) code send a SIGKILL signal on its own on most version of pytorch 1.6.0, 1.7.0, 1.7.1?",stale/low urgency,2,stale/low urgency,1,0,54823,,,,0,e197681ab99703a5cc739ee91aac03be9b9772ad016c70cb41009f4c609cf1db
thread:issue:54829,issue #54829: c++ libtorch config mismatch with python version,stale/low urgency,2,stale/low urgency,1,0,54829,,,,0,56f6980682bb1471a110a941eed5bd4077f9e39acc1ca51f0ecd6448da2fb346
thread:issue:54841,issue #54841: [testing] test_reference_numerics_extremal_clamp_cpu_bfloat16 fails on ci build with GCC5.4 and Python 3.6,stale/low urgency,2,stale/low urgency,1,0,54841,,,,0,b3b5f7260cac1b18d16e43f888548d7b3a0cdef75cd56d2eedc0b419fbaed883
thread:issue:54843,issue #54843: Inconsistent behaviour with `weight` argument in CrossEntropyLoss and BCEWithLogitsLoss,stale/low urgency,2,stale/low urgency,1,0,54843,,,,0,4e0c2feea1cee4f6c903771c6c7622b1051fe3964c6f7fe00f73413828b011ba
thread:issue:54861,issue #54861: Occured error in loss.backward() when using sparse=True in Embedding layer,stale/low urgency,2,stale/low urgency,1,0,54861,,,,0,162ecc13ce5820ec8bfcbc08f29a569cbd0327a2101daa6edcdc7989047fd781
thread:issue:54871,issue #54871: Not possible to save dataloader in C++,stale/low urgency,2,stale/low urgency,1,0,54871,,,,0,784127126a166e2ab4de124868d5272429944c6ed712cfb58264596831500eb5
thread:issue:54874,issue #54874: Get a thread safe copy of torch::nn::Sequential object,stale/low urgency,2,stale/low urgency,1,0,54874,,,,0,dfeab7016bfe54d58b3e68ac4fc15733fd5b26632cb618cbb6921399cb2abcd1
thread:issue:54897,issue #54897: Default generated meta functions for inplace operations don't report errors,stale/low urgency,2,stale/low urgency,1,0,54897,,,,0,31722dfce658aed29e203e24f5634509c3967e6a995e54fa26e17bd910bc9c95
thread:issue:54899,issue #54899: [fx] forward references in annotations do not produce correct source code,stale/low urgency,2,stale/low urgency,1,0,54899,,,,0,9828c5934893c311211cf1e5d1f6598cc59a7c333772b6721a85fb9d2a541207
thread:issue:54905,issue #54905: Allow CUDA build without requiring a physical GPU device.,stale/low urgency,2,stale/low urgency,1,0,54905,,,,0,9e87f7ea62890201c3f0f1acf8e0902bec9a5178d3bc3cf30c896e86c91c627c
thread:issue:54906,issue #54906: Out variants for Convolution ops,stale/low urgency,2,stale/low urgency,1,0,54906,,,,0,ce7b54679c79ca894e266160f03e35302e273dea78cab5ed38957cc7079380c4
thread:issue:54913,issue #54913: FX tracer doesn't support returning a new instance of `object` subclass from module forward,stale/low urgency,2,stale/low urgency,1,0,54913,,,,0,8d626b9f52ba8c2a8f480e4a5a7184525b6099fff6738700206052ec7f5e29e4
thread:issue:54938,issue #54938: [tests] `cumprod` OpInfo tests take long time to run (around 1min),stale/low urgency,2,stale/low urgency,1,0,54938,,,,0,b4bd1f585b46148bc06044519328b5bda5108b1b15ee0abb270417835fb64d0d
thread:issue:54941,issue #54941: Regression in Python arg parser performance.,stale/low urgency,2,stale/low urgency,1,0,54941,,,,0,aa987149f060db767b725dfe83086f5cb1c6ed966e2ca7e808be9bfe63027c5a
thread:issue:54942,issue #54942: Optimizations to TORCH_CHECK change inlining behavior.,stale/low urgency,2,stale/low urgency,1,0,54942,,,,0,08edfcdd58343a69f1b84c7508935db2cbc62847c7b19c1cebdb163a465984c0
thread:issue:54943,issue #54943: Structured kernels have increased TensorIterator overhead.,stale/low urgency,2,stale/low urgency,1,0,54943,,,,0,f251fb51ef9eb20a89b034f8773604a296a1be4dfe898157724b2feb8fff46c5
thread:issue:54947,issue #54947: Results surprisingly differ on different backends,stale/low urgency,2,stale/low urgency,1,0,54947,,,,0,adfbd38a04ab2a94a455fd06cf16fabcfea6e46ebcd89d37ccc90f6c593150b9
thread:issue:54979,issue #54979: test_quantize_fx.py test_resnet_18_dpp test failure,stale/low urgency,2,stale/low urgency,1,0,54979,,,,0,055d4907ffa0e5a4659ecc2e6240085414b565abe1bf8a308ee0a62ad98152a7
thread:issue:54983,issue #54983: Accept objects with `__float__` wherever regular `float`s are accepted,stale/low urgency,2,stale/low urgency,1,0,54983,,,,0,ecc7a5ccba6d695162677882fa6890821ab442f00812783edf0473e5e35b278f
thread:issue:55005,issue #55005: [TensorExpr] Provide a safe API to request the N-th loop for a given Buf.,stale/low urgency,2,stale/low urgency,1,0,55005,,,,0,7275fb21d10681d2517b5e9b56bda0691e2fbaced69ba6fbb630744393490edf
thread:issue:55030,issue #55030: Add Minimal Gated Unit (MGU),stale/low urgency,2,stale/low urgency,1,0,55030,,,,0,e0c20df68cce9ad2d5b1cebd7c62703779e408b5b698e30fad32e26e8ff34808
thread:issue:55045,issue #55045: 16 bit accumulation for Convolution,stale/low urgency,2,stale/low urgency,1,0,55045,,,,0,c54dc72774d0bce5ac10a5d9ac5f3d677318c658043a3fca789af5c5316400eb
thread:issue:55083,issue #55083: [JIT][Testing] Split up test_jit.py and test_jit_py3.jit,stale/low urgency,2,stale/low urgency,1,0,55083,,,,0,205edf55050f7c42c729a69f9691aee31bc5ff5c4ccf0bef54c186aa9acdec17
thread:issue:55120,issue #55120: Why isn't pip finding the correct pytorch dependency?,stale/low urgency,2,stale/low urgency,1,0,55120,,,,0,04b3142566cab728af86ee0dcb44d6fa110dcae15d8faee400d6336d9ba78bed
thread:issue:55143,issue #55143: Add suport to torch.gather for negative indices,stale/low urgency,2,stale/low urgency,1,0,55143,,,,0,5eb56fee6f94e5922475bb4fa863fae57457035c3e156f5d14e20acd1207b65f
thread:issue:55156,issue #55156: multiprocessing function cannot pass cuda objects use when calling inside from a DDP process,stale/low urgency,2,stale/low urgency,1,0,55156,,,,0,a557b58cff6392b754107abac17e1e0b65ae0860cad70595d95267cb5bdf6490
thread:issue:55161,issue #55161: Internal assert failed: `iter.device(arg).is_cuda()`,stale/low urgency,2,stale/low urgency,1,0,55161,,,,0,64a15edc9bc2950e2ba5e8f97cf5e0c5171338b5102881a19087b725c734a625
thread:issue:55191,"issue #55191: `torch.tensor(..., device='meta')` doesn't work",stale/low urgency,2,stale/low urgency,1,0,55191,,,,0,4b38e086519224e129b5ec041f3050497ec8d4495ccc99c1406b021eb5a143a1
thread:issue:55192,"issue #55192: Consider adding meta:1, meta:2 devices",stale/low urgency,2,stale/low urgency,1,0,55192,,,,0,9875386848dfde82b2824ba034764bfc29fe2d5b430fe439b13c1327252ceb27
thread:issue:55198,issue #55198: [FX] to_folder breaks with qualified type name,stale/low urgency,2,stale/low urgency,1,0,55198,,,,0,529a2ff434570e2422f3d03896fcadcd3b0e4e33871789b960eadd6f92c59e50
thread:issue:55228,issue #55228: Requesting a feature to calculate attention weights.,stale/low urgency,2,stale/low urgency,1,0,55228,,,,0,8f69db5d5717df7c9707b03ea306fa44bd137c148af057045ef90c86a21546c3
thread:issue:55230,issue #55230: Profile Optionals and Optional[Tensor] specifically,stale/low urgency,2,stale/low urgency,1,0,55230,,,,0,21444b7a1bb248efca85d7e32e57b6eef7de6dd7ae73624f3d4f29f682103394
thread:issue:5524,issue #5524: Redo torch.nn.functional docstring strategy,stale/low urgency,2,stale/low urgency,1,0,5524,,,,0,115f476ab5a2ef42c6c320eb39fb1f629deaea3cb1662d17652416a778a55d02
thread:issue:55260,issue #55260: Cuda memory leak check is somehow unstable with repeat_test_for_types,stale/low urgency,2,stale/low urgency,1,0,55260,,,,0,80091725fcbcbe561f5510ae31d9f93891e72937ac43916454b6b85a720cedbd
thread:issue:55261,issue #55261: Batched multi_dot / chain_matmul + let it accept a tensor instead of tuple,stale/low urgency,2,stale/low urgency,1,0,55261,,,,0,5bac26c1022b8fb04a7ce0abfcbc07a0ea92d17703a6875bef208b4a4cd634ee
thread:issue:55274,issue #55274: [Feature Request] PowerNorm,stale/low urgency,2,stale/low urgency,1,0,55274,,,,0,05918ae8f90d6d094bd3bdbb7e325a534d134d0c358306fee10501c8f44dc988
thread:issue:55277,issue #55277: Small model occupies too much GPU in CUDA11.1 + Torch1.8.1 but is normal in Torch 1.6 + CUDA10.1,stale/low urgency,2,stale/low urgency,1,0,55277,,,,0,b63ef17ec689d5fdf4ddc66cc0372f106aac048bf22e590d4a4169b133cad636
thread:issue:5528,issue #5528: TestNN.test_data_parallel takes 10G of memory,stale/low urgency,2,stale/low urgency,1,0,5528,,,,0,011edc3634343c521db943876bfbe44b5b2245c69d40087da945d2500e118011
thread:issue:55286,issue #55286: endpoint=False for torch.linspace and torch.logspace,stale/low urgency,2,stale/low urgency,1,0,55286,,,,0,e7c14cf5d7b9fa792c1a3d732d37e645b128b74eed55f903da825489074051f8
thread:issue:55289,issue #55289: Feature Request: Add a rounding mode to round,stale/low urgency,2,stale/low urgency,1,0,55289,,,,0,c1eb4a1ba3dfb15a1d5f1546b48dc1e9dc0662516c75e3c43ee815d87ecff53d
thread:issue:55304,issue #55304: nn.DataParallel should raise error when provided with list of tensors,stale/low urgency,2,stale/low urgency,1,0,55304,,,,0,59619493c10add758ccc59992f28f7ca1c134efa73af89cd5a429569bdb9eac2
thread:issue:55329,issue #55329: Split test/cpp_extensions/setup.py per orthogonal extension module and/or do Python build from ninja too,stale/low urgency,2,stale/low urgency,1,0,55329,,,,0,25b9aedc4cd276c249ab9568644c5802f2e93c7668576f78951c8c296239df61
thread:issue:55330,issue #55330: CI not surfacing some failures (-Werror?) on PR,stale/low urgency,2,stale/low urgency,1,0,55330,,,,0,4741c74f13163deacd9268441fc3d466d4b6a38695b548fb2a20fc4cc21d727b
thread:issue:55331,issue #55331: Tests should be runnable without run_test.py,stale/low urgency,2,stale/low urgency,1,0,55331,,,,0,55c7b0d3120a6a474c39d1f9e9b6e54a763529172b56b6ed45e6c8be531d0bbf
thread:issue:55355,issue #55355: simple matrix multiplication yields wrong result on Ampere (3080),stale/low urgency,2,stale/low urgency,1,0,55355,,,,0,f1d1c03b3da893b79200130ae128c552504d5598c886e5e047d563287c21cb39
thread:issue:55356,issue #55356: torch.allclose does not allow different types for comparison,stale/low urgency,2,stale/low urgency,1,0,55356,,,,0,c8bf30799233a058f2ed5f3bcb74890af41689efea07120e9ef43c93fe3894eb
thread:issue:55363,issue #55363: it is lack of compatibility of CUDA_HOME (or other env),stale/low urgency,2,stale/low urgency,1,0,55363,,,,0,36288569f68e0776e900cb3f1bfe13c1d20a633dc597beb03be9bf80c75d83e1
thread:issue:55374,issue #55374: [RFC] Extend Autocast to CPU/CUDA with BF16 data type,stale/low urgency,2,stale/low urgency,1,0,55374,,,,0,857d757b6b58d916aba0a3933065c1d1a3ce746f669153922e14c29abefa6434
thread:issue:55394,issue #55394: [FX][testing] Test tracing into all the standard `nn.Module` instances,stale/low urgency,2,stale/low urgency,1,0,55394,,,,0,f268bbffdaee88aab02de2de91f212b6f937792a28094e91c1a24488e93c7387
thread:issue:55395,issue #55395: [FX][testing] Test tracing into all the standard `torch.nn.functional` instances,stale/low urgency,2,stale/low urgency,1,0,55395,,,,0,72519293f65b165deacd5a8b1e50a3195a3b4b4efb0376f3b1a2946bf0685e87
thread:issue:55397,issue #55397: [FX][testing] Run symbolic tracing tests over torch benchmark in CI,stale/low urgency,2,stale/low urgency,1,0,55397,,,,0,668a008f12816f6f64ec74a1deb0a2095ade396efaa3cd031c9c389f48548bfc
thread:issue:55399,issue #55399: [FX][testing] Test symbolic tracing of detectron2,stale/low urgency,2,stale/low urgency,1,0,55399,,,,0,dc88adb7d2253f83f9b0c5e8678bdf755e6a6293dbce6f5fb431f0a46d773bbf
thread:issue:55400,issue #55400: [FX][testing] Add symbolic tracing tests for torchtext,stale/low urgency,2,stale/low urgency,1,0,55400,,,,0,f99740ba3537a15ea0bfc38836452ed69f16e6e7e5c9069a664f4ddaf19afc4c
thread:issue:55401,issue #55401: [FX][testing] Add symbolic tracing tests for torchaudio,stale/low urgency,2,stale/low urgency,1,0,55401,,,,0,145d416831d7cc325ecc8cd2d32d6c251a6686e36accb2fe9472524f6bcfdba9
thread:issue:55437,issue #55437: Error building docker image: No module named 'typing_extensions',stale/low urgency,2,stale/low urgency,1,0,55437,,,,0,2e86efac7bee1d8e343293df3ae8582d21c9de12769a0646f243d89abe141e8b
thread:issue:55440,issue #55440: Have the ability to disable `__torch_function__` dispatch for torch.nn.functional functions,stale/low urgency,2,stale/low urgency,1,0,55440,,,,0,04950f6f5c551b342bc05116365dc76a12f22a9bd1a8c6b1972b4eed52860979
thread:issue:55442,issue #55442: More AVX2 vectorization support for half (float16),stale/low urgency,2,stale/low urgency,1,0,55442,,,,0,c0225a36495ffa657430723c17cd1a94fd526918eadd2552012e1ab5dfba9a1b
thread:issue:55448,"issue #55448: Errors occur when operating broadcast with error log: Got completion with error 12, opcode 1, len 32547, vendor err 129",stale/low urgency,2,stale/low urgency,1,0,55448,,,,0,2f19d4ca17e4582b158f2509cdc19324dec330fe91147997a83d4d6c6201444d
thread:issue:55450,issue #55450: [DOC] Incorrect image for the confusion matrix,stale/low urgency,2,stale/low urgency,1,0,55450,,,,0,78c070807928a3b8c0a8efb5d1dfc6b06f484aa22a6d7b164126f9685f760d32
thread:issue:55453,issue #55453: C++ sparse_coo_tensor ignores TensorOptions argument,stale/low urgency,2,stale/low urgency,1,0,55453,,,,0,69df018009973244cc0e01889f59dcb2ec6137769be147eea3954f905916ac99
thread:issue:55454,issue #55454: Include the cuda headers in the libtorch pre-built packages,stale/low urgency,2,stale/low urgency,1,0,55454,,,,0,1837dfb6b8d653c7e086b21a3b61b2ea58abc11c6337e4842d8633a07f063f7f
thread:issue:55457,issue #55457: [feature request] TupleSequential module or an option for Sequential for unpacking tuples (i.e. support varargs),stale/low urgency,2,stale/low urgency,1,0,55457,,,,0,d33f5d19bb0625d1684e83acc062b2dc779c4023404cc1c356d27d510a98d14a
thread:issue:55461,issue #55461: Cross-compiling for cortexa9 processor,stale/low urgency,2,stale/low urgency,1,0,55461,,,,0,b40577c6af927bcc8b4c234be421b93af853a1cf7498aa2e06055d31bfe2910e
thread:issue:55464,issue #55464: Support quantized functional linear/conv operators with 32b output,stale/low urgency,2,stale/low urgency,1,0,55464,,,,0,0e456e4b087cb9ef1615badfb88aa61e05176eebcfcbc5d7d014a58a8868b64d
thread:issue:55544,issue #55544: [FX] See if we can type-annotate attribute accesses on proxy,stale/low urgency,2,stale/low urgency,1,0,55544,,,,0,03bc0ad14e3bd41a851ba9f973d6121ede34a54890c2ddc321a86110ff5769c6
thread:issue:55549,issue #55549: `embedding_bag` has unexpected behavior when given `offsets` that are not monotonically increasing,stale/low urgency,2,stale/low urgency,1,0,55549,,,,0,2887bd1db5daf5b91b9ed0f563712dec40b1ac07d61cb3d650ae5f8d9f459e0e
thread:issue:55557,issue #55557: Gradcheck failure for `torchaudio.functional.phase_vocoder`,stale/low urgency,2,stale/low urgency,1,0,55557,,,,0,6a719fff2869478c0b39c15cd3b93228e06f0f64a9c7aa544d6cf8817c695e5d
thread:issue:55583,issue #55583: Improve test runtime of distributed tests.,stale/low urgency,2,stale/low urgency,1,0,55583,,,,0,29a35a62de7e195d5a790f3b6fa8e9a2d92203a19335c1fda7770502b7a02a95
thread:issue:55607,issue #55607: pytorch inference lead to memory leak in cpu,stale/low urgency,2,stale/low urgency,1,0,55607,,,,0,7f69f37985536a2abb2bc20a3e4f0f5e0937fbebeae6c4715d35e7cbe60c0ff3
thread:issue:55609,issue #55609: requires_grad does not get propagated properly when using the JIT compiler,stale/low urgency,2,stale/low urgency,1,0,55609,,,,0,797b5b3d67d5ed714f294525e87db00b999c67c2ecf4f9fdf6f332a7d431be2e
thread:issue:55611,issue #55611: ATen/native/sparse headers not in LibTorch or pypi package,stale/low urgency,2,stale/low urgency,1,0,55611,,,,0,21d59793a77e4a53ff58c2834b17a83e2297f2e793002200905210cfd28ff077
thread:issue:55629,issue #55629: torch.arctanh not implemented for torch.float16,stale/low urgency,2,stale/low urgency,1,0,55629,,,,0,d1973d3c576cfd68b830e12e4bb654a53c3a13255fc7abb86e07f18f5327e7d6
thread:issue:55643,issue #55643: at::globalContext().hasCUDA()'s forked tongue tells lies,stale/low urgency,2,stale/low urgency,1,0,55643,,,,0,6cc71d5a99b74986e164bbee14fbbb7baab8c33e98f228b79819eeb5845c0f82
thread:issue:55668,issue #55668: GraphModule.to_folder generates code with syntax error (imports in the middle of a class) [torch.fx],stale/low urgency,2,stale/low urgency,1,0,55668,,,,0,f2e0c71cb5f0a7e3e17544a23238ad78fb4ea54ef8fcbda6cfef6d2be030e57a
thread:issue:55694,issue #55694: symbolic tracing does not support patterns with multiple arguments?,stale/low urgency,2,stale/low urgency,1,0,55694,,,,0,f8311e2e8f1ccebaea796759f15a1ad58cce87f26b3c4b09d30a937f07119a76
thread:issue:55696,issue #55696: [TensorExpr] IRSimplifier fails to optimize useless stores,stale/low urgency,2,stale/low urgency,1,0,55696,,,,0,639c52d7dd07197e2c51131f272e3909e166c7502a7805d016c1159a4733ce30
thread:issue:55736,issue #55736: RuntimeError: CUDA error: CUBLAS_STATUS_INTERNAL_ERROR when calling `cublasCreate(handle)`,stale/low urgency,2,stale/low urgency,1,0,55736,,,,0,5e20209dd87afad6678829d7de5283e7aaf81300cb362c3303f977738faac9f0
thread:issue:55751,issue #55751: [docs] algolia docsearch in pytorch docs,stale/low urgency,2,stale/low urgency,1,0,55751,,,,0,7a29837f4301d601efe1be37febb275139a4ef7a2afd1931c03f2207258abb28
thread:issue:55752,"issue #55752: [question] Influence of divisibility of B, C, T by 16 on Conv1d (and Conv2d perf) with CuDNN, including presence of padding",stale/low urgency,2,stale/low urgency,1,0,55752,,,,0,26b6775c1d875444f9274bd343ff6a2b59a3fc347f9e0094e742d03791ef7121
thread:issue:55755,"issue #55755: Some float16 inputs to CPU matmul are supported, others aren't",stale/low urgency,2,stale/low urgency,1,0,55755,,,,0,888442a1215c20acf96b7c233155e8a650b7926e881c3281d7b5353a601f4120
thread:issue:55756,issue #55756: Cuda OOM after several steps for DataParallel model,stale/low urgency,2,stale/low urgency,1,0,55756,,,,0,c2c5c2f7613cc1c401af4e828e0da2a7b0fe46149853fa790ca95db685485658
thread:issue:55764,issue #55764: clang format on OS X ssl verification failure,stale/low urgency,2,stale/low urgency,1,0,55764,,,,0,4a0ab3f9fc9ae81e0c20795a07b0070fdeb4d86263f0ed711c8cb7001ba8ede4
thread:issue:55784,issue #55784: pybind11 Tensor type caster forces reference count bump,stale/low urgency,2,stale/low urgency,1,0,55784,,,,0,603f001095d11559e56567a91664fefda9c1bb83d0de5a2913056404b541fe5e
thread:issue:55804,issue #55804: Add torch.nn.Conv2D correctness test,stale/low urgency,2,stale/low urgency,1,0,55804,,,,0,1c0fbb7b6cc5031d77315c744bfa0225138987469daac8d10210ce8779cd02a1
thread:issue:55865,issue #55865: Don't define __assert_fail on systems using `musl-dev` headers,stale/low urgency,2,stale/low urgency,1,0,55865,,,,0,a3cdbfc731b418fc3554e9005858730599de87de17f584a026b5ac4722e073ac
thread:issue:55869,issue #55869: We should perhaps prevent linking against LibTorch DLLs with the wrong configuration (debug vs release),stale/low urgency,2,stale/low urgency,1,0,55869,,,,0,3300ccf412d9e34758909f12b5a6d076073daea919c84e99d5b34f827ed66570
thread:issue:55876,issue #55876: [rfc] Trigger callback when backwards begins for DDP with custom autograd function,stale/low urgency,2,stale/low urgency,1,0,55876,,,,0,a70dcab4d8456a88bb02f1c491797fc6dba550585f12750b658a41d635a513f0
thread:issue:55879,issue #55879: Vulkan backend on desktop platforms,stale/low urgency,2,stale/low urgency,1,0,55879,,,,0,aae70ee2e44272696d7ef5b8bd34bd3d88b3192a795daec95d3d6a73d0ba3e40
thread:issue:55884,issue #55884: [Bug] RuntimeError: could not create a primitive on Xeon,stale/low urgency,2,stale/low urgency,1,0,55884,,,,0,c4a9548fce77d8eae8b13d9858cac75b1b0c1cbca396ce8f17e2e86a9decf42f
thread:issue:55902,issue #55902: Warn users when cross entropy is called after softmax,stale/low urgency,2,stale/low urgency,1,0,55902,,,,0,e0b9c5372c429c444253abee1ec54284e81a7771eba6520b5b3a2fb3221c483a
thread:issue:55907,issue #55907: test_variant_consistency_eager_addbmm fails on both cpu and cuda,stale/low urgency,2,stale/low urgency,1,0,55907,,,,0,00c7a614306ce7c7b7afd37a7edc46930f69972bc3d363e56b7f0d06e4de9c70
thread:issue:55910,issue #55910: Allow variable intermediate hidden dimensions for stacked RNN/LSTM layers,stale/low urgency,2,stale/low urgency,1,0,55910,,,,0,c8838964d2e7d17003cf05c69d7e1b2589d9b428465e2ac11e3a6593d1b98eb6
thread:issue:55912,issue #55912: Support float_qparams in quantized_clone,stale/low urgency,2,stale/low urgency,1,0,55912,,,,0,7e0a8ec85397ec95a9ec9923ae65839b575937f57b2bed8f7ffe99a04488143e
thread:issue:55933,issue #55933: [NNC] Replace ComputeAt,stale/low urgency,2,stale/low urgency,1,0,55933,,,,0,443d23e3cbb19401d29df595f12ab73a4f3f0869680b2de1c62e24c85f5abd89
thread:issue:55941,issue #55941: [Tracking] Remove unneeded BC duplicates from native_functions.yaml,stale/low urgency,2,stale/low urgency,1,0,55941,,,,0,839708ebe99ddc3dff9dd2f21914d6082a4709ac0c0c5a9f5a770f77456afee9
thread:issue:55944,issue #55944: Sparse-sparse matrix multiplication only works with torch.sparse.mm(),stale/low urgency,2,stale/low urgency,1,0,55944,,,,0,13687f9bde3aad3bd624233fd4fb987b10753cc5532dbfcf4f3df952991fe346
thread:issue:55945,issue #55945: Quantizable LSTMCell does not work correctly.,stale/low urgency,2,stale/low urgency,1,0,55945,,,,0,112c98bda32bcf3069b825db3ea01dcfb85352ec381b726177838828500a9eea
thread:issue:55948,issue #55948: Deprecate torch.stft returning real-valued tensors and torch.istft accepting real-valued inputs,stale/low urgency,2,stale/low urgency,1,0,55948,,,,0,1a98cb9f495bbe54cec485fae39ec9acb13bfd1fbd11e1f306f3b2dba1b78304
thread:issue:55951,issue #55951: Update linspace and logspace to throw an error when steps is not provided,stale/low urgency,2,stale/low urgency,1,0,55951,,,,0,5c6a4fa00bf808db71e40b28687bb5a5d6b7864df78c169611ff776a9f847e9b
thread:issue:55952,issue #55952: Compiler warnings tracking,stale/low urgency,2,stale/low urgency,1,0,55952,,,,0,fc6d115f7bf1ba0c6f5d56e0e6662989cf74efcf3d595ebdebff0b73b50a0096
thread:issue:55964,issue #55964: Finish deprecating torch.range,stale/low urgency,2,stale/low urgency,1,0,55964,,,,0,d7478b7b0757629f2b86a16c86e8492182dd2f02faf67f6c104406e400178b6a
thread:issue:55967,"issue #55967: Cannot init, destroy, and then re-init process groups",stale/low urgency,2,stale/low urgency,1,0,55967,,,,0,cfc65d28ad3d8785092f7d78dec224084bb67d9b6d7718377d92cfcdb37f0ec5
thread:issue:55981,"issue #55981: LazyEmbedding, an embedding layer with a dynamically sized vocabulary",stale/low urgency,2,stale/low urgency,1,0,55981,,,,0,02dd93cc311e5c85903a0b58dd3b70633fba910d16f7639344c450f4c69b2a35
thread:issue:56012,issue #56012: Failure in complex CUDA numerics tests for sigmoid,stale/low urgency,2,stale/low urgency,1,0,56012,,,,0,a404f6d1a30caa274bd8245c03f7e55c183a7b61e0e7f2d4e90d39acbd2658c8
thread:issue:56023,issue #56023: [Opinfo] Better ErrorMsg for test_out with wrong shape,stale/low urgency,2,stale/low urgency,1,0,56023,,,,0,76ede20be7391b5bd19e426cd4668c042336874a4ab3442f6a2ce18bc1454e76
thread:issue:56030,issue #56030: Add Dirichlet Multinomial to PyTorch Distributions,stale/low urgency,2,stale/low urgency,1,0,56030,,,,0,49372882d3ef034cf4cc956cea975061aa0fccce5cc1d86a3e8159f0f53dbb0d
thread:issue:56038,issue #56038: [Meta] PyTorch features build/test matrix,stale/low urgency,2,stale/low urgency,1,0,56038,,,,0,942ebdfbff7cb41c7a172ea1ee9eac7098ab091a2b4b505ce3de963691e7f449
thread:issue:56064,issue #56064: Optionally include padding_idx items in the EmbeddingBag reduction,stale/low urgency,2,stale/low urgency,1,0,56064,,,,0,7654ab88a94064d107ceede75cdfa4007afcc708e3bfc651a1153e49cff7b9dc
thread:issue:56067,issue #56067: A more flexible torch.hub search strategy,stale/low urgency,2,stale/low urgency,1,0,56067,,,,0,594ea0100303c47b570b3d42ade97491594f5e1148d973e0ebe0d2058b55f298
thread:issue:56100,issue #56100: [package] implicit externing can miss cases where the stdlib depends on further stdlib modules,stale/low urgency,2,stale/low urgency,1,0,56100,,,,0,84820c251eb62f85522ca39bfc80c260b81871f2aea239262b62cce3962fa451
thread:issue:56110,issue #56110: caffe2/utils/signal_handler.cc is failing to symbolize libtorch_cpu.so,stale/low urgency,2,stale/low urgency,1,0,56110,,,,0,149820622e4302d40ba75f991914205ed67b79242c8cc001fe2c136ddbf1d859
thread:issue:56136,"issue #56136: Please add a tab of ""all"" packages in `Docs` option",stale/low urgency,2,stale/low urgency,1,0,56136,,,,0,b9e990e98d969249ab17955fb93215f8ddce664d9da9e83e60ba4b3a7da2fc43
thread:issue:56144,issue #56144: libtorch static linking results in undefined references to onnx_torch and caffe2::EmbeddingLookup,stale/low urgency,2,stale/low urgency,1,0,56144,,,,0,927cf1a77465b6f1c72ea25932a8d7ef90d33befe33ca92315ac5a4dc673d148
thread:issue:56161,"issue #56161: [rfc] Add a ""core only"" build flag",stale/low urgency,2,stale/low urgency,1,0,56161,,,,0,e3d355da1bd508634563b88b52dd7cf495eecb5aeb5124ebc07a255f5f1d6d2d
thread:issue:56177,issue #56177: torch.gradient not throwing error when spacing to be equal to 0,stale/low urgency,2,stale/low urgency,1,0,56177,,,,0,67cb68cb89c9357cafe32686001f235c9737445a58dc545e553347864c7a7199
thread:issue:56183,"issue #56183: scripting LSTM module generates a script that has no .code, no .graph, and cannot be executed.",stale/low urgency,2,stale/low urgency,1,0,56183,,,,0,5d5540976e858969154f7d9e485a7f3767138ac74ca7958058c9888d8be57f62
thread:issue:56187,issue #56187: AVX512 and Vec512,stale/low urgency,2,stale/low urgency,1,0,56187,,,,0,adf5cb3fe36909f05a9e645e26fe33bca8f9dc9288616343ac828494d96b0e1e
thread:issue:56191,issue #56191: [FX] Issues with names of functions in user packages,stale/low urgency,2,stale/low urgency,1,0,56191,,,,0,6e072b95172effd1593a675cdf3c7308c6a8fd72a2f177c6eaaad0857a305fe4
thread:issue:56202,issue #56202: Tolerance for non-determinism operators in gradcheck,stale/low urgency,2,stale/low urgency,1,0,56202,,,,0,41e50e62798b43f7c70bb66906be64540476796c847a2a27abf3824c80dcb8cd
thread:issue:56224,issue #56224: Error message regarding Padding of Conv2d needs improving,stale/low urgency,2,stale/low urgency,1,0,56224,,,,0,ce01c350711d9798fc24e80b5dc1c3143a682ed7f4389fe538d1444bd2caeb3e
thread:issue:56231,issue #56231: Half precision support for torch.sparse.mm,stale/low urgency,2,stale/low urgency,1,0,56231,,,,0,96af6cedd0d4653ea563a83191d5f387d90cd54f6331cf858bc716166d10124d
thread:issue:56235,issue #56235: `test_cholesky_solve` gradgradcheck fails sometimes,stale/low urgency,2,stale/low urgency,1,0,56235,,,,0,709414736f20560dbda7df4214434dc6df41e08ea30fa881f2006e1f567a78e4
thread:issue:56237,issue #56237: optim.Adadelta: local variable 'lr' referenced before assignment,stale/low urgency,2,stale/low urgency,1,0,56237,,,,0,bb92f32c624ba45f808b3ea0122d2b2bb4456b484732298bae6c104339687ceb
thread:issue:56246,issue #56246: Silent incorrect running with zero padding for Conv1d,stale/low urgency,2,stale/low urgency,1,0,56246,,,,0,16e62cf8a0b05577d7f48b1953eb56dd0d50414294c48e9750a4ec2f513ab318
thread:issue:56262,issue #56262: torch.nn.quantized.functional.conv_transpose1d/2d/3d support,stale/low urgency,2,stale/low urgency,1,0,56262,,,,0,4383fe309cbef3f5b5e4c3a33c50df6941eafe43c3df8d1fba0a82224ba59cb2
thread:issue:56263,issue #56263: Reflect padding_mode should be supported for Conv3d,stale/low urgency,2,stale/low urgency,1,0,56263,,,,0,8306ae0afc8f9b27372604d2894ba081c5df2eb130d2b712ccbe45dc9d741620
thread:issue:56275,issue #56275: cumprod gradgradcheck fails in fast_mode=True,stale/low urgency,2,stale/low urgency,1,0,56275,,,,0,d7a772bb9ea9ed5568161362d5558a59e0d3162739e496ecb57d7195231b6f82
thread:issue:56285,issue #56285: qnnpack uses deprecated pthreadpool APIs,stale/low urgency,2,stale/low urgency,1,0,56285,,,,0,57f6fecc3792223b5b1940d54156fd9f6f3c8c369c4865ca38537892c9a23109
thread:issue:56328,issue #56328: Why does output_padding have constraints in ConvTranspose1d?,stale/low urgency,2,stale/low urgency,1,0,56328,,,,0,4b9f5e4a2e288f8bd91b98ed780a31b71c57cede62b60fdba79bb27ef64bff01
thread:issue:56329,issue #56329: Change `other` to `src` in torch.Tensor.scatter_add_,stale/low urgency,2,stale/low urgency,1,0,56329,,,,0,8760d219b07e0ea2c39ef4aaaabd33477a29f3bc527876b8bbf377357c318a5a
thread:issue:56330,issue #56330: Type checking doesn't match actual type in error message,stale/low urgency,2,stale/low urgency,1,0,56330,,,,0,3dd9626639a9234849aead45f75d9f212604bb66ae46afd5081e3f7043cb64d2
thread:issue:56333,issue #56333: Unable to pip install PyTorch on M1 Mac - Errors,stale/low urgency,2,stale/low urgency,1,0,56333,,,,0,dc3f119fb6fec52e6b0dd3cf0bdeb3bf9151d2d2793f9e944d77418c1476f37f
thread:issue:56350,issue #56350: logcumsumexp (and maybe other cum* ops) has divergent CUDA and CPU out= behavior,stale/low urgency,2,stale/low urgency,1,0,56350,,,,0,958a6841a0e03e51d7d19c2a257ed0302f7d318ab2c17726fff3b7dd6bb467dc
thread:issue:56357,issue #56357: [NNC] All NNC APIs should return bool flag,stale/low urgency,2,stale/low urgency,1,0,56357,,,,0,d0178ab5e7099acb7eb89e0746b5405b48abe586f4878fcb15f57817ae03fbd6
thread:issue:56360,issue #56360: Support for stereo audio data in from torch.utils.tensorboard.SummaryWriter,stale/low urgency,2,stale/low urgency,1,0,56360,,,,0,651055888e3661c22a0a698fa99d108bb5c47669d1594320ab18523055cf390a
thread:issue:56370,"issue #56370: Acquiring ""is_grad_enabled()"" inside an autograd function",stale/low urgency,2,stale/low urgency,1,0,56370,,,,0,217dcfcc0fd3a550da5c5b572a35ef921dcfb0804724b560b13ea4e784fe69ba
thread:issue:56396,issue #56396: Tests in CI are run from the test/ directory,stale/low urgency,2,stale/low urgency,1,0,56396,,,,0,eae7cd694b7eb9d4df12e94e94cf5fac843148446179b01b6473ca394eea716a
thread:issue:56399,issue #56399: Figure out what mojo FB common/process/StackTrace.h has that we don't,stale/low urgency,2,stale/low urgency,1,0,56399,,,,0,3e73dd7901abb3026830f9c458024f029683aaa494814ae762f7ad9bc366de8d
thread:issue:56460,"issue #56460: I meet an error assert key in deserialized_objects when I torch.load(pthname),and pth file is trained on multi gpu",stale/low urgency,2,stale/low urgency,1,0,56460,,,,0,998a1336447812c583a29247918185817ac5c903e6ce85c7eff299894370af52
thread:issue:56464,issue #56464: Scatter tests missing when passing a reduction operation.,stale/low urgency,2,stale/low urgency,1,0,56464,,,,0,a3e1dab0164e1de0841980f92d17253f9f7c5870d70f955d1e6a57fab9a23533
thread:issue:56465,issue #56465: Provide a pkg-config file,stale/low urgency,2,stale/low urgency,1,0,56465,,,,0,075d8c6b41ac8f15a5a5b9dffcaf2a49bb6d85f65908290359e59a21906ad696
thread:issue:56482,issue #56482: unsafe_reclaim_from_nonowning is not that unsafe,stale/low urgency,2,stale/low urgency,1,0,56482,,,,0,44e06b72f23d160bd1cafea29ffd3b2659954e9a885c50e3428decf60722edde
thread:issue:56483,issue #56483: Add compute_residuals flag for torch.linalg.lstsq,stale/low urgency,2,stale/low urgency,1,0,56483,,,,0,c08ecebbeee9b33b33a5d736c0f3df264eb6537dd63655204114b7b75ac5b297
thread:issue:56489,issue #56489: [torchdeploy] ar: _bz2module.o: No such file or directory,stale/low urgency,2,stale/low urgency,1,0,56489,,,,0,52c6d58a73b8912c4c67b011eeab21f0f1c55860fc2b830ebfff77f1bc8feb5f
thread:issue:56500,issue #56500: Marking variable not required in backward calculation if it is not needed,stale/low urgency,2,stale/low urgency,1,0,56500,,,,0,b8cf66f787e484691c0a2feebdfaa73930a70d85e552f0c750252df7f640d85c
thread:issue:56522,issue #56522: [cpp op] TORCH_LIBRARY schema doesn't respect keyword only,stale/low urgency,2,stale/low urgency,1,0,56522,,,,0,3ee75299af590788013222d8021720be54a5f6191a0c0ae6d4a779deee989fc4
thread:issue:56525,issue #56525: Can't deepcopy memory format objects,stale/low urgency,2,stale/low urgency,1,0,56525,,,,0,2a051146fc7ecfebf8bbbcfafbcd653b5e4eb6b8e91ae8e964e07093c58572f6
thread:issue:56542,issue #56542: CrossEntropyLoss target shape expectation is inconsistent with BCEWithLogits,stale/low urgency,2,stale/low urgency,1,0,56542,,,,0,c036d83f1b4acdf5a1a987db7711380d258514be90ad885e1970934a70cd1401
thread:issue:56583,"issue #56583: All thread stack trace dumps interleave output in distributed tests, giving garbled output",stale/low urgency,2,stale/low urgency,1,0,56583,,,,0,efe80001c33c1e7bb0da6efbdd56989eb663f62abcfa6580c4bd6ef1dcbc7dee
thread:issue:56634,issue #56634: [package] Module name reported in error message does not always match what is needed to extern/mock it,stale/low urgency,2,stale/low urgency,1,0,56634,,,,0,cfd7f25b9694836013c7c8e5120bf1d0ae828fe6bddb16702d77b3f6d71a85b7
thread:issue:56671,issue #56671: {h / v / d}split methods are missing,stale/low urgency,2,stale/low urgency,1,0,56671,,,,0,a7fdfd55b48ee24a0dbfde0a4ba97109ce9e3f7bc11341793c47ce8e99839984
thread:issue:56688,issue #56688: Torch FX does not work with torchvision model,stale/low urgency,2,stale/low urgency,1,0,56688,,,,0,18e31ca25ab4b1380320aafafee347a3907f752674816f05359a5da64d9b9f29
thread:issue:56691,issue #56691: [structured] at::cpu_unchecked functions,stale/low urgency,2,stale/low urgency,1,0,56691,,,,0,6b4e9babfa69f548ff07663725c49d82b447b4baa5fbd212c09979feb941248d
thread:issue:56694,issue #56694: PyTorch 1.8.0 / 1.8.1 CUDA memory corruption on oddly sized tesnor sections of strided tensor,stale/low urgency,2,stale/low urgency,1,0,56694,,,,0,eb9f9a7e57be7e28dc9db43835a5419b72fb3166d71d7a138c6beead702d2f0f
thread:issue:56698,issue #56698: Convolution2D may have a numerical error.,stale/low urgency,2,stale/low urgency,1,0,56698,,,,0,0fb96b4656a9046dcaa7be8ce0fc4af14dec8d2a5fb99f3083ece8eb945ca224
thread:issue:56707,issue #56707: [TorchScript JIT] `dtype` propagation pass (meta-issue),stale/low urgency,2,stale/low urgency,1,0,56707,,,,0,4b9343c557f341b804ef3bc957f1abbcdd99fd71d87c7e1cb31df09b02ceafa1
thread:issue:56710,issue #56710: Hard fail build when there is no CUDA but `USE_CUDA=1`,stale/low urgency,2,stale/low urgency,1,0,56710,,,,0,58747e3f5cb72df36c53b40b1857e05dd685b585e7e284a1d956b85e74fa04c9
thread:issue:56772,issue #56772: The multi-fc losses calculating in DistributedDataParallel.,stale/low urgency,2,stale/low urgency,1,0,56772,,,,0,a411fa00c5abe14bd095a6d2db64e9eaf4d26c6c875cbd5652cb41dbae58cccd
thread:issue:56774,issue #56774: Support `expand_dims`,stale/low urgency,2,stale/low urgency,1,0,56774,,,,0,25d941471ef4da51defc4da1ba83df69b4d14c9399d8f0890d0ca928586831ad
thread:issue:56777,issue #56777: Error when building PyTorch from source in CentOS Linux 7,stale/low urgency,2,stale/low urgency,1,0,56777,,,,0,1e848d1dcc9435b8dca010f00c9e493b6bdbea6509f6fbb1d57af5042b766e79
thread:issue:56778,issue #56778: can't find user's fuser method in quantization.fx.fusion_patterns.py when run fx graph qat.,stale/low urgency,2,stale/low urgency,1,0,56778,,,,0,546699d5f8e2c904584b282de51372f9ad3bea392ffbf9181e1a72d41375b5f4
thread:issue:56781,issue #56781: Named Tensors in C++ Is Undocumented,stale/low urgency,2,stale/low urgency,1,0,56781,,,,0,8cccea07adb70431cbf81f910fb10696c098bb2a2c6e054a4576f1e99b6399cc
thread:issue:56805,issue #56805: Implement random SeedSequence,stale/low urgency,2,stale/low urgency,1,0,56805,,,,0,a70afdd086c1eb6407195c0569196c04022ecbec3bab06bd4f0be2d6d73e4a40
thread:issue:56809,issue #56809: Autograd engine worker thread initialization fails with python 3.9 debug build,stale/low urgency,2,stale/low urgency,1,0,56809,,,,0,6e8ed3e10157e5bd3eaf1a6d2657fc19d9da0d671ffaae3ac728616141d96f33
thread:issue:56836,issue #56836: Remove single device constraint from ProcessGroupNCCL profiling,stale/low urgency,2,stale/low urgency,1,0,56836,,,,0,f49a9322b865b74a35487b4be617fddb20555d4986e0dc773426e5b10740681d
thread:issue:56838,issue #56838: Comprehensively test NCCL's `get_future()` API,stale/low urgency,2,stale/low urgency,1,0,56838,,,,0,c59f4b239a27d293483fcca6407262c802ae7172b791dfd6609d460d3a62dc6b
thread:issue:56843,issue #56843: Add clipnorm parameter to optimizers,stale/low urgency,2,stale/low urgency,1,0,56843,,,,0,ea4bbacb0c848bb1bd0f1aa75a7e2bd29aedafb583d56060cdd16a33cfd6e9ef
thread:issue:56862,issue #56862: Non-symbolic FX tracer,stale/low urgency,2,stale/low urgency,1,0,56862,,,,0,f0131c4c1b7f2e9290eb74ae88215a4ca7dc0210a8ea703b9f72b484146a8c7a
thread:issue:56881,issue #56881: [FR] integral discrete distributions should support dtype=int64,stale/low urgency,2,stale/low urgency,1,0,56881,,,,0,ca3985c94ee644e16ccc79e655c69957b55f4f4db5dfbcd4aeb166ce9f45ba6a
thread:issue:56896,issue #56896: masked_select is x3 slower than reshaping and index_select,stale/low urgency,2,stale/low urgency,1,0,56896,,,,0,c41d785628c9c658094ecb3600e2f652ff3ddaffad69e691cd74c5c01c7d478c
thread:issue:56901,issue #56901: Static prebuilt libraries contain shared libraries,stale/low urgency,2,stale/low urgency,1,0,56901,,,,0,a7e264ae2da752ef5d35079221eb7215faafc275c4925e4914cf97aa714f8edd
thread:issue:56926,issue #56926: CUDA 11 tracking issue,stale/low urgency,2,stale/low urgency,1,0,56926,,,,0,face03b493426360c1db9a5f4e2d4be8040faca80456e2f011d46b45af4891d0
thread:issue:56938,issue #56938: remove the dependence on node names in fx graph mode quant,stale/low urgency,2,stale/low urgency,1,0,56938,,,,0,afaced81c8f62e6dbafd89083af8923c313c3065a6fa12286bc8af37025dc0b2
thread:issue:56978,issue #56978: Do padding of weight and activation tensors to match optimized backend implementation,stale/low urgency,2,stale/low urgency,1,0,56978,,,,0,bc27406222c5747cf08ca11268ae1f165b41d39b776e1a94ca2c871406744ca2
thread:issue:56979,"issue #56979: apex-0.1-py3.6-linux-x86_64.egg/apex/amp/wrap.py"", line 28, in wrapper return orig_fn(*new_args, **kwargs) RuntimeError: CUDA error: an illegal memory access was encountered",stale/low urgency,2,stale/low urgency,1,0,56979,,,,0,405dd0b89f6910b346b32149b1e66e383d4759f607a135edcba7449e96730824
thread:issue:57005,issue #57005: about torch install,stale/low urgency,2,stale/low urgency,1,0,57005,,,,0,644799d96834941860f6d4c99ab175b12544369d23dfd26e66f4a6d715e80aae
thread:issue:57018,issue #57018: Attempting to concatenate scalar tensors throws a runtime error,stale/low urgency,2,stale/low urgency,1,0,57018,,,,0,0e7e7ed78e468c67928741ce0a9d7d79f49a5cb34d270e77191f3a0f9f8642d8
thread:issue:57059,issue #57059: Get rid of weak_intrusive_ptr,stale/low urgency,2,stale/low urgency,1,0,57059,,,,0,b5a355fee53aace8adc7c1ffaba7ba1a97f8c04ce651dc7e962b22716a332ec8
thread:issue:57065,issue #57065: Static Linking Pytorch,stale/low urgency,2,stale/low urgency,1,0,57065,,,,0,d248120dd95da5bd9dd9043fb41c5a44d1f54edfd84e2b94af7cd85d18058521
thread:issue:57095,issue #57095: [FR] unflatten support empty shape if unflattened dim has size 1,stale/low urgency,2,stale/low urgency,1,0,57095,,,,0,07df9e74f108c1450a4d7414706d970125dccd1f1350e5c86540a7d945b009eb
thread:issue:57108,issue #57108: Compilation failure in caffe2/core/plan_executor_test.cc with GCC 5.4,stale/low urgency,2,stale/low urgency,1,0,57108,,,,0,6fd11a0de7d74754839c77324d4f328446d3055cb55edc8c6f60112c0dc6544c
thread:issue:57112,issue #57112: :attr: does not behave as expected.,stale/low urgency,2,stale/low urgency,1,0,57112,,,,0,3efe2e2493fd827509c3270264ff446372a539a488135e0971b4a1e4fcc5b603
thread:issue:57117,issue #57117: MultiheadAttention.out_proj.weight is not explicitly initialized,stale/low urgency,2,stale/low urgency,1,0,57117,,,,0,6353b646e5b766fdcb3c2e1e2c7a23fe71ec582ba3d97965083572b4a2d35ec7
thread:issue:57123,issue #57123: LibTorch ships with two identical library files: libdnnl and libmlkdnn,stale/low urgency,2,stale/low urgency,1,0,57123,,,,0,16d1e735d8a3f30f97235cbb01e9f3afd1e72e3c0b9a9a8f6fbd28dcc6bb04fa
thread:issue:57124,issue #57124: Allow `ScriptFunction`s to be symbolically traced,stale/low urgency,2,stale/low urgency,1,0,57124,,,,0,f12081498ca7eeb04657da07d082aa98cd8b65863a481fd804dc525ad42b4ad5
thread:issue:57136,issue #57136: Intermittent segmentation fault when using rpc_async/rpc_sync with CUDA tensors,stale/low urgency,2,stale/low urgency,1,0,57136,,,,0,9cd04a6366c385d27f98dc3914df88623c78c44f045fc39af1d05127e7318a29
thread:issue:57157,issue #57157: Using register_full_backward_hook with target module for intermediate activation,stale/low urgency,2,stale/low urgency,1,0,57157,,,,0,e9efbd795ee5f88c20cc3f76a4140f1612012cb6c8c4d72c859f8523d7d30b1e
thread:issue:57158,issue #57158: Increasing RTT when using rpc_async in a PS architecture setup,stale/low urgency,2,stale/low urgency,1,0,57158,,,,0,cc4668d20207e08e2d5803e8314a1462f173d83631a2aa2903fd30e0ecfef759
thread:issue:57185,issue #57185: Segmentation fault with ITIMER_REAL,stale/low urgency,2,stale/low urgency,1,0,57185,,,,0,5a947a6af577b02087848272abcc810226e8a92f6473118eecbf775b66ce462e
thread:issue:57230,issue #57230: Implementation of Self Attention vs Encoder Decoder Attention Causing Early Evaluation for PyTorch/XLA,stale/low urgency,2,stale/low urgency,1,0,57230,,,,0,836410f7165966f4fdfd0f17a1a8c08bdba25b35042f43047d176abe493c9022
thread:issue:57272,issue #57272: Consistent treatment of non-differentiability in linear algebra operations,stale/low urgency,2,stale/low urgency,1,0,57272,,,,0,a41a02494db29b1e9417b93e6c1174b0af6bb011a04287123e00a9066c0bc5ac
thread:issue:57278,issue #57278: Attributes consistency in Sampler and DistributedSampler,stale/low urgency,2,stale/low urgency,1,0,57278,,,,0,24aa7d87c3257c7881514c2cbc2e726b42be802bb0c15c37c35277be788cbb85
thread:issue:57301,issue #57301: Avoid code repeat in create sample inputs for sort/msort,stale/low urgency,2,stale/low urgency,1,0,57301,,,,0,b1af154a4dd6c61d7b787d4464fbf1e4475e0f5496380bb0899979928213e92d
thread:issue:57326,issue #57326: Renaming Autograd dispatch keys to ADCreateGraph,stale/low urgency,2,stale/low urgency,1,0,57326,,,,0,5d685c32c19c93fa88774f45d751872635d2d8752ab6b5a1981649a8dc6d5dbc
thread:issue:57358,issue #57358: Missing complex autograd support for some operators,stale/low urgency,2,stale/low urgency,1,0,57358,,,,0,460d9b6b8b0d8de9bb0e7bb46fd9fc22a4f06f3d4e928741dd4e1acdd2be1879
thread:issue:57359,issue #57359: torch.return_types does not exist,stale/low urgency,2,stale/low urgency,1,0,57359,,,,0,e6ff03c64f549dd21cf1aa89535079ccd192a153565722ffc8664704dfe3f5e7
thread:issue:57369,issue #57369: Delay errors of inference tensor to backward pass?,stale/low urgency,2,stale/low urgency,1,0,57369,,,,0,a5d6b2284aac735223f9a1d5ef92fd9bbb5631ee673166dd212e6345d0a4eb88
thread:issue:5737,issue #5737: [feature request] Callback on learning rate drop in torch.optim.lr_scheduler.ReduceLROnPlateau,stale/low urgency,2,stale/low urgency,1,0,5737,,,,0,0512db70f72cb109560f3ec1753e3f36123cfe87840f8cfa97f6fe81a1f57d63
thread:issue:57370,issue #57370: gen_proto_typestubs_helper.py is slow,stale/low urgency,2,stale/low urgency,1,0,57370,,,,0,b6ce2675887bcf621357e608041295798dd43425acf0b79d5e59eb6964553c51
thread:issue:57380,issue #57380: Asserts all tensor are defined in dispatch wrapper,stale/low urgency,2,stale/low urgency,1,0,57380,,,,0,8aa59501516291d2afa9f3a0da72bdf9a7cf3f62c64a6af53bde4a5888318faf
thread:issue:57392,issue #57392: Simplified API for custom inplace & view kernel,stale/low urgency,2,stale/low urgency,1,0,57392,,,,0,d743cbfa23d2c04ad63a19f7cc8286e3f0d455674f77c7732753a4fa10bdb950
thread:issue:57404,issue #57404: [lint] Make sure that all CMake options have a corresponding output in the summary,stale/low urgency,2,stale/low urgency,1,0,57404,,,,0,dc627e1c1d181e5409396e90bd2cc3cdd7eee7038dfdc29aa7fc206008514cc2
thread:issue:57406,issue #57406: nn.Module custom setattr leads very confusing behaviors,stale/low urgency,2,stale/low urgency,1,0,57406,,,,0,04848ea0b6d9e151cc2853be7a78ab7408dea98d9e817b4d1e3ae4edd5d1e52e
thread:issue:57415,issue #57415: torch 1.8.1+cuda crashes when setting breakpoints.,stale/low urgency,2,stale/low urgency,1,0,57415,,,,0,50a2ad83e1dc59ba3c7d1141fb2a9122670a541cbefb631e96fe37abff75ace0
thread:issue:57423,issue #57423: internal compiler error compiling PyTorch master on Fedora 33,stale/low urgency,2,stale/low urgency,1,0,57423,,,,0,b7839e78d8ee1d477c95077a24134329409a796a9d36592e1ac9d3646de089c4
thread:issue:57440,"issue #57440: I have just wasted 3 solid days trying to build PyTorch, without success",stale/low urgency,2,stale/low urgency,1,0,57440,,,,0,e90c44d187ed748c16a8d759618dccfb94aa7673cd2a810629c9eca1530eb971
thread:issue:57447,issue #57447: `torch._dirichlet_grad` returns `nan` value on cuda device,stale/low urgency,2,stale/low urgency,1,0,57447,,,,0,6d199e2f5becfb9f2a779b2d6551b753d2ca0b939c178461ba0364a9130cfb33
thread:issue:57475,issue #57475: CUDA RPC stream synchronization does not work with @rpc.functions.async_execution,stale/low urgency,2,stale/low urgency,1,0,57475,,,,0,f26ecfd33b8f13ff55ad4ffc6b19e9206dc24ffe37c5b566585a5f0312b0407c
thread:issue:57495,issue #57495: Please remove this assertion - it triggers on valid use cases,stale/low urgency,2,stale/low urgency,1,0,57495,,,,0,d8a2d43133e1b30607de50226741b4dd0ee24320839dadc8b2794c37d90d12c9
thread:issue:57502,issue #57502: [structured] Fail if CompositeExplicitAutograd is in dispatch table of structured_delegate function,stale/low urgency,2,stale/low urgency,1,0,57502,,,,0,926a6fcbf5799f675c201b071e65a062ae4b23c087effac0bf4ee3e9a57cafe0
thread:issue:57503,issue #57503: [new feature] Adaptive gradient clipping,stale/low urgency,2,stale/low urgency,1,0,57503,,,,0,d249b2f3ad69983bdaf07747db21cccd40a71039a3782c1e7dc1ec53adfb906d
thread:issue:57507,issue #57507: Cleanup usage of IS_PYTHON_3_9_PLUS in autograd engine,stale/low urgency,2,stale/low urgency,1,0,57507,,,,0,bd6abc64790c00bd3b5f37693460ccddb53e006fff089110826b94e2c5e1f9ba
thread:issue:57533,issue #57533: test_grid_sample (from TestNN) fails on POWER,stale/low urgency,2,stale/low urgency,1,0,57533,,,,0,5b453b662d58864410b34bbd768e9109fbf294296cff97f458d024f9535dd740
thread:issue:57547,issue #57547: Add distributed testing for CUDA aware MPI,stale/low urgency,2,stale/low urgency,1,0,57547,,,,0,5494f01468afc6b95edcc4d0ebe41d2a4cea7af046c8c648176478dc7c738662
thread:issue:57593,issue #57593: [RFC] Provide an API for Structural Performance Tips in DDP,stale/low urgency,2,stale/low urgency,1,0,57593,,,,0,47729496cac92c00a55ba346b22917caba36860baa3154ee3f3adc6b0ce405d7
thread:issue:57611,issue #57611: [perf] 10x improvement on element-wise operations with manual broadcast,stale/low urgency,2,stale/low urgency,1,0,57611,,,,0,d7a7ae4d78ffeab70fb37fb1b4681443635566ff1ceaa4b6fc9180b1964b4f81
thread:issue:57631,issue #57631: Pruners' compute_mask returns tensors with dtypes that are not consistent with each other,stale/low urgency,2,stale/low urgency,1,0,57631,,,,0,b1bcee38d79c646115b457e2c19cccbd3272beae5fefd9292d8516ba7491101d
thread:issue:57646,"issue #57646: get_future() should be documented once it is enabled for gloo, mpi backends",stale/low urgency,2,stale/low urgency,1,0,57646,,,,0,9c693525dfa5f6ede85f0ab3f33add7560b3854121eb7ff811791ea8cbd3dc54
thread:issue:57690,issue #57690: torch.cdist returns high diagonal values with CUDA,stale/low urgency,2,stale/low urgency,1,0,57690,,,,0,0288091e411512ded03d53e469d96a4a093d73d489590d9ba60452efdacc26d5
thread:issue:57691,issue #57691: Unify `std::getenv` usages,stale/low urgency,2,stale/low urgency,1,0,57691,,,,0,58bb2f9f4be15ead8932e21b6dc5216adca6d9a73724f1114c88f842c2b12b73
thread:issue:57712,issue #57712: [Docs] nn.init._calculate_fan_in_and_fan_out(),stale/low urgency,2,stale/low urgency,1,0,57712,,,,0,5bc07b313822a0d3012e5174e3a07545b22a140b5d5902d7ec41fece4610285b
thread:issue:57721,issue #57721: nn.TransformerEncoder cannot deal with large negative value even when this value is masked by src_key_padding_mask,stale/low urgency,2,stale/low urgency,1,0,57721,,,,0,bb6157f67c0d12d74fdaae68b8981a1306c7a4c7043568660ef1e82e3189e5d1
thread:issue:57742,issue #57742: [structured] Meta-impl split implies redundant dtype tests,stale/low urgency,2,stale/low urgency,1,0,57742,,,,0,b0fc44a88f8a637063509852ca68d0225db6de1e57e3db421e4f79e95ee75b1a
thread:issue:57759,issue #57759: Show commit hash for master branch docs on pytorch.org,stale/low urgency,2,stale/low urgency,1,0,57759,,,,0,38f04efb5037faf133b12bce7b69840c42e61a57e058666544b58b8dabfa96df
thread:issue:57773,issue #57773: BFloat16 CUDA GEMM ops unsupported on Nvidia P100 (SM_60) on CUDA 11.3,stale/low urgency,2,stale/low urgency,1,0,57773,,,,0,38c5c53303df9747eb480dfdeb7d26377172d2f07129fc2f57d9bb09d95106cc
thread:issue:57776,issue #57776: [rfc] Build a debug tool to help users find which parameters are unused during DDP training,stale/low urgency,2,stale/low urgency,1,0,57776,,,,0,1ea9cde898a9c458dffe96e7566440af1b7a6998472004a9a48b5644ddd4ca3a
thread:issue:57787,issue #57787: Add CI automation tests for Infiniband for DDP,stale/low urgency,2,stale/low urgency,1,0,57787,,,,0,b629234aa22efb779d13fa00afaa7dfb178f73aba7571a4069939f6606b66db3
thread:issue:57843,issue #57843: Torch.save doesn't make new file,stale/low urgency,2,stale/low urgency,1,0,57843,,,,0,cb7a9c386c2127ff26aab53a86aff6cacb6e16210cbe4b4138523ec1d5002166
thread:issue:5790,issue #5790: Add hookable weights,stale/low urgency,2,stale/low urgency,1,0,5790,,,,0,3649da0c7fa84c5d78383fdbfd46d79b77c807c309bb6cb252b050db25b5e221
thread:issue:57921,issue #57921: Backward pass for a nn.Conv2d with half-precision on Quadro RTX 8000 leads to CUDNN_STATUS_INTERNAL_ERROR,stale/low urgency,2,stale/low urgency,1,0,57921,,,,0,08a13376e84c4026af166516784f4192b8fbb3bf1a634f65ca3a801a46cbbce2
thread:issue:57949,issue #57949: [structured] Clarify if NoNamesGuard is actually needed in impls or not,stale/low urgency,2,stale/low urgency,1,0,57949,,,,0,cf95e00afe731d5946e5e27a93722565575ffbebc425e69f158d6ae74f9bd928
thread:issue:57957,issue #57957: Add a few tests to make sure new dispatch keys for backends are added properly.,stale/low urgency,2,stale/low urgency,1,0,57957,,,,0,f9d81aad6ce0b0deb966fc33337fd6f0212d6ed342ad985ee38e398d83d73109
thread:issue:57960,issue #57960: Use the same CUDA stream for all RPCs within the same dist autograd context,stale/low urgency,2,stale/low urgency,1,0,57960,,,,0,b0b1358fd9c7168c5d644ce0c2fa806b4fafdaceec7dce3c05928a8f3dc567d6
thread:issue:57987,issue #57987: `type: ignore` everything,stale/low urgency,2,stale/low urgency,1,0,57987,,,,0,940a6c18c5414ffcf75cab958a8756e1bfe2b567be3817aa36321db828cfd5a7
thread:issue:58015,issue #58015: Support a dist.group_like API,stale/low urgency,2,stale/low urgency,1,0,58015,,,,0,3f8b248a1716a5152c54dde924d6a5f4285354cf36cbfdcda1173ac465f68467
thread:issue:58016,issue #58016: Eliminate potential double device check,stale/low urgency,2,stale/low urgency,1,0,58016,,,,0,7130f52d105077c160fdb4a3fbcaf47ae9f89b265b11cdbc9ee734bf9b0f2447
thread:issue:58037,issue #58037: linalg.eigh and linalg.cholesky UPLO flag breaks in backward,stale/low urgency,2,stale/low urgency,1,0,58037,,,,0,4cd543f08e5de2dca1adb10746d0fd0ef19ae78541e1ce0e899600c1052848a3
thread:issue:58051,issue #58051: quantize_per_tensor not symbolically traceable with FX if scale+zp are proxied,stale/low urgency,2,stale/low urgency,1,0,58051,,,,0,4929d754f6d83e19d8339bd3d68cd78a2273766eb72adf50098d5f21c062b788
thread:issue:58054,issue #58054: PyTorch master failed to build with glog 0.5.0.,stale/low urgency,2,stale/low urgency,1,0,58054,,,,0,59b3845be9ea5afa6a926c3c2dfe48583f4c5c9798f9ccc039c026563adb8b4c
thread:issue:58087,issue #58087: TensorList upfront checks parameter types in argument parsing and throws an error unexpectedly,stale/low urgency,2,stale/low urgency,1,0,58087,,,,0,be24a164eaa53cf96f6ed666a0f8d30ac9a923eb9ec936b8cf18ecd51f715bee
thread:issue:58091,issue #58091: Follow-up updates to inference mode python bindings,stale/low urgency,2,stale/low urgency,1,0,58091,,,,0,2aa600358880cf589296b807f828e1039d7abe8f1ca1a6464e994a5790b69432
thread:issue:58093,"issue #58093: TrilinearBackward takes 98.4% of total computational time, is this to be expected?",stale/low urgency,2,stale/low urgency,1,0,58093,,,,0,35383a4be5c416ef74bba7485e3e883b79c7397eea2188daac42e2e413e48135
thread:issue:58124,issue #58124: Python API binding code generation does not need to pack TensorOptions for `xxx_like` generators,stale/low urgency,2,stale/low urgency,1,0,58124,,,,0,c8a04e650397a26128073e673a1534ae9c72034de51275d2be873aace499ebb6
thread:issue:58127,issue #58127: Save checkpoint error,stale/low urgency,2,stale/low urgency,1,0,58127,,,,0,e5cbe5ddb144bb327689189dcefa95c6fe2d549d9c58edf26098003039e06b7f
thread:issue:58139,issue #58139: [docs] Strange torch.unique function signature,stale/low urgency,2,stale/low urgency,1,0,58139,,,,0,70b80d64ff555a551eeb4413982f58ac5483646919d03879a2252ce593748b69
thread:issue:58155,"issue #58155: ""uncorrectable NVLink error"" making the tests fail",stale/low urgency,2,stale/low urgency,1,0,58155,,,,0,6ace8a1bdb38de281763f2fd8cfea25c6f53ad5a3882e5d9aea6114c0207c451
thread:issue:58167,issue #58167: Allow `Union` as a `Dict` key,stale/low urgency,2,stale/low urgency,1,0,58167,,,,0,33898c2c5c48a82b860da5c79756445d3c01606f1e808a44833989c430381795
thread:issue:58171,"issue #58171: Error while converting model for NNAPI ""Unsupported node kind ('quantized::batch_norm2d')""",stale/low urgency,2,stale/low urgency,1,0,58171,,,,0,d948d199a8361d9e0457241a3296c594956029d1f0de5a5838c6ec15f0b2be41
thread:issue:58172,issue #58172: Mobile Android: Could not run 'aten::quantize_per_tensor' with arguments from the 'Vulkan' backend,stale/low urgency,2,stale/low urgency,1,0,58172,,,,0,bef903eed446b6512f58b11a8c3482a820f3e283523407d4fc0df71c05f68f76
thread:issue:58218,"issue #58218: Torch.nn.DataParallel training model, the output of the model becomes list type, the number of lists is batch size",stale/low urgency,2,stale/low urgency,1,0,58218,,,,0,d3221c9ec82b514d825ae6ed8b6f093df4ed627eb92752e678b48c07e5371ce4
thread:issue:58238,issue #58238: torchdeploy doesn't prevent Obj from being used on wrong interpreter,stale/low urgency,2,stale/low urgency,1,0,58238,,,,0,2f44014712c431ac7d174e8f44b6d9777e9722938aeeb5fc49ab309571263c37
thread:issue:58267,issue #58267: Pytorch build issues on PowerPC,stale/low urgency,2,stale/low urgency,1,0,58267,,,,0,2be54959937254b5eb88ba9c789162e09f45919c120c78d9bdc181a5c687c7ac
thread:issue:58343,issue #58343: Add the features like Keras,stale/low urgency,2,stale/low urgency,1,0,58343,,,,0,ddc83444db62f0d927d814f527cfc39f9088f88639016dca714d2dd1ec0a84fb
thread:issue:58351,issue #58351: need sampled softmax loss function to train model to get user vector and item vector with a large dictionary,stale/low urgency,2,stale/low urgency,1,0,58351,,,,0,1b44cf3caccf1f845117507d3ea8b445ef5748463947691e3b2d684ea6a2bd01
thread:issue:58352,issue #58352: Broadcasting multiple tensors to all procs in distributed training,stale/low urgency,2,stale/low urgency,1,0,58352,,,,0,42110bbc44533ee72e1f7b6ca8aa2692cb957814fd2b57afc6d258346fc7bdf6
thread:issue:58371,issue #58371: Tests do not pass even though build successful,stale/low urgency,2,stale/low urgency,1,0,58371,,,,0,8809852ec3a941d75b8b839f44ec81202610c7c6276769723cc1034df0b311ab
thread:issue:58380,issue #58380: Clarity of error message in einsum regressed in performance improvements,stale/low urgency,2,stale/low urgency,1,0,58380,,,,0,9d91263c4e9f11c5e7259395a9281cf941d9fac35e47e08093a945ff5f651c36
thread:issue:58385,"issue #58385: A ""sanitizer"" mode for the CUDA caching allocator?",stale/low urgency,2,stale/low urgency,1,0,58385,,,,0,563a8a53d429786c1e60c542007ee0ba7f2f05a9cabd22c808c90c42606bf799
thread:issue:58386,issue #58386: Better shared memory allocation under Docker,stale/low urgency,2,stale/low urgency,1,0,58386,,,,0,ad108228377be2cdeacd935f392c32144472626efb3277640ecf558afe47bbf3
thread:issue:58393,"issue #58393: In Android, the memory used by the tensor or model cannot be recycled. Is there any way to solve it?",stale/low urgency,2,stale/low urgency,1,0,58393,,,,0,0eff3408d88c18a051ceae1cbcbce6ec6a151336297db8fce08f16b614575159
thread:issue:58395,issue #58395: [DOC] Add info about differentiability of functions,stale/low urgency,2,stale/low urgency,1,0,58395,,,,0,8860bbf21cedd249815ce5c01c14cb1c60363153a32e7eb718290f3c1aeb85f5
thread:issue:58402,issue #58402: fx OpInfo tests don't work with nn.functional test names,stale/low urgency,2,stale/low urgency,1,0,58402,,,,0,cbed69910ef1ebfd9e99036a1e9952ce40cdb93864fa86bb9192f7f99da6f394
thread:issue:58442,issue #58442: TSAN issues in autograd,stale/low urgency,2,stale/low urgency,1,0,58442,,,,0,084a99f34115685c6ba6f6b06478d295ac66685e066ab8736e949363bd41a046
thread:issue:58447,issue #58447: TSAN issues in at::RecordFunction,stale/low urgency,2,stale/low urgency,1,0,58447,,,,0,bdde4adf1ee8cf5a094f0104eb86db3ca1ada58ae9335400a1b798e2f649d00f
thread:issue:58471,issue #58471: Why the same training code running on different GPUs (V100 vs P100) brings a significiant difference in outputs and thus train loss / model accuracy?,stale/low urgency,2,stale/low urgency,1,0,58471,,,,0,1abf664fa84dae23a5ecfb533a241b66325e0d50f77fad66c2f2c5c31f7dd2a4
thread:issue:58474,issue #58474: nn.Upsample result mismatch in 1.1.0a0+828a6a3 and 1.9.0,stale/low urgency,2,stale/low urgency,1,0,58474,,,,0,44a45747ddf18787ff90c58dd33a13b69f9da4744f03911dcf979d688e8218f2
thread:issue:58479,issue #58479: Make SavedVariables aware of pyobjects,stale/low urgency,2,stale/low urgency,1,0,58479,,,,0,6ecbfe030b075153cd51b59905abc883ef12e906dd9971c484468184df4c95be
thread:issue:58522,issue #58522: DDP grads dont have parity with local training when grads are undefined,stale/low urgency,2,stale/low urgency,1,0,58522,,,,0,a2975ce1e814202a42933e229efab2b131b811abbed61c7c52a942cc233393b4
thread:issue:58533,issue #58533: Migrate wheels to manylinux2010,stale/low urgency,2,stale/low urgency,1,0,58533,,,,0,f90fa93b91f9c38bff92ce859d3c05f2b489db95d441a8f1c92bfc071ad2b655
thread:issue:58548,issue #58548: Pixel shuffle support for PyTorch Mobile NN-API,stale/low urgency,2,stale/low urgency,1,0,58548,,,,0,441aed66193bd0aec38c06d02a7151ba205d378077be17b56e19ace33584a32a
thread:issue:5857,issue #5857: Change THCudaCheck to suggest that device-side asserts likely mean that you have out of bound indices,stale/low urgency,2,stale/low urgency,1,0,5857,,,,0,7a66ae571c6663f816c39f004629d67f1160ef3604fab4f0972701af5eb32a84
thread:issue:58585,issue #58585: Calling benchmark.Timer with default `num_threads=1` disables parallelism permanently,stale/low urgency,2,stale/low urgency,1,0,58585,,,,0,be017da7a6cc543419345f3b8459698a0822e4814c1990b13fc5d5d990260cc0
thread:issue:58591,"issue #58591: link error on BatchNormImplBase<D, Derived>::pretty_print when using icpc",stale/low urgency,2,stale/low urgency,1,0,58591,,,,0,96dba43db36a2792c88df0525ba632995240eb5c00f4ca6520e26c4ffb62dd02
thread:issue:58596,issue #58596: Some cuda objects take very long time to build,stale/low urgency,2,stale/low urgency,1,0,58596,,,,0,0bd4495bfb281bd5c28386c7404a729b44dd6b08f4daf3af5fe1021e0acb399b
thread:issue:58617,issue #58617: [rfc] Bug developers about release notes for new APIs,stale/low urgency,2,stale/low urgency,1,0,58617,,,,0,3e1ab71578ed8be21966c1531a9fc754ee50a499f54e2e4825bdcf475fca407c
thread:issue:58626,issue #58626: GPU 0 context created on GPU 1 worker when using pin_memory=True,stale/low urgency,2,stale/low urgency,1,0,58626,,,,0,ead6a12af2a84606dd27861363c33d2873f63aef6ee06ddbc21223d579480600
thread:issue:58657,"issue #58657: Improvement suggestions for the error message ""Expected batch2_sizes[0] == bs && batch2_sizes[1] == contraction_size to be true, but got false""",stale/low urgency,2,stale/low urgency,1,0,58657,,,,0,f548e2e97c9579159e253b4112f0806a1e4d574487b60d0eeb6e9428b2377d45
thread:issue:58670,issue #58670: Ability to enabling/disabling cuDNN and cuBLAS API logging in PyTorch API directly,stale/low urgency,2,stale/low urgency,1,0,58670,,,,0,15e588619586aad59ae9aacdd205165250ce11596de843837b81cd2b055aa769
thread:issue:58671,issue #58671: torch.load with dill is unable to unserialize from buffer,stale/low urgency,2,stale/low urgency,1,0,58671,,,,0,cae58a17ebb223f3557d1e53104e7ee3318385857c7bf41aac6a95768043a6c4
thread:issue:58672,issue #58672: Werror should include -Wmissing-prototypes,stale/low urgency,2,stale/low urgency,1,0,58672,,,,0,a3756a102eb910e3e7a29cf0466f0cfb8157d76c43c3e43fe12257c84969abc2
thread:issue:58720,issue #58720: `pickle_save`/`pickle_load` are undocumented,stale/low urgency,2,stale/low urgency,1,0,58720,,,,0,b6ca5a1a00da128a1d86901ffcef3d90d999310f124567181010e108dfa0d844
thread:issue:58735,issue #58735: CSR construction: safe_get_attr_string suppresses real errors,stale/low urgency,2,stale/low urgency,1,0,58735,,,,0,f822f492ad2bf4e1cd19e035d72fb3334a569b3c96f2fe3949446f67ae056bda
thread:issue:58758,issue #58758: precision/consistency issue in `linspace`,stale/low urgency,2,stale/low urgency,1,0,58758,,,,0,0a49d8d93d458a16c3351329e698a0f1020094fd5b29cdaf03ff535196b7e2ed
thread:issue:58765,issue #58765: [codegen] generated inplace/out= wrappers don't have input checks,stale/low urgency,2,stale/low urgency,1,0,58765,,,,0,cf5ddfc30583b0626b31b373121a56058ccef490a724f5161c5c2a2620a3df20
thread:issue:58791,issue #58791: [nnc] Add more optimization for conditional stmts in loopnest transformations,stale/low urgency,2,stale/low urgency,1,0,58791,,,,0,09919b837e0d5b5ad2faded2cf58b5d1713d3a992e6592f784402301fe92bccf
thread:issue:58812,"issue #58812: Transferring tensor to the gpu and converting dtype in a single call to .to() is slower than first transferring, then converting.",stale/low urgency,2,stale/low urgency,1,0,58812,,,,0,f1fba47aedc490c7c41cdb8b1d1a36a4e9312b9161f552bea71cc9e61b43a4f3
thread:issue:58813,issue #58813: Single-Process Multi-GPU is not the recommended mode for DDP,stale/low urgency,2,stale/low urgency,1,0,58813,,,,0,531929e4e3bcb20e13c6c613d5c8f17f596ebb841b9b25d72a7bc3916250dca2
thread:issue:58814,issue #58814: TensorIterator for sparse layouts,stale/low urgency,2,stale/low urgency,1,0,58814,,,,0,ddbc0634a39811aed1b225be38dbbb5b2d60cab356b24b9909d333b58797297d
thread:issue:58820,issue #58820: [docs] Tensor.bernoulli_ formatting is hard to read and UX inconsistent with the function variant,stale/low urgency,2,stale/low urgency,1,0,58820,,,,0,97a107a89df7fe0b1782e659a53b5b2e3ed98a4a9044d70e08af1522693769ca
thread:issue:58841,issue #58841: [bug] torch.topk sometimes supports `float16` and sometimes doesn't,stale/low urgency,2,stale/low urgency,1,0,58841,,,,0,d9436e37e1456f8b528a865e463af891a5d763724823bfd03639b72f32862233
thread:issue:58846,issue #58846: Simplification of pruned models,stale/low urgency,2,stale/low urgency,1,0,58846,,,,0,2a4d5088e0be8d73f749f80b97ed5c2eb4608b266590ccfe76d3010b32c8d6fe
thread:issue:58847,issue #58847: Dataloader Rerunning with num_workers=0 may give better error trace,stale/low urgency,2,stale/low urgency,1,0,58847,,,,0,91be40df633fca01d4ca72d0c2c2702967e15edde4ad462b536119309177e06c
thread:issue:58876,issue #58876: [deploy] Enable `torch.distributed.rpc` Python bindings in deploy,stale/low urgency,2,stale/low urgency,1,0,58876,,,,0,913f957a3540b95b5d48f54ffa8421c4eda3aaa14d75dfffad64c22298012005
thread:issue:58930,issue #58930: [local lint] Shellcheck quicklint file filter is incorrect,stale/low urgency,2,stale/low urgency,1,0,58930,,,,0,8c5b855b136c29b6d5eafc54a047326c24a40c1c708fe494dff502bdcaa504cd
thread:issue:58944,issue #58944: Use PYTHONHASHSEED during pytorch build to avoid nondeterminism,stale/low urgency,2,stale/low urgency,1,0,58944,,,,0,84a4583ddc756119071e5faea7369ae646a0e2a16d42a5e2a6ab27f34a81b6d7
thread:issue:58969,issue #58969: Improper use of quantization API for MHA should fail fast,stale/low urgency,2,stale/low urgency,1,0,58969,,,,0,235db2a6c6c20ad0e740616db74e1089705cce377235f8c1f6218f7a79fb9b67
thread:issue:58973,issue #58973: First-class Java API,stale/low urgency,2,stale/low urgency,1,0,58973,,,,0,aa287358d6a7558bee9e016cbeff34a61d0b69bf08fe260dbf8ae0bc5c4245d8
thread:issue:58979,issue #58979: gradient layout contract,stale/low urgency,2,stale/low urgency,1,0,58979,,,,0,350599e32375f3169ed849d4a4efde8254300c7163ce48076fa422e0cf279410
thread:issue:58980,issue #58980: libtorch conflict with cxxopts and cause memery leak,stale/low urgency,2,stale/low urgency,1,0,58980,,,,0,c543f502dafd27d9f072c28bc00f51241f2bb29304e54def9562b6d89a076088
thread:issue:5900,"issue #5900: Build Fails on Gentoo with CUDA 9.1, GCC 6.4, Python 3.5",stale/low urgency,2,stale/low urgency,1,0,5900,,,,0,fd91d2bb98ad8f6d8470fb98082950aef46a14056f45c3430a628683c1fb859d
thread:issue:59071,issue #59071: test_bottleneck_cuda fails without error message,stale/low urgency,2,stale/low urgency,1,0,59071,,,,0,6873529ee910abeb8057abe0e641d56acc4199ef881f3f2a3ae7d8afdfc5f824
thread:issue:59074,issue #59074: c++ use pybind11 to import torch free(): invalid pointer,stale/low urgency,2,stale/low urgency,1,0,59074,,,,0,4ca0d9144739bc391308138859abd8d96310a4cc27be731d854e2da978b27db1
thread:issue:5912,issue #5912: Conv-RNN combination slow in backward pass,stale/low urgency,2,stale/low urgency,1,0,5912,,,,0,7d9ca46963eadeb25ad03414fd41c539e806ae67d3f3c7bbbf862eaed94c6d70
thread:issue:59137,issue #59137: [autograd] `fill_` gradgradcheck raises RuntimeError,stale/low urgency,2,stale/low urgency,1,0,59137,,,,0,937a22524eba87154492412ebcacfa74fbb74e5765701ada60152e057bad68e0
thread:issue:59140,issue #59140: No index type check of torch.tensor in array manner,stale/low urgency,2,stale/low urgency,1,0,59140,,,,0,92bac71a8f6d0bedfb3e624c3eb9d108aca30cb2dd168fd23398d909c5fa59d8
thread:issue:59185,issue #59185: Add supports_nnc metadata to OpInfos,stale/low urgency,2,stale/low urgency,1,0,59185,,,,0,73cadfdb5ff16666a83ebc565bf4234d89e681ec29bdb44022333f19abf2743f
thread:issue:59216,issue #59216: Add a BlendLoss class in torch.nn and a blend_loss function in torch.nn.functional to allow blending of a list of loss functions,stale/low urgency,2,stale/low urgency,1,0,59216,,,,0,424db294af0e2ca08f285d237a50abd3aa6db029f79c7b67afe607c96e2d84fd
thread:issue:5924,issue #5924: Install doesn't work with spaces in directory,stale/low urgency,2,stale/low urgency,1,0,5924,,,,0,c6a841b3830a647b99836ce10048a45bf3e942d7d55de28d45f1e2e75e66d260
thread:issue:59243,issue #59243: Can't install pytorch =1.7 with python 3.8.5 on Raspberry Pi,stale/low urgency,2,stale/low urgency,1,0,59243,,,,0,533c98ec3912794263e73293a4a4e2ecdacf5ee91b6745b29c614c1dce48fbf2
thread:issue:59247,issue #59247: Support complex numbers in `at::nan_to_num`.,stale/low urgency,2,stale/low urgency,1,0,59247,,,,0,e9d1ddc815b61a430e5d14a16791a48f5f5648c1e4f6805571466cece6d644ca
thread:issue:59256,issue #59256: ProcessGroupGloo creation crashes when world_size > 150,stale/low urgency,2,stale/low urgency,1,0,59256,,,,0,31de335afda2d8e3b4a9d7d440f97c29e7badf34dec93c1d402a82902d465111
thread:issue:59262,"issue #59262: Possibly alignment issues on NEON vectorized ops, Jetson platforms",stale/low urgency,2,stale/low urgency,1,0,59262,,,,0,a6a338ad09454f8f24a6a429696ac5bea8fe2de2d7c3f0323b010da9263417ba
thread:issue:59263,issue #59263: Ubuntu 20.04 GeForce RTX 3070 / NVIDIA-SMI 460.73.01 Driver Version: 460.73.01 / GPU Visibility Issue,stale/low urgency,2,stale/low urgency,1,0,59263,,,,0,be8cca1e5c827b7d6e3505547d9318e39726be403f9216a1a825bc2e5d95c890
thread:issue:59309,issue #59309: CUDA memory error when using torch.device('cpu'),stale/low urgency,2,stale/low urgency,1,0,59309,,,,0,67b619721c1261663908d1bc1c628f67e2cb1ad6591eab0e95c36a27346f70f3
thread:issue:59356,issue #59356: pytorch build failure on arm64,stale/low urgency,2,stale/low urgency,1,0,59356,,,,0,e1f0d9796efc20a0b489375c4ce4c4837384e7c98ed395d74ff8d128f28ea2d2
thread:issue:59386,issue #59386: Energy-efficiency benchmarks?,stale/low urgency,2,stale/low urgency,1,0,59386,,,,0,eba8837286e8775162cc2c17979453f733efc33d335962c4dbddac63841d710d
thread:issue:59418,issue #59418: Libtorch segfault in packed GRU evaluation with cuda batch_sizes,stale/low urgency,2,stale/low urgency,1,0,59418,,,,0,797d4139663dfd5572f3cd7e1d93af96b2fdd3ff6a6e7792609ab4f7f29c2a06
thread:issue:59437,issue #59437: pytorch/manylinux-cuda102 support for aarch64,stale/low urgency,2,stale/low urgency,1,0,59437,,,,0,a5d10126585916ac80e190889b2724e27108a1d85a60037c5f464d142f2c2a47
thread:issue:59438,"issue #59438: DataLoader worker (pid(s) 18056, 20540, 4512) exited unexpectedly",stale/low urgency,2,stale/low urgency,1,0,59438,,,,0,1f2e5c3d59280a07452a0a7747b25995f5dfbe54d1254c625effca40505c61f0
thread:issue:59439,issue #59439: NaN values on torch.nn.functional.conv2d (aarch64),stale/low urgency,2,stale/low urgency,1,0,59439,,,,0,c176c8957f9471b9bb8ee2ac40c118a2daa002da2cde9ed56df40ba751bbf37d
thread:issue:59457,issue #59457: Error compiling PyTorch: `/usr/bin/ld: cannot find -lmagma /usr/bin/ld: cannot find -lnvToolsExt`,stale/low urgency,2,stale/low urgency,1,0,59457,,,,0,f0bd5f2fcfd365408854eeafe3bc1badf62121dec52f1011405162ac4138781e
thread:issue:59493,issue #59493: ctc loss bug,stale/low urgency,2,stale/low urgency,1,0,59493,,,,0,10b543f8ef10bdefcdda0b7ca98c7c804cf47a1a94d396f79ccb29ba432d97fb
thread:issue:59515,issue #59515: Conv1d with large batch size and half precision in cuda returns incorrect result,stale/low urgency,2,stale/low urgency,1,0,59515,,,,0,f71a4cab6072b2ec0d9307e4d2735f78412add58dbd2b99ad3ac94d2591ff5d3
thread:issue:59531,issue #59531: Failed to enable libuv,stale/low urgency,2,stale/low urgency,1,0,59531,,,,0,ab408a2d226501176586aa94b091246bd274a6ffb10069f7f523fb414370e247
thread:issue:59542,issue #59542: Build Pytorch with Aten?,stale/low urgency,2,stale/low urgency,1,0,59542,,,,0,1a37cec78564d873f04624540ec0e4cdb0d33a97639848b54a8224486c9e07fe
thread:issue:59546,issue #59546: Multiprocessing model evaluation gets stuck,stale/low urgency,2,stale/low urgency,1,0,59546,,,,0,111a5b69e8b2e2bb46b64855b08eb4a4cb3202adacb9e96167edf69b0a8e5dce
thread:issue:59575,issue #59575: [clang-tidy] check for iteration over unordered data structures,stale/low urgency,2,stale/low urgency,1,0,59575,,,,0,bc65a812c5dc907517cca9b3b677b551151cd5701c5b97f605efdf8234fee4b9
thread:issue:59592,issue #59592: Build failed with TBB,stale/low urgency,2,stale/low urgency,1,0,59592,,,,0,0194d45bd8c08376aec161a14ce4905507ba295c034048ad059227edb4c06ad1
thread:issue:59617,issue #59617: torch.split() infer -1 entry from the other split sizes,stale/low urgency,2,stale/low urgency,1,0,59617,,,,0,4175392a495e75eeb33dd23d942d2e2a195b916f61546103d0f12a2544b5e76a
thread:issue:59627,issue #59627: Example code for is_storage missing,stale/low urgency,2,stale/low urgency,1,0,59627,,,,0,21b3a7d4cc229c47225ef98043cccae4e061a86821a5de8dc728bb5523fee089
thread:issue:59692,"issue #59692: cuda streams run sequentially, expected to run parallel",stale/low urgency,2,stale/low urgency,1,0,59692,,,,0,a3560eba18d4c43c5cf5fada88cafb703e39995fe6bb451d32472f0d76ce8b65
thread:issue:59730,issue #59730: Quantized conv2d with dilation and groups much slower than float32,stale/low urgency,2,stale/low urgency,1,0,59730,,,,0,1a86af2ae55f503870471393783b1494c9fb24e511d365171192c37a5e7ca82b
thread:issue:59743,issue #59743: m.fallback(torch::CppFunction::makeFromBoxedFunction<&my_fallback>) gives bad error message,stale/low urgency,2,stale/low urgency,1,0,59743,,,,0,ff1d251865be0baca0b1218d47ef8bac00f2b855abc37af33e864278c6b506f2
thread:issue:59778,issue #59778: CUDA Memory Error: PyTorch doen't allocate memory even though it's available,stale/low urgency,2,stale/low urgency,1,0,59778,,,,0,a7167e9da771ca5276d2e7098002180bf7c1c73b38d0388aa0609d910d57d8fe
thread:issue:59783,issue #59783: [Feature Request] Deterministic implementation for AdaptiveMaxpool1d.,stale/low urgency,2,stale/low urgency,1,0,59783,,,,0,8d59f36209f80dbba30098e501c04b314d3d68b74b8ea39c0887322dddb2f114
thread:issue:59822,issue #59822: TransformedDistribution's log_prob gradient is inconsistent w.r.t. cache_size,stale/low urgency,2,stale/low urgency,1,0,59822,,,,0,7b6176705894d102d868070f35304d069ce5d85b6615620367b16715d7287d2a
thread:issue:59854,issue #59854: Misreported linting messages,stale/low urgency,2,stale/low urgency,1,0,59854,,,,0,9892a4e139d2c69c6f3ba43e52a3b43d72251cf6f7123c12d7d27b7c1140af5f
thread:issue:59875,issue #59875: [ROCm] test_gather_stress_cuda is flaky,stale/low urgency,2,stale/low urgency,1,0,59875,,,,0,48a242a552549ebfb02f18aed7b99b844d592ade82ea7246f3c13b1375266af5
thread:issue:59896,issue #59896: Autograd engine current graph_task should be in the TLSState,stale/low urgency,2,stale/low urgency,1,0,59896,,,,0,24fb57c46c4b5505cd2ccb82d8773c0fbb97768b16a3f8be786bb05a47b0fc0d
thread:issue:59925,issue #59925: Implicit module registration permits silent programming errors in torch.nn.Module,stale/low urgency,2,stale/low urgency,1,0,59925,,,,0,bf240f7825a78ef22fe9a685e0fb8de91631fa117837b7ca98e1b6f5e1caf847
thread:issue:59935,issue #59935: Error /usr/local/lib/libopenblas.so: error adding symbols: File in wrong format while building pytorch for ppc64le,stale/low urgency,2,stale/low urgency,1,0,59935,,,,0,85a2b3c74a69c564c9d8efee7345990f4d850bdcca9cd2c3d9bcfb7788a6d0d5
thread:issue:59936,issue #59936: Can't perform any operation on Vulkan device - macOS M1,stale/low urgency,2,stale/low urgency,1,0,59936,,,,0,fa5867171212ff7d95f03b3975e51a92b90f97bef7fe8f192974b4fb67c77bb0
thread:issue:60008,issue #60008: Dynamic quantized LSTM does not support output projection now.,stale/low urgency,2,stale/low urgency,1,0,60008,,,,0,adff7ab0cb641073fab6efd3e33b764bc5d08fd83184b2eeaeb2716642dd5e05
thread:issue:60028,issue #60028: RReLU and PReLU do not propagate input strides to outputs,stale/low urgency,2,stale/low urgency,1,0,60028,,,,0,cccc6eaf3a60bfd59cc7bc2fb0908f042ae41fcbbe03a081f3abf09cc646f747
thread:issue:60063,issue #60063: Replace native NHWC BN kernel,stale/low urgency,2,stale/low urgency,1,0,60063,,,,0,c49882188996dd231a33085c0f6de1f624b76e555564e06f766ec2383cf2cea9
thread:issue:60069,issue #60069: Symbolic trace with *args,stale/low urgency,2,stale/low urgency,1,0,60069,,,,0,906a0bdf8c99523c7abb32ab5b3c7dd9f4944b5d7508e7f4f8b1852d09e3b2df
thread:issue:60083,issue #60083: caffe_translator.py Unable to convert Power Layer,stale/low urgency,2,stale/low urgency,1,0,60083,,,,0,3ae27b6c11874d86757bec82372130cdc2dcd26a16f2ac25907a9feb600f3066
thread:issue:60092,issue #60092: `memory_format` flag for `tensor.new_empty()`.,stale/low urgency,2,stale/low urgency,1,0,60092,,,,0,2002ee15fc6c48f1b70c56da852bda1654b7444c8702de4bc4f96e744be85493
thread:issue:60096,issue #60096: Tutorials that require `pip install` not loading correctly in Colab,stale/low urgency,2,stale/low urgency,1,0,60096,,,,0,5a5da8fc35ff1505bc95974113c9ef83a1dbc7547ad39f1bb7aadee08f111167
thread:issue:60102,issue #60102: Allow packaging of classes defined in a notebook,stale/low urgency,2,stale/low urgency,1,0,60102,,,,0,a94a645aba10f20664c9800a724ff80e4f15acc9ca93340ef80d2af95a3ef4b0
thread:issue:60146,issue #60146: [torch.profiler] enhancements + corrections,stale/low urgency,2,stale/low urgency,1,0,60146,,,,0,68052958422b34287d33928ae2f70f2ff732e856a256ee27fabdaa9a863cdce3
thread:issue:60156,issue #60156: PEP-585 type annotations not supported for class-level annotations,stale/low urgency,2,stale/low urgency,1,0,60156,,,,0,39001714e21946677a0faa38e7f130878c3bc13af63db1f4e0ba0a269d5cf3ae
thread:issue:60159,issue #60159: Some ops throw runtime errors when called with complex tensors that require_grad but work when requires_grad=False,stale/low urgency,2,stale/low urgency,1,0,60159,,,,0,1a9a98d449e9e167c4312f4de7910939451d7faa57c2cae8ab8bd55b8974885f
thread:issue:60164,issue #60164: Training goes wrong with amp and no_grad since pytorch 1.8,stale/low urgency,2,stale/low urgency,1,0,60164,,,,0,15a5f9ab72249b38c0b73df01ab5b07ed39590ccb43dbd70b346d0331dea35fb
thread:issue:60165,issue #60165: torch.load non backwards compatible on Transformer between 1.8.1 and 1.9.0,stale/low urgency,2,stale/low urgency,1,0,60165,,,,0,df803cec4fa7a85a1fdb41cb7f8437938a1ebcaeb0ebae29e54f56e47f694cd9
thread:issue:60210,issue #60210: Quantized model using `boolean_dispatch` not picklable,stale/low urgency,2,stale/low urgency,1,0,60210,,,,0,4c2e2bc66cca750eed81eac744dc708aba2d4ac533f162011914c87136de5475
thread:issue:60264,issue #60264: Subtensor operations like `tril/triu` to have an option to respect the strides of the input,stale/low urgency,2,stale/low urgency,1,0,60264,,,,0,7f88a5151123585ed6927cc70c949eb9469ef9ec7443d5b560a28554d5b71a02
thread:issue:60277,issue #60277: Sparse CSR tensor should not accept equal column indices in the same row,stale/low urgency,2,stale/low urgency,1,0,60277,,,,0,e365f1e0f1ab2bc11c1e12f0d107b39944a474aed140c7d70ecee2e05e5bb841
thread:issue:60288,issue #60288: provide hashsum for downloads,stale/low urgency,2,stale/low urgency,1,0,60288,,,,0,0da8f30043658a6ea1ad81108cb6ab6e5f1868d46704ffd8a9a811aca97f4170
thread:issue:60306,issue #60306: Divergent code is needed to record usage streams on different TensorImpl types,stale/low urgency,2,stale/low urgency,1,0,60306,,,,0,b1db75d1959520520323f1f88fe6398054eaf0ac66a1375ab13091203d28a407
thread:issue:60343,issue #60343: What changes we need to make in metrics calculation and visualization part when we use Distributed Data Parallel for distributed training,stale/low urgency,2,stale/low urgency,1,0,60343,,,,0,9c07ff8f1524ffa2447e34b9d36a4d0d92a4db98ae06feaaebba814e1a9acfae
thread:issue:60372,issue #60372: Docs for `Installing C++ Distributions of PyTorch` needs update,stale/low urgency,2,stale/low urgency,1,0,60372,,,,0,906efd95481ee695b9b79d7a938b5e3f06ecd9501e404f99fa47f60797399fdf
thread:issue:60381,issue #60381: broken source link / missing anchors,stale/low urgency,2,stale/low urgency,1,0,60381,,,,0,ce0c08f16b9f69c08347ca7201dce1dacd38fbbdaef0f8a42124da645375917e
thread:issue:60400,issue #60400: Using inbuilt function name `input` as variable names in examples,stale/low urgency,2,stale/low urgency,1,0,60400,,,,0,de5634c74810e0a4ceac28ee0adfd33f8d69584f12fd664ce4417d61572a2eed
thread:issue:60401,issue #60401: Need workaround to support multiprocess CUDA tensor sharing on Jetson Platforms,stale/low urgency,2,stale/low urgency,1,0,60401,,,,0,aed557490d1555b54f09a6c5fb29a78eb8f74af632bab4b17d73675594bb3e04
thread:issue:60426,"issue #60426: Output of non-inplace, non-CompositeImplicitAutograd op has TensorImpl > 1 or StorageImpl use_count != 1",stale/low urgency,2,stale/low urgency,1,0,60426,,,,0,32883e53e58862e800f4a94e03025f538069f6fe17b93ce1dd654d9389f936cc
thread:issue:60431,issue #60431: Implement simple view methods for TensorAccessor/PackedTensorAccessor,stale/low urgency,2,stale/low urgency,1,0,60431,,,,0,8837a7e50d6d99832051f7abf1fdd1392dd57e79bc4743ef57eba84cfa850eca
thread:issue:60448,issue #60448: Enable large file support,stale/low urgency,2,stale/low urgency,1,0,60448,,,,0,07a0dda2561dec465d7877ba320bf5a62fc057ea07396e41f51859bcee854961
thread:issue:60459,issue #60459: [Mkldnn] has_bf16 check only works on Linux for tests,stale/low urgency,2,stale/low urgency,1,0,60459,,,,0,f9947c5399c8234c7923aa044736850408f8908171a5d03dc6b8feebd429b050
thread:issue:60520,issue #60520: [docs] torch.log_softmax is undocumented,stale/low urgency,2,stale/low urgency,1,0,60520,,,,0,86c0835054e6d7888b89b51691e1176c722654722b4c38b5ada4f30961d9084e
thread:issue:60548,issue #60548: Ensure warnings relate to user code with stacklevel,stale/low urgency,2,stale/low urgency,1,0,60548,,,,0,f64e9233e32b348d6c2d4ca3c5284740ed0c2edb22d7df9da0d7852ceb3e9118
thread:issue:60561,issue #60561: nn.CrossMapLRN2d is missing from docs,stale/low urgency,2,stale/low urgency,1,0,60561,,,,0,b526ad48b99a1bd8cc45baefc41da943f9daa8f512348183f953248e5e795b07
thread:issue:60674,issue #60674: Turn on -Winfinite-recursion in OSS CI,stale/low urgency,2,stale/low urgency,1,0,60674,,,,0,ddef0cec4d09e8739f4370721b6183191a8f5593cb20198be154b01142c142a8
thread:issue:6071,issue #6071: Repeated 'python setup.py install' with clang leads to -lcpuinfo not found,stale/low urgency,2,stale/low urgency,1,0,6071,,,,0,b72107f2b33aef6a236c6e58744951668c4426e28957606fb09870362206f915
thread:issue:60724,issue #60724: bilinear interpolate is very slow under mixed precision training mode.,stale/low urgency,2,stale/low urgency,1,0,60724,,,,0,97ad0bf12ee7755795948ab7677a64ed5f5fc643f9729824eb5d73413d08dc7a
thread:issue:60748,"issue #60748: Record file/line number when creating test data, and then report it in backtraces associated with this data",stale/low urgency,2,stale/low urgency,1,0,60748,,,,0,5739371e1b385b0abc58d17e851141f650d961b35844f2e60d982875bc7d6b8c
thread:issue:60766,"issue #60766: _foreach_maximum returns ""incompatible type"" mypy error unexpectedly",stale/low urgency,2,stale/low urgency,1,0,60766,,,,0,80fcd1c561f743558280a1ab674d612cf12237a097c79b8d0462bca65b62356d
thread:issue:60796,issue #60796: AccumulateGrad short circuits before calling variable hooks when the incoming grad is None,stale/low urgency,2,stale/low urgency,1,0,60796,,,,0,665913e6b18cc0ccba13dac07afdf3ff0cda8e11e1846c91e325b9d06f716d81
thread:issue:60802,issue #60802: Bug in `pytorch/benchmarks/tensorexpr/pt_engine.py`,stale/low urgency,2,stale/low urgency,1,0,60802,,,,0,be70fb29e9a6c83ef24f10be6dcb2451502268f7fc2cae08c86b7a1fe749166a
thread:issue:60825,issue #60825: Where OpInfo doesn't handle cases where one of the inputs is a scalar,stale/low urgency,2,stale/low urgency,1,0,60825,,,,0,8493603e030006857681083f26b49afdac4b9de35c7d8da342ef75b6274eb065
thread:issue:6083,issue #6083: [feature request] DC-ASGD (Delay Compensated Asynchronous Stochastic Gradient Descent),stale/low urgency,2,stale/low urgency,1,0,6083,,,,0,1b8d5f2a358bd063006395b372c850f723e0d9496a383d9693b50b55f9544122
thread:issue:60834,issue #60834: complie error when i use python setup.py build on arm computer,stale/low urgency,2,stale/low urgency,1,0,60834,,,,0,0cbf0b6c32c4a3fb768353d4c7aa90969b071dc19e09b4f7819bc2f2816543c7
thread:issue:60843,issue #60843: PyTorch's last 1 year update became unfriendly due to the lack of examples,stale/low urgency,2,stale/low urgency,1,0,60843,,,,0,826d4cef717ab5afa9df3eb69726489eb209f39c6ce3521f7b684227a859c9f7
thread:issue:60848,issue #60848: Fail to build with gcc11,stale/low urgency,2,stale/low urgency,1,0,60848,,,,0,2ecd27ad4c4e6c16ca23b51999b5430868b5c247e6a6060b822510645b8ef9dc
thread:issue:60911,issue #60911: [elastic launcher] redirects/tee support for global rank,stale/low urgency,2,stale/low urgency,1,0,60911,,,,0,2d18d980aed8042fd8d1bc006d63ff09651a9bb3d2bd2525809227c8f1ced351
thread:issue:60916,issue #60916: Scipy 1.7.0 may cause some test failures,stale/low urgency,2,stale/low urgency,1,0,60916,,,,0,fac88277db5220df24ddf2439b3caafe4fa793345499be3d83751ab32b73d192
thread:issue:60924,issue #60924: [rfc][local lint] Create lint runner as a setup.py target,stale/low urgency,2,stale/low urgency,1,0,60924,,,,0,dcdd9ddac4751318e9b8de27489ebbb1694046f05db6860ab66c40a88155cbb1
thread:issue:60932,issue #60932: Is Python version of Docker image on DockerHub downgraded?,stale/low urgency,2,stale/low urgency,1,0,60932,,,,0,a1b5b7310a6146f68be25c2948a8c4315237173c839e75054df82eedc1b8cdfb
thread:issue:60933,issue #60933: jit.load error when changed Folder Name,stale/low urgency,2,stale/low urgency,1,0,60933,,,,0,14388684856a633cb43569347b6f7b7ade4ea47fff446bd259d63c06d7c2a83d
thread:issue:60939,issue #60939: Functional multi_head_attention_forward softmax get nan for fp16 mode,stale/low urgency,2,stale/low urgency,1,0,60939,,,,0,6d028209b4680d3ad07d085f5da696f742718734ee1fad5be3d65e22e1dc8cdb
thread:issue:60941,issue #60941: Incorrect dtype cast for binary ops w/ mixed ComplexFloat + Double operands,stale/low urgency,2,stale/low urgency,1,0,60941,,,,0,825ff223530973e6bbec744b87b29635b67f03fbb96f474e5c0931ba852ba37c
thread:issue:60950,issue #60950: Singular build setting CLANG_CXX_LANGUAGE_STANDARD has different values,stale/low urgency,2,stale/low urgency,1,0,60950,,,,0,a3a7f128c9a6abc962a5d8c258cad40d88b7fa442614f2e07eac090fcee58f37
thread:issue:60956,issue #60956: Naming convention for variants of operations that never short circuit,stale/low urgency,2,stale/low urgency,1,0,60956,,,,0,2d5618f7e283dee7f1e16da28feb6e53d253f7d83ffa391815a3a36374697059
thread:issue:60977,issue #60977: [package] pattern usage can potentially create broken packages,stale/low urgency,2,stale/low urgency,1,0,60977,,,,0,4e35dc2379674140372f1b0189f3ec4a0bfdddd8abc457fce7d546fabf0a8e09
thread:issue:60991,issue #60991: [NNC] APIs needed to bridge scheduling gap in reductions,stale/low urgency,2,stale/low urgency,1,0,60991,,,,0,f7e053367ecfe4edb4ae5e9190c1f0251bab3589ffc3a96a997a7c297be675dc
thread:issue:60997,issue #60997: Dispatcher doesn't handle ops with an empty list of tensors (e.g. `torch.cat()`),stale/low urgency,2,stale/low urgency,1,0,60997,,,,0,b91d1c3b1c7cbb28bf982f919155e85b35ebba635fa918cb1f6cd67adc5ef23f
thread:issue:60999,issue #60999: Merge seemethere/upload-artifact-s3 and driazati/upload-artifact-s3,stale/low urgency,2,stale/low urgency,1,0,60999,,,,0,93e142346835efd98fe7ec10d30b74e04a5c1406d80668bcc4c1039f079a346f
thread:issue:61030,issue #61030: [Docker] Incompatible torchvision for 1.9.0-cuda10.2-cudnn7-runtime tag.,stale/low urgency,2,stale/low urgency,1,0,61030,,,,0,e795d3b8e1c8c83767acad45cb77d24e77fa7dc168b682870e23a3b15a194161
thread:issue:61046,issue #61046: Audit use of __ARM_NEON__ define,stale/low urgency,2,stale/low urgency,1,0,61046,,,,0,99fffabc6aec5ed61b9851f37c9803dc357700efa890bc027d591791273c883f
thread:issue:6107,"issue #6107: Well documented, safe method to deserialize model parameters from untrusted sources",stale/low urgency,2,stale/low urgency,1,0,6107,,,,0,d05a8682c0b1d055cc199eeeae26da1a7181fefd9bd0ca7acdb9ac6dbd958ef0
thread:issue:61110,issue #61110: Incorrect exponential calculation on Jetson devices with float32 dtype,stale/low urgency,2,stale/low urgency,1,0,61110,,,,0,9c83f519e0585f764ce997f6f05d40ab8b1550b696ceb5fac3e708e949e144c1
thread:issue:61117,issue #61117: flatten renames in place,stale/low urgency,2,stale/low urgency,1,0,61117,,,,0,7078bec8b88812358cf1aca47fbf307862e925b905bc7d3139dee107fe56797e
thread:issue:61122,issue #61122: Make copy_ use dispatcher,stale/low urgency,2,stale/low urgency,1,0,61122,,,,0,d5928896cdc004af8518e57444a5f3622c27bccabec624ca1cf4d5bb8e99e116
thread:issue:61136,issue #61136: rename(**{}) should throw no error,stale/low urgency,2,stale/low urgency,1,0,61136,,,,0,a569e9b118d928cbe384ccf820a183131ed1620589444505c3be96c07a753578
thread:issue:61149,issue #61149: `pad(mode='reflect')` beyond `input.size`,stale/low urgency,2,stale/low urgency,1,0,61149,,,,0,10f496356e4a1072e5cb878f0156c1cbd2d4c1423fc9a24ed15f3cc255168b53
thread:issue:61167,issue #61167: Torch .pth model export to ONNX returned empty graph,stale/low urgency,2,stale/low urgency,1,0,61167,,,,0,727f68eb95c521b51bc804358d007602e425d43bd03eb4e0795c101cf7eb7be1
thread:issue:61173,issue #61173: Increased memory usage with AMP,stale/low urgency,2,stale/low urgency,1,0,61173,,,,0,4d38a527dc68f22c08509d52ee467c3da116491bad16fb77366db021731fa99f
thread:issue:61182,"issue #61182: Feature Request: torch.functional.interpolate to quietly ignore ""align_corners"" when mode is set to ""nearest""",stale/low urgency,2,stale/low urgency,1,0,61182,,,,0,10784be9601093d11741901692f719d06b3415afc09fc5374a825248088dd780
thread:issue:61186,issue #61186: possible SummaryWriter memory leak,stale/low urgency,2,stale/low urgency,1,0,61186,,,,0,10f7778b804ca6627f327add088b36126e2aaa0069922c63f48a7111568738b3
thread:issue:61222,issue #61222: The syncbatchnorm can not be employed in the MODEL that the gradients including in the loss function,stale/low urgency,2,stale/low urgency,1,0,61222,,,,0,096b66cc42dff054ae3b04c671cc9f2f71eae59ad81fc668c37a72c166e2c2ae
thread:issue:61233,issue #61233: [typing] ModuleDict ctor has wrong signature,stale/low urgency,2,stale/low urgency,1,0,61233,,,,0,4abb92139ed13d41ca901e4798c39da12dde32226fd1e30cefafef43235fa9b8
thread:issue:61244,issue #61244: error: invalid conversion in vec256_bfloat16.h,stale/low urgency,2,stale/low urgency,1,0,61244,,,,0,96f8ed6b573b4fa8b4404f0795be0bda91f061b05e8237eb95a417b6eba80b05
thread:issue:61246,"issue #61246: torch.quantization.fx.fuser forget to pass ""additional_fuser_method_mapping"" to fuse method",stale/low urgency,2,stale/low urgency,1,0,61246,,,,0,451c56fc34f86d26816d28f378089f0e4d5140e4bf0c51dcd1abb020b536ddf7
thread:issue:61261,issue #61261: libtorch_cpu.so is getting too large.,stale/low urgency,2,stale/low urgency,1,0,61261,,,,0,3780a7dcbf9683eaf085822b5e11b071cc2d3ccf5eade8f897bf6efb2494ca9d
thread:issue:61281,issue #61281: DataLoader with IterativeDataset throws an error when providing a BatchSampler,stale/low urgency,2,stale/low urgency,1,0,61281,,,,0,9f25722c54b1bc03217eddc1bcf451eca0a19981e085088ba1796d8228e75251
thread:issue:61289,issue #61289: Zero seed in DistributedSampler,stale/low urgency,2,stale/low urgency,1,0,61289,,,,0,0a61a6baba5eb277a9c276774a976f3a11e002e20fee0bf7fb590457537df511
thread:issue:61292,issue #61292: torch.linspace tensor support,stale/low urgency,2,stale/low urgency,1,0,61292,,,,0,f6c81cf26cf8d476a697dc4c706cdbfdc7ab672b8a7454017458e3a11cc467c1
thread:issue:61309,issue #61309: The class weights implementation for loss functions is incorrect,stale/low urgency,2,stale/low urgency,1,0,61309,,,,0,d747af3a60aab7b94cad43c3bda991a7d853e6d919beb10713704628f97c997c
thread:issue:6135,issue #6135: Test suite should test implementations,stale/low urgency,2,stale/low urgency,1,0,6135,,,,0,5369ab4e513fba3e909ff5a859a73ff0f8a19cecc2bc0352f0c172a13007a847
thread:issue:6138,issue #6138: [feature request] SIGNUM an optimizer that takes the sign of gradient or momentum.,stale/low urgency,2,stale/low urgency,1,0,6138,,,,0,c99d11b5c09e5598aef56c50b0cc3bf174a61dbdf8ef5fce32704874c8d8d8a2
thread:issue:61384,issue #61384: Merge fork of upload-artifact-s3 to a single commit,stale/low urgency,2,stale/low urgency,1,0,61384,,,,0,aaa88db9bdd2d9c2a2633c4cc641fbe9bee3fd6c03abdd1b39e20c102e012ace
thread:issue:61410,issue #61410: requires grad get lost during transform.,stale/low urgency,2,stale/low urgency,1,0,61410,,,,0,c6d78aeecedab42d7c825c04e4e8fb4208ad74310fdd0f14159eee9c1bd51b97
thread:issue:61452,issue #61452: New module: Split log softmax with loss,stale/low urgency,2,stale/low urgency,1,0,61452,,,,0,b8a2a067c2053346394e83b138ced7ffaed38b57b343e1e4eaf28f8bb76eecbb
thread:issue:61457,issue #61457: using DistributedSampler occur RuntimeError: Expected a 'cuda:0' generator device but found 'cpu',stale/low urgency,2,stale/low urgency,1,0,61457,,,,0,e45a0b4d21d6e0acb327a9f89c5c732ad1350bc2f7efd855c9ade469a3ef82f9
thread:issue:61468,issue #61468: Version is set to 1.8.0 in the torch tag v1.8.1,stale/low urgency,2,stale/low urgency,1,0,61468,,,,0,f861a68d9200747f5b404ff69eab30b17c0264bddc3e69f6a8c4f981c0701d74
thread:issue:61485,issue #61485: CompositeImplicitAutograd ops should not call out= variants of operators,stale/low urgency,2,stale/low urgency,1,0,61485,,,,0,94095e41d4679d5a2f18e6704db89de27b53f99500d3db582911991c1e1b879c
thread:issue:61503,issue #61503: UCI Data Sets,stale/low urgency,2,stale/low urgency,1,0,61503,,,,0,b8979dd3d095f7987ba7d29d64396978e23c7ed7da79879b57c63f64a7079a46
thread:issue:61509,issue #61509: [feature request] Exact euclidean distance transform,stale/low urgency,2,stale/low urgency,1,0,61509,,,,0,ba600e1de50e54502f3b98b9628c5ec3c9ee2301605ea2ae8303e7e73a4b66e6
thread:issue:61510,issue #61510: What is find_package(Torch REQUIRED) doing that a manual include/glob doesnt?,stale/low urgency,2,stale/low urgency,1,0,61510,,,,0,f8ded057dd85571122f9109264948042dac743f62c990706784fbae3896a7638
thread:issue:61513,issue #61513: Adding 'differentiable' section to each function docstring,stale/low urgency,2,stale/low urgency,1,0,61513,,,,0,ae56c13723cf39eb72c4fe01634d263a234d903666d0405a4339253236249e0e
thread:issue:61515,issue #61515: Buggy page with overlapped tabs,stale/low urgency,2,stale/low urgency,1,0,61515,,,,0,50b66f0a777b9e30678810d801cf2c0b272ef94394e3779b1855ae60b596763c
thread:issue:61516,issue #61516: MemoryError with pip Install even with --no-cache-dir option,stale/low urgency,2,stale/low urgency,1,0,61516,,,,0,64c99f76e09e56a77950b57535480d9fb90b589823e57818e8ce745e654fa093
thread:issue:61525,"issue #61525: I have graphic card of P4, but torch.cuda.is_available() always shows false",stale/low urgency,2,stale/low urgency,1,0,61525,,,,0,8c4117cb735dbcabcdfc3523c344a1b9ec6995ddf7163693e5a67c15921e389f
thread:issue:61538,issue #61538: The use of epsilon in torch.nn.functional.pairwise_distance leads to incorrect minima of the distances,stale/low urgency,2,stale/low urgency,1,0,61538,,,,0,4ee6246870dbcf2c92a8a7a06522419b925e58f0e2f3b727f700bf2f1ac03cef
thread:issue:61544,issue #61544: `__torch_function__` always records default kwargs as kwargs,stale/low urgency,2,stale/low urgency,1,0,61544,,,,0,393be566c801f72f3c404bd0571de08e18a4ffff3d7f1178e999f8601e7b370f
thread:issue:61574,issue #61574: [nn] Passing dtype to `_stacklevel` argument in `log_softmax` silently works,stale/low urgency,2,stale/low urgency,1,0,61574,,,,0,456d3c795fc98335c5ad02a554019b25916a8d7afb724d45dd8ab46d334173c1
thread:issue:61591,issue #61591: Add `map_location` arg for c10d collective comms,stale/low urgency,2,stale/low urgency,1,0,61591,,,,0,621d1eca74534ee93f69cde1146e1fb0a8d0f468bd7723da9161f81132c1a677
thread:issue:61614,issue #61614: Possibly flaky test: TestMultiprocessing.test_fd_sharing,stale/low urgency,2,stale/low urgency,1,0,61614,,,,0,350ff127c50059047c5e1fa24fb54dd80f3c97a48de2e8ebb2a3ccdaa6b80a8e
thread:issue:61619,"issue #61619: When 'trapezoid' is called with an empty tensor input, it does not produce an output with requires_grad",stale/low urgency,2,stale/low urgency,1,0,61619,,,,0,c178bb797fcdad4c6367260dd065f937184f5ddf2d42f29ce32abebc23779676
thread:issue:61636,issue #61636: Enhance Distributed `get_future()` docs,stale/low urgency,2,stale/low urgency,1,0,61636,,,,0,7f4002cdf30bfc5b263f3a5d20be4b46078aa224d3fedf0486342c08099b1bf2
thread:issue:61645,issue #61645: `channel_shuffle` output is sometimes aliased with its input,stale/low urgency,2,stale/low urgency,1,0,61645,,,,0,628f350da6ac51ff5367fcc3d22a0bae23c0c6f32e5538d4fa4a926e9bd8481c
thread:issue:61649,issue #61649: Move `tensordot` into `torch.linalg`,stale/low urgency,2,stale/low urgency,1,0,61649,,,,0,1f0c5722d8f75bbb28b12b57dd0dafb8c7df4fe16092ef573bd17cf5ad5b42f9
thread:issue:6165,issue #6165: Note about unusual stride situations in dev docs / make it easier to test for this in the library,stale/low urgency,2,stale/low urgency,1,0,6165,,,,0,30106004109ebd589b1a0871eb4d365e64a5a11e6ffe34f6b2336ab80d71e287
thread:issue:61659,issue #61659: Python program locks up when using some innocuous code inside mp.spawn(),stale/low urgency,2,stale/low urgency,1,0,61659,,,,0,eb8ef2660b1d7f83226815498028dc2793b250a92fda8ca97cfb206d57ae0448
thread:issue:61661,issue #61661: Return `Future` instead of `Work` in `notify_join_context()`,stale/low urgency,2,stale/low urgency,1,0,61661,,,,0,d09f7dc5a78933feb908af13fee55e334be1dc3ae589b33a1230f0e8a4c463b8
thread:issue:61662,issue #61662: `super().__init__()` usage for `_Joinable`,stale/low urgency,2,stale/low urgency,1,0,61662,,,,0,6a8bea723740b0b71b64cb18779d171632ba2fec3ce4be483c459daf5a39d674
thread:issue:61665,issue #61665: Default Collate doesn't work for subtypes of ndarray,stale/low urgency,2,stale/low urgency,1,0,61665,,,,0,eb8258ec14d28bf9439b4134dc760b97751c4552ff17e54d54dd8ddaae869288
thread:issue:61695,issue #61695: miss header file while using python setup.py install to install pytorch with anaconda env,stale/low urgency,2,stale/low urgency,1,0,61695,,,,0,3cba1a08d5b6b1936931c77a79a6388325929bf83b5d537214dac5d5d223a2cc
thread:issue:61718,issue #61718: ATen codegen is always re-running for Makefile generator,stale/low urgency,2,stale/low urgency,1,0,61718,,,,0,81a6836b13b53e0e51f2738f1ee3bb70b3c6a5afc0c7e642444b4a496dbdec8a
thread:issue:61727,issue #61727: ```torch.distributions.Categorical``` unintended ```log_prob``` gradient w.r.t ```probs```,stale/low urgency,2,stale/low urgency,1,0,61727,,,,0,97e1170fa87568bfd4bda98069125cc40f8b016543b3546c47a104898e501797
thread:issue:61730,"issue #61730: Can't specify compiler, leading to SUPPORT_GLIBCXX_USE_C99 failed",stale/low urgency,2,stale/low urgency,1,0,61730,,,,0,2f1be3bed9032bfd763a1ea31be212181e939f7415b5891eba4109f1f8005e65
thread:issue:61758,issue #61758: `make_tensor` tracking issue,stale/low urgency,2,stale/low urgency,1,0,61758,,,,0,a3323115636b64971099ef5dfec556bc3679ede7006bd2d34b33790f8bb1bec4
thread:issue:61772,issue #61772: Can I set the algorithm of torch.nn.Conv to CUDNN_CONVOLUTION FWD_ALGO_GEMM by myself?,stale/low urgency,2,stale/low urgency,1,0,61772,,,,0,bf5d99d118f23c846aca0b18b3294dfd437522fc0ef2782424207b51640bf0e3
thread:issue:61781,issue #61781: Letting `_allgather_base` to support multiple tensors as inputs and outputs,stale/low urgency,2,stale/low urgency,1,0,61781,,,,0,c05f92156986d1840ed8fa12ee24e78f08b4fb7c62507e05a23c172f7a502be3
thread:issue:61816,issue #61816: Better argument names for torch.atan2 and other math functions,stale/low urgency,2,stale/low urgency,1,0,61816,,,,0,0e87e73e660b58473a790389fc646483ead25ffb007663ee8683466e2047ea08
thread:issue:61825,issue #61825: Improved OpInfo dtype testing,stale/low urgency,2,stale/low urgency,1,0,61825,,,,0,3c5216b916bae0492e523a71cc7b9b86861c4610fb007b1bc6c495a6fda84e18
thread:issue:61827,issue #61827: Bail out if `-std=c++` setting is detected during build,stale/low urgency,2,stale/low urgency,1,0,61827,,,,0,c6f929da4fd270928be990e2a404d26dc7aa3b932fc041802e4f3ff8ce60a982
thread:issue:61835,issue #61835: Broadcasting documentation diverges from implementation,stale/low urgency,2,stale/low urgency,1,0,61835,,,,0,938b39dd1f7bcc79a070a720ffaa193b87ed941d04ddfbb416970f0bdf1d4b0a
thread:issue:61839,issue #61839: died with <Signals.SIGSEGV: 11>,stale/low urgency,2,stale/low urgency,1,0,61839,,,,0,7e43aff3d8b8fd2770887be55e5a48ede6cc811c2b664c6ac08c8d725f893672
thread:issue:61843,issue #61843: Investigate fairscale's use of storage APIs and find alternatives,stale/low urgency,2,stale/low urgency,1,0,61843,,,,0,65de1fc353d4c91f0a6da0f0c6b6a2936417ea9248217acceed50f51f1051c71
thread:issue:61890,issue #61890: The matrix multiplication operator can't get correct results on 3090 !!,stale/low urgency,2,stale/low urgency,1,0,61890,,,,0,3477808c0fbf3f183731890f91f12911706efb3b6981b6ffa969816cd386b1ec
thread:issue:61909,issue #61909: Support matrix operations between complex and real tensors,stale/low urgency,2,stale/low urgency,1,0,61909,,,,0,2d3cf19ba9d1844676a1560a2e4b1542da6de1ff3dcce3d3b523721072dc9aa5
thread:issue:61919,issue #61919: RuntimeError: !ref.requires_grad()INTERNAL ASSERT FAILED,stale/low urgency,2,stale/low urgency,1,0,61919,,,,0,be04a4ad87121c6e3036ac6e0f79f73d90e790bc65658d3e212e097b467317bd
thread:issue:61923,issue #61923: Calling `torch.save()` twice writes a corrupted file,stale/low urgency,2,stale/low urgency,1,0,61923,,,,0,56d3ae1b5647e36feece0a3be395298998c50fdb8630c6517a82c491ec1fb974
thread:issue:61930,issue #61930: NNAPI delegate library cannot be built on MacOS,stale/low urgency,2,stale/low urgency,1,0,61930,,,,0,1de32ebe680d5e2f253df94cbfd33a73f3508950a7e14ead34752f41e619abc6
thread:issue:62011,issue #62011: Enable clang-tidy on all of master,stale/low urgency,2,stale/low urgency,1,0,62011,,,,0,f8a8acd48bb425cf59aedd8cd6e1684a9dcb61de60094e00c0e31724c995c13d
thread:issue:62016,issue #62016: ppc64le build fail: invalid conversion from Bfloat16 in functional_base.h:180,stale/low urgency,2,stale/low urgency,1,0,62016,,,,0,bce55be0fd3e31290ebb770e2f8a463b13cb8e1fb32fa1412cf94c23282ab2ac
thread:issue:62026,issue #62026: Should we be upcasting integral types to int64 in torch.sum and torch.prod?,stale/low urgency,2,stale/low urgency,1,0,62026,,,,0,79fbfdd047769e66b788ead3f665c1d135bca4d6b37c11c05a91ba57a24c8703
thread:issue:62031,issue #62031: Caffe2 tensor to ATen conversion doesn't initialize PyTorch CUDA state,stale/low urgency,2,stale/low urgency,1,0,62031,,,,0,e6651ff15bf77fdd3ad84b4009df8a5a88c81516bfe4f0448c2ac66e028681f4
thread:issue:62034,issue #62034: cudnn_batch_norm_backward is extremely imprecise for some input shapes,stale/low urgency,2,stale/low urgency,1,0,62034,,,,0,93a88236d744dd87290fb5aa04f09e0e3a5aeba18b646a9bceab661174e81a17
thread:issue:62066,issue #62066: Memory Leak Found in Persistent DataLoader,stale/low urgency,2,stale/low urgency,1,0,62066,,,,0,908fa4b3e1b6809119aa61463ddc0224c29e4c4343f999f9ef60ad7e68a30fd5
thread:issue:62130,"issue #62130: nn.DataParallel not working on A100 with nvidia-driver 470.42.01, Cuda 11.1, Debian 10",stale/low urgency,2,stale/low urgency,1,0,62130,,,,0,b0d27d52e09c6973ca3e9e59e64767b2616b1a782f9d617a6dfeee8509fec74e
thread:issue:62141,issue #62141: Dedupe code in functional optim classes,stale/low urgency,2,stale/low urgency,1,0,62141,,,,0,31bea87d9b420915d0baa1245a375835fa6ce69597cda5a6b95875d9e8b5ef4a
thread:issue:62162,issue #62162: Clarify sparse COO tensor coalesce behavior wrt overflow + how to binarize a sparse tensor,stale/low urgency,2,stale/low urgency,1,0,62162,,,,0,aa37e0c2080333ce2eaa21236cfcdf91c6d18eec113a8c0df25b1196e3592c89
thread:issue:62168,issue #62168: Error installing from source,stale/low urgency,2,stale/low urgency,1,0,62168,,,,0,ee88de241af973e8ae625cc6203244534633699023c11414ee28b235abebeea5
thread:issue:62178,issue #62178: Wrong implementation of method log_prob in torch.distributions.negative_binomial,stale/low urgency,2,stale/low urgency,1,0,62178,,,,0,a537cdc0986c59f3834c3a9aad7ce7bea260de85d7d6c4c177e867213704e37a
thread:issue:62255,issue #62255: Support `torch.linalg.trace`,stale/low urgency,2,stale/low urgency,1,0,62255,,,,0,3908d5ab72f98e1149983084a20526bebb4b1c608297dda432af10d3ca2a96fb
thread:issue:62265,issue #62265: `periodic-pytorch-linux-xenial-cuda11.3-cudnn8-py3.6-gcc7-build` started to fail,stale/low urgency,2,stale/low urgency,1,0,62265,,,,0,73fc24a726b051755010cb63a291a29433ab60c1490571d19f75dc2f98e05946
thread:issue:62271,issue #62271: Refactor serialization tests to use device parametrization,stale/low urgency,2,stale/low urgency,1,0,62271,,,,0,af32fcad930988bdf2e43f9e9f422b0c5a7c859ab8788e2dcade89d5019c266e
thread:issue:62273,issue #62273: Serialization map_location silently ignores xla/mlc/meta (any serialization mechanism that skips storage),stale/low urgency,2,stale/low urgency,1,0,62273,,,,0,419e0cc5cdc3edb15ea79621e67736a5e19d9c6d5246ee2fa09e110a19ecb50a
thread:issue:62343,issue #62343: maybe_wrap_dim should have an example attached to it,stale/low urgency,2,stale/low urgency,1,0,62343,,,,0,2fc5df7126298d277f83ed0da6c946a841a92a2fbcb5024d9987f29f0aa2a981
thread:issue:62350,issue #62350: NotImplementedError: Could not run 'torchvision::nms' with arguments from the 'CUDA' backend.,stale/low urgency,2,stale/low urgency,1,0,62350,,,,0,c85b7ad0801711e0b26b416a33dfb2e471c333b47f2e9010b85cacb41a9de446
thread:issue:62354,issue #62354: Some way to specify expected failures for OpInfo-based tests,stale/low urgency,2,stale/low urgency,1,0,62354,,,,0,a0f1a8249af69f2d2abb2c4eb747c73d2a10f5c4948a5be40f44f7001a5d2bdb
thread:issue:62395,issue #62395: [FX] Does PyTorch want to support for dynamic control flow for torch.fx?,stale/low urgency,2,stale/low urgency,1,0,62395,,,,0,a44573e463a7ccba98882eb41426b0a123ed1e3537b324c3bd850002c7e24300
thread:issue:62399,issue #62399: Understatement about how running_var is computed in BatchNorm,stale/low urgency,2,stale/low urgency,1,0,62399,,,,0,26844796f8887c2638591e2d8ec6cbdd7355e0075c165ddd3325ae1ae14cfa03
thread:issue:62446,issue #62446: [docs] Can't get header links to h3 on docs pages,stale/low urgency,2,stale/low urgency,1,0,62446,,,,0,ea8064df6c02c9fb064ed410c9377fa2c4bdf287fe22872b5e4fd913834e1de3
thread:issue:62449,issue #62449: Review replacing test/test_namedtuple_return_api.py with an OpInfo-based test,stale/low urgency,2,stale/low urgency,1,0,62449,,,,0,3f9ab1b8271663833a5b0702047e9a8ad8d3a2ba5ac58aa284eeff098a50ee1d
thread:issue:62473,issue #62473: Sparse updates to logits in distributions.Categorical,stale/low urgency,2,stale/low urgency,1,0,62473,,,,0,e1593a818bc6350abd0aad7eef9ef85136fa4d2a0a62cff758f28adb6663f0fa
thread:issue:62500,issue #62500: meta tensor `set_` seems fishy,stale/low urgency,2,stale/low urgency,1,0,62500,,,,0,927548da9ee2ce7a95f204d716f6d8844258cc51649d1010ce05688ee7151797
thread:issue:62545,issue #62545: Incosistency with args for `nn.functional.max_poolNd` vs `nn.MaxPoolNd` functions,stale/low urgency,2,stale/low urgency,1,0,62545,,,,0,c3b3b3f09e29768a9e6df949c6b6ea7fb057b0a938a8ee3f454f167871ddf263
thread:issue:62554,issue #62554: `torch.fx.node.normalized_arguments` does not pass `normalize_to_only_use_kwargs`,stale/low urgency,2,stale/low urgency,1,0,62554,,,,0,1616025ca3cdb511676ef00738e3bcd198cc69561fd418ed917e9779e8c2502f
thread:issue:62566,issue #62566: `c10::optional<T>` operators should delegate to corresponding operators on `T`,stale/low urgency,2,stale/low urgency,1,0,62566,,,,0,68a819392234fc77033f5bcb588b97d6e2cfcd121c08ec06461b5d71ee08daa9
thread:issue:6257,"issue #6257: [feature request] adding a nonzero element ""in-place"" in sparse tensor",stale/low urgency,2,stale/low urgency,1,0,6257,,,,0,52ec9b7281cec248bbdf385c0c53306e038cebe874ebe0fd550d9787c2c9ba8a
thread:issue:62588,issue #62588: static library kineto_LIBRARY-NOTFOUND not found.,stale/low urgency,2,stale/low urgency,1,0,62588,,,,0,8a17c019c5aa07defb26226a78e70a1cc3082b4e14bcf073d7042b2c073dcf4f
thread:issue:62591,issue #62591: Unnecessary python bindings and documentation for internal functions/ops,stale/low urgency,2,stale/low urgency,1,0,62591,,,,0,976ffc3ee772ba5e295cc87b53a73abce4f0f8bf7ab8e3806ec331ee161bc41b
thread:issue:62602,"issue #62602: ""test_events_wait"" is flaky on ROCm",stale/low urgency,2,stale/low urgency,1,0,62602,,,,0,7bfecb6e4a0c33db5d26efbbf26545ec3f277d72ddc4baa2757d35987a0baa18
thread:issue:62606,issue #62606: Guidance on where to add documentation,stale/low urgency,2,stale/low urgency,1,0,62606,,,,0,0ce09cb25c4b248a5b671fd2fc855f63879f1c30305f04a85db9642d381f614c
thread:issue:62630,issue #62630: Instructions for CMAKE_PREFIX_PATH seem to be broken,stale/low urgency,2,stale/low urgency,1,0,62630,,,,0,cb1b8088e7235be40bde0e014460a0b40f86bfcd6e61ee9f5246732cff8c16d1
thread:issue:62640,issue #62640: add gamma to CosineAnnealingWarmRestarts so max lr can decrease in cycles,stale/low urgency,2,stale/low urgency,1,0,62640,,,,0,deb5a3dd5241a93f6d2d11b8c448b57703d646a9c07e022662f30a01dfa29841
thread:issue:62648,issue #62648: Feature request: Generate Sobol points directly on the GPU,stale/low urgency,2,stale/low urgency,1,0,62648,,,,0,28dc7f4c53bf9fa14749c3891365c0875df337f17257079818b313f488dc3371
thread:issue:62650,issue #62650: Error while importing torch: libtorch_python.so: undefined symbol: PyThread_tss_alloc,stale/low urgency,2,stale/low urgency,1,0,62650,,,,0,0689e9f01851f384ec5d90f6d6da9743280b1804b0980e2a4aaaf4377f0bad25
thread:issue:62654,issue #62654: Checkpointing without re-entrant autograd,stale/low urgency,2,stale/low urgency,1,0,62654,,,,0,130fa640c517a3b6a6290ef3983cb53d6c27ad66eaaf6715cb5bcc6e98658fa1
thread:issue:62655,issue #62655: Make KernelFunction::makeFromUnboxedFunctor infer KernelFunctor from input argument,stale/low urgency,2,stale/low urgency,1,0,62655,,,,0,fd78496431fc1a9af74095d69aa597d1b230e49f37811f5e8eacb5a0e4ceef5d
thread:issue:62656,"issue #62656: Building pytorch from source results in ""an object with that name is already defined"" error on import",stale/low urgency,2,stale/low urgency,1,0,62656,,,,0,11b1827b0769bd2bc526eaacb77ae027919108bab707b46c750c6c652e5361d0
thread:issue:62661,issue #62661: Rewrite PyTorch CUDA backend in Triton,stale/low urgency,2,stale/low urgency,1,0,62661,,,,0,5f0742afdb3c5838c57b65463823b76d964412c616510e779c763515c8453f0d
thread:issue:62672,issue #62672: Add Convolution support for lazy tensor,stale/low urgency,2,stale/low urgency,1,0,62672,,,,0,100e3eb3092c636e51e1b8567bdf4865e8e172453615ba05ed9dcdcb0723e8f6
thread:issue:62673,issue #62673: Make printing to stdout / stderr in tests in CI an error,stale/low urgency,2,stale/low urgency,1,0,62673,,,,0,a48640d6caffb825481a1dcae5939fac452c5c33a02bc1aed3c3642255ad337a
thread:issue:62695,issue #62695: Use fast gpuAtomicAdd for FP16 data type in CUDA kernels,stale/low urgency,2,stale/low urgency,1,0,62695,,,,0,d8d6288363c51a020fa60ab22b55de40effaec958203afcae4dcf5fd5a710753
thread:issue:62699,issue #62699: Can not parse the caffe2 pretrain model. google.protobuf.message.DecodeError: Error parsing message,stale/low urgency,2,stale/low urgency,1,0,62699,,,,0,996982216bcc4d02aade42edd429ad6d1f886f25c406a6d363c5ff965a56a5e2
thread:issue:62742,issue #62742: Add a way to edit CI env variables from PRs,stale/low urgency,2,stale/low urgency,1,0,62742,,,,0,01ea35c3cae89e39bfec1c232d86b027693a7f15832e63fa6d31c21aa29c0d3f
thread:issue:62788,issue #62788: PyTorch(with mkldnn) in inference mode using non-optimal prop_kind for convolution,stale/low urgency,2,stale/low urgency,1,0,62788,,,,0,8b5107cbb9e552d234c4ea3897e180843953553c223b1131a0bd856856af796b
thread:issue:62790,issue #62790: Remove c10d::kDefaultFirstBucketBytes,stale/low urgency,2,stale/low urgency,1,0,62790,,,,0,377f60861db1a6518ebd6215b20e79b44fdf60960d908234f8238647346d3cbe
thread:issue:62845,issue #62845: allow specification of variadic arguments in native_functions.yaml,stale/low urgency,2,stale/low urgency,1,0,62845,,,,0,ed276157fcd98b89f182c3792bec1c709feec90d87a594d43a49687339dfbdc9
thread:issue:62849,"issue #62849: slice_embed, select_embed like diag_embed",stale/low urgency,2,stale/low urgency,1,0,62849,,,,0,c7a54624b7a0d03854b81026746df82eb16e29ff88650c4ec7484a912a8074e5
thread:issue:62862,issue #62862: Convert unset variables in `Reducer::Timer` to use c10::optional,stale/low urgency,2,stale/low urgency,1,0,62862,,,,0,8089cc28ea0e0c734650f817bd09fbe6448ac36e47dd0a52a906fa27c89aa791
thread:issue:62865,issue #62865: Multiplication of non-int torch.tensor with string gives unexpected TypeError,stale/low urgency,2,stale/low urgency,1,0,62865,,,,0,60ca00822b4fbc97dc04decee528bd258386b049b16a3cd07819bbd8f3d36c1c
thread:issue:62876,issue #62876: Key already registered with the same priority: uv,stale/low urgency,2,stale/low urgency,1,0,62876,,,,0,d1a1a891620089f35884903ddd2a9f1d97310984e6636b6262667eb5a760d897
thread:issue:62884,issue #62884: Conjugate fallback dispatch key should be per-backend,stale/low urgency,2,stale/low urgency,1,0,62884,,,,0,a08e5212aac24cf3d304a208c49648aa693571830602030e1672b96b24160b6a
thread:issue:62888,issue #62888: Add a mechanism to not universally dispatch to python for all for all operations when using __torch_dispatch__,stale/low urgency,2,stale/low urgency,1,0,62888,,,,0,589bbd9b63b6d51b558e92c26867f6ed76cdc2cc57b73e9d4148da568c7eb6d8
thread:issue:62917,issue #62917: Use `linecache.lazycache` in torch.fx,stale/low urgency,2,stale/low urgency,1,0,62917,,,,0,6bb7b83ecf53e9ba97e55d1db96800a6a88d69a0c2dd941543a1491fabe7d89b
thread:issue:62925,issue #62925: [Testing] memory_format decorator,stale/low urgency,2,stale/low urgency,1,0,62925,,,,0,d7269717cd805f9766043c6df97afe6252da8f62cbd56a67c32c5fe1278d3507
thread:issue:62926,issue #62926: cudnn_convolution_add_relu fails under basic conditions,stale/low urgency,2,stale/low urgency,1,0,62926,,,,0,381bcc61907865cb3084063080b04b98cee4397fa6911c6fe2c1c53f967e03cd
thread:issue:62933,issue #62933: Compilation error with gcc11 and libstdc++11 (thread_id != operator removed),stale/low urgency,2,stale/low urgency,1,0,62933,,,,0,8a4cc78401b84b8174f1a4b6fdf3acc09c317d724d52954fa5baa92f63c8eddd
thread:issue:62934,issue #62934: `tools/autograd/derivatives.yaml` doesn't support methods on optional tensor,stale/low urgency,2,stale/low urgency,1,0,62934,,,,0,dc333bd4d1a9d5e8ff6980d432deb5b0e28eb90fcfcd211389f5d05712a9d1ea
thread:issue:62955,issue #62955: Make axes selection keyword arguments in torch.diagonal and torch.transpose consistent,stale/low urgency,2,stale/low urgency,1,0,62955,,,,0,309ca1cbbc306ed98820f79a02e1be6950dc03371d8d65ce6bba76245618bd20
thread:issue:62973,issue #62973: Audit use of `C10_UNUSED`,stale/low urgency,2,stale/low urgency,1,0,62973,,,,0,b164858ff2fd50690d48d844921235ff6d6a63ee095aeb4c1871f94446ac47cf
thread:issue:62989,issue #62989: Doc page for functions with overload is not properly indented,stale/low urgency,2,stale/low urgency,1,0,62989,,,,0,83e7e379d701fa46b5f7dc305d7db532a076797461badb198433fa9f0e638a53
thread:issue:630,issue #630: Add Peephole connections for LSTMs?,stale/low urgency,2,stale/low urgency,1,0,630,,,,0,81448f9a695740f707c8d25ebb6c7cba4ce65f579463c1cbf0ea5992ba9ee0d5
thread:issue:63004,issue #63004: ignore_index does not work as described with nn.CrossEntropyLoss,stale/low urgency,2,stale/low urgency,1,0,63004,,,,0,5f516de4ed187c4ac51e839edeb8d75099f00be490c01a927b13549b8e2e4c6f
thread:issue:63016,issue #63016: grid_sample should not require grid to be the same dimension as input.,stale/low urgency,2,stale/low urgency,1,0,63016,,,,0,1ce9bc9f975eaa1455fe50b0ae8d292a54152e12bd87fe889ece49d886cabed6
thread:issue:63041,issue #63041: [RFC] Should DDP support custom reduction logic for registered module buffers?,stale/low urgency,2,stale/low urgency,1,0,63041,,,,0,a905abb2ffe6d2c65351e7d724bbd458033b7e54bfb5cf4a650a602284fe8908
thread:issue:63057,issue #63057: Find out why TestSparseCPU.test_softmax_cpu_float64 grad and input dtypes are different,stale/low urgency,2,stale/low urgency,1,0,63057,,,,0,dfd3c12bc0d6d41048d4aaefe52e30603af31f76f3a0b806e215fd47b9e6ec85
thread:issue:63069,issue #63069: CUDA error: invalid configuration argument for torch.sparse tensor backward,stale/low urgency,2,stale/low urgency,1,0,63069,,,,0,ae5822c96da183cc379be66765befca42c61683abade6affb34d59fb35ce5441
thread:issue:63076,issue #63076: Tracing models all the way down to basic functions,stale/low urgency,2,stale/low urgency,1,0,63076,,,,0,31841a9f63aebf3e2a36ebf3c55a983c833f33884efa872ec521e8f6edbbbc61
thread:issue:63078,issue #63078: torch.cuda.amp fails with torch.sparse.softmax,stale/low urgency,2,stale/low urgency,1,0,63078,,,,0,4acc4e37ab05bd15e8f994d75fbe677f9f1ea854fd83d63009071273331af765
thread:issue:63082,issue #63082: script/build_android nn::Module support,stale/low urgency,2,stale/low urgency,1,0,63082,,,,0,6744d3740701f54aa9f3cbbaeb4617de741190b50ca175d0efe6c1df596851ad
thread:issue:63113,issue #63113: Implement half_to_float flag for both _log_softmax and _softmax operators,stale/low urgency,2,stale/low urgency,1,0,63113,,,,0,17d36fbbf70e254f97a7f2d8cf0f512f24985496705d1105af72c3c8e3aa885d
thread:issue:63138,issue #63138: Sign in slogdet is set to requires_grad = False even when using complex numbers.,stale/low urgency,2,stale/low urgency,1,0,63138,,,,0,ff9410231c7b96242bb955733bde3deb634d30f70a2ee73196a506d4a930f3f8
thread:issue:63140,issue #63140: [documentation] torch.distributed.elastic: illustrate how to write load_checkpoint and save_checkpoint in Train Script,stale/low urgency,2,stale/low urgency,1,0,63140,,,,0,9326badf08dbc68726e8ed7b36bf27a11eac4aca693623c9fd6290a95512f54a
thread:issue:63188,issue #63188: ROCm build documentation lacks needed dependencies,stale/low urgency,2,stale/low urgency,1,0,63188,,,,0,c0453ddf8f8a97a3e2f3e8d414a9340b587c436c2978af36db4d4d1efe979108
thread:issue:63205,issue #63205: split up test/test_autograd.py,stale/low urgency,2,stale/low urgency,1,0,63205,,,,0,73e19698e7fc3924a7e7050762ecc9f4912332f398684c4b1f6d1e76ded88927
thread:issue:63207,"issue #63207: document our compatibility with upstream packages (e.g. languages, compilers, tools)",stale/low urgency,2,stale/low urgency,1,0,63207,,,,0,938126a58c0bd56bdb86f1b457c93f2fc7bbd203904ca3c8c890ff4a8aff4b46
thread:issue:63211,issue #63211: Memory leak in `ddp_zero_hook.py` hooks,stale/low urgency,2,stale/low urgency,1,0,63211,,,,0,64a5b3e923f2c1dd02c1a995dcd9e3851290a651fada6ae84fdea43dcfd29947
thread:issue:63217,issue #63217: Torch.nn.functional.interpolate() function memory release problem,stale/low urgency,2,stale/low urgency,1,0,63217,,,,0,3f61c43bec9ec6873739eaadc82086e232212d63ab9968da6ad6e63a64ae9a71
thread:issue:63223,issue #63223: [bug] torch.distributed.elastic logging: Failed to print the first info statement,stale/low urgency,2,stale/low urgency,1,0,63223,,,,0,be9e8097f00dc61460efcbb4576f984023738476bf8c4440b57d80657a52c93a
thread:issue:63226,"issue #63226: cuda.is_available returns True in pycharms python console, False in code",stale/low urgency,2,stale/low urgency,1,0,63226,,,,0,ea784f1b74b7d0a3346cf6bac9bbbec479ad025eabb7d46f9d8bb9e7bead7c68
thread:issue:63257,issue #63257: `help(torch.finfo)` doesn't give useful information even though class has documentation,stale/low urgency,2,stale/low urgency,1,0,63257,,,,0,b6dd2bc60e0297ae0a4e1c87f4d139c620eb01893589edab8c5727a3b55801c2
thread:issue:6328,issue #6328: [feature request] Include libomp support (macOS),stale/low urgency,2,stale/low urgency,1,0,6328,,,,0,5e1326e8792fc9631800818912e155805bea5a11c1611ca359e35b5826d7b6f5
thread:issue:63282,issue #63282: Serialising `torch.bool` generates a warning about `np.bool` being deprecated.,stale/low urgency,2,stale/low urgency,1,0,63282,,,,0,ae057c9987ab7e0b2dca7ae4781616e399a309b34f826257e223ced64d96672d
thread:issue:63290,issue #63290: DistributedDataParallel documentation,stale/low urgency,2,stale/low urgency,1,0,63290,,,,0,b1588aa0585a51388e9108b8a48697c11f336d550ffc85ddb48778019e841b61
thread:issue:63291,issue #63291: [typing] new_ones has wrong signature,stale/low urgency,2,stale/low urgency,1,0,63291,,,,0,0c704f8e6f12363efa533d5cf12200dda7652bff96b96d0a3be8a9d6342c3a3e
thread:issue:63297,issue #63297: Named tensor in tracer,stale/low urgency,2,stale/low urgency,1,0,63297,,,,0,c177aac8258ec40fb8d6b2f6d8fe7f10cf04efbfeb4c33ec32fdb696ea503269
thread:issue:63299,issue #63299: Channels last mode doesn't work with inputs with 0 batch dimension,stale/low urgency,2,stale/low urgency,1,0,63299,,,,0,2dce5ffa1f780a4377542398fae6d3da73ad69e5bbfad3106f00249d8338bf56
thread:issue:63305,issue #63305: Alternate Model Loading - Android,stale/low urgency,2,stale/low urgency,1,0,63305,,,,0,968ff171216d0b97f8523eb7d09755d996a07b126f7cffe072b4d6ed9acfb1f7
thread:issue:63313,issue #63313: Automatic mixed precision works worse for 3D neural networks.,stale/low urgency,2,stale/low urgency,1,0,63313,,,,0,24e54ce7d71f0fe3164aa4e934d69d5da5b7cceeeaf5fbc042669cc569355e1b
thread:issue:63337,issue #63337: Property based testing like AFL,stale/low urgency,2,stale/low urgency,1,0,63337,,,,0,c8d751e7389425e4e17f80a5eb0b8c98ef891e1afd02381f08f4edbd23a0bf88
thread:issue:63356,issue #63356: [python codegen] correctly plumb TensorOptions defaults through the python binding layer.,stale/low urgency,2,stale/low urgency,1,0,63356,,,,0,4ceed8f0d6915481cee7412deb275eecc677034135f4ed19e32818e5a1031ef8
thread:issue:63441,issue #63441: OpenCV causes backpropagation to get stuck,stale/low urgency,2,stale/low urgency,1,0,63441,,,,0,767c4304ab2ab83d208e9f5db350d540deece428fda14ceed54790fb03548e32
thread:issue:63475,issue #63475: C++ Inference error causes Identity Module,stale/low urgency,2,stale/low urgency,1,0,63475,,,,0,e738f6ad3e3a75cb7eb015d3101a2b467d97d0bf72cfdd27370940c3963c20ca
thread:issue:63477,issue #63477: Supporting Pytorch for Custom Compiler Backend,stale/low urgency,2,stale/low urgency,1,0,63477,,,,0,3321544e0eb652c693b5030056d3fc77d78569fb00104db296171016650e4873
thread:issue:63479,issue #63479: Cannot include extension.h under Windows - linker error THPVariable_Wrap,stale/low urgency,2,stale/low urgency,1,0,63479,,,,0,9d2bf1538467572c0a0cb24f9267f3ac3dffae323a514d2428110f46df67042f
thread:issue:63488,issue #63488: MultiStepLR with different gammas for each parameter group,stale/low urgency,2,stale/low urgency,1,0,63488,,,,0,1c2b525c522ffa211eba81492829c32f198488bd8b0f122715b252952aba3cf0
thread:issue:63490,issue #63490: [FX] Data type of target should be validated on Node construction and lint,stale/low urgency,2,stale/low urgency,1,0,63490,,,,0,98b9d015c8b7abd1ba4c577ba97ba8cb704f309820e564e5ad8533f76262628f
thread:issue:6350,issue #6350: worker assignments in torch.utils.dataloader.py,stale/low urgency,2,stale/low urgency,1,0,6350,,,,0,6d859144da4077bc2a52d7a2ed5fb6b3f961b9d185644723a0f655e7782c8d0c
thread:issue:63555,issue #63555: pytorch 1.8.1+cu111 used much more CPU RAM than pytorch 1.8.1 after run `import torch`,stale/low urgency,2,stale/low urgency,1,0,63555,,,,0,8e8a1cf90602eb731552bf386cd8ddcb0844c82855781022a1f68e08d4b33732
thread:issue:63558,issue #63558: universal binaries for libtorch on mac (x86_64+arm),stale/low urgency,2,stale/low urgency,1,0,63558,,,,0,e2f6084a201f079bfe3b00a5ddc869cd170ab9421d4fad9efca1cf6669e1651a
thread:issue:63559,"issue #63559: Functional grid_sample: example of padding_mode=""reflection"" description error",stale/low urgency,2,stale/low urgency,1,0,63559,,,,0,1f491fb43cfae2734bd82fe81981edf2bd193a660036bbdec6dcc0cfb5b0c051
thread:issue:6357,issue #6357: [feature request] Unpooling layer in Caffe2,stale/low urgency,2,stale/low urgency,1,0,6357,,,,0,2fd8665dcbb89f7b9d699287f19f5e7f7ec3ef3279f230887358727dfdd61fd0
thread:issue:63574,issue #63574: compile_commands.json doesn't contain nvcc invocations,stale/low urgency,2,stale/low urgency,1,0,63574,,,,0,61c9087294fc3b313d0b8f192745f1f2367869b5afb9120c3d23fd12005470da
thread:issue:63623,issue #63623: Expose a SavedTensorsHooks nn.Module for users to register saved tensors hooks,stale/low urgency,2,stale/low urgency,1,0,63623,,,,0,3321c97bd244062ef6f00057e69aa20b2820289e8ea004bdb152df5fe14b911b
thread:issue:63624,issue #63624: Raw saved tensors can survive the deletion of the underlying SavedVariable object,stale/low urgency,2,stale/low urgency,1,0,63624,,,,0,56117185885ea23fe38b819cc2966864471e0234dcf47e1cdcd220a5b0c7db11
thread:issue:63644,issue #63644: LeakyReLU and Elu use more VRAM than needed,stale/low urgency,2,stale/low urgency,1,0,63644,,,,0,726a3fdd6bb8b6ea503e2d30f7c3573abd8f1722a7b57f5be019539c1c737811
thread:issue:63645,issue #63645: Create a new OptionalArrayRef template class to replace Optional<ArrayRef>,stale/low urgency,2,stale/low urgency,1,0,63645,,,,0,1876c97353334140c30d5e2f0e9e7a377947528162bbfc1d748e5ebd561bc9d5
thread:issue:63720,issue #63720: Torch.save with zip serialization causes memory bloat,stale/low urgency,2,stale/low urgency,1,0,63720,,,,0,02d7629ec6f32299bd48c66bcfc921be4c9cf98f4e2f4c9e9418b0989d404d2b
thread:issue:63781,issue #63781: [rfc] Hardcoded Target Determination,stale/low urgency,2,stale/low urgency,1,0,63781,,,,0,bb558591d4721f30f8a63eebebef97fdacf22967d2de9798aa73c06395156871
thread:issue:63802,issue #63802: [torch.distributed.launch|run] Hangs on SIGINT when using a TCPStore backed rdzv_backend,stale/low urgency,2,stale/low urgency,1,0,63802,,,,0,9a6349a80d1ddaf17cf77d9429eeda96d4f6570f48668cd6f0e265b9c1a78ee6
thread:issue:63815,issue #63815: [FX] Documentation for Graph.flatten_inps and Graph.unflatten_outs,stale/low urgency,2,stale/low urgency,1,0,63815,,,,0,0b5b779a940a006c03099680cbaeabf3d4dc62d55bf144207233f7eba534dfcf
thread:issue:63837,issue #63837: `torch.scatter` doesn't fail correctly on CUDA (memory overlap),stale/low urgency,2,stale/low urgency,1,0,63837,,,,0,91e732dd145e20d5cb7b9823d85712e165a7560f2f08468885f5cc46c258e98e
thread:issue:63846,issue #63846: [torch.distributed] Make dynamic_rendezvous log handler configurable,stale/low urgency,2,stale/low urgency,1,0,63846,,,,0,61f2e5fa9c24588ac8999c34dadc0e23e69fc1d4774b8113da1b33dd6ed334a9
thread:issue:63849,"issue #63849: Installed successfully the torch-1.9 and torch_xla-1.9. RuntimeError requested XLA_GPU:0，but available devices are [CPU:0, XLA_CPU:0 ]",stale/low urgency,2,stale/low urgency,1,0,63849,,,,0,e79f0425654ed9d22f2bb6020b63e4a5910f9fd82f0453cb7e7f08eebb39c835
thread:issue:63855,issue #63855: Include tensor shape in its default print formatting,stale/low urgency,2,stale/low urgency,1,0,63855,,,,0,ca491a2071411ccde8eeac07a4a01d9a24d58d1228783e9bfc123d91aff3b1c6
thread:issue:63897,issue #63897: Scalar construction in C++: PY`torch.tensor(7)` !~= C++`at::tensor(7)`,stale/low urgency,2,stale/low urgency,1,0,63897,,,,0,9252ad39c84776839bc5f42f7d29ed166b538800491d4eb832a12eb85963cdb5
thread:issue:63911,issue #63911: Add efficient symmetric tensor representations,stale/low urgency,2,stale/low urgency,1,0,63911,,,,0,ce9370d8d6abc891888bc5a994b8b0880328569d6427ce7311a22f451bdca82e
thread:issue:63919,issue #63919: quicklint mypy step fails with assertion,stale/low urgency,2,stale/low urgency,1,0,63919,,,,0,0adec0511a56ff3c69d2f9e74f20cd6a73168d200cba9772b912a401915e055c
thread:issue:63920,issue #63920: quicklint clang-tidy step fails with many unrelated errors,stale/low urgency,2,stale/low urgency,1,0,63920,,,,0,120953a0a4370ed1456b47f5ca267301c2147e176e8d8394e26d7902cdda3b46
thread:issue:63921,issue #63921: More efficient colon backwards in advanced indexing,stale/low urgency,2,stale/low urgency,1,0,63921,,,,0,e8c247fd5cf27a0ad4933bbf496a7269f2f92a44963745dcb108858c3042b255
thread:issue:63935,issue #63935: Inconsistent behaviour of index put when assigning from other type,stale/low urgency,2,stale/low urgency,1,0,63935,,,,0,c6ed11509cac669da304457b4c537551daeb8f08a5a0e4be9c7e6943751e2e6a
thread:issue:63970,issue #63970: [LTC] Investigate a testing utility to make sure correct lowering has happened,stale/low urgency,2,stale/low urgency,1,0,63970,,,,0,dbfa13a42406b428a6dd0ac8bdde6764cf0f13d67d5fcf357c82030d2445fde6
thread:issue:63971,issue #63971: benchmark.Compare raises: TypeError: object of type 'NoneType' has no len(),stale/low urgency,2,stale/low urgency,1,0,63971,,,,0,788b06aecc15198cc5e4d17b83c9a73d84a7160ce94aaed5f19cf827809484e5
thread:issue:63972,issue #63972: bazel build warning: Artifact 'torch/csrc/api/include/torch/version.h' is duplicated,stale/low urgency,2,stale/low urgency,1,0,63972,,,,0,4cbebf4dec0bdcc8700cf1055877a3705f067dc2c0c4c145e6748d8c084a41dc
thread:issue:63992,issue #63992: Direct inversion and linear systems solutions for small matrices,stale/low urgency,2,stale/low urgency,1,0,63992,,,,0,e413162a9a4222faba9cc1c380cd70f500bdd596861da2b471078ac2f5f6ce6f
thread:issue:63998,issue #63998: cat out= variant memory_format inconsistent for cpu and cuda,stale/low urgency,2,stale/low urgency,1,0,63998,,,,0,f9d620c283efbbdab5ab78fda046ea6137e25e0fe1d3f394d0d2609100dd5f12
thread:issue:64006,issue #64006: Missing device and dtype description in torch.nn.MultiheadAttention,stale/low urgency,2,stale/low urgency,1,0,64006,,,,0,641aee9898ba8708b0ae38392c291209b40c83375ca08ba220a1969ccc6b0d5c
thread:issue:6402,issue #6402: [caffe2] How to freeze a layer?,stale/low urgency,2,stale/low urgency,1,0,6402,,,,0,950ac79d829944161927da57417ea642eca5f6450d26cc892ec59d46990c3532
thread:issue:64025,issue #64025: RuntimeError: cuDNN error: CUDNN_STATUS_INTERNAL_ERROR,stale/low urgency,2,stale/low urgency,1,0,64025,,,,0,c6fb356e107feea47bf4b2c8e96b1362347769b120c6816e001d7f7ba63d1db3
thread:issue:64041,issue #64041: torch.unique acting up for a binary tensor,stale/low urgency,2,stale/low urgency,1,0,64041,,,,0,7c7fe798f270b24ac1e389b1810d7cf8f31f048aa84c9006e02c00dd45e97ce1
thread:issue:64043,issue #64043: Uncleared memory use after torch.load(),stale/low urgency,2,stale/low urgency,1,0,64043,,,,0,5fe5d9d8b79f3182544248239edd388acb2f10fa59a8ea1346ec39bd737d5c71
thread:issue:64048,"issue #64048: Be able to use pytest to run a ""core"" set of tests",stale/low urgency,2,stale/low urgency,1,0,64048,,,,0,2041730e7d22181a114511cb371fe2ca02399a3e95dc07ccaf010ca323e67555
thread:issue:64079,issue #64079: torch.equal does not support sparse tensors,stale/low urgency,2,stale/low urgency,1,0,64079,,,,0,86cc3960d3bb38160a297de78f4213cf18c53a707147a103ccf4262075aa2c0d
thread:issue:6408,issue #6408: [caffe2] Prefetching blobs for memory optimization,stale/low urgency,2,stale/low urgency,1,0,6408,,,,0,b7aed4fcdca2835f0bb9a08ab8e34e0d7b783452a1d60884d10108fefed8c0eb
thread:issue:64106,issue #64106: modulefinder_determinator is incompatible with imports into the test/ directory,stale/low urgency,2,stale/low urgency,1,0,64106,,,,0,4a523eecc58b3ce27572a0d300e03fdfcc633a712b0e0726ff9174ba4676a949
thread:issue:64107,issue #64107: as_tensor and negative strided np arrays,stale/low urgency,2,stale/low urgency,1,0,64107,,,,0,23657a88be4cb1591eb34232423772ea9638034253578afd5a60b6d8f3802ed7
thread:issue:64124,issue #64124: ciflow tracking meta issues,stale/low urgency,2,stale/low urgency,1,0,64124,,,,0,b771f57ee45c56d5fd24a30a2fc588c4cfb598072705cb106836844193d7053a
thread:issue:64133,issue #64133: nn.MultiheadAttention does not belong to activation function,stale/low urgency,2,stale/low urgency,1,0,64133,,,,0,e02efcfe637a145221436f674758ca212e23fd224f8bd921098ffc9dbf075b40
thread:issue:64138,issue #64138: Longer int and normal Floats conversion to FloatTensor is strange,stale/low urgency,2,stale/low urgency,1,0,64138,,,,0,541e87f5b142335640788ce4875be426507dd891904b856789dba22130c46b72
thread:issue:64145,issue #64145: F.conv2d: confusing error message when using uint8 input instead of float32,stale/low urgency,2,stale/low urgency,1,0,64145,,,,0,4c0bbf622bffdd9e80925a9bab89ed8fe1dbeef84cc05cb3d1e1016333833cdb
thread:issue:64186,issue #64186: Complex recurrent layers produce NaN as grad,stale/low urgency,2,stale/low urgency,1,0,64186,,,,0,17565ad71340d11bf39a41ed240bd0cfd4495f54c2db801bd93c379d40c679a3
thread:issue:64187,issue #64187: Uniformly test that defaulted Tensor arguments appropriately handle __torch_function__,stale/low urgency,2,stale/low urgency,1,0,64187,,,,0,0a106943dc5fc7ab904a4f12bffc66539172a972cc2e639bae184875d5b2dec5
thread:issue:64206,issue #64206: Document how to generate Pybind bindings for C++ Autograd,stale/low urgency,2,stale/low urgency,1,0,64206,,,,0,5ae15e5cbdb04c49b66544b79dc6be0f8375225067d6c64db2d0e003ae219586
thread:issue:64284,issue #64284: FR: Record results of OpInfo reference tests and detect when numerics of an operator change,stale/low urgency,2,stale/low urgency,1,0,64284,,,,0,91e8d00905d1d1cff0650775c5296c86e35d76592847b66c13d312b0060085a3
thread:issue:64292,issue #64292: torch.empty_like not taking None as layout?,stale/low urgency,2,stale/low urgency,1,0,64292,,,,0,2e0470560bdc4e4a95aa322d87808d7be70fc58f09af3b18380c60ad2328948d
thread:issue:64306,issue #64306: GradGrad of max_pool2d fails with empty batch dimension,stale/low urgency,2,stale/low urgency,1,0,64306,,,,0,867410089b826cff227a94b6bd6cf4a63ec5cc52737b27de5af29b790430ac99
thread:issue:64308,issue #64308: Building from source results in no member named 'cerr' in namespace 'std',stale/low urgency,2,stale/low urgency,1,0,64308,,,,0,1a58e2b874956bd7f7fbe39a2239b6284dfb05599b37dad5c38c0cc9e62b6749
thread:issue:64311,issue #64311: issues flutter in android studio,stale/low urgency,2,stale/low urgency,1,0,64311,,,,0,3b3040f028aa1b982cf43ee477dc218ba3da8fc2ae4f9a10619a12a2aba38938
thread:issue:64384,issue #64384: [doc] incomplete `get_num_interop_threads` doc,stale/low urgency,2,stale/low urgency,1,0,64384,,,,0,0bbd0b6fee39ab47b41dbfc0df6e0c21b55fb3e26fa18bda8835bea930940837
thread:issue:6442,issue #6442: [feature request] dropout1d,stale/low urgency,2,stale/low urgency,1,0,6442,,,,0,a7b22486ac051c55043989b0714bb98851c1053d51247d5149a8a9a9ed41de55
thread:issue:64426,issue #64426: Add support for ROCm MIOpen miopenConvolutionForwardBias and miopenConvolutionBackwardBias,stale/low urgency,2,stale/low urgency,1,0,64426,,,,0,64a3c658f5a11e4332dbd7e7f7fafdb2d63f52ce1a8869412d74c00de75bffe3
thread:issue:64458,issue #64458: Rename C++ Serialization APIs,stale/low urgency,2,stale/low urgency,1,0,64458,,,,0,e0139e0f1b9740b4b1b76fbc164e8ce5146ce830a36dfab67645079c156e1fbb
thread:issue:64477,issue #64477: torch.onnx.export crash while export quantized model,stale/low urgency,2,stale/low urgency,1,0,64477,,,,0,99b9902997e0b8a2642a3e2f50886011f9bbe9d54a46f688788f493f3f6a5528
thread:issue:64491,issue #64491: TypeError when using torch.cuda.list_gpu_processes() on Windows with the WDDM driver,stale/low urgency,2,stale/low urgency,1,0,64491,,,,0,69fd98fa37fb5f5c771110ff45522d1bdcc199e485f7ec360a81545a35f823f6
thread:issue:64497,issue #64497: FX tracing on Fairseq-based roberta fails.,stale/low urgency,2,stale/low urgency,1,0,64497,,,,0,b3506347f272095f3bffe92fbb0bfdc3ac7ad1ce0c857bfa3b2fc605c925fa47
thread:issue:64502,issue #64502: [numpy] torch.nonzero is similar to np.argwhere not np.nonzero,stale/low urgency,2,stale/low urgency,1,0,64502,,,,0,e22fdd519de8ddcc555a9300f0bb7d8d83653bc1e0996224e6d8b28b47028411
thread:issue:64509,issue #64509: Return attention weights in nn.Transformer,stale/low urgency,2,stale/low urgency,1,0,64509,,,,0,f1abb949ca8e97e5c3d601fd27ea4a943cb39dc63fd93c1a85f11e0e6b2b8133
thread:issue:64525,issue #64525: nn.TransformerEncoder - all nan values issues when src_key_padding_mask provided,stale/low urgency,2,stale/low urgency,1,0,64525,,,,0,fdb9b5150454d09f01b10f60234ada999e1b75a224365baf5f801833f35999df
thread:issue:64527,"issue #64527: Asking for ""git submodule update"" when building a specific version",stale/low urgency,2,stale/low urgency,1,0,64527,,,,0,4a615138e41aa2a256e076ded48fb72f33bd33e2b43083b88b4099c51744e841
thread:issue:64539,issue #64539: [LTC][Documentation] Python examples in API_GUIDE.md don't work,stale/low urgency,2,stale/low urgency,1,0,64539,,,,0,087092d5f010983c8311c4e363aba2a2cc141d855b832edd4dfdaa9d59237b23
thread:issue:64541,issue #64541: [docs] Version switcher to not reset home page unconditionally,stale/low urgency,2,stale/low urgency,1,0,64541,,,,0,4d53320c7344cb3d1a30e7c6a0ae7b23f7643174b2bdbf31e342a5bca20eddf1
thread:issue:64548,issue #64548: Interpreter initialized state changed by torch dispatch,stale/low urgency,2,stale/low urgency,1,0,64548,,,,0,0fac45dababd8271cffa07795d318f355e73d7c120eb94d980188298b890b63f
thread:issue:64635,issue #64635: 'LSTM' object has no attribute '_flat_weights_names',stale/low urgency,2,stale/low urgency,1,0,64635,,,,0,5d3c5e7301a406f78773e8d656096474575003f0169475d9bb9c51b268c76ac7
thread:issue:64637,issue #64637: Evaluating on single GPU (DDP),stale/low urgency,2,stale/low urgency,1,0,64637,,,,0,31c2d8031d21c727de6877942756d809820ea654f2ecf59e0ac8a9099e4b4bfb
thread:issue:64644,issue #64644: Improve meta tensor testing,stale/low urgency,2,stale/low urgency,1,0,64644,,,,0,387f4e988cc58e77160042ef57ed27c7915b4dcde8d1f9f24573e24183fa1c31
thread:issue:64645,issue #64645: Compilation from source fails,stale/low urgency,2,stale/low urgency,1,0,64645,,,,0,c693e0ed98842044df24bba741ae96dd71bf2c517110065a22c0580235a43884
thread:issue:64690,issue #64690: _disabled_torch_function_impl is unimplementable (must be special cased),stale/low urgency,2,stale/low urgency,1,0,64690,,,,0,2efb006ecc873b3d7dc8da90261a6d9eaa14f60192c71d9a923e7137a6a15540
thread:issue:64692,issue #64692: Remove the dependency with CUDA_LIBRARIES in TorchConfig.cmake,stale/low urgency,2,stale/low urgency,1,0,64692,,,,0,abfaf62adae8c434990d2712d4a7ec7721b3f5e34f859c49abcbd8b4116689fe
thread:issue:64709,issue #64709: Add out= resize warning to cat and maybe_native_stack,stale/low urgency,2,stale/low urgency,1,0,64709,,,,0,97058778aaa67e4783739435c247c1c484685b9b9d85e73abd88b05381054f4b
thread:issue:64711,issue #64711: Improve Docs builds to not have to push to a repository to update,stale/low urgency,2,stale/low urgency,1,0,64711,,,,0,492f13ce0d24aa08f5b42e2e406896c4762a4c95de989aea704c475c23402080
thread:issue:64743,issue #64743: Named Tensors to support custom unify functions,stale/low urgency,2,stale/low urgency,1,0,64743,,,,0,e62b27366c69855e6653084e468ad7be3e030d4c5f4e04109435c24b28889298
thread:issue:64744,issue #64744: Consolidate ProcessGroup allreduce_coalesced and allreduce,stale/low urgency,2,stale/low urgency,1,0,64744,,,,0,bb45331ee27cd91a7bc56ac98b2591acecaecb606c48ab72a67d42b632ed5ffa
thread:issue:64745,issue #64745: Consolidate ProcessGroup allgather_coalesced and allgather,stale/low urgency,2,stale/low urgency,1,0,64745,,,,0,f061608b24750e4515190d854c5154cb727a36f71b531c5689d0334fa148bdd7
thread:issue:64766,issue #64766: Ability to explicitly close/dispose a DataLoader,stale/low urgency,2,stale/low urgency,1,0,64766,,,,0,290848c5c6b086c572e4f9acf5563cbff373ebc2d6d56b8c0b26cb366025f224
thread:issue:64818,issue #64818: Inconsistent NaN handling by cholesky between CPU and CUDA,stale/low urgency,2,stale/low urgency,1,0,64818,,,,0,b0a72a974796c7bcaf383956709e95a70be526b9453ed46bf9e7b9dc41035fec
thread:issue:64845,issue #64845: Track Reliability of Test Reporting and Uploading,stale/low urgency,2,stale/low urgency,1,0,64845,,,,0,5ad047512b17e6afd535ba03acdb88d16d26ceba59e86833ee9a5edebc386e16
thread:issue:64864,issue #64864: Base parallelism on CPU count available to process rather than system total,stale/low urgency,2,stale/low urgency,1,0,64864,,,,0,09a5afc2ad82467d5a2f5770b892ab9574a64fd125c00bd0345787a5c930bd89
thread:issue:64868,issue #64868: CMake outputs lots of warnings in CI,stale/low urgency,2,stale/low urgency,1,0,64868,,,,0,c6f7b707057bebf5d4189562ba72ebe0c37fb1e549ee1bffd10284b504b6ec3d
thread:issue:64874,issue #64874: I am unable to detect torch.Tensor._version change during __torch_dispatch__ call,stale/low urgency,2,stale/low urgency,1,0,64874,,,,0,39fcaf65238ba542f060a3e24dd20cf4ea479725efc267123ef0244df4e1f1a4
thread:issue:64931,issue #64931: native functions should not be allowed to take in a `grad` argument,stale/low urgency,2,stale/low urgency,1,0,64931,,,,0,c716859a7a8c00f4df887873d5e4ea310d90618bcd09f6ec947ada00a1957e54
thread:issue:64932,issue #64932: [RFC] TorchStore - A Shared-Memory Tensor Store,stale/low urgency,2,stale/low urgency,1,0,64932,,,,0,d34afdcdbe33a486f260d4694e4dbea389c86a21f66bf302d83630fbc2004895
thread:issue:64939,issue #64939: BC CI error message should link to some information about how to squash the warning,stale/low urgency,2,stale/low urgency,1,0,64939,,,,0,a9a5b1cf115bd288402a89c9bdf9eba13ad4714afecdea0bba8049c371326ae4
thread:issue:64997,issue #64997: FX failed to normalize op `block_diag` and `broadcast_tensors`,stale/low urgency,2,stale/low urgency,1,0,64997,,,,0,1e6bf19d4bf5b7b75e469da9a0ca10bcfa49880df687b9209f6e50ea8b11706a
thread:issue:65002,issue #65002: Traceback tensor mode,stale/low urgency,2,stale/low urgency,1,0,65002,,,,0,b901008265a565df6c3678750d61b69a89fde970885e3f08b57210055b3f4a52
thread:issue:65022,"issue #65022: [AWS EC2 P3DN, EFA is enabled] Torch RPC tensorpipe/common/ibv.h:172 """": Operation not supported",stale/low urgency,2,stale/low urgency,1,0,65022,,,,0,fa810b6d6e528def5026bd8b2ce8b62df1d282f7f91d2d49e3a933c2c999e684
thread:issue:65038,"issue #65038: [pytorch v1.9.0] for torch sync.pipeline in a single GPU, passing tensor layer by layer is better or using the skip-connection API is better?",stale/low urgency,2,stale/low urgency,1,0,65038,,,,0,21a45be2ff455afd7595f99c697f5136b0217977b02bf05074a7fe3ae68a06a1
thread:issue:6505,issue #6505: [Caffe2] Onnx to Caffe2 Blob error,stale/low urgency,2,stale/low urgency,1,0,6505,,,,0,97103a391e3c66de0cd1913d10c6506318b1470e472ca6bcef68ec29cd981846
thread:issue:65050,issue #65050: torch.cross precision problem,stale/low urgency,2,stale/low urgency,1,0,65050,,,,0,9b5c9c4ff06cfc0f93e0b62d5ca124d11616fb4c018a091f3c8b4147ab9a4c46
thread:issue:65062,issue #65062: Gloo and TensorPipe depend on different version of libuv,stale/low urgency,2,stale/low urgency,1,0,65062,,,,0,a8bcfef5430419f9a258e0275c02ca8135d2f41f3b23f3fbfc9dfe0c55cd1506
thread:issue:65067,issue #65067: [DataPipe] Add Test Cases to Ensure Correct Behaviors When IterDataPipes Reset,stale/low urgency,2,stale/low urgency,1,0,65067,,,,0,9b8812b8dfa3aaa50ae63b6136a6c00f5a5454d6fabf00347df30145993d8598
thread:issue:65069,issue #65069: torch.linalg.cholesky does not raise a RuntimeError when the matrix is not symmetric,stale/low urgency,2,stale/low urgency,1,0,65069,,,,0,3ce56935eb682ceae4be87a1b3878543ab078b33de35a619dc2dd8afb3c6698f
thread:issue:65083,issue #65083: THPVariable_subclass_dealloc (i.e. tp_dealloc for torch.Tensor) should handle Python error state gracefully,stale/low urgency,2,stale/low urgency,1,0,65083,,,,0,55e232037a1707321f299b457e200e409a8afb1a231a73be112033c4cca36377
thread:issue:65096,issue #65096: [FX] `passes.split_module` doesn't have documentation,stale/low urgency,2,stale/low urgency,1,0,65096,,,,0,3e4e0082cca92088ee91624e37dc92ca3b174cbf845210d2d3048dd8f7e541ba
thread:issue:65100,issue #65100: [Bug] [CUDA IPC] CUDA IPC memory cost,stale/low urgency,2,stale/low urgency,1,0,65100,,,,0,1d7d49ed827d54f9d9969e005651eec9510fe137444f85b58953038eaa09d0d5
thread:issue:65140,issue #65140: profiler_legacy `_parse_legacy_records` deduplication logic has false-positive.,stale/low urgency,2,stale/low urgency,1,0,65140,,,,0,d97f827e265060501cb1678da697331166156286e2b546ce239e201f46d1faf0
thread:issue:65153,issue #65153: Unification of model initialization methods / naming across domain libraries + support of skip_init,stale/low urgency,2,stale/low urgency,1,0,65153,,,,0,b07b64a9b6915129ef1ab6b68948a77f3f477d712200388fefcbfd43a4e8777a
thread:issue:65161,issue #65161: error: ‘_mm512_set_epi8’ was not declared in this scope when build from source on AMD 3600 CPU,stale/low urgency,2,stale/low urgency,1,0,65161,,,,0,c55a5c404eaa3ba9550db0d3ac142596df73e8eb28616477c836e4c29352a46e
thread:issue:65163,issue #65163: Parametrization cannot be parametrized,stale/low urgency,2,stale/low urgency,1,0,65163,,,,0,8da743f119ef279ab54df5dfe5753e42949cfffc8581dde302e79aa889bad806
thread:issue:65166,issue #65166: Review the Tensor and Operator Basics Onboarding page,stale/low urgency,2,stale/low urgency,1,0,65166,,,,0,9b8f853a213c6de9b384ea798b862f971b93d655cf47c856184954cbe312bb25
thread:issue:65198,issue #65198: Multi-processing leaking file descriptors,stale/low urgency,2,stale/low urgency,1,0,65198,,,,0,e95631dd1a1db53aa47e24e9bcd8f8cc15c04220b5d120db9973b539ab8ce4c6
thread:issue:65215,issue #65215: _disabled_torch_function_impl should not be necessary for __torch_dispatch__,stale/low urgency,2,stale/low urgency,1,0,65215,,,,0,54a4003e5560d221dbdef5a04a7ab484e1e02235913bab71cd03ed5e55e9548c
thread:issue:65216,issue #65216: Add function that retrieves a batch from a DataLoader,stale/low urgency,2,stale/low urgency,1,0,65216,,,,0,c5871dcaa6f8e6f00e5903b6cc9c9d5e979f8afe58eb412dac61a8b496791cc2
thread:issue:65229,issue #65229: at::parallel_for created max_threads for inputs larger than GRAIN_SIZE,stale/low urgency,2,stale/low urgency,1,0,65229,,,,0,9f46ea376158844761232fd2a4ed15764e8929ed493a22a801130a0bcc840e9a
thread:issue:65230,issue #65230: RuntimeError: Trying to create tensor with negative dimension -1741885395: [-1741885395],stale/low urgency,2,stale/low urgency,1,0,65230,,,,0,1c5e2584160fd7d6e33a6c02c36aadac597a09801661e9129a68fbbaa2bea977
thread:issue:65253,issue #65253: TestCppExtensionJit's test_crash_handler fails for force_on_cpu job on GHA,stale/low urgency,2,stale/low urgency,1,0,65253,,,,0,bff1152a2a23c6ef1fd1ef0010f74622d3c44416018ee63da551a66f97a8fd14
thread:issue:65261,issue #65261: Add an `all_paths` method to PackageExporter,stale/low urgency,2,stale/low urgency,1,0,65261,,,,0,ac89de73e10ef3c1762aed2cfdb306240dbc4215c9efe8d32c6416e45e69c20a
thread:issue:65301,issue #65301: Anomaly detection: Error detected in CudnnRnnBackward0,stale/low urgency,2,stale/low urgency,1,0,65301,,,,0,b84bbddb24b912bd0bdc2e34cca5ede78bc40cb6ab74cfbc67524e6a65afbbb7
thread:issue:65306,issue #65306: [numpy] Add `iscomplexobj` and `isrealobj`,stale/low urgency,2,stale/low urgency,1,0,65306,,,,0,8876630829cfe836665ada5e9b8598686c663ba1fe9d9dacbc87254532691b8e
thread:issue:65313,issue #65313: openSUSE Build: CMake Generate step failed. Build files cannot be regenerated correctly.,stale/low urgency,2,stale/low urgency,1,0,65313,,,,0,938fe023b475f0ebe93394bebc3e7974db7cafd4206c81fd159c68e8ff9c9e6a
thread:issue:65329,issue #65329: How can I re-weight a sample based on both class weights and instance weights?,stale/low urgency,2,stale/low urgency,1,0,65329,,,,0,2b54efcdece73fc68857ea1c0b85a28ddd96b3f1d1ba8b5a9b85ecf094ac8e96
thread:issue:65339,issue #65339: Make it easier to accurately reflect storage/view relationships in meta/wrapper tensors,stale/low urgency,2,stale/low urgency,1,0,65339,,,,0,f96eb354e9419c33c6a70e418813ce30734cd5107d00499a0f1f659a79c58710
thread:issue:65375,issue #65375: Make python_error use c10/util/Exception.h C++ backtrace machinery,stale/low urgency,2,stale/low urgency,1,0,65375,,,,0,9d189fc916afaf5519d0cb842acb2e6e84e3961c93cfdfb82e8a5d81373139a1
thread:issue:65386,issue #65386: [LTC] Setup a subset of TorchBench for the current implementation,stale/low urgency,2,stale/low urgency,1,0,65386,,,,0,649de0608244960e2499ff36b3f27e18fecdda1126733f59efcf2ef4bbc207be
thread:issue:65400,issue #65400: torch.sparse.sum on scalar sparse tensor fails when dim is specified,stale/low urgency,2,stale/low urgency,1,0,65400,,,,0,05c47e9170f2a1737047c06497d22e7e37faa061975274dee3a9fc30d152dbff
thread:issue:65418,issue #65418: Sending tensors from short-lived multiprocessing process fails on Linux,stale/low urgency,2,stale/low urgency,1,0,65418,,,,0,9c3cbd36d9f303df36e95d39befe44e133ea436d96ea77a9b7899c2135d6493e
thread:issue:6542,"issue #6542: 【Train issue with caffe2 detectron】Aborted at 1523501813 (unix time) try ""date -d @1523501813"" if you are using GNU date *** PC: @ 0x7fa35103733a (unknown)",stale/low urgency,2,stale/low urgency,1,0,6542,,,,0,e781540d9dd85475180e08acd27d8e000927221368f5e552ddd872a81383f02b
thread:issue:65423,issue #65423: Enable is_contiguous slowpath for torch_dispatch,stale/low urgency,2,stale/low urgency,1,0,65423,,,,0,5abbfc497864e64eb47ddc541e0bc3e8d952166f61f64a80cdefab5677729c6d
thread:issue:65426,issue #65426: Add a `__torch_pre_dispatch__` that will live at the top of the dispatcher,stale/low urgency,2,stale/low urgency,1,0,65426,,,,0,73001aadf010630a141f2fa8f3a797de00371c3608051e2f6fcadef692b43918
thread:issue:65428,issue #65428: module 'torch.distributed' has no attribute 'optim',stale/low urgency,2,stale/low urgency,1,0,65428,,,,0,3a72dbe08bb6880b4f2df3d46e8283f27e4ecee12660296f1c6e387703a08081
thread:issue:65436,issue #65436: Make DDP uneven inputs work with custom buffer reduction,stale/low urgency,2,stale/low urgency,1,0,65436,,,,0,a2e714490d2730887fb895be44b880ee764f9bcbd9ebc441a46a76f40cedd48a
thread:issue:65446,issue #65446: libtorch compile problem. How to get the correct protobuf version? what PROTOBUF_VERSION <3011000 and 3011004 <PROTOBUF_MIN_PROTOC_VERSION?,stale/low urgency,2,stale/low urgency,1,0,65446,,,,0,26ce06fbfd6e3db60bdc7e53474b7f089ec3773ad191b3b349f41caea8d02ecb
thread:issue:65447,issue #65447: No support for torch.trace on CPU for float16 tensors,stale/low urgency,2,stale/low urgency,1,0,65447,,,,0,4f23779f5389bf68383a67c94adc79fcdc12255a538dc04e36064dcebea70c2d
thread:issue:65448,issue #65448: Magma : Intel MKL Errors,stale/low urgency,2,stale/low urgency,1,0,65448,,,,0,1e8010002222e3910af1909e9db38aaf60cbf60c7578ea5c70b29eee4cf0d227
thread:issue:65475,issue #65475: building pytorch from source without conda,stale/low urgency,2,stale/low urgency,1,0,65475,,,,0,322ed49cf7037d291909a67de33c3f0024d2a3bae648f8de7e55eee1d1d7a82d
thread:issue:65494,issue #65494: Feature request: add bool dtype support to CPU abs(),stale/low urgency,2,stale/low urgency,1,0,65494,,,,0,24f42036823ecf1292a3834e3375a1730e682d918752493f2136511105908e3c
thread:issue:65510,issue #65510: nn.conv1d padding='same',stale/low urgency,2,stale/low urgency,1,0,65510,,,,0,4042d30a6e4e7a618a3711fc317fb4eecad206b700fc5008cccda7964efc5347
thread:issue:65512,issue #65512: Performance problems of eigh operator on CPU,stale/low urgency,2,stale/low urgency,1,0,65512,,,,0,e24571eb9ba6d7829e8c529c5fd6e721de23e6a630b044aa55bf2f4f3e69adbe
thread:issue:6552,issue #6552: [Caffe2] - Run on different Cuda streams,stale/low urgency,2,stale/low urgency,1,0,6552,,,,0,347d76945b0f33513e3520778a219e3f71135c82b1274287e96171e3de3ebd2e
thread:issue:65520,issue #65520: Improve input checking for running_var of nn.functional.batch_norm on CPU/GPU,stale/low urgency,2,stale/low urgency,1,0,65520,,,,0,5a4f2da58a787998b34a0e43c43d84031092cbfe1b14eb119568ecb0ac0d42c3
thread:issue:65522,"issue #65522: pytorch framework tests using make_tensor hangs with pytest's boxed exec option ""--forked""",stale/low urgency,2,stale/low urgency,1,0,65522,,,,0,240f8cf7b87b7a8f10c3a2e5aa9f97b5ef73924ad2d110bd4cf13072d6547cbb
thread:issue:65526,issue #65526: [docs] Better explain signature of toch.randn,stale/low urgency,2,stale/low urgency,1,0,65526,,,,0,1f137f872c097b472242567ce008044cc2d7c9c01ca128e89ee9d74224803459
thread:issue:65528,issue #65528: I cannot use x.to(GPU) or x.cuda(GPU),stale/low urgency,2,stale/low urgency,1,0,65528,,,,0,14bed98692388aedd55f80616e5d2b311ab0cc13ea4b773f22d7a229c8df8cd7
thread:issue:65536,issue #65536: Pass backward flags such as retain_graph to context of custom torch.autograd.Function,stale/low urgency,2,stale/low urgency,1,0,65536,,,,0,edc7fde8a22d1e021652a8116ea64b8870cfeb8665f0c4a3a6841bb8bc0ed80d
thread:issue:65569,issue #65569: Support 'meta' device for 'shard_parameter' API,stale/low urgency,2,stale/low urgency,1,0,65569,,,,0,71835ce2672042e1d0cf725c282e4642b58cf940ef3859cc3ceb82cd08ac74f0
thread:issue:65572,issue #65572: PyTorch setup.py should confirm with user if the user changes build flags,stale/low urgency,2,stale/low urgency,1,0,65572,,,,0,fa574356ef12285995ca83c6a497cb5a71718499466dea9fc162daa99d54f0c6
thread:issue:65576,issue #65576: [LTC] Code-gen IRs and TorchScript lowerings needed by running full TorchBench,stale/low urgency,2,stale/low urgency,1,0,65576,,,,0,6449d8d0c1804216af7a62c4334018f35d86abc04ff87dfb2114883f8f68336a
thread:issue:65577,issue #65577: Support cat() for meta tensors,stale/low urgency,2,stale/low urgency,1,0,65577,,,,0,7ec0b14ba7c2371f9c9fb8778aa1f9b6db9f93e01906ebee34975763739bd867
thread:issue:65604,issue #65604: pytorch serializes entire tensor when you try to pickle a slice,stale/low urgency,2,stale/low urgency,1,0,65604,,,,0,7111c4235d6064504be304cf2543fd237ccb08e6bbf21284d4d92c608e63c2f3
thread:issue:65619,issue #65619: Integration of __torch_dispatch__ with masked tensor semantics,stale/low urgency,2,stale/low urgency,1,0,65619,,,,0,679e8805358305bb5ae993e6c1ad2b7246f81e11df36eb637e34f3f1b270707c
thread:issue:65622,issue #65622: Transformer modules should not hard-code the activation function,stale/low urgency,2,stale/low urgency,1,0,65622,,,,0,539c2e335e1b0494fe5393c5da88f4736a1ac96049f2cb1d33a4ba14bc33bb3e
thread:issue:65628,issue #65628: Add torch.distributed.shard package,stale/low urgency,2,stale/low urgency,1,0,65628,,,,0,b91f57528a7b8add916350c4cb4cf7df63d783b450a4371db8c132bed3ea1153
thread:issue:65637,issue #65637: Test ZeRO on gloo backend for GPU,stale/low urgency,2,stale/low urgency,1,0,65637,,,,0,b76c82bf26c236df140531f9b3f3ee03fd7e0db003c729aa3ec55f1299c556ce
thread:issue:65640,issue #65640: [LTC] Memory growing up during training until OOM,stale/low urgency,2,stale/low urgency,1,0,65640,,,,0,1e4220e9c3aa3218820b43683ced079388fd984537607ad21a5e6e169811831d
thread:issue:65646,issue #65646: [LTC] Building unit tests on Debian results in double defined _GLIBCXX_USE_CXX11_ABI flag,stale/low urgency,2,stale/low urgency,1,0,65646,,,,0,15192a82d84a9d46a08caf78f1b26fa04043d703ee057805c22a08dbef2b460a
thread:issue:65650,issue #65650: Does ZeroRedundancyOptimizer support re-split partition on load_state_dict?,stale/low urgency,2,stale/low urgency,1,0,65650,,,,0,cc34c56aea48ae1046e1d14cef035df00e8b4e42b7e2745850a969f221735345
thread:issue:65657,issue #65657: Transpose channels when exporting to ONNX,stale/low urgency,2,stale/low urgency,1,0,65657,,,,0,948027621cccdfb3cfb54d86de4262c5bf97a7a975bcd3a4e3e2c30752e2e540
thread:issue:65661,issue #65661: [BUG] ConvTranspose with out_channels=1,stale/low urgency,2,stale/low urgency,1,0,65661,,,,0,25519e846e6fe0fc5463edcd8cab339220448e04dbbeb7ee3a630d2d4694362f
thread:issue:65663,issue #65663: Model Evaluation Returns Nan Values Sometimes for the Same Input,stale/low urgency,2,stale/low urgency,1,0,65663,,,,0,568c683f42467691be49191509433b4a137aaf03f2f48ffa37658be55d13ef14
thread:issue:65664,issue #65664: C10dProcessGroupSerialization.test_process_group_as_module_member is flaky,stale/low urgency,2,stale/low urgency,1,0,65664,,,,0,e070ccada5de692a76f54e7656b12578ac0bd9cae59cc9cdc001debf081965e5
thread:issue:65665,issue #65665: Pipe method for torch tensors,stale/low urgency,2,stale/low urgency,1,0,65665,,,,0,8715f659bb3c8986dcfe7c89f70eb7df5eaa75679d815faa6b15529dc56f4317
thread:issue:65670,issue #65670: Memory Leak in Distributed RPC nightly,stale/low urgency,2,stale/low urgency,1,0,65670,,,,0,9e20eaaa46eda677f8d3e1d1b7fb99ca7a275fcff1aa8f16c0067ba7ba0db4b4
thread:issue:65673,issue #65673: [libtorch]incorrect sigmoid result on arm chip(rk3326),stale/low urgency,2,stale/low urgency,1,0,65673,,,,0,6769d33b033256f6eda7d8d7921428fe676bb265af47e47a1770a41a120a0fc5
thread:issue:65683,issue #65683: `torch.nn.functional.l1_loss` fails gradgradcheck for complex inputs,stale/low urgency,2,stale/low urgency,1,0,65683,,,,0,b7110e7a2376ea2dac4eb8fa86c3295e03b3a7eba7951f23fee8fcf16b92d25f
thread:issue:65687,"issue #65687: repeat_interleave changes memory format from ""channels last"" to ""contiguous""",stale/low urgency,2,stale/low urgency,1,0,65687,,,,0,52d14ff11181e05a4e33defab1763a6d542708e444cee52ddebaf84f8d8023d2
thread:issue:65688,issue #65688: Parametrization silently disables RNN weight updates,stale/low urgency,2,stale/low urgency,1,0,65688,,,,0,14ac32cfe25eddcbae462edda812ba84080082d8d6ecff474e73a80a4f092cd1
thread:issue:65694,issue #65694: Plain Tensor serialization does not save the content of `__dict__`.,stale/low urgency,2,stale/low urgency,1,0,65694,,,,0,89d08ffb78103fa772c10ed89e93777f2e1265f9737895766291c49125f78b58
thread:issue:65698,issue #65698: torch.onnx.export error,stale/low urgency,2,stale/low urgency,1,0,65698,,,,0,73ea56b27feb5c7e9c1a7a86b760d19db83bda7f49719806529f54e219617634
thread:issue:65702,issue #65702: Indicate support for more general use of `pos_weight` in binary cross entropy,stale/low urgency,2,stale/low urgency,1,0,65702,,,,0,2ce982c1ac21c82932115142cecd77c6c335bfdc125ca73cf0543358e1aaf794
thread:issue:65718,issue #65718: Migrate current Windows CI scripts off of batch,stale/low urgency,2,stale/low urgency,1,0,65718,,,,0,540eed2fc96caa03e84e39ed5756d24cd96028dbc8b10d1632f13aeaec7c57ba
thread:issue:65739,issue #65739: Sobol state API,stale/low urgency,2,stale/low urgency,1,0,65739,,,,0,40a08e351f4653732c5799c6667bfd46553bf84b871680b7e5729ccf7b4f6bb5
thread:issue:65752,issue #65752: Deprecate torch.cross with optional dim,stale/low urgency,2,stale/low urgency,1,0,65752,,,,0,11af04eec297b648e648b6ed46c11c419a42d60f35a9c2ab4ae6b811492fdfd3
thread:issue:65753,issue #65753: [POC] Convolution-relu fusion via torch function and one-step lazy evaluation,stale/low urgency,2,stale/low urgency,1,0,65753,,,,0,fff4aa9dc4b3092fca3b86ca5e0907c2d0a087cbe00be612b6c05481d156780c
thread:issue:65774,issue #65774: Non-persistent modules,stale/low urgency,2,stale/low urgency,1,0,65774,,,,0,ceb37cab6af863dd756a12232ce20e568fd664a4aa9545c3f8cf7a9f93536ec9
thread:issue:65811,"issue #65811: Beta distribution yields wrong results for a,b >1 and x=0 or x=1",stale/low urgency,2,stale/low urgency,1,0,65811,,,,0,5bed4eccd264e56980c013dc595e916e4cdd7834d87c6c90786b91c1e8c72aff
thread:issue:65825,issue #65825: Incorrect reference to previous kernel in the warning displayed when overriding a previously registered kernel for the same operator and the same dispatch,stale/low urgency,2,stale/low urgency,1,0,65825,,,,0,8a7cd6650df8ee793df0042d86efd9ca9ff6162bf541d1ecf9f085399cbdf5d3
thread:issue:65837,issue #65837: NVFuser Examples to Help with Integration with Lazy Tensor Core,stale/low urgency,2,stale/low urgency,1,0,65837,,,,0,99ba9d8477d379e42f1d94d6fa27b27da73777dc1508ad0309645d5f90872ce2
thread:issue:65881,issue #65881: torch.hub.load can confuse external python package with local python package.,stale/low urgency,2,stale/low urgency,1,0,65881,,,,0,6a0317ce6328a5d8368dbe199977708a14458f8f60d7aa1272d83e1616bde183
thread:issue:65890,issue #65890: `pad_packed_sequence` is not exactly the inverse of `pack_padded_sequence`,stale/low urgency,2,stale/low urgency,1,0,65890,,,,0,2c3ee97504b4b96f5e0b25c467b2968a7d7ca8cc2717ce129ac740ef76626f79
thread:issue:65910,issue #65910: Deprecation: Remove nn.functional.sigmoid,stale/low urgency,2,stale/low urgency,1,0,65910,,,,0,8f75a81f25df9f518fe4e5ea5972c8fbc60baf1d37dc1ac922a5234a4ea9b15c
thread:issue:65936,issue #65936: [POLL][RFC] DataParallel Deprecation,stale/low urgency,2,stale/low urgency,1,0,65936,,,,0,77689ee240233d86936c28092fdec0c72cff4a7b656fe0a3b5aad17eab744757
thread:issue:65949,issue #65949: torch.unique signature is not descriptive,stale/low urgency,2,stale/low urgency,1,0,65949,,,,0,e0e1e4bc26edb6559dc997cef0d7be50f186ec29fd27f59cbe0e2f920a6c938f
thread:issue:65950,"issue #65950: Storage `is` should work if the storages ""are the same""",stale/low urgency,2,stale/low urgency,1,0,65950,,,,0,c3cd231d2efcd731e310c1c908b050c9782c2b3e6ef1c81bb9b6f95a345756b1
thread:issue:65984,issue #65984: libtorch: collate_fn equivalent,stale/low urgency,2,stale/low urgency,1,0,65984,,,,0,2c3517f1248af41716b5bbcf57cd272c2a9abe77d428a3b6eb04ec681bd1c9db
thread:issue:66026,issue #66026: `__torch_dispatch__` can result in returning `None` for an op that should return Tensors.,stale/low urgency,2,stale/low urgency,1,0,66026,,,,0,8a8c4705fe42ba92cd2256abbcefdbf7824ae0560747ae53ebe51c29b67ff5b0
thread:issue:66066,issue #66066: Finishing OpInfos: test_autograd.py,stale/low urgency,2,stale/low urgency,1,0,66066,,,,0,4b6a056d3755290dfbb20d33bb6513aee210522f5cb66874f79b109388057545
thread:issue:66076,issue #66076: Bad Input to torch.nn.NLLLoss causes CUDA Error,stale/low urgency,2,stale/low urgency,1,0,66076,,,,0,186cfa3755a8680259ed6ce22929a1fa6ddc8b1691648f79da6255ca4e7fe4cd
thread:issue:66083,issue #66083: [feature request] Allow broadcasting for F.huber_loss and potentially other losses,stale/low urgency,2,stale/low urgency,1,0,66083,,,,0,b9a6c49c43fc5e4ac33b3b07e002ecc7400e87fddc2905ffce4622603be5eb38
thread:issue:66116,issue #66116: Removal of BufferedShuffleDataset,stale/low urgency,2,stale/low urgency,1,0,66116,,,,0,602152679f6812fa684743344d49928565ba39f152b7f3126f8a0aec7755ff00
thread:issue:6613,issue #6613: [caffe2] benchmark performance for different operators,stale/low urgency,2,stale/low urgency,1,0,6613,,,,0,12dc5854d198ad13d89fc7971e21080ae817d7b3d2e9d9415c07a2a13cd19803
thread:issue:66162,issue #66162: Cuda Out of Memory - MMDMatic,stale/low urgency,2,stale/low urgency,1,0,66162,,,,0,b004493b2361b48978c0bc1d000a05ee29dde8fbf6b5d5573a353a214bafb765
thread:issue:6618,issue #6618: [Caffe2] CUDNN_STATUS_BAD_PARAM Error with the LRN layer while trying to run the code using CUDA. The training works fine on CPU,stale/low urgency,2,stale/low urgency,1,0,6618,,,,0,711b2262dcb33f0f633938f527b974ab4aec929f2df68a49b9fa4db76b069613
thread:issue:66196,issue #66196: Specifying USE_VULKAN=0 in launching build_android.sh does not disable Vulkan,stale/low urgency,2,stale/low urgency,1,0,66196,,,,0,211cecbc83b5a45740f7e2086be6ecd708adc0ce19d80fd85309fc22c640ed1f
thread:issue:66197,issue #66197: `torch.fx.replace_pattern` doesn't work with untraceable wrapped functions,stale/low urgency,2,stale/low urgency,1,0,66197,,,,0,45712c1fc7beb7de6f2b0fcba253f69ecade88c08904a09fdd6a4ab6424e9218
thread:issue:66199,issue #66199: Move the `queue_callback()` API out of `Variable._execution_engine` and into a public API,stale/low urgency,2,stale/low urgency,1,0,66199,,,,0,d3fe5a87e33456ea4ed5335206328734bdc4c9d3b79d8e481ae9d52ce76ab143
thread:issue:66200,issue #66200: An exported onnx model can't reduce on dim with value of 0 if 'keepdims' is false,stale/low urgency,2,stale/low urgency,1,0,66200,,,,0,91e01ca16629cd3d48329b77b872b457da829af22e38e9230508cf01a0249e93
thread:issue:66208,issue #66208: Add LTC-TS support for ops used in SGD optimizer,stale/low urgency,2,stale/low urgency,1,0,66208,,,,0,023eaadcff078e31e0467c8ba6d50d99ddc23519244e5d78a9ca23ccdf26941c
thread:issue:66223,issue #66223: Data Loader tests hang when run in ASAN test job,stale/low urgency,2,stale/low urgency,1,0,66223,,,,0,f0130fc391874a213add45d66f8d7a4e53ad8c0eae0a00de6c6dc7985526e8ff
thread:issue:66224,issue #66224: [feature request][quant] Support FakeQuant qconfigs in `test_module_init`,stale/low urgency,2,stale/low urgency,1,0,66224,,,,0,5eb3c919ffb85029a27d3c91f843c9b48700fead24448df5b1713cf8bcaccbd9
thread:issue:66249,issue #66249: Latency issue with torch.sin,stale/low urgency,2,stale/low urgency,1,0,66249,,,,0,f30e7149a28dc6e1f5bd9250405d1840e25926f7e62b8efb52e2cfdefc63d542
thread:issue:66251,issue #66251: Should the drop_last parameter of Dataloader be mutually exclusive with the batch samplers?,stale/low urgency,2,stale/low urgency,1,0,66251,,,,0,f98635a16842e999a7a4827f0b22a6f5e7da65ef72edee2d14a475ebfa72bd11
thread:issue:66271,issue #66271: 1.10.0-rc1 fails to compile with gcc 6.4.0,stale/low urgency,2,stale/low urgency,1,0,66271,,,,0,df099b4c3ee1031f553559b769d051e3fe2ba669a4e1c5b19b37797c37c35c1a
thread:issue:66284,issue #66284: Memory leak issue with pytorch_java_only 1.9.0 and libtorch 1.9.0+cpu,stale/low urgency,2,stale/low urgency,1,0,66284,,,,0,c56f47ea057ef6f042344736efd8c84d6ba32039f4803caaca411c3b7e0e51ea
thread:issue:66305,issue #66305: Modify convolution kernels for ops,stale/low urgency,2,stale/low urgency,1,0,66305,,,,0,f78da20ed69514c0f6734528fb27919990ac9303c521acdfd8f668f3b386bcd8
thread:issue:66341,issue #66341: Invalid handling of out args with type Tensor[] in native_functions.yaml,stale/low urgency,2,stale/low urgency,1,0,66341,,,,0,fbf30a45fdf2cb768fd35f0dfcd4dafd3ea32a6ae73dd2919554d33e39a7e7df
thread:issue:66357,issue #66357: OpInfos disabled for batched forward grad computation,stale/low urgency,2,stale/low urgency,1,0,66357,,,,0,79a4a778ee04a5fbda2fd09d487a3ba8e8f905b93134132ff665eea0f9762e6f
thread:issue:66359,issue #66359: question on installing GPU-enabled Pytorch,stale/low urgency,2,stale/low urgency,1,0,66359,,,,0,9132df0031a0c8c8a4dae9460c63d8ca4ed65a7f6e00b8fcb66e1495b2d1f629
thread:issue:66364,issue #66364: Hope to obtain authorization for Chinese translation of official tutorials.,stale/low urgency,2,stale/low urgency,1,0,66364,,,,0,17f12c03b88dfc6877cb6eef18238f309b6c7dd1f05ff7dbafd46f5786f95e2f
thread:issue:66366,issue #66366: Feature Request: Add constant padding_mode for grid_sample,stale/low urgency,2,stale/low urgency,1,0,66366,,,,0,f4fb99706d7f22025f7dfca96169e512c8e2753b12670a2ccf131dfd0a7528bc
thread:issue:66403,issue #66403: [opinfo] Confusing interface for `ops` decorator,stale/low urgency,2,stale/low urgency,1,0,66403,,,,0,90245edad941086801d1e6dd9555fc26e3da67248b5a7d4627a13974a927f7a2
thread:issue:66410,issue #66410: CIFAR10 doesn't work on M1 MacBook,stale/low urgency,2,stale/low urgency,1,0,66410,,,,0,cc73831ffcd842ceb4aeef86dd2d3c2ae04670397c2d13e9d002c6a37adc0ffa
thread:issue:66460,issue #66460: Inconsistencies in LSTM outputs when processing sequence stepwise on CPU,stale/low urgency,2,stale/low urgency,1,0,66460,,,,0,c52244d561647a10c9a8f62bd62ec4bbcf09d040a54670ac9fb0db3d1b49df1b
thread:issue:66473,issue #66473: nccl comm init needs a global barrier.,stale/low urgency,2,stale/low urgency,1,0,66473,,,,0,9b5362f2442b79dfdcf11c8c379b6c9e678743d2179936df027b00ac4c82efed
thread:issue:66476,"issue #66476: 0INTERNAL ASSERT FAILED, We don't have an op for aten::eq but it isn't a special case",stale/low urgency,2,stale/low urgency,1,0,66476,,,,0,9bf8715e7813349f9826f6178e08ec390f1d0c96a8ecf573b83f1021e0fda493
thread:issue:66482,issue #66482: BAR1 memory of GPU is not released when main process is killed.,stale/low urgency,2,stale/low urgency,1,0,66482,,,,0,2364d9d69197d3f0a44ad5851e89c5bbeb564a1b4091b39e5ea1b5cd1c69b5cd
thread:issue:66489,issue #66489: The uniform operator param names in the C++ impl use python keywords,stale/low urgency,2,stale/low urgency,1,0,66489,,,,0,e236e2e8a9bc8d502e377837f3a7970c474d7964f9e347b338758b97f1548731
thread:issue:66563,issue #66563: Give a better error message when REGISTER_DISPATCH is used in improper context,stale/low urgency,2,stale/low urgency,1,0,66563,,,,0,8e8a35b171627fcc9ae4012a7fd5b32166f2682b3f92d4f86bcac3768701fd05
thread:issue:66565,issue #66565: Bazel target all_tests improperly reports failures on CPU-only (non-CUDA) build,stale/low urgency,2,stale/low urgency,1,0,66565,,,,0,6d5398f7cf71ea58f93fdefa79fe733b92df637ec54e721dcd7369ab05d411c5
thread:issue:66573,issue #66573: tools/amd_build/build_amd.py should fail if any file fails to write,stale/low urgency,2,stale/low urgency,1,0,66573,,,,0,759069af88361ef458399c1c567f9a8d6ba3fef6f75ac61ec5e3ff239749adeb
thread:issue:6660,issue #6660: [caffe2] GANs,stale/low urgency,2,stale/low urgency,1,0,6660,,,,0,fe93a1f028d01ac83afb941eef35c3643c731805bfc391e2ee76c3475e3a289b
thread:issue:66678,issue #66678: Adding modern linear transformer variants to `transformer` module,stale/low urgency,2,stale/low urgency,1,0,66678,,,,0,63ce559f0c570d2a6c1ef9dd6d306c4f61b7a943939d1d2d299ebc6136df6e85
thread:issue:66755,issue #66755: Domain Transformation APIs for LibTorch and LibTorch-Lite,stale/low urgency,2,stale/low urgency,1,0,66755,,,,0,519d0c2db6bb8f808b5ca2080ffcad27e0a2413dc7354409d9af75aa84c21d30
thread:issue:66775,issue #66775: Feature request: FFT operations on Metal,stale/low urgency,2,stale/low urgency,1,0,66775,,,,0,63c668b5797c6585d9942817f0fba5c9556074e509bd3ad0dd35fd087b64e500
thread:issue:66787,issue #66787: loading large model not finished after 16 hours,stale/low urgency,2,stale/low urgency,1,0,66787,,,,0,66326d60770357cb52d2f03fe3bafabed7e551fe4b7fc7b59bee990869486153
thread:issue:66806,issue #66806: Missing doc for `torch.segment_reduce`,stale/low urgency,2,stale/low urgency,1,0,66806,,,,0,797f95804f5acaa5f393e1534615805154be4af86b7549c81a5495622e51dc36
thread:issue:66858,issue #66858: CUDAgraph error while capturing,stale/low urgency,2,stale/low urgency,1,0,66858,,,,0,fb699cb02bae3cb9421b036df8fb2d18ab30fdcf7cc5134e1c5c90661055aa48
thread:issue:66868,issue #66868: torch.triu behaves differently when diagonal out of range,stale/low urgency,2,stale/low urgency,1,0,66868,,,,0,5d503d8065afc7060da46f5e98a6ff49c6cfde1759b8db8759fbcf5d306091a6
thread:issue:66874,issue #66874: Error using _stateless version of Module,stale/low urgency,2,stale/low urgency,1,0,66874,,,,0,7de739d7835efa8dd4b43b764e50e68579ea32c7f5e1e8583960e30aada534c5
thread:issue:66875,"issue #66875: Make the evaluated value of function f(x) accessible from `torch.autograd.functional.jacobian(f,x)`",stale/low urgency,2,stale/low urgency,1,0,66875,,,,0,a541c31cde002f8c0b03f1d2971eac7454c74e13b537157e68328afb70baa929
thread:issue:66902,issue #66902: Modify Dr. CI so it could detect runner disconnection failures,stale/low urgency,2,stale/low urgency,1,0,66902,,,,0,59d951879bea1197c959256b87f92ddf0fa698c6f8899780bfab1ce107f583e2
thread:issue:66907,issue #66907: torch.chunk return type may not be documented correctly,stale/low urgency,2,stale/low urgency,1,0,66907,,,,0,ccec900191f8d6992036a45a1de1e9ace32f994e2243013eb4f8d7b2b5c2baf6
thread:issue:66921,issue #66921: readme not update,stale/low urgency,2,stale/low urgency,1,0,66921,,,,0,353565372a45ce0ce6bd82355ec99e99684f6f9313d0a2683dd03f707e3a3327
thread:issue:66924,issue #66924: [BUG] Inconsistent initialization on different machines (tensor.uniform_()),stale/low urgency,2,stale/low urgency,1,0,66924,,,,0,f1d808cb410cb01a9d3c81a44d33bfcd30aac1780190031636e52ab2e8c327ab
thread:issue:66936,issue #66936: Lifetime issues when recording external CUDA streams with the caching allocator,stale/low urgency,2,stale/low urgency,1,0,66936,,,,0,ed02bacbf4288e0b7351b4aa673b46e6a10de389801273365f044e10c259789f
thread:issue:66963,issue #66963: Significantly difference in execution time when convolution is run as nn.Conv2d and as nn.Sequential,stale/low urgency,2,stale/low urgency,1,0,66963,,,,0,c7bbe7e56d2f324722f19739204d5f725e826dc0cd349ee0ad384b8d0d881e88
thread:issue:66992,issue #66992: Bug? :Run torch.unique twice get different running time?,stale/low urgency,2,stale/low urgency,1,0,66992,,,,0,b132715fea37f3c2c5c065e542671f542cd4c2923edfefd77fe3965da2c6bbdd
thread:issue:67008,issue #67008: kwonly arguments without defaults don't work with test_overrides.py,stale/low urgency,2,stale/low urgency,1,0,67008,,,,0,d5af40c57fe5d944ced2d4be1f34ca142fec048983b8a81f244edee2bdc5cdba
thread:issue:67011,issue #67011: Importing numpy interacts with `tensor.sum` perf,stale/low urgency,2,stale/low urgency,1,0,67011,,,,0,ec8be281a35831f0f9a83125d01b51a0a9aec858a51a68b2a7f8733d90fa7f3a
thread:issue:67030,issue #67030: fx graph mode quantizing error,stale/low urgency,2,stale/low urgency,1,0,67030,,,,0,0ff279772519e019150df886b442f0d707731fb66ba05227d58d7e8d582c50b1
thread:issue:67071,issue #67071: Forward results vary depending on batch size on A100 machine,stale/low urgency,2,stale/low urgency,1,0,67071,,,,0,ae03f90fddcf72d5c45c9850c3259744c9589c8593c85e96037410dc3901b100
thread:issue:67092,issue #67092: Sparse matrix multiplication (torch.sparse.mm) NotImplementedError,stale/low urgency,2,stale/low urgency,1,0,67092,,,,0,29284d68d768701395413537c2d6e4f257efd86372d1eeb7f7922e2895d5f469
thread:issue:67136,issue #67136: addition of loss function RMSE in the torch.nn,stale/low urgency,2,stale/low urgency,1,0,67136,,,,0,c59853cf60d0d5fa6878d7eaf0679b13fe72436fb05c2d6141ad9d5e79315847
thread:issue:67158,issue #67158: Make streams used for NCCL operations configurable,stale/low urgency,2,stale/low urgency,1,0,67158,,,,0,6b2c20a20ff4c4179ba85ebe66f684a7f8c6f962cc0fcabfc951b06c67fc2045
thread:issue:67168,issue #67168: Multiply two named tensor causes RuntimeError,stale/low urgency,2,stale/low urgency,1,0,67168,,,,0,88c963c2955b6a5bc1b1eb7da0009fc7e443a1cdcede1c1bda3eebfbfca40f5e
thread:issue:67187,issue #67187: Expose `isDifferentiableType` to python,stale/low urgency,2,stale/low urgency,1,0,67187,,,,0,a80835a004922a9efc5e8f4810fbaffeada2eecb14d711a3e647711154acfe32
thread:issue:67208,issue #67208: TestForeachCUDA.test_binary_op_tensorlists_fastpath__foreach_add_cuda_bool and TestForeachCUDA.test_pointwise_op_fastpath__foreach_addcmul_cuda_uint8 fail intermittently,stale/low urgency,2,stale/low urgency,1,0,67208,,,,0,fc2039bfcca55d673b9831ca6f53996d95403df29609f9af099662f60371a68d
thread:issue:67240,issue #67240: Feature: Add derivative for channel_shuffle,stale/low urgency,2,stale/low urgency,1,0,67240,,,,0,cedb6f2baf0cb825dd079138a177648649270aaeb2bf50b32eb50cc578ce08fb
thread:issue:67279,issue #67279: register_dispatch_key should provide a way to tell if a native function has a no-op meta kernel,stale/low urgency,2,stale/low urgency,1,0,67279,,,,0,5b3fe19680f7c33f569571d64ff325d0eed0884fcc4044ad80c90525522c2459
thread:issue:67284,issue #67284: Failed to build with master.,stale/low urgency,2,stale/low urgency,1,0,67284,,,,0,154990b67ba716f90b75a660eb8ee2bfe8f8d1c029b95bd57a1c45d70c503eab
thread:issue:67312,issue #67312: ProcessGroupNCCLTest.testSequenceNumInit is failing on 4xlarge,stale/low urgency,2,stale/low urgency,1,0,67312,,,,0,a171839fb574813f1c73d82f84cced546748cd4b67f639affc7dc452f100b8bb
thread:issue:67333,issue #67333: Feature request: Complex Number Support for Special Functions,stale/low urgency,2,stale/low urgency,1,0,67333,,,,0,6a870c39c1d890e2873f5e93e5c0b0843142fa9559f5dd17c5e0511914046b77
thread:issue:67350,issue #67350: [FX] autowrap_functions doesn't work for module-scoped functions,stale/low urgency,2,stale/low urgency,1,0,67350,,,,0,463f77e3ee6e9c60e08c36aad77d4cf3fdd233be906008d90383b6b140a41ffc
thread:issue:67364,issue #67364: [FX] FX torchvision tests fail with torchvision 0.10,stale/low urgency,2,stale/low urgency,1,0,67364,,,,0,109598689e5248cf95b454795cee3be1ae74abe52eca2073ed159db72ae10c64
thread:issue:67377,issue #67377: [LTC] Replace mv with addmv in ts_native_functions.yaml,stale/low urgency,2,stale/low urgency,1,0,67377,,,,0,e681c8073f0d7a375078c0d17333571905f653b842581178b941cdef6cb821d9
thread:issue:67387,issue #67387: Missing doc for torch.autograd functions,stale/low urgency,2,stale/low urgency,1,0,67387,,,,0,fb19cedaadfde7b7c3fa68d2d08cc84344f28aa3d3dab632bdaa6a0b81051c0d
thread:issue:67388,issue #67388: Missing doc for torch.cuda functions,stale/low urgency,2,stale/low urgency,1,0,67388,,,,0,b1eb7354357a75874034242bd48ebe2f5d293a1a0875949483ceb2947dbf2251
thread:issue:67390,issue #67390: Missing doc for torch.distributions functions,stale/low urgency,2,stale/low urgency,1,0,67390,,,,0,62789b3c4e07592fbf464cd35d0d7b41d2e48c9d2c9e3b33af68d4aff6a3d2db
thread:issue:67402,issue #67402: [Documentation] Incomplete FX module,stale/low urgency,2,stale/low urgency,1,0,67402,,,,0,70bf287c7103186959e15745f0404b66904f1466a1746f3980d51e6f050cd274
thread:issue:67439,issue #67439: Exception in thread when using dataloader,stale/low urgency,2,stale/low urgency,1,0,67439,,,,0,11c5ad608bf46c8c6ea32712449646914b90fd3b45c1da2eaf61d101d65d6448
thread:issue:67457,"issue #67457: Reductions on zero-size dims: 1) by accepting a custom default value, 2) if tensor has another non-reduced zero-size dim",stale/low urgency,2,stale/low urgency,1,0,67457,,,,0,fcfba3424870ccae94205456653355bb2af91f258740993a46f43c8555a95dd8
thread:issue:67463,issue #67463: `test_forward_mode_AD` hangs for `nn.functional.cosine_embedding_loss`,stale/low urgency,2,stale/low urgency,1,0,67463,,,,0,b61f57394818f595f08d604775216228aca08111d03d4078f6ebfe4674c1685d
thread:issue:67470,"issue #67470: integer (and possibly float16) matmul fails test_noncontiguous_samples on CPU (and all backward dtypes, too?)",stale/low urgency,2,stale/low urgency,1,0,67470,,,,0,d34b7d806efffa623d31ad742b51e18a7e82f43d74525ac94f7e862aa50a5f93
thread:issue:67481,issue #67481: Batchnorm2D does behaves differently with different shapes when batch_size=1,stale/low urgency,2,stale/low urgency,1,0,67481,,,,0,e9f80ed928efc1e795dd37e7a19ec0b008e77df6acb951925a32d4cfb0aee225
thread:issue:67482,issue #67482: Improve docs coverage testing,stale/low urgency,2,stale/low urgency,1,0,67482,,,,0,03c8542cd14b0cc809c57c2f9be8ead4f226f1e01b8108569b1026ca2504f06e
thread:issue:67493,issue #67493: XLA not being tested in TestAutogradDeviceType,stale/low urgency,2,stale/low urgency,1,0,67493,,,,0,1bf2b6946777c6e6fbb0396943661e2217b6ec8ddc058b7298de313477e07d73
thread:issue:67535,issue #67535: [LTC] `UncachedCompile` numbers increases for several iterations until fixed,stale/low urgency,2,stale/low urgency,1,0,67535,,,,0,530975441d23f1e9e13acc5869705ed5716ec928d3a361b3e66c63bdc018ab84
thread:issue:67540,issue #67540: LTC asynchronous executor is not synced with python code and ignores context manager.,stale/low urgency,2,stale/low urgency,1,0,67540,,,,0,313328213198547d09d5d4b8a23afc29127e86b615069ed1d52da45b0c2a378f
thread:issue:67543,issue #67543: [FX] Support `int` and similar variants in FX,stale/low urgency,2,stale/low urgency,1,0,67543,,,,0,abcca2d4b8ae4d40988b03e6f6c2de8433048e810da7a43a3462b34e56fac827
thread:issue:67565,issue #67565: Adding a new kwarg to a torch.nn.functional function breaks FC for JIT,stale/low urgency,2,stale/low urgency,1,0,67565,,,,0,c47c5521f3234c54c505be3b069698798dcb3fda799434cf33cc9a66f9574598
thread:issue:67571,issue #67571: [discussion] Have default size/shape equal to a scalar,stale/low urgency,2,stale/low urgency,1,0,67571,,,,0,82b12b7bc6cf536bdec67a0119f44bc8686665ee878bc4ee7af6d0e6d3414bde
thread:issue:67594,issue #67594: Feature request: set_grad_enabled(*mods_or_params) as a safe context manager,stale/low urgency,2,stale/low urgency,1,0,67594,,,,0,84ed2976025dc6df1638abfdc6e666cff85ccc4d6ae0a72eaf4e0bdf73a5f309
thread:issue:67597,issue #67597: torch.kthvalue (and maybe torch.quantile) should support argument descending/largest (=True) like topk/sort,stale/low urgency,2,stale/low urgency,1,0,67597,,,,0,fe3c3b0a88b956acdbb5da63e4bdc0b1977050aa4787275459433848066075e9
thread:issue:67598,issue #67598: unnecessary warning with autocast,stale/low urgency,2,stale/low urgency,1,0,67598,,,,0,ad63b98eead34e16480a5c67a63e19191bab08fc40f3c97cafb4167eee6ef388
thread:issue:6760,issue #6760: [Feature request] LayerNormLSTMCell and LayerNormLSTM,stale/low urgency,2,stale/low urgency,1,0,6760,,,,0,6421f34d90b93fd297d0e9c0e75ac29e841a37cd34719526e526bdf44612b93d
thread:issue:67602,issue #67602: [feature request] Add no-op set_epoch on Sampler class,stale/low urgency,2,stale/low urgency,1,0,67602,,,,0,c172a08f0234e65610d0e98af4c97519764c2c4c6c120b00be88994de593368c
thread:issue:67642,issue #67642: Is there a pre-built pytorch package for CUDA 10.1?,stale/low urgency,2,stale/low urgency,1,0,67642,,,,0,b4aa21eb6f24608c154fed074f39ee04df4c3b6d09c7a7b12712bab44c385c3c
thread:issue:67651,issue #67651: MultiHeadAttention in quantizable seems incorrect with batch_first=True,stale/low urgency,2,stale/low urgency,1,0,67651,,,,0,52ddc50e3c8d317a38d0d4c79955a6d23f354197d8fb2d64e4ac21bae4c4e081
thread:issue:67653,"issue #67653: TSAN issue in autograd ""set_next_edges""",stale/low urgency,2,stale/low urgency,1,0,67653,,,,0,ef9be42edb021696e358f9bab88465515e4219016b3171f1319dec724f4a1d22
thread:issue:67662,"issue #67662: ""self.graph.owning_module not set for purity check"" error when trying to remove a node from torch.fx.graph",stale/low urgency,2,stale/low urgency,1,0,67662,,,,0,a78fdfa6c735612e29b5bb5f326efcfe58b2a790a2df52ceee520dcb8aeab02d
thread:issue:67680,issue #67680: OOM with a lot of GPU memory left,stale/low urgency,2,stale/low urgency,1,0,67680,,,,0,cfe06fc0b79175cb9da9020df3c9fb0e34b9b34222df2d6e98c44b0341d09acd
thread:issue:67682,issue #67682: Feature request: bfloat16 support for CUDA matmuls,stale/low urgency,2,stale/low urgency,1,0,67682,,,,0,3c4d4172daf9c27bd80ea4c5486591c79bec3fdef594163ff91eed628e04985b
thread:issue:67684,issue #67684: [feature request] torch.clamp on BoolTensors,stale/low urgency,2,stale/low urgency,1,0,67684,,,,0,79c1d66080660a7cc523cda5fe47e682c9856bb6dede420a2702618fe5b0ad2c
thread:issue:67687,"issue #67687: Trying to quantize and save a pre-trained deberta model, where i get a runtime error",stale/low urgency,2,stale/low urgency,1,0,67687,,,,0,641f10fbee7bd9c3930bf1575ff23377ac5f87d04eaca4570dd676af82ecbefe
thread:issue:67745,issue #67745: A `sieve` operation for separating values of a tensor into two tensors based on a condition.,stale/low urgency,2,stale/low urgency,1,0,67745,,,,0,8081c0d048c703520c356518a82a6f83b46c4ce93d77da5d2711c068f73127bf
thread:issue:67751,issue #67751: Remove logic for constructing symbolic shapes in Profiler,stale/low urgency,2,stale/low urgency,1,0,67751,,,,0,459337651d1638d0337ad83d07866a4b2e92d3bf788060c4b48c231793fc9e2a
thread:issue:67758,issue #67758: Building libtorch from source requires too much RAM,stale/low urgency,2,stale/low urgency,1,0,67758,,,,0,d8958e62c6b8d65fe2955c74e65cc6ce9e0f255b4f369c0500d3ccb814695dc2
thread:issue:67772,issue #67772: Functions are rendered incorrectly,stale/low urgency,2,stale/low urgency,1,0,67772,,,,0,adad7056b0da6bbd2da241f34d1cb0082d145dce38cc86a49c3e6d513786be05
thread:issue:67773,issue #67773: Instability in `test_input_weight_equalization_activation_values` test for random test values.,stale/low urgency,2,stale/low urgency,1,0,67773,,,,0,9a7c190d470a0025687b2b8c777f0fd28f51c5a3cdbfee3f55eac839bc22587c
thread:issue:6783,issue #6783: [caffe2] IfOp,stale/low urgency,2,stale/low urgency,1,0,6783,,,,0,bcd0cf6c678dd6f326375d59470351e4556d50d4946e4bcab50140e85a841518
thread:issue:67879,issue #67879: AMP does not improve performance,stale/low urgency,2,stale/low urgency,1,0,67879,,,,0,b671f2fc58eea1a2e1925e9fa5f324784d9bcd9d29ef0c74db80aa7fcc3f5fda
thread:issue:67913,issue #67913: Tracker: ModuleInfo-based testing,stale/low urgency,2,stale/low urgency,1,0,67913,,,,0,7584aee6c8064cc9427a1a4f0cd04cbd88e619ae778f33d4dc1c6d6bbf8ca0f0
thread:issue:67937,issue #67937: Add support for LTC ops needed by backwards/autograd/autocast,stale/low urgency,2,stale/low urgency,1,0,67937,,,,0,fe420ab7304653c1ef866a0521bad1509bdd32623e8a712bac62e38cde6634a9
thread:issue:6794,issue #6794: [Caffe2] TensorProtosDBInput AttributeError,stale/low urgency,2,stale/low urgency,1,0,6794,,,,0,b6e7088cc65a2390cb0f100a1e968bd883a0a98409c598c82844172826523ef8
thread:issue:67968,"issue #67968: unresolved external symbol ""__declspec(dllimport) struct _object * __cdecl THPVariable_Wrap",stale/low urgency,2,stale/low urgency,1,0,67968,,,,0,8be5c8df175cdb207c7db52439ca70996ea89ca87f23eb493c17d98a7c206d0f
thread:issue:67978,issue #67978: c10::CUDAError,stale/low urgency,2,stale/low urgency,1,0,67978,,,,0,363c5f98058d208b63a7a5e2e273a0fe8fc7b36dea827cca2e9eac21ddd83592
thread:issue:67979,issue #67979: Feature Request: Support prelu_cuda for BFloat16,stale/low urgency,2,stale/low urgency,1,0,67979,,,,0,099095e37f0d242f76683f7492a650c613362eb75e6f95f2f757a69dce45dc4d
thread:issue:67989,issue #67989: the abnormal display of bottleneck of resnet in tensorboard,stale/low urgency,2,stale/low urgency,1,0,67989,,,,0,9a7fc95f249786f799cc9beebac4c1d4ce948fd2e499585bfc16d99aadd835a2
thread:issue:67999,issue #67999: Improved Transformer and MultiHeadAttention design,stale/low urgency,2,stale/low urgency,1,0,67999,,,,0,c2a58ac456b799fff21367455cbbc6d4c21edb1d43569f8a41b3c5411d371006
thread:issue:68009,issue #68009: Properly document the `to.dtype_layout` overload of `Tensor.to`,stale/low urgency,2,stale/low urgency,1,0,68009,,,,0,526ff82c64620ecc341b64bc8530001fbf499280b2dcdc9e2aace6ba6fc8f226
thread:issue:68017,issue #68017: MacOS CI result is not in /README.md,stale/low urgency,2,stale/low urgency,1,0,68017,,,,0,0e1d92751bceae58932d1b5b47f6e0c05e199847dfca15a1ec186acd6bc93ebd
thread:issue:68046,"issue #68046: register_ Hook causes the CUDA out of memory, and remove() is useless",stale/low urgency,2,stale/low urgency,1,0,68046,,,,0,b579dddf4b51126a67aca605181b24e8d7ea1f728ac2e2bea5414721969dfa25
thread:issue:68055,issue #68055: Release official Python 3.9 images,stale/low urgency,2,stale/low urgency,1,0,68055,,,,0,635549bc6f32ed32d3c06982ef132c436855d8d709ad8bd338d13d5bc8bf0e39
thread:issue:68058,issue #68058: Some lr-schedulers docs seems to have typos/missing information,stale/low urgency,2,stale/low urgency,1,0,68058,,,,0,0249a6d03b5e53e35611955af5e95216e95ee84dcc8f72e503008e05f4e28647
thread:issue:68102,issue #68102: Can not guarantee reproducible for `nn.ReflectionPad2d`,stale/low urgency,2,stale/low urgency,1,0,68102,,,,0,3bbabf444824f9ce92287bd2018bf834cba33c150f41d90ca8c85b6f5e634718
thread:issue:68163,issue #68163: wsl2 install failed from source code,stale/low urgency,2,stale/low urgency,1,0,68163,,,,0,c158202f65f0e5c4ec268af7712ca7a27665e3d58d2d7f53ab9d29ef0ad8daac
thread:issue:68165,issue #68165: Upgrade NCCL2.11.4 in the new PyTorch,stale/low urgency,2,stale/low urgency,1,0,68165,,,,0,144c8295e2973ce84d1ae21a23f204820d9605210cc22e79317cc8e177e7448d
thread:issue:68169,issue #68169: torch.histogram: Sum of PDFs is num of bins instead of 1,stale/low urgency,2,stale/low urgency,1,0,68169,,,,0,627ee6d64133753997279f9d791ae64c553af073c2da19f69a6ef38a9b7724c1
thread:issue:68179,issue #68179: torch.nn.Linear wrong type annotation for bias,stale/low urgency,2,stale/low urgency,1,0,68179,,,,0,a799c4229329f0d09ad45fe9e87de1cdc924f4e2c355aec79eeee84783080e49
thread:issue:68190,issue #68190: [FX][docs] Document gotcha about `training` flag,stale/low urgency,2,stale/low urgency,1,0,68190,,,,0,4564091dd46465fc3befd94efeffb08047868af36aa38c0c3808a5991ebc1e8d
thread:issue:68193,issue #68193: clang-tidy shows false positive failure on PRs that rename files,stale/low urgency,2,stale/low urgency,1,0,68193,,,,0,ce7749034a6422b1f283523e3b055a05af66124af5d8ea25f01050bce4f4aaa4
thread:issue:68208,issue #68208: Make `torch.Tensor.view` support autograd for appropriate cases,stale/low urgency,2,stale/low urgency,1,0,68208,,,,0,ee94c7f2b2793411c4f6226d9fd17a332b4065d252f440f32006dabfc3d6bfba
thread:issue:68258,issue #68258: Option to configure TorchElastic multiprocessing/api.py PContext.close() timeout,stale/low urgency,2,stale/low urgency,1,0,68258,,,,0,2b18f2998849a67b22bf0bd8576df43d80bb7af42c964ac30a1226e6afdfaf10
thread:issue:68272,issue #68272: Query CUDA version LibTorch has been compiled with,stale/low urgency,2,stale/low urgency,1,0,68272,,,,0,35de34048be1eb8f3b0f2c3e6aac35b5aada03b7ab55738a4c90fef3daff1e31
thread:issue:68301,issue #68301: [FX] [BUG] Tensor.{inplace_method}_(.) is eliminated as dead code,stale/low urgency,2,stale/low urgency,1,0,68301,,,,0,82f51fe5f172def1292ae0efc7172c36b4b2a5d7213e44ec27aeaa55b673e4ec
thread:issue:68315,issue #68315: [docs] Strange signatures for torch.autocast,stale/low urgency,2,stale/low urgency,1,0,68315,,,,0,7dbc92688e1cc4ef4654b985ce71682c00859783c6a57957279ac55d6a304026
thread:issue:68323,issue #68323: sparse.mm: CUDA error: internal error when calling `cusparseSpGEMM_workEstimation [...]`,stale/low urgency,2,stale/low urgency,1,0,68323,,,,0,87ab0d0e97d70f8cf54146c9f8a4060297845f85e12fdf8998e4cb345c13089b
thread:issue:68331,issue #68331: Casting real parameter to complex during forward produces warning on backward,stale/low urgency,2,stale/low urgency,1,0,68331,,,,0,2a3741b19a4610f60a065dca8ef39cc210fae90ff4e8174f85d9df7a54e31054
thread:issue:68337,"issue #68337: `nn.functional.max_unpool{n}d`: shape checks fail with `output_size=(C, ...)`.",stale/low urgency,2,stale/low urgency,1,0,68337,,,,0,7d9f407ae9077dd87369bb74ef8bbf0a53e822abcec8303423e0f99b698a750c
thread:issue:68352,issue #68352: `Delta` not defined in the docs of `HingeEmbeddingLoss`,stale/low urgency,2,stale/low urgency,1,0,68352,,,,0,37c2bac457e885ea54559dfce384456135b28b48a154ffdb65aff35285b97c9c
thread:issue:68377,issue #68377: Misleading Error Message from nn.RNN when Passing Incorrect Data Type,stale/low urgency,2,stale/low urgency,1,0,68377,,,,0,2042ceb3d07dfcde661a61cac36632431ef838d5e536ef9b52ee53be492d5cb7
thread:issue:68394,"issue #68394: INTERNAL ASSERT FAILED at ""../aten/src/ATen/MapAllocator.cpp"":323, please report a bug to PyTorch. unable to mmap 68 bytes from file </torch_530808_7992>: Cannot allocate memory (12) ---------------------------------------------------------------------------",stale/low urgency,2,stale/low urgency,1,0,68394,,,,0,6bab9e4c8500e81c6cd7a4750515bfeb4048e714407120087f06d5c02749dad6
thread:issue:68405,"issue #68405: Windows multi machine multi card training, card initialization in GLOO communication.",stale/low urgency,2,stale/low urgency,1,0,68405,,,,0,3dab0da7010ab25a99b6ab4637bd6b90f453f0c75cdbd541246759cc71d0d476
thread:issue:68420,issue #68420: `nn.functional.max_unpool(2|3)d`: failing shape check for correct inputs (with `dilation > 1`) with specified `output_size`,stale/low urgency,2,stale/low urgency,1,0,68420,,,,0,38a283843e571432baf150a56eb96ceef9b81c8c0f998edca46d723bd7131081
thread:issue:68422,issue #68422: There is a problem with GPU memory reclamation,stale/low urgency,2,stale/low urgency,1,0,68422,,,,0,6ce496b9996e6549c75464d01fa65b1f2f42b483cab6364502bc1b06f018c3d0
thread:issue:68423,issue #68423: `torch.utils.checkpoint.checkpoint_sequential` is not optimal,stale/low urgency,2,stale/low urgency,1,0,68423,,,,0,f8305d1957936ae9fab3747f5c2002e147ad2a56e2f341e0be694a5b19b4a600
thread:issue:68429,issue #68429: istft gradcheck fails on ROCm,stale/low urgency,2,stale/low urgency,1,0,68429,,,,0,3a0ae5b935b7d567f8314641cc7c5ea06b9f24ee7e546f1d2aa9e64a465628e4
thread:issue:68430,issue #68430: Strides issue on unsqueezed channels last tensor,stale/low urgency,2,stale/low urgency,1,0,68430,,,,0,a2d9a5d0bbce31484dbeae62e8c383486ea3c164131fb71e0b669188f348ed09
thread:issue:6845,issue #6845: [caffe2] global average pool in caffe2,stale/low urgency,2,stale/low urgency,1,0,6845,,,,0,02a378b09cf920a48adbcf09f0c29509584b1e48305676eac56df1abf9e90dbb
thread:issue:6846,issue #6846: Jetson TX1 Caffe2 installation fails on latest JetPack release,stale/low urgency,2,stale/low urgency,1,0,6846,,,,0,a0051a724acad6d6a64c7e964de4ba3500dfa8400eea8cef1a58f75890fa8f67
thread:issue:68464,"issue #68464: Changing the type of dtype argument in softmax, log_softmax, and softmin is possibly BC-breaking.",stale/low urgency,2,stale/low urgency,1,0,68464,,,,0,9eabacffc62ed4d6aaaf646e9ce342d45b218e0f760540e7581c53c1cc2ab83c
thread:issue:68474,issue #68474: [LTC][BE] Fix all compile warnings,stale/low urgency,2,stale/low urgency,1,0,68474,,,,0,f71fa8bc0fbfc960de21f5463c8e96c35a9d769c0442e9f870e197414c12651b
thread:issue:68475,issue #68475: [LTC][BE] Make the compile flags more strict,stale/low urgency,2,stale/low urgency,1,0,68475,,,,0,69eaa824b194342c508673996a18a8305ae0a85a2d2ba2227b8c8c865eaed0a0
thread:issue:68498,issue #68498: Review disabling tests workflow,stale/low urgency,2,stale/low urgency,1,0,68498,,,,0,63c13b388fb1aaec28498ae0113892ce3b6d45b7c28e10a3e460cea4762fa494
thread:issue:6850,issue #6850: [feature request] More `index_*_` functionality and/or lambda functionality,stale/low urgency,2,stale/low urgency,1,0,6850,,,,0,47615baeaad22bc67f244ac77a46df8a5cd5cdf29130407a73f6b9378cc466ad
thread:issue:68507,issue #68507: Multiprocessing hang and queue short circuiting,stale/low urgency,2,stale/low urgency,1,0,68507,,,,0,ac8ee8de54f863e158bff4b88599c8e872db7dcbc1795d4fa99759507110898e
thread:issue:68512,issue #68512: 'replicate' padding in convolution is 77 times slower on cpu than 'zeros',stale/low urgency,2,stale/low urgency,1,0,68512,,,,0,b594d3fccf78609f65b060565345e210a15a99c31d67c9f66ad5076fe1d415dd
thread:issue:68513,issue #68513: torch.fx cannot trace torch.Size() properly,stale/low urgency,2,stale/low urgency,1,0,68513,,,,0,db5cad025c2914a65d30d1028602c6eed7b5ffbe80db4af876ea58d3a23f8c9a
thread:issue:68519,issue #68519: dataloader will miss batch data when num worker>0,stale/low urgency,2,stale/low urgency,1,0,68519,,,,0,da1e35c07d922ae2ff63407cb39e28292fcd7c669064002c1444bc6019e14062
thread:issue:68553,issue #68553: [distributed] elastic/agent/server/api.py could react faster on exit / Ctrl+D / Ctrl+C,stale/low urgency,2,stale/low urgency,1,0,68553,,,,0,a5bd5ded32008b4e3e02a451465c849b915803e76a5e93bdc0d02f1a0cf430f4
thread:issue:6857,issue #6857: [caffe2] Understanding WorkFlow for Training and Testing,stale/low urgency,2,stale/low urgency,1,0,6857,,,,0,e20278019366cd8dc56eb18695f13ffa3cdc3c490d6e45c40a039119da14f440
thread:issue:68575,issue #68575: [ux] Print tensor's dtype by default,stale/low urgency,2,stale/low urgency,1,0,68575,,,,0,097b1cda58f5d176f7178a1cf2a4bd0e26a7e8f0c4d102055bdaddf3fb0d1e43
thread:issue:68610,issue #68610: `torch.unique_consecutive`: passing positional optional arguments results in empty tensors,stale/low urgency,2,stale/low urgency,1,0,68610,,,,0,aa917db6e0d527e98afdc78d5ae1b8e959d146cb9e0e048e983a9faae28c7a44
thread:issue:68622,"issue #68622: torch.{tensor, Tensor, LongTensor, ...} isn't captured under `enable_python_mode()`",stale/low urgency,2,stale/low urgency,1,0,68622,,,,0,d18346bd3866280bdfc6f9cfcef7125e978617a341288b95e9f4b34f3819c05d
thread:issue:68655,issue #68655: [ux] F.binary_cross_entropy (and maybe other losses) to auto-cast bool category mask to float,stale/low urgency,2,stale/low urgency,1,0,68655,,,,0,0bd2de14efaff9f6528c1154f83a7c368a653589c34eda0c45997595703aed58
thread:issue:6868,issue #6868: [Caffe2] Android NNApi integration bugs.,stale/low urgency,2,stale/low urgency,1,0,6868,,,,0,813bccc7fc04bffb990aaf8cb39f1fde657af8166e40a7adcbd56da7b6421963
thread:issue:68696,issue #68696: Is it possible to use torch.linalg.cholesky for cholesky decomposition of a huge matrix that doesn't fit in memory?,stale/low urgency,2,stale/low urgency,1,0,68696,,,,0,2f034e1eebada278c81dbc304c56acccb14b131e910631bf4a13d1aadafdbf51
thread:issue:68701,issue #68701: Tensor precision manifests differently between CPU and GPU.,stale/low urgency,2,stale/low urgency,1,0,68701,,,,0,9e1ef285ff845bead598683d55c90d03b7807a9ccb75e5a49b166dd21e3dcd74
thread:issue:68703,issue #68703: Feature request: Extend `torch.eye` for creating a batch of eye matrices,stale/low urgency,2,stale/low urgency,1,0,68703,,,,0,3e3f313892d21cf7cd0db221989e3fb961910577119729a9c0bf04811ff1f782
thread:issue:68746,issue #68746: Add kl_divergence between Normal and Laplace distribution.,stale/low urgency,2,stale/low urgency,1,0,68746,,,,0,0db4a472b3ea7e03c351aa0bb48a26fa2fd55bd439cf76906f7c59c9d882dee3
thread:issue:6875,issue #6875: [Bug] VS reports unresolved external symbol from caffe2 observer,stale/low urgency,2,stale/low urgency,1,0,6875,,,,0,1f9eeadab5ccc05874fdd5f015a69b30c624073c22228ae774a39712bddc8870
thread:issue:68752,issue #68752: `softmin` and `softmax` operators support different input dtypes based on whether the `dtype` kwarg is passed,stale/low urgency,2,stale/low urgency,1,0,68752,,,,0,91b0220a19d020ad9562449cb3d2c1fd7a9ade950885e29e6c89a3fff8dff630
thread:issue:68781,issue #68781: [docs] Clarify DDP activation checkpointing support,stale/low urgency,2,stale/low urgency,1,0,68781,,,,0,4d026482a15c6d8d3629b2666caf20ff4886e11670ba331f6c1f50a5d3489f5c
thread:issue:68803,issue #68803: [DDP] Verify ignored parameter names in debug mode during init,stale/low urgency,2,stale/low urgency,1,0,68803,,,,0,01a42efc2cd87f73a02cf2e205605bf4de52d4572df4a0c87146a8862faf9aa4
thread:issue:68861,issue #68861: [In Gunicorn & multiprocessing environment] Cannot re-initialize CUDA in forked subprocess,stale/low urgency,2,stale/low urgency,1,0,68861,,,,0,3f614664c2250222706efa1937f0feee31b0d5e1a11d72366eb7ecca00c6cb12
thread:issue:68892,issue #68892: fusion in fx graph mode did not take care of direct attribute access,stale/low urgency,2,stale/low urgency,1,0,68892,,,,0,5907062e53e345d9622a43b2fc0419c669df4ca0bf73125abb04762e98fe555e
thread:issue:68909,issue #68909: Bug about Dropout CUDA Kernel,stale/low urgency,2,stale/low urgency,1,0,68909,,,,0,21417ef72727915d3986813f4ee5074d0f88b153fca79e5a7e4a518674f94cbd
thread:issue:68916,issue #68916: Crash loading fairseq based model,stale/low urgency,2,stale/low urgency,1,0,68916,,,,0,4cfd84490cfc23a0be94360a23bd3c4cd19770257b5688421296b8956f5e9a40
thread:issue:68923,issue #68923: [cpp extension] making it possible to parallelize the building process,stale/low urgency,2,stale/low urgency,1,0,68923,,,,0,90ef235c18cb9621b76ebbb0f4dd0bceed6d027ca85aca2a635e58cbdb0f266c
thread:issue:68925,issue #68925: How to implement `bucket_by_sequence_length` with IterableDataset and DataLoader,stale/low urgency,2,stale/low urgency,1,0,68925,,,,0,c3d8a1285456eb9c24e914ce95278a050f2b566818a21f92405e0f887c189cf2
thread:issue:68931,issue #68931: Dead lock happened ran pytorch 1.9.0 cuda11.2 on xeon gold 6326 cpu,stale/low urgency,2,stale/low urgency,1,0,68931,,,,0,1e588f235548347892d64fc81af8720e616158fc44ef86862b9e592c944adea5
thread:issue:68935,issue #68935: repeat_interleave hangs in a forked subprocess,stale/low urgency,2,stale/low urgency,1,0,68935,,,,0,919b231fd1ef9d978bf8bb3eec7b4d8ddb30ba53fc822da9314daf4396b59d15
thread:issue:68944,issue #68944: Pytorch to ONNX model conversion,stale/low urgency,2,stale/low urgency,1,0,68944,,,,0,1d3def22f58a23b4e7b940ecd91788f4dba04063fb58552f23e379bf68c4d345
thread:issue:68959,issue #68959: Deterministic implementation for grid_sampler_2d_backward_cuda,stale/low urgency,2,stale/low urgency,1,0,68959,,,,0,a8b635b10f6f70128dad9d23ae72005c847de543bbe12cce615043c63a8ffe29
thread:issue:68961,issue #68961: Deterministic implementation for upsample_bilinear2d_backward_cuda,stale/low urgency,2,stale/low urgency,1,0,68961,,,,0,376e58a89368cf6c2d02d783835d077840540f48c3b84a3fa5054e0fa2fa12d1
thread:issue:68966,issue #68966: `torch.ldexp` generated tests fail on call to `torch.mul`,stale/low urgency,2,stale/low urgency,1,0,68966,,,,0,ca41ae5ce6d0dddaf0c9fe37133ff04ff6a3b3a0881f3cdea24bb01b85fbf7cb
thread:issue:68967,issue #68967: [feature request] torch.argminmax,stale/low urgency,2,stale/low urgency,1,0,68967,,,,0,d630bb1d8c811043c0511304bcdc425569092b8f9b04ebe1eb37ba98eca83e72
thread:issue:69009,issue #69009: Mamba does not respect `cpuonly` when creating a conda environment from YAML file,stale/low urgency,2,stale/low urgency,1,0,69009,,,,0,a9e33241b2b8988eee3cc8e927b6b01c15f553c555dc9d2fa5eae1368274dfcd
thread:issue:6901,issue #6901: [Caffe2] [feature request] Dilations in grouped convolutons,stale/low urgency,2,stale/low urgency,1,0,6901,,,,0,2c9a65e405fabc3529d3ee2509c98e905d62e364524e8bb04bdb133e0c1f0ce4
thread:issue:69013,issue #69013: libtorch_cuda links against wrong libnccl.so,stale/low urgency,2,stale/low urgency,1,0,69013,,,,0,150880ccf7892bbe427910154710fb6e972b086c01458a638a6474ed80d09bc6
thread:issue:6902,issue #6902: [Caffe2] Flatten Layer in caffe2,stale/low urgency,2,stale/low urgency,1,0,6902,,,,0,efc45119a2aafc0840f7bb41e6e42799a94885100881222888df4244012c701d
thread:issue:69025,issue #69025: [subgraph_rewriter] Support skipping matching for some ops,stale/low urgency,2,stale/low urgency,1,0,69025,,,,0,7391e0440c12f1353ea4cc975e41bd5de925fd781292e4344a29ca80590040ee
thread:issue:69063,issue #69063: make: Makefile: No such file or directory,stale/low urgency,2,stale/low urgency,1,0,69063,,,,0,84e1ccfe8f7f4b2540c84f2ed18b5843187f0c48aa426a46b643ccef166a8b18
thread:issue:69072,issue #69072: Libtorch dll versioning,stale/low urgency,2,stale/low urgency,1,0,69072,,,,0,c01cd10074f9cf962670db6aa31207b3699f2ca1ced5e3d53f9e547a5302af28
thread:issue:69080,issue #69080: Generalized matmul,stale/low urgency,2,stale/low urgency,1,0,69080,,,,0,159c5f244ad478226ed5c80dc27ad1d5986fa95aa46aa906cfc4b2a399583804
thread:issue:69086,issue #69086: Confusing value of `TORCH_CUDA_ARCH_LIST`,stale/low urgency,2,stale/low urgency,1,0,69086,,,,0,6a8cd0df3469ddfca5e88004ea1eb3f517d539b5adedb6c70d9da5da2a6d048c
thread:issue:69095,issue #69095: [FX] `concrete_args` with unpacking breaks `fx.Interpreter`,stale/low urgency,2,stale/low urgency,1,0,69095,,,,0,ddb7ac4d017cdbc0606191e24de625b45ab1b42bf94b303201d058c67069747a
thread:issue:69158,issue #69158: Fusing module with multiple hooks causes `RuntimeError: OrderedDict mutated during iteration` in fuse_known_modules,stale/low urgency,2,stale/low urgency,1,0,69158,,,,0,073c645c3ead4fbd690e8d36b3aa67068e15d444f076fe36b79e9ed59b5bfd2d
thread:issue:69197,issue #69197: RendezvousConnectionError when use C10d on multi nodes,stale/low urgency,2,stale/low urgency,1,0,69197,,,,0,b5c5c33862acb53c970ca86714641c22b7dd06052e76c0326f82196fcf930f08
thread:issue:69199,issue #69199: Protocol buffers library mismatch.,stale/low urgency,2,stale/low urgency,1,0,69199,,,,0,86fe0941e4da17d47cc8d041f5376cd30ec641fe9ee0ef8770ba042e723ad7a4
thread:issue:69202,issue #69202: CPU only mode with cudatoolkit 11.3,stale/low urgency,2,stale/low urgency,1,0,69202,,,,0,e1df89be7f4ed9f5ee162b50e6874a9c57550e49ded80aa5397ab84cec0d4214
thread:issue:69207,issue #69207: Function(s) expecting a tuple argument don't accept generators,stale/low urgency,2,stale/low urgency,1,0,69207,,,,0,e6f42a88ac78e65cd2d7254b59c0606c17902436b7d611f3dafe1f43ac78a932
thread:issue:69213,issue #69213: torch.stack should have a pin_memory parameter,stale/low urgency,2,stale/low urgency,1,0,69213,,,,0,4cec8276c6818cee60aa77012a4bd4e4f19672c618e1691ce130dee17f350cad
thread:issue:69233,issue #69233: [ROCM] elementwise kernel launch,stale/low urgency,2,stale/low urgency,1,0,69233,,,,0,57e78f3460bf408dc36a712caf0e3af722f9adfb10738b77c5a61b36d6d2f83c
thread:issue:69281,issue #69281: Implement reparameterized sampling for LKJCholesky distribution w.r.t. concentration parameter,stale/low urgency,2,stale/low urgency,1,0,69281,,,,0,fdb2852ef94795a0bbab75cd19f8578fef678cee6caa1e6500d9d9ed69d73249
thread:issue:69286,"issue #69286: subprocess.CalledProcessError: Command '['cmake', '--build', '.', '--target', 'install', '--config', 'Release', '--', '-j', '32']' returned non-zero exit status 1.",stale/low urgency,2,stale/low urgency,1,0,69286,,,,0,c2b5b28d24b2ed5614282c138a73ad1a7906071e397c11de2a78498d289985db
thread:issue:69287,issue #69287: `out=` variant for `conj` unary operator,stale/low urgency,2,stale/low urgency,1,0,69287,,,,0,7d323527fba885a419ede7e1ad0a2950508f624341de1c0ab52dcfcf73e8630c
thread:issue:69288,issue #69288: Inplace and `out` variants for `positive` operator in PyTorch,stale/low urgency,2,stale/low urgency,1,0,69288,,,,0,6444fea92300631020a433c15803055e68ba421e8cf0b727788f8dc75e5ac7ff
thread:issue:69290,issue #69290: Deprecation warnings generated when including header files in C++17 code,stale/low urgency,2,stale/low urgency,1,0,69290,,,,0,813dd215f96b137949daeb74664981091a6809e7640ec7e07e0451c028a2e993
thread:issue:69321,issue #69321: Poor Transformer inference scalability on CPU,stale/low urgency,2,stale/low urgency,1,0,69321,,,,0,89fc9be1b86023b168fe38652136463662b455a16b807eca78cc80733acfe299
thread:issue:69325,issue #69325: `nn.functional.fractional_max_pool2d` and `nn.functional.fractional_max_pool3d` produce incorrect output on non-contiguous inputs,stale/low urgency,2,stale/low urgency,1,0,69325,,,,0,a2a6dda247c9cc9bef6e32b2b9bc20a6933bb20437d3c51f6f0367d42e685157
thread:issue:6934,issue #6934: Feature request: SSIM/MS-SSIM,stale/low urgency,2,stale/low urgency,1,0,6934,,,,0,931fcf826d17f2470a5f1973f8d2afff48c2069fce200aa2afb4abcae3d3686f
thread:issue:69348,issue #69348: `torch.sspaddmm` should broadcast the input tensor,stale/low urgency,2,stale/low urgency,1,0,69348,,,,0,2667a22c38f7c1faf72e8dbeaf03559251f7e6a68de5a7415ee75c9f2da6feba
thread:issue:69360,issue #69360: 1.10.0 failed to build due to missing TensorBody.h,stale/low urgency,2,stale/low urgency,1,0,69360,,,,0,d93ddd3378dfa8abf7128de1d9417a7db082e98650ded3749b9cee977aede85a
thread:issue:69363,"issue #69363: Citation request for ""probabilities for each class"" in the doc's description on cross-entropy",stale/low urgency,2,stale/low urgency,1,0,69363,,,,0,fa90ac3ed3037391442c1dfb697f43eb24ac2952debb227bc254eec3bf3ecdd6
thread:issue:69408,issue #69408: `torch.hstack` should raise an error when tensor is 0 dimensional,stale/low urgency,2,stale/low urgency,1,0,69408,,,,0,3e80310b35e722dc72903934e36b955ddaa898ac89cff9af02804ea7b58921ad
thread:issue:69433,issue #69433: `torch.transpose` should raise an error when indexing 0 for 0 dimensional tensor.,stale/low urgency,2,stale/low urgency,1,0,69433,,,,0,ba622108f1cd417ca5f7112e432939dd4e6c1018c7dc556aba9fc0090a645d4c
thread:issue:69442,issue #69442: The formula for KL-divloss is wrong in the document,stale/low urgency,2,stale/low urgency,1,0,69442,,,,0,8843fa0662d0abe21392288a4fd376fb47f3ef0ebc19af91b1622642990b3d79
thread:issue:69451,issue #69451: Exponentially scalable modified Bessel function of the second kind,stale/low urgency,2,stale/low urgency,1,0,69451,,,,0,30d97a8d856ea4c3329e4113308d3396aaf0c58bb2eadebd20c0a7e3ddd3a83d
thread:issue:69452,issue #69452: Complete elliptic integral of the first kind,stale/low urgency,2,stale/low urgency,1,0,69452,,,,0,e11cab75dace3997545528eb038f7223881d51d3e05510be018fcf43e4aaf1ec
thread:issue:69453,issue #69453: Jacobian elliptic functions,stale/low urgency,2,stale/low urgency,1,0,69453,,,,0,8c604613f661797af85d0a85c37c05909a6eb7592a44b5081ce625c80d769ebe
thread:issue:69471,issue #69471: Include Declarations.yaml in Libtorch distributions,stale/low urgency,2,stale/low urgency,1,0,69471,,,,0,b9cb83bb8f17a8fb6fe5fe85003d6f97d789071f796b0a686a5a5916774cbe84
thread:issue:69505,"issue #69505: with the same environment ,pytorch 1.8 worked but 1.10 can't work",stale/low urgency,2,stale/low urgency,1,0,69505,,,,0,7c83dd194ac281b99bf543dc4b058350d28d46075ec9ab0e5c9ab1136f7de99f
thread:issue:69516,issue #69516: `torch.sparse.softmax` and `torch.sparse.log_softmax` do not support negative dim.,stale/low urgency,2,stale/low urgency,1,0,69516,,,,0,05262659fc4fc104e4f68ffca5d732ceda5b7422b37b3be0e0619dd7f0eeb214
thread:issue:69520,issue #69520: race condition of agent's TCP store on exit_barrier(),stale/low urgency,2,stale/low urgency,1,0,69520,,,,0,a09785c89705a11ecb0b4d8be8f8d2c2c4b8af04d6a808a5a70642f54e8e5959
thread:issue:69524,issue #69524: [feature request] Accepting python scalar inputs for torch.minimum/torch.maximum and friends,stale/low urgency,2,stale/low urgency,1,0,69524,,,,0,eaddec2e353edb59d32ee61e3aa472c310446672af61ff20643bb46cbad3c2b9
thread:issue:69530,issue #69530: Extend composite compliant testing to backward formulas,stale/low urgency,2,stale/low urgency,1,0,69530,,,,0,14d675de9c317676ff06f19ba6a7732ee9a8296b1af2b7bb5b25de90b273537d
thread:issue:69577,issue #69577: [subgraph_rewriter] Add match_filter for the matches,stale/low urgency,2,stale/low urgency,1,0,69577,,,,0,3fb0b47e3edf09a38e9269c01d0b6dfdba6fcd5263fa354adc709a9bd4ec4aea
thread:issue:69610,issue #69610: [Question] How to extract/expose the complete PyTorch computation graph (forward and backward)?,stale/low urgency,2,stale/low urgency,1,0,69610,,,,0,b34f25a36df58017c0aed7ede5ef3b36c35d23abe6b0d04a1260422ba7f41867
thread:issue:69627,issue #69627: c2r fft input generation,stale/low urgency,2,stale/low urgency,1,0,69627,,,,0,4df18794239b16eca65b791035921e8057c5a3c78e52fa8aa2aa5116b5f8688c
thread:issue:69654,issue #69654: Fused AlphaFold 2 modules,stale/low urgency,2,stale/low urgency,1,0,69654,,,,0,503128c467ad0e491c8e2f066f2919ae3d0de24a6395b56744d84b44ea477ebd
thread:issue:69663,"issue #69663: After updated to pytorch1.10.0 cuda11.1, NCCL is not available",stale/low urgency,2,stale/low urgency,1,0,69663,,,,0,3a2f33cd58475fec83ff3b8cf668d0331c271d557fd0fa533d19f8c65f5e28c3
thread:issue:69665,issue #69665: No dtype check for zero sparse tensor!,stale/low urgency,2,stale/low urgency,1,0,69665,,,,0,f8576ddd398d1e4467c88cefa1dfa5283bc6c0c39938144a6fff5b388802133b
thread:issue:69676,"issue #69676: [LazyTensor] model after to(device), traced IR changed and lose type info",stale/low urgency,2,stale/low urgency,1,0,69676,,,,0,fa33fb6c89ddee9a4e7ec4951a5f3540866dee91fbfef0e966c8b41b8e67ebdd
thread:issue:69685,issue #69685: Wrong PyTorch version in Docker image,stale/low urgency,2,stale/low urgency,1,0,69685,,,,0,b951998e00b388179a4aacb9ae230ce8feef6198b3cdde00b1ad984c4fdab530
thread:issue:69714,issue #69714: Possibly out of date error in autograd codegen,stale/low urgency,2,stale/low urgency,1,0,69714,,,,0,175810cbb790d918ba166c216adf147547fdbf325c867c9dd0446b57ce992215
thread:issue:69721,issue #69721: Alternative to keras StringLookup in Pytorch,stale/low urgency,2,stale/low urgency,1,0,69721,,,,0,f6d826329f97987de2064fd24792fc0334e8a4e236c45577e1c48c843c3051ea
thread:issue:69753,issue #69753: last_epoch parameter of CyclicLR and OneCycleLR is not the number of epochs,stale/low urgency,2,stale/low urgency,1,0,69753,,,,0,bb4f98af1be11b1d411ba5581b32bdfecfec90f85d683890aeedbcdc46777aad
thread:issue:69756,issue #69756: Introduction of nn.LazyRNNs,stale/low urgency,2,stale/low urgency,1,0,69756,,,,0,3978e706526b6e78c1f8f49600c9f6ab88ede99b2d56ab6781501e74ee57d6f3
thread:issue:69772,issue #69772: RuntimeError: t == DeviceType::CUDAINTERNAL ASSERT FAILED when trying to calculate gradients,stale/low urgency,2,stale/low urgency,1,0,69772,,,,0,e707f504053f44df7253755a4e1ae9f02a6963d947849509d7a52e223c0c7fc9
thread:issue:69800,issue #69800: Incomplete error message at tensor indexing (when indexing with set),stale/low urgency,2,stale/low urgency,1,0,69800,,,,0,72b9cb59aeb916196a6d886ca59e3b3405d8b80a0f2bd251777e957599b4143a
thread:issue:69804,issue #69804: typing error in the signatures of `torch.unbind` when getting them using `torch.fx`.,stale/low urgency,2,stale/low urgency,1,0,69804,,,,0,12e0be05bb5dca9f763916ef3c4743eebcb1c9fb8246cfdc8af82cb2ed70c47e
thread:issue:69831,issue #69831: Incorrect error for integer `out=` dtypes when a float is expected.,stale/low urgency,2,stale/low urgency,1,0,69831,,,,0,0f888cf95513bdbe06821de0559e0ba64cd08a7947872a756d3878eb422eaaa5
thread:issue:69858,issue #69858: Feature Request: Bayesian Personalized Ranking Loss,stale/low urgency,2,stale/low urgency,1,0,69858,,,,0,71050fad6ac2525576955e0da4bf289a2a9d6ce484b21ba65686af2aa72711c9
thread:issue:69893,issue #69893: subclassing torch.Tensor,stale/low urgency,2,stale/low urgency,1,0,69893,,,,0,51e1ea2dfaac279279e29529aa7e7f4f36d582e6513926afe7b0bc70277e3719
thread:issue:69900,issue #69900: Port MarginRankingLoss to TensorIterator,stale/low urgency,2,stale/low urgency,1,0,69900,,,,0,9e4f6dcf3aaf3d6622daa09f93dd645a3f6ea1a21d222cd439d1242bbadb1507
thread:issue:69912,"issue #69912: torch.nn.functional.ctc_loss with invalid input produce NaN or infinity gradient, while the batch entries are fine",stale/low urgency,2,stale/low urgency,1,0,69912,,,,0,a68710f94609b4a3c05056ef0c8173352aab73ddf6fa1a633d5c91e59cebf21d
thread:issue:69921,issue #69921: Context manager to enable/disabled TensorFloat32 on demand,stale/low urgency,2,stale/low urgency,1,0,69921,,,,0,2b016f2f252758acd18159da127d5cd6fedcc3cdd93499f3db4db7b881a325c0
thread:issue:69932,"issue #69932: Would pytorch like some free multi-factor authentication (MFA) tokens from Google & GitHub, via the OpenSSF?",stale/low urgency,2,stale/low urgency,1,0,69932,,,,0,d15ed1246ce18c8e2d215ea74ce04fadb60818168295ba38b1101d8a31463ea5
thread:issue:69984,issue #69984: Docs for torch.nn.MSELoss are confusing,stale/low urgency,2,stale/low urgency,1,0,69984,,,,0,d1017d9a2a35323fd7ab510cafe7435c181f675af32f3f0e404b9afc64b4d0c6
thread:issue:70040,issue #70040: Error in `torch.cdist` documentation,stale/low urgency,2,stale/low urgency,1,0,70040,,,,0,2205a7876efbca1a9a7746db0b6d86cee458891db1a9aa9deda778b8c48316e2
thread:issue:70095,issue #70095: Add support for a `default` arg in `ModuleDict.pop`,stale/low urgency,2,stale/low urgency,1,0,70095,,,,0,6765d2fab0a37c9e8d05191121661a635660f63fbab68f6e95e905eaa05c6c56
thread:issue:70100,issue #70100: Feature Request: dim parameter of torch.nn.functional.normalize should accept tuples,stale/low urgency,2,stale/low urgency,1,0,70100,,,,0,c50e178a434922a9f9f0a1ecbaeffc1f53c90172b7a8898e16dc5b5add5d0e52
thread:issue:70138,issue #70138: build_ios.sh prevents iOS.cmake from configuring ios deployment target correctly,stale/low urgency,2,stale/low urgency,1,0,70138,,,,0,987571dff363ec1aa749cf5b00b43d9597b5b0af0da6afb32daf111377eb5f53
thread:issue:70158,issue #70158: pin_memory *still* destroys custom containers,stale/low urgency,2,stale/low urgency,1,0,70158,,,,0,dc4fe062c34382a383ad0dec039b9d2967d03a224de3365b9d2349ccf001d1b4
thread:issue:70166,issue #70166: [feature request] Autocast module and function wrappers,stale/low urgency,2,stale/low urgency,1,0,70166,,,,0,9ca16fe07d47da1ad20828d9ea3f49fe6adc46c48e5ba201a4a296e70b32a443
thread:issue:70170,issue #70170: EmbeddingBag allows out of index ranges.!,stale/low urgency,2,stale/low urgency,1,0,70170,,,,0,31bb72a0d03865c5f588c6e08f51b5a54eb0c49ecd43a92f5dee0a2522fa16ba
thread:issue:70171,issue #70171: Channels last performance problem,stale/low urgency,2,stale/low urgency,1,0,70171,,,,0,3b4b4a0fbb054545ac926d1d786cd8f0ef5d3dc1baf3e306b07e07d7798ea965
thread:issue:70180,issue #70180: Can you provide the torch.trt module to directly convert the pytorch weights to tensorrt?,stale/low urgency,2,stale/low urgency,1,0,70180,,,,0,b54aa5436944211ae2011fa09e62eebfb73b8fc751373967b27d4e4fb8cfd555
thread:issue:70184,issue #70184: Optimization: convolution_backward doesn't always need to call .contiguous on certain inputs,stale/low urgency,2,stale/low urgency,1,0,70184,,,,0,8a22cefef355921da12ce7b3cdabe3a1d17be80e0f5c3ea545222f44b2896fc7
thread:issue:70199,issue #70199: Missing Docker image for 1.10.1,stale/low urgency,2,stale/low urgency,1,0,70199,,,,0,660eabb99718f9341a4e18de1f2a480732b82148730e526811d02f4d575ae53d
thread:issue:7020,issue #7020: [Caffe2] LT and GT cannot be exported to ONNX,stale/low urgency,2,stale/low urgency,1,0,7020,,,,0,2170c21a70da8829ef7150c9fe452930af9f9be5efd9682cff768a67377d1d5e
thread:issue:70222,issue #70222: Grad strides do not match bucket view strides,stale/low urgency,2,stale/low urgency,1,0,70222,,,,0,02bc8619b63ebca88404c9445aab0505489c6f479965630d4c2091ae580f0b28
thread:issue:70238,issue #70238: `TensorIterator`: refactor `build_ternary_op` to match binary versions,stale/low urgency,2,stale/low urgency,1,0,70238,,,,0,06d3a2a7813058a60fe3b2e077a5bfca04db07e21390ed241cf10b0bef540cce
thread:issue:70239,issue #70239: `pytorch` hangs during interaction with `ray` package,stale/low urgency,2,stale/low urgency,1,0,70239,,,,0,5d9390ed0a6e32a33274dd841936cd9c14d38e2f563f12d513c9dd8f4bd5d570
thread:issue:70243,issue #70243: Unable to compile PyTorch when libcudart_static.so is not available,stale/low urgency,2,stale/low urgency,1,0,70243,,,,0,8995b416433a075ed3e48026f57df56da25170735f9ea481c0fdd5204f627c4d
thread:issue:70265,"issue #70265: Shape parameter inconsistency in torch.Tensor.view, torch.reshape, torch.Tensor.reshape",stale/low urgency,2,stale/low urgency,1,0,70265,,,,0,17f2bfe111327524e21de4b638ea5bb9ade277dff850719675aaefda888848b9
thread:issue:7032,issue #7032: [feature request] norm argument for RNNCells,stale/low urgency,2,stale/low urgency,1,0,7032,,,,0,25eef6baceaac2ab6e82e1eb68f1ed763cfa7101f87f4df15581f6719b89f29d
thread:issue:70342,issue #70342: Empty or NaN data pollute gradient even if they are not involved during backward,stale/low urgency,2,stale/low urgency,1,0,70342,,,,0,43922f39e0bac73bfe51a6148fdb97672bd0d225cb2df77c9c4812becba5f6fe
thread:issue:70346,issue #70346: Add hints for gradient long time overflow when using torch.cuda.amp,stale/low urgency,2,stale/low urgency,1,0,70346,,,,0,7cd46316b5198b0bf0d200c38b34bd872544c57896cf085a94b91f899d8c5ca4
thread:issue:70348,issue #70348: nan return by nn.CrossEntropyLoss when all the labels are ignore_index in torch 1.11,stale/low urgency,2,stale/low urgency,1,0,70348,,,,0,77035fa18c35903756e120a74d59660eca8b4ee56b3f37532225eb1673753548
thread:issue:7035,issue #7035: [Compilation] how to disable caffe2?,stale/low urgency,2,stale/low urgency,1,0,7035,,,,0,69c5b50b876f3c3cb481fdd79bf303e50c41c20657171ca53d652c2fd0fffd08
thread:issue:70352,issue #70352: Rprop Optimizer: UnboundLocalError: local variable 'step_size_min' referenced before assignment,stale/low urgency,2,stale/low urgency,1,0,70352,,,,0,479ff184ebfcfb92c5c515d49e73947682742cdd333527d4ff9b6d62d1f6459b
thread:issue:70386,issue #70386: AT_ASSERT fail with DataLoaderOptions().drop_last(),stale/low urgency,2,stale/low urgency,1,0,70386,,,,0,4db44287696694263f22aaa31532649be341f6b55827fe7b48d8633d4f02c3b7
thread:issue:70388,issue #70388: RuntimeError: tensor has too many (>25) dims when permuting tensor with GPU backend,stale/low urgency,2,stale/low urgency,1,0,70388,,,,0,8a69486dc4c18f53898357eb250bc4cae3a37b1222930ffcce966bf066f3a661
thread:issue:70391,issue #70391: linalg.lstsq INTERNAL ASSERT FAILED,stale/low urgency,2,stale/low urgency,1,0,70391,,,,0,59a21056d13e2187593c05edaba092d5ec50803c878c59005d7b4fd606f92d71
thread:issue:70392,issue #70392: third_party/breakpad/ compilation failure,stale/low urgency,2,stale/low urgency,1,0,70392,,,,0,1b24b8d9e0d007e0fc84b085dd6e7d981018dd5e017e37acab7f9162d883b9a8
thread:issue:70394,issue #70394: Intel MKL FATAL ERROR: This system does not meet the minimum requirements for use of the Intel(R) Math Kernel Library.,stale/low urgency,2,stale/low urgency,1,0,70394,,,,0,f9bbc461306621cf43debeadefaab8757d7b943d91c2c4e8265c4d3c318a471f
thread:issue:70397,issue #70397: `torch.empty_strided` works when the stride is negative!,stale/low urgency,2,stale/low urgency,1,0,70397,,,,0,bb51ba50cca39136ad67f4e869c0f890127fde26c909e59170a083d300fadab1
thread:issue:70398,issue #70398: `torch.broadcast_to` can create tensor with negative dimension.,stale/low urgency,2,stale/low urgency,1,0,70398,,,,0,f6f794516f3f391a6a7649188701f7dd578cc62b84022c4114717be993fa71e2
thread:issue:70413,"issue #70413: PyTorch crashes without an error message, when running this code snippet with torch.tensor subclassing & forward hooks (Not sure what the exact cause is, but the code snippet reliably causes it)",stale/low urgency,2,stale/low urgency,1,0,70413,,,,0,337cad40ad2da4f19533d3491fe8d629d94f88859f6f534c5a16ba4bfd3b3b5c
thread:issue:70416,issue #70416: libtorch cuda use too much system memory,stale/low urgency,2,stale/low urgency,1,0,70416,,,,0,0dad89288425df00e0c78f3096ccdb4c1778035fe1eb52b80d0d461caa0c8b47
thread:issue:70419,issue #70419: LazyModules `cls_to_become` field exposes implementation detail,stale/low urgency,2,stale/low urgency,1,0,70419,,,,0,f6f45cdedd9860d6365e3b6690a4b4e04e0806c33fde89285e2b956c1237f926
thread:issue:70446,issue #70446: Exporting the operator prim_DictConstruct to ONNX opset version 13 is not supported,stale/low urgency,2,stale/low urgency,1,0,70446,,,,0,fe28820a5bfdfa0a2a3845dbe3fca226c4cdc84a791beefc5ffe5fc110807771
thread:issue:70447,issue #70447: The same code can not be reproduced on multiple GPUs with dataparallel,stale/low urgency,2,stale/low urgency,1,0,70447,,,,0,4aa81ed1b52b8a47b7ad269315d14513189af10ea65f5f785d545e7254d68577
thread:issue:70449,issue #70449: [FX] Detect attribute mutation during tracing,stale/low urgency,2,stale/low urgency,1,0,70449,,,,0,203157a3fcc747561fafb7cfc1b9c217415cae16b4d6a313f3053138412a2452
thread:issue:70454,issue #70454: Building docs locally fails,stale/low urgency,2,stale/low urgency,1,0,70454,,,,0,f631220eb9a062681bb6b82c98d0db8060605035e1fd3f69632a7a97b06e8054
thread:issue:70455,issue #70455: Make it possible to remove all hooks on a specified module without needing the hook handles,stale/low urgency,2,stale/low urgency,1,0,70455,,,,0,8b61d0765491153a12bc7210a8fde251034e5ccbf1694dbc981ffda95408631f
thread:issue:7047,"issue #7047: ""Undefined symbols for architecture arm64"" when linking libcaffe2.a for iOS",stale/low urgency,2,stale/low urgency,1,0,7047,,,,0,d180e8099865041a03a957e92b2563e4027f7ba9ade18e79c9c85dade5ba3bde
thread:issue:70480,issue #70480: Dynamic quantified Conv2d have accuracy issue,stale/low urgency,2,stale/low urgency,1,0,70480,,,,0,3f3ee43ed3f9a5a2006c9aef63a13387b91d1086b3a5dc07fcbbdf6efa4c73e1
thread:issue:70487,issue #70487: Fusion of Convolution and BatchNorm,stale/low urgency,2,stale/low urgency,1,0,70487,,,,0,aeedb2bd0a9614b47e7ca4475975ffd8c8e96ece709c3ad890011b9bd97a344b
thread:issue:70490,issue #70490: Cuda sync mode input checking is wrong for non-string/int inputs.,stale/low urgency,2,stale/low urgency,1,0,70490,,,,0,0c4edc9ac8ed863849d0d3cb2e2d97c07f5f14d8a13bd9e807d12d54653fbdc6
thread:issue:70498,issue #70498: Multiple invalid summaries in torch.nn documentation page,stale/low urgency,2,stale/low urgency,1,0,70498,,,,0,a1690ae6565a6ef4564f37a637a93bed79828ec561cb9d7201d0dc04369184da
thread:issue:70500,issue #70500: Adding the new Phish activation function,stale/low urgency,2,stale/low urgency,1,0,70500,,,,0,8a64db2b2d023552c96a77296dee0575825368431d59d57a667f8fe6a831f2ab
thread:issue:70505,issue #70505: `UninitializedParameter.to(device='meta')` creates a zero-sized meta tensor instead of remaining uninitialized,stale/low urgency,2,stale/low urgency,1,0,70505,,,,0,dfa8223249c41ed32ef7c274c6fff040b6abd01830e4d02ab84e4b481bd3459e
thread:issue:70513,issue #70513: Conda Repodata.json file not found in Pytorch channel,stale/low urgency,2,stale/low urgency,1,0,70513,,,,0,b66c80554eaee2f37d7b132bc3b9e46e71a5e4ce3f9d05dfd579c08ae7339ef9
thread:issue:70521,issue #70521: tmpxft_00008487_00000000-6_THCStorage.compute_86.cudafe1.cpp:(.text+0x60b): additional relocation overflows omitted from the output when build PyTorch 1.8.2 from source,stale/low urgency,2,stale/low urgency,1,0,70521,,,,0,8280efda2941ae79895c93e049023bc066a5eb51cc9bdeb521d619368d36a31e
thread:issue:70529,issue #70529: BatchNorm on variable-length sequences or batches,stale/low urgency,2,stale/low urgency,1,0,70529,,,,0,7ab904cbf3cc1d36b1080357ae919f6f3a912b6b76cbc687c3147f2b7eb6e420
thread:issue:70540,issue #70540: cumcount / cumulative count,stale/low urgency,2,stale/low urgency,1,0,70540,,,,0,eb18d05951d3f966ee17ac5f25d22b2040b095b416c663d7a3f78af6274ffc90
thread:issue:70555,issue #70555: [feature request] Exponential moving average (EMA) of a tensor across a dimension,stale/low urgency,2,stale/low urgency,1,0,70555,,,,0,0d90983606ad769e1eb245d2de70df9bcee18f8b282764b991f23eed676f5e34
thread:issue:70654,issue #70654: [RFC] UCC integration in ProcessGroupNCCL,stale/low urgency,2,stale/low urgency,1,0,70654,,,,0,db7fcefa8bb32bcebc6b78ae9962d1c302025a332c3f874c0bfe8e63ec22012d
thread:issue:7066,issue #7066: LBFGS contribution,stale/low urgency,2,stale/low urgency,1,0,7066,,,,0,e8279aec8417a753685f6db40ea3abab1ccb1a77a688a23b1e3289b8876172b7
thread:issue:70669,issue #70669: CUSOLVER_STATUS_EXECUTION_FAILED when using the torch.logdet(),stale/low urgency,2,stale/low urgency,1,0,70669,,,,0,c9d9b9a05d67fa273828bcdcb3b6cd6ce3aadbe429e93e4c3ab36d6bd5c3e833
thread:issue:70701,issue #70701: AdaptiveAvgPool1d - RuntimeError: shmem_size <= sharedMemPerBlockINTERNAL ASSERT FAILED,stale/low urgency,2,stale/low urgency,1,0,70701,,,,0,bffdba9513155ec19877b42b1041f882ee1bb814014cc5f116c17b7adfb9d638
thread:issue:70735,issue #70735: named tensor doesn't work with deepcopy,stale/low urgency,2,stale/low urgency,1,0,70735,,,,0,0c04d426093fde0ebf87f598c8925bb0a474e67aec218692b14be971574b3c65
thread:issue:7078,issue #7078: Fail to import Caffe2Backend,stale/low urgency,2,stale/low urgency,1,0,7078,,,,0,8b00972ffb88ce2d207ea3d6e1c94bc2c1af0410b6898e027953292d7fccae87
thread:issue:70870,issue #70870: Add flag for functional.Jacobian to return output as well,stale/low urgency,2,stale/low urgency,1,0,70870,,,,0,5fb82e5c2995639c2782e949b2d07d02691dd8fee6dfdf62cd70e5c2a709db56
thread:issue:70892,issue #70892: Memory leak while training model generated by torch.fx.symbolic_trace() in data parallel mode,stale/low urgency,2,stale/low urgency,1,0,70892,,,,0,5329e7d896cf1f05b4d28e274931ad4dba122f1b66324a21f7b9619042fcd554
thread:issue:70901,issue #70901: Accuracy problem of `torch.batch_norm_gather_stats_with_counts` when `running_mean` is half tensor,stale/low urgency,2,stale/low urgency,1,0,70901,,,,0,256f85b5f662350e6c7ee1d2420c0f2a3b7a0baef20542f9bcc9fc189d6d1fc3
thread:issue:70926,issue #70926: Feature Request: Implement `torch.sparse.spdiags` and `torch.sparse.diags`,stale/low urgency,2,stale/low urgency,1,0,70926,,,,0,34821dde56e8d4373df98c6a0d29efb353b24b182ff4db1a48cd2d6fe1ea624f
thread:issue:70940,"issue #70940: ThreadLocalState::setThreadLocalState is not setting the ""enabled"" flag of SavedTensorDefaultHooks",stale/low urgency,2,stale/low urgency,1,0,70940,,,,0,2064a48ecad8108e1683d7f7c56359ca89ac89cac3e9c23d2ca54e76cbc9024b
thread:issue:70970,issue #70970: argmin/argmax incorrect doc for the first form,stale/low urgency,2,stale/low urgency,1,0,70970,,,,0,bfe8547b07f95495d00a6895b89a6a04d8ecda317c8e8b8fefdd19b5b822a429
thread:issue:70995,issue #70995: Rewrite tests in test_nn to not depend on LAPACK,stale/low urgency,2,stale/low urgency,1,0,70995,,,,0,a09a9b15bffd1fbe4bd862065b5edc512e74edc481486bbd5d59dfd1a07f8f13
thread:issue:71022,issue #71022: installation of pytorch `cpuonly` from conda with `nomkl` installs `mkl`,stale/low urgency,2,stale/low urgency,1,0,71022,,,,0,12cae9402fe2a1647ddd64f71347a5231a732a3f84e389c103143b0b0865e913
thread:issue:71028,issue #71028: Skip LSTM quantization by default in get_default_qconfig_dict and get_default_qat_qconfig_dict,stale/low urgency,2,stale/low urgency,1,0,71028,,,,0,194de1f826d2f42b5f626719a5ef588f62629d750e8e75013ca16cb8eabafccd
thread:issue:71029,"issue #71029: remote failure INTERNAL ASSERT FAILED at ""../torch/csrc/distributed/rpc/rref_context.cpp"":389",stale/low urgency,2,stale/low urgency,1,0,71029,,,,0,bd6d219f50c290d3fd5d3aba421bc9f4c99e634961206e2d7af981c47f7ef4b1
thread:issue:71058,issue #71058: `torch.Tensor.where` cannot work when `y` is float,stale/low urgency,2,stale/low urgency,1,0,71058,,,,0,e21d4a307199883f725bdadd958e981812f6458e5e9f1068cb8df8c490cce1ee
thread:issue:71059,issue #71059: `torch.scatter` will return random value when `input` is empty tensor,stale/low urgency,2,stale/low urgency,1,0,71059,,,,0,6a7d83e9009e5e8a3b0ac7bc5f87046bcbe96dc73615fcf04d18cb6f1bb949f3
thread:issue:71069,issue #71069: C++ torch::nn::Sequential clone() method overwrites child module names,stale/low urgency,2,stale/low urgency,1,0,71069,,,,0,002a942e0361a893d4dc8407839dfa902f4bedc59e1578ca9690f59e2a2c584b
thread:issue:71071,issue #71071: Feature Request: torch.special.ellipe,stale/low urgency,2,stale/low urgency,1,0,71071,,,,0,d0aed0a84abcc6088433983a3b26f2f361e4645ad4d361d12f4c3cfe5395f63b
thread:issue:71076,issue #71076: Error in `torch.Tensor.logit` documentation,stale/low urgency,2,stale/low urgency,1,0,71076,,,,0,48700349d1a29caf27f8ddcb357e9dd76514f6e6a593480946b3cc09a6dd1c19
thread:issue:71078,"issue #71078: `torch.nn.{Constant,Zero}Pad` unexpectedly fail",stale/low urgency,2,stale/low urgency,1,0,71078,,,,0,8b79db7b447f5939175516e59b337f6e08f1e71e75e9b3b7f40ebc7080a27604
thread:issue:71084,"issue #71084: `torch.{inverse,cholesky}` have wrong shape check of square matrices",stale/low urgency,2,stale/low urgency,1,0,71084,,,,0,fd17efac57d0d0f81679080bea004b705f286feaa36335c6b9f7e4b71ae4c746
thread:issue:711,issue #711: Feature Request: Easier to extend base RNN implementation,stale/low urgency,2,stale/low urgency,1,0,711,,,,0,98d710b083a8045395eba97f28856489696089f9038243c8a4812fab039ed270
thread:issue:71134,issue #71134: Slowdown in torch.distributed.new_group when scaling to large clusters.,stale/low urgency,2,stale/low urgency,1,0,71134,,,,0,75a6f9d55ae159ac145c7f97e6c7e07244fa896c96c2f4fe6669ffcdfb6800c2
thread:issue:71152,issue #71152: clang format hash mismatched for linux64,stale/low urgency,2,stale/low urgency,1,0,71152,,,,0,58910852f1d17af6c48dac027e474a2367e93bd7f2a7b7554903b244f14a3f50
thread:issue:71156,issue #71156: Slow backward for matrix multiplication of two sparse COO tensors on CPU,stale/low urgency,2,stale/low urgency,1,0,71156,,,,0,bca84bbf16af18d5758c3cb60645b5000f77dcd6e48da5c84bc8db2126c6058e
thread:issue:71203,issue #71203: Keys of a `ModuleDict` cannot have the same name as existing `ModuleDict` class attributes.,stale/low urgency,2,stale/low urgency,1,0,71203,,,,0,30b5aa8a3292eff88c81f309e0406ad4404fcc0c530a3d246a79a96cc70cd72b
thread:issue:71204,issue #71204: `torch.diag` unexpectedly fails,stale/low urgency,2,stale/low urgency,1,0,71204,,,,0,246981dea03514b23bb08e94558b9321fb7de816af67b5efc1b1cafd784c4f51
thread:issue:71205,issue #71205: Possible security issue of `torch.hub.load`,stale/low urgency,2,stale/low urgency,1,0,71205,,,,0,772c09ce85eb89d6477130a0ba04be36c126b840cd63f2b58c8cf3ae06cd79b0
thread:issue:71211,issue #71211: Lack of type check in `nn.functional` APIs,stale/low urgency,2,stale/low urgency,1,0,71211,,,,0,a926cc54f04f9609be922015b19f51520823d165d08d0a87eade8311bea25142
thread:issue:71228,issue #71228: 1.10.11 fails to compile libtorch_cpu with -fopenmp,stale/low urgency,2,stale/low urgency,1,0,71228,,,,0,5083797bac85d9a39fe0b2a2ca6e62beb25f4a1f60d8a8baa4641bb44977c992
thread:issue:71266,"issue #71266: ""Memory Leak"" when creating an iterator from a tensor",stale/low urgency,2,stale/low urgency,1,0,71266,,,,0,d0da142347065f78a2d205054c49cf4e60c1c6cd6047da7705f5bc3865ba9cf6
thread:issue:71300,"issue #71300: Kernel fusion for Gather, Apply, Scatter (GAS) model",stale/low urgency,2,stale/low urgency,1,0,71300,,,,0,823d6d8489f1fb0a425e9707abfca08cc0b6d2884311dcb7b13ac55882b2c1af
thread:issue:71301,issue #71301: Why AMP make backward speed more slow?,stale/low urgency,2,stale/low urgency,1,0,71301,,,,0,53c4002756e88f20b9f78ca6b11d9625acac2c3bc3cf8bc04618e9006daf8995
thread:issue:71305,issue #71305: Convert a tensor with type caffe2::TypeMeta to std::vector,stale/low urgency,2,stale/low urgency,1,0,71305,,,,0,f9a875904be23efbcb021bea8eed4f83a6e2d684eebd81d1370a76a3f49ee87c
thread:issue:71317,issue #71317: [torch.deploy] Using zipped torch modules in torch.deploy interpreter,stale/low urgency,2,stale/low urgency,1,0,71317,,,,0,ad32ff90ec1dec194d64891805e494f7619ada80557c2682659a13cc455e94c6
thread:issue:71322,issue #71322: Allow users to pass use_reentrant=False to checkpoint_sequential,stale/low urgency,2,stale/low urgency,1,0,71322,,,,0,e971331458362831c86e234bf29191d7dad2748dcfd659c8d4a356578a8d765f
thread:issue:7133,issue #7133: [Caffe2 warpctc] How to use the offered warpctc?,stale/low urgency,2,stale/low urgency,1,0,7133,,,,0,3e985db3422672c645ae65b9b026d06b1a746ef6dff4e2d080d3950ca6613d5f
thread:issue:71330,issue #71330: [RFC] Gossip SGD (as a DDP Communication Hook),stale/low urgency,2,stale/low urgency,1,0,71330,,,,0,ec25e3a793e58cba90d26a4e65751535e11b9e8cce5357d9016416ca6fa04737
thread:issue:71331,issue #71331: Change the order of checks for tensor indexing errors,stale/low urgency,2,stale/low urgency,1,0,71331,,,,0,281dd4e9c26f2adf3702f20e54f783ad03e42cfd670f3827b699a74be468a387
thread:issue:71340,issue #71340: locally installed PyTorch upgraded or superseded on windows/macos conda tests,stale/low urgency,2,stale/low urgency,1,0,71340,,,,0,8f79a446c8dbcf62b7fa1649d539b3a64c17ed3e7a3fc5c7e1091ecfa3870029
thread:issue:71364,issue #71364: make pytorch support different hardware acceleratioin configuration,stale/low urgency,2,stale/low urgency,1,0,71364,,,,0,bcaf31c1c7ff0be21bfba2f4eb5a3c65f14a607b57c33a3a3983de8bb1b1964a
thread:issue:71366,issue #71366: nn.Batchnorm1d input shape notation inconsistency,stale/low urgency,2,stale/low urgency,1,0,71366,,,,0,46a9cd5d8f04c8cf25926b063d72e5545bb91f90ffba7e95875dbd5eb453083e
thread:issue:71367,"issue #71367: F.cross_entropy do not have a deterministic implementation, adding deterministic support for this operation.",stale/low urgency,2,stale/low urgency,1,0,71367,,,,0,6427d30254e95dfb9442d907e6ad2974cbdd8767677c551e343a912d3d48f281
thread:issue:71377,issue #71377: is_alias_of support for storageless tensors,stale/low urgency,2,stale/low urgency,1,0,71377,,,,0,622e686005f36212812a9a045c748be839d2cf3b04bb60fa1ca41fe3fac0ec3a
thread:issue:71383,issue #71383: Allow to shutdown persistent workers,stale/low urgency,2,stale/low urgency,1,0,71383,,,,0,1851e30eeb6b039549bbddc27c86167187f8b163935531182115f383d3eb46ff
thread:issue:71396,issue #71396: a lot nightly builds are canceled due to VM errors since Jan14,stale/low urgency,2,stale/low urgency,1,0,71396,,,,0,8b7e9bb8f48874005e8abdf505a8b6d2c5a2a05dbf846884cb2fef82f333f098
thread:issue:71407,"issue #71407: Push to fork failed with cryptic ""refusing to allow a Personal Access Token to create or update workflow `.github/workflows/run_torchbench.yml` without `workflow` scope""",stale/low urgency,2,stale/low urgency,1,0,71407,,,,0,3b3ff9fcc1d0b548653286213efd0c893965e1c718b3a70010ac2058863b1c68
thread:issue:71414,issue #71414: Gamma distribution returns some wrong extreme values,stale/low urgency,2,stale/low urgency,1,0,71414,,,,0,100ef857a4b7b814904cdecaf85be68603c9ea3b16a98fbdf2c890f7fadc3ea8
thread:issue:71471,"issue #71471: [pytorch1.5.0] subprocess.CalledProcessError: Command '['cmake', '--build', '.', '--target', 'install', '--config', 'Release', '--', '-j', '6']' returned non-zero exit status 2.",stale/low urgency,2,stale/low urgency,1,0,71471,,,,0,fb8091f9a41797d6ed9c78bae96d4723ef18a60f5f7d1f79256ba522df55733f
thread:issue:71472,issue #71472: Feature request: Add complex support to `torch.nanmean`,stale/low urgency,2,stale/low urgency,1,0,71472,,,,0,bd7eec50fb89af50474729953350d12db5bc53852c4e0147d7533402e34874e3
thread:issue:71473,issue #71473: pytorch installation error with cuda!!,stale/low urgency,2,stale/low urgency,1,0,71473,,,,0,5281a28a341b1b1cf2e6b4d97fd90eaff3ed8225426f1acfba4370337268e1f1
thread:issue:71475,issue #71475: The signature of `torch.nanmedian` in the doc is wrong,stale/low urgency,2,stale/low urgency,1,0,71475,,,,0,07a20b7938f1aaa167a4f009317496bd1f60b735a56fc2ce139f0fc850232979
thread:issue:71495,issue #71495: Memory Leak in PyTorch 1.10.1,stale/low urgency,2,stale/low urgency,1,0,71495,,,,0,18b496ba6e94787faa314825ceaf85b2df3d5ac6f4d3f61bfc81cf31b87c7a69
thread:issue:71518,issue #71518: Compilation instructions are not exhaustive: <<parameter packs not expanded with ‘...’>> on Fedora 35/CUDA 11.6,stale/low urgency,2,stale/low urgency,1,0,71518,,,,0,fe896c1c5af2730ca355f66e7977eb99e789678999d3e904b4d84d69665435aa
thread:issue:71541,"issue #71541: `torch.sub` behaves differently with `add`, `mul`, `div`",stale/low urgency,2,stale/low urgency,1,0,71541,,,,0,e2ff8ad6e2bafec3d08aa601da5894fb9c045f1f73226582a107df483df1b604
thread:issue:71543,issue #71543: Initialize DataLoader workers in parallel,stale/low urgency,2,stale/low urgency,1,0,71543,,,,0,2fc5363b006470079387f9e1143cf29c9cce4648d80e5598018bab2ab71bbed4
thread:issue:71552,issue #71552: [docs] Tensor.uniform_ docs are not clear about whether from/to boundary values are included in sampling or not,stale/low urgency,2,stale/low urgency,1,0,71552,,,,0,f98f75884d53b2dc227308accef16c84c5b9c173f27657c1656b8b2b3fe76bd1
thread:issue:71575,issue #71575: Better Engineering: test_..._mem_overlap in test_torch.py should be ported to ErrorInputs,stale/low urgency,2,stale/low urgency,1,0,71575,,,,0,e57c8b6ef074ed65412e3ff2275ab43ab5738208602414f89f384b6e8a6283b4
thread:issue:71629,"issue #71629: `torch.{max,min}` have strange error message when `input.numel()==0`",stale/low urgency,2,stale/low urgency,1,0,71629,,,,0,55a5deb390616c7101aeec232db662179a4aa377c7610cf5b7e30db6ced622a2
thread:issue:71630,issue #71630: More than 4 Dimensions for Channel Last Memory Format,stale/low urgency,2,stale/low urgency,1,0,71630,,,,0,297af4db0cffd522c758eaed43403585617bf9e47ea939c22ac3260c6899956b
thread:issue:71633,issue #71633: Negative Exponents of Int tensors result in output of zero,stale/low urgency,2,stale/low urgency,1,0,71633,,,,0,2bdf23068c1976087cd4993c4e732b96ca2b80cadafa7be9dfeaca4151b19b1c
thread:issue:71635,"issue #71635: `torch.nn.functional.{instance, batch}_norm` trigger INTERNAL ASSERT FAILED when input is empty tensor with `complex32`",stale/low urgency,2,stale/low urgency,1,0,71635,,,,0,2b4154b1a672a2d5c7edf3b4040492c0aeb058f90baa0132686ad925a7c680b0
thread:issue:71660,issue #71660: Build release binaries with USE_GLOG=ON by default,stale/low urgency,2,stale/low urgency,1,0,71660,,,,0,f74ec4a0f56693dab26c98681d446b2297b81ce97baf4295e2e08c3507a4be9e
thread:issue:71671,issue #71671: Many APIs `INTERNAL ASSERT FAILED` when promoting `complex32` dtype,stale/low urgency,2,stale/low urgency,1,0,71671,,,,0,63805833c0a81fdb80b407c22cf05bc8021f3196a01f509cb9f481d051ffc0f2
thread:issue:71678,issue #71678: torch.bmm backward with sparse input,stale/low urgency,2,stale/low urgency,1,0,71678,,,,0,203ba58867cb4620ee4d64b104152910e0874d8aabd8975e4ee187ebc2c96a79
thread:issue:71682,issue #71682: _GLIBCXX_USE_CXX11_ABI=0 does not work when building from source code,stale/low urgency,2,stale/low urgency,1,0,71682,,,,0,f48df9743542d099599fb31b40e7d3e288f9be0db5891c57f94d44ad90cfff05
thread:issue:71689,issue #71689: torch.distributions.categorical.Categorical does not work with 0 batch size,stale/low urgency,2,stale/low urgency,1,0,71689,,,,0,05cf8595a3f685c53feb4c9fb4616a56a2ba3c9705b760102db0166592a5100e
thread:issue:71697,issue #71697: Libtorch dlls delayed loading,stale/low urgency,2,stale/low urgency,1,0,71697,,,,0,d18ceaba62e8b90aca5cf61e83402273b56b82921e5be292842c22a55d95ab11
thread:issue:71698,issue #71698: [libtorch]can not save a vector<int> to AutogradContex->saved_data.,stale/low urgency,2,stale/low urgency,1,0,71698,,,,0,9c7592389bce03ffca5a86cef800498127bc7a1238ea8f4ab5a3e44b3ba24024
thread:issue:71715,issue #71715: [FX] Support call_method in NormalizeArgs,stale/low urgency,2,stale/low urgency,1,0,71715,,,,0,e9f24a19a95b3e547c8a57cf3652a29d37ae58fbfaff229b9102b03293302c2d
thread:issue:71721,issue #71721: Improve performance of index for quantized ops,stale/low urgency,2,stale/low urgency,1,0,71721,,,,0,5bea39a054dc88f8368508ef47c5fe1ecf46d1a23793f34487df0aed0442648f
thread:issue:71760,issue #71760: Adapting the citation style according to GitHub's CFF,stale/low urgency,2,stale/low urgency,1,0,71760,,,,0,e2c33b62af2dc9ed43e5ed694186735f56b3399972a8dbe1ed8eedabbb2750a6
thread:issue:71774,issue #71774: matmul returns uninitialized memory for int64 tensors with inner dimension of zero,stale/low urgency,2,stale/low urgency,1,0,71774,,,,0,e6e7957bad1f49cb49cb8d341c55d472069e312a8c4a51bf340e53480ff47de8
thread:issue:7179,issue #7179: [caffe2]when i do CMAKE,stale/low urgency,2,stale/low urgency,1,0,7179,,,,0,a61cc9f2291c8d7a26becaee7762ce1f4d740ec1d116d3b575f5bc225760cb72
thread:issue:71872,issue #71872: Clarify the behavior of DataLoader sampler and batch_sampler parameters,stale/low urgency,2,stale/low urgency,1,0,71872,,,,0,37f36553edd5bd87268f983563f2d3ec01872f4519606efa1528b0e5a4cb58fe
thread:issue:71877,issue #71877: Add `output_size` argument to `Upsample` forward method (just like for `ConvTranspose` Modules),stale/low urgency,2,stale/low urgency,1,0,71877,,,,0,6090be60bb795917c4595addc1eb8a390328ec16420c46bdad507f820b30580e
thread:issue:71889,issue #71889: Make scale/zero_point accessible from TorchScript traced module for QFlaotFunctional modules,stale/low urgency,2,stale/low urgency,1,0,71889,,,,0,23c78e0ff6ad1ea620f73ee23819c2e7947feee3776642318a664ae81284c926
thread:issue:71896,issue #71896: CUDA Graph API Improvement,stale/low urgency,2,stale/low urgency,1,0,71896,,,,0,29018aa24561b92cd9cdfa86048a6834795e020c7e8be74d4ef5e3a0e32f098a
thread:issue:71918,issue #71918: Support mixed python scalar/tensor types for torch.clamp's min/max args,stale/low urgency,2,stale/low urgency,1,0,71918,,,,0,171bf3ea30e49b9b7d36a65b791f66cddec84216ebc98556d306194891cef3a3
thread:issue:71919,issue #71919: Graph Mode Quantization does not keep NamedTuple Information,stale/low urgency,2,stale/low urgency,1,0,71919,,,,0,851bbc36202d851ab29eb4cf069a946ea2dbc170234d7c076d42cd76cbcc0638
thread:issue:71951,issue #71951: Multivariate normal defined by eigendecomposition,stale/low urgency,2,stale/low urgency,1,0,71951,,,,0,b7d6ddee67c11f6d02f0b37f5fd1736907adbbbe2c2ee1ca434ba6adced52128
thread:issue:71978,issue #71978: from_blob / make_tensor support MemoryFormat,stale/low urgency,2,stale/low urgency,1,0,71978,,,,0,73acc204d81f0736095aad170565e8004e41e74d3dd797a9448dff891c7057b9
thread:issue:72022,issue #72022: [PT-D] To make the ShardedTensor Reshard more generic,stale/low urgency,2,stale/low urgency,1,0,72022,,,,0,6815a28d30b688d215bfc8ad3c7a5e27872c7d06d29260da372d93a7f37ca3f5
thread:issue:72041,issue #72041: RuntimeError: Trying to backward through the graph a second time (or directly access saved tensors after they have already been freed). Saved intermediate values of the graph are freed when you call .backward() or autograd.grad(). Specify retain_graph=True if you need to backward through the graph a second time or if you need to access saved tensors after calling backward,stale/low urgency,2,stale/low urgency,1,0,72041,,,,0,e931ff0b5178baada65346401aab0a496304258dee1fffcfdad173d00fa023e0
thread:issue:72046,issue #72046: Building PyTorch with Vulkan backend don't work,stale/low urgency,2,stale/low urgency,1,0,72046,,,,0,9a333d4e10348106ce2ec65ed8f45fe9cffb404ed3e723e27da2060e7276fcdc
thread:issue:72055,issue #72055: index_add : Inconsistent between CPU and CUDA,stale/low urgency,2,stale/low urgency,1,0,72055,,,,0,853f902d6b4958b197be5b25de4f4df13214a77e5dfe6e60cec81ca9e9ae774b
thread:issue:72061,"issue #72061: `svd_backward`: does not handle inputs of rank `r < min(m, n)`.",stale/low urgency,2,stale/low urgency,1,0,72061,,,,0,4825e7f725b041364757213c689e286bd898a1e973c13f8991bcfe7c83c843c7
thread:issue:72065,issue #72065: Batched sparse-sparse matrix multiplication/ sparse torch.einsum,stale/low urgency,2,stale/low urgency,1,0,72065,,,,0,86fc04744d22ad99ab5ac27490e676d7fe489ca5cbfba5d1acd34fe40f006aeb
thread:issue:72117,issue #72117: Force PyTorch to clear CUDA cache,stale/low urgency,2,stale/low urgency,1,0,72117,,,,0,5076caa40bca0705b7ebbd3f027c137138bfc8873bb1a61f7a0ee15f5dd2d667
thread:issue:72134,issue #72134: Add support for `complex` `mean` to the `normal` operator,stale/low urgency,2,stale/low urgency,1,0,72134,,,,0,5bcafe92c525dcb5615547557de2c00bcb86966f1af268cb8124d6a9be779ad9
thread:issue:7214,issue #7214: Do not put system paths in RPATH,stale/low urgency,2,stale/low urgency,1,0,7214,,,,0,ad6b27b6e81d6e43382b9d45ae830c0176bff7108c370c85a01f4502ddd3940d
thread:issue:72179,"issue #72179: Mechanism for Tensor subclasses to ""disable autograd""",stale/low urgency,2,stale/low urgency,1,0,72179,,,,0,501888a73254f212de85b0702974e8b803e309443a5171e7e49a80e8d9632299
thread:issue:72194,issue #72194: `TestCase.assertEqual` has `equal_nan` default to `True`,stale/low urgency,2,stale/low urgency,1,0,72194,,,,0,bf91dc9ba36a0cb09aeac6ba4bf33260b7a7fa52848524790171b85968a890b6
thread:issue:72240,issue #72240: recurrent neural network module,stale/low urgency,2,stale/low urgency,1,0,72240,,,,0,03f5bcf0bbb6fe2cd4f9ed5fd342e8267bf372c063ae9e9c1a0e68a69eb4907a
thread:issue:72257,issue #72257: `torch.hasnan`,stale/low urgency,2,stale/low urgency,1,0,72257,,,,0,a900b14bd736631de160121526255ec911d8abb72c6492e4aff58360b1009dbb
thread:issue:72263,issue #72263: Removing deprecated `cpp_custom_type_hack`,stale/low urgency,2,stale/low urgency,1,0,72263,,,,0,5f41fea2f5697f8c2b0888c6d0fc48d81c15fde49c6bcc63582328c28642e823
thread:issue:7227,issue #7227: [Caffe2]Install problem,stale/low urgency,2,stale/low urgency,1,0,7227,,,,0,4a9cf599b96098426c7d0d2789efe2ccd46646656861703552c8f2336daec36a
thread:issue:72283,issue #72283: Add Sparsemax function to libtorch,stale/low urgency,2,stale/low urgency,1,0,72283,,,,0,07c02c04320239d34f74cb6e9b58485c3ec9600aa28c87e1065d45e81b0c0c0c
thread:issue:72317,issue #72317: Feature request: a mode to disallow calling prototype or beta features,stale/low urgency,2,stale/low urgency,1,0,72317,,,,0,7d63db86dc72ddf8f132e2bf7470253f81e9b6adb204019379f31eaf317397cb
thread:issue:72332,issue #72332: Ability to assign to `tensor.require_grad` might lead to bugs,stale/low urgency,2,stale/low urgency,1,0,72332,,,,0,3a958c111c7a84557f99cf5fea7a142b89da638161b11e72b51c1b20540c231a
thread:issue:72393,issue #72393: Some system-installed headers are mistakenly used.,stale/low urgency,2,stale/low urgency,1,0,72393,,,,0,8a70e119abeb04675e4806e119de197763c29e77606a5b7e49f76ec9a8ebc2e9
thread:issue:72408,issue #72408: Matrix multiplication is 30 times slower for integers than floats on CPU,stale/low urgency,2,stale/low urgency,1,0,72408,,,,0,855993f5201e11252eacb227d8bb16df86627c7bdc21a406cfe923c8d17a0016
thread:issue:72421,issue #72421: Feature Request: A simpler decorator for disabling mixed precision,stale/low urgency,2,stale/low urgency,1,0,72421,,,,0,eb478d24867bcc520238daa7ba3c25e9c61478c00c4e14163dcf90c767ca6660
thread:issue:72448,issue #72448: Use Module `__getattr__` in `torch.ops` and friends,stale/low urgency,2,stale/low urgency,1,0,72448,,,,0,f8aea7d9132dd436decf3a1ec2c64f1d19d04058ab8a21b97b32b583fd830cf4
thread:issue:72450,issue #72450: Memory parity with JAX attention,stale/low urgency,2,stale/low urgency,1,0,72450,,,,0,656a8fffb1007f102c91982465f6e7e53128544d8fe823bf42818e8486e63f19
thread:issue:72452,issue #72452: Re-raise the exception when the `forward` of a parametrization raises,stale/low urgency,2,stale/low urgency,1,0,72452,,,,0,0a288548a03bbb43967743bd16c18858840aaedec67db2a2b2da17e32473aba6
thread:issue:7250,issue #7250: feature request: support for new/future hardware accelerators,stale/low urgency,2,stale/low urgency,1,0,7250,,,,0,f169403a9dddc2e0864836789755d94a068530c0f156f42c7d010fe26b2a6e86
thread:issue:72537,issue #72537: Feature: Add tril_embed and triu_embed,stale/low urgency,2,stale/low urgency,1,0,72537,,,,0,9f795bfa46bb43b255db4defda135d32f775d1259f909c5cc9c79d4f2c02a20c
thread:issue:72545,issue #72545: Split up torch.distributions docs into multiple pages,stale/low urgency,2,stale/low urgency,1,0,72545,,,,0,3fc075365747c25765ede8bd250005e7f3b8337bae077f0bc36cdc0845dde76a
thread:issue:72555,issue #72555: Enforce quotas on CI users,stale/low urgency,2,stale/low urgency,1,0,72555,,,,0,fc8274a184b1f781402139a014a6413d07ad003dfb59d90679598010248b9469
thread:issue:72580,"issue #72580: Setting a list of Modules as an attribute does not work like setting a Module as an attribute, and throws no warning",stale/low urgency,2,stale/low urgency,1,0,72580,,,,0,1669d4832eaa3bda6b3659b09911da331031b0beb6a41154f8a027c1f1c61b6a
thread:issue:72581,issue #72581: Zero-copy on shared memory of NVIDIA Jetson devices,stale/low urgency,2,stale/low urgency,1,0,72581,,,,0,8fe863758ba763326852b0f002f31f1a4138f1c8c44f7eafe988de486606c492
thread:issue:72582,"issue #72582: Avoid using thrust:: directly, use THRUST_NS_QUALIFIER:: instead",stale/low urgency,2,stale/low urgency,1,0,72582,,,,0,34431f4076ffe178a33048df95af94523deb8c3d3c88b5cc73190b7f01d2d81b
thread:issue:72643,issue #72643: upstream `apex.normalization.FusedRMSNorm`,stale/low urgency,2,stale/low urgency,1,0,72643,,,,0,f9189e718f96d4f4538f27816e4dc080e6fa42cddedeae98ebc883b8e1cffe9c
thread:issue:72650,issue #72650: Some inplace ops don't raise on incompatible shapes and meta device,stale/low urgency,2,stale/low urgency,1,0,72650,,,,0,2ccbc6ad98360d5e1ded21700d51dba2a4406bfebc62573c09b35571a359ffbd
thread:issue:72672,issue #72672: nn.functional No-batch-dim support should have OpInfo examples,stale/low urgency,2,stale/low urgency,1,0,72672,,,,0,a0465c31b95cee61d11996430e8a9cf1ce0dbafeff98eebe2177d47beb4bc4d3
thread:issue:72709,issue #72709: Feature: Use iterative refinement algorithm from cuSOLVER for the least-squares solver on CUDA,stale/low urgency,2,stale/low urgency,1,0,72709,,,,0,0d224082bdad13a91d6beb4b8b88e0b84995a3b03f5ce6fa091ca523a13ae48e
thread:issue:72711,issue #72711: Feature request: Implement `gelsd` backend for `linalg.lstsq` via `linalg.svd`,stale/low urgency,2,stale/low urgency,1,0,72711,,,,0,936b0ddb6fe9f99418b186f2f51ba9c20ac71a0e326537afabcc916eca8055e8
thread:issue:72714,issue #72714: Add `pct_end` parameter to `OneCycleLR`,stale/low urgency,2,stale/low urgency,1,0,72714,,,,0,812238d5a374e78319e95a7f40e610d5dc55850581cd2c6182d1c5e17b4c1a95
thread:issue:72737,issue #72737: Pattern Matching with Tensors,stale/low urgency,2,stale/low urgency,1,0,72737,,,,0,72715e48e4c11c95f11b0f6f79ca27f83a1bd27539ed5684d5e854d55c9f4197
thread:issue:72749,"issue #72749: pytorchmergebot doesn't react to comments left from ""files"" tab",stale/low urgency,2,stale/low urgency,1,0,72749,,,,0,37219e556420aeac8ddb22d2a9814f3ddb06fe9818141c25dfca816c525b8bdc
thread:issue:72766,issue #72766: Feature Request: Deterministic MaxPool3d and AvgPool3d,stale/low urgency,2,stale/low urgency,1,0,72766,,,,0,331dc47daee070e0b6d76d475094fb2d54dcece23d91e2ae997b85654add649f
thread:issue:72768,issue #72768: Add NCCL and MPI version printing to torch.utils.collect_env,stale/low urgency,2,stale/low urgency,1,0,72768,,,,0,0e793a3694c694d9ec548c1243660a5fbf5e025b2cc97914383b3ebbfeeb7b55
thread:issue:72784,issue #72784: Torch version in docker container does not match tag,stale/low urgency,2,stale/low urgency,1,0,72784,,,,0,bddda7dcbb7303e7010bc46e01fe04338a61b2d4f9b743c5778000535fd57d03
thread:issue:72788,issue #72788: Docs bug: type annotations for linspace (and logspace) start and end arguments is wrong,stale/low urgency,2,stale/low urgency,1,0,72788,,,,0,95d25613bf36bdda1f774793f80ff9ccfa7dbcb66dd3e7c099ec9eabca5db484
thread:issue:72791,"issue #72791: why ram memory surges while loading model, with change in torch load device from CPU to GPU",stale/low urgency,2,stale/low urgency,1,0,72791,,,,0,ad073a7cffa0eaaa9725f4a08f3a9a27a8f9dd6ee6bb2970354a58dcb349ea64
thread:issue:72814,issue #72814: Standardize Naming for Workflows/Jobs,stale/low urgency,2,stale/low urgency,1,0,72814,,,,0,d9b97580534916a98d1297be833082b3b4a4915492ed5954ff29e9954914e4f8
thread:issue:72821,"issue #72821: `{Batch,Instance}Norm{1,2,3}d` works when `num_features != C`!",stale/low urgency,2,stale/low urgency,1,0,72821,,,,0,c1f6865ef7c45cc66476788939b67b25e0c5a176169f55e1b8cde2f6e6d4ced9
thread:issue:72850,issue #72850: N-dimensional Convolutions,stale/low urgency,2,stale/low urgency,1,0,72850,,,,0,ed027e9946de14f95fb2250043fbfe040ddcb652ef7bf2d6b8ec0112719bf418
thread:issue:72880,issue #72880: Bazel fails in an obscure way if submodules are not initialized,stale/low urgency,2,stale/low urgency,1,0,72880,,,,0,8b33cc8309df070407fce00b167cfad44515025de28d92ecbdba747f06688b8c
thread:issue:72897,issue #72897: Hanging Validation,stale/low urgency,2,stale/low urgency,1,0,72897,,,,0,12447e85fb843dac4d682227d4ccd6ffdf1b0eaa24278ba19202e9718d0b46af
thread:issue:72909,"issue #72909: `torch.svd_lowrank` should set the default value of `q` as `min(6, m, n)`",stale/low urgency,2,stale/low urgency,1,0,72909,,,,0,2f109d377e649fce4220cdc8c7e9a8a52c030e3345db3969fbce0a5191841611
thread:issue:72915,issue #72915: Is the current behavior with addcmul and integer dtypes intended?,stale/low urgency,2,stale/low urgency,1,0,72915,,,,0,bd40e1e1de8b5e5babb35cf15e0ae5ceccbd8e70322373956e420ee85f9f3841
thread:issue:72933,issue #72933: Why there are 8 flavors of iOS build jobs for every commit,stale/low urgency,2,stale/low urgency,1,0,72933,,,,0,dd273aafa1d5eb9fceef20652193aa40254857a91b758c4633e9fcf6bed68092
thread:issue:73003,issue #73003: pin_memory hangs instead of throwing,stale/low urgency,2,stale/low urgency,1,0,73003,,,,0,8e7ae5986fd7102abdf6852cc121a91a8ac0c023892dd44ebbcd623987f7f854
thread:issue:73007,issue #73007: Feature: support better rendering for ..deprecated Sphinx directive,stale/low urgency,2,stale/low urgency,1,0,73007,,,,0,f073109fb63e929cc44799fe4b70ab1597a8f405f02646f2b30276139222d068
thread:issue:73016,issue #73016: Unable to build and use libtorch function via pybind11: undefined symbol error upon import,stale/low urgency,2,stale/low urgency,1,0,73016,,,,0,4093b924ce79eae6ac021320a8d2f44bd0cd9a57ea89b9ca0aeab852173e67fa
thread:issue:73034,issue #73034: Is it intentional that PyTorch linux binaries aren't manylinux1 compliant?,stale/low urgency,2,stale/low urgency,1,0,73034,,,,0,c8549012ee3700503140537e7f597d88739b9d06607855fd84cdfc14ccf5325f
thread:issue:73051,issue #73051: Cannot install latest pytorch into Docker on Apple M1,stale/low urgency,2,stale/low urgency,1,0,73051,,,,0,1786bd08a3f154909ca7206d835aed6a5c61acf8d698598d8c10a3faf5a005ac
thread:issue:73062,"issue #73062: Clang Compilation Error: more than one constructor applies to convert from ""ptrdiff_t"" to ""c10::Scalar""",stale/low urgency,2,stale/low urgency,1,0,73062,,,,0,4e1d5015290e0b595184b811f21a14aba77d7b95cc3535d4c8d141b72a0926a5
thread:issue:73065,issue #73065: `index_fill_` accepts wrong dtype for meta tensors,stale/low urgency,2,stale/low urgency,1,0,73065,,,,0,32daba3f218f5281903487b756711427b4edb75e219977702689180ac13f9d66
thread:issue:73070,issue #73070: `enumerate_support` for continuous distributions,stale/low urgency,2,stale/low urgency,1,0,73070,,,,0,daafb14f0d10cc9f7534909a367eea448544edb552bcfaebe185883bf73d6ae5
thread:issue:73083,issue #73083: Improve test_test_history.py,stale/low urgency,2,stale/low urgency,1,0,73083,,,,0,7c95ca09c9ec9b633e79b4af16e155b89bf51820fb68aeb1c5da76638261ccf5
thread:issue:73121,issue #73121: make c++ logger preamble meaningful,stale/low urgency,2,stale/low urgency,1,0,73121,,,,0,9384d409fc4696d58ea6e89355b11c6970db7385c78b5107fded2fa3236ad495
thread:issue:73131,issue #73131: Support writing tensorboard traces to AWS S3 (and other cloud storage services) in profiler,stale/low urgency,2,stale/low urgency,1,0,73131,,,,0,d46b7d6c533712e53813eaab69950fbe5ed697827fc5a180eb98f3e750394ab0
thread:issue:73134,issue #73134: Test functionalization doesn't run,stale/low urgency,2,stale/low urgency,1,0,73134,,,,0,368e3d671eed7be8f0834eabf03dff11183eba22f6199f3339500cbb8f1f4a35
thread:issue:73137,issue #73137: `create_graph=True` results in grad_fn error for differentiable functions,stale/low urgency,2,stale/low urgency,1,0,73137,,,,0,39135071329ed09e58fd00c7c49b1f6477c1a942de613784110d57b2aef24384
thread:issue:73141,issue #73141: libtorch: `DistributedRandomSampler` uses the same random order in every epoch,stale/low urgency,2,stale/low urgency,1,0,73141,,,,0,fcc816cd5b02759b98e6306b9b2b938e1029b25b2d737acdcd2f539b317f9902
thread:issue:73150,issue #73150: How do we handle metadata-modifying in-place operators (like `squeeze_`) with `__torch_dispatch__`?,stale/low urgency,2,stale/low urgency,1,0,73150,,,,0,629917fc2b313707be8390b7e2b2cf3ceee0faba18383c7707c9b19f5540bfcf
thread:issue:73160,issue #73160: Preprocessing function for backend coreml is not registered,stale/low urgency,2,stale/low urgency,1,0,73160,,,,0,8b0c56e9508006447ce7c6c1f15982fcd05b8055c0f0d55125885eee6aa7f07e
thread:issue:73161,issue #73161: native_batch_norm and native_layer_norm have strange epsilon behaviors,stale/low urgency,2,stale/low urgency,1,0,73161,,,,0,4e4a90fc62f86e3e9c8dcc9572f00dc889f413eeb194780d3bd5e3718fa5c771
thread:issue:73171,issue #73171: libtorch need operator= in torch::Device,stale/low urgency,2,stale/low urgency,1,0,73171,,,,0,193276bcf8dbb8039a32e5fdde2a47e4dc1bea15475a382f66a5867791594955
thread:issue:73180,issue #73180: Segmentation fault in _sobol_engine_ff_,stale/low urgency,2,stale/low urgency,1,0,73180,,,,0,912548eabdc07364cb5600411b31ed7bab4bccf4f7c99c28ef0b92098a734d8b
thread:issue:73181,issue #73181: Segmentation fault in _sobol_engine_initialize_state_,stale/low urgency,2,stale/low urgency,1,0,73181,,,,0,63400932a5762da97fe281a2ae0abda2cde975c8c2083dfcc4a897147630d42c
thread:issue:73182,issue #73182: Segmentation fault in _sobol_engine_scramble_,stale/low urgency,2,stale/low urgency,1,0,73182,,,,0,f544c83f689abf519e3b88adbd71b23bb0998b3faae53756aef715e5cbe0d017
thread:issue:73196,issue #73196: `torch.pow` errors out on specific input,stale/low urgency,2,stale/low urgency,1,0,73196,,,,0,edb1c643add951cd1b0bf445bc1635e8ae2198d0d4a2a8925c8173d8ba49628b
thread:issue:73218,"issue #73218: This is not completely a bug, but something that will be amazing if can be taken care of regarding nn.DataParallel.",stale/low urgency,2,stale/low urgency,1,0,73218,,,,0,64ba1d1fddf26fda5f58d8645bc9a644a6ab00144b754d6ed5892250b9fcb5c3
thread:issue:73222,issue #73222: Having rrelu functional + module take a generator object to match native functions entry,stale/low urgency,2,stale/low urgency,1,0,73222,,,,0,f58ffbe1b94f608d5cec79b7031b2dbe6c26f3eeb779443187d7d2fc777f5b74
thread:issue:73268,issue #73268: Add detection of interned submodule of externed module in PackageExporter,stale/low urgency,2,stale/low urgency,1,0,73268,,,,0,a59b14e73a7f2a7444717b9687e70c59a716e165515508117d7b1a42a5edfbb6
thread:issue:7330,issue #7330: Issue when importing both retro (from OpenAI) and torch,stale/low urgency,2,stale/low urgency,1,0,7330,,,,0,403b5e6c069be7b251981989b99693e0e152495a2ff3342b143783d657b72552
thread:issue:73304,"issue #73304: fx.symbolic_trace docs says that it ignores different values of concrete_args, but it doesn't",stale/low urgency,2,stale/low urgency,1,0,73304,,,,0,86e10edf5a5c0517043f9f8b6b649d0c1075febab2f0da4fdc72d2fa4c4bffe7
thread:issue:73355,issue #73355: Review and refactor the way libcublas static libraries are linked,stale/low urgency,2,stale/low urgency,1,0,73355,,,,0,f931cf998eed7137563af354eb1568fb707cdb35ccba075995cce5335239a52c
thread:issue:73359,"issue #73359: Pytorch Typing, for Tensor type annotations",stale/low urgency,2,stale/low urgency,1,0,73359,,,,0,5a0b8d63a381abb04e3c389d4e41a2d78c2148c05c2fe22e690579470562b485
thread:issue:73380,issue #73380: elastic/rendezvous: _matches_machine_hostname doesn't resolve hostnames fully,stale/low urgency,2,stale/low urgency,1,0,73380,,,,0,7d63ad92796aa89f2565582cf7fc26ba3a3a365ac21dccf51591d5661af5babc
thread:issue:73394,issue #73394: Make it easier to figure out if packages need to be interned/mocked/externed,stale/low urgency,2,stale/low urgency,1,0,73394,,,,0,d5edb414ba0d6648e71cec97abf0d6113b9f0f0d4fb9dbbbc84e9183f2ba8493
thread:issue:73413,issue #73413: Conv3D consumes lots of memory on Mac with Apple Silicon,stale/low urgency,2,stale/low urgency,1,0,73413,,,,0,005c1c3049ec0e1e3c1a54815cc1c8cb308b3b74e540270598e4bdeaa7c0ae84
thread:issue:73417,issue #73417: MaybeEncodingError: Error sending result,stale/low urgency,2,stale/low urgency,1,0,73417,,,,0,76fe27c7e447e096367d5072a4ecafcdd0f7b10a49408f21347e2b0a3788e9ed
thread:issue:7342,issue #7342: [feature request] [PyTorch] More flexible optimizer API,stale/low urgency,2,stale/low urgency,1,0,7342,,,,0,82496a33ce2d3d1f710cf6fd82b0f9b2752394e217b44df618944ab023bd11cd
thread:issue:73424,"issue #73424: ""munmap_chunk(): invalid pointer"" interaction error with pytorch (< 1.10), pybind, and cv_bridge",stale/low urgency,2,stale/low urgency,1,0,73424,,,,0,1886a4b0de0d85b67c625fd6dc83b94521baef9f915dd73e1cf26452e6d687e7
thread:issue:73425,issue #73425: ufunc codegen support for dtypes that are supported on CUDA but not CPU,stale/low urgency,2,stale/low urgency,1,0,73425,,,,0,1b232dd3ba45604c25a3c1ea82fbb56b536fb84bdfe932ddce70d329434bae05
thread:issue:7343,issue #7343: [memory leak] [PyTorch] .backward(create_graph=True),stale/low urgency,2,stale/low urgency,1,0,7343,,,,0,869641ad836cee78ee436937d435e831526468fa481c3c1d881eec6a698e7f5e
thread:issue:73461,issue #73461: `TestCommonCUDA.test_noncontiguous_samples_pca_lowrank_cuda_float32` fails on A100 due to TF32 operation in `svd_lowrank`,stale/low urgency,2,stale/low urgency,1,0,73461,,,,0,15c8fc078303174fdbf274ba5918ddefef84b840d8fe491eb3348b23d74617ad
thread:issue:73466,issue #73466: [feature request] torch.hub.load_state_dict_from_url to be replaced by a new good general download-a-file function and to also support local paths and google drive links / private github release links,stale/low urgency,2,stale/low urgency,1,0,73466,,,,0,cf82c172b28f5ae0de92407253c260fe000f6cb71cc00fd649b1ef1ef19083fb
thread:issue:73469,issue #73469: torch.fx failed when tracing functions from other Libs.,stale/low urgency,2,stale/low urgency,1,0,73469,,,,0,cd2ca609e34c941b07b381d1f9ee830ad86f1a4ffa2a987443d6713e8fb25b60
thread:issue:73475,issue #73475: different results on each batch_size in torch==1.10.2+cu113 on RTX 3080,stale/low urgency,2,stale/low urgency,1,0,73475,,,,0,c853bf28703a0eea6f7fecafb7663f8d5d908227c9d5f2c58ea44bb4db16ef09
thread:issue:73484,issue #73484: Rename `keep_vars` in `nn.Module.state_dict`,stale/low urgency,2,stale/low urgency,1,0,73484,,,,0,f52ff89889ca43c698dd5d8203f15599afa09f12dcb0e36e73911b0b4c099e77
thread:issue:73501,issue #73501: `index_copy` has different index behavior with `index_fill`,stale/low urgency,2,stale/low urgency,1,0,73501,,,,0,646c5331697f8adfa2eb8864dfe70a4f64fa7b4fcc4f682a4936c9d56b882a5f
thread:issue:73504,issue #73504: `ge` and `div` behaves differently when converting an overflow number,stale/low urgency,2,stale/low urgency,1,0,73504,,,,0,838e75c5bb23dca281771f016839563300974b8abcbf26a4244f49c14f175a49
thread:issue:73505,issue #73505: Figure out what to do with functions that take both Tensor and TensorOptions,stale/low urgency,2,stale/low urgency,1,0,73505,,,,0,61733a764d429971f1b1eba39770c071e2e797e39d11dd50a767565e58ee18a7
thread:issue:73506,issue #73506: Idiom for extensible string printing for TensorImpl subclasses,stale/low urgency,2,stale/low urgency,1,0,73506,,,,0,c272670c5f952ed38b0e7b1e75e0d42184fbe7cb97924bb4a02a24c229be2934
thread:issue:73507,issue #73507: Implement SiLU method for the QuantizedCPU backend,stale/low urgency,2,stale/low urgency,1,0,73507,,,,0,fbe8cb90ff8ecf75b57955c211a8c55e35d17c41259487f027985b991cda2dd2
thread:issue:73517,issue #73517: Teach tools.codegen.api.translate about IValues,stale/low urgency,2,stale/low urgency,1,0,73517,,,,0,faaf6b12f0be7208744a461063ecfb358e730d29b912bd2533de68edbcf6cf3c
thread:issue:7352,issue #7352: Check for F2C convention (for blas) at runtime,stale/low urgency,2,stale/low urgency,1,0,7352,,,,0,bee724a3bc3d96479d32bf90bfa08552630fb1a3180dafbaa7340ded2a53f14d
thread:issue:7353,"issue #7353: Windows MAGMA binary requires explicit linking against MKL LAPACK, or it will silently give incorrect results",stale/low urgency,2,stale/low urgency,1,0,7353,,,,0,ebe50412beaf3e499018b34c7eff14c9409d0aad8b5296a8d77467dabc1b8e9c
thread:issue:73531,issue #73531: Mention docker build process in RELEASE.md and automate building those for release,stale/low urgency,2,stale/low urgency,1,0,73531,,,,0,744941d8da453d114543cdbe3e601615c2184fe76a52f222b7479676af1021eb
thread:issue:73554,issue #73554: Build failure using GCC 11.2.0,stale/low urgency,2,stale/low urgency,1,0,73554,,,,0,7766334ce0ecc8482b65786c94a390f14080fce4b43b3426d535395105d68c17
thread:issue:73563,issue #73563: Pytorch Installation from source fails,stale/low urgency,2,stale/low urgency,1,0,73563,,,,0,68a3067dfcf4fbc844cd853dd0f3086f0fc20e6994312fc26c0dfd839eb94d98
thread:issue:7358,issue #7358: DataParallel on list inputs,stale/low urgency,2,stale/low urgency,1,0,7358,,,,0,bda509c2c67a357638635e730f3616afbc55a207b443da34808e66003f314113
thread:issue:73585,issue #73585: Add optional log_scale argument for torch.distributions.Normal,stale/low urgency,2,stale/low urgency,1,0,73585,,,,0,a205111f61a96f4ac43c7fa913858ac08db237eccc90b463830464ca57de53de
thread:issue:73600,issue #73600: Add a section in DDP tutorial to explain why DDP sometimes is slower than local training and how to improve it,stale/low urgency,2,stale/low urgency,1,0,73600,,,,0,dd8b717b67f1a90c9671c0f677bcdb6c799443be930a679bbb2184e0739c880b
thread:issue:73638,issue #73638: addcdiv is failing the ASAN test for zero divisors,stale/low urgency,2,stale/low urgency,1,0,73638,,,,0,a1d026a853413d94fe66edf4a4ffc63953a4b906dfc70fb9eab920f38b0e5551
thread:issue:73646,issue #73646: Pure C binding/wrapper with libtorch for inference applications,stale/low urgency,2,stale/low urgency,1,0,73646,,,,0,0564064be368c5efbbc21e372be05ff432488fd89121f501b2cfbc365bfc83f4
thread:issue:7365,issue #7365: nn.DataParallel fills None grads with 0,stale/low urgency,2,stale/low urgency,1,0,7365,,,,0,b60a1be0d2fe48e428ce7ad1006f6da54cd8dbc1a1452711bd51c5ba9341e7f1
thread:issue:73665,issue #73665: Nice way to override string representation of tensor subclasses,stale/low urgency,2,stale/low urgency,1,0,73665,,,,0,e2d21abf691c9640f119eae02a67d30d89cdd27b26164ad4912e0d95e5bda1c4
thread:issue:73683,issue #73683: test_broadcast_coalesced_nccl fails on A100 GPUs,stale/low urgency,2,stale/low urgency,1,0,73683,,,,0,721e3b4a2317061301c3587d853ebc055e5565be338254d28678d3fb69b49358
thread:issue:73699,issue #73699: How to get tolerance override in OpInfo-based test?,stale/low urgency,2,stale/low urgency,1,0,73699,,,,0,4bad8c31930d61feb993454499b787d2b5e35482551adc1e514afd2329038d03
thread:issue:73709,issue #73709: DropIT: Dropping Intermediate Tensors for Memory-Efficient DNN Training,stale/low urgency,2,stale/low urgency,1,0,73709,,,,0,22ed4cb643317683dbed9e5b4b4d7c7e949db4e379ce5bf45ad1cec49698ee01
thread:issue:73721,issue #73721: Audit exception rewrapping to ensure stack traces are preserved,stale/low urgency,2,stale/low urgency,1,0,73721,,,,0,1bfd9e6f72426e70534fabec4bb8995b29891b141e28635ea5b9fba241a4662f
thread:issue:73722,issue #73722: Why is distributed RPC using the default pickler?,stale/low urgency,2,stale/low urgency,1,0,73722,,,,0,5788928c153645c8a724704016a99c09edaea22eef8bf75d3049785d8bb8eee2
thread:issue:73738,issue #73738: torch.ao.quantization.ReuseInputObserver should also reuse the dtype of the input,stale/low urgency,2,stale/low urgency,1,0,73738,,,,0,787821d0875f683db5786384d073a460e473329a366e08070086d995a60549f1
thread:issue:73755,issue #73755: CI: Bake as many dependencies as we can in the AMI (windows),stale/low urgency,2,stale/low urgency,1,0,73755,,,,0,84b67f1d930633300ec76bbbe7e2562a37d80502e004ce74f5de860a4d560182
thread:issue:73825,issue #73825: TypeError: cannot pickle 'torch._C._distributed_c10d._ProcessGroupGloo' object,stale/low urgency,2,stale/low urgency,1,0,73825,,,,0,cb345cd5e9f57ae111ceceef15b63151ad373f1e9c172c7d3e3a2fbbf96d4034
thread:issue:73826,issue #73826: Better Error report in torch.distribution.*.sample (when passing a non-iterable),stale/low urgency,2,stale/low urgency,1,0,73826,,,,0,fe89ccc48f2648e17e59e9b31acbf04c483e2b4a32b9d4d68dc8db17f6e5a860
thread:issue:73829,"issue #73829: Run libtorch examples, export error ""undefined reference to xxx""",stale/low urgency,2,stale/low urgency,1,0,73829,,,,0,54f86dd531c297ebfca4f7e919835ce4a91e67283c7fcfb1250d7a3d36cf18db
thread:issue:73832,issue #73832: Legacy sparse tensor constructor (e.g. torch.cuda.sparse.FloatTensor) silently ignores device kwarg,stale/low urgency,2,stale/low urgency,1,0,73832,,,,0,a05d914a5caf42d54d05fd3c500102eaad014fd7f05148a0714247a91fe80303
thread:issue:73840,"issue #73840: max_pool1d(): argument 'dilation' must be tuple of ints, but found element of type int at pos 1",stale/low urgency,2,stale/low urgency,1,0,73840,,,,0,b2fea2513ff1a11d5a918c2f48fc4d7b65b8b2297197bcdbf85c12a82cc00af3
thread:issue:73841,issue #73841: max_pool1d(): `RuntimeError: [enforce fail at CPUAllocator.cpp:68] . DefaultCPUAllocator: can't allocate memory`,stale/low urgency,2,stale/low urgency,1,0,73841,,,,0,fe2fc3940f8cf3f731dfcb3193669a62ee460e7d5cc7d011093f6b982a7f8227
thread:issue:73848,issue #73848: torch.distributions.multinomial.Multinomial (an example mistake of docs)?,stale/low urgency,2,stale/low urgency,1,0,73848,,,,0,63843b06d890c31d966ca8bda8f47d5692a0b9085f8602f557324c30d95ccc43
thread:issue:73860,issue #73860: tensor subclass and `__torch_function__` performance issues,stale/low urgency,2,stale/low urgency,1,0,73860,,,,0,a5c4f315ef6094d4415b79a16f59ca145fe1b12ee0cd671afaa9bd1093058a74
thread:issue:73883,issue #73883: Possible issue with memory allocation.,stale/low urgency,2,stale/low urgency,1,0,73883,,,,0,922e456546328ebd24200ef46300c7316b8b7edc357210bbf799b02734a9f2fd
thread:issue:73909,issue #73909: SummaryWriter reports encoding error,stale/low urgency,2,stale/low urgency,1,0,73909,,,,0,06f72d1939583cb7f9c8542ccf30973f6a3c8538ddfda5eb1c702cdfbe043cf7
thread:issue:73916,issue #73916: Invalid code inrandom_ kernel,stale/low urgency,2,stale/low urgency,1,0,73916,,,,0,9420d49ebcf810de02f1a4c2f6c170e1b21c9852fa99c465f82047f10cf331cf
thread:issue:73920,"issue #73920: pytorch/pytorch:1.8.1-cuda10.2-cudnn7-devel docker container contains cudnn 8.2, not 7.x as the name implies",stale/low urgency,2,stale/low urgency,1,0,73920,,,,0,88da5a3714031ba948a7ac25d172c543e78e68e3ebab018e5bd3018c4454b7f5
thread:issue:73924,issue #73924: torch.cuda.amp: Remove SPMD DDP doc portion,stale/low urgency,2,stale/low urgency,1,0,73924,,,,0,49faa0ab4edac0342aa788504e35234855321cb2ccc9b4c49fa05e2f71eae9fd
thread:issue:73930,"issue #73930: User raised TypeError from __torch_dispatch__ gets turned into generic ""unsupported operand type"" message",stale/low urgency,2,stale/low urgency,1,0,73930,,,,0,e0900aadeeae7a277b0a645fd20c75d1215080ae9eeae3f8903f1269fd0e9e27
thread:issue:73941,"issue #73941: [Quant] Framework observes weight in convert, changing numerics",stale/low urgency,2,stale/low urgency,1,0,73941,,,,0,e9e9bd0b3c5dc7f932fee57e70547d9ed0bd7bf31e1de8f75d997485b9828c21
thread:issue:73985,issue #73985: Release note bot is not sending messages on PRs anymore,stale/low urgency,2,stale/low urgency,1,0,73985,,,,0,8ff903fa290675c543b8ce0462cbbd468b75b1b6d11d241df27cdf115cc3cbc2
thread:issue:73989,issue #73989: Invalid example tensor: detectron2_maskrcnn,stale/low urgency,2,stale/low urgency,1,0,73989,,,,0,2601d11d2f915537aa5543adecaf6745275a353c3b9f7d396de556ccbceae989
thread:issue:73992,issue #73992: [FX] Tensor constants are not lifted to attributes in direct Graph construction,stale/low urgency,2,stale/low urgency,1,0,73992,,,,0,133ba817468b6ac027a799fa0da78021832c923ee73838b6c619844e8858544d
thread:issue:73994,issue #73994: Lazy: add support for index ops,stale/low urgency,2,stale/low urgency,1,0,73994,,,,0,ed3bd1985cd339ff80506d3eca9dfa535b90a5ae403591684b767d1cd3755dcd
thread:issue:74014,issue #74014: torch.utils.data.Dataset combined with pycuda issue,stale/low urgency,2,stale/low urgency,1,0,74014,,,,0,f2902cb0f29d0f08cd95417a8c4adbe41422599f58e2e019baeef65ee822c20f
thread:issue:74022,issue #74022: Inconsistent implementation on SWA,stale/low urgency,2,stale/low urgency,1,0,74022,,,,0,89d127f9572db4c1b40650d1b1cf3439879b6d9741443fab2a9a62280f9fdff3
thread:issue:74034,issue #74034: [torch.onnx] ONNX export failed on adaptive_avg_pool2d because input size not accessible not supported,stale/low urgency,2,stale/low urgency,1,0,74034,,,,0,8bd1887d26bfa438c55b3aad8686b4a30bbe391dcfd18ee52f48951915a14162
thread:issue:74052,issue #74052: assertEqual gives confusing error when comparing tuple with Tensor with Tensor,stale/low urgency,2,stale/low urgency,1,0,74052,,,,0,40da4bd4b070de4a435633b755d808970aa4f9c1513c6e8596988d9e3bef7ac1
thread:issue:74058,issue #74058: torch.fx.symbolic_trace is non-deterministic,stale/low urgency,2,stale/low urgency,1,0,74058,,,,0,8ff595290151282b899032b83a8f552dda301e83518ac0a968b447d9c4405b6f
thread:issue:74101,issue #74101: C++ Context::setDeterministicAlgorithms default 2nd arg not defined in header,stale/low urgency,2,stale/low urgency,1,0,74101,,,,0,978a0c7448e3fd40ef04b520971e8072703cd24a6c2231194098865895c9d2c4
thread:issue:74112,issue #74112: torch.jit.script does not custom state_dicts,stale/low urgency,2,stale/low urgency,1,0,74112,,,,0,9ba1530bcb2f47290690a26391899119cd5935f2c180aabbf97ac4b3c13a1a62
thread:issue:74134,issue #74134: No module named 'pygame': soft_actor_critic,stale/low urgency,2,stale/low urgency,1,0,74134,,,,0,0acb2d8a78dafc624d58d9804c36d445313ca705981dc5c5f784713fa42b3901
thread:issue:74138,issue #74138: `torch.set_printoptions` overwrites settings of its own previous calls,stale/low urgency,2,stale/low urgency,1,0,74138,,,,0,a34d5e2684432d83f02fc55973fa58018977ad6477783e329b7fd6e11c926eed
thread:issue:74143,issue #74143: Tracing model parameter shapes without instantiating the model parameters,stale/low urgency,2,stale/low urgency,1,0,74143,,,,0,fc04e859856384028e4db0e3822b56258dcdbceee3a0a2c958c184a437a83763
thread:issue:74146,issue #74146: RuntimeError: Connection reset by peer when backened by NCCL,stale/low urgency,2,stale/low urgency,1,0,74146,,,,0,6aaad28bdab573dec1201d1e85cdca7c4c60e7eb3c6f7676343e4291084578e7
thread:issue:74152,issue #74152: All-caps names for pages are confusing,stale/low urgency,2,stale/low urgency,1,0,74152,,,,0,a19163af76ad1ac2041b1eb1d70d8494e6ac9d3476fc06cdf79ca7552f6c1d7d
thread:issue:74155,issue #74155: [Android Studio] DefaultCPUAllocator: not enough memory: you tried to allocate 280166432 bytes,stale/low urgency,2,stale/low urgency,1,0,74155,,,,0,76f8cfd0a0ca7b5a32a7bdca05e8d85bee6c21442fa22d5b0f0bdc3c5f36fe05
thread:issue:74167,issue #74167: Slower performance of `torch.mm` method with sparse CSR tensor,stale/low urgency,2,stale/low urgency,1,0,74167,,,,0,4cfe3c802b496d82051fbe0184da4edda1a10bd5c24d950ccf4a024267ba11dc
thread:issue:74185,issue #74185: [DDP] Parallelize initialization collectives,stale/low urgency,2,stale/low urgency,1,0,74185,,,,0,fdfda57c17a83db0278e713a7c6e081a608d6c5a083df078b5e79b11da6c32a0
thread:issue:74187,issue #74187: [docker] test_corrcoef_cpu_complex64 fails on CPU build,stale/low urgency,2,stale/low urgency,1,0,74187,,,,0,ae0b5de596b2e7c28d1b0bc78d7fe8e675afe09c08dd2c61e1da0f4c3755b400
thread:issue:74218,issue #74218: Cannot get op through FX when using nn.Sequential,stale/low urgency,2,stale/low urgency,1,0,74218,,,,0,976466fcd472a9bdba66995a910d1ae2649fd37378c2c8bf91b4045dc4e31865
thread:issue:74222,issue #74222: Autograd API to get saved for backwards tensors for an autograd graph,stale/low urgency,2,stale/low urgency,1,0,74222,,,,0,af5d0abb378ed7be252a6f627c9a1f818907a27e433ac25a8fbc7d496062cfc2
thread:issue:7423,issue #7423: Multi queue for dataloader when workers > 1,stale/low urgency,2,stale/low urgency,1,0,7423,,,,0,c46ceb12a06965afcd3199c73e7528458059ed278cbe91d297aa31f14a27d26b
thread:issue:74235,issue #74235: RuntimeError: CUDA error: unspecified launch failure,stale/low urgency,2,stale/low urgency,1,0,74235,,,,0,d2da099b8315628440a2e7cae613fcb6c848c675d2400ab597f3f1e808a219d7
thread:issue:74248,issue #74248: dir() on torch.ops.aten doesn't work,stale/low urgency,2,stale/low urgency,1,0,74248,,,,0,aee41faff0660bbce9f13664e2d59d49b31fbc2bedb66843f59775982e25a451
thread:issue:74256,issue #74256: Create secure credential storage for metrics credentials and associated documentation on how to regenerate them if needed,stale/low urgency,2,stale/low urgency,1,0,74256,,,,0,ef9220b47f5d5637e4fa7d7ec6d42c7c9ff3a6734c5635472f918de0a0ad0b6c
thread:issue:74281,issue #74281: Move torchbench workflow to `workflow_dispatch`?,stale/low urgency,2,stale/low urgency,1,0,74281,,,,0,f8225874204316bd573d0795a36c0f8fa7e32fa416f943b05ee449698f2a9b90
thread:issue:74295,"issue #74295: TypeError: new_empty(): argument 'size' (position 1) must be tuple of ints, not list",stale/low urgency,2,stale/low urgency,1,0,74295,,,,0,fa053b28b2ff2cc82b5c2c8b8c4bcc7c34db3a88f7bc42bed43fac757b0f6f37
thread:issue:74296,issue #74296: Time Gated Lstm,stale/low urgency,2,stale/low urgency,1,0,74296,,,,0,283ef3f0382df34ff3e5d612772c2d5967fbcd8401291f80496bb68a9c1171e7
thread:issue:74299,issue #74299: Gradients tests are very time consuming,stale/low urgency,2,stale/low urgency,1,0,74299,,,,0,2a42cebc1cb634cbf20cab2dcb305c698b4841392b25415fe8518c895221aa12
thread:issue:74300,issue #74300: repeat_interleave is not meta friendly; not transform friendly either,stale/low urgency,2,stale/low urgency,1,0,74300,,,,0,e75d1804ed320b039e3a7f2836a0280a90ddc4e9b2e4ca4c7d0ddc7c784034ba
thread:issue:74301,issue #74301: Error in lobpcg when using largest=False,stale/low urgency,2,stale/low urgency,1,0,74301,,,,0,6c4c4e2000974f1a02f47c6efb165602fca023763958d8fc7959d6dd27effd74
thread:issue:74320,issue #74320: Make OpInfo repr more useful,stale/low urgency,2,stale/low urgency,1,0,74320,,,,0,e615bd1c8158f0d1167610b5f4fc35d1b34118cef2a2f7160b062a1c8a2ec4de
thread:issue:74328,issue #74328: instantiate_device_type_tests is misnamed,stale/low urgency,2,stale/low urgency,1,0,74328,,,,0,65e959683851ef7f4c45d2d9c92c591e3bcc04455109358a759bb4cc0add1509
thread:issue:74329,issue #74329: CompositeImplicitAutograd addr implementation does not have correct behavior for bool (meta gives wrong dtype),stale/low urgency,2,stale/low urgency,1,0,74329,,,,0,bb0c76e8b422576744833ed431ef48fedfb1c46a2538e2d0cddc4e4d6d04712c
thread:issue:74337,issue #74337: None returned from data loader causes debugging difficulty from collate function,stale/low urgency,2,stale/low urgency,1,0,74337,,,,0,b30b9a11a3c76f09b6b97d9e26de6c53b89c19b21a8b40fb5db57c3c294ce13d
thread:issue:74368,"issue #74368: Can't download torch vision models for older versions via torch.hub.load: ""multiple choices""",stale/low urgency,2,stale/low urgency,1,0,74368,,,,0,b6ab9b96f714c9060a4f467d35ee271154672c50fc3772957115f931dfa0cae8
thread:issue:74384,issue #74384: Package and Deploy: Generate External Python Registration files at runtime [WIP],stale/low urgency,2,stale/low urgency,1,0,74384,,,,0,0b48a3dfd89acb6e1ae65c79a27fe65b90f8133d45a597020f2e4345767541dc
thread:issue:74389,issue #74389: torch.package unpickling transforms: ModuleNotFoundError: No module named 'torch._C._linalg'; 'torch._C' is not a package,stale/low urgency,2,stale/low urgency,1,0,74389,,,,0,22c6442e1980ed75c82f3eca0d1a4ed8acc1bac8e2cbc0fc5837e79a8478a820
thread:issue:74391,issue #74391: Can this cudaDeviceSynchronize call be removed?,stale/low urgency,2,stale/low urgency,1,0,74391,,,,0,b0a43376de3e4c51d93b4f3b470764d2697aae93d8cf34dd40a7adc4bdc3dc16
thread:issue:74393,issue #74393: Quantized cannot inference with cuda,stale/low urgency,2,stale/low urgency,1,0,74393,,,,0,e296f144cbc8bb9d44b41639e8d7a1edf8d31645b0ffd2cc927cab1c10c5a2c8
thread:issue:74398,issue #74398: _run_ninja_build failing with clang,stale/low urgency,2,stale/low urgency,1,0,74398,,,,0,a544764a3e71b3e717259c09602c867ece38a5b0e9bde2b89db9072e64da1800
thread:issue:74411,"issue #74411: UserWarning: size_average and reduce args will be deprecated, please use reduction='mean' instead.",stale/low urgency,2,stale/low urgency,1,0,74411,,,,0,fd3d44cb254a27cd046e99455db02d7071b4bf1ef1c31b63581c0e1928a66e87
thread:issue:74421,"issue #74421: can not build pytorch, failing due to missing _ctypes module",stale/low urgency,2,stale/low urgency,1,0,74421,,,,0,3d006c305bacd171946466d6c86a64da1440be6e6843dd1fe616776fd3ff3251
thread:issue:74442,issue #74442: kron has unnecessary and undocumented dependence on memory layout,stale/low urgency,2,stale/low urgency,1,0,74442,,,,0,97c728bd1db88cd259c306083dfe6a6926bb24c41f6def7281171bbce87b1071
thread:issue:74454,issue #74454: torch has no attribute sparse_csr_tensor,stale/low urgency,2,stale/low urgency,1,0,74454,,,,0,9385ec535350172fa6aa151e4ca0012f8b655cbd82efa2a07599f6c47a1117eb
thread:issue:74485,issue #74485: dir(torch._VF) doesn't work,stale/low urgency,2,stale/low urgency,1,0,74485,,,,0,4f5ea44e41f560750c1bffa1b68974af182bf5eba1bf63e755cbc5a2bb1e7220
thread:issue:74491,issue #74491: __torch_function__ is hitting private functions in some cases,stale/low urgency,2,stale/low urgency,1,0,74491,,,,0,90149c97ef49efeb6a978993a328736f6f93526aac612fb62a58f71583b553f7
thread:issue:74500,issue #74500: One should be able to query if a DecorateInfo is a xfail or skip,stale/low urgency,2,stale/low urgency,1,0,74500,,,,0,01ddc61efab74b6f3bd9703dac54f9b92001066a8d31a6b895c864878220550d
thread:issue:74503,issue #74503: Compilation error on M1 Mac,stale/low urgency,2,stale/low urgency,1,0,74503,,,,0,05a319cbc8fd53821471d6d434742baaf3700e7a7da5aae72531d6a271c40571
thread:issue:74514,issue #74514: Figure out a lint for directly returning a nullptr instead of raising python_error,stale/low urgency,2,stale/low urgency,1,0,74514,,,,0,665d25cc76b620d779219c0f5893045b915d033ed064d01926ba9a8f343ca2d4
thread:issue:74531,issue #74531: Feature Request: torch.testing.make_scalar (make_tensor for scalars),stale/low urgency,2,stale/low urgency,1,0,74531,,,,0,b5485f9f95a23c468d558ba99a3f1178ac076c7a1384d9a8b669957c0f01d4b9
thread:issue:74539,issue #74539: Making PyTorch more pythonic: the PyTorch functions should be inspectable.,stale/low urgency,2,stale/low urgency,1,0,74539,,,,0,3b65bc82d7ee386ec955363f86edb8ad5e9d91d5df465bc370ca8443f42cd04f
thread:issue:74540,issue #74540: No factory functions for strided quantized tensors,stale/low urgency,2,stale/low urgency,1,0,74540,,,,0,b58acda33e9752ca32c48b27e802e0cc1b3520efebfd3ab828746972bdc5802c
thread:issue:74544,issue #74544: Many inplace operators are not being tested for variant consistency (test_variant_consistency_eager),stale/low urgency,2,stale/low urgency,1,0,74544,,,,0,a912445f611b1d03c5e37cdb82fa637b03302953c570715cbab2651bf4b1c4f7
thread:issue:74554,issue #74554: Specifying left-right-padding as tuple for asymmetric padding,stale/low urgency,2,stale/low urgency,1,0,74554,,,,0,5048ff77e1a1d5a9be850dc38fed8c501a9b624c43965ab49cb38d93598026d9
thread:issue:7457,issue #7457: [feature request] torch.nn.DataParallel should work nicely both for cpu and gpu devices,stale/low urgency,2,stale/low urgency,1,0,7457,,,,0,12fba86c49e855b3a8e7af7fcf7fc1fc51f31db0e84e36f416cacf3eb826ef48
thread:issue:74604,"issue #74604: Training, Forward / backward pass with _different_ batch-size, no speedup observed when backward pass has smaller batch-size",stale/low urgency,2,stale/low urgency,1,0,74604,,,,0,28ae4a608a342341d4ef6199f628d24592377d0cac37b52c49ee251dfe9fff00
thread:issue:74613,issue #74613: OpInfo request for `nn.functional` and `unbind`,stale/low urgency,2,stale/low urgency,1,0,74613,,,,0,9ef5b349e0104032cf876cdb4a7d12e7a2454a605b21ee3ec5eccaa840ab03e0
thread:issue:74614,issue #74614: compilation error with PyTorch v1.11 for CPU on ppc64le,stale/low urgency,2,stale/low urgency,1,0,74614,,,,0,07f7ee92a3be239593208b6bea6b64868f2c45d4cf0513edba55c2a9b31b579f
thread:issue:74627,issue #74627: Feature request: INT4 format support,stale/low urgency,2,stale/low urgency,1,0,74627,,,,0,d8796357aae77071db2730d7ae020cc68393e26e4e6d6fd2bb4881a0d0071dfb
thread:issue:74634,issue #74634: Sampling Issue With Distributions,stale/low urgency,2,stale/low urgency,1,0,74634,,,,0,37041efc989ad254949086254907f003cb2c9dea0ecf42b4dddda9cc7bb8b4d2
thread:issue:74642,issue #74642: Some easy way to add xfails to OpInfos,stale/low urgency,2,stale/low urgency,1,0,74642,,,,0,771c366c0a1e9eaffc982322e9e39e143bd65a47551c79c8c2fe65d1a5563bda
thread:issue:74669,issue #74669: String dtypes for torch Tensors,stale/low urgency,2,stale/low urgency,1,0,74669,,,,0,67fdc8a11c5c50039018657905ff535d9b4d5e6308379d3f1577aba2aae71618
thread:issue:74677,issue #74677: `check_batched_forward_grad` fails for `torch.norm` and related ops,stale/low urgency,2,stale/low urgency,1,0,74677,,,,0,b8703daf6e131a5b0a6b901158906fa8472446cf29c6b50b83701ca0ecf1b301
thread:issue:74679,issue #74679: Embedding isn't determinstic on linux-xenial-cuda11.3-py3.7-gcc7,stale/low urgency,2,stale/low urgency,1,0,74679,,,,0,32bc6c1b21e4cb61e623dfc8ef9ea866242ceb1338daae788acfceb3335a1388
thread:issue:7471,issue #7471: Make c10d/FileStore cache file descriptor,stale/low urgency,2,stale/low urgency,1,0,7471,,,,0,b5c883c49741564358476aa2a8797cb990b7ec83b0e3bf7267649cdd1bc6d789
thread:issue:74739,issue #74739: GradScaler support FP16 gradients?,stale/low urgency,2,stale/low urgency,1,0,74739,,,,0,eb12f5672fa076c1718755739e1f9f6eee8fd1bf7ef738db6d70f9b910181b05
thread:issue:74748,issue #74748: [jiterator] Jiterate Complex Ops,stale/low urgency,2,stale/low urgency,1,0,74748,,,,0,25732e32d66545b6c9abd872cf9eb4e1e4b07c1fd5e14f758930e5697d2abb88
thread:issue:74788,issue #74788: [Structured] Make it possible to override create_out implementation on a per dispatch key basis (structured sparse),stale/low urgency,2,stale/low urgency,1,0,74788,,,,0,298cdcc69561dfe384d24d1832548a0d0648d443b7c7ebf95e22458a931954fc
thread:issue:7480,issue #7480: [caffe2] How to use multiple CPUs?,stale/low urgency,2,stale/low urgency,1,0,7480,,,,0,0c00a5df972b43da06dbeebc67cf76bf476edcf27a64de8bc819fab1596b34fa
thread:issue:74842,issue #74842: error in quantization by quantize_fx.prepare_fx,stale/low urgency,2,stale/low urgency,1,0,74842,,,,0,f51b258588f00da5be921579d2f08823b147c84fa333d7ffe54a84c6089c1645
thread:issue:74855,issue #74855: Enable dtype keyword argument for to_dense method,stale/low urgency,2,stale/low urgency,1,0,74855,,,,0,decc8bc48d0c8876d33ed9b6229aecda7a5148e488623c02f0f2bf2e8c98cb2b
thread:issue:74876,issue #74876: Failure to set number of threads on AWS Lambda,stale/low urgency,2,stale/low urgency,1,0,74876,,,,0,51592032da7a867d9629a6fec87d8e4ac373829a379913af4095818d751132fc
thread:issue:74884,issue #74884: Problematic ASGD Optimizer,stale/low urgency,2,stale/low urgency,1,0,74884,,,,0,a99e84809b0e957594c5051d4f4a059d267ec642b9fcbb6b671649bb5b3204dc
thread:issue:74896,issue #74896: Install PyTorch from source on power machine,stale/low urgency,2,stale/low urgency,1,0,74896,,,,0,fbffa3d8501b7360a98f4d3a398083ea3cb2c9ba0f847344eb3f10951532d577
thread:issue:74901,"issue #74901: torch.fx.wrap will not work, when encapsulate the code",stale/low urgency,2,stale/low urgency,1,0,74901,,,,0,d79f30fe5f8d888d17366675a3490f22892731673e52a207b22de822a4a52b9a
thread:issue:74907,issue #74907: [Proposal] Use batched oprations to accelerate PowerSGD,stale/low urgency,2,stale/low urgency,1,0,74907,,,,0,de5b69f77d2ce912267096ea84f6a56afbeceac0ed906be4590ddebfe8000236
thread:issue:74911,"issue #74911: Include finfo(dtype).[min, max, eps, tiny] in the extremal test case",stale/low urgency,2,stale/low urgency,1,0,74911,,,,0,392a05f23a940b481d0a81f62f9e8f75fa8d06657b6fb4609a7b2f3fb7b2deec
thread:issue:7493,issue #7493: Build from source with anaconda [caffe2],stale/low urgency,2,stale/low urgency,1,0,7493,,,,0,4409e6743dee17152afc5bb58ac6e8eaaf6c11c2c11b158f83157d1739ee30d0
thread:issue:74935,issue #74935: [RFC]A suggestion of channels last memory format implementation for 3D tensor,stale/low urgency,2,stale/low urgency,1,0,74935,,,,0,2e7fa28539f942b1c942ae7696683f1d95fa3806bd1381d9e653003ed31533c3
thread:issue:74937,issue #74937: [docs] RandomSampler has unrendered back-ticks,stale/low urgency,2,stale/low urgency,1,0,74937,,,,0,80171f0ba40fc69f2de3ef3e31d14fd1434b7fe1495a6bfbdcd636aae40fd79e
thread:issue:74940,issue #74940: Feature Request: Add J0 J1 J2 H0 H1 H2 Bessel functions,stale/low urgency,2,stale/low urgency,1,0,74940,,,,0,71977b7b520efa396c5df153d19bb12887c339e534746c4fd5ce1840115e66b3
thread:issue:74942,issue #74942: Store SourceDataset in MapDataset using pointer,stale/low urgency,2,stale/low urgency,1,0,74942,,,,0,7f0aa67a3156f828b42da82e6fc065e402d24118a112eed4b26f195667b1bb84
thread:issue:74951,issue #74951: [feature request] `coalesce` to support `dim` argument.,stale/low urgency,2,stale/low urgency,1,0,74951,,,,0,6df1b68976c743b20efb52b7b3c1ce2634b0e312630614b54fe7eddcab75e3fe
thread:issue:74952,issue #74952: [feature request] Make `index_select` parallel.,stale/low urgency,2,stale/low urgency,1,0,74952,,,,0,c98d4ed9ddaed16086094f9d45fa2641644b1fe81d011f300955de42907df822
thread:issue:74953,issue #74953: LSTM quantization fails if proj_size > 0,stale/low urgency,2,stale/low urgency,1,0,74953,,,,0,ee328a987527359cd96b832a58a69722f7c8b5490d605e9792ffad674dfa9af9
thread:issue:74991,issue #74991: Tensor Subclass not preserved for Tensor subclasses created via inheritance of a TensorImpl subclass,stale/low urgency,2,stale/low urgency,1,0,74991,,,,0,5ca001e6042c4d33b14ee12e14dc7d2b2e0cb6b78ed7417e58049936a5fb18f2
thread:issue:75025,issue #75025: torch.cuda.init() unstacks existing CUDA contexts,stale/low urgency,2,stale/low urgency,1,0,75025,,,,0,b986e4380b953c5106c1471a6955d9f96ad69abbf0c4d113b9707d070cc91687
thread:issue:75090,issue #75090: torch.fx deepcopy bug,stale/low urgency,2,stale/low urgency,1,0,75090,,,,0,709d10e2948f4b801896ffd7d1c4ef35677068d03fa02989f39e117aad81e9c8
thread:issue:75129,issue #75129: [BE] add documentation for adjust learning rate when going to distributed training,stale/low urgency,2,stale/low urgency,1,0,75129,,,,0,476458a49c75ed95b12b0233044503b1b04a016f82b7c91e60cf829eda5c83f5
thread:issue:75147,issue #75147: Dataloader hangs. Potential deadlock with `set_num_threads` in worker processes?,stale/low urgency,2,stale/low urgency,1,0,75147,,,,0,908b1dad8b9373682e6137bf879f00446292bb160952bed66f296170379988ea
thread:issue:75158,issue #75158: Offical libtorch (pytorch c++ frontend) docker image,stale/low urgency,2,stale/low urgency,1,0,75158,,,,0,140da02179e98cc206028365786b23abba7bb18eff66a9b351d9f7ab8d185afd
thread:issue:75166,issue #75166: Disable Python mode (torch dispatch mode) inside of mode-induced __torch_dispatch__call,stale/low urgency,2,stale/low urgency,1,0,75166,,,,0,65230a977402a8459bb4b6ccbcf1d74f5eeab8a3ac95ac6506b3f3d39e3f1f0f
thread:issue:75172,issue #75172: Runtime configuration to disable TORCH_WARN temporally?,stale/low urgency,2,stale/low urgency,1,0,75172,,,,0,56340e8f67f3a5ebda49e799be1d63f61163c8bef6493901746fed3981fc17f0
thread:issue:75173,issue #75173: LowRankMultivariateNormal doesn't work with 0 diagonal,stale/low urgency,2,stale/low urgency,1,0,75173,,,,0,02a9f23857525213f28c972570e3713d0dc748934ddda45a0b071f67db7bf9ad
thread:issue:75180,issue #75180: Add BUILD_LAZY_CUDA_SPARSE option,stale/low urgency,2,stale/low urgency,1,0,75180,,,,0,172ea3d6b9887d7346cabf04d77fd3995f864408df52ae4f6d1cce41849d76b9
thread:issue:75184,"issue #75184: PyTorch source code compile fail after ""Built target fbgemm_avx2""",stale/low urgency,2,stale/low urgency,1,0,75184,,,,0,c99be3348256a00eadb641befa1584572bb6f4b0236e2764849cb7f10b2e4e9e
thread:issue:75198,issue #75198: `torch.cuda.get_device_name` fails to identify RTX 3090 Ti,stale/low urgency,2,stale/low urgency,1,0,75198,,,,0,1de41f1d80d931a95829b280b1ef7c9340a732d33415415ff6d67cdd725f16bc
thread:issue:75213,issue #75213: Improve wording in _store_based_barrier logging and identify ranks that have not joined,stale/low urgency,2,stale/low urgency,1,0,75213,,,,0,53108021ca9aafd398d8e0d3141dde3e1436ae5935a45f5ce187294f7a202d33
thread:issue:75251,issue #75251: Potential runtime optimization of Mish activation,stale/low urgency,2,stale/low urgency,1,0,75251,,,,0,a686065e7f15bf1f3cca1b571f020ef28922375115417641fc8037336f377d31
thread:issue:75342,issue #75342: `torch.fx.operator_schemas.normalize_function` is too permissive,stale/low urgency,2,stale/low urgency,1,0,75342,,,,0,6c488285ae1fe65a55eec4a150dc92548b5d3515b748a2f6852e82c71fd025c9
thread:issue:75371,"issue #75371: NVFuser bad ""reshape"" performance",stale/low urgency,2,stale/low urgency,1,0,75371,,,,0,c51c0175a8a97d0cca2e87430b7433c4a7ce3c8716256d317f92e2c3e48399f7
thread:issue:75391,issue #75391: Pytorch linalg test failure with cuda 11.6,stale/low urgency,2,stale/low urgency,1,0,75391,,,,0,b8e0d7e6cb67182a8512ad5ab919885a3677bfc0d29454fc563d207fe5d1d930
thread:issue:75427,issue #75427: `torch.cuda.is_bf16_supported()` seem to not work properly,stale/low urgency,2,stale/low urgency,1,0,75427,,,,0,9fac0ef927cfda4e99d3fa4a155d535669c8649d2ea3e5070a4d9b5ccb3b23b6
thread:issue:75432,issue #75432: Rollup: Top forward-over-reverse formulas,stale/low urgency,2,stale/low urgency,1,0,75432,,,,0,dc97c78940b6e4f4fb5133a0b772fb08a99aa9aef5714b22b02e1cf1b26a1ff1
thread:issue:75440,issue #75440: [quant] PackedLinearWeight::apply_dynamic_impl does not handle ReLUFused template argument,stale/low urgency,2,stale/low urgency,1,0,75440,,,,0,ed2a2f4c9ab7f1bf100dcc2d18406eac2ee9fc883f4bc0ecc6da26d198e84c1c
thread:issue:7546,issue #7546: [Caffe2] Build broken on macOS High Sierra: can't find sys headers in /usr/local/include.,stale/low urgency,2,stale/low urgency,1,0,7546,,,,0,dcd54611fdf178e31418cbad157e55c7a3a842b8124bcfe0d387e14f97a76543
thread:issue:75462,issue #75462: NotImplemented confusion between __torch_function__ and __rpow__ (and other dunder magic methods),stale/low urgency,2,stale/low urgency,1,0,75462,,,,0,31d9fb8f036ebabf5dfadb811eeda1e417a59bef6516f94c4e669510c5ca85a2
thread:issue:75474,issue #75474: [NVFuser] call kernels with informative names (e.g. pow_mul_add),stale/low urgency,2,stale/low urgency,1,0,75474,,,,0,3c0fbff9142187ab06a0c448032e2bb156ee45d6a9a8628b76370d9b50740161
thread:issue:75480,issue #75480: IndexError: Caught IndexError in replica 0 on device 0.,stale/low urgency,2,stale/low urgency,1,0,75480,,,,0,98dd335fd7c279ae656b609ed1521a0772b3c03241ccce4dea6529b048feb1ca
thread:issue:75494,issue #75494: `torch.pinverse` on CUDA tensors produces non-optimal output for certain type of (invertible) matrix on torch > 1.7.1 but not on torch <= 1.7.1,stale/low urgency,2,stale/low urgency,1,0,75494,,,,0,b03df2d1f57e86e96f73637bd34fc983faa36b13585355e31a6923edd5f6caa9
thread:issue:75495,issue #75495: Forward AD convolution fails for the empty backend,stale/low urgency,2,stale/low urgency,1,0,75495,,,,0,cb54aad8bf88294d6ebd72b816f098048b73890bd3a21151f61f727799b24b16
thread:issue:75500,issue #75500: RuntimeError: cuDNN error: CUDNN_STATUS_EXECUTION_FAILED,stale/low urgency,2,stale/low urgency,1,0,75500,,,,0,4d53a71e86c62beb4a4e22909334546ccd19805e4ea31ff7f6dc7ae262570ec6
thread:issue:75503,issue #75503: make_dual errors out when primal is a Tensor and tangent is a subclass Tensor,stale/low urgency,2,stale/low urgency,1,0,75503,,,,0,284c4621b1e5a59daf8e1c8becda8e0836c9dc39b792e6daa27c2b0c20c41e3f
thread:issue:75510,"issue #75510: [doc] view appears to mean different things, `view/reshape` vs `transpose/permute`.",stale/low urgency,2,stale/low urgency,1,0,75510,,,,0,08e9be2494d53b21504e62b67ddd7eec5d61ab2a722eca746bc5ad75f384c5e8
thread:issue:75512,issue #75512: Connection closed by peer when using dist.isend in gloo backend,stale/low urgency,2,stale/low urgency,1,0,75512,,,,0,9ec1be9f0e546fe7ca32ab2e4875f925a3b42d5f01b5d1eca5fd6c66427885d4
thread:issue:75549,issue #75549: `torch.linalg.lstsq` raises `CUBLAS_STATUS_EXECUTION_FAILED` for large `B` in CUDA tensors,stale/low urgency,2,stale/low urgency,1,0,75549,,,,0,91aa0b176edf66e9a86084bb668e2158323908fadab393b61e33fb0e2e05d96b
thread:issue:75554,issue #75554: torch.overrides testing is not catching people adding new kwargs and not passing on to handle_torch_function,stale/low urgency,2,stale/low urgency,1,0,75554,,,,0,ac820110dd89b26a269d6eafb0d70117f6575d670a6e27075cfabac45c179eb6
thread:issue:75562,"issue #75562: clang-tidy ""error: do not use const_cast"" cppcoreguidelines-pro-type-const-cast is counterproductive",stale/low urgency,2,stale/low urgency,1,0,75562,,,,0,2c340413fb6df8cf6b7bcce5006502e497fe0784510a2b8de4a66642f4ccf1a1
thread:issue:75565,issue #75565: make lint should advertise make setup_lint,stale/low urgency,2,stale/low urgency,1,0,75565,,,,0,a621fbc23f02961b6d9735bc52e0defb0150167e895974031a80d8767b8d938d
thread:issue:75568,issue #75568: mypy typing strategy for Tensor-likes (`__torch_function__`),stale/low urgency,2,stale/low urgency,1,0,75568,,,,0,b8f23f2a0e5891da7613dc1b1f01d2924ceed53d3de8556d91f0252d28edf0b6
thread:issue:75577,issue #75577: Cannot use socks5h proxy because of urllib: `urlopen error Remote end closed connection without response`,stale/low urgency,2,stale/low urgency,1,0,75577,,,,0,6e107b7d862c49bbeedce9dcb43bb8c35761814eddcd024616809f1eeafe9023
thread:issue:75582,issue #75582: Add `balance` flag to `random_split`,stale/low urgency,2,stale/low urgency,1,0,75582,,,,0,b3bbf7803c4bc897c98369bc12c72d0b739775b97c7793e63a9a920bcf2d387c
thread:issue:75625,issue #75625: [ONNX] Support unit tests in scripting that we already support in tracing,stale/low urgency,2,stale/low urgency,1,0,75625,,,,0,d579ce04beeeead7c30508da93c3785c74c380993ad2a183af2a47556b0dec92
thread:issue:75652,issue #75652: Op segfaults with ForwardAD and Subclassed Tensor as Tangent,stale/low urgency,2,stale/low urgency,1,0,75652,,,,0,f771f4f2068b7f2b00b955c782b84a263b023d1c650d744611910078d2c3e79e
thread:issue:75659,issue #75659: Misleading documentation for cholesky_inverse,stale/low urgency,2,stale/low urgency,1,0,75659,,,,0,aaff12fa6aa7335a1ff45064dde28a513371e93c2ab484251ca2f33195bce0ab
thread:issue:75662,"issue #75662: multiprocessing and torch.tensor, Cannot allocate memory error",stale/low urgency,2,stale/low urgency,1,0,75662,,,,0,047e27638427d8d2ffb4fde76b2643063fa0bb228cc84e866e1788d3c4f39765
thread:issue:75667,issue #75667: Implement histc for bfloat16 on CPU,stale/low urgency,2,stale/low urgency,1,0,75667,,,,0,f9e6815bc089f4668bc2bbe5581af4802160c40ede4726b12bfa13a60ab69e0a
thread:issue:75680,issue #75680: Addressing skips in OpInfo nn.functional.binary_cross_entropy_with_logits,stale/low urgency,2,stale/low urgency,1,0,75680,,,,0,d770b49b542b36b388bdd54ef37df17677a881961329614c5acc397636422168
thread:issue:75701,issue #75701: Einsum should have an `out=` parameter,stale/low urgency,2,stale/low urgency,1,0,75701,,,,0,269f8a810117e48513ad1b4da30070f39bfff87eec7fc56f24e1d0524c16a25a
thread:issue:75703,issue #75703: Let's host NVIDIA dependencies in our own S3,stale/low urgency,2,stale/low urgency,1,0,75703,,,,0,7215ac6f16e42142cf7f51d32b9eba7e34a420bb8552b0fe385b55280d04f179
thread:issue:7571,"issue #7571: the latest version of caffe2, Which file is the function loadToNCHW()",stale/low urgency,2,stale/low urgency,1,0,7571,,,,0,ff9dc60fa11476652f0a49e9bdd9081f5eba58598d96f9d624b1f415cb4ceb15
thread:issue:75721,issue #75721: Sparse tensors: indexing with `index` and/or being able run `sparse_tensor[index]`,stale/low urgency,2,stale/low urgency,1,0,75721,,,,0,e99f42bad46c956e516768cd73f8f1bfe8a48d38d2b4738091ff779e93e19a85
thread:issue:75733,"issue #75733: `torch.sum, prod, cumsum, cumprod, sparse.sum` INTERNAL ASSERT FAIL",stale/low urgency,2,stale/low urgency,1,0,75733,,,,0,68ce0c81002bce692cf024403e138610c6264bea2198942733abae16254f63ec
thread:issue:75737,issue #75737: torch.device missing doctring,stale/low urgency,2,stale/low urgency,1,0,75737,,,,0,fdedcbf8c458d8ace755470f0a25923acac54c0108da0e00634722952f9f55dc
thread:issue:75773,issue #75773: Handle noncontiguous inputs in distributed backend layer,stale/low urgency,2,stale/low urgency,1,0,75773,,,,0,10302ab8571ca8b8b65c315dcffa0347a4099afabc916e7c24a8b3611987074c
thread:issue:75785,"issue #75785: [ONNX] Expected quantizer->qscheme() == kPerTensorAffine to be true, but got false.",stale/low urgency,2,stale/low urgency,1,0,75785,,,,0,3c27acce0b7c2809a14ba5a2be3dd50e0d9aff32723a314af4839594bbc0cfb0
thread:issue:75798,issue #75798: Fix workaround `__module__` used to appease public binding checks,stale/low urgency,2,stale/low urgency,1,0,75798,,,,0,1b94eca40708ef3f4a0f1fbcfe2824913a13a610a5050bb975acb1cac38ddd05
thread:issue:75895,issue #75895: add -D_GLIBCXX_ASSERTIONS in debug mode,stale/low urgency,2,stale/low urgency,1,0,75895,,,,0,91065d87a73a04fe43532445d2d09b1946ab6ee5b69bf2e9de54aef84363947e
thread:issue:75904,issue #75904: Move _SKIP_PYTHON_BINDINGS to native_functions.yaml,stale/low urgency,2,stale/low urgency,1,0,75904,,,,0,c3307576a2d37383b0ad7d779c6a0e57191b3671d8fedbe19d95b35f3ae979ba
thread:issue:75909,issue #75909: Add batching rules for `{view}_copy` operators,stale/low urgency,2,stale/low urgency,1,0,75909,,,,0,bd4625fb9311853ae046f9807584b5f55429104653eed1e5beaec22c42b781a8
thread:issue:75911,issue #75911: 'python setup.py build' failed but succeed using 'pip install -v .' which calls 'python setup.py build'.,stale/low urgency,2,stale/low urgency,1,0,75911,,,,0,0d64bdca2a79f56ce27b7029842e972bbc20df6e7c30409dc3a43fde7fce2964
thread:issue:75923,issue #75923: aten::_softmax.out doesn't work with non-contiguous Tensors,stale/low urgency,2,stale/low urgency,1,0,75923,,,,0,0539c53e32550ce31e7afed8c84c9c9fb0dbe9f4275e7d1bad51905fac5c12aa
thread:issue:75925,issue #75925: autogen-58 microbenchmark fails on NNC gpu fusion,stale/low urgency,2,stale/low urgency,1,0,75925,,,,0,eb06c2f1ace613d820c1df9ea9fbb481dfb742da224671fbe26c5d78e1394aac
thread:issue:75926,issue #75926: Disable TracerWarnings on NVFuser opinfo tests,stale/low urgency,2,stale/low urgency,1,0,75926,,,,0,d70ffd8276495fc46fa1fd9db7acd6512a9e3d692b0c3bf845cc4a816c4e6371
thread:issue:75940,issue #75940: Add nesting of nested Tensor,stale/low urgency,2,stale/low urgency,1,0,75940,,,,0,34b2ac78da8af66137a833e6da70e9de9b1c60e8d6d75fb15befac0b90899f20
thread:issue:75986,issue #75986: gql_mocks.json has really long lines,stale/low urgency,2,stale/low urgency,1,0,75986,,,,0,28a99283b4bd8932d5054e4ec5eee9396d3d195d83a8333e63e89d84d810ba0e
thread:issue:76007,"issue #76007: Build a default NVFuser comparison callback, e.g. for use with torchbench",stale/low urgency,2,stale/low urgency,1,0,76007,,,,0,225d1921b6b6b5803e204ec86b61b2ca5e55dfd3212270eede8faa0b1bb6f24b
thread:issue:76011,issue #76011: backcompat tests in test_nn.py are slow,stale/low urgency,2,stale/low urgency,1,0,76011,,,,0,605391a6c07e5f8a65a3a5e2d62b68d6c950e2fad7f68c02fcf97ff48dc8e06b
thread:issue:76025,issue #76025: Numerical instability: matrix multiplication got different results on cpu and gpu,stale/low urgency,2,stale/low urgency,1,0,76025,,,,0,2763a7e0d505f1815827c6999c6c3026fdb3630350e094ec611d00f38c87bbd6
thread:issue:76028,issue #76028: A bug in instructions for building PyTorch with ASAN,stale/low urgency,2,stale/low urgency,1,0,76028,,,,0,7bf1d800c80da585c7e3cc741f99ecc797d985baa8ee453fd53905d629d84bfe
thread:issue:76030,issue #76030: Dirichlet with small concentration,stale/low urgency,2,stale/low urgency,1,0,76030,,,,0,1d4eeed447e0f25af1d75ecd4dae6aefd095cb3a6a57dc13789064f1d5c122ac
thread:issue:76037,issue #76037: __torch_function__ and generator input hazard,stale/low urgency,2,stale/low urgency,1,0,76037,,,,0,a632d5da5e73039957d4ea48d479c3bd556fcf1653e2a261f71a334c039af4f1
thread:issue:76040,issue #76040: SSL certificate error: urlopen error [SSL: WRONG_VERSION_NUMBER] wrong version number (_ssl.c:1091)>,stale/low urgency,2,stale/low urgency,1,0,76040,,,,0,aeeafd643f568fd4a45497944657085866b345cf44b07ffbbe8f3c18f24fbd8c
thread:issue:76043,issue #76043: RuntimeError: bucket_count == per_bucket_sizes.size() INTERNAL ASSERT FAILED,stale/low urgency,2,stale/low urgency,1,0,76043,,,,0,8e47124ebb24b411c55352b28af9add93d210a2f7cf0e8c5dbd144a2d9f58404
thread:issue:76047,issue #76047: NNC failing opinfo accuracy tests,stale/low urgency,2,stale/low urgency,1,0,76047,,,,0,15a3ed4f81b3ed38d7bbb91ffc3a108ac941a8c17faae7d824f4a7527175cd3e
thread:issue:76069,issue #76069: `init_process_group` hanging on HPC multi-node system w GPU,stale/low urgency,2,stale/low urgency,1,0,76069,,,,0,1df9e1d036b5fd10ae18081e103e9c54b57a181f7596eb389adae9c56d52e148
thread:issue:76070,issue #76070: NVFuser microbenchmark classifier - hash on memory formats,stale/low urgency,2,stale/low urgency,1,0,76070,,,,0,64df18c2a0378899e15ab694c76bbdec1e319d9d79c6975291ec7c957fa7808d
thread:issue:76071,issue #76071: HIPFFT_EXEC_FAILED when using AMD GPU run FFT,stale/low urgency,2,stale/low urgency,1,0,76071,,,,0,557d64d5047364105fc48ded321fbe9ff332fbbf729a7a2a1f2831e01cfbb408
thread:issue:76082,issue #76082: Eliminate uses of deprecated `FindCUDA.cmake`,stale/low urgency,2,stale/low urgency,1,0,76082,,,,0,bf6c42aca322348ec8545c2512d05c09cfc03bf32b309d8a9f992e18e2556d29
thread:issue:7610,issue #7610: Inconsistent interactions of PyTorch tensors and NumPy ops,stale/low urgency,2,stale/low urgency,1,0,7610,,,,0,9c6fac5499281c40b201d35c9bad564d520efa32e99b8551485f94eb8a102b69
thread:issue:76105,issue #76105: Failed to build on Ubuntu 18.04 due to bad MPI linker flags,stale/low urgency,2,stale/low urgency,1,0,76105,,,,0,15cd9d431c88fbc969008f8546b0d253419d16799661a848d25be3c48bbaa63d
thread:issue:76106,issue #76106: Allow any operation that takes a Storage to also take a contiguous Tensor instead,stale/low urgency,2,stale/low urgency,1,0,76106,,,,0,f240d83adb83150116a064acdd242bf4c3e7ce34dfd10771214b9f0d9e1d543b
thread:issue:76108,issue #76108: RuntimeError: [enforce fail at C:\actions-runner\_work\pytorch\pytorch\builder\windows\pytorch\caffe2\serialize\inline_container.cc:300],stale/low urgency,2,stale/low urgency,1,0,76108,,,,0,d38f8935cd70c78f6279f0b34c58fd987653d406d65ce9fb6d4fa69a8203dabe
thread:issue:76127,issue #76127: run_test.py option to write out failed tests,stale/low urgency,2,stale/low urgency,1,0,76127,,,,0,2f3d74cfdba0d7d893f289857f272a37c58cce4e714d0bba1541a44f8d7ca78d
thread:issue:76128,issue #76128: test_license_for_wheel always fails on my local dev copy,stale/low urgency,2,stale/low urgency,1,0,76128,,,,0,74f186de24017d835ed19feabab8dc7d5855775b0d4afe9effa638cd9694e554
thread:issue:76138,issue #76138: Coverage test is only checking packages and not all submodules,stale/low urgency,2,stale/low urgency,1,0,76138,,,,0,cceff07dd0373e374c4598d7243f6809782e9bcd14ed4635f9041cffe3a1037e
thread:issue:76160,issue #76160: test_wishart_log_prob fails locally for me,stale/low urgency,2,stale/low urgency,1,0,76160,,,,0,d1a8ab758941a0768ccff8bf72bded0f03903b0d7b11dbde325ae99a45ca67f0
thread:issue:76166,issue #76166: at::real and at::imag as methods,stale/low urgency,2,stale/low urgency,1,0,76166,,,,0,6caa3cbd7faa956c3545c5c0fff6306a1cb726b07b157f248d69846d7440c1e8
thread:issue:76175,issue #76175: Many dispatch keys do not print to string correctly,stale/low urgency,2,stale/low urgency,1,0,76175,,,,0,270cc68576740a4a79defdd2b18b535211748e6c13c39d5c6f81115502622e13
thread:issue:76186,issue #76186: fx: cannot find module <built-in method matmul> when using apex.amp,stale/low urgency,2,stale/low urgency,1,0,76186,,,,0,b7c3825dd8afee0fd0f439b18039a9109c973916782ccb15eff215edd4dc788b
thread:issue:76206,issue #76206: Update NCCL to 2.12,stale/low urgency,2,stale/low urgency,1,0,76206,,,,0,a8c12dd0e3173f759636f6cadecafff2eb0d83de436c12f8f729f92ac4bf3ce3
thread:issue:76225,"issue #76225: RuntimeError: bucket_count == per_bucket_sizes.size()INTERNAL ASSERT FAILED at ""/opt/conda/conda-bld/pytorch_1646755853042/work/torch/csrc/distributed/c10d/reducer.cpp"":980, please report a bug to PyTorch.",stale/low urgency,2,stale/low urgency,1,0,76225,,,,0,5e13f744bdf786c2a779ecae0fe54acf71a2606c9e1809a6b2950f67c7cf4d35
thread:issue:76232,issue #76232: Observing a strange behavior - Row parallelism,stale/low urgency,2,stale/low urgency,1,0,76232,,,,0,ee2022812c40a3be5f1eb6115125b76455fd176321ecfa3619597eec3a76c68b
thread:issue:76244,issue #76244: Initial integration of ZenDNN as backend into PyTorch,stale/low urgency,2,stale/low urgency,1,0,76244,,,,0,9003f6e0d3601e9d9134c5c133a7122379b7a8ce773fa77448312dbb23f1508b
thread:issue:76288,issue #76288: Multi-GPU distributed training reports errors,stale/low urgency,2,stale/low urgency,1,0,76288,,,,0,15a755a94ad81145b0584dc4666c7f5855c18ebac00f147de4f86f6cf0d469b9
thread:issue:76294,issue #76294: torch._remove_batch_dim is interceptable by __torch_function__ / batch tensors don't print correctly,stale/low urgency,2,stale/low urgency,1,0,76294,,,,0,72bc6307098a711ea024428ddb449cb08fe43b8d306d7f1f76d2be3c63db7606
thread:issue:76300,issue #76300: Unexpected _LinAlgError appeared only on my device,stale/low urgency,2,stale/low urgency,1,0,76300,,,,0,04e6aa8a6be7308c391ca73bdd920f996c4b5402342c841200c1ef16f3f6a5da
thread:issue:76304,issue #76304: Quantization fails when padding parameter given as string,stale/low urgency,2,stale/low urgency,1,0,76304,,,,0,e327311a56e5d85feec7acbc1def6f8f014093f8b34a3ea3bff9ed2d22cd7c06
thread:issue:76323,issue #76323: [Checkpoint] Add module documentation,stale/low urgency,2,stale/low urgency,1,0,76323,,,,0,5f2995f4db1a76b260dd42a591bcfc96c26c9721491159ae3c5f67f843e66152
thread:issue:76326,issue #76326: [checkpoint] Support models with different cross-rank metadata,stale/low urgency,2,stale/low urgency,1,0,76326,,,,0,bb5b2db21508db0ae545db83009454a798ef74fa770ec081e443957d7f40e6c9
thread:issue:76327,issue #76327: [checkpoint] Add extension points to avoid the default serialization behavior,stale/low urgency,2,stale/low urgency,1,0,76327,,,,0,58e1397d5fa8b364e5138a604a694ff087153da2fec4f26ff8f0568b5aaf0900
thread:issue:76331,issue #76331: [checkpoint] Avoid loading whole tensor when resharding,stale/low urgency,2,stale/low urgency,1,0,76331,,,,0,1616212c4e18af5f71a1485f50524707d25ef9951e55b6bc6a7ceb966973521d
thread:issue:76338,issue #76338: torch.bucketize doc typo on the left boundary when 'right=True',stale/low urgency,2,stale/low urgency,1,0,76338,,,,0,1fc43257542ce5841ab2ac1649a61e500a24ca3b7d2695ba5486614d8c4b9ba7
thread:issue:76347,issue #76347: `index_select` allows negative `index` for sparse but not for strided `self`,stale/low urgency,2,stale/low urgency,1,0,76347,,,,0,47c8dddc3506ab2291ecb4baefa4468e12a950d2cf14f8322f4cb803b78512cf
thread:issue:76362,issue #76362: Whether 'targetSize' in inferExpandGeometryImpl needs to be checked when it is less than 0,stale/low urgency,2,stale/low urgency,1,0,76362,,,,0,bd40cdccd8b86bce4f0cf3074877bb5777a5843cc09db096a01812a978ab0450
thread:issue:76383,issue #76383: Suggestion to throw a UserWarning when a user forgot .eval() mode during inference,stale/low urgency,2,stale/low urgency,1,0,76383,,,,0,7b09a27ef43387de6f6adf2e00666c42bf2239b11721097de2c5bec7914d9ef0
thread:issue:76441,issue #76441: [checkpoint] Switch away from pickle-base serialization,stale/low urgency,2,stale/low urgency,1,0,76441,,,,0,0794c3a50c40c3d059499cba4284caa4b112485c17b2e4309af91425fddb050a
thread:issue:76451,issue #76451: [checkpoint] Extension hooks to support logging and telemetry,stale/low urgency,2,stale/low urgency,1,0,76451,,,,0,766040369400b3383e6af9f8cf1483af02909cd67f655d91fb2b8c2ea006bc54
thread:issue:76465,issue #76465: Pyre type checking fails,stale/low urgency,2,stale/low urgency,1,0,76465,,,,0,7fad3aabe11bd76bb6354d6bc5574752a3a56beb180744da861f0c5462239f99
thread:issue:76487,issue #76487: `F.interpolate` uses incorrect size when `align_corners=True`,stale/low urgency,2,stale/low urgency,1,0,76487,,,,0,6226969d5c51e5573fbccfae32c81bf1cebc38ae37987063443fc564450224f7
thread:issue:76491,issue #76491: [NVFuser] Automated generation of microbenchmarks,stale/low urgency,2,stale/low urgency,1,0,76491,,,,0,8d1a0c138a01d261582097ef16b4b2c7e99dfb10a8d2b8201ec3bf4396bc75f9
thread:issue:76510,"issue #76510: Docs source link broken for all methods defined on Tensor in _tensor.py, e.g. `Tensor.register_hook()`",stale/low urgency,2,stale/low urgency,1,0,76510,,,,0,6780a7fc8ef21f8bede48eae28704a93d07fcd3986d6e67fcdb1449e21df8ab5
thread:issue:76519,issue #76519: Fix layout of masked output when all sparse dimensions are reduced,stale/low urgency,2,stale/low urgency,1,0,76519,,,,0,d48d77e5b49879b0b62b775fc4d53bb316d5b7d1065e99d79f7b89a6b1e24a04
thread:issue:76522,issue #76522: `torch.clamp` does not distribute gradients as element-wise`min/max` do,stale/low urgency,2,stale/low urgency,1,0,76522,,,,0,485a515d10081f955ff52e7e91f9715340fe7fa680dd89b7151f4a01a097fca3
thread:issue:76557,issue #76557: Pip packaging and publishing improvements in pytorch wheels for better integration with poetry,stale/low urgency,2,stale/low urgency,1,0,76557,,,,0,eaa86c88564ed5f9cc80cd58bf55d07f914feded643219299ed1c1096065d2a9
thread:issue:76585,issue #76585: `torch.linalg.cond` has different results for tensor requiring autograd,stale/low urgency,2,stale/low urgency,1,0,76585,,,,0,f11a7c70934cd348d30e3d843448f64dc654ad5383c913d02de159010fb2b466
thread:issue:76609,issue #76609: AttributeError: '_thread._local' object has no attribute 'rel_tol' (cannot use TestCase.assertEqual from other threads),stale/low urgency,2,stale/low urgency,1,0,76609,,,,0,d065a258b2104f8c047add9634fa33e4a684bd56212a88f63415320d53d085f3
thread:issue:76611,issue #76611: OpInfo CUDA bfloat16 support detection is buggy,stale/low urgency,2,stale/low urgency,1,0,76611,,,,0,a65558ba4b854465caf496cdcebd87beef54fabf6b5ce7432784c4e3af3a97e2
thread:issue:76612,issue #76612: OpInfo incorrectly advertises lu_solve support on CUDA even when compiled without magma,stale/low urgency,2,stale/low urgency,1,0,76612,,,,0,6f52be4ff26a22fae7ae4dbf241538f8c782025d6da7adf7e5778bdf11af1ebb
thread:issue:76613,issue #76613: MeanVarianceNormalization,stale/low urgency,2,stale/low urgency,1,0,76613,,,,0,0953162e7a4dc11ff1ecb1c1a2590fb6b8d0492217cfeaab063961edd128e2af
thread:issue:76618,issue #76618: pylint segfault,stale/low urgency,2,stale/low urgency,1,0,76618,,,,0,4e62e93405b488fb4977ee4a5551decd8027bd545741aebd4397f50235fadbb8
thread:issue:76627,issue #76627: `Tensor.logit`'s signature in doc misses `eps` argument,stale/low urgency,2,stale/low urgency,1,0,76627,,,,0,905b25c12fb8adbd9dd2257175153ed9760f8f3f301f747db52304c85267c9c2
thread:issue:76636,issue #76636: WeightNorm: Add reset_parameters Linear override,stale/low urgency,2,stale/low urgency,1,0,76636,,,,0,0fd67436e4dfffef2c78a0f99ef9288e596709602a6c44189add7118b77e5728
thread:issue:76638,issue #76638: Design API for accessing sparse tensor indices,stale/low urgency,2,stale/low urgency,1,0,76638,,,,0,0a7c1d3455b1b005a55d4f5101f192ced8d80f331ade0e5de37f1e23de8bf67d
thread:issue:76646,issue #76646: `torch.nn.HuberLoss` backwards unexpectedly fail,stale/low urgency,2,stale/low urgency,1,0,76646,,,,0,159f7b9dbc45e48eb19ce97c731fe7c558d1f54e581a8f912d4dcaaac3156da9
thread:issue:76649,issue #76649: Add `ldl_unpack` functionality,stale/low urgency,2,stale/low urgency,1,0,76649,,,,0,c2209443a62f3025f44116a66fd63b3ee8e35bcd8ce046b7c23627d02b696b8b
thread:issue:76656,issue #76656: Clarify dependency on NumPy (related to maskedtensor?),stale/low urgency,2,stale/low urgency,1,0,76656,,,,0,44f921d6fae2ea746673e0280e380d5f58d597d14ac46cf9020cb8076130e2d3
thread:issue:76689,issue #76689: Performance bad on ARM AArch64 for PyTorch C++,stale/low urgency,2,stale/low urgency,1,0,76689,,,,0,f7b7d68947c3dc6998c7175779ec1de9986a5c21214b64ec59e132f8621ce676
thread:issue:76705,"issue #76705: Torch `x += y.bmm(z)` is faster than `x.baddbmm_(y, z)`",stale/low urgency,2,stale/low urgency,1,0,76705,,,,0,f8932b697c8554cb5e45b883145c5ea91a78066472ab27d330ba1d65e4e4f1a6
thread:issue:76709,issue #76709: `reshape` for distributions.,stale/low urgency,2,stale/low urgency,1,0,76709,,,,0,2e1bfb34d5a661c75735cfeb4cc37a27b0f85f2b8c6e090e4af44489c33b4cbc
thread:issue:76732,issue #76732: Adding Polyloss to `torch`,stale/low urgency,2,stale/low urgency,1,0,76732,,,,0,144e7d40664fc6108a65237916b8875a18ffedf22c4dce903f18b8628d903db1
thread:issue:76733,issue #76733: test_python_reference_meta_functions takes too long to run,stale/low urgency,2,stale/low urgency,1,0,76733,,,,0,46bb303d0f596c3378bbefe6e1eaf9a9b34fbe8aa6dad0cd95fcc3f13fc0315c
thread:issue:76734,issue #76734: __torch_function__ callers should always pass kwargs,stale/low urgency,2,stale/low urgency,1,0,76734,,,,0,94a32395b40072e6d1115ba8a0589d02b685b13c60d486f6df34a04669b3c8f0
thread:issue:76743,issue #76743: Supporting torch.tensor.apply_ over GPU,stale/low urgency,2,stale/low urgency,1,0,76743,,,,0,c94770e29bf98e64c32b2f5a30643bc563955a5c1db4762379003f5e124bfb3e
thread:issue:76745,issue #76745: [checkpoint] Handle overlapping storage during save and load,stale/low urgency,2,stale/low urgency,1,0,76745,,,,0,4399416d526d3ad88b2ce392173b0c6a125058596bf9b896c69ad9df1c5ed2ce
thread:issue:76747,issue #76747: [checkpoint] Stable file format for checkpoints,stale/low urgency,2,stale/low urgency,1,0,76747,,,,0,40e37f8b50356d3b3a1f3c2394fb96eb9ba58e08bbacf78d03e29d5c6013ccc2
thread:issue:76750,issue #76750: Bug in dataloader iterator found by mypy,stale/low urgency,2,stale/low urgency,1,0,76750,,,,0,5530a7f77a64406c6a4743d8273ae075674b172bd9934e240253b83e5882c99a
thread:issue:76751,issue #76751: Default repr of __get__ methods in __torch_function__ is bad,stale/low urgency,2,stale/low urgency,1,0,76751,,,,0,8eed198b51e2f6e4173fadab36345364d671baa364707c5a7f043c1a7aa6f29c
thread:issue:76755,issue #76755: Quantization in Libtorch,stale/low urgency,2,stale/low urgency,1,0,76755,,,,0,f088de801507922ed6dd3a94eebcc0d5b291bba53227a00fe8a1e4b83af4a4e6
thread:issue:76779,"issue #76779: Strange warning from `matmul(..., out=...)`",stale/low urgency,2,stale/low urgency,1,0,76779,,,,0,c9b1cb94d7a0e1ed02ee1bb89dd3c53ae571b9bea25f4eb104a78b212903cb6a
thread:issue:76780,issue #76780: `gradcheck` should support the comparison of NaN,stale/low urgency,2,stale/low urgency,1,0,76780,,,,0,74dc00191b85530e9bb149308fe4947a5b3f4d3b1064cf381566321f69eb5e5c
thread:issue:76783,issue #76783: `gradcheck` fails for `torch.trace`,stale/low urgency,2,stale/low urgency,1,0,76783,,,,0,97edf66a52bd1d5b6edfa1756d27f2b415c5e117749a8df20d745e7aeeda61f2
thread:issue:76786,"issue #76786: `cumprod, prod` will backward fail if `dtype` argument is different than the dtype of input tensor",stale/low urgency,2,stale/low urgency,1,0,76786,,,,0,4b71a2766541191b4bcd652a461c156140ce309ff3f359da095b768560e64abb
thread:issue:76804,"issue #76804: torch.add bool x bool allows integer alpha, inconsistent with other dtype type checking",stale/low urgency,2,stale/low urgency,1,0,76804,,,,0,d740b273f9b650203152a452ea8f532f699f8aa864e56716f4b97e9d5b1fc1e5
thread:issue:76827,issue #76827: TYPEIGNORE lint run locally disagrees with CI,stale/low urgency,2,stale/low urgency,1,0,76827,,,,0,ca79ac2b99e237118e6537a0b03a3f3d86764c528d4e4e4aada43f5cec61de5f
thread:issue:76838,issue #76838: Windows CUDA TTS tracking task,stale/low urgency,2,stale/low urgency,1,0,76838,,,,0,d6f713698bee385fd728cf241e2e10d3db496952905c6985c0e018f3e351f64c
thread:issue:76844,issue #76844: Unify torch.ops argument parsing code with PythonArgParser,stale/low urgency,2,stale/low urgency,1,0,76844,,,,0,dbd858bd2dabf0b014f2bc77e9def99af197f33ebd0eda7f537e2b1b55ddf115
thread:issue:7686,issue #7686: Only one thread is used on macOS (super slow on CPU),stale/low urgency,2,stale/low urgency,1,0,7686,,,,0,a851a6b8a5e4668d7935d6eef80e6b984b35b1d8a2cdba352c742152afdeb940
thread:issue:76871,issue #76871: PrimTorch binary refs do not handle CUDA + CPU scalar tensors correctly,stale/low urgency,2,stale/low urgency,1,0,76871,,,,0,048483d920fbf64f5cffc06fd9eb0e987a8b62de367c2e0765b9a57378dba018
thread:issue:76887,issue #76887: Multiprocessing DataLoader hangs on exception inside iterator when using a simple queue and a producer thread,stale/low urgency,2,stale/low urgency,1,0,76887,,,,0,a7d9b9d1f879d66f7ed2b742b5af80eecff7a72e0f76a60d7974b9b775367177
thread:issue:76891,issue #76891: Unwanted behavior with some in-place operations on CPU,stale/low urgency,2,stale/low urgency,1,0,76891,,,,0,e487d91aac9dadea03ef6dc374e979b71cf35931dbc40305cfed5fd947e7f2bd
thread:issue:76906,issue #76906: torch.nn.functional.linear sometimes incorrectly accepts arguments of the different type,stale/low urgency,2,stale/low urgency,1,0,76906,,,,0,b01f67c420e29960558feefece0efb7c206e7396c65bf8546bbb77a010be3d02
thread:issue:76920,issue #76920: Avoid Self-loops on Module Creation,stale/low urgency,2,stale/low urgency,1,0,76920,,,,0,07a61895a6de3aa659c634244c56286871d2f9a21be159c4fadce0409a85d3f5
thread:issue:76933,issue #76933: NVFuser opinfos - check for CudaFusionGroup in the graph,stale/low urgency,2,stale/low urgency,1,0,76933,,,,0,c2fd631807cf9c26adddbcc03d6e01d306651b08d77c68e071a0073d5e93b7c9
thread:issue:76936,"issue #76936: Private API for accessing all ""internal"" attributes on Tensors",stale/low urgency,2,stale/low urgency,1,0,76936,,,,0,32f7183622b8d0b32954a4e42dd579195f69d17c300c674ede8d2f2402ddf162
thread:issue:76944,issue #76944: primTorch references don't handle scalar x scalar inputs correctly,stale/low urgency,2,stale/low urgency,1,0,76944,,,,0,a694c9e845e3d11a75e3f994542249cfffa5a064cda6a07d4f9825e28365b2f0
thread:issue:76966,issue #76966: [bug] `NATIVE` and `OMP` `parallel_for` implementations are inconsistent.,stale/low urgency,2,stale/low urgency,1,0,76966,,,,0,cc374b450ca7d25797b564c2745c7da63df6edf5fbde94dd47dbd77404f3cf85
thread:issue:76972,issue #76972: Display EC2 information,stale/low urgency,2,stale/low urgency,1,0,76972,,,,0,bfd61247352c9d61ee4254e5b733af93a86e4c085cfdb89f4d9b965e41ae2677
thread:issue:77016,issue #77016: Where is fx2trt fx to tensorrt tool?,stale/low urgency,2,stale/low urgency,1,0,77016,,,,0,11d83eea0ff431a8634f935ef2cc62fdc214ed0be141711bc8bba7c42b4e730f
thread:issue:77046,issue #77046: torch.stack test_conj_view and test_neg_view are failing after 77043,stale/low urgency,2,stale/low urgency,1,0,77046,,,,0,1e9401a558846a5512dfd9bfff36604da1339a54c2a4db3685968d14b7d37660
thread:issue:77047,issue #77047: Stop calling sizes/numel/dim/is_contiguous on undefined tensors,stale/low urgency,2,stale/low urgency,1,0,77047,,,,0,42676d5ee73ca9642ac0ae6300610ebcea20fe0439ed6e895ff5d426f41cc6be
thread:issue:77049,issue #77049: Peak GPU-memory usage extremely huge when sorting with torch.sort,stale/low urgency,2,stale/low urgency,1,0,77049,,,,0,2f0b072bfd7c0fe83673a88e11369b3ea8b6ca45885855f110afeb3f4100b055
thread:issue:77050,issue #77050: ENORMOUS OVERHEAD from mp.get_context('spawn'),stale/low urgency,2,stale/low urgency,1,0,77050,,,,0,83cb3ebee8b1b3bd8efa613b35f0e926954e332c45e944f5f12af8b49185d070
thread:issue:77052,issue #77052: Building from source results in broken __version__,stale/low urgency,2,stale/low urgency,1,0,77052,,,,0,4afc375c5c14021c720c0a90ede52be60d093bbe2b249d1553aaf048aa5949b3
thread:issue:77053,issue #77053: Incorrect documentation in ``gumble_softmax`` function.,stale/low urgency,2,stale/low urgency,1,0,77053,,,,0,8dca21abc983c465de0b858c10af101a2c632594123a325c0b46db8b681d27e3
thread:issue:77067,issue #77067: Unable to continue adding modules to `nn.Sequential` after using `del` method,stale/low urgency,2,stale/low urgency,1,0,77067,,,,0,f51c697f4dcba3066b5028200bd9f2c548bc750700757cfaa1d3fb8248b080c9
thread:issue:77118,issue #77118: __name__ on OpOverload should not contain period,stale/low urgency,2,stale/low urgency,1,0,77118,,,,0,67c4143b7ea08fd4237bf0e7c97e387e3f4ef1f5783b7243f512f965c0113a15
thread:issue:77159,"issue #77159: A somewhat cryptic error message (for newcomers) - ""Cannot re-initialize CUDA in forked subprocess"" - report and suggestion for a possible solution",stale/low urgency,2,stale/low urgency,1,0,77159,,,,0,8e4068cbcece7dd4fecc02be6d6e12ddad3524380fd338e79cf7cac95bd5d0ab
thread:issue:77161,issue #77161: SymInt shouldn't be in dynamic_type.h,stale/low urgency,2,stale/low urgency,1,0,77161,,,,0,c1051bd0af631393a536ae82fba4e2236e96803339ad7d59faabdd6f41d7582d
thread:issue:77165,issue #77165: Automate cleanup of header includes,stale/low urgency,2,stale/low urgency,1,0,77165,,,,0,b152d720cff7b4907121003c7bed2cbc7da09671dafb9e7db97f90f9d489e903
thread:issue:77216,issue #77216: [primTorch] Reduction references don't return views consistent with their original operators,stale/low urgency,2,stale/low urgency,1,0,77216,,,,0,4d4d706a6f512a87328a86fb50fa9d627d57b9a2bd4704ccf3b7f6947816317e
thread:issue:77230,issue #77230: fast `gradcheck` fails when outputs that do not require grad precede outputs that do,stale/low urgency,2,stale/low urgency,1,0,77230,,,,0,af61b027e7402953cc91fcabc6345fdbcaf638530e1e9fa5070991e445693505
thread:issue:77231,issue #77231: `torch.scatter_add` will succeed when the `index` is a complex tensor,stale/low urgency,2,stale/low urgency,1,0,77231,,,,0,4ff9b39fbce6a46ead59b15b881f7e24e85ade2b91d56dd80c9c7c0dedd5bec7
thread:issue:77241,issue #77241: libtorch1.8 torch::sigmoid is wrong,stale/low urgency,2,stale/low urgency,1,0,77241,,,,0,a4aeeea55a05c2b5431734d6b52785885e17c8ba456eb1dac003e646fb93e54e
thread:issue:77251,issue #77251: The codegen unconditionaly generate code even when it is not going to be used,stale/low urgency,2,stale/low urgency,1,0,77251,,,,0,907fd9fbe992e84d70170b6ad9847a087dbbef985f765a3f8ac4b6de00afd875
thread:issue:77253,issue #77253: lintrunner not working,stale/low urgency,2,stale/low urgency,1,0,77253,,,,0,8b18ba9b415602d1dc36460b6f219da7b6095e71ba07cfb6be3e018722de2dd9
thread:issue:77256,issue #77256: CrossEntropyLoss computes SoftMax always across the second dimension,stale/low urgency,2,stale/low urgency,1,0,77256,,,,0,ffcaa70a701392ea72dd6b01e567db5ada1b2979d2c82112f9988f9234f2ccf8
thread:issue:77265,issue #77265: Subclasses with unwrapping `__torch_dispatch__` impls as parameters,stale/low urgency,2,stale/low urgency,1,0,77265,,,,0,2d84a6ab13672a2e6af9d1e0eefcbcd1045ee633233e1f1290474a9c5713f933
thread:issue:7729,issue #7729: [Caffe2][Caffe] Caffe to Caffe2 ParseError,stale/low urgency,2,stale/low urgency,1,0,7729,,,,0,d2dd7979a2d486de6ec146f886270c775eda468ad9bd6467488ac29a84e7a0bf
thread:issue:7733,issue #7733: Inserting a tensor into a python dict causes strange behavior,stale/low urgency,2,stale/low urgency,1,0,7733,,,,0,5d3ce566227e4807470ef841a7ec8757c6daa7146119812db85c1baaa18269d1
thread:issue:77340,issue #77340: Disable issue doesn't disable multiple dtypes correctly,stale/low urgency,2,stale/low urgency,1,0,77340,,,,0,e03746a5730bbe8ebf6db05c4ef61b935935103bce825d6d2c76be0a9f8aa84f
thread:issue:77380,issue #77380: microbenchmark-style tests,stale/low urgency,2,stale/low urgency,1,0,77380,,,,0,fce36ce966a22ad51b5e2878593c50f2a8f62767a5fa285418ab53c7cda82604
thread:issue:77397,issue #77397: Large numerical inconsistency for `torch.einsum` on RTX30 series GPU.,stale/low urgency,2,stale/low urgency,1,0,77397,,,,0,df0c43871a3a531c9033a49a94af9960257ab9952f3cb5e7ccdd02f7ac9ca8a8
thread:issue:7740,issue #7740: Conv3D can be optimized for cases when kernel is spatial (probably),stale/low urgency,2,stale/low urgency,1,0,7740,,,,0,bebee7a49c140c57ed06c766ed39a42e077821e3fa45878227737b3969bc819c
thread:issue:77411,issue #77411: Allow force building with/without AVX,stale/low urgency,2,stale/low urgency,1,0,77411,,,,0,a47edb507f34d0da312a1fa969f9d9e033ada0b4cbdabc39e266ed98b24124e1
thread:issue:7745,issue #7745: [Caffe2] Align Element-Wise Ops Broadcasting to Numpy,stale/low urgency,2,stale/low urgency,1,0,7745,,,,0,183af58b6b253ce8cec565d2f82de78bcef99ed0a8d68374ca9cbde447858584
thread:issue:77454,issue #77454: More clarity in doc for `torch.cuda.Event.record`?,stale/low urgency,2,stale/low urgency,1,0,77454,,,,0,8253afd69e87241eddb247deecee6345fdc7ea046e56ddd8feb247f551c60cbd
thread:issue:77463,issue #77463: conda CPU installation for LTS fails with UnsatisfiableError,stale/low urgency,2,stale/low urgency,1,0,77463,,,,0,0f3346479b0e5d946ab993bbeabec95fe62d74e3e3f6580c2606eb466179a7b4
thread:issue:77514,issue #77514: `max_pool1d` can succeed when padding is negative for tensor requiring grad,stale/low urgency,2,stale/low urgency,1,0,77514,,,,0,f71c5fa82c4df9d53587227a636b6176801aea33f884f9d54664dcc8ceae9182
thread:issue:77515,issue #77515: Inplace Bool API + `sum` will trigger INTERNAL ASSERT FAILED,stale/low urgency,2,stale/low urgency,1,0,77515,,,,0,d90af56f1bfa97f85d9f48a63d0e3b3597a31af61cfbc5650d78556bde811e5e
thread:issue:77529,issue #77529: softmarginloss should use `log1p` and has an incorrect out= behaviour.,stale/low urgency,2,stale/low urgency,1,0,77529,,,,0,d3b863a071977f0ec555647e9a3041d3a6d383d52e110e7fc2457f763f31465f
thread:issue:77537,issue #77537: batch Kronecker product,stale/low urgency,2,stale/low urgency,1,0,77537,,,,0,04afe5e8039a04965bde0c57d822974e6e0ecd8d9e0b644fc435a92e0421c652
thread:issue:77538,issue #77538: Failed to run on iOS - Couldn't find an operator for `aten::conv1d`,stale/low urgency,2,stale/low urgency,1,0,77538,,,,0,e3eecc31cd67d486e2a7fc2c85fac84018fe085742d88072146d5d2f36022deb
thread:issue:77544,issue #77544: Modernize LoggingTensorMode,stale/low urgency,2,stale/low urgency,1,0,77544,,,,0,3f0064898cfe8b0a21a69084bb32ce54f1051a331d69fc93593a7544356bfe22
thread:issue:77546,issue #77546: Build check for AVX512 fails with AMD CPU and march=native,stale/low urgency,2,stale/low urgency,1,0,77546,,,,0,fb03d68bb84b1505eb03f5bcd5f784b9672cba22d68ba8e15d15edfbf9668975
thread:issue:77548,issue #77548: lintrunner doesn't give good error message suggesting lintrunner init,stale/low urgency,2,stale/low urgency,1,0,77548,,,,0,06367aa02b798487d4baf03d2cf8c9365001a3c7119c4998d33fb2a6b77921b4
thread:issue:77552,issue #77552: Functional Jacobian does not work with Torchdiffeq,stale/low urgency,2,stale/low urgency,1,0,77552,,,,0,6748f6d592c7dcb0c65b0f47c7b876b15e758ba1655a38a62f0bada7a9a52087
thread:issue:77576,issue #77576: `stateless.functional_call` doesn't work with `nn.DataParallel`,stale/low urgency,2,stale/low urgency,1,0,77576,,,,0,8e1f1564877f9b78bc19049b39b64424f2cd2299b795aaa0f2a6e2a893e7a5b9
thread:issue:77589,issue #77589: How to handle __module__ attribute for Public API bindings,stale/low urgency,2,stale/low urgency,1,0,77589,,,,0,f125d678079e24209da08d8c2f04320b895b62400a2bd5c18ba6e2aaed0ccf34
thread:issue:77614,issue #77614: PyTorch fails to build on gcc 12 due to gloo,stale/low urgency,2,stale/low urgency,1,0,77614,,,,0,58aba51e725269c5952f03e3a99dcb1a879bbfe8b8a2c448ddf70eb631cea38d
thread:issue:77646,issue #77646: Werror=nonnull in dataloader.cpp (part of tests),stale/low urgency,2,stale/low urgency,1,0,77646,,,,0,63bae8212885934896748e109cae84ee61e34daf6898f3914c7b74332d8a5b57
thread:issue:77668,issue #77668: Extend BC test to test for __torch_function__ overridability,stale/low urgency,2,stale/low urgency,1,0,77668,,,,0,06c8e7e9f677e640b2421f988bdba7a5ccbbd11eea54fd677f5ffd523de84561
thread:issue:77675,issue #77675: PrimTorch refs do not match argument naming with their PyTorch counterparts,stale/low urgency,2,stale/low urgency,1,0,77675,,,,0,cb4024cb3c7cb783b93fdc4a1af45a04b395005f300573c6a2f123ad8638d1e6
thread:issue:77724,issue #77724: FSDP: enhanced shared parameter support,stale/low urgency,2,stale/low urgency,1,0,77724,,,,0,6389c46ecb7720474815932c9d38498737dd63cfbe9dd83386bca3acdf1675d9
thread:issue:7773,issue #7773: [feature request] Add Local Contrast Normalization,stale/low urgency,2,stale/low urgency,1,0,7773,,,,0,5dfd49be4e1a4c9b8018f5296a116d9c83d24dd0b462472de414e2ba37d4781c
thread:issue:77733,issue #77733: Adding Vulkan Support,stale/low urgency,2,stale/low urgency,1,0,77733,,,,0,0e90a7b05bc8fa7a1a903c357c04e83120a0d2763659280a1d6acfe6bd1e199b
thread:issue:77736,issue #77736: TimeSeriesDataset retrieve columns,stale/low urgency,2,stale/low urgency,1,0,77736,,,,0,ea2d5d88d4e717602c2c59354ba40a5f8735310db95fa00b3fca41c490319627
thread:issue:77737,"issue #77737: torchrun leads to `ModuleNotFoundError: No module named 'tensorboard'`, but python -m torch.distributed.launch is ok",stale/low urgency,2,stale/low urgency,1,0,77737,,,,0,35375ff6fd5de342edda5f85f9b569b140c0ee402e3938c911d9330aee3fc34c
thread:issue:77738,issue #77738: net_observer_reporter_print.h missing,stale/low urgency,2,stale/low urgency,1,0,77738,,,,0,7cf20baf9d68b7a58672e1e0f6f8ef87531de861f3c891fc19fc8196e0fb5bdc
thread:issue:77801,issue #77801: `torch.utils.benchmark.examples.blas_compare` can not be parsed by Python-3.7 runtime,stale/low urgency,2,stale/low urgency,1,0,77801,,,,0,cfa85a4c1fb8b82b02738d3fb89bfc69f282877076520cfcf49652f42cc4e72e
thread:issue:77808,issue #77808: `Could not start gRPC server` flakiness in XLA tests,stale/low urgency,2,stale/low urgency,1,0,77808,,,,0,70c85ae6b4aa14e0c0dd1c0d0165c30c6052c3f97258fb7f0332cf6774ca5236
thread:issue:7781,issue #7781: [Caffe2] Fail to build after upgrading to cuda 9.2,stale/low urgency,2,stale/low urgency,1,0,7781,,,,0,c2679225b622a8b89b4782b87f4d090188ee377c48bd3f54b72d0879e645ba9b
thread:issue:77821,issue #77821: cannot convert to channels last format for conv2d conv3d hybrid model,stale/low urgency,2,stale/low urgency,1,0,77821,,,,0,574843751126eaf4ac6aed1dfd045b1d8ecbbd65c03097069a324027c7cd22af
thread:issue:77838,issue #77838: Sporadic convolution error with dilation=0,stale/low urgency,2,stale/low urgency,1,0,77838,,,,0,af2f69b776f576f2f92fc2c77c3177544211e105a85d1e0f9d9968ad7d55307c
thread:issue:77839,issue #77839: BUG: reference count leak when using `THPLayout_New` and `THPMemoryFormat_New` (static analyzer reports),stale/low urgency,2,stale/low urgency,1,0,77839,,,,0,0b83c815f378f77bae7346485b0e792b4e178247831be4617afa1a8cfe03d663
thread:issue:77844,issue #77844: Conv2D with large different number of input and output channels gives a CUDNN_STATUS_INTERNAL_ERROR,stale/low urgency,2,stale/low urgency,1,0,77844,,,,0,c6c620c819c6743c980674cf34bfab7dfa94933081d4a7ab5475f36f61fb0d4c
thread:issue:7786,issue #7786: [feature request] Simple and Efficient way to get gradients of each element of a sum,stale/low urgency,2,stale/low urgency,1,0,7786,,,,0,5d94a25afc98d433a7d6da47aee778532ff9e74852f6684bf1aea4052e2b7d7e
thread:issue:77880,issue #77880: Make the appropriate backend `DimensionNode` visible to LTC core,stale/low urgency,2,stale/low urgency,1,0,77880,,,,0,0ba29db8353704a942da19710fe639602cc022087136686b18d35202a8bc8f13
thread:issue:7789,issue #7789: detectron net create error,stale/low urgency,2,stale/low urgency,1,0,7789,,,,0,0860e1115ad0586425436e2faed61a2c5fac34ac7297367c45efd3b3314e6c3f
thread:issue:77894,issue #77894: Floating point exception in channel_shuffle,stale/low urgency,2,stale/low urgency,1,0,77894,,,,0,cf153351657ef10dcdf888978ff51281f3b8679ab0f6f6f59cdfd5422e7e6382
thread:issue:77899,issue #77899: Floating point exception in native_channel_shuffle,stale/low urgency,2,stale/low urgency,1,0,77899,,,,0,1df89b233a57bc7f59e8a5c457e8be18a5993818e1ee6f27f6332ea61293fbe5
thread:issue:77900,issue #77900: Floating point exception in slow_conv3d,stale/low urgency,2,stale/low urgency,1,0,77900,,,,0,990f58423fe0f9900a0aa6817b291a8d66e22ccabc30604c5ed1d6a0886dc792
thread:issue:77901,issue #77901: Heap corruption in slow_conv_transpose3d,stale/low urgency,2,stale/low urgency,1,0,77901,,,,0,d62b41110767d7dc02dbaeee7b4f49bcf1dfa9681bac662323a45cdb1145cf49
thread:issue:77939,issue #77939: Fails to compile with GCC 12.1.0,stale/low urgency,2,stale/low urgency,1,0,77939,,,,0,e6858077f30dc61eb1573d2a00555b7921ac92171f9f4344a1b6986be557f357
thread:issue:77946,issue #77946: [failing test] test_foreach::test_binary_op_scalarlist_fastpath,stale/low urgency,2,stale/low urgency,1,0,77946,,,,0,8ac01861713b5497c405521753574d2a8d33449910579c6962dc82cd033d9ac9
thread:issue:77951,issue #77951: `topk` returns different results with the same input twice in cuda,stale/low urgency,2,stale/low urgency,1,0,77951,,,,0,e6a05aa42ea512ccb0e7533449dd681e5b6e38b913063bad2f9f1f57f1fdcc31
thread:issue:77955,issue #77955: NaN tensor values problem for GTX16xx users (no problem on other devices),stale/low urgency,2,stale/low urgency,1,0,77955,,,,0,3507238ac2a3a0b389091a290894eb205b7062eb6e40b659d95f31ac8158fcf0
thread:issue:77962,issue #77962: Operating on boolean torch tensor and numpy array casts to `unit8`,stale/low urgency,2,stale/low urgency,1,0,77962,,,,0,637f8e5cae415f6f5146d26a9a99e2e0ac13cdfb0fceb8ffe86dca89861463cd
thread:issue:77963,issue #77963: `logaddexp2` fails to backward,stale/low urgency,2,stale/low urgency,1,0,77963,,,,0,84ab27eba849adc1184787316d34bfe0e522d5d9514870adaa256e65bc737d89
thread:issue:77967,issue #77967: Inference Tensors should not be allowed to hold `grad_fn`,stale/low urgency,2,stale/low urgency,1,0,77967,,,,0,23184015e53de7ec277d689a031c8e3e216479b024f25113f64f007fcc413d71
thread:issue:77973,"issue #77973: forward-mode support for ""logically composite"" operators",stale/low urgency,2,stale/low urgency,1,0,77973,,,,0,45a78e134819c4570bad9c9c863131393e3e472583485fdf850780338a3671ed
thread:issue:77981,issue #77981: Override sym_sizes to create LTC IR for SymIntNode,stale/low urgency,2,stale/low urgency,1,0,77981,,,,0,1de06923917daa67a7ecd68f43b2f5d273b20560de8e9d737508163dec14c819
thread:issue:7801,issue #7801: Checkpointing is slow on nn.DataParallel models,stale/low urgency,2,stale/low urgency,1,0,7801,,,,0,3b192b3ef9a92cc6392d711422396ac397b86a786ab997c90e182f7b216acb15
thread:issue:78018,issue #78018: Three memory copies of every dataloader cpu tensor,stale/low urgency,2,stale/low urgency,1,0,78018,,,,0,ce92ae2965d11aecd0050a406c0c46fb5a3ef3949042bbfd33c74d2df660ffab
thread:issue:78047,issue #78047: DDP multi host with single GPU each.,stale/low urgency,2,stale/low urgency,1,0,78047,,,,0,0e8ad47275c05471fe794bf59b6e05378e9ab031329780aa491e7f15470ff82a
thread:issue:78053,issue #78053: toleranceOverride should override atol and rtol even when explicitly specified in a test,stale/low urgency,2,stale/low urgency,1,0,78053,,,,0,7232e31493b7ee2051493869e8e3052e49e65f5fc7f26d717f2dffca9bf62be7
thread:issue:7806,issue #7806: OOM Exception when using torch.nn.grad.conv2d_weight (apparently because CuDNN backwards is not used),stale/low urgency,2,stale/low urgency,1,0,7806,,,,0,e0dce757a6a819b9c275d28b2d82c3e56add306da4aff41ec76a40056d2e9ae5
thread:issue:78061,issue #78061: Python memory allocator called without holding the GIL when running torchrun under Python debug version,stale/low urgency,2,stale/low urgency,1,0,78061,,,,0,5b581fd89d2791ab7ccb792e3f83424f5a17d23783870a3b94794402976f95eb
thread:issue:78064,issue #78064: nn.CosineSimilarity returns value larger than 1,stale/low urgency,2,stale/low urgency,1,0,78064,,,,0,f47c3029be4e2b072e7922d7a880ecd7dc45fd216aee3021362920d6631c445f
thread:issue:78067,issue #78067: Installation on Jetson target board,stale/low urgency,2,stale/low urgency,1,0,78067,,,,0,342e63499eec0f6405b83f25662cbb200b004de07072cc020ada4bf66cb27cfa
thread:issue:78070,issue #78070: [primTorch] Meta function for item creates a dummy value,stale/low urgency,2,stale/low urgency,1,0,78070,,,,0,487f79924fea0e8e45b56a60ad1cdfcef584132b9f94c0ddd09ca558bc315e56
thread:issue:78075,issue #78075: torch.multiprocessing.spawn raise PicklingError inside a decorator,stale/low urgency,2,stale/low urgency,1,0,78075,,,,0,da8afd6699f9328a8408ff41ff6258665babc5b4184f1ebe05fc04dc0abec427
thread:issue:78122,issue #78122: Segfault in bincount,stale/low urgency,2,stale/low urgency,1,0,78122,,,,0,4d8ab3daf1ceeb33facc2757de73e3f44904e84a352ab29687b41009bc2d96eb
thread:issue:78125,issue #78125: Segfault in grid_sampler_3d,stale/low urgency,2,stale/low urgency,1,0,78125,,,,0,b21add69174cae81def486ab3d380baa25c3eb112461e53bb2145b93713a58ca
thread:issue:78126,issue #78126: Segfault in max_unpool3d,stale/low urgency,2,stale/low urgency,1,0,78126,,,,0,25d16808650b7701e7318297511a0198c96880ac6a4d589b65a8d1bc71faadb2
thread:issue:78127,issue #78127: Segfault in torch._C._nn.reflection_pad2d,stale/low urgency,2,stale/low urgency,1,0,78127,,,,0,ff4dffcd63df88dc47d808cf4a01de3ca40a6a48004125bff0dfa6f311ea2785
thread:issue:78128,issue #78128: Segfault in torch._C._nn.thnn_conv2d,stale/low urgency,2,stale/low urgency,1,0,78128,,,,0,8a7daf97f5083cca6cc943bd61473eb1ad848a34f31c90393a01954efdd4dbdc
thread:issue:78129,issue #78129: Segfault in _embedding_bag_forward_only,stale/low urgency,2,stale/low urgency,1,0,78129,,,,0,b7cec4d58e14fd340c28a8cb7f912d4b95d60e7b92caf86eebe29cdd43a0d180
thread:issue:78130,issue #78130: Segfault in _grid_sampler_2d_cpu_fallback,stale/low urgency,2,stale/low urgency,1,0,78130,,,,0,06c6d73c245e19a819b12c9851c8a269487929fca023f98b5affda40dcde17d4
thread:issue:78131,issue #78131: Segfault in _pad_packed_sequence,stale/low urgency,2,stale/low urgency,1,0,78131,,,,0,9af9dc323fe3cc034054e9f262650999522c3c4331f6da334912a7e543dfe82a
thread:issue:78151,issue #78151: `ctc_loss` will backward crash,stale/low urgency,2,stale/low urgency,1,0,78151,,,,0,8f6b52d5c03e3bdec98d32182e692765b487170c6e6a56437c92b86464e4881a
thread:issue:78153,issue #78153: `pack_sequence` crash,stale/low urgency,2,stale/low urgency,1,0,78153,,,,0,a5ab8dfa3dbf066dfb20da1df9407084e1969a19a3ee74f64ce2adb25de99415
thread:issue:78157,issue #78157: [Optimizer Overlap] Custom optimizer registration,stale/low urgency,2,stale/low urgency,1,0,78157,,,,0,694ac1182e6c57e5ac736ca7f7922038266dfd291cb29b785760f7b516d64693
thread:issue:78159,issue #78159: [Optimizer Overlap] Parameter group support,stale/low urgency,2,stale/low urgency,1,0,78159,,,,0,e0f8886058d8a9a4666c8dc0cbece66fdb778cbc7483b856281b9fb191990dfb
thread:issue:78170,issue #78170: [chalf] reference_testing: low quality test for fast growing ops,stale/low urgency,2,stale/low urgency,1,0,78170,,,,0,f8f165ede43ef706b13c68e35bead78aaa99719c60e3bf8be92bd685d5445a77
thread:issue:78185,issue #78185: test_to (__main__.TestTorch) fails with multiple gpus,stale/low urgency,2,stale/low urgency,1,0,78185,,,,0,91b700659c57c594d5901176aa105d297aefb7cc59832a8c0481200af7ccabb2
thread:issue:78201,issue #78201: nn.Sequential causes fx.replace_pattern to not find any match.,stale/low urgency,2,stale/low urgency,1,0,78201,,,,0,29a17c554c289165625c8642bb912edd3977dfad7b30ea3a0050e65866374f2e
thread:issue:78210,issue #78210: Performance with MPS on AMD GPUs are worse than CPU,stale/low urgency,2,stale/low urgency,1,0,78210,,,,0,4b8504c96789003bea3cc273cba866052fa7fe36e74c8b011528057bc9bf99c7
thread:issue:78248,"issue #78248: USE_NATIVE_ARCH flag causes nvcc build failure due to ""'arch=native': expected a number""",stale/low urgency,2,stale/low urgency,1,0,78248,,,,0,866095da8b0ab389cba1b3cd3e7665d19dc4a45bd3d93bd7bdfec27a27da8e42
thread:issue:78249,issue #78249: `max_unpool2d` is not deterministic,stale/low urgency,2,stale/low urgency,1,0,78249,,,,0,b32da41b90270ccd2159baddc9a14b99b923e49c9091e012582f61740800d05a
thread:issue:78253,issue #78253: Any plan to add Noam scheduling?,stale/low urgency,2,stale/low urgency,1,0,78253,,,,0,06384546eb0e8223d6978632e304fe433d60fec8e72dc1df1fa41a07ffa1dcfa
thread:issue:78255,issue #78255: Floating point exception in _conv_depthwise2d,stale/low urgency,2,stale/low urgency,1,0,78255,,,,0,efe0be725e3984b4d95031d89bc1253ef29feb1004d6dbb4fa693887a28c50db
thread:issue:78260,issue #78260: [CI] Detect when tests are no longer running from CI,stale/low urgency,2,stale/low urgency,1,0,78260,,,,0,e4004d54302bb663aa7710f9861c3c997b21bc19cdcb4067b3bd09f7855eaa67
thread:issue:78261,"issue #78261: pytorch-android-lite use its own libfbjni.so, which is not compatible with any other version at all..",stale/low urgency,2,stale/low urgency,1,0,78261,,,,0,d9354ae329a0c9f80ea362e4833200aa4bcf81549ba14ebbdc949d7a07544930
thread:issue:78274,issue #78274: Memory allocation errors when attempting to initialize a large number of small feed-forward networks in RAM with shared memory despite having enough memory,stale/low urgency,2,stale/low urgency,1,0,78274,,,,0,126354d0d8fa0f89c6fb8c37bee928fbef8e7235f86d670623bdf88b69c4228c
thread:issue:7834,issue #7834: torch.Tensor.new() disappeared in 0.4 doc,stale/low urgency,2,stale/low urgency,1,0,7834,,,,0,c820d237a79f60f26294c3485a618a1dbcc324c4391c88e7617df6eff315e8d8
thread:issue:78346,issue #78346: `gradcheck` fails for `torch.distribution.transform` APIs in forward mode,stale/low urgency,2,stale/low urgency,1,0,78346,,,,0,050b5c7933117c079c9cd33b3d42a2ad095e194ea3b9481c3afa5a7362f909bb
thread:issue:78367,issue #78367: Split up and reorganize RPC tests,stale/low urgency,2,stale/low urgency,1,0,78367,,,,0,e225d811400efdec0d5a5f39f8aa796cb6904d074f5012f46bbe41b435534760
thread:issue:78413,issue #78413: torch.angle differs from np.angle for -0.,stale/low urgency,2,stale/low urgency,1,0,78413,,,,0,8704c87f8fe4e929d3d034d86252edbb30f3bf70d0efc9c0a3b127f0f60ff3ac
thread:issue:78422,issue #78422: Permutation of Sparse Tensor,stale/low urgency,2,stale/low urgency,1,0,78422,,,,0,459d0efa9b4adfea5299220dbda386b77323271a45c62315a3a9b222269e4072
thread:issue:78435,issue #78435: fx.Tracer with param_shapes_constant=True not working for RobertaForMaskedLM,stale/low urgency,2,stale/low urgency,1,0,78435,,,,0,5083cd7f50198858800e4c0301ea468dc77963fac7e7ae4fa9811c09c96a5b84
thread:issue:78450,"issue #78450: Softmax, LogSoftmax are over parameterized",stale/low urgency,2,stale/low urgency,1,0,78450,,,,0,139ec32bc9f35ac5ac4fae8c1e9481fb94db55507c85c4d4a0a60507040d521c
thread:issue:78482,issue #78482: RuntimeError: Event device type CUDA does not match blocking stream’s device type CPU,stale/low urgency,2,stale/low urgency,1,0,78482,,,,0,8c8bfdbc4efc03c0a37229069c99dff26d08cee93a318c74a80a4a227e090d2c
thread:issue:78484,issue #78484: torch.lerp: discrepancy between CUDA and CPU (with extremal inputs),stale/low urgency,2,stale/low urgency,1,0,78484,,,,0,eceec52414efc558020be634e64e1b0984407e61f7fee42b87afffc4e342f034
thread:issue:78507,issue #78507: Parallel execution of multiple unrelated statements written sequentially,stale/low urgency,2,stale/low urgency,1,0,78507,,,,0,e1462f755cda489fcf786ca2fb04c981d3eaf20552a9edf419a56c877d08b5ec
thread:issue:78513,issue #78513: clear input shape declaration on pytorch model inputs and outputs,stale/low urgency,2,stale/low urgency,1,0,78513,,,,0,dd916611bb04c23af82e3bf4b5d03ea26bf646d5724176eeb9ccbd6e34a08cc0
thread:issue:78518,issue #78518: Exponentiating floating number with cuda tensor is slow,stale/low urgency,2,stale/low urgency,1,0,78518,,,,0,a2441b5b4e287adcaad45e7e62d1f132387cd1c27da4b2ca5c0c18a68ee6bcfc
thread:issue:78519,issue #78519: test_python_dispatch fails on DEBUG=1,stale/low urgency,2,stale/low urgency,1,0,78519,,,,0,28ce1646ae2343d3f1bd14a4e18bb4376b370db1e7fffb2c9cfba9544326cb6f
thread:issue:78530,issue #78530: LibTorch cannot be used without nvcc,stale/low urgency,2,stale/low urgency,1,0,78530,,,,0,33efa8a02bbed9d5d9cc2d0eafda98dec9b391d9c16c38b21c5fde73e2a54454
thread:issue:7857,issue #7857: Feature Request: Logistic Distribution,stale/low urgency,2,stale/low urgency,1,0,7857,,,,0,25e949250eade0f8eba0c80cfcc261ccbc126f37ba52ff57d40caec45a094574
thread:issue:78581,issue #78581: RFC: Improve the performance and usability of linear algebra on CUDA devices,stale/low urgency,2,stale/low urgency,1,0,78581,,,,0,05ed31fdc9e1bb756e146e12718edd2ebaa511c128415c83137b46e20ee5fe12
thread:issue:78624,issue #78624: linalg.pinv_singular tests are slow,stale/low urgency,2,stale/low urgency,1,0,78624,,,,0,44756336cf22280ff925b810eb7a91c7f7835772818fad3a68e9bdb763932b1c
thread:issue:78634,issue #78634: Debug job does not build in debug mode,stale/low urgency,2,stale/low urgency,1,0,78634,,,,0,5021691a7733cad97614318abbfcebadfab438247e76d31c3e6dd8fe5bfc0580
thread:issue:7865,issue #7865: [PyTorch] weight tensor dimension assumption,stale/low urgency,2,stale/low urgency,1,0,7865,,,,0,a8b0fd5c9730b201a85b9da0a72bbd5e7ff8b6de6772b4873171bb2927e60c3d
thread:issue:78656,issue #78656: Allow batch_norm_backward_elemt and batch_norm_gather_stats_with_counts handle 0 counts,stale/low urgency,2,stale/low urgency,1,0,78656,,,,0,4e17a6055a0ab28a0c997caafd312fa5fdab19f476e0c0e9535abbfd9848d972
thread:issue:78708,issue #78708: BuildExtension does not choose correct CUDA installation,stale/low urgency,2,stale/low urgency,1,0,78708,,,,0,57fef7a8d2e1d5761f244d2d79888b8a8611d40488c057df5e915a425db60270
thread:issue:78720,issue #78720: ValueError during `yaml.dump(dtype)`,stale/low urgency,2,stale/low urgency,1,0,78720,,,,0,8a0385063fe166e2c0dff644fed49cbafc17da2ebc5bd6232e5539ae1607a2c4
thread:issue:78722,issue #78722: AlBert quantization,stale/low urgency,2,stale/low urgency,1,0,78722,,,,0,5e5f1f736ea914eb0f4e2e79fc97717bd6cf6ddc80252a6bd0ba259bba744d84
thread:issue:78729,issue #78729: [Better Engineering] Make OpInfo-based test failures easy to reproduce,stale/low urgency,2,stale/low urgency,1,0,78729,,,,0,e8ecc2d8dd1d919695ca7cd8148432357f0a4d4f690923303ae85068ccec8e25
thread:issue:78737,issue #78737: New c10 constants,stale/low urgency,2,stale/low urgency,1,0,78737,,,,0,006a8d2b27ebdbfa9316cc89ea13250a499cefef01c58427dea1c48e1f78b5c6
thread:issue:78741,issue #78741: Test approximation and numerical stability of numerical operators,stale/low urgency,2,stale/low urgency,1,0,78741,,,,0,ec43f10ff555c64fa26888062f601eab07e5330b8493a865b98933bbe08936ad
thread:issue:78742,issue #78742: Add a gallery of examples with sphinx-gallery,stale/low urgency,2,stale/low urgency,1,0,78742,,,,0,6d5009918c4d7044255bab1d5792ac223f1501f842ff1d6feb8631cbdbe81d78
thread:issue:78743,issue #78743: Expose docs from the yaml for each torch.Tag in Python,stale/low urgency,2,stale/low urgency,1,0,78743,,,,0,e8cd611b192cbb95e7261d4690155fd6c6c03d444fb5ee1f752e285d84e51b6e
thread:issue:78744,issue #78744: Unable to programmatically update models using references from model.named_modules()...requires additional parsing,stale/low urgency,2,stale/low urgency,1,0,78744,,,,0,046d1a7b19992440ca650c4294c84c84933b7469c623be26dd24eba382110164
thread:issue:78754,issue #78754: Getting NotImplementedError when trying to implement E2E support for `prim::is_nested` Op in torch-mlir.,stale/low urgency,2,stale/low urgency,1,0,78754,,,,0,4fbba208ba3a54583cb61d761ae1629d006346f96f8499587102a22dae6b34b0
thread:issue:78759,issue #78759: Add `inplace_view` tag for `resize_`,stale/low urgency,2,stale/low urgency,1,0,78759,,,,0,3cc1c925f17eb6e4df6c883c74497ad87fff1f8b3e1b581df68d0a236e5743aa
thread:issue:78761,issue #78761: Extend tag testing for aliases,stale/low urgency,2,stale/low urgency,1,0,78761,,,,0,1563d2bc4aa04024642aadad88ba42630633b45e58eda6df37f2cc5e2ebd8ed4
thread:issue:78786,issue #78786: Deprecate hardtanh type promotion behavior.,stale/low urgency,2,stale/low urgency,1,0,78786,,,,0,dd17e55e39cdd8e96bc6f25ced643d25efc4ca194ef447727705ebc01852feff
thread:issue:78800,issue #78800: Efficiency of unary operations on CPU for large tensors,stale/low urgency,2,stale/low urgency,1,0,78800,,,,0,c7a1c9e36ffcf93bd8505bb30021512e8f354a386131da13f24c5a48bf81c88d
thread:issue:78808,issue #78808: Feature request: Integer system decompositions,stale/low urgency,2,stale/low urgency,1,0,78808,,,,0,13b4be5904c1619ca891cd0a3a68c78800f59b7af48d07d9e39e65d48339637d
thread:issue:78809,issue #78809: Feature Request: Hessenberg and Schur decompositions,stale/low urgency,2,stale/low urgency,1,0,78809,,,,0,769cd4de42f8dea493f5ee58b4be0bdd76e97850c2a2d32ca4e7fac79f61bd8b
thread:issue:78845,"issue #78845: [Modes] no_dispatch is not the same as DisableTorchFunction, causing differences in modes",stale/low urgency,2,stale/low urgency,1,0,78845,,,,0,29baad30d7f10b6c2718a88b8d25f3a5e5eb5c55dcefa1cb2ec146b9ccf3595d
thread:issue:78871,issue #78871: [RecordFunction] Hold a durable schema reference,stale/low urgency,2,stale/low urgency,1,0,78871,,,,0,d96d41df9cef8208d396aa418c80d7e69e6659a6b89028ed9206ef03dbd74535
thread:issue:78876,issue #78876: Remove const from function return type if returning const value,stale/low urgency,2,stale/low urgency,1,0,78876,,,,0,bce432e139873071ca098a7dfb4679690b1f58af0311f1f5291c4c350af86895
thread:issue:78880,issue #78880: fatal_signal_asan_no_sig_test in current master hang.,stale/low urgency,2,stale/low urgency,1,0,78880,,,,0,10d90054f69e366a6c053ee567524dd15ee820f85e623b7bfb206b8e005927be
thread:issue:78882,issue #78882: Cannot build master on AWS cluster: error: ‘__fatDeviceText’ was not declared in this scope,stale/low urgency,2,stale/low urgency,1,0,78882,,,,0,ca4dc48426492d3232ef2686ab4d5bd4d57ff3b6f0dbcf84b339cdcb5ed28633
thread:issue:78884,issue #78884: Abnormal GPU memory usage when using CUDA tensors with multiprocessing,stale/low urgency,2,stale/low urgency,1,0,78884,,,,0,99e0e22f314686411d173990013c0bf92d9541aaf8f2b0d8a8d2f4a8d6311439
thread:issue:7890,issue #7890: [feature request] batch_first of RNN hidden weight for Multi GPU training,stale/low urgency,2,stale/low urgency,1,0,7890,,,,0,051cfe6cb970bed45f599570b5f3f748bae874b2f6e5eae93ebadd9544abd2b7
thread:issue:78921,issue #78921: Do we really need sampler for IterableDataset?,stale/low urgency,2,stale/low urgency,1,0,78921,,,,0,c350e0cb718b9af44d83b1c5b6c091422d350728994ad49b52643058b03e015c
thread:issue:78929,issue #78929: Add type() support for mps backend,stale/low urgency,2,stale/low urgency,1,0,78929,,,,0,1cd4371a937093397d5f8b9c153b134ee219ec8d43067c81f52de282b8a98277
thread:issue:78954,issue #78954: Can we have Additive Attention?,stale/low urgency,2,stale/low urgency,1,0,78954,,,,0,89a47139e767461e8b09a8f58dff66d02d1d415c16064bf7b25bd6dd82a8b679
thread:issue:78961,issue #78961: [forwardAD] torch.no_grad has no effect under forward_ad,stale/low urgency,2,stale/low urgency,1,0,78961,,,,0,570507885c8cb6757ac9e84f0fa9ab1b3ef0a777baafc629ed44c89b03245e7f
thread:issue:7897,"issue #7897: [caffe2] build from source, cannot find my cudnn7.1",stale/low urgency,2,stale/low urgency,1,0,7897,,,,0,2ca44dc58a5cc45f46139b7b42343105c002193a27bd41a26ba66307d9c0a32a
thread:issue:78987,issue #78987: DataLoader leaking resources?,stale/low urgency,2,stale/low urgency,1,0,78987,,,,0,8e2985ef4e762b85b2088005b2f843299160f264d1d397f6a02355abc106c524
thread:issue:7900,issue #7900: Deprecate torch.Tensor,stale/low urgency,2,stale/low urgency,1,0,7900,,,,0,ed37210db1fddbd63959f6b99c7f7a1718d5a3f6fac102824368336f76fd0d6a
thread:issue:79004,issue #79004: Doc on index of CPU Device seems wrong,stale/low urgency,2,stale/low urgency,1,0,79004,,,,0,445574146f1940227e109c327a65cf73e9481d03fa6479e4f101965a8466e4b6
thread:issue:79013,issue #79013: Multi30k can't be downloaded the destination domain can't be reached,stale/low urgency,2,stale/low urgency,1,0,79013,,,,0,76fea2223d64693b745916b5a13c410c8f89a3a7cf9d22f53a52b5146b9a7ad9
thread:issue:79016,issue #79016: Corner cases of ShardedTensor checkpoint when using TorchRec,stale/low urgency,2,stale/low urgency,1,0,79016,,,,0,f02fa877e11f491df0aa3f5d4dde79b55d2195d9a9d6acf2c93942b1cbe58620
thread:issue:79018,issue #79018: Random functions should infer device from user-specified Generator,stale/low urgency,2,stale/low urgency,1,0,79018,,,,0,64079c5be09e61813ef26405f4bd4848c12e49d92173921162169d30e1921ed5
thread:issue:79020,"issue #79020: When setting sizes and strides on a tensor subclass in `THPVariable_make_wrapper_subclass`, also make offset symbolic",stale/low urgency,2,stale/low urgency,1,0,79020,,,,0,c3b79c69045f59f7383366dade1f44d490a553732bc1eaa15ad047b07b0d4589
thread:issue:79021,issue #79021: Building PyTorch from Source with BUILD_LAZY_TS_BACKEND_ON,stale/low urgency,2,stale/low urgency,1,0,79021,,,,0,8971255112a52c1b156ff83d7e5877e35a7612c13e38dfecd76d86ac5bef7f39
thread:issue:79030,issue #79030: Devirtualize `sym_sizes`. It still has to work for python tensor subclasses and LTC/Xla,stale/low urgency,2,stale/low urgency,1,0,79030,,,,0,75add6932c41e3a531360259eadf8618e9a601e33b16fbd8a1e72a70cdc78eba
thread:issue:79031,issue #79031: Implement SymbolicIntNode interface for lazy (i.e. lazy::SymbolicIntNode),stale/low urgency,2,stale/low urgency,1,0,79031,,,,0,1faf3efb2dc7e517c62e58e9c1b0b92e161172ea9fe39e8498b327f579b46241
thread:issue:79032,issue #79032: Add a test that shows that lazy_ir reuse breaks SizeNodes,stale/low urgency,2,stale/low urgency,1,0,79032,,,,0,2ad966433f5482dd8d79c239c0ca723a8fe3a3277ea10e3d4697285ec1a3b891
thread:issue:79050,issue #79050: [MetaIssue] Investigate if we should be reusing primtorch formulas for `is_dynamic`,stale/low urgency,2,stale/low urgency,1,0,79050,,,,0,d0227f97f581927d344f0302e2b326730896e2332bc42eeb65a4396ddf794398
thread:issue:79052,issue #79052: Mirror and implement `SymbolicIntNode` API for `SymInt` so we can trace in C++,stale/low urgency,2,stale/low urgency,1,0,79052,,,,0,033d1958959556655fc9a8462bb042d689de5e3e3fd651ab05087f216927f4e5
thread:issue:79053,issue #79053: [MetaIssue] Propagating SymInts through Autograd,stale/low urgency,2,stale/low urgency,1,0,79053,,,,0,f09753956b8a8af81374c77d4349f3d1b9ebaf66e483e914be34fd97b6b5475a
thread:issue:79073,"issue #79073: Multi-node, Multi-GPU set up tutorial for Slurm cluster",stale/low urgency,2,stale/low urgency,1,0,79073,,,,0,b1eb660d3df0814a529f7f6f231301039a1fecb50a02245cf4e98dc1fd12fa6b
thread:issue:79083,issue #79083: CI workflow creates too many tags in RSS feed,stale/low urgency,2,stale/low urgency,1,0,79083,,,,0,14c7f6f60180cd4ca15325b32e5ee8981f7bce93e134e87f6ee15f451e38df82
thread:issue:79089,issue #79089: Improve PrimTorch testing for view consistency.,stale/low urgency,2,stale/low urgency,1,0,79089,,,,0,053b3c6bbf6240c7eda9a274e5d4481d9dee51dd809337b3f885a9fec43071ca
thread:issue:79091,issue #79091: Error with Named Tensors and multiple threads,stale/low urgency,2,stale/low urgency,1,0,79091,,,,0,fa8c07465e26f676f254b8ec53754f564cf2af64383bd36fb74b95844c7d688d
thread:issue:79118,issue #79118: About the source code of,stale/low urgency,2,stale/low urgency,1,0,79118,,,,0,8cc5553828f57f5b88a37af5617b404472938ff41d9ae3c98499c087dd4167ff
thread:issue:7912,issue #7912: [Caffe2] Operators of Detectron module not registered/compiled when built on windows,stale/low urgency,2,stale/low urgency,1,0,7912,,,,0,364e8d58f3b0532fe543d8c2364812baf1560403aa7526a508c3c79b94b90c97
thread:issue:79120,issue #79120: Hide or fuse TupleConstruct / TupleUnpack from tensorboard graph,stale/low urgency,2,stale/low urgency,1,0,79120,,,,0,05fc54425e6408ebcc426709025bf6201df5ef430df69a64a96f4cbf64540eed
thread:issue:79130,issue #79130: torchvision.models.mobilenetv3 can't save pre-trained model to custom dir?,stale/low urgency,2,stale/low urgency,1,0,79130,,,,0,6358cbbd58001a53b0920557cee68d1f85881934396cbba0caaff69666de9539
thread:issue:79137,issue #79137: [AUTOGRAD] codegen to use sym_sizes for ops w/ symint overloads in derivative formulas,stale/low urgency,2,stale/low urgency,1,0,79137,,,,0,b335bcaf778b65c4ab189089b64de520a31847b6cac52549d4e5841c89e3cc37
thread:issue:79138,issue #79138: [AUTOGRAD] support implicit reductions with SymInts in autograd.,stale/low urgency,2,stale/low urgency,1,0,79138,,,,0,ad67da20d2a1d1a50160acc86d359651482e7118394c6948c333de16f2c08e76
thread:issue:79145,issue #79145: Redundant info are saved when using torch.save to save part of torch.tensor,stale/low urgency,2,stale/low urgency,1,0,79145,,,,0,acd1432cf7985df9f3438e431c74c5df5d02bb3c2c0f71e9ed652c135b55e633
thread:issue:79164,issue #79164: PyTorch/XLA's DDP XLABackend is broken by upstream change,stale/low urgency,2,stale/low urgency,1,0,79164,,,,0,aa25c5924191081b6dd933c3e3ccb59b9cfa99f5ad485a06157d36e74b8e1afa
thread:issue:79171,issue #79171: DistributedDataParallel `static_graph=True` fails to handle unused parameters,stale/low urgency,2,stale/low urgency,1,0,79171,,,,0,59be464c20a2acf396e922ede3c83b89b23bd5d26f48596780cfe67298946847
thread:issue:79177,issue #79177: _make_elementwise_unary_reference and other function factories in torch._refs don't set __name__ correctly,stale/low urgency,2,stale/low urgency,1,0,79177,,,,0,d2c2a27a552f247924ab8df1f0c87a4856c51106c8bad77155601625cf2afd09
thread:issue:79195,"issue #79195: [bug] fill_, masked_fill_ : fill ops allow lossy downcasting of fill value",stale/low urgency,2,stale/low urgency,1,0,79195,,,,0,2018a8d70a544925289c260d10501cd4f02846d298a96de84faa78199a4e4e4a
thread:issue:79202,issue #79202: [bug] Device dispatcher can choose CPU path for CUDA tensors.,stale/low urgency,2,stale/low urgency,1,0,79202,,,,0,9ca91f475698ee1e52ea60d212ed66794c2f4b19fe682cfc64ee08c73af19e29
thread:issue:79222,issue #79222: scripted fft Convolutions are faster than nn.Conv1d with large kernels,stale/low urgency,2,stale/low urgency,1,0,79222,,,,0,f10e78c5226d0d808988cb8b6b2f8191d66e07bcc9348237f91bbe910e269c48
thread:issue:79244,issue #79244: [NVFuser] bad performance on pyhpc_equation_of_state,stale/low urgency,2,stale/low urgency,1,0,79244,,,,0,1128782cd95bc6a3fd5137170d210cea27cb6f0dcca2685c6b30a4f83fdc8b36
thread:issue:79246,issue #79246: [NVFuser] bad performance on mobilenet_v2 and mobilenet_v3_large,stale/low urgency,2,stale/low urgency,1,0,79246,,,,0,9eb7459a6579989f93b9a2d8c7e23956b4cbb0e2f73c16a33b70af57931bad0e
thread:issue:79261,issue #79261: [NVFuser] bad performance on pyhpc_isoneutral_mixing,stale/low urgency,2,stale/low urgency,1,0,79261,,,,0,02b4b92097ac8b209332fc23da371555b25dbaad346b20e1518a55e9e77ae3f6
thread:issue:79272,"issue #79272: PyTorch leaks a macro definition called ""CHECK"" in the C++ version",stale/low urgency,2,stale/low urgency,1,0,79272,,,,0,8ef91d5878bcdc286f8c91c73223379e5f922860eb61ffa0bf52658db96aef9b
thread:issue:79299,issue #79299: PyTorch gets stuck when using an NVLink/A6000 and more than two GPUs,stale/low urgency,2,stale/low urgency,1,0,79299,,,,0,7e6dfbc8cc42fc3ff453aad72a95976979683104566a4781a6b3b809f18a5893
thread:issue:79307,issue #79307: `prepare_qat_fx` docstring doesn't run,stale/low urgency,2,stale/low urgency,1,0,79307,,,,0,0cc2a0129f89fb7faa35bcabeff587e9b6ce3251889bed3a3058be73d56f424f
thread:issue:79325,issue #79325: [NVFuser] hitting fallbacks on demucs (from torchbench + lazy tensor),stale/low urgency,2,stale/low urgency,1,0,79325,,,,0,13c051199c95044754a8f1928709ec2d9c264022600b6199622601ba9a9ebf7d
thread:issue:79336,issue #79336: Batches are being duplicated from go http call,stale/low urgency,2,stale/low urgency,1,0,79336,,,,0,f16242e2f424ab4131c5f12c0f4981ead4ae91ae474fd1ca1fe7b2d0bf64075d
thread:issue:79349,issue #79349: PyTorch get positive log_prob of a multivariate normal distribution,stale/low urgency,2,stale/low urgency,1,0,79349,,,,0,4f2347b25e7d10dbbe51120e96cebc9db77133fd3a4a5f34ab82d0d35bf22f52
thread:issue:79351,issue #79351: [bazel] ability to run gpu tests on gpu machines in RBE,stale/low urgency,2,stale/low urgency,1,0,79351,,,,0,0b0e732c095f768a2ecd8a1985ef73dbfc65ae251bb77d215b898740ac08ac7b
thread:issue:79355,issue #79355: [bazel] build spams warnings,stale/low urgency,2,stale/low urgency,1,0,79355,,,,0,9d17d5e09c094949747dad47f6778017fd4c37780e25f3a759013b497f681b12
thread:issue:79359,issue #79359: [build] No documented way to install C++ binaries for pure-python development of pytorch,stale/low urgency,2,stale/low urgency,1,0,79359,,,,0,ed81901ce357231db9cb7aeea9e0fa5107588ec539202cc5f2fdb7e68451e059
thread:issue:79382,issue #79382: torch failure to open libcuda.so.1 on macOS,stale/low urgency,2,stale/low urgency,1,0,79382,,,,0,03fce4eea46de10f88f89b0e5b811fad35741d89d3ab13d4f2ebe44434947a24
thread:issue:79387,issue #79387: caffe2_nvrtc is produced even when it won't be used,stale/low urgency,2,stale/low urgency,1,0,79387,,,,0,984694b3c5141241422137b659a335faec730c309695454e2abea7be173aa811
thread:issue:79395,issue #79395: SymInt equality tests are unsound,stale/low urgency,2,stale/low urgency,1,0,79395,,,,0,fcee34f0df7f019c143612d32986f385ef85866ebf9884dca8ba576c981dbe20
thread:issue:79407,issue #79407: Conda enviroment,stale/low urgency,2,stale/low urgency,1,0,79407,,,,0,622637957557078b0182384f956d2a29b2c2088514ad858f4b1c017ffee26c97
thread:issue:79421,issue #79421: Mismatch in clang toolchain lead to binary incompatibilities on M1 between torch and torchvision,stale/low urgency,2,stale/low urgency,1,0,79421,,,,0,14b614186a3bfdd44b11e5c354d17f3d668f5b0d3a021886e7a154f39fa89904
thread:issue:79425,issue #79425: API for accessing SymIntNode mandates refcount bump even when it is unnecessary,stale/low urgency,2,stale/low urgency,1,0,79425,,,,0,19fa3d65e103dff7c932c0f3d0212e1af4b17e50076f9258fa10d93e1b88529b
thread:issue:7944,issue #7944: Better error message in DataChannelTCP::_receive,stale/low urgency,2,stale/low urgency,1,0,7944,,,,0,dd520ea2044bd30bb3eab566c49c953206057cffa1a8a8435ceaa4ad13ac060e
thread:issue:7945,issue #7945: [Caffe2]How to convert Caffe's mean file to Caffe2's ?,stale/low urgency,2,stale/low urgency,1,0,7945,,,,0,04f14bec514c2509bdff47283e465154e95819d86d631aaa6685af654299e789
thread:issue:79453,issue #79453: [LTC] Make `torch::lazy::BackendImplInterface::ExecuteComputation` takes `ComputationPtr` instead of `Computation`,stale/low urgency,2,stale/low urgency,1,0,79453,,,,0,957662b961618f638daf587bfe2927300e9cc0596552dcfd7a95a8fa67365759
thread:issue:7946,issue #7946: how to use Softmax when do segmentation,stale/low urgency,2,stale/low urgency,1,0,7946,,,,0,ea8a9345ed2657a58909ae1da97eecb7c55fda158c9b45f1d47932d63c553294
thread:issue:79464,issue #79464: Iteration # 1-offset in DDP logging,stale/low urgency,2,stale/low urgency,1,0,79464,,,,0,3748a803d067ce7960ef1ec92b0949aa5900caf18b98bdf562b5696e2c5461b6
thread:issue:79467,issue #79467: Add autograd support for dispatch passable c10d ops,stale/low urgency,2,stale/low urgency,1,0,79467,,,,0,4d93fc9e783354c1101daf986a3ee9a203165d880841a5da1ead58f52dd32ff2
thread:issue:79468,issue #79468: Ensure the guards in distributed_c10d.py wrappers get executed in the replay of the graph,stale/low urgency,2,stale/low urgency,1,0,79468,,,,0,0f089e0b7f1ee956bf76f58ae32a2d3c11b1bb5530b6f868846a3cab55bd3ced
thread:issue:79469,issue #79469: Add a new _broadcast_coalesced op for DDP,stale/low urgency,2,stale/low urgency,1,0,79469,,,,0,8599581f0686c008ac3f8d6c358cf29406322f7fc34460b1813e3d92725561f1
thread:issue:79518,"issue #79518: Display a ""reference"" link for ops that points to primTorch implementations",stale/low urgency,2,stale/low urgency,1,0,79518,,,,0,1bf68480c312469ae8e52d0d823816175723f1e0ee1693025e15323e05557ef0
thread:issue:79542,issue #79542: Unable to use a parameter with torch.sparse_coo layout with DDP,stale/low urgency,2,stale/low urgency,1,0,79542,,,,0,81f59059ca2594549f7c9ad1ea94c6266723203dddb43c6c69505e4b6d530a64
thread:issue:79563,issue #79563: __torch__dispatch does not return new output in inplace function,stale/low urgency,2,stale/low urgency,1,0,79563,,,,0,f2902499d1a6a549baba531d7834914a8b84eef412603bb2c20df12294f5517f
thread:issue:79592,issue #79592: [LTC] Introduce a `MetricsReport` python binding and allow backend to add their report as string,stale/low urgency,2,stale/low urgency,1,0,79592,,,,0,a0d7a9e9d066f12e6ecd32ec5ae9aa240d180e5f51fbb59c0986ed55eea74db0
thread:issue:79604,issue #79604: Some unit tests are failing,stale/low urgency,2,stale/low urgency,1,0,79604,,,,0,f04be3601e3f2bf64f8f33c56746427d842333db1de24b319ae7f8a8b49d8ac0
thread:issue:79606,issue #79606: PyTorch Preview (Nightly) version number does not comply with Conda conventions,stale/low urgency,2,stale/low urgency,1,0,79606,,,,0,5c0dffc0df439b2de2cbf10db5b85c03daebc3facbe4352e03f968bbeb48f318
thread:issue:79629,issue #79629: Missing the time unit in duration time of DDP logging,stale/low urgency,2,stale/low urgency,1,0,79629,,,,0,bd44073e7fdc735b7c80362fe155a0fbbf1df62b5e9f24130cc86c2d0af4b33b
thread:issue:79647,issue #79647: tensorboard SummaryWriter.add_graph fails when model uses empty tuples,stale/low urgency,2,stale/low urgency,1,0,79647,,,,0,93a9d479c1999548e02181d2a94ecb5c559ecdbde3828a53156d81eb2e9696d9
thread:issue:79705,issue #79705: gradgradcheck fails for torch.native_layer_norm,stale/low urgency,2,stale/low urgency,1,0,79705,,,,0,3c1e05f2b2afa6cb5537f8da9d4bba58a3682a122a793dfa5c942587d9625400
thread:issue:79709,issue #79709: ONEDNN testing is not done properly in quantization codebase,stale/low urgency,2,stale/low urgency,1,0,79709,,,,0,733f25420d602443e702ce5a25cc7e4970a4309bbb91a88a376bfce338d6c315
thread:issue:79739,issue #79739: quantization: misleading backend config for linear_dynamic_fp16,stale/low urgency,2,stale/low urgency,1,0,79739,,,,0,25b0844b70dac5c6250a2c8c344c3ab2e169f1927a7c75de54d38d7d4ae5a51a
thread:issue:79787,issue #79787: Improve clarity by making sharding a static nightly update,stale/low urgency,2,stale/low urgency,1,0,79787,,,,0,402b43135e176950dabe89d1a2ab7e6f0515fc18083cd9bca01cc26ff78fa779
thread:issue:79788,issue #79788: Parameter.__deepcopy__ doesn't preserve view relationships,stale/low urgency,2,stale/low urgency,1,0,79788,,,,0,a5a8f9466a9079551efff80b29234ced29f2877302f79ef010f8276ce8d9cb81
thread:issue:79847,issue #79847: Multi-node training meets unknown error,stale/low urgency,2,stale/low urgency,1,0,79847,,,,0,59ebb561c6738a65f35801ba4783d0c915ea4a592692817150e768e7409d1bc7
thread:issue:79848,issue #79848: Automatically calculate output_shape of sequential model (or any other fCNN),stale/low urgency,2,stale/low urgency,1,0,79848,,,,0,3fbba966fe31e15336d4d65607afde9bca395a9e8e8467b78db0c00452159e2c
thread:issue:79851,issue #79851: [bazel] [ci] `//:module_test` CUDA error: CUDA driver version is insufficient for CUDA runtime version,stale/low urgency,2,stale/low urgency,1,0,79851,,,,0,d9e50495dc853d4cf658d09bd5b8851e9f29871ff10615e70aa4138fb0b5aeeb
thread:issue:79853,issue #79853: [bazel] [ci] `//:lazy_tests` Could not run 'aten::mul.Tensor' with arguments from the 'Lazy' backend,stale/low urgency,2,stale/low urgency,1,0,79853,,,,0,14cbd61b1efb0a785edd600b151ac8ebdef55426ebdf2c71db54f7d811543e08
thread:issue:7986,issue #7986: [Caffe2] place some nodes of the same network on GPU and others on CPU?,stale/low urgency,2,stale/low urgency,1,0,7986,,,,0,e575644713f420f88b6044f67514744f90cbc2269a38a0a0d80d66fdfdd046a6
thread:issue:79867,issue #79867: All {view}_scatter variants should support all (or most) dtypes,stale/low urgency,2,stale/low urgency,1,0,79867,,,,0,f2b986510c56cf8cb26efbbc7da25e3bb5a0cc8f3be6d3fab5bfb16a3e2f92c2
thread:issue:79893,issue #79893: Write lint for isGreen,stale/low urgency,2,stale/low urgency,1,0,79893,,,,0,044c08cef0830e49722ba364563ee70980b131256f74ebb9d47b2b25205dafdd
thread:issue:79895,issue #79895: Modify update-viable-strict GHA to use internal version of checkout,stale/low urgency,2,stale/low urgency,1,0,79895,,,,0,248a000e003d911686982682c3653ee4654ecde9955c65bc4280fc2885dfc9aa
thread:issue:79977,issue #79977: [CI] Do we run all cpp tests on CI?,stale/low urgency,2,stale/low urgency,1,0,79977,,,,0,d2fbc9392b1aae2f1a60176b1c06efdc003495135fab9bcb9e9afe08cecca690
thread:issue:79997,issue #79997: Comprehensive documentation for Tensor indexing?,stale/low urgency,2,stale/low urgency,1,0,79997,,,,0,945bfef92138657b516eaff51f89b850da139455701f4ab5b9f833edef33a065
thread:issue:80022,issue #80022: memory leaking when doing all_to_all_single communication,stale/low urgency,2,stale/low urgency,1,0,80022,,,,0,69602a4b953ec23f7fc6a55c1d36470277afb36621371dea0d929b054862e7f3
thread:issue:80061,issue #80061: pow CUDA tensor raised to CPU scalar tensor result can't backward properly,stale/low urgency,2,stale/low urgency,1,0,80061,,,,0,f1fe5982a263a7838920d00b269f8f629ec237f7460022e51e53bf1ed0b83c38
thread:issue:80067,"issue #80067: FakeTensor: Support torch.tensor([FakeTensor, 0])",stale/low urgency,2,stale/low urgency,1,0,80067,,,,0,6b8497ce69328b6add4a5bc2ba6d7ef0e275908e3ecd7fe99b58f8be50be701f
thread:issue:80080,issue #80080: Complex support in DDP,stale/low urgency,2,stale/low urgency,1,0,80080,,,,0,c6d141348c9b0cf2226c3536b25c81bd8ffbbd876a6f20ec92f140a753c34841
thread:issue:80118,"issue #80118: Allow a user provided ""test name - test time"" mapping file work with pytorch's test sharding mechanism",stale/low urgency,2,stale/low urgency,1,0,80118,,,,0,074e62a74078a3acab20c2364e36e1f21bee30bf0b50fd695e4265222f9b8de8
thread:issue:8012,issue #8012: [Caffe2] VideoInput/LMDB Reader to Standalone Predictor Question,stale/low urgency,2,stale/low urgency,1,0,8012,,,,0,0785ede14f040470efed8b982bc307a70afc809b95e6fc5cd71df1173b100663
thread:issue:80132,issue #80132: Expose more MAGMA backends for solve_triangular,stale/low urgency,2,stale/low urgency,1,0,80132,,,,0,afe488b30756c232935e8b1b404dfc2604c27687ec22a2fb11ebbc0fbe658cd9
thread:issue:80134,issue #80134: [feature request] Add support for a custom DatasetFetcher in DataLoader,stale/low urgency,2,stale/low urgency,1,0,80134,,,,0,ebc3a99dcdad7920d9c31fa173e876b35ba79bd5fcb0e38792c19844192d8cd7
thread:issue:80154,issue #80154: [primTorch] No _refs support for torch.Tensor.requires_grad.__get__,stale/low urgency,2,stale/low urgency,1,0,80154,,,,0,d638ab74691194d01510141e5d6a67ca9914a893d981a00bf607e64588a85cc3
thread:issue:80167,issue #80167: AMP step() enforce synchronization,stale/low urgency,2,stale/low urgency,1,0,80167,,,,0,817650397103e78003a4187b5f4b093ee74c360bc00acf11e6c30fe4e32bfd9b
thread:issue:80172,issue #80172: Torch.fx: add reporting of the name of a module not found during tracing,stale/low urgency,2,stale/low urgency,1,0,80172,,,,0,cbc1d409d2c4cb7ad5b6e545a2008564f2fdb62a04ad525fdf7c38aa0a390210
thread:issue:80189,issue #80189: [NVFuser] Investigate models without any fusion groups found,stale/low urgency,2,stale/low urgency,1,0,80189,,,,0,e974175cf0e9ebada8ad1b5d402a01fb188cadba09ca67d9f1cfc90a2b284bc4
thread:issue:80204,issue #80204: `max_unpool` gives wrong gradient when `indices` has duplicate,stale/low urgency,2,stale/low urgency,1,0,80204,,,,0,280aaf1cdb52babcaf8d965420efc8707a90ef209e3ac51fdc3087e5454f86d7
thread:issue:80206,issue #80206: `soft_margin_loss` gives wrong gradient when `target` with dtype uint8,stale/low urgency,2,stale/low urgency,1,0,80206,,,,0,25ff8413f48d16baaf6fc7a29c726ca71ebf9b75dcfff18fdcff5e8a559818a9
thread:issue:80221,issue #80221: OpInfos for torch.ops.aten operations,stale/low urgency,2,stale/low urgency,1,0,80221,,,,0,f3f2533fc222b340207f3ca1df3ce1c525e1d0d54b43313dc0358937846d5ce9
thread:issue:80226,issue #80226: Support tensor subclasses as `UninitializedParameter`s,stale/low urgency,2,stale/low urgency,1,0,80226,,,,0,0bb0d948aa70307565d39788653648771a2b7fc4ec0f1516a5b9b6170cc652d7
thread:issue:80230,issue #80230: [Prims+NVFuser] Aten2Prim refs tracking items,stale/low urgency,2,stale/low urgency,1,0,80230,,,,0,ccf5f0d8606c5383c18a157fa72b9be35165251638857fd4e3ed864761c30b97
thread:issue:80231,issue #80231: [Prims+NVFuser] Supports 0-sized inputs,stale/low urgency,2,stale/low urgency,1,0,80231,,,,0,8cf8db35167fecde59d1f937fe848b0d1bedff7e30fd36eac7a3292ebc5362ca
thread:issue:80238,issue #80238: test_conv_backend tests OOMing in 10.2 slow_gradcheck CI,stale/low urgency,2,stale/low urgency,1,0,80238,,,,0,ddd5391382b935bb21f7fce4dd0c3bc138b1c2adf69308e4d8b955c244da08be
thread:issue:80241,issue #80241: Change c10d APIs in ProcessGroup to accept const std::vector<at::Tensor>&,stale/low urgency,2,stale/low urgency,1,0,80241,,,,0,24aa8da7583ddd67f08654014c03397736e655490382ea3e82281c0dca52178c
thread:issue:80242,issue #80242: Write some torch.distributed.nn.* tests for the new dispatcher passable ops,stale/low urgency,2,stale/low urgency,1,0,80242,,,,0,cedbb7df3428393bc6c959bf9ea43c5421445dd8c2af54c8774bc9e7a80dc305
thread:issue:80256,issue #80256: [complex] dropout and it's variants should support complex tensors,stale/low urgency,2,stale/low urgency,1,0,80256,,,,0,64f34741f138cdc86bb7fc0a88f8ee27e31ea053fdf605412509a24dd4245891
thread:issue:80259,issue #80259: Cudnn batch norm kernel (batchnorm_bwtr_nhwc_semiPersist) gets blocked by overlapping NCCL all_reduce calls,stale/low urgency,2,stale/low urgency,1,0,80259,,,,0,56db97059cb6f27b8397a6e7a29fd6d94d0fde32a2c5bf51583c6de4f4ac0aba
thread:issue:8028,issue #8028: Build error : mpi/mpi_gpu_test.cc.o: undefined reference to symbol '_ZN3MPI8Datatype4FreeEv',stale/low urgency,2,stale/low urgency,1,0,8028,,,,0,0865f8d108aec8a4efb1b1a04b6be2ab72dea68f4e9639dedb79d014acb94f53
thread:issue:80296,issue #80296: Sort tensors inplace,stale/low urgency,2,stale/low urgency,1,0,80296,,,,0,e9a663ca6fe160fd0f41e4f4d6138608e15b34611eefbce06acb58256f0ad359
thread:issue:8035,issue #8035: [PyTorch] EmbeddingBag comparison vs Embedding fails w/ small max_norm on CUDA,stale/low urgency,2,stale/low urgency,1,0,8035,,,,0,aeb0c529bb6076a6f2af80fcb6440537e32361a55209d293e68d81e3f7874e16
thread:issue:80372,issue #80372: [BE] Update ProcessGroupWrapper tests to test other collective message,stale/low urgency,2,stale/low urgency,1,0,80372,,,,0,49098f25879f0f28a23f190e268c616a0658940de6cdd184f07e9cc67ba02796
thread:issue:80377,issue #80377: Modify _add_docstr to also set the correct module for the APIs,stale/low urgency,2,stale/low urgency,1,0,80377,,,,0,bf2d46cc45ba224d669ac0c35a5c790c286dfd1b77d66d1d850e43d9fcc69b90
thread:issue:80411,issue #80411: Tracker: Slow gradcheck failures possibly indicating incorrect gradients,stale/low urgency,2,stale/low urgency,1,0,80411,,,,0,1e98fb02072be2d74976a74c5c704c6008206067bc734f9150af75e57d8a0464
thread:issue:80417,issue #80417: [c10d] Async object-based collectives,stale/low urgency,2,stale/low urgency,1,0,80417,,,,0,31d6b424dbadb6faf0139eb3c8b89380ee07dad0176c4724344fd82c63033d25
thread:issue:80420,issue #80420: [DDP] output_device argument appears completely unused,stale/low urgency,2,stale/low urgency,1,0,80420,,,,0,47f9fcd646a5c5928b264b946410619c543e2ef6379b5ef6ad41cbcb5ad44b2b
thread:issue:80427,issue #80427: Torch fx print line number of each node,stale/low urgency,2,stale/low urgency,1,0,80427,,,,0,87544d5ab25697d2c32084db504a932f0cc06ae526522bd35923bbba488f818a
thread:issue:8044,issue #8044: [PyTorch] Windows CI CUDA mem leak check on BN tests are flaky,stale/low urgency,2,stale/low urgency,1,0,8044,,,,0,5d66769d7e82f8af4734e104023b8d3118b48ac84926c39c6b707564d7557e00
thread:issue:80458,issue #80458: Revisit OpInfo samples for nn.functional.max_poolNd,stale/low urgency,2,stale/low urgency,1,0,80458,,,,0,cfd68802c003063fa7ce031a982e7c82211b8c351318edf2a25b0cbaf9be2061
thread:issue:80494,issue #80494: [bug] libtorch bug in nn::MultiheadAttention and nn::Transformer,stale/low urgency,2,stale/low urgency,1,0,80494,,,,0,f531bb141202cdc756c8c949e8cd6ea606a3a7469050a4c5085ac3861d5f9db4
thread:issue:80541,issue #80541: [Prims+NVFuser] Prims with missing NVFuser ops,stale/low urgency,2,stale/low urgency,1,0,80541,,,,0,e2ebb407e375205ce6b128cc1ec54a7f9bf9a86bf0f334f92f65bd12e55dff6a
thread:issue:80551,issue #80551: NVFuser should extend caching to remove necessity for PrimTorch's executor to Provide Tensor Contiguity Info,stale/low urgency,2,stale/low urgency,1,0,80551,,,,0,e2fa6cee358429fd79afef23c1073fc092b0b22333c1a3edfcfeed34817efbe1
thread:issue:80574,issue #80574: `bmm_sparse_cuda` kernel for `bfloat16`,stale/low urgency,2,stale/low urgency,1,0,80574,,,,0,73377304a438d2b6a0356b53a604abbd0340515acd75a174fd3976594e937a92
thread:issue:80577,issue #80577: 2-dimensional arange,stale/low urgency,2,stale/low urgency,1,0,80577,,,,0,2e997206261759c424f487138a6eda5b6ab15ee0d1b6fc9aa5c6b15df6bf5b8a
thread:issue:80580,issue #80580: `torch.ops.aten.find` inconsistent with `str.find`,stale/low urgency,2,stale/low urgency,1,0,80580,,,,0,5669cff3a1703208ea677d542030d2d8f7241b900c8fbff1bdc3317a49b53ca5
thread:issue:80594,issue #80594: RuntimeError: DataLoader worker (pid 22822) is killed by signal: Aborted.,stale/low urgency,2,stale/low urgency,1,0,80594,,,,0,a586b392e4a49f5c15240fab67072006086174904bf50ba96a8a656292b9caeb
thread:issue:80605,issue #80605: pad_sequence and pack_sequence should support length zero tensors,stale/low urgency,2,stale/low urgency,1,0,80605,,,,0,709275dd6f990c342d0af1a1173b12fb89e88e4fdebc471dadbee8a423e99d2a
thread:issue:8062,issue #8062: CrossEntropyLoss mishandles weights,stale/low urgency,2,stale/low urgency,1,0,8062,,,,0,b162b4ea92d66bf304df2e1224fae47b61a639acc874b66543ae4ba5418094af
thread:issue:8069,issue #8069: Document torch.acos() behavior near -1 and 1,stale/low urgency,2,stale/low urgency,1,0,8069,,,,0,8d1bda2b3d3a23707a458202ff93c63e969017ccdf2dc1b38ed60bd29adcc53e
thread:issue:80738,issue #80738: Output for `aten::_native_multi_head_attention` appears inconsistent with entry in `native_functions.yaml`,stale/low urgency,2,stale/low urgency,1,0,80738,,,,0,64f407d03ae517b4b1a7eb495202cd908d7d74a790e908df2c2895f8eb139365
thread:issue:80742,issue #80742: Add TorchDynamo as a submodule to Pytorch?,stale/low urgency,2,stale/low urgency,1,0,80742,,,,0,765365f66aca0461941bad48bb45683b28a01053b5dbc3be43a63d8fe8aa5f46
thread:issue:80762,issue #80762: `atan2` will gradcheck fail when `other` is a tensor with `int8` dtype,stale/low urgency,2,stale/low urgency,1,0,80762,,,,0,3d9647ba436ea2f60518936e90b104ca164258560dc261abc82f695193fd1436
thread:issue:80765,issue #80765: Let torch.utils.tensorboard support multiprocessing,stale/low urgency,2,stale/low urgency,1,0,80765,,,,0,e3fad60825786f5255b3068be70eff9c4baa554f176e36cf82f3032f09d3fe1c
thread:issue:80771,issue #80771: Anaconda is not a package manager,stale/low urgency,2,stale/low urgency,1,0,80771,,,,0,ee62e5cf3a8471e601bc8c96c0f370fcfb4537c9f94125d9ca3183653fdeb555
thread:issue:80776,issue #80776: `torch.inverse()` crash in cuda,stale/low urgency,2,stale/low urgency,1,0,80776,,,,0,74dfcab04771510ce7e4e927ceea4d5d16378ba4b686464474ff153a024a449e
thread:issue:80803,issue #80803: `hardshrink` gives wrong gradient for 0 input when `lambd` is 0.,stale/low urgency,2,stale/low urgency,1,0,80803,,,,0,ae3e91db17bd552bd5dda6d5a197639e7642b668f2bb2ec95df802b79b1c2277
thread:issue:80804,issue #80804: `torch.renorm` gives wrong gradient for 0-valued input when `p` is even and `maxnorm=0`.,stale/low urgency,2,stale/low urgency,1,0,80804,,,,0,000d5fb55fb8cc2981e8f3cecf50b8d4cd50c95bffe2204e4fa9af67f6b4edf6
thread:issue:80808,"issue #80808: The result of doing a dot product between two vectors, using einsum, depends on another unrelated vector",stale/low urgency,2,stale/low urgency,1,0,80808,,,,0,df7b85176a7dcd03c745d7ce895bae6b2a5990531308dc99daedac7b36aa7f96
thread:issue:80827,issue #80827: Inconsistent computation of gradient in MaxUnPooling,stale/low urgency,2,stale/low urgency,1,0,80827,,,,0,b7a3dbda667429131298a61864fd69cb3d46d830cea20a1b550d60697398d695
thread:issue:80851,issue #80851: [bug][nvfuser] Applying nvfuser to the model leads to runtime error,stale/low urgency,2,stale/low urgency,1,0,80851,,,,0,8c58de3f1565e6d80e8d5124eb256b29e5d4f6633818a8cdae0c86237b6d91b8
thread:issue:80857,issue #80857: Compatibility List,stale/low urgency,2,stale/low urgency,1,0,80857,,,,0,4186561e4c7b9bfc7d2672f058a523861dcf9915a4b0c14186067fc3672a0081
thread:issue:80874,issue #80874: C++ extensions inject a bunch of compilation flags,stale/low urgency,2,stale/low urgency,1,0,80874,,,,0,b3dfe6968a5ee859f8f1dbea4c2310eb4d27c782a9840568ca467a042cb52991
thread:issue:80903,issue #80903: make_fx doesn't work with truly dynamic argument functions (e.g. fx.Interpreter),stale/low urgency,2,stale/low urgency,1,0,80903,,,,0,b61afa37825806b70625abdc4a61b5f80ec311be4ec1b0b501f2a6087e18b5c1
thread:issue:80921,issue #80921: [primTorch] `|` operator does not work with FakeTensor in _refs,stale/low urgency,2,stale/low urgency,1,0,80921,,,,0,2b3932de1ceabb5654168a4b85628cac72aa9350e495b0afa30b29d3f3b1221d
thread:issue:80929,"issue #80929: Interpolation artifacts when using nn.interpolate, trilinear mode for 3D label images",stale/low urgency,2,stale/low urgency,1,0,80929,,,,0,5c73c9dbbb9ea306f2755d4ce143d44739666a1fd2a14176654cd96a6e75f4a7
thread:issue:80932,issue #80932: Position embedding aware global circular convolution,stale/low urgency,2,stale/low urgency,1,0,80932,,,,0,770d27579ba4bb1ada509ea08551b488dacce9b52354f0ce2129cbf4078fb7ba
thread:issue:80940,issue #80940: Wrong example of sliced computation in doc page Numerical Accuracy,stale/low urgency,2,stale/low urgency,1,0,80940,,,,0,e8bbb50f01886bc1b546320498fa7c2fc54aeea4bbbc305b7a410251b0ac2bd2
thread:issue:80942,issue #80942: [LTC] OOM on mnist example,stale/low urgency,2,stale/low urgency,1,0,80942,,,,0,9c874e54242386486da701efa041d525de5d66ad5e54e98ace9dd9ba8e1ed1a0
thread:issue:80951,issue #80951: ExpandedWeights can't handle modules with tied weights,stale/low urgency,2,stale/low urgency,1,0,80951,,,,0,d4cfb477e436c7d83238e915173748915ff3c6d4d20ec5ef3ee86ef740497aed
thread:issue:80954,issue #80954: ExpandedWeights sometimes fail silently and doesn't compute .grad_sample attribute,stale/low urgency,2,stale/low urgency,1,0,80954,,,,0,65be4f5979572ac8dbb674b593018fe00e0d5fe9def4503b429317192d2fc231
thread:issue:8097,issue #8097: manager.cpp:64: undefined reference to `shm_open (when building with GCC 5.x (sic)),stale/low urgency,2,stale/low urgency,1,0,8097,,,,0,6286fa2f341633a2fd44b5c9f6ac2127e3e06ff021df297e214cce16eac1bdee
thread:issue:8098,issue #8098: Please provide wheel package for windows on PyPI,stale/low urgency,2,stale/low urgency,1,0,8098,,,,0,812b08e9da65b5af1e197f99a68e9a60e6029e59f4405528760750cecf727556
thread:issue:80985,issue #80985: Nested tensor: Support Noncontiguous Buffer,stale/low urgency,2,stale/low urgency,1,0,80985,,,,0,4885a4b8f92a981db6ad04948cb69f44983cb3bf432c3f009495256e8e6ae012
thread:issue:80986,"issue #80986: [Prims+NVFuser] nvFuser running into ""Tensors of type SparseTensorImpl do not have strides""",stale/low urgency,2,stale/low urgency,1,0,80986,,,,0,0b8513ebe563b1f1bf8969f28774efa5488797e246abdcfabc9d8e90a7dc7c0f
thread:issue:8104,issue #8104: load_state_dict unexpectedly does not load Tensor to buffers that currently have None value,stale/low urgency,2,stale/low urgency,1,0,8104,,,,0,15045a49276f395c18f629ddca018ee908ab4aae36764587bbe84071f3aed5a8
thread:issue:81046,issue #81046: three typing inconsistencies on Tensor methods,stale/low urgency,2,stale/low urgency,1,0,81046,,,,0,6145ee3339b181c7e426fd99fc4a864f5abdda74106c14d3f968640c1b5253b8
thread:issue:81065,issue #81065: [Releng] Improve the tutorials release process,stale/low urgency,2,stale/low urgency,1,0,81065,,,,0,15bcebfd711bf0c1b296e7bf5e25e37612c14b7822744bfe6f2de33c9327a617
thread:issue:8107,issue #8107: Export CC is ignored when I build pytorch,stale/low urgency,2,stale/low urgency,1,0,8107,,,,0,c6bbb1832cf2c3d241bfda65b041afa3056e07279554df06e0a8452e6ba15646
thread:issue:81084,issue #81084: [ONNX] Exporting the operator `::svd` to ONNX opset version 13 is not supported.,stale/low urgency,2,stale/low urgency,1,0,81084,,,,0,4dbc4a58c55c5625e10f43080b7291ad61a2e34ededa7910e4523de38a2fe1c5
thread:issue:81110,issue #81110: CI: Run cpu tests in parallel processes?,stale/low urgency,2,stale/low urgency,1,0,81110,,,,0,3bbc534dfbff366a38a74716a23ca9df625c836b3bd071dfe4a81ebce978f995
thread:issue:81140,issue #81140: libprotobuf version compatibility,stale/low urgency,2,stale/low urgency,1,0,81140,,,,0,a37746ae973e255702a85b27d857aa9576ae7fb0e5ea3baf55c147cd1acd9c51
thread:issue:81162,issue #81162: SparseAdam performance issue during optimizer step,stale/low urgency,2,stale/low urgency,1,0,81162,,,,0,a7413eebded3210284ca7d33cc45c6c4ceaaf82961f7504696496499f32fe30e
thread:issue:81167,issue #81167: move bazel files out of pytorch repo root,stale/low urgency,2,stale/low urgency,1,0,81167,,,,0,2190f3afb1527081c443e950899c31e784cfcbc86b38acda7652e6666fd9ee00
thread:issue:81172,issue #81172: build libtorch with the same mkl as Matlab,stale/low urgency,2,stale/low urgency,1,0,81172,,,,0,d06a3f40168c0c3eb2559d1fc9bdbd10d450d9246006c905c0b4526f8b190d21
thread:issue:81186,issue #81186: grad not preserved during copying or pickling,stale/low urgency,2,stale/low urgency,1,0,81186,,,,0,0758d494ec4ac28c27e4f7ed858eb6ced414ffe6cc61b18d2797dfd5fab325d4
thread:issue:81229,issue #81229: torch.fx.node.map_aggregate and torch.utils._pytree.tree_map do the same thing,stale/low urgency,2,stale/low urgency,1,0,81229,,,,0,3e5c423eb1d328339dce7c3ab011b6ac552e16b030eddb154c94cda0dbba01d9
thread:issue:81240,issue #81240: Functionalization and fake tensors failure in torture test,stale/low urgency,2,stale/low urgency,1,0,81240,,,,0,dc6d04c14fc27a553db83a8503ee8a4b91239c40c8ab536d87f1caa3c01f10fb
thread:issue:81244,issue #81244: CapabilityBasedPartitioner does not work correctly with mutating operations,stale/low urgency,2,stale/low urgency,1,0,81244,,,,0,01c47e3a5c1ef7a689a4de68ff705a66ae2d5af3ee2f08ab2000547f6e74f183
thread:issue:81259,issue #81259: Refactor linter adapters to avoid code duplication,stale/low urgency,2,stale/low urgency,1,0,81259,,,,0,34d77d7f9e579f6f177046adb57aabed85f80134ce1d335a0f6bd3edf1db73bb
thread:issue:8126,issue #8126: PyTorch multiprocessing using single CPU core,stale/low urgency,2,stale/low urgency,1,0,8126,,,,0,a625b4db62c8b48bc4e09334c510b7116fbe78c8b5865cb6b34debdf13dd6dd6
thread:issue:8128,issue #8128: Cmake is getting permission denied when installed system wide,stale/low urgency,2,stale/low urgency,1,0,8128,,,,0,8cef85610878398d92e3556b3817d015bddc89e9dc78fa2c9d31771df63b1ed6
thread:issue:81287,issue #81287: Runtime error in Libtorch cpp project (Didn't find engine for operation quantized::conv2d_prepack NoQEngine),stale/low urgency,2,stale/low urgency,1,0,81287,,,,0,53ff09225abff6deb49511902ac453285360d6fc968a63753dec9fe1d0c6d983
thread:issue:81297,issue #81297: Files downloaded with torch.hub should respect umask,stale/low urgency,2,stale/low urgency,1,0,81297,,,,0,51935940d46e7400ef5d9a07ddcff89c4c62adcb59037fe0c12b168640e8d0d1
thread:issue:81307,issue #81307: [Prims+NvFuser] Non-fusible ops Tracker,stale/low urgency,2,stale/low urgency,1,0,81307,,,,0,089672b937302dc0a9eb99393ff68295de52e8a5af656eee4b27e73f968628ae
thread:issue:81317,issue #81317: Incorrect results for mean or sum kernels on aarch64 when building with gcc-7,stale/low urgency,2,stale/low urgency,1,0,81317,,,,0,e2604b2a48d234b07e5e04b144b02cef7396cd604387597e86ae1b7eafdfdb64
thread:issue:81323,issue #81323: `torch.overrides.get_testing_overrides` does not function as intended for native tensor methods/operations,stale/low urgency,2,stale/low urgency,1,0,81323,,,,0,3ef699387c4e6a451e2372cae458f2bff745e720ef193b477735baa4b54c4467
thread:issue:81358,issue #81358: [Prims+NvFuser] Issue with aten.where.ScalarSelf,stale/low urgency,2,stale/low urgency,1,0,81358,,,,0,6b10b75e4b56473c65c37e071bdcc4d2af0bb338c62b4a9557fd5f42981bdc5f
thread:issue:81381,issue #81381: quantization: QConfigMapping should be easy to print,stale/low urgency,2,stale/low urgency,1,0,81381,,,,0,814cce3adffc5b5a036a98c8ade7d880c799e21d0b0ff2ba8ebcf4443d12c077
thread:issue:81385,"issue #81385: ""Attempted to resize a view tensor to a larger size. This is not allowed in the functionalization pass"" reported on non view tensor",stale/low urgency,2,stale/low urgency,1,0,81385,,,,0,677658de7aabfc7144a96b3a78e32905a5905cdec06961331ce7adcb1dca4481
thread:issue:81405,issue #81405: Implement shape/size functions for nestedtensor,stale/low urgency,2,stale/low urgency,1,0,81405,,,,0,a9b1d7c1af2f2e4d8801dfc20c32f06f8215916a7728c7744f03d9dfda5f4681
thread:issue:81412,issue #81412: num_worker and prefetch_factor in DataLoader do not scale,stale/low urgency,2,stale/low urgency,1,0,81412,,,,0,ca8df8d3a3af0c7afeabe926aeafb572acf3d6d92ee3393caf15ddb6defe7f41
thread:issue:81417,issue #81417: Add a check to detect mutation of the inputs during backward,stale/low urgency,2,stale/low urgency,1,0,81417,,,,0,5a99bef42de29938ca24b3adbcead00fe076aac0113e706501df7b37259df18f
thread:issue:81448,issue #81448: torch.utils.checkpoint optimization opportunity,stale/low urgency,2,stale/low urgency,1,0,81448,,,,0,53c2f9f824688d33d0347e2caa673cbb18c926a65db25038ee1928ab7f1d94e7
thread:issue:81459,issue #81459: Missing corner case handling in ATen ctc_loss implementation,stale/low urgency,2,stale/low urgency,1,0,81459,,,,0,79364532b99dce81370172442e902f15f14d9fa70ced763f0b01f5116e386859
thread:issue:8153,issue #8153: GRU is implementation of GRU v1 draft rather than final GRU paper algo,stale/low urgency,2,stale/low urgency,1,0,8153,,,,0,2348825a33e8f98e7951e3ca0f3255749015363c1d85bff561821611676398c8
thread:issue:81531,issue #81531: CapabilityBasedPartitioner doesn't support horizontal (vertical?) fusion,stale/low urgency,2,stale/low urgency,1,0,81531,,,,0,493e64422db072cdc7f9c79557b7cc46915ff350278ef1836308fe98b032e23a
thread:issue:81539,issue #81539: CapabilityBasedPartitioner treats non-compute ops inconsistently,stale/low urgency,2,stale/low urgency,1,0,81539,,,,0,57ccdb610b6a7dd26351016ce78f6850179a84fe996bf4b9e0091b532418ce39
thread:issue:81543,issue #81543: Linking pytorch libraries causes sstream behavior to be overridden globally,stale/low urgency,2,stale/low urgency,1,0,81543,,,,0,93c48285e2b6549a4ebf2459095a5444da2526d4c6fdb6be39abb31510405224
thread:issue:81544,issue #81544: Torch does not build with Lazy TS disabled,stale/low urgency,2,stale/low urgency,1,0,81544,,,,0,512b2996748070c71ae6e4e4c1ebd627215916a2504b8ba2d367edf74b6bef53
thread:issue:81565,issue #81565: __getitem__ is returned as an OverloadPacket instead of an OpOverload in __torch_dispatch__,stale/low urgency,2,stale/low urgency,1,0,81565,,,,0,18577d6e7371ddc8ef02806195b171c0ddd2a953f993957d287ba20d6cc8b30a
thread:issue:81568,issue #81568: Improve interaction of PyTorch downstream libraries and torchdeploy,stale/low urgency,2,stale/low urgency,1,0,81568,,,,0,ff7c4f57dfaba7d4eeeb5ffa816f66ffdfc5532081440e26befae8490a7659c5
thread:issue:81620,issue #81620: PyTorch 1.12 cu113 Illegal Memory Access or Internal Error instead of Out of Memory cases,stale/low urgency,2,stale/low urgency,1,0,81620,,,,0,17b7dc16f7d0940aed8d650e58b86a13fd78bcaf05a1345a87e6f230a6fd7a9e
thread:issue:81625,issue #81625: [bug] the output shape from torch::mean and torch::var is different in libtorch,stale/low urgency,2,stale/low urgency,1,0,81625,,,,0,d2bc75d01dee0fe5c2c9daf3baeac3131de055e7ca767a0c560f7c3f088abe56
thread:issue:81648,issue #81648: `sparse_coo.to_dense()` produces different results between CPU and CUDA backends for boolean non-coalesced inputs.,stale/low urgency,2,stale/low urgency,1,0,81648,,,,0,da5be10c67990973d07ebfa7489152a3f6ab90ad91382d12a44a63152a8ebc78
thread:issue:81649,issue #81649: Idiom for PrimTorch refs for Tensor methods,stale/low urgency,2,stale/low urgency,1,0,81649,,,,0,b155a4ddde18c2439a8587b8d8859071f3bda1d1391bd387656c3ea380a76be0
thread:issue:81650,issue #81650: RFC: auto-generated plain Tensor argument only sparse primitives,stale/low urgency,2,stale/low urgency,1,0,81650,,,,0,83076f5fd221d9b169626b4dbd9267db121a781d9ea13b6c6ecbec7f1d845635
thread:issue:81654,issue #81654: [packaging] Conda install missing python local version label (+cu123 or +cpu),stale/low urgency,2,stale/low urgency,1,0,81654,,,,0,da9c9f9ae1acd4910ad468bb1a691137ae17006d987e5d1b7f724791b65aa9a3
thread:issue:81667,issue #81667: [Tracker] AO migration of quantization from `torch.nn` to `torch.ao.nn`,stale/low urgency,2,stale/low urgency,1,0,81667,,,,0,e0631fb8ccc9fce74c92cb9bea92d4c397f74454aa61f54625c6d4a34a48c8ad
thread:issue:81669,issue #81669: Register refs for CompositeImplicitAutograd ops as decompositions,stale/low urgency,2,stale/low urgency,1,0,81669,,,,0,be12b5282b87c327ac2ceedc4fc3448ee8c97f775ecb3a13475bffffe45d7cb4
thread:issue:81682,issue #81682: Python operator registration API for subclasses,stale/low urgency,2,stale/low urgency,1,0,81682,,,,0,2f4144891503b834b8244c06e8f290fd28daf12440665c21d72da2381eb50ea8
thread:issue:81717,issue #81717: PyTorch Embedding Op with max_norm is not working as expected,stale/low urgency,2,stale/low urgency,1,0,81717,,,,0,f328b35bdf53b9ec906a0c3e869a9d518c04c2a87e3cc20eee21d65119a4a0db
thread:issue:81722,issue #81722: Inconsistent naming convention for end of enum in DispatchKey,stale/low urgency,2,stale/low urgency,1,0,81722,,,,0,39cca3d926409e9daed24bb1cfcbbb73c8d5380dcd555182dadbd7866947c5f1
thread:issue:81750,issue #81750: Modernize logging tensor in torch.testing._internal,stale/low urgency,2,stale/low urgency,1,0,81750,,,,0,7c6fa1d4cbe344a76bfd1a97aca538c954704d2c9a800ae41e8d8c327df5e4db
thread:issue:81768,issue #81768: TestTagsCPU.test_tags__refs_constant_pad_nd_cpu_float32 flaky with dynamo & pytest,stale/low urgency,2,stale/low urgency,1,0,81768,,,,0,3dc0ce6195daba8d2bc8bd7d814b66754573cfbc4a77e8f7d99a31f90f3337eb
thread:issue:81801,"issue #81801: Rename DispatchKey Dense/Sparse/etc to DenseFunctionality/SparseFunctionality, use original name for alias",stale/low urgency,2,stale/low urgency,1,0,81801,,,,0,c79e807eee404715a4798b85e96c4d06ae555d215920c115d6e0187009418fac
thread:issue:81856,issue #81856: Compatibility with newest MKL,stale/low urgency,2,stale/low urgency,1,0,81856,,,,0,d7a75eab86fc2d8e017a5bf64c8777b1fa97f9a4021c7a3793fd746a6d8d21e6
thread:issue:8186,issue #8186: [Caffe2] Successive in-place operators cause RuntimeError of gradient operator versions,stale/low urgency,2,stale/low urgency,1,0,8186,,,,0,df3c4595b3dd827d6b8376d6ee0d1d6bda2a13ca98d89bfc1dfa128a636999c9
thread:issue:81876,issue #81876: dtype mismatch when after using auto mixed precision,stale/low urgency,2,stale/low urgency,1,0,81876,,,,0,ef385a763af43f2020fd58796991b0d53c3200eadb53abd57ea970a1a1c62618
thread:issue:81883,issue #81883: RuntimeError: CUDA error: no kernel image is available for execution on the device,stale/low urgency,2,stale/low urgency,1,0,81883,,,,0,7d5d80585f1bc0d7b51834fb764d9fa54faf747910d60abb0eda7608372ec7f8
thread:issue:81899,issue #81899: Dependency header directory is not properly expanded in the utils.cpp_extention in ninja mode,stale/low urgency,2,stale/low urgency,1,0,81899,,,,0,099c519276c3645676a9685709a2f5ccbd251e48f02c86b093e2a9aa5376fc50
thread:issue:81932,issue #81932: Make sure we always redispatch through a dispatcher for all SymInt ops,stale/low urgency,2,stale/low urgency,1,0,81932,,,,0,e3cf2e6dda43af1f97821d79a7482c45b10b3a805d08fc224b8a3e4aeec5002b
thread:issue:81935,issue #81935: implement sym_numel,stale/low urgency,2,stale/low urgency,1,0,81935,,,,0,8de46c3c67bfd7cdfc3449af3f6b3eb50397440ee3747da23fbbe2d1da0d8787
thread:issue:81937,issue #81937: Add more autograd tests with symints,stale/low urgency,2,stale/low urgency,1,0,81937,,,,0,4521412923cf1327304160d90878acc557c13fc3fbb6c93da9537344d839a254
thread:issue:81938,"issue #81938: Devirtualize sym_sizes, virtualize sym_sizes_custom",stale/low urgency,2,stale/low urgency,1,0,81938,,,,0,dfa37e7438553dbd2fd6848e69644ccf280bed8e4d7a28d06e9ffc7cffcce3ea
thread:issue:81943,issue #81943: Investigate if it's okay to throw a RuntimeError instead of TypeError here : https://github.com/pytorch/pytorch/pull/79560/files#diff-415017bcad4fa6cd6d3dfe5f6ea1caffcd7122b46b8c1e4825f7d889efc80a62R1816,stale/low urgency,2,stale/low urgency,1,0,81943,,,,0,06b91a76cfd855d2e870cacc32c3a37366f2feb8cff938b8ce9142f834cfb927
thread:issue:81945,issue #81945: Investigate adding Dockerfile linter hadolint to CI,stale/low urgency,2,stale/low urgency,1,0,81945,,,,0,65190eb816501690bad7485193b6154d2cc573ef3f735421e04f58807b102e16
thread:issue:81955,issue #81955: Investigate adding shell linter/checker to CI,stale/low urgency,2,stale/low urgency,1,0,81955,,,,0,3a72ce22ebc2450b8e32e0928de7006c81063b2fb52bdf0b61e9bc5b6e452811
thread:issue:81959,issue #81959: Overloading multiple signatures for a single ref,stale/low urgency,2,stale/low urgency,1,0,81959,,,,0,8a5c23e98793fcc4e0b84ece722547c6e154eb6015d1a54166d410c4720c7061
thread:issue:81979,issue #81979: test_sparse_spdiags_cpu_bool fails on my local working copy,stale/low urgency,2,stale/low urgency,1,0,81979,,,,0,3886b24a06da7b2afe49577ca85f58a54d7fd463bcc2584284aa8c7204ff669f
thread:issue:81982,issue #81982: test_sparse_matmul_cpu_complex128 fails on my local copy,stale/low urgency,2,stale/low urgency,1,0,81982,,,,0,53519345c907885c21869e3f3fee949de5efeaa7f043c917e38dd9b21b32ac67
thread:issue:81983,issue #81983: backwards compatibility ALLOWLIST is misused,stale/low urgency,2,stale/low urgency,1,0,81983,,,,0,3a814f34c4c79f8ebbb723c54996489abd2aeb9fc98e7de703f6d26718f74590
thread:issue:81985,"issue #81985: nn.InstanceNorm and nn.GroupNorm are affected by padding, so they need to masking",stale/low urgency,2,stale/low urgency,1,0,81985,,,,0,a6c65ad3c8f541ad55e0f739d6c462c45e130c6832727d33820349bdedec1ab8
thread:issue:81988,issue #81988: Unify c10::Event and at::cuda::CUDAEvent,stale/low urgency,2,stale/low urgency,1,0,81988,,,,0,ac8132890f5bc797ce960ce1a441e10c8f03a1c80e35293b4b8dc74267c4154f
thread:issue:81996,issue #81996: linspace cpu and sometimes cuda is wrong on integral types,stale/low urgency,2,stale/low urgency,1,0,81996,,,,0,bfa649ae2c711f7d50d550edf93ce71197b751e5cabd1d6f021fad7ed61397b5
thread:issue:82033,issue #82033: [PyTorch/XLA] Improve the XLA PR landing process,stale/low urgency,2,stale/low urgency,1,0,82033,,,,0,0c021cae8bf4f48f8667b1bf487788c04a1afbe02fd872952cd937c392f4e1cb
thread:issue:82041,issue #82041: [Misleading] The doc started using Tensorflow terminology in the document to explain how to use the Pytorch code.,stale/low urgency,2,stale/low urgency,1,0,82041,,,,0,71cf774a860d1e20b18da1ae83cd8315d7a048b3a33f430652fc1c72fe6ced14
thread:issue:82053,issue #82053: Inconsistent implementation of quant_utils:: ChooseQuantizationParams compared with fbgemm:: ChooseQuantizationParams,stale/low urgency,2,stale/low urgency,1,0,82053,,,,0,4fcbbbf8e66c9c6de5406f939f9cb121b3c724d929dd700bc3e661557c4789d5
thread:issue:82061,issue #82061: Documentation for torch.cuda.Event(blocking=True) is wrong,stale/low urgency,2,stale/low urgency,1,0,82061,,,,0,8828a8cacf58b71d3c458444e64e7a72500c47a54b082808f898bebd8b65f1db
thread:issue:82073,issue #82073: CUDACachingAllocator should be cuda memory merge/compact friendly,stale/low urgency,2,stale/low urgency,1,0,82073,,,,0,8222b8f93da561d4195da56225bf9d61761f2b79b68e0d8660bcec5e09258603
thread:issue:82076,issue #82076: Autocast documentation examples would break,stale/low urgency,2,stale/low urgency,1,0,82076,,,,0,0b912eb7df3bf779ba491ec833bdb0efba49fac65da84467b944259326bd2e02
thread:issue:82081,issue #82081: Could be clearer that Cross Entropy takes logits as input,stale/low urgency,2,stale/low urgency,1,0,82081,,,,0,fc8fb55b00347dcca3917ac302012a52185a79909cbe5b02b55c8c12618b33b7
thread:issue:82088,issue #82088: linear.matrix_power is not composite compliant,stale/low urgency,2,stale/low urgency,1,0,82088,,,,0,45f911d7f0fbe0db4777d365651b88dcce3c2b252a0b73664bdf093fce658e9d
thread:issue:82091,"issue #82091: RuntimeError: [1] is setting up NCCL communicator and retreiving ncclUniqueId from [0] via c10d key-value store by key '0', but store->get('0') got error: Timeout waiting for key: default_pg/0/0 after 1800000 ms",stale/low urgency,2,stale/low urgency,1,0,82091,,,,0,a15e661d64b709e4794a0dd971b853b266881b1b2e2e129202b41ef1fe4ad0ae
thread:issue:82093,"issue #82093: When using libtorch v1.10.2, calling at::slow_conv_dilated3d directly returns wrong results on cpu backend",stale/low urgency,2,stale/low urgency,1,0,82093,,,,0,bfe1055c6409040dc1ccd47bdf99230846c95e121138c2d80c0213a4ca757796
thread:issue:82095,issue #82095: torch.concat type hints fail for keyword argument,stale/low urgency,2,stale/low urgency,1,0,82095,,,,0,1b83c3f67fe2c82c94f080a8dd6027c8c4894ca856f83f4d81379d3303fe433d
thread:issue:82132,issue #82132: External libraries cannot have a requirements.txt that needs to install a cpp_extension,stale/low urgency,2,stale/low urgency,1,0,82132,,,,0,3d8b600f3723493b8cceef88119814ca240749aab05dcfac966a9ee953bf202a
thread:issue:82140,issue #82140: Stop manually binding sparse factory functions,stale/low urgency,2,stale/low urgency,1,0,82140,,,,0,a6988329599a9cde32852b27184156708b0666e272d527985996b681008e3995
thread:issue:82145,issue #82145: set_grad_enabled not respected when running on a web server,stale/low urgency,2,stale/low urgency,1,0,82145,,,,0,bec672d6fb6e45b13be755aadca6e7234cb62fc7710831f066691f6ed08af3df
thread:issue:82156,issue #82156: [Checkpoint] Support multiple unpack in saved tensor hooks,stale/low urgency,2,stale/low urgency,1,0,82156,,,,0,30c6542353e777e2c5c3e5f45a1b042b1da3958b1d81d236d1bc34379e4d8341
thread:issue:82185,issue #82185: DataLoader: `pin_memory` should respect object attributes before object collection type,stale/low urgency,2,stale/low urgency,1,0,82185,,,,0,51e007df7a24dce2ff53b0660f8e5c04c3ad0ce7970b75d2cbd0b0502aefe50c
thread:issue:82197,issue #82197: Provide error handling for ops that don't yet support Dynamic Shape,stale/low urgency,2,stale/low urgency,1,0,82197,,,,0,328fe1ab5cb3367c6c4207e40eaf6d72b481d127119cef900b1b6e01763ebf4a
thread:issue:82200,"issue #82200: model.to(device) takes time forever on A40-8Q, NVIDIA. cuda11.1, torch1.9.1.",stale/low urgency,2,stale/low urgency,1,0,82200,,,,0,ae41c7d5deabf905237fc08ea77d45a1045df9b2ad0b63a86623454ea613ab51
thread:issue:82212,"issue #82212: RuntimeError: ""reflection_pad2d"" not implemented for 'Half' in autocast enabled region",stale/low urgency,2,stale/low urgency,1,0,82212,,,,0,60974a49eef98750e985a625bd16a23a72b5ae15f56e2f8c3b3932730fba9298
thread:issue:82219,issue #82219: [feature request] DataLoader to accept num_threads argument to auto-set number of threads for OpenMP / intra-op parallelism,stale/low urgency,2,stale/low urgency,1,0,82219,,,,0,fd10eaf9613eda1c0adfcb30b094dac76575aff3882548dc84a1cde66c82cd74
thread:issue:82230,issue #82230: logspace and linspace off by one on cuda for integer dtypes for some inputs,stale/low urgency,2,stale/low urgency,1,0,82230,,,,0,9e20c0f01a3e9fa0e9b9ae9b4a717fb44b49762524dcc0c59dc3468813e1fcee
thread:issue:82239,issue #82239: primtorch refs should be composite compliant,stale/low urgency,2,stale/low urgency,1,0,82239,,,,0,4368acc8aa718fd4a74aff455bb571e52d6a9f32564538d02105141f5ddb067b
thread:issue:82242,issue #82242: logspace inconsistently casts inputs to int before performing computation,stale/low urgency,2,stale/low urgency,1,0,82242,,,,0,b9a83f826b3033f4cbc60d9d5728f3c9ec1cf58e0ed0f484b16a4569507e42b7
thread:issue:82259,issue #82259: [Reproducibility] Make tests say when unusual environment variables are set that change behavior of the test,stale/low urgency,2,stale/low urgency,1,0,82259,,,,0,32c13397ecdde9015f0c47ebef2e65739cd34931b689a18a8a40eeb632b2fb81
thread:issue:82276,issue #82276: DEBUG=1 env var doesn't actually set DEBUG preprocessor macro,stale/low urgency,2,stale/low urgency,1,0,82276,,,,0,c7281081b599b7dfe2026f120772f56e749b3e7220ac02bdb5c4b5a340b9129a
thread:issue:82303,issue #82303: Race condition between torch.tensor's view and /= (/= returns incorrect result),stale/low urgency,2,stale/low urgency,1,0,82303,,,,0,0f5d2352f79de6cea6be32a20e8d8952daefe2a46b9b2c64796c64eb170a4bbd
thread:issue:82316,issue #82316: Add more Vulkan operations,stale/low urgency,2,stale/low urgency,1,0,82316,,,,0,dcc341eba0d1f905cecf800f3438882af809a52b79f1e1af619274ab810a8614
thread:issue:82357,issue #82357: ufmt and flake8 lints race,stale/low urgency,2,stale/low urgency,1,0,82357,,,,0,b453bfcaced1333af07bf623f57d8d49f3be0dce0bdeb268aee0de0774985f27
thread:issue:82377,issue #82377: masked_scatter_ is very lacking,stale/low urgency,2,stale/low urgency,1,0,82377,,,,0,3d2483b9a7cca22a3c3524d18b3e08a1c98a36d66dfa7108d63c7a42015a513d
thread:issue:82382,issue #82382: Problems in built-from-source pytorch with USE_DEPLOY=1 in Ubuntu,stale/low urgency,2,stale/low urgency,1,0,82382,,,,0,12468cc29777f7d4c21ef8d4f53a3a53d3537af13f8500f8a3e6d908952520d7
thread:issue:82397,issue #82397: Bilinear interpolation with antialiasing is slow in performance,stale/low urgency,2,stale/low urgency,1,0,82397,,,,0,0b26c567ac669e5d2dbddfd3e644729857e8bd42f0861a0109cf4e7086d526ec
thread:issue:82419,issue #82419: The torch::deploy document is not updated,stale/low urgency,2,stale/low urgency,1,0,82419,,,,0,c7f472e1e057f55308cdd8f97a2290878ab6f3e6dd83a9850c73c5fb92ed8240
thread:issue:8242,issue #8242: Different behavior of LSTM and LSTMCell implementation,stale/low urgency,2,stale/low urgency,1,0,8242,,,,0,01c90f078854d82141f61a44eba24496b7fbbf36fa6dc37f4fef73c160b3174f
thread:issue:82430,issue #82430: Minor inconsistency in description of `attn_output_weights` in MultiheadAttention docs,stale/low urgency,2,stale/low urgency,1,0,82430,,,,0,a6fcb5b9df97d9db662d001f7c929430ffab7de405128ad0c5e71911104bc832
thread:issue:82465,issue #82465: Does torch.utils.checkpoint compatible with torch.cuda.make_graphed_callables?,stale/low urgency,2,stale/low urgency,1,0,82465,,,,0,a5481aafb89aaa1175bf81e65c929cfba081dd57237a30ad72b022f0dd6077e7
thread:issue:82479,issue #82479: Cloning conjugate tensor in torch_dispatch context produces non equality.,stale/low urgency,2,stale/low urgency,1,0,82479,,,,0,e36957cfddd610de46f9cfd3e27d468fa60399eed84fd9f6b4f7e8cbb31e75ec
thread:issue:82494,issue #82494: Implement torch.clamp() on sparse tensors with SparseCPU backend,stale/low urgency,2,stale/low urgency,1,0,82494,,,,0,af5027ea2b6fff4d400ec797067e45544406a3577ce5ffe43d2058d5e7b63456
thread:issue:82534,issue #82534: Use NestedTensor in RNN models,stale/low urgency,2,stale/low urgency,1,0,82534,,,,0,03bbc41d54d530800a7741f4827c3134ba10b83ed3e5846860ca0693f5527de8
thread:issue:82542,issue #82542: Is there Doc that explains how to call an extension op in another extension implementation?,stale/low urgency,2,stale/low urgency,1,0,82542,,,,0,a2068a4f74d5fbe0e1d3a276202e230fc39cef27bd92f6a44b6b420ca7aedda1
thread:issue:82545,issue #82545: Incorrect CPU implementation of CTCLoss backward step,stale/low urgency,2,stale/low urgency,1,0,82545,,,,0,2828781ab30b6223d2104b15165d37428f1ed0f57a51d702870b00bc7389d369
thread:issue:82577,issue #82577: RFC: Add flag for RNN decomposition to all RNN modules,stale/low urgency,2,stale/low urgency,1,0,82577,,,,0,60342be3fd9aeb53c9083afbc23364e7f421ed798c03103e996f3774437e570e
thread:issue:82583,issue #82583: DataLoader parameter pin_memory_device should accept torch.device type,stale/low urgency,2,stale/low urgency,1,0,82583,,,,0,5db7fe436111cc1da492eef4c917f7861a24e42930c7060e52b4a34204e7e931
thread:issue:82610,issue #82610: contiguous() not work for rank 1 length 1 tensor.,stale/low urgency,2,stale/low urgency,1,0,82610,,,,0,822f19cae495368cb499fb6994b804980f7016f94e5ffd8d46637c26d8c8b641
thread:issue:82669,issue #82669: unittest.subTest and way to selectively mark subTests as expected failures,stale/low urgency,2,stale/low urgency,1,0,82669,,,,0,0e405174bac1672917368551f3c33f2e43950af9ff789dbb0fd47e54e926b218
thread:issue:82687,issue #82687: Move nested-tensor tutorial from prototype,stale/low urgency,2,stale/low urgency,1,0,82687,,,,0,5431835a49a22c43b5c58b8c2f81b1b47fa7b64f5600526d9c818e800e0f8f32
thread:issue:82710,issue #82710: build fail when using lto with gcc,stale/low urgency,2,stale/low urgency,1,0,82710,,,,0,cca22140faa567a60a310b35c7def52b23b252c423be8fb2a1b4667c30c5fc55
thread:issue:82727,issue #82727: DecompositionInterpreter creates invalid graphs for FX graph modules created with torch.fx.symbolic_trace,stale/low urgency,2,stale/low urgency,1,0,82727,,,,0,d2dae6b85eefa1aae631c60471dd697f8212eae9b81f61e1147bb87cae079232
thread:issue:82751,issue #82751: Refactor how errors decide whether to append C++ stacktrace,stale/low urgency,2,stale/low urgency,1,0,82751,,,,0,7ac73aa3708ea713709b31090e80bd36c6c8e247aee6067cc619ac5007f253f0
thread:issue:8276,issue #8276: [Installation]: Support conda/pip install with ppc64le(power8),stale/low urgency,2,stale/low urgency,1,0,8276,,,,0,080e0a67969a012456cc91fd9a9beacfdcde46fa482ec805a6e396f1e2632007
thread:issue:82764,issue #82764: Subclass of Tensor doesn't support __format__,stale/low urgency,2,stale/low urgency,1,0,82764,,,,0,6f7dffc0009e08ee16222b5d6076ea285a7f4288b984a38e7eb0cf21c8a6bd3c
thread:issue:82789,issue #82789: [LibTorch] the C++ api needs detailed error reports like pytorch,stale/low urgency,2,stale/low urgency,1,0,82789,,,,0,15b1bdcecead62f9e4dc5a4a422cb6cb1003378ef8e6eb6b7883af0d79a43d45
thread:issue:82793,issue #82793: Profiling results on CPU is not reliable,stale/low urgency,2,stale/low urgency,1,0,82793,,,,0,4b171a04a6fd56ece8362d0a511f09ac3debb109803698d9b9ffd90b094f4ed8
thread:issue:82802,"issue #82802: [ROCm] build instruction is haphazard missing information unclear, build does not work",stale/low urgency,2,stale/low urgency,1,0,82802,,,,0,2cc54bc005d9efb0ce4074a56823a7fda312e1f4bf64f4860cac6131e82e65f9
thread:issue:82823,issue #82823: getDLContext in DLConvertor.h cannot be found,stale/low urgency,2,stale/low urgency,1,0,82823,,,,0,e8c632ad86da781292f9884e245a1703c23ee09fe4e6de40b2c71a0416998d39
thread:issue:82843,issue #82843: Tensor operation hangs when used with multiprocessing,stale/low urgency,2,stale/low urgency,1,0,82843,,,,0,2f55823534efa22a10085be5c9f6eac5a400b4df599c6a97ebbecb4248a4c25e
thread:issue:82871,issue #82871: GPU arch 8.6 is not covered by the `TORCH_CUDA_ARCH_LIST = All` option,stale/low urgency,2,stale/low urgency,1,0,82871,,,,0,410a6b958a136dfa47c4f1b239dafc3d837bfaed1cbf2e93c30448cbeb8013ea
thread:issue:8290,issue #8290: [feature request] MPI init_method for torch.distributed,stale/low urgency,2,stale/low urgency,1,0,8290,,,,0,116c24912ccc097c8927174746615a85121fefe87fdc0f80a6892767accc0ce4
thread:issue:82919,issue #82919: Adding a warning of non-compatibility with forward hooks for the fast path of TransformerEncoderLayer,stale/low urgency,2,stale/low urgency,1,0,82919,,,,0,437a9341ca22cf1243a028de3b52e701e990351a8df5422600c920328ab48abd
thread:issue:82960,issue #82960: torch.bitwise_xor argument 'other' documentation wrong,stale/low urgency,2,stale/low urgency,1,0,82960,,,,0,2cd78888903aeaa7d2fc2c70f653d802f1972d5ff395218d131dcdc277a8ee12
thread:issue:83015,issue #83015: Incorrect tensor conversion to m1 MPS.,stale/low urgency,2,stale/low urgency,1,0,83015,,,,0,d1ab082407afa050a0334e570cda0bc91d7068a423bc9d5cd642f91c6cc9923d
thread:issue:83019,issue #83019: TestCommon.test_dtypes error message is confusing,stale/low urgency,2,stale/low urgency,1,0,83019,,,,0,949436a6f777bbc9207611e8447fbc4e718fb32b544b02176617bd70907c4376
thread:issue:83024,issue #83024: New PR template suggests a pattern that does not close PR,stale/low urgency,2,stale/low urgency,1,0,83024,,,,0,44297b0be61f024fcea8d8b043a104c42b374595ebb8a106829468ecf9cd4b6e
thread:issue:83032,issue #83032: Support for CSR Tensor with NN layers,stale/low urgency,2,stale/low urgency,1,0,83032,,,,0,d0ff0d275d9b09e4877f1430c3a88160a38c3676a04a6af313d654dce8483f64
thread:issue:83060,issue #83060: torch.empty_strided argument 'size'and 'stride' documentation wrong,stale/low urgency,2,stale/low urgency,1,0,83060,,,,0,a3a9b8e9cbc974774a34a019d94be7e0fa8f5528f89e8685b85a8792c8dc7a8d
thread:issue:83064,issue #83064: Updating the LTS version of the torch (1.8.2 -> 1.10.2\1.11.2?),stale/low urgency,2,stale/low urgency,1,0,83064,,,,0,7e3f92596509febdad9025f5561f0d1cf6232beb9393b17f78471cccbcccf439
thread:issue:83074,issue #83074: DPP training incompatibility with checkpoint and detach,stale/low urgency,2,stale/low urgency,1,0,83074,,,,0,17454f6711d05e7078342988791d4ec7debc9732ebdfd1a7e12aca93cea86971
thread:issue:83081,issue #83081: [CI] Split up periodic.yml into forward-fixable.yml and periodic.yml,stale/low urgency,2,stale/low urgency,1,0,83081,,,,0,69b93000438544b7472eed8b0f15faa8ead63614680993c570d247bd0999d0da
thread:issue:83082,issue #83082: [CI] Create periodic fuzzy testing for PyTorch build flags,stale/low urgency,2,stale/low urgency,1,0,83082,,,,0,fbe363155561fb6bcd643f0e6437b9dc6583cd5b84a065d8248707481f56a389
thread:issue:83135,issue #83135: torch.nn.functional.avg_pool{1|2|3}d error message does not match what is described in the documentation,stale/low urgency,2,stale/low urgency,1,0,83135,,,,0,ebf84563fa1409c8326c9f9fb204463da05a3988d597bc7a82c50d79385337ba
thread:issue:83148,issue #83148: Cannot call CUDAGeneratorImpl::current_seed during CUDA graph capture,stale/low urgency,2,stale/low urgency,1,0,83148,,,,0,300c730ce5a1347a49e8c1b7c6dd54b5957bea81b390b26822177df1e2da28f5
thread:issue:83152,"issue #83152: When padding is big int, torch.nn.functional.fold runs too long and can't return result",stale/low urgency,2,stale/low urgency,1,0,83152,,,,0,fa8ad5602acbf80359d27e009761d4689fb9a7a2ad5bdb0997ce67b932acfbd5
thread:issue:83157,issue #83157: Fail to install torch for source,stale/low urgency,2,stale/low urgency,1,0,83157,,,,0,8f0e449bf3d0f19b46250d4b231e12f8f8d67be96f749e8f42ad2c4af7aef437
thread:issue:83169,issue #83169: torch.nn.functional.softplus / torch.nn.Softplus parameter beta can be set to zero,stale/low urgency,2,stale/low urgency,1,0,83169,,,,0,01865f54ef9e6f512dae1152ecd1dd83651d0b82ffe38dcde99b8689348ffd50
thread:issue:83175,"issue #83175: torch.nn.GRU runs long time, when num_layers is large",stale/low urgency,2,stale/low urgency,1,0,83175,,,,0,3be00f6c5ba337be3e6711b33652b2d63a6ca775e7e05c57a18ddaf4fade87bd
thread:issue:83193,"issue #83193: module 'torch.distributed' has no attribute 'pipeline' - macOS, PyTorch 1.12.1",stale/low urgency,2,stale/low urgency,1,0,83193,,,,0,ba766665ff8b19c5d7ef3b86d8f4cbc343ae7254945678e8c2a77ec57b608e36
thread:issue:83214,issue #83214: torchgen.model.FunctionSchema.parse fails with following ops' schema,stale/low urgency,2,stale/low urgency,1,0,83214,,,,0,e80cf8160f17f03714a069699f71dcbe34de2140b213f1a4468ff23cda12e2dd
thread:issue:83221,issue #83221: torch.nn.InstanceNorm{1|2|3}d doesn't verify the value type of parameter num_features,stale/low urgency,2,stale/low urgency,1,0,83221,,,,0,6ff22daf57cca4e4216df33a0471b5f4d1de36487a78066727e9b883ac481e20
thread:issue:83229,issue #83229: torch.nn.MaxUnpool2d get negative size tensor,stale/low urgency,2,stale/low urgency,1,0,83229,,,,0,6c0ad3e93af8a5a4601a798e7a70efce1802300375109f54aa85d99e5c2deb58
thread:issue:83232,issue #83232: torch.nn.PixelShuffle error message wrong,stale/low urgency,2,stale/low urgency,1,0,83232,,,,0,29d575ca7fa4e0eb991e8182b29f921793bb5f4e77e5eaff52a994fe343933d4
thread:issue:83234,issue #83234: torch.nn.ReplicationPad{1|2}d supports more input dimension than are written on documentation,stale/low urgency,2,stale/low urgency,1,0,83234,,,,0,ec0a09572965188f6db0d8915639be1b7acb6cfa8ca2854f778d9353a5a5b526
thread:issue:83238,issue #83238: The type of parameter 'p' in torch.nn.TripletMarginLoss wrong,stale/low urgency,2,stale/low urgency,1,0,83238,,,,0,3084bd3f0dbde82cea4eec0b1e1ea42f2fae37fcb67d5aeea905b9058ffa8641
thread:issue:83244,issue #83244: torch.nn.Upsample's error message is inconsistent with the documentation,stale/low urgency,2,stale/low urgency,1,0,83244,,,,0,c66e1883c3bc5dd804ebe198cf1e820ce0451583fa3ed8c0de75751e9d9cf064
thread:issue:83245,"issue #83245: During DDP training timm densenet121, mobilenetv2(v3) models do not save state_dict correctly.",stale/low urgency,2,stale/low urgency,1,0,83245,,,,0,5151d1d8b5aa0939bc29e5ecd9862d64b2e931a7dc0ab7c88ef6989888cccae8
thread:issue:83257,issue #83257: PyYAML not listed as a dependency,stale/low urgency,2,stale/low urgency,1,0,83257,,,,0,6cf1fa076b714ea45c37fc80228554cf7b177123c261f475db3a2da38508e7e8
thread:issue:83272,issue #83272: bmm operator in bfloat16 has low TFLOPS for some tensor shapes with CUDA 11.6,stale/low urgency,2,stale/low urgency,1,0,83272,,,,0,6475b49204afe0b6085c18362b882ad4b39a8977e4d3e531f9b2521fa542bc00
thread:issue:8328,issue #8328: how can I fetch the result of output's Blobs_namet after I sum two numpy array via brew.sum,stale/low urgency,2,stale/low urgency,1,0,8328,,,,0,f3a508bd71341bb90704b6e25f97a5be5deedec5f59dea80cc0b3559b5bc93e7
thread:issue:83281,issue #83281: addcdiv_ (in and out of place) not implemented for torch.float16 and cpu,stale/low urgency,2,stale/low urgency,1,0,83281,,,,0,d950eab40f54cbeddce37a05a9e06adaab5f56d2576634d87463bef2069e8856
thread:issue:83292,issue #83292: TorchVision testing in CI + test_fx,stale/low urgency,2,stale/low urgency,1,0,83292,,,,0,c481c9824c47d7aabd3215b091dbc9c59b230b33f9373ba490dd12abb27bd7b7
thread:issue:83313,issue #83313: torch.Tag doesn't have accurate mypy info,stale/low urgency,2,stale/low urgency,1,0,83313,,,,0,d60d8f71103f52556fb685dfc96c4401276de0f6fe5c7391638c4b883fb53246
thread:issue:83315,issue #83315: dataparallel function doesn't work,stale/low urgency,2,stale/low urgency,1,0,83315,,,,0,31ff20d93b88b11eba745b9dff178f832cb06646245dd6cbcf66f89a03a196d7
thread:issue:83320,issue #83320: Torch1.10.2 is slower than torch1.9.1,stale/low urgency,2,stale/low urgency,1,0,83320,,,,0,95e22e1f2b0a007adf6e3e3af9f7b9d2fdf92dd45b89f87f58799337ad3fbfea
thread:issue:83376,issue #83376: Complex-Valued Gaussian distributions,stale/low urgency,2,stale/low urgency,1,0,83376,,,,0,9202f9545a1542bae293120eac22347448f6c2a6a50b3977c4c2f8a21bf18a16
thread:issue:83388,"issue #83388: zero-numel tensor has ""RuntimeError: strides[cur - 1] == sizes[cur] * strides[cur] INTERNAL ASSERT FAILED"" in multi-thread.",stale/low urgency,2,stale/low urgency,1,0,83388,,,,0,50df9bd69064b99aefa0b49d19fc6b89ba961a5329ccd43a61f6c77e650a2fed
thread:issue:83392,"issue #83392: How to turn off determinism just for specific operations, e.g. upsampling through bilinear interpolation?",stale/low urgency,2,stale/low urgency,1,0,83392,,,,0,5b478fff9df09abc454b04bb9da0f9b0e60bc1721c6f73d5ecd5fdbfe5ed5ce1
thread:issue:83493,issue #83493: Calling torch.linalg.cholesky on a CPU tensor requires compiling PyTorch with LAPACK.,stale/low urgency,2,stale/low urgency,1,0,83493,,,,0,1e2d93dbb431f3dc8252515730f6e370e3140af1a6cfdcfab1c175d69bbd359a
thread:issue:83509,issue #83509: Build and Run QNNPACK on X86,stale/low urgency,2,stale/low urgency,1,0,83509,,,,0,1556d064566d3e9d593e23f3785acc747d3862a462952d123054d8aec8758f87
thread:issue:83510,issue #83510: Cdist backward dependent on compute_mode,stale/low urgency,2,stale/low urgency,1,0,83510,,,,0,af0dbead9810d38f8f58d8d9fa271dfb5fff5d2df5fd291afb08a7426c33ed12
thread:issue:83521,issue #83521: quantize_per_tensor/quantize_per_channel operators should honor the quant_min/quant_max from observer,stale/low urgency,2,stale/low urgency,1,0,83521,,,,0,56090b17af91971c14584ae52107572c5652ffa66fd5e86af7b8885206a15ef2
thread:issue:83537,issue #83537: Add a new argument `check_inf=True` (by default) or check_pos_inf / check_neg_inf to anomaly mode,stale/low urgency,2,stale/low urgency,1,0,83537,,,,0,1a3d71038fc8cfcacff97c3cf695222d84445468c83e682de12b4dddf3c9e0d9
thread:issue:83540,issue #83540: I have the same issue as @samgelman on my MacOS.,stale/low urgency,2,stale/low urgency,1,0,83540,,,,0,d4efa8b608babca915ece59fc26c32572848e37459de234a39b228e58eb71593
thread:issue:83564,issue #83564: KL-divergence of two Generalized Dirichlet distributions,stale/low urgency,2,stale/low urgency,1,0,83564,,,,0,be5816b55debddf6b6178cc0429030c0c36ff4ad86c81ac8d056358754b1b950
thread:issue:83579,issue #83579: Conv1d: NNPACK SpatialConvolution_updateOutput failed when batchsize or padding is too large,stale/low urgency,2,stale/low urgency,1,0,83579,,,,0,b1d646031327dbdc220390d1a1892de908b1970258ee4a895d8dd328f95d396d
thread:issue:83583,issue #83583: Silent promotion of bool to int in the dispatcher,stale/low urgency,2,stale/low urgency,1,0,83583,,,,0,9a79b4325afb96f82c0ec66212cb2f60d7ae9e6f81d2903132df6fa571b15527
thread:issue:83624,issue #83624: Unhelpful error message from torch.linalg.ldl_factor,stale/low urgency,2,stale/low urgency,1,0,83624,,,,0,5137589cc381636457317064425db0eb309b0d004ae48609a3a1f8fe9a36465d
thread:issue:83657,issue #83657: [Discussion] Add custom device,stale/low urgency,2,stale/low urgency,1,0,83657,,,,0,2c744837ee067aac4c1f58af2e90a00e87c6ddaf65f8df2c700d4a55d4b1d14e
thread:issue:83672,issue #83672: quantization: unexpected casting of tensor min and max to int in histogram observer,stale/low urgency,2,stale/low urgency,1,0,83672,,,,0,eb97c17518951449053b3d08a6bf28eccbcadea3788a4fef16b91195d75326b5
thread:issue:83714,issue #83714: Build from source failed on MacOS 10.6 with CUDA 10.1,stale/low urgency,2,stale/low urgency,1,0,83714,,,,0,9ee7cf8c8e273f977dff446f63640b2a51db10505c209831806b9b25bccf640e
thread:issue:83726,issue #83726: nvfuser + prim stack generated illegal PTX code on hardware with sm <= 70,stale/low urgency,2,stale/low urgency,1,0,83726,,,,0,31959e5a80a394e2505ca74d12eaff878642e61b7f34497fc7773ce6b112c3bc
thread:issue:83733,issue #83733: BCELoss results in autocast CUDA warning,stale/low urgency,2,stale/low urgency,1,0,83733,,,,0,082b744b2c2f17090111cd51e176a224c31e16293a6a98b0490f019488457ec5
thread:issue:8374,issue #8374: [Feature request] Add torch.multiprocessing.Pipe,stale/low urgency,2,stale/low urgency,1,0,8374,,,,0,5ed46a8670899952e05d824024d83048d74d5aabf9c66e6badbb46c7a784ee78
thread:issue:83749,issue #83749: [Nested Tensor] Update TestCase.AssertEqual,stale/low urgency,2,stale/low urgency,1,0,83749,,,,0,4fcf8c7f650540efdf8c21f719dfeb65c85486ccee601c8c81e2e048412fc23a
thread:issue:83764,issue #83764: Missing header file,stale/low urgency,2,stale/low urgency,1,0,83764,,,,0,f11c2c06750bdeff1ebe9aa05a43abad44c119cb7be82e2ad9a2eaee4c7f8f75
thread:issue:83817,issue #83817: [feature request] Add new device type works on CPU,stale/low urgency,2,stale/low urgency,1,0,83817,,,,0,6b5843aba7ee80cd3f101bc4b79a7106b5bc76a631e004077d9ce3fe9a7b18a4
thread:issue:83826,issue #83826: test_lazy spuriously fails if LAPACK is not installed,stale/low urgency,2,stale/low urgency,1,0,83826,,,,0,3150e21bb8927adc85a2555757dc58c19259c978062155e2338eff72e3dfc597
thread:issue:83851,issue #83851: torch fx cannot trace assert for some cases,stale/low urgency,2,stale/low urgency,1,0,83851,,,,0,f8434272fee6cb36dcf1f2786a7ef91b1dbc8aec53a67b68a79b836d761b707c
thread:issue:8386,issue #8386: Properly release NCCL resources,stale/low urgency,2,stale/low urgency,1,0,8386,,,,0,52bf05f926ac1451439b7fea25ade861b981200f7b74f8226573ca2505c6adfa
thread:issue:83902,"issue #83902: Bug in batch names with matmul (result tensor has names=('i', 'i', 'k')).",stale/low urgency,2,stale/low urgency,1,0,83902,,,,0,381da562f27c9baaa6103d3b34c6dbe924a1ccdf3f1b8db748293ecb7eb88735
thread:issue:83910,issue #83910: Tracking nested tensor functions with backward kernels registered in derivatives.yaml,stale/low urgency,2,stale/low urgency,1,0,83910,,,,0,e4455b8e47feb955331358f8c7e8eca045f90e9da4f1d7a36c5dbb1541d4f0c3
thread:issue:83914,"issue #83914: RAM not free when deleting a model in CPU? worse after inference, is there some cache hidden?",stale/low urgency,2,stale/low urgency,1,0,83914,,,,0,3c4e58ae821b2d091344b1817e3608cd48eea0919477efad7dc92b9a86de91ed
thread:issue:83923,issue #83923: Support primtorch view ops in functionalization,stale/low urgency,2,stale/low urgency,1,0,83923,,,,0,6cba8b16a81ce6cc374b4a049620398e1707d1f56e043b7f0a20480be59a2270
thread:issue:83929,issue #83929: ModuleNotFoundError: No module named 'torch.ao.quantization.experimental',stale/low urgency,2,stale/low urgency,1,0,83929,,,,0,d9221ae0660adf857224e114aa247367a6e3153e9ae2286a53264b619a91a9df
thread:issue:83931,issue #83931: primTorch: support refs and decompositions when ATen and Python disagree,stale/low urgency,2,stale/low urgency,1,0,83931,,,,0,1d2dbf16e861d9e12166d531bbb5206f501f3b89a9c7eae236d05c41655fc20f
thread:issue:83932,issue #83932: Hitting rate limits for pytorchbot token,stale/low urgency,2,stale/low urgency,1,0,83932,,,,0,130e62e3f702d292d2f465fb5b133feff040c5eedbc5c6ab3eeacb5ba7b1fa6f
thread:issue:83941,issue #83941: empty_quantized should probably be new_empty_quantized,stale/low urgency,2,stale/low urgency,1,0,83941,,,,0,4289522d71d4945705556dea4f214b15ade189e580f7e7c113e8f9295d08fbef
thread:issue:83947,issue #83947: Are PyTorch Android nightly builds getting automatically published,stale/low urgency,2,stale/low urgency,1,0,83947,,,,0,94efc6d84d5d5de2f6bd5b9ac4a227d3b31d98af85440f5f32b5e70a03a9f3b3
thread:issue:83977,issue #83977: pytorch 1.12.1 doesn't build with ffmpeg 5.0,stale/low urgency,2,stale/low urgency,1,0,83977,,,,0,c3398c075595ea9cfab4a2b75f35176584fcdbdd52bee9ce875fbd43af579f68
thread:issue:83979,issue #83979: Undefined reference in libtorch_cpu.so `...std::__cxx11::basic_string...`,stale/low urgency,2,stale/low urgency,1,0,83979,,,,0,7d83e933e0f6d4082484353940aaf78b1e7a8c50d59ebe710f2f74f582584033
thread:issue:84009,issue #84009: [Nested Tensor] Enable Nestedtensor to work with OpInfos,stale/low urgency,2,stale/low urgency,1,0,84009,,,,0,4fe85e9f2c22fc5e4e688a1fe8f5c8e0f57cc6b457d84d47bbb9c94e18e92318
thread:issue:84014,"issue #84014: fill_ OpInfo code not used, also, doesn't test the case where the second argument is a Tensor",stale/low urgency,2,stale/low urgency,1,0,84014,,,,0,02d5862b539caa213fd182db44bf3c12f38cf17e84c82d13ed1e38ceaf176ab6
thread:issue:84044,"issue #84044: hipErrorNoBinaryForGpu, but reversed",stale/low urgency,2,stale/low urgency,1,0,84044,,,,0,a3aa520a676d01cd68dfaa5e491804cc741b299064ac7f819e4fe74abc3c6253
thread:issue:84045,issue #84045: COREMLTOOLs/NNPACK Python Issue,stale/low urgency,2,stale/low urgency,1,0,84045,,,,0,dd42a222999ada531843d7967d0270573c4693bcc732b6774604d8870467a55e
thread:issue:84050,issue #84050: I cannot install pytorch by Bad CRC-32 for file 'torch/lib/libtorch_cpu.so',stale/low urgency,2,stale/low urgency,1,0,84050,,,,0,b05a3109b81ef6a7f181b8717e118be7725761d038f4798fa6a84685766df92f
thread:issue:84052,issue #84052: Index out of bounds Error with PerChannel Quantization,stale/low urgency,2,stale/low urgency,1,0,84052,,,,0,07bd51e85ab0ddb4d1edd1f9d590996eb45faf6be80edcf8a39ac33c8639bcbb
thread:issue:84064,issue #84064: Better error message for qlinear_prepack,stale/low urgency,2,stale/low urgency,1,0,84064,,,,0,ed1112cb93af3508a1ed82a46af16d4fd08ae6fcf8b0d8fd4ff779aab657583e
thread:issue:84140,issue #84140: Please include virtual/physical batch sizes in the tutorials,stale/low urgency,2,stale/low urgency,1,0,84140,,,,0,be01e2ca79db4639f9b03136f6d83aa8b42839b226b888d2240b7050993df9f1
thread:issue:84148,issue #84148: [Quant] Reference get_default_qconfig_mapping in docs/tutorials,stale/low urgency,2,stale/low urgency,1,0,84148,,,,0,e40ed63409b6aab6bc39fc879fa41228c19e61bcba82b4aba0f23386e7334b82
thread:issue:84178,issue #84178: Support setting strides on quantized weights of Embedding,stale/low urgency,2,stale/low urgency,1,0,84178,,,,0,a5ade07789c1e7d122e92031e39411abdfc0c9e9e3d4df7758b3430bfa40ab8c
thread:issue:84193,issue #84193: Attach execution time to each node in an fx trace,stale/low urgency,2,stale/low urgency,1,0,84193,,,,0,8e570f2187732cfd5a32458f45e806d0e8ba1b4a9d0266e196998227ca36a321
thread:issue:84194,issue #84194: Pytorch gets small bias on the result of different types of divisors while doing floating point division.,stale/low urgency,2,stale/low urgency,1,0,84194,,,,0,1847e514f2f197430273504372d11eecdc9b408b855378cd6bfb0ef579dd4795
thread:issue:842,issue #842: BCELoss doesn't accept LongTensor targets,stale/low urgency,2,stale/low urgency,1,0,842,,,,0,cec136d58f1d117ac4551fbff3e60ffce64d0716578edd3154b7802fd725b646
thread:issue:84202,issue #84202: Gradient value calculation error in MultiLabelMarginLoss,stale/low urgency,2,stale/low urgency,1,0,84202,,,,0,3ccdb5f818d008b143db8f43880f1c96b950139c9d4fb4d146efaa747cbaf7d8
thread:issue:84247,issue #84247: Ensure ops account for offsets and strides,stale/low urgency,2,stale/low urgency,1,0,84247,,,,0,dccad0d9770fd90ca94483915531d75b5a97f0b324b6bba4d093940aaaf3ba43
thread:issue:84259,issue #84259: would you like upload to the cpp libtorch to vcpkg package repo?,stale/low urgency,2,stale/low urgency,1,0,84259,,,,0,2fd494440578f7d14d494caf7c390edb9cad13aaf37aadb5eaf7049f74056a1b
thread:issue:84265,issue #84265: torch.Tensor.to.dtype_layout overload is not available in Python,stale/low urgency,2,stale/low urgency,1,0,84265,,,,0,85b3c3f085aa7b5ace66811533be34e8387ac5f3f53163e4433e8b6134516fe2
thread:issue:8430,issue #8430: batchnorm2d track_running_stats,stale/low urgency,2,stale/low urgency,1,0,8430,,,,0,3d2e9c8ca2b57722aeeae15d8a3158a360abf89fcf75741c4536ff20c7db4f6f
thread:issue:84309,issue #84309: Unable to run a single convolutional layer in different CUDA-contexts,stale/low urgency,2,stale/low urgency,1,0,84309,,,,0,f4863368066fd77b78834815fa986fb4cb69d0f5121bd8cec624a97e45684fc9
thread:issue:84311,issue #84311: DecompositionInterpreter creates invalid graph,stale/low urgency,2,stale/low urgency,1,0,84311,,,,0,5adbb183bd843c8604fd1c95da08f351187bda9243b22fbd5743840fd2f803a3
thread:issue:84316,issue #84316: Torch.fx tracing bug with dictionary.update calls on input,stale/low urgency,2,stale/low urgency,1,0,84316,,,,0,a218e79d511903fadc846adf921788b1d2619b93a7f34d627caf2bd03ced5873
thread:issue:84335,issue #84335: Should enable skipped tests for `to` OpInfo,stale/low urgency,2,stale/low urgency,1,0,84335,,,,0,b511a838917a391f964e378fa3b5b564e556f9f639d5ccae0e24cdad72350f79
thread:issue:84336,issue #84336: torch 1.12.1 cuda 10.2 runs slower than torch 1.8.2 cuda 10.2,stale/low urgency,2,stale/low urgency,1,0,84336,,,,0,39c30b0b7e3888d01764aff19906d7fa29428cfbf1f43fa59ad0832adf4a613b
thread:issue:84346,issue #84346: fix ATen tests that do not compile,stale/low urgency,2,stale/low urgency,1,0,84346,,,,0,a85e57042934d43096fbe990148e94b88b4704a7ea058457aad32e892e8033c6
thread:issue:84348,"issue #84348: RuntimeError: ""slow_conv2d_cpu"" not implemented for 'Half'",stale/low urgency,2,stale/low urgency,1,0,84348,,,,0,7209ac6bd5635032fa8067a1adc9ccc67757de6a1d9810256a5d46bfc90d3cf4
thread:issue:84353,issue #84353: torch.Size should convert all elements to ints,stale/low urgency,2,stale/low urgency,1,0,84353,,,,0,bfd811f2534dfba9d58da8f10f614b7c3e032a5dc1dc9dd30ae7eb8ffc192c5e
thread:issue:84412,issue #84412: Odd type-casting behaviour in prims.div,stale/low urgency,2,stale/low urgency,1,0,84412,,,,0,1a615c758259b6b2b515d83ec63c03d76395e7194e7259d67d437d1fb0f9701b
thread:issue:84414,issue #84414: several questions about pytorch DDP,stale/low urgency,2,stale/low urgency,1,0,84414,,,,0,56d70cae027390922f38afc5d821cbf7107a4081bfc198a97fbca90213ab0064
thread:issue:84415,issue #84415: Deepcopy of FX graph fails with nested make_fx and constant tensors,stale/low urgency,2,stale/low urgency,1,0,84415,,,,0,20c0a5e29fd935ed8ea4f4f0f44e229074bee29bf8599a33f7af2db4ec26bbf2
thread:issue:84445,issue #84445: Strange cuda illegal memory allocation error,stale/low urgency,2,stale/low urgency,1,0,84445,,,,0,b5b01239f46403510444027657a615397f5c9450b9e05356e7bcd05685c8036c
thread:issue:84468,issue #84468: [c10d] Support a public API to retrieve default process group,stale/low urgency,2,stale/low urgency,1,0,84468,,,,0,8024aa334264850f63c347fcd0b2df4fedadc5c20ca406bb29634efcc6a1d1ac
thread:issue:84473,issue #84473: Install LibTorch by Conan or other C++ package manager,stale/low urgency,2,stale/low urgency,1,0,84473,,,,0,276b63930425baedc136ee8cc501e45c9b179157a7c77499b42eb7228bda1054
thread:issue:84510,issue #84510: [NVFuser] RuntimeError: ref_id_it != replayed_concrete_ids_.vector().end() INTERNAL ASSERT FAILED,stale/low urgency,2,stale/low urgency,1,0,84510,,,,0,3a64c2a45275767dcce50bc366ca04b41b1f93651f670a766bac33df6188db38
thread:issue:84515,issue #84515: Torch.FX work with autograd.Function,stale/low urgency,2,stale/low urgency,1,0,84515,,,,0,ff78907299031e5aac3b87e919a9f2b29b3101aa69b127f724a881756880a9af
thread:issue:84529,"issue #84529: test_prims.py:test_nvfuser_no_args_cuda, memory leak",stale/low urgency,2,stale/low urgency,1,0,84529,,,,0,461d44c20283188f54532db816f8ba8f8a5fd6d424fdd061b28b42baaf08c5f8
thread:issue:84530,issue #84530: `tensordot` not working for dtype int32 and lower when there is only 1 element in the given axis,stale/low urgency,2,stale/low urgency,1,0,84530,,,,0,60393c0f7dd3cc145df9efa8287fe26a13edea31cd333c6ffc5d18ba9501fc92
thread:issue:84537,issue #84537: Minimal example for torch.optim.SparseAdam,stale/low urgency,2,stale/low urgency,1,0,84537,,,,0,510367907422da5c28e9c93b5103ce856e0003335ed27d59ca5f3cafe0776b7d
thread:issue:84538,issue #84538: OpInfo tests should compare gpu to cpu implementations,stale/low urgency,2,stale/low urgency,1,0,84538,,,,0,8c6fa4044f93a2ec6b34a273c42958f3419b927c703e42905e684ff56dd5798b
thread:issue:84539,issue #84539: list of tensors can't be converted to a torch tensor while list of lists gets easily converted to a pytorch tensor,stale/low urgency,2,stale/low urgency,1,0,84539,,,,0,cb4d37773904f9c95fd365b66dc1c6e5fa2154a61b9bedd87b4cbb2141387791
thread:issue:84560,issue #84560: [optim] asgd : handling of complex params as real params (NaN vs inf),stale/low urgency,2,stale/low urgency,1,0,84560,,,,0,caccc602a82ca584de5392ee82ea506a1f40daa11a61ca9a2b503afe6f84266d
thread:issue:84597,issue #84597: Accept SymInts and SymFloats For Scalar Inputs,stale/low urgency,2,stale/low urgency,1,0,84597,,,,0,5accf4f268c0013ff3edd3e010cbcdba923658232f93ca3235a98b3a18dfcc91
thread:issue:84616,issue #84616: Autogenerated out functions are missing at::cpu:: and co bindings,stale/low urgency,2,stale/low urgency,1,0,84616,,,,0,433396073163197c641fef6701651bfcbee101374878aef488c6c6b0772b14b3
thread:issue:84620,issue #84620: torch.hub.load local model,stale/low urgency,2,stale/low urgency,1,0,84620,,,,0,66b8fa97abbafc52bda232aee6bbba7e5bd0b1e1b6289082f3bfb5d614cfa382
thread:issue:84630,"issue #84630: torch.nn.functional.interpolate fails on some degenerate shapes, but passes on others",stale/low urgency,2,stale/low urgency,1,0,84630,,,,0,4a3d2b100c57b0c796294ba61eecdc6ea23c8bfc2dd51bd1e8217884096957e9
thread:issue:84691,issue #84691: test_public_bindings is not robust to various build options,stale/low urgency,2,stale/low urgency,1,0,84691,,,,0,07f5144bea0a60ed8324bf39ddb8650545d545c9a4a63d5f5b125345448dc64d
thread:issue:8471,issue #8471: Does the caffe2 have the depthwise separable convolution oprater ?,stale/low urgency,2,stale/low urgency,1,0,8471,,,,0,b38a7b49b28cd1a9aacd5d3a6e768fdd49189453535172d870893417d1b9069f
thread:issue:84710,issue #84710: About the different ways to print models,stale/low urgency,2,stale/low urgency,1,0,84710,,,,0,7b6e9f3f735a42f32e0b9add77c51818158babdc372c668bfc40ed769d448293
thread:issue:84717,issue #84717: Test aten decompositions match their alias information,stale/low urgency,2,stale/low urgency,1,0,84717,,,,0,2cfc0b1a00e291957528ef3bd2fffc8f5f24fabb8b3520460d9bbaeaecd184af
thread:issue:84721,issue #84721: PyTorch-DirectML RFC,stale/low urgency,2,stale/low urgency,1,0,84721,,,,0,afa89d3eacf99bf77d2dded03b3263f7d25134b075b043bc80c1a99c69d7ac29
thread:issue:8479,issue #8479: An error occurred while creating a new notebook.,stale/low urgency,2,stale/low urgency,1,0,8479,,,,0,2e4a99bf9f1e195248ca1764f1d66facfccd183f9f15cfa5d0acb5eb5105d36f
thread:issue:84805,issue #84805: compiling failed from source,stale/low urgency,2,stale/low urgency,1,0,84805,,,,0,0651853dee83f471964ea8ed599dc82e7e5e8db10e05f2640902d0800aee07b9
thread:issue:84813,issue #84813: MPS: allow selecting specific MTLDevice by registryID via environment variable,stale/low urgency,2,stale/low urgency,1,0,84813,,,,0,cf6b350b900a5cd433571ff5c13fa3330a34c6c20e4466b8413ab08f6e106289
thread:issue:84835,issue #84835: Batch multiplication for torch.sparse matrix multiplication,stale/low urgency,2,stale/low urgency,1,0,84835,,,,0,a5f22811d1b98bf8e71ab5df9f558369e72ffae6103c3cd465b3a6506f56392f
thread:issue:84840,"issue #84840: TypeError: finfo(): argument 'type' (position 1) must be torch.dtype, not HFProxy",stale/low urgency,2,stale/low urgency,1,0,84840,,,,0,e4a7317de2df00e46c46e3c6fd2756265e8a82addc336c06a47a56ccbe5602bc
thread:issue:84932,issue #84932: Separate doc and binaries build,stale/low urgency,2,stale/low urgency,1,0,84932,,,,0,92a0d74d636056aa39af7e704c43471ad8e9efbdad5066ea53e6493a17c5a74e
thread:issue:84944,issue #84944: test_warp_softmax_64bit_indexing_cuda_float16 takes ~147GB of CPU memory and is very slow,stale/low urgency,2,stale/low urgency,1,0,84944,,,,0,f6e6511dc354d0f487a09e9548bd379cad59a9ffa7be96ce18fdc5c362254b1f
thread:issue:84972,issue #84972: Add unit tests for test decorators,stale/low urgency,2,stale/low urgency,1,0,84972,,,,0,c8cf956afa7f48d9257f9840d636af3a83d0516340a3b2b2c33f9d172ff73879
thread:issue:84994,issue #84994: Cuda tensor is zero when passed through multiprocessing queue,stale/low urgency,2,stale/low urgency,1,0,84994,,,,0,d69c46cabaf79bdb926168b7bdcfbe4aa8065eabd3c9e36004527458ef7926f9
thread:issue:85058,issue #85058: Selectively sync internal Meta discussions / posts to dev-discuss.pytorch.org,stale/low urgency,2,stale/low urgency,1,0,85058,,,,0,43f6206be51abbd12a882d765f2d20783826f4d5d4a4bbc0fd570ecbac404378
thread:issue:85078,"issue #85078: libtorch create a tensor is very slow, who can tell me why",stale/low urgency,2,stale/low urgency,1,0,85078,,,,0,2b54b457c09bc786b13a3b1ac2d3b81ec7b7ba423da319d971b814fdc91e4a67
thread:issue:85081,issue #85081: primTorch/nvfuser: have a way to check that refs are added to __all__,stale/low urgency,2,stale/low urgency,1,0,85081,,,,0,c59ec40180e70d5ba82c22a6e337298e18f823c432676f692daa676830d3ab27
thread:issue:85082,"issue #85082: [NNC] loop vectorization fails, `Ramp` and `Broadcast` undefined",stale/low urgency,2,stale/low urgency,1,0,85082,,,,0,f7d6faa2f8d2fed91a709e15a381922d5bed56c50d4cd833f577f2dde0c88527
thread:issue:85096,issue #85096: Update `use_deterministic_algorithms` documentation and tests to include `nn.functional` counterparts for all `nn` modules,stale/low urgency,2,stale/low urgency,1,0,85096,,,,0,8b43e9efcd7f3aee58ab0c94cc5209fa6c33ab17406d9b885e071398447e7db2
thread:issue:85098,issue #85098: install libtorch cxx11 ABI as default in PyTorch pip installation,stale/low urgency,2,stale/low urgency,1,0,85098,,,,0,4e64d613677af489b7babf27ee3c42f0865df6d182457717e2835e953c965b98
thread:issue:85105,issue #85105: Python dispatch for PyOps needs to respect tensor subclasses,stale/low urgency,2,stale/low urgency,1,0,85105,,,,0,94571740ddcf217ce5e94d83891efb9cf9b7dbbdb8053cc92166f66fe9677bd9
thread:issue:85121,issue #85121: make_traced() doesn't respect setting the seed,stale/low urgency,2,stale/low urgency,1,0,85121,,,,0,a59232ccb49178e7aa7550b41cab5e9d5e35e6fafb123b399676c9481cbeb229
thread:issue:85122,"issue #85122: PT Dispatcher confusing error message ""There were no tensor arguments to this function""",stale/low urgency,2,stale/low urgency,1,0,85122,,,,0,4d248463d0f9739d32d3fd5b4d84c47926b7b4ad4e00ca02566e9613853abdb5
thread:issue:85147,issue #85147: torch::quantile performance?,stale/low urgency,2,stale/low urgency,1,0,85147,,,,0,ee957349c484268b08383c74c137659bc7f4e0746aeaa79928d570c303612203
thread:issue:85201,issue #85201: performance between manually created graph and CUDAGraph.replay,stale/low urgency,2,stale/low urgency,1,0,85201,,,,0,f54633aa3928a4ba3d4f70421f83d4b8ff6f91d9a9b88ca2fc56b0264f43ee73
thread:issue:85227,issue #85227: topk returns different results with the same input in cuda and cpu,stale/low urgency,2,stale/low urgency,1,0,85227,,,,0,328db92635e632fdd69f61ebff24b2e685c4e5f7f44fc83558f736138c6782cf
thread:issue:85229,issue #85229: TracerWarning: Output nr 1. of the traced function does not match the corresponding output of the Python function.,stale/low urgency,2,stale/low urgency,1,0,85229,,,,0,c24b391f157b0f07a5c0a80be2a7e062310d02155bd12311ffafe7f6f822442e
thread:issue:85258,"issue #85258: nvFuser support for {ceil,floor,round,trunc}(int)",stale/low urgency,2,stale/low urgency,1,0,85258,,,,0,2ac5052b2272d4770c9b04fca176b02f0824d17bb34438bb70a702b5af872f33
thread:issue:85300,issue #85300: torchrun substitutes host names for IP addresses,stale/low urgency,2,stale/low urgency,1,0,85300,,,,0,d9f1ac9f220f30b9b4e00e01330d674b0afb193f378480358f187e9870e23c97
thread:issue:85302,issue #85302: Remove `TypedStorage` and use only `UntypedStorage`,stale/low urgency,2,stale/low urgency,1,0,85302,,,,0,b311ea66530c6a2b185f7c62351c3e0c50e396510a2c8bc7e9bab4461abe81e1
thread:issue:85351,issue #85351: Point community docs to master,stale/low urgency,2,stale/low urgency,1,0,85351,,,,0,5d3a7ea35b7fe3ea9ea12eec4399bc2a7762eda05f6cd20faf7d94775c8aad4b
thread:issue:85375,issue #85375: Inconsistency between geometric distributions,stale/low urgency,2,stale/low urgency,1,0,85375,,,,0,510556501759dcfb8d118e696cc6a7883c94f1f81aa68a9e8903736590f38501
thread:issue:85377,issue #85377: CI fails for test_compare_cpu_nn_functional_embedding_cuda_float32 which is not reproducible locally,stale/low urgency,2,stale/low urgency,1,0,85377,,,,0,e51ee6d4a34feeac69da8105c3ee868f88e5311a5bcc634a88b8ad1b723e1b2a
thread:issue:85387,issue #85387: very strange speed of torch.bmm with specific tensor shape,stale/low urgency,2,stale/low urgency,1,0,85387,,,,0,6fcb9d3de0abbe4340d60c51a6dca41f721577e5ab80bdfe180a518e74b3e5ea
thread:issue:85390,issue #85390: Please put back missing rocm builds of Torch Vision.,stale/low urgency,2,stale/low urgency,1,0,85390,,,,0,433da5031682b9001d24c15bd51a44a39990a9da7e0d6812fc3754a15e64e24a
thread:issue:85392,issue #85392: please report a bug to PyTorch. Expected Object but got PyObject,stale/low urgency,2,stale/low urgency,1,0,85392,,,,0,85bc1d0e406235f2c9b6a4e89b7b6f2abfd92e021a3f6ad5d3405db9fae92954
thread:issue:85397,issue #85397: torch.nn.utils.prune.remove reorders the parameters of a module unexpectedly,stale/low urgency,2,stale/low urgency,1,0,85397,,,,0,64c611cb8653dfc19340886a0004c43c4a2dff49a4986ebe8daa1c1ccfd11f7c
thread:issue:85438,issue #85438: Prim Output Spec is Not Always Consistent With Eager,stale/low urgency,2,stale/low urgency,1,0,85438,,,,0,faab493f9c309a2c7a3ac50996760b244029eaa321001b1c507e4f1cc24af610
thread:issue:8546,issue #8546: Q: how to generate my own pb files to C++ Predictor,stale/low urgency,2,stale/low urgency,1,0,8546,,,,0,d4319abc8be41ea9da23c1b0229503501a03ae1a47ca4b28b15f293412f11cb8
thread:issue:8548,issue #8548: [caffe2] [feature request]Doese caffe2 support conv_nd with group?,stale/low urgency,2,stale/low urgency,1,0,8548,,,,0,27b42089ab81deaed1e6ed5a5fa93adc15928590d938f0fde4290e0b19ae608c
thread:issue:85485,"issue #85485: `max_pool2d_with_indices(self, ...)` shouldn't need to save `self` for backward",stale/low urgency,2,stale/low urgency,1,0,85485,,,,0,15b92ad99ea8a5da3bd11bf454c417018399f783cbf53ba01939bcf8532f2535
thread:issue:85499,issue #85499: Execute smoke test for Better Transformer feature,stale/low urgency,2,stale/low urgency,1,0,85499,,,,0,ce184b3a8df723bbb1eecd9ef986e0493dcf7ee954687983d3bec35a2e474108
thread:issue:8550,issue #8550: caffe2: Does caffe support the model running on different device (CPU and GPU) same time ?,stale/low urgency,2,stale/low urgency,1,0,8550,,,,0,bc80603613871036a097560540662a5496b72b8bf61241c28df63ededbb4d9df
thread:issue:85520,issue #85520: Implement `rand_like` ref and implement nvfuser_impl for `uniform` prim,stale/low urgency,2,stale/low urgency,1,0,85520,,,,0,3e550e9e88085bb1fd61ff61e4ebaaa0ae8178b56317fe39478ff7f0b843c360
thread:issue:85533,"issue #85533: [functorch] transforms like jacrev, jacfwd, grad, etc don't work with BatchNorm",stale/low urgency,2,stale/low urgency,1,0,85533,,,,0,3e89e15ed332bebd0d353d331092ed7781bc8138b690aa063b8f8ac7c9f54c3d
thread:issue:85544,issue #85544: NvFuser single mode changes the output,stale/low urgency,2,stale/low urgency,1,0,85544,,,,0,0aad36a2673e8372591658bbf29250523d404103fe2efeb292a379559f38fca9
thread:issue:85558,issue #85558: torch.Tensor.transpose().contiguous() on dimension of size 1 gives wrong stride,stale/low urgency,2,stale/low urgency,1,0,85558,,,,0,95b58d1d88f0b2b20cbc660b6c6402ea70f5ceb5c4ef046346e5b96a593ab8aa
thread:issue:85590,issue #85590: Solve default argument induced include cycles by not using defaults / moving the defaults to inl,stale/low urgency,2,stale/low urgency,1,0,85590,,,,0,5d4f232fedf04568ab98d17525e8cca444292b1bfb308caa77ca16f2c72f9087
thread:issue:85604,"issue #85604: Multiple GPUs get ""errno: 98 - Address already in use""",stale/low urgency,2,stale/low urgency,1,0,85604,,,,0,31137bb20481e19eed0254cd9ae2edce823bfe919a09a265e5343dbfe5de14c4
thread:issue:85606,issue #85606: Creating NumPy array with `dtype=object` of PyTorch tensors fails,stale/low urgency,2,stale/low urgency,1,0,85606,,,,0,d6e7d18de4d8fe146d82887eb6aedfb12ca7d007b772d7f4366f02b8a997170e
thread:issue:8561,issue #8561: cleanup BLAS detection,stale/low urgency,2,stale/low urgency,1,0,8561,,,,0,a6f92a4a02cbbf0a5e5afc03ef34e9871b91fd82669a4b5f131bb94377889f1d
thread:issue:85618,issue #85618: Build from source failed with error of different gpu architecture (compiler shows sm_30-related error but I use sm_86 GPU),stale/low urgency,2,stale/low urgency,1,0,85618,,,,0,565996bcdfddcaeec4f20fa1ef44946d2e43597e37507a51ed5739cd16682064
thread:issue:85621,issue #85621: Gloo DDP SocketTimeout error on Windows,stale/low urgency,2,stale/low urgency,1,0,85621,,,,0,20434c3d2222774c0dbcc2bc283a2eeb0e4535f545d7e57931128432c52835f9
thread:issue:85625,issue #85625: How to install pytorch with cuda 11.7 in anaconda envirment?,stale/low urgency,2,stale/low urgency,1,0,85625,,,,0,628a3608644e874b71e79020fc1b5ab880dc4818be98fde03b9aeb261ffd5da8
thread:issue:85642,issue #85642: Could not run 'aten::native_batch_norm' with arguments from the 'SparseCUDA' backend. using batch_norm,stale/low urgency,2,stale/low urgency,1,0,85642,,,,0,ef802efb3b0cc4175357417015f3be22e83aa0dbf0a0210580fbe7985101fea7
thread:issue:85656,"issue #85656: [functorch] conv.{1, 2, 3}d should raise errors",stale/low urgency,2,stale/low urgency,1,0,85656,,,,0,a2472bf843944d1ebc2ca7a28e01fa6b78588a6f54bf9a828c8a15f08ff9d9ed
thread:issue:85745,issue #85745: [Quant] Remove or clarify the meaning of Nones in QConfig/BackendConfig,stale/low urgency,2,stale/low urgency,1,0,85745,,,,0,cbd2b0da9d7b1f03f890edeba3d7a2be8052fc58dc33c3f6302b092195fafdc0
thread:issue:8577,issue #8577: Update tests to no longer spew debug info,stale/low urgency,2,stale/low urgency,1,0,8577,,,,0,573cf7648e5784b9d58b2635fcb81c5f0231ed86d887724102d467fae3a67da7
thread:issue:85775,issue #85775: Timed out receiving the shared seed from the distribtued store on Rank 2,stale/low urgency,2,stale/low urgency,1,0,85775,,,,0,91b971b92c9adb1d3c840ae89e21dfb9c71e31f7b927ba5e4cb30e1ec4a1557a
thread:issue:85791,issue #85791: nn.CrossEntropyLoss overflow with FP16 and minibatch,stale/low urgency,2,stale/low urgency,1,0,85791,,,,0,e5d3f03db4faf8cb7b3d706ccc26c83d0ebb0482d4b012ba26f860c7b2fa0d40
thread:issue:85815,issue #85815: string interning for dispatcher operator names,stale/low urgency,2,stale/low urgency,1,0,85815,,,,0,c5bf824dedf56cabb6c4b4edd9bffd52285d730b7c888bc31eb7603714af3787
thread:issue:85841,issue #85841: Setting the cuda device when using start_processes in Jupyter on Ampere leads to CUDA reinitialization error,stale/low urgency,2,stale/low urgency,1,0,85841,,,,0,dcfb4b58b7d729bcaa074cb7b6c7ea22d20108cdc0b83601b39151c81502f2d4
thread:issue:85851,issue #85851: Print a warning when user specifies a qconfig for some node and the qconfig is not supported by BackendConfig,stale/low urgency,2,stale/low urgency,1,0,85851,,,,0,b0dfa13e10d00921c1702354f6a5bacf6d9fb3e077f85365b9435ea1e536b734
thread:issue:85889,issue #85889: High occupation on GPU 0 when converting Tensor to multi GPU,stale/low urgency,2,stale/low urgency,1,0,85889,,,,0,9eeca7782864695ded3f7738106d5b21628c358d77554f0d90eb1b5863303dcb
thread:issue:8602,issue #8602: Come with a better strategy for TensorArg (error reporting),stale/low urgency,2,stale/low urgency,1,0,8602,,,,0,095a347e03d3c546bdcdb8128f23a77fcf01f34b400d5c28cbbd48502c935a53
thread:issue:86020,issue #86020: Functorch memory_efficient_fusion gives wrong output batch size,stale/low urgency,2,stale/low urgency,1,0,86020,,,,0,9d7bf4d7a0187853a977ed445fbc0d2b351ad6c0ffc7af1c22628f9fadca4f0c
thread:issue:86060,issue #86060: [MPS] Pin_memory on MPS backend raises exception of setting wrong device storage,stale/low urgency,2,stale/low urgency,1,0,86060,,,,0,0b8d0797c6971dfd71b109a9bad67d3609a2577d29cad8f9fe69dc35f5d0671c
thread:issue:8607,issue #8607: [Caffe2] Wrong prediction with simple FF,stale/low urgency,2,stale/low urgency,1,0,8607,,,,0,945072e346dea2f12a98ecc9b1ee5e742101f38776192c1cb1b35f46372d367f
thread:issue:86074,issue #86074: torch.remainder and torch.fmod produce wrong results,stale/low urgency,2,stale/low urgency,1,0,86074,,,,0,252aafa46673fc11373fddb6758cff3589007a7b79ffd03a7e8d1c86c7e67925
thread:issue:86097,issue #86097: [Activation Checkpointing] Investigate pin_memory for CPU offload,stale/low urgency,2,stale/low urgency,1,0,86097,,,,0,6e235468abaf3189e90645aaf7eaaaa17b89c3032bfae6933fde5b1250b8ba97
thread:issue:86111,issue #86111: When will the torch.sparse module be usable?,stale/low urgency,2,stale/low urgency,1,0,86111,,,,0,30bd5a9af6383ec8cf0ee9a2d8286b24502e23dd17ed779998eb1009beeeaa47
thread:issue:86128,issue #86128: [TorchDispatch] Scalar Only Inputs Gets Matched To Tensor Schema,stale/low urgency,2,stale/low urgency,1,0,86128,,,,0,aded82c526e70554e1f042ad35129b14d0146a502d86597f78e1c1d2b8925c8e
thread:issue:86131,issue #86131: Debuggability++: Share instructions for building exotic CI configurations,stale/low urgency,2,stale/low urgency,1,0,86131,,,,0,cc00b6563fcc7f2061261360912e70ef78137ffeff98ff46b1126c05697b5f95
thread:issue:86192,issue #86192: fmt/src/os.cc: error: unknown type name 'error_code'; did you mean 'std::error_code'?,stale/low urgency,2,stale/low urgency,1,0,86192,,,,0,e9483c87413f9f9ab37b56574addb85c36232169e1a52045a247911316074989
thread:issue:86194,issue #86194: path in WORKSPACE,stale/low urgency,2,stale/low urgency,1,0,86194,,,,0,54e284727f98eab0ca6d2d29c3848ab356816ba2cb3a52407a804b585c97af12
thread:issue:86204,issue #86204: How to perform unstructured interpolation,stale/low urgency,2,stale/low urgency,1,0,86204,,,,0,c8544fd2857f4b691027d3b03eab4ed44ac4fef75f3b8bd9c6f8f5128d7e3611
thread:issue:86225,issue #86225: [Feature] Dispatching PyTorch Distributed Collectives,stale/low urgency,2,stale/low urgency,1,0,86225,,,,0,82a3454a258684210f6c71819c7450ff04cde231aa9ab0056f5a511f0de42bc9
thread:issue:86265,issue #86265: TORCH_WARN is executed just once per set of parameters,stale/low urgency,2,stale/low urgency,1,0,86265,,,,0,dbfe3ec34f342cbb586ab65704b00a64d45fb19b51e8c166ce18c547139d0771
thread:issue:86270,issue #86270: Autocast with BF16 on CPU slows down model more than 2X,stale/low urgency,2,stale/low urgency,1,0,86270,,,,0,746aed37c7398e739e40b21b6b211f17ca5b04871593550b5be84459c0a39370
thread:issue:86271,issue #86271: AUTOGRAD is not working on IOS,stale/low urgency,2,stale/low urgency,1,0,86271,,,,0,20371a48e6d5d6d8381811dfe295c64cff032c0b4a7fb936b425fdf529be1731
thread:issue:86274,issue #86274: TransformerEncoder/TransformerDecoder has same initial parameters for all layers,stale/low urgency,2,stale/low urgency,1,0,86274,,,,0,0853f4b4d43dcebbd39364372d9324ef281c78ab543a161216d2531fa9dec2ad
thread:issue:86298,issue #86298: AOT Autograd Device Partitioning,stale/low urgency,2,stale/low urgency,1,0,86298,,,,0,dd195386773dca09a381ba8485c8a3e920c6bb738f656df856f1d9fa33560168
thread:issue:86315,issue #86315: Feature request: Tests for `int` should be tests for `numbers.Integral`,stale/low urgency,2,stale/low urgency,1,0,86315,,,,0,9257db2213da30dd8fa90fe8a851d5c53cac2a2072aac3fc87ae6f49ba244754
thread:issue:86326,issue #86326: `torch.special.round` doesn't support the same dtypes as `torch.round`,stale/low urgency,2,stale/low urgency,1,0,86326,,,,0,faab115a9afbdccda74bc74b2be359b219ca3a11465c37e4c9f307454d45f521
thread:issue:86339,issue #86339: Add `maximize` support to LBFGS optimizer,stale/low urgency,2,stale/low urgency,1,0,86339,,,,0,f94e10ff3db4144700c40ff4cbc57ff5ba88c5e08f4780ced312f9ba479b37e4
thread:issue:86340,issue #86340: Add complex support for SparseAdam and LBFGS optimizers,stale/low urgency,2,stale/low urgency,1,0,86340,,,,0,d21773af05420e1425c7637dc267de199eab89930066cf16bbddc5c7b8cfa5c0
thread:issue:86356,issue #86356: `conv_transpose` is not similar to `nn.grad.conv_input` when `output_padding` is passed with non-default values.,stale/low urgency,2,stale/low urgency,1,0,86356,,,,0,6f066bb37d60f622c7b82308188c569e4b93a097ce2a114363259a12727ff82b
thread:issue:86381,issue #86381: OpInfo Tests To Validate that All Operators Are Being Tested With Strided Tensors,stale/low urgency,2,stale/low urgency,1,0,86381,,,,0,d8da36502f97f57a6212c94436974a3ab39f8405989e4553dcdefe10c41732f0
thread:issue:86444,issue #86444: Dedicated function for shallow_copy_and_detach,stale/low urgency,2,stale/low urgency,1,0,86444,,,,0,9041ecc33b52757a5f2ee68e275bf70646e347fc4a5a730c0ed55599c1d9ad81
thread:issue:86449,issue #86449: torch.cuda.empty_cache() is not working,stale/low urgency,2,stale/low urgency,1,0,86449,,,,0,2b62de51588f75989416b9375980b391a7b50bfbf86679389c5ac86fd9e24cc5
thread:issue:86465,issue #86465: Wrong results with torch.linalg.inv on batched matrices when using cuda,stale/low urgency,2,stale/low urgency,1,0,86465,,,,0,a7120174be49ae5ba599b85a0c78bf0de94fdb4b3a7b69d24d1f42f83314c231
thread:issue:86467,issue #86467: torch.tensor obj automatically moved to shared memory upon Process launch,stale/low urgency,2,stale/low urgency,1,0,86467,,,,0,05747f772e8c2939ceb86a3b525876a18844999476967f7f36063e50675d654c
thread:issue:8648,issue #8648: Cannot allocate memory Error from operator,stale/low urgency,2,stale/low urgency,1,0,8648,,,,0,1fbaf8b742473d1d8e9315bb3225a49b60fd99dbd12f97f33fa868a00717f2de
thread:issue:86493,issue #86493: `mem_get_info` reserves memory and can not be destroyed / deallocated.,stale/low urgency,2,stale/low urgency,1,0,86493,,,,0,83e12c84acf620a815bd60b7277eafa57bde0bb6981217887100a5fff5e6eff3
thread:issue:86494,issue #86494: Automatic broadcasting for batch addition for sparse tensors,stale/low urgency,2,stale/low urgency,1,0,86494,,,,0,384e4cff1dcc4047179d03124a61290b8a7e686f94cc045d34b05f796e898ca5
thread:issue:8650,issue #8650: Cannot use IterOp at runtime CUDA&CPU,stale/low urgency,2,stale/low urgency,1,0,8650,,,,0,fec3248c97b5f6de698ec0aa5a5e86c99c5be32c47c83e2765ce6ba0dbd27c61
thread:issue:86529,issue #86529: [NvFuser] JIT model with `mul+atan+sgn` will access illegal memory on cuda when computing gradient,stale/low urgency,2,stale/low urgency,1,0,86529,,,,0,394dd437a8658accacefdb9e92ffefddb6fd25cb39db48a3d573fb8e5983f964
thread:issue:86544,issue #86544: Compile failed at allreduce without gloo,stale/low urgency,2,stale/low urgency,1,0,86544,,,,0,c911cc7e5df3ff8f8ff130c26934ab26163f42eaa3e070d32db30729f0b63fa5
thread:issue:86548,issue #86548: About autocast,stale/low urgency,2,stale/low urgency,1,0,86548,,,,0,b4498d51a6452770561325d9b3558b7ab93ad2648c235d17194a1538fb6ec31f
thread:issue:86551,issue #86551: `MultiLabelMarginLoss` will return incorrect values in JIT after the first run on cuda,stale/low urgency,2,stale/low urgency,1,0,86551,,,,0,26a1c71f19ef057ae1125c7d377ec054a452d88871242dd2561daa70cdc55a4c
thread:issue:86554,issue #86554: [NvFuser] would change the output for some inaccurate dtype,stale/low urgency,2,stale/low urgency,1,0,86554,,,,0,7d1da147b1054a1396f2f94986e0198c79bed9548d6793ad34912bfe60ed83cc
thread:issue:86563,issue #86563: [functorch] colab links on functorch 0.2.0 website should be linked to a permalinked version of the colabs,stale/low urgency,2,stale/low urgency,1,0,86563,,,,0,7c2d433af2dab916db0a191f4e35a40c73af586642ec50454f23465664e8e113
thread:issue:86601,issue #86601: NVFuser batch norm with prims: internal assert failure from test suite,stale/low urgency,2,stale/low urgency,1,0,86601,,,,0,bd6a7271cc1a4c4f2e2c2ab4bfdb14497263bcdea0348d8ecc36cb7143a7ee53
thread:issue:86644,"issue #86644: [NvFuser] INTERNAL ASSERT FAIL ""ScalarType should be static for Tensors in fusion for amp optimization""",stale/low urgency,2,stale/low urgency,1,0,86644,,,,0,ada952a208cb1bd254ff8b73e426fae67019a92d53c2e312aaf7e1c0d0e289c4
thread:issue:86660,issue #86660: libtorch make failed,stale/low urgency,2,stale/low urgency,1,0,86660,,,,0,0e5693174ab58fae5e1e8bd168d794d310a6949ece78c04b957efef3e46cbb9b
thread:issue:86676,issue #86676: Pytorch built for Jetson errors if CUDA is not found,stale/low urgency,2,stale/low urgency,1,0,86676,,,,0,d6a4cd426668545ee5a6ea82541cd45f5efa389e04e62d15428f28a92cb89324
thread:issue:86694,issue #86694: Feature request: Deterministic test input generation,stale/low urgency,2,stale/low urgency,1,0,86694,,,,0,9106a0de96364e3e8af200d3924d0ec6789fee502ab977be88fb677f054d2894
thread:issue:86717,issue #86717: NVFuser `FusionRootMappingMultipleBroadcast_CUDA` raises exception on sm_80+,stale/low urgency,2,stale/low urgency,1,0,86717,,,,0,0be556f417e8b8be7b4957cfc2ef8b490fbe6ed9e81dcc8e25140f7ab5119def
thread:issue:86782,issue #86782: Poisson sampling on GPU fails for high rates,stale/low urgency,2,stale/low urgency,1,0,86782,,,,0,67d7c332e4b3e7b534a6953bc074c46cfe49632607e54590b1c0186f19267b45
thread:issue:86798,issue #86798: TF32 conv_transpose2d with groups has bad precision compared to fp32,stale/low urgency,2,stale/low urgency,1,0,86798,,,,0,2e6e4403ef140745cc459bd085eeaf88edec70479a19cfc05b6f89683b4f61f5
thread:issue:86814,issue #86814: Expanding the parameters of `torch.svd_lowrank`,stale/low urgency,2,stale/low urgency,1,0,86814,,,,0,97531d2b4a64371c3af00453e0b03219fcd382f899f65a5efa6d4389b598c781
thread:issue:86819,issue #86819: Could not run select_backward [vmap] [dlrm] [functorch],stale/low urgency,2,stale/low urgency,1,0,86819,,,,0,8104c7b081412a469bda773026254cc1ae9b11ec9fa9e0734406f206965c6a0f
thread:issue:86848,issue #86848: .view(dtype) on a quantized tensor throws SegmentationFault,stale/low urgency,2,stale/low urgency,1,0,86848,,,,0,13657a21643f41444ea0f027b6ad484091830368d36686f42a60807ccf2e4782
thread:issue:86888,issue #86888: Broadcasting add for nested_tensor,stale/low urgency,2,stale/low urgency,1,0,86888,,,,0,0069f61ce286bb007f5a20c365a7f1e89b3fc81c01a9a1852ecc92f60300c1a2
thread:issue:86918,issue #86918: torch.cat on empty tensor is bogus,stale/low urgency,2,stale/low urgency,1,0,86918,,,,0,0862efa1b34805c08111ad38d177a1a466ed174801cd79d41d252adde565f5f9
thread:issue:86965,issue #86965: Upgrade to a newer llvm-openmp version to avoid `/dev/shm` pollution,stale/low urgency,2,stale/low urgency,1,0,86965,,,,0,910bc64b83ad12557d1225e021fefa98f4136657ee62888536acb93fe8e6b9fa
thread:issue:86968,issue #86968: Drop deprecated behavior from NumPy-style `T`,stale/low urgency,2,stale/low urgency,1,0,86968,,,,0,f037cde9020038dbd01334726b231ef33cfcf07b972e8d834804585259d0a789
thread:issue:86989,issue #86989: Error: unknown architecture `armv7-a;' and Error: selected processor does not support `command' in ARM mode,stale/low urgency,2,stale/low urgency,1,0,86989,,,,0,4eae89b5f104e298e64d9fd1a40b1f48cb0a56586ce9a85a01415a4c079d3735
thread:issue:87003,issue #87003: Multiprocessing DataLoader pickles multiprocessing.Queues incorrectly,stale/low urgency,2,stale/low urgency,1,0,87003,,,,0,dbd62630334c5c039a3f27c44edb7345e186a7551b689176458e46e59197760d
thread:issue:87019,issue #87019: Rewrite `narrow_copy_dense_cpu_out` using `copy_` and `narrow`,stale/low urgency,2,stale/low urgency,1,0,87019,,,,0,7a0c1cb8844a54e1812c70ae65ae66cf541d3dffa9581380736d7a0ca64fab53
thread:issue:87031,issue #87031: Improve Readability of error(s) when provided unexpected keyword arguments.,stale/low urgency,2,stale/low urgency,1,0,87031,,,,0,ee40f858528e0fba72a30f9afffb93e465953ce95fe45e7e3b5f03757d948b09
thread:issue:87041,"issue #87041: Segmentation fault: 11 when running ""import torch"" on Mac OS X",stale/low urgency,2,stale/low urgency,1,0,87041,,,,0,7e64858b46f413eb0574f1a6ac88eb314eab2b1dd1bab4de6680ed3afc47a77c
thread:issue:87056,issue #87056: Pipe conveys inconsistent value in GPU env,stale/low urgency,2,stale/low urgency,1,0,87056,,,,0,def75a014c813348d8908866149d9f4bb14c9f00d00724e40c476754a347713d
thread:issue:87076,issue #87076: Nandense layer for missing values,stale/low urgency,2,stale/low urgency,1,0,87076,,,,0,b69047b4dd6efe685eeb866db241f1a38bfc1b80178c62ad39b5a279ce7069cf
thread:issue:87080,issue #87080: cppextension host compiler check ignores executable symbolic link in CUDA bin directory,stale/low urgency,2,stale/low urgency,1,0,87080,,,,0,af3111683b0c60bf69eb035e196a69d37ac038cd294479c7047499b8f9d263b1
thread:issue:87087,issue #87087: Failing periodic tests: test_dense_mask_index_cpu (__main__.CpuTests) & est_expanded_reduction_cpu (__main__.CpuTests),stale/low urgency,2,stale/low urgency,1,0,87087,,,,0,9a2474548c61e5ffeae1d2949acd5180db04d49d26c0e8abff6c9e536e978e9d
thread:issue:87090,issue #87090: MPS memory usage significantly higher than on CPU,stale/low urgency,2,stale/low urgency,1,0,87090,,,,0,4afc2df9a7eff7cd2d8a1b34a382dfa947cd2e48f7420e6cfef81270887e52fd
thread:issue:87126,issue #87126: Bug in Histogram Observer Implementation,stale/low urgency,2,stale/low urgency,1,0,87126,,,,0,616dfc9c5271e3c9235ec4cf1ada14e4f5d8c5f4a4e93710903c0b1ee6f9b638
thread:issue:87129,issue #87129: [MPS] sum on a size=1 dim is ~5x slower than squeeze,stale/low urgency,2,stale/low urgency,1,0,87129,,,,0,f847b627f2a3c35c07ec39cf854a52fd0b24c2ec3d6ae33381fdfe255d975589
thread:issue:87159,"issue #87159: torch/csrc/utils/python_arg_parser.h:424:94: error: format ‘%ld’ expects argument of type ‘long int’, but argument 7 has type ‘int’",stale/low urgency,2,stale/low urgency,1,0,87159,,,,0,ce0b59c61c41b6e5928fc2d66a789084b164f180b4dce96e9063d1c84e9c59cd
thread:issue:87163,issue #87163: New APIs for cuda graph inspection and manipulation,stale/low urgency,2,stale/low urgency,1,0,87163,,,,0,70350f786a1dcc34d934205d7e759617b9bb44037a85cc7fa21c7ac0278a09f7
thread:issue:87170,issue #87170: torch.linalg.cond gives inconsistent results on CPU/CUDA,stale/low urgency,2,stale/low urgency,1,0,87170,,,,0,f2e69c8330fdaf0870f037597741d783dcca926f7e7ec1b14792c6f7f56fa8f9
thread:issue:87178,issue #87178: The autogenerated out variants via `autogen:` do not check that the dtype of the `out` kwarg via `canCast`.,stale/low urgency,2,stale/low urgency,1,0,87178,,,,0,0bd6bd84f1a32f003e0f793331ebfa70e32b3b379ed6585f7ec846c54c2ba87c
thread:issue:87196,issue #87196: Missing `docker` directory in `tools/`,stale/low urgency,2,stale/low urgency,1,0,87196,,,,0,83252359f878ca471b420e532abfacad8f60af93fdf4c6a9e36449ddb8977065
thread:issue:87222,issue #87222: `custom_jvp` and `custom_vjp`,stale/low urgency,2,stale/low urgency,1,0,87222,,,,0,50ee73d4340856372074cb4f6b2ac8ae44855ecddb1e21e680600de4555536b1
thread:issue:87267,issue #87267: The installation commands given on the pytorch website will not install properly,stale/low urgency,2,stale/low urgency,1,0,87267,,,,0,93bb9f3f03998ceb0074cb17a81881514a5f221303f6e744989798a99dde221f
thread:issue:87276,issue #87276: torch.save throws ValueError: ctypes objects containing pointers cannot be pickled,stale/low urgency,2,stale/low urgency,1,0,87276,,,,0,192c5823510b0521ad187b6068b9490acfabafa834b0a6f05a2541c7eb5fb64e
thread:issue:8730,issue #8730: [caffe2] AffineChannelOp,stale/low urgency,2,stale/low urgency,1,0,8730,,,,0,9e6c5fde89dd403bd1bf642fe1ab33c2f9e27d28fb4d0ee81b09b50c764e993b
thread:issue:87350,issue #87350: OpenCL 3.0 support: support every GPU on earth through rusticl,stale/low urgency,2,stale/low urgency,1,0,87350,,,,0,e8d262993793cba5067f3df42531ee6309991168212e63a54d5e5a592c6db458
thread:issue:87352,issue #87352: RuntimeError: Tensors of type TensorImpl do not have numel,stale/low urgency,2,stale/low urgency,1,0,87352,,,,0,ef825ac87bcb4dc9430b2326e8a8fd8ec4091b53d2a393843af0f1079f191b19
thread:issue:87353,issue #87353: Speed of torch.istft,stale/low urgency,2,stale/low urgency,1,0,87353,,,,0,d495174b2dc6760c5268b23dd056326f9133ebaf4a2991a6280f0d261407b11b
thread:issue:87388,issue #87388: Document dist.new_subgroups,stale/low urgency,2,stale/low urgency,1,0,87388,,,,0,6861a3ffd74b4b4dab2e2748f795544c3840faab7902067cbb66fa13cc1569ae
thread:issue:87389,issue #87389: Installing PyTorch with BUILD_SPLIT_CUDA=ON and CUDNN fails on linker error,stale/low urgency,2,stale/low urgency,1,0,87389,,,,0,2c9220aba328b7a4a209d7f8a7f24034d0b33d499ce185690439ccc4b63d2773
thread:issue:8741,issue #8741: [Feature Request] Add to() method for optimizers/schedulers,stale/low urgency,2,stale/low urgency,1,0,8741,,,,0,660a86d76fafb3883bad7ca4f5cdd93171f4e32738b7867180f579e8777f1905
thread:issue:87433,issue #87433: index_select() applied in sparse tensor can't backprop,stale/low urgency,2,stale/low urgency,1,0,87433,,,,0,65a1f79f03f9ca1bfe89b68ec3d3ec1c3bbb334504efb9ed0db7c0c3550a5cdc
thread:issue:87449,issue #87449: pytorch could not build from source with cudnn 8.0.5,stale/low urgency,2,stale/low urgency,1,0,87449,,,,0,39850995a7acbdf325fe582421d79d421fd8875ac854eae1c598d7ce9c9f9404
thread:issue:87451,issue #87451: Some operations do not keep `channels_last` memory format which yields accuracy drop,stale/low urgency,2,stale/low urgency,1,0,87451,,,,0,657c92b7d6d455f91e1d748d1f1bfaf982f08927b6653ece9021dfc7368b8cc5
thread:issue:87468,issue #87468: Categorical fails simplex validation after its own normalisation on CUDA,stale/low urgency,2,stale/low urgency,1,0,87468,,,,0,c93b0cf1c36b28db2a44f4831377686373409d5f41d5612d026a536c4c8b15df
thread:issue:87494,issue #87494: Missing docstring for resize_as,stale/low urgency,2,stale/low urgency,1,0,87494,,,,0,e91ae536c53db269d330650c3203d66f81c2f56cc492294cfe1c24faa394c252
thread:issue:87501,issue #87501: Synchronize domain builds to be executed after core build have completed,stale/low urgency,2,stale/low urgency,1,0,87501,,,,0,a0261fe7f70b3b0390af185e17b4252476b045f1942a6cdcf3191997ee98c5d2
thread:issue:87514,issue #87514: invalid_arguments.cpp is busted,stale/low urgency,2,stale/low urgency,1,0,87514,,,,0,56ddf0a7250064ef01151fb379663887135bd77471f6de6d395b9a386c3f7559
thread:issue:87539,issue #87539: RAM leak when copying tensor from cpu to cuda,stale/low urgency,2,stale/low urgency,1,0,87539,,,,0,a8f18b7f16c4455e6969879121bbdc878c14cf1cf9ebc08bc78d058f637c4b52
thread:issue:87551,issue #87551: `bmm` will return wrong result on cpu with in-place,stale/low urgency,2,stale/low urgency,1,0,87551,,,,0,36fa9120fac081c8adfa64660195e7460b8ac519decc7bc3f70ef52adcb868b8
thread:issue:87555,issue #87555: Improve `c10d::ReduceOp` & `torch.distributed.distributed_c10d.ReduceOp`,stale/low urgency,2,stale/low urgency,1,0,87555,,,,0,8b2454c7be2353116dabe024ac6fef920473d2bc266aa47af9eebab3a7195b46
thread:issue:87559,issue #87559: diagonal of Jacobian matrix,stale/low urgency,2,stale/low urgency,1,0,87559,,,,0,4cea631f6eb5a7fbd6d4cccd13c3a0c99de3e1b1fbdae4558888e6074530d5c5
thread:issue:87571,issue #87571: Testing insufficient to catch incorrect dispatch key for bernoulli.p re functionalization,stale/low urgency,2,stale/low urgency,1,0,87571,,,,0,ec228777d95f49b1285af93cf071e5f0a5b40d92abe12d3f2c5960256f11a384
thread:issue:87589,"issue #87589: #error ""Expected GLOO_USE_CUDA to be defined""",stale/low urgency,2,stale/low urgency,1,0,87589,,,,0,9db6e7a5416bdd441aac85cf139a334bb1313a386ef118bf7587dfaa6f885a1d
thread:issue:87634,issue #87634: Add tests for ProcessGroup cpp extensions,stale/low urgency,2,stale/low urgency,1,0,87634,,,,0,cec8373d21a4fc2b662acf1fa1d1ccef34b69a7844164c30c026174920489ded
thread:issue:87642,issue #87642: `libtorch_cpu.so` is exposing some LLVM symbols,stale/low urgency,2,stale/low urgency,1,0,87642,,,,0,ca5cc907ec7d7fca38554fda643eeaed1474867c214f06553967e28865bba8b1
thread:issue:87696,issue #87696: Enable PostLocalSGDOptimizer on CUDA tensors,stale/low urgency,2,stale/low urgency,1,0,87696,,,,0,5fd0bc800019bfbeffcde9b4373cd0dc405faa5fd648ad12460e17a7031eaaac
thread:issue:87701,issue #87701: pytorch/pytorch cpu official Docker images,stale/low urgency,2,stale/low urgency,1,0,87701,,,,0,774861307d0735c74c8369825d7ea288d8cc2d338b47f809f5d803d7f3cfaea8
thread:issue:87706,issue #87706: C++ Adagrad optimizer doesn't initialize parameter state,stale/low urgency,2,stale/low urgency,1,0,87706,,,,0,08f803b1e38e03c091712e599a453439ba8b11ab878ed97d9bf77b2b6b9c4157
thread:issue:87755,issue #87755: Autograd precision for CONV + BN between pytorch version 1.11.0 and 1.12.1,stale/low urgency,2,stale/low urgency,1,0,87755,,,,0,27cd7cda9a62c8dc1866543cf1f49cf7c71d00e486e216fafdbdcedb67c50d2e
thread:issue:87777,issue #87777: Improvements to fuse optimization,stale/low urgency,2,stale/low urgency,1,0,87777,,,,0,5d22f4dcba2bbf5fde1c911d96cb0cd86d18020497b0ba02892f1dca3466a95b
thread:issue:87782,issue #87782: Libtorch windows binaries publishing,stale/low urgency,2,stale/low urgency,1,0,87782,,,,0,252a421b059c469cde628b4bf39a9684b1562423feb1f5f9b563b389392dd996
thread:issue:87794,issue #87794: CUDA error: operation not permitted when stream is capturing (2 GPUs),stale/low urgency,2,stale/low urgency,1,0,87794,,,,0,4e6dcc36cb5333ea7285a46035a57b5b5e5e52f560952c6e50e10f9edfab5ba6
thread:issue:87841,issue #87841: `max_unpool3d` will trigger an assertion fail under compute sanitizer,stale/low urgency,2,stale/low urgency,1,0,87841,,,,0,59931aa228208d2dac65432873329d4d204099bea07789f885f91fd2b254d025
thread:issue:87861,issue #87861: RuntimeError: unable to mmap 29764 bytes from file </torch_10182_3020184674_63991>: Cannot allocate memory (12),stale/low urgency,2,stale/low urgency,1,0,87861,,,,0,7101f54c525c59beab5804220e1b0a3bd756dfc377957e1e444504a270edffa2
thread:issue:87902,issue #87902: Permute,stale/low urgency,2,stale/low urgency,1,0,87902,,,,0,99c67e5b24eb0e2878a97f48f43bde87b5494ff276d3a9c66df33085f2bff3bd
thread:issue:87955,issue #87955: AttributeError: '_MultiProcessingDataLoaderIter' object has no attribute 'next',stale/low urgency,2,stale/low urgency,1,0,87955,,,,0,d8f1f2deb3eddc758f128b9d3ab0d5fb6fd88dc5e4039e19e4037946794cdcf2
thread:issue:87956,issue #87956: Autograd doesn't stop executing backward graph early enough in situations involving set_,stale/low urgency,2,stale/low urgency,1,0,87956,,,,0,c18576f02106bf0ccbf1f73886b741eb8437203cd894f5287a421c6e289669f0
thread:issue:87957,issue #87957: VS2022Preview ParallelCommon.cpp.obj : fatal error LNK1161: invalid export specification,stale/low urgency,2,stale/low urgency,1,0,87957,,,,0,6ed4da2f47c89e634411ece703d8ef2fc461d1fbbeb9e6f908a0d10e328ff785
thread:issue:87979,issue #87979: amp with `bf16`: backward happens in `f16` when using `@torch.cuda.amp.custom_bwd`,stale/low urgency,2,stale/low urgency,1,0,87979,,,,0,44c98a8b938d2b78fee6229045a90a0add89dc609923c872075885fe6901984e
thread:issue:87992,issue #87992: torch.rand(...) is not consistent for large shape dimensions across GPUs (with the same random seed),stale/low urgency,2,stale/low urgency,1,0,87992,,,,0,b1969c04d6d2e37847380073ddc54847f37528b81e842991b3f95e53cb587065
thread:issue:88004,issue #88004: link error happen when intergrate libtorch to other tool,stale/low urgency,2,stale/low urgency,1,0,88004,,,,0,06520ebfc0976d37a168c6885cdd30324d25c17647aab3d9ae24d63b660d7c2b
thread:issue:88027,"issue #88027: getting error error: namespace ""cub"" has no member ""Debug"" when try to build v1.8.2 with CUDA 11.6",stale/low urgency,2,stale/low urgency,1,0,88027,,,,0,e2c71f2fc855262ffef40df5b4fffaca550d2bfe423a171b8fe0d72f51be4213
thread:issue:88053,issue #88053: Different behaviour in sparse matmul,stale/low urgency,2,stale/low urgency,1,0,88053,,,,0,ab8fa2f27ced68b7369a5bf088a2dafa408c5239296661fb27704446dbeb4a71
thread:issue:88062,issue #88062: Cannot import `traverse_dps` from torch.data.utils.graph,stale/low urgency,2,stale/low urgency,1,0,88062,,,,0,a647567d180651f43b85287caeef9335bf52c127f1bb813d306e6c9fa79edf36
thread:issue:88081,issue #88081: out of memory with pytorch version after 1.8.1,stale/low urgency,2,stale/low urgency,1,0,88081,,,,0,e36dcf1ce0d01fe4fbd0b3dd0f505c240be525f0166eb69b5f553595e0632f58
thread:issue:88136,issue #88136: Placing LSTM model on bfloat16 on GPU causes error,stale/low urgency,2,stale/low urgency,1,0,88136,,,,0,c638ddef48e570abfb62abd05bf889a15eb8cc2fc547d117715eb5ce8acf9b72
thread:issue:88142,issue #88142: Refactor `torch.return_types.topk` to behave like a `namedtuple` or a `dict`,stale/low urgency,2,stale/low urgency,1,0,88142,,,,0,1ad5b66f64b864cef8eda8e795df052238f119927f9b3ee20d830a4d7b47322a
thread:issue:88147,issue #88147: Conv2d is not deterministic when input tensor has different strides,stale/low urgency,2,stale/low urgency,1,0,88147,,,,0,4e52c31e7348f745038e332b566c809ea5dabc0130db450470c08ab947048448
thread:issue:88185,issue #88185: `torch.mm` Trigger RuntimeError with UndefinedBehaviorSanitizer,stale/low urgency,2,stale/low urgency,1,0,88185,,,,0,a63d929708486dd2f7e90dbeb407a3c05c6face48d2e05a8c7db67f00987f8b1
thread:issue:88189,issue #88189: Moving tensor to GPU by .cuda() gets stucked when AMD Secure Encripted Virtualization (SEV) is activated,stale/low urgency,2,stale/low urgency,1,0,88189,,,,0,e454f6fd1cd3c07d6114e5964b11d4759bbed75ce441009aea8e60dd59b96bb3
thread:issue:88191,issue #88191: `torch.nn.RReLU` not reporting `lower > upper` on CUDA,stale/low urgency,2,stale/low urgency,1,0,88191,,,,0,59d8437547644847f158d108d43fcdc5b2cf6a735fded6b13cff2b9e431e51ed
thread:issue:88192,issue #88192: [docs] torch.is_neg/torch.Tensor.is_neg not documented,stale/low urgency,2,stale/low urgency,1,0,88192,,,,0,82745016b7927cc00a416714f3561af0c48a7e4e2f12fd978ec23fb3ea3811f9
thread:issue:88194,issue #88194: Add a config option to raise errors instead of warnings in nvFuser integration,stale/low urgency,2,stale/low urgency,1,0,88194,,,,0,c09e517f862a7fe90e643f6eb262bce8a73c72d413313c457b983c4897f72783
thread:issue:8821,issue #8821: Storages still use legacy printing,stale/low urgency,2,stale/low urgency,1,0,8821,,,,0,52d62ccca9aeb80daf1a879c78603822643aadd03bd16f4a2e81c5bf8f1f2a9f
thread:issue:88227,issue #88227: Enable `torch.topk` to support `stable` flag,stale/low urgency,2,stale/low urgency,1,0,88227,,,,0,356b4cad6900e06eb06dbce4bdb82f214b163e4c5b32c471c47ae3602ece263c
thread:issue:88265,issue #88265: build: failure when upgrade oneTBB to 2021.7.0,stale/low urgency,2,stale/low urgency,1,0,88265,,,,0,3470f824ac07d6707b7af15047346bb5dc8a1c581ed3a934b36c9fd2b135db51
thread:issue:88304,issue #88304: AttributeError: module 'tensorboard.compat.tensorflow_stub.io.gfile' has no attribute 'MakeDirs',stale/low urgency,2,stale/low urgency,1,0,88304,,,,0,4544d7a26ed443dfcd2eaa84815c0e164df5b85571279ffcdca2674082d41c14
thread:issue:88309,issue #88309: ASAN shard 4 started to OOM after unrelated commit,stale/low urgency,2,stale/low urgency,1,0,88309,,,,0,88d5e45ddbd991346c14a404c4a07da9d7eb9d4ce0a5125a8bd1e2225861f3af
thread:issue:88320,issue #88320: `torch.load()` cannot load data saved at non-zero position in a file (`failed finding central directory`),stale/low urgency,2,stale/low urgency,1,0,88320,,,,0,c78018f7d8b43967235d0ac52df8feb2befa1b41de282cbd8699ac1e374efd91
thread:issue:88327,issue #88327: MSE documentation is weak,stale/low urgency,2,stale/low urgency,1,0,88327,,,,0,a3eba4b66e6308f7e647452e3369d8a12612f43581622d421f505a96e71974b4
thread:issue:88380,issue #88380: cuDNN error (CUDNN_STATUS_NOT_SUPPORTED) for torch.nn.functional.grid_sample(),stale/low urgency,2,stale/low urgency,1,0,88380,,,,0,a0f2b80f6efc4db9101b8a4cb18e79b10558505f9329f9b1b5c1638880887731
thread:issue:88389,issue #88389: Whether to support libtorch source code compilation of C++11 ？,stale/low urgency,2,stale/low urgency,1,0,88389,,,,0,efec6476bde37eddf1595da728bb59d0d4bfd86e7905d0e7760b1f237758fa1b
thread:issue:88408,issue #88408: Unit test with `--subprocess` command doesn't respect the `-k` filter flag and runs all available sub tests,stale/low urgency,2,stale/low urgency,1,0,88408,,,,0,1623a9b7691947cb1d79fda5f567c64b57e5a504bb2f1dc89319dafa1a060b07
thread:issue:88410,issue #88410: benchmark cache persist,stale/low urgency,2,stale/low urgency,1,0,88410,,,,0,80f437a7753c350520d07aa5120654dbefab7fe715d30406308c160e652f15a0
thread:issue:88423,issue #88423: view_as_real and split_with_sizes links in Tensor Views docs are broken,stale/low urgency,2,stale/low urgency,1,0,88423,,,,0,a410e29b1c34418f0a87c690fa3af52ba923da5845bba928ea267d0489ecec3f
thread:issue:88474,issue #88474: Mixed precision training fails due to NaN in batch norm running_mean,stale/low urgency,2,stale/low urgency,1,0,88474,,,,0,04dcc9719c5e7ec9e175f27100e2933394ff10b710d924cf318868d7e423f5ef
thread:issue:88505,issue #88505: quantization: error message when using `convert_fx` on a model on cuda should be better,stale/low urgency,2,stale/low urgency,1,0,88505,,,,0,83654186cf3dcd9c9b72a2e1bcad3b0e53d033aa0f4b3ba02bca36b236f51595
thread:issue:88563,issue #88563: `nn.functional.embedding_bag` Trigger out-of-bound Read under Compute Sanitizer,stale/low urgency,2,stale/low urgency,1,0,88563,,,,0,8d1f491e0c20fb485d9560d2e5b667eaea5f15b878123cf11384935bcb01e583
thread:issue:88565,issue #88565: What causes CPU to degrade when I load the weight with torch.hub.load(),stale/low urgency,2,stale/low urgency,1,0,88565,,,,0,f11acbc703afbc6b3bb68cb816b302b29bea021e4f033a6cf3b4580d0ba1ad86
thread:issue:88579,issue #88579: [Quant] Validate FixedQParams observers in eager mode,stale/low urgency,2,stale/low urgency,1,0,88579,,,,0,9c39630efbd1fd8eff857da693d7add2fd1b582cb57680b1bc1a65aac384678b
thread:issue:88591,issue #88591: `pack_padded_sequence` not compatible with deterministic mode it calls `torch.scatter`,stale/low urgency,2,stale/low urgency,1,0,88591,,,,0,359642873d2be16f27507a10e627d68e9c9f3cecf0fa1174c9dea14664b8e1c1
thread:issue:88598,"issue #88598: Despite having aten::diag_embed.out, torch.diag_embed doesn't support out= argument",stale/low urgency,2,stale/low urgency,1,0,88598,,,,0,b58898241c2c56ae1c3561c938469eaae6610b12718e215b4af8f0e321278ace
thread:issue:88612,issue #88612: Bernoulli uses legacy contiguous memory format,stale/low urgency,2,stale/low urgency,1,0,88612,,,,0,f276e0b2aa6559b7115925bba291c5f6d8fbd1990b61a0868cbfc4253bc27a7b
thread:issue:88614,issue #88614: The libtorch test TestScalarTensor.TestScalarTensorMPS fails on Apple Silicon,stale/low urgency,2,stale/low urgency,1,0,88614,,,,0,0a7e2f9abaa7209ca0b793d4751f14b9f6e49afc27080501cb28049fdcceba40
thread:issue:88616,issue #88616: The libtorch test SequentialTest.ModuleForwardMethodOptionalArg fails on Apple Silicon,stale/low urgency,2,stale/low urgency,1,0,88616,,,,0,69d28f357abd7ba0a706f2b07b66f4584356c407d21ae7fd2d3835e4d888c55f
thread:issue:88621,"issue #88621: [FSDP] FSDP produces different gradient norms vs DDP, and w/ grad norm clipping creates different training results",stale/low urgency,2,stale/low urgency,1,0,88621,,,,0,2934dbd1d8f7e9d027b415621a19056b63728b54491d8c66df269ab32ad14304
thread:issue:88639,issue #88639: CUDA unknown error after suspend during debugging,stale/low urgency,2,stale/low urgency,1,0,88639,,,,0,d550c11ca6f6e0575f0090a0038763f5270dfef8d7844d36dbb6101b4762bbd5
thread:issue:88643,issue #88643: pack_sequence() always fail after set_default_tensor_type to CUDA,stale/low urgency,2,stale/low urgency,1,0,88643,,,,0,6e68b33b2b0d4209574e5cc2923e3d5f284470e5e9ed7e730f7af439b0617ac8
thread:issue:88691,issue #88691: forward AD for _euclidean_dist,stale/low urgency,2,stale/low urgency,1,0,88691,,,,0,e31f51c5d90bcd3a0947210ab0c71bc2b2331bfe7e6e15eadf3e01013707ed7d
thread:issue:8871,issue #8871: Use target_compile_options to set warning flags,stale/low urgency,2,stale/low urgency,1,0,8871,,,,0,194a1702447e0560bf2b2409222204d6a13bd695e349ee1c059e3c18454654d0
thread:issue:8874,issue #8874: Issues with dynamically created grad_fn for views,stale/low urgency,2,stale/low urgency,1,0,8874,,,,0,1fc55f6baa78e3411dff3a930df8dffc4e1416290b0f47bff54d070db326f4b5
thread:issue:88775,issue #88775: Batched Random Number Generators,stale/low urgency,2,stale/low urgency,1,0,88775,,,,0,f7e566e581d887a5c43667c1dcdba0e30f1fce860324c2125ccf7d69fc574d9a
thread:issue:88800,issue #88800: Quantization error between fake-quantized model and quantized model using the new observer,stale/low urgency,2,stale/low urgency,1,0,88800,,,,0,c08ba96d3244f3cbff07b333443ff4aea0e7bd309cbb34a509c3863446e33142
thread:issue:88802,issue #88802: ImportError: libcupti.so.11.2: cannot open shared object file: No such file or directory,stale/low urgency,2,stale/low urgency,1,0,88802,,,,0,c5f8573990e4a22d5b6a7b7952c4645edd1164fc3d340c3951d542d6e7051043
thread:issue:88805,issue #88805: [feature request] Get/set fastmath CPU bit (and some other FPU flags?),stale/low urgency,2,stale/low urgency,1,0,88805,,,,0,84e65310bad43301d0c86f9fafa6a090520190d049191cb2244fd5e685855e4a
thread:issue:88810,issue #88810: Return the attention weights using the Transformer Encoder class.,stale/low urgency,2,stale/low urgency,1,0,88810,,,,0,94921aa63ec7d252259c1644d2e5bfea9a775bbec8f06944364e8842e20e44d8
thread:issue:8886,issue #8886: [Caffe2] Dose caffe2 support the function similar to itersize of caffe,stale/low urgency,2,stale/low urgency,1,0,8886,,,,0,20a48118eb02d7b6058e1f99c2df435503d916effbeb545fedf3244cd23b950c
thread:issue:88880,issue #88880: Add alphatensor support for faster matrix multiplication?,stale/low urgency,2,stale/low urgency,1,0,88880,,,,0,86877072bac8b619fd511a41c79c257bd5372dfaeadff0529e465efb03850cc8
thread:issue:88883,issue #88883: Investigate why `test_aot_autograd_symbolic_exhaustive_masked_median_cpu_float32` is flaky,stale/low urgency,2,stale/low urgency,1,0,88883,,,,0,5049b459916f351f44409dfa6ad48480140aca0c2d2b22c86e7905f75629201c
thread:issue:88893,issue #88893: RuntimeError: derivative for aten::mps_max_pool2d_backward is not implemented,stale/low urgency,2,stale/low urgency,1,0,88893,,,,0,52da002e75ea8e840fc539dad499399c12471951f6af2347f865523a0322f614
thread:issue:88902,issue #88902: MPS test_numpy_ref_mps_nn_functional_group_norm_mps_float32 is flaky?,stale/low urgency,2,stale/low urgency,1,0,88902,,,,0,ba55245dd6ff76af1bcaf72c3ec0ab989a6eda859720054832864ab43bde559c
thread:issue:88931,issue #88931: INTERNAL ASSERT FAILED. Missing scalar type infromation.,stale/low urgency,2,stale/low urgency,1,0,88931,,,,0,dfeff28290482d51f02c46f8017166262f4b79a90f18d18abbc08bc216ba1ad2
thread:issue:88941,issue #88941: `torch.linalg.lstsq` Trigger RuntimeError under UndefinedBehaviorSanitizer,stale/low urgency,2,stale/low urgency,1,0,88941,,,,0,d9d7864b799a9039a00b046b1fd5d3ffc483cccdeea1559d680d864e4e0dda0b
thread:issue:88942,issue #88942: `torch.svd_lowrank` Trigger RuntimeError under UndefinedBehaviorSanitizer,stale/low urgency,2,stale/low urgency,1,0,88942,,,,0,17f3058056c08465a5c9ff43a25079f7171b7c2a143bc0ffc92f8bbf90fdb43d
thread:issue:88943,issue #88943: `torch.vander` Trigger RuntimeError with UndefinedBehaviorSanitizer,stale/low urgency,2,stale/low urgency,1,0,88943,,,,0,03f1f905a40d0654c9d473bd3763f33de04a324dd3b1fe7349f763cf5f9000af
thread:issue:88944,issue #88944: `torch.topk` Trigger RuntimError under UndefinedBehaviorSanitizer,stale/low urgency,2,stale/low urgency,1,0,88944,,,,0,dde37ce142f9d28e31fe5ab66be6e8b388ef0b24eb0bdc82f9780c60098ce53c
thread:issue:88945,issue #88945: `torch.linalg.eigvals` Trigger RuntimeError under UndefinedBehaviorSanitizer,stale/low urgency,2,stale/low urgency,1,0,88945,,,,0,4724216fd47da185c5eebbbdef372baf36b61e9ea843d2fe6156057a6f527b41
thread:issue:88947,issue #88947: `torch.Tensor.msort` Trigger RuntimeError under UndefinedBehaviorSanitizer,stale/low urgency,2,stale/low urgency,1,0,88947,,,,0,97b8e03e4516f708ff7bd7fdbf60f5bb888f14fec5fb1a519f6132415522bacb
thread:issue:88949,issue #88949: `torch.set_rng_state` Trigger RuntimeError under UndefinedBehaviorSanitizer,stale/low urgency,2,stale/low urgency,1,0,88949,,,,0,a480663456b046f80ac972a2e0f2556f61edaa7304188a76e86f84891f76b6ee
thread:issue:88950,issue #88950: `torch.nn.functional.embedding_bag` Trigger RuntimeError under UndefinedBehaviorSanitizer,stale/low urgency,2,stale/low urgency,1,0,88950,,,,0,30ba784da781047a16d7969d35e34db7a4b4399d56acefb6c849c7d20773f480
thread:issue:88968,issue #88968: M1 runner i-090e1df32b6f48a20 run out of disk space,stale/low urgency,2,stale/low urgency,1,0,88968,,,,0,b72b20ff49e791548bb209ebefe4f1f462c64080388512e019e9b1fd7ff99051
thread:issue:88991,issue #88991: nvprims.native_batch_norm doesn't support fake tensor inputs,stale/low urgency,2,stale/low urgency,1,0,88991,,,,0,380a3e9884e2e2bd7e51ed50310141d95ec7ee7d458430f0d37efb366f75efd9
thread:issue:88992,issue #88992: Incorrect version in the instructions on official website,stale/low urgency,2,stale/low urgency,1,0,88992,,,,0,ff19370d284bcc0531c294ddfb9a44c9a7afd9b7b6a0c137cbf0300f17ee9849
thread:issue:8902,issue #8902: [Caffe 2] Does caffe2 support Galaxy 5S with Android 6?,stale/low urgency,2,stale/low urgency,1,0,8902,,,,0,71925fbeb50c2c93cab8236f43e77a046866db866d6634883cde4a459a6b9fda
thread:issue:89051,issue #89051: caffe2_interface_library CMake macro prevents linking to LibTorch as a transitive dependency,stale/low urgency,2,stale/low urgency,1,0,89051,,,,0,402a84186a616deccb6a4a2858ec0abc634fc7d2368f489657dd87439e5d2f0c
thread:issue:89060,issue #89060: Extend test_proxy_tensor tests to support ops test non floating point types,stale/low urgency,2,stale/low urgency,1,0,89060,,,,0,9b45786eaf4d439bb43aed1fb44297a5203915b1ac37120aa73efcec340d5dc0
thread:issue:89068,issue #89068: view_copy out= does not reshape zero element tensors,stale/low urgency,2,stale/low urgency,1,0,89068,,,,0,b4f4d83b990cba5e5781c06b7ff6f65a65e38aec8146512e4fd77b217e2f4c27
thread:issue:89080,issue #89080: Unable to backprop through dense weighted sum of sparse_coo_tensors,stale/low urgency,2,stale/low urgency,1,0,89080,,,,0,1f96ff8e0c12ff12dc7a6cb8a0c88ca24fedd1c1bedcfc04b04afb6fcd658b4f
thread:issue:8913,issue #8913: Multiprocessing Self Test Error,stale/low urgency,2,stale/low urgency,1,0,8913,,,,0,c13fbc400a02717844c9c8c39145400db41a8d5bf330fe1680d6ae0c34421c9f
thread:issue:89158,issue #89158: Support disallowing calls to certain instance methods in TorchDynamo,stale/low urgency,2,stale/low urgency,1,0,89158,,,,0,c3c7900cb4e46b423d2c9a74c0a34f1866d93b3244341c605e9bea4d26393fe3
thread:issue:8921,issue #8921: [feature request] More methods for PackedSequence,stale/low urgency,2,stale/low urgency,1,0,8921,,,,0,2240eb07520c6382e1c4875bfae774de95232246cf9d0bf1407fe2730fa286ae
thread:issue:89218,"issue #89218: `torch.nn.LayerNorm` Abort with ""invalid device ordinal"" Error",stale/low urgency,2,stale/low urgency,1,0,89218,,,,0,d7c804202923b07ea281d60c3c75c20def36f8eac6bcfa87f3bf08c80e71d21e
thread:issue:89245,issue #89245: sm_80 support,stale/low urgency,2,stale/low urgency,1,0,89245,,,,0,2dfb95392da95248614754df456b560134b5db0819485abc3ba9456d95adcb67
thread:issue:89254,"issue #89254: `torch.nn.ReplicationPad2D` Report ""invalid configuration argument"" Error under Compute Sanitizer",stale/low urgency,2,stale/low urgency,1,0,89254,,,,0,a09dd0adcd117c6ecd01e6a4b0d49fac757114206e96c75146afeab1d13b7734
thread:issue:89275,"issue #89275: torch.addcdiv: input, tensor1, and tensor2 parameters should be of the same type",stale/low urgency,2,stale/low urgency,1,0,89275,,,,0,6958c1338fd4282eb6d23f79a02603b94a01958a03690cf7413e037c3f0a9614
thread:issue:89293,issue #89293: fbgemm_avx512 build failure,stale/low urgency,2,stale/low urgency,1,0,89293,,,,0,66a14bbd595c0f0a32aae27af18b9e72130d42b09a26103ca647410f3e0bb7bd
thread:issue:89301,issue #89301: Meta implementation for copy_ is wrong,stale/low urgency,2,stale/low urgency,1,0,89301,,,,0,d7d3e647e55e50d883d092f05f78f730a6ad0a4a19f04ebd506280343d230daa
thread:issue:89303,issue #89303: code sharing for fundamental ops in quantization,stale/low urgency,2,stale/low urgency,1,0,89303,,,,0,9b22366caa9f33ae2b03ec12d5d72e8f2fe383622dc1af3057251134d746d6eb
thread:issue:89320,issue #89320: `masked_fill` with `FloatTensor` mask will never mask but fails silently.,stale/low urgency,2,stale/low urgency,1,0,89320,,,,0,22997325c3a820e10c45dfd8ab14d22d0ee6fa0a2afdb6edfa023a6622526063
thread:issue:89336,"issue #89336: Conv2d error on M1 mac, RuntimeError: NNPACK SpatialConvolution_updateOutput failed",stale/low urgency,2,stale/low urgency,1,0,89336,,,,0,03d84ddd0632e0e5441446d216174072fc1660979787f6069afbde9e1eb6d59b
thread:issue:89342,issue #89342: torch.mm: Exceptions thrown on the CPU and GPU are inconsistent,stale/low urgency,2,stale/low urgency,1,0,89342,,,,0,abd79a6dca6d591146a8995f270954c643ce20ebc1597511e2dbfdaa61a3a277
thread:issue:89344,issue #89344: AdaptiveAvgPool1d failed in the lower version,stale/low urgency,2,stale/low urgency,1,0,89344,,,,0,9a038352e431db2153d6fe13c27ce759614084175ae4642c5dd0cd51a6b55c74
thread:issue:89357,issue #89357: Quantizable LSTM has different behavior than LSTM in bidirectional setting,stale/low urgency,2,stale/low urgency,1,0,89357,,,,0,d86b1b6d328b6c65f31b3fc7593d251d9ae6a16faa6efcd09aa5079baf32111c
thread:issue:89369,issue #89369: RuntimeError: CUDA error: device-side assert triggered,stale/low urgency,2,stale/low urgency,1,0,89369,,,,0,f503e5caee057287f1295ad79c804bad99e85d81d79ce2a201616ad7c86d6065
thread:issue:8937,issue #8937: [caffe2] Is there any method to implement learning rate scheduler?,stale/low urgency,2,stale/low urgency,1,0,8937,,,,0,376408b0ffea831af7b673338a7f59d73211b900946f4fa4c40552822b8a10c6
thread:issue:89381,issue #89381: torch.nn.ReplicationPad1d:The description of the exception information thrown is not accurate,stale/low urgency,2,stale/low urgency,1,0,89381,,,,0,49dfb348371c2b20cc9374d698bb779b87bbb299f4da9c7261d1e716cc76425e
thread:issue:89393,issue #89393: Edge case: CPU bool abs is not supported,stale/low urgency,2,stale/low urgency,1,0,89393,,,,0,30ad1310e908c3ddf4cdf2de87f1fe322f172b5f688b295bc6a46fffe09ef8d5
thread:issue:89394,issue #89394: torch.nn.TransformerEncoderLayer missing exception description information.,stale/low urgency,2,stale/low urgency,1,0,89394,,,,0,3070a53e524a40d4c665bbdede36d5f20243debac0dd0cf953f943ee6fe9bc64
thread:issue:89408,issue #89408: torch.nn.MultiLabelMarginLoss has different performance on CPU and GPU.,stale/low urgency,2,stale/low urgency,1,0,89408,,,,0,04af47b45097b21478ce1600fb629d41a4bbfabb9d36d925eabfe51c5b35ee4c
thread:issue:89416,"issue #89416: When the torch.masked_select operator passes in the same parameters, it behaves differently on CPU and GPU.",stale/low urgency,2,stale/low urgency,1,0,89416,,,,0,1caeddc750b177a247e5793380706cf8154c337aef32adb6095cddde26989ebe
thread:issue:89417,issue #89417: OSError: libcublas.so.11: cannot open shared object file: No such file or directory,stale/low urgency,2,stale/low urgency,1,0,89417,,,,0,e00893890d4e6dc858ebc6bffef9b8cdc1872c4e73b3e0439bc813d2a7ed2d9f
thread:issue:89418,"issue #89418: torch.floor_divide: The dividend of torch.floor_divide is set to 0, but it can still run on the GPU.",stale/low urgency,2,stale/low urgency,1,0,89418,,,,0,0370b5668fb85dc2fc9c5f2a833b7c7d13be3540f5ac0d57d2b5ed96966f5689
thread:issue:89419,issue #89419: torch.equal can still run successfully when the parameter types are different.,stale/low urgency,2,stale/low urgency,1,0,89419,,,,0,090f8d816ff3048b2ef773c0b10851f892dbe1b0d43c3c74cc84bf0f5881df4b
thread:issue:89420,issue #89420: Edge case: torch.baddbmm supports double x int8 x int8 inputs on CPU but not CUDA,stale/low urgency,2,stale/low urgency,1,0,89420,,,,0,ad1c7589c9bd47708212f5be6d272052900eec3b3beb6c4ef67dd1d41756241f
thread:issue:89421,issue #89421: fx.wrap is ignored with make_fx proxy tensor tracer,stale/low urgency,2,stale/low urgency,1,0,89421,,,,0,222755a77124b44d447a6a5ac5e612711f23f123ca00d9576e0723fffd7861e4
thread:issue:89438,issue #89438: Allow `low` and `high` to be tensors in `torch.randint`,stale/low urgency,2,stale/low urgency,1,0,89438,,,,0,b14909a5ede65d515bcbc15b183096dd3690603bdeda5dfcdefe4fc3a3a71e20
thread:issue:8944,issue #8944: [caffe2] UnicodeDecodeError when running LeNet,stale/low urgency,2,stale/low urgency,1,0,8944,,,,0,e38c261d89920f8638832a6499bd33e9a897d2ff9d4b7101cab61a7c8db950fb
thread:issue:89490,issue #89490: build: cmake: ability to disable -Werror* (-Werror considered harmful),stale/low urgency,2,stale/low urgency,1,0,89490,,,,0,a0623245ea73817d9310052433e498efb9b1e78c00c108ae3ec7c840d168fc2d
thread:issue:89574,issue #89574: Finish deprecation of autograd decorator over class objects,stale/low urgency,2,stale/low urgency,1,0,89574,,,,0,a2800c228465a179ba9ed470390da646148a1d499c3af9ad14e330eb6bd4d0b7
thread:issue:89597,issue #89597: Caching a model's weights and state_dict to disk to save RAM,stale/low urgency,2,stale/low urgency,1,0,89597,,,,0,dffaf8230645ee6a7bdf8ecc2b3fa6ad6a2a0350ad76f73fb257d3826f670430
thread:issue:89609,issue #89609: NVFuser failing masked.{amax|amin|sum} extremal and correctness tests,stale/low urgency,2,stale/low urgency,1,0,89609,,,,0,f97a2601a0d57fdc65d85dd2e9c3dbf37802e8df050c8fd08ef7729e0191a80b
thread:issue:89634,issue #89634: [ONNX] torch.onnx.export can not export the grad of conv when the op is in CPU,stale/low urgency,2,stale/low urgency,1,0,89634,,,,0,a7229509b0e13b6769ddcfa535c0e4b0ba4401011faadbf45e2f9a7fdb649adf
thread:issue:89675,issue #89675: [ONNX] torch.onnx.export snapshots the grads as constants in onnx when op is in cuda device,stale/low urgency,2,stale/low urgency,1,0,89675,,,,0,6061ccd35cb94f8f331882bef490502d03e40b9efbf4e5dd0d4cceaf119d4d0a
thread:issue:89684,issue #89684: `positive_semidefinite` constraint fails on CUDA 11.7,stale/low urgency,2,stale/low urgency,1,0,89684,,,,0,78516e97cd6461f0b28ed768c166e677830554405b0d274b0412cbfacb3fffe2
thread:issue:89686,issue #89686: MultiProcess tests fail when run on nodes with 1 GPU,stale/low urgency,2,stale/low urgency,1,0,89686,,,,0,9137683a04b1e79dc6e45aebbe9cf1189de2ed7ab98519c9a68829e88752c355
thread:issue:89688,issue #89688: Enable NCCL for PyTorch on Windows,stale/low urgency,2,stale/low urgency,1,0,89688,,,,0,e1bb71668f95ad59df6c4389f2331444fd1d6f66afb7b9cd5a1e01bb788e5b64
thread:issue:89730,issue #89730: torch.addbmm throws different exception differences on CPU and GPU.,stale/low urgency,2,stale/low urgency,1,0,89730,,,,0,ee34627aada7eb4770f7c72c31ea95b4dd5063c4b402cba20436af95f334ad20
thread:issue:8979,issue #8979: Check failed: error == cudaSuccess unspecified launch failure [caffe2],stale/low urgency,2,stale/low urgency,1,0,8979,,,,0,e4ec2d29da57ae87de79900ca43861dc63d6181daf889d55e298cad5946d8451
thread:issue:8982,issue #8982: parallelize_bmuf_distributed_test intermittently hangs,stale/low urgency,2,stale/low urgency,1,0,8982,,,,0,a4985381540847e2f6b6f158550e3567ee0a6f92e1bfee11b3392d396fc1b7df
thread:issue:8984,issue #8984: [Caffe2] segmentation fault,stale/low urgency,2,stale/low urgency,1,0,8984,,,,0,0ce802b7628f7a4040bd9d25d6d4c32a59099cbf538110229ac014c8c669ce15
thread:issue:89907,issue #89907: nn.CrossEntropy/nn.NLLLoss : Request for option to specify invalid ignore_index for perf. optimization,stale/low urgency,2,stale/low urgency,1,0,89907,,,,0,cab3e52e42addc6bfebf54a2f9e7bd9504ab64d90f193ce72c16cef45b04ffed
thread:issue:89942,issue #89942: Random K compression hook in PyTorch DDP,stale/low urgency,2,stale/low urgency,1,0,89942,,,,0,829b966cf82e111a1871f4a6fc7f8ecc4d4ea2053f0e116b39c5ea9050752a04
thread:issue:90000,"issue #90000: Can not use x=torch.tensor(b), to create a Tensor out of a List[List[Tensor]] (A List of Lists of Tensors)",stale/low urgency,2,stale/low urgency,1,0,90000,,,,0,f95b7c156c938a419045fbf0162a452d2ec328418180612dc6e8f7610f1bfc91
thread:issue:9003,issue #9003: cmake error,stale/low urgency,2,stale/low urgency,1,0,9003,,,,0,07c0a12e43960a1f4445613ee31d2b35d6a233fdfe5193261ae18e7f2f2fe01f
thread:issue:90057,issue #90057: [GradScaler] Inconsistent scale values across different GPUs caused by uneven inputs for AMP DDP training,stale/low urgency,2,stale/low urgency,1,0,90057,,,,0,9c3273862c5ddfdf385dcf83ac2a3effe0fbcc4f8b472b00dea8f6fdb4302b75
thread:issue:90065,issue #90065: forward-mode AD formula for torch.add (and possibly others) accidentally upcasts float32 to float64,stale/low urgency,2,stale/low urgency,1,0,90065,,,,0,11e230f293118449b98ba81a92458fd129a3102b424f1c50ce2ea037057f308c
thread:issue:90114,issue #90114: documentation need to be as per python version,stale/low urgency,2,stale/low urgency,1,0,90114,,,,0,6f0398e5247ece0fa97dadf079b3ee639e6d11abd906819332c98af07bd80604
thread:issue:90126,issue #90126: torch._dynamo.exc.BackendCompilerFailed: compile_fx raised TypeError: tqdm.__init__() got an unexpected keyword argument 'desc',stale/low urgency,2,stale/low urgency,1,0,90126,,,,0,c0093b1cf7c1961fdba25b37efbf276cfa38d1f574c3146c1825a9fe69862e5d
thread:issue:90147,issue #90147: [Feature Proposal] Extend torch hub to better support cloud serving and edge deployment,stale/low urgency,2,stale/low urgency,1,0,90147,,,,0,3a0504ad24efdc44f53a63a5da7322bc654cba00114cdbe68ffb0af741009055
thread:issue:9016,issue #9016: SimpleMetaNetDefInitializer intermittently hangs,stale/low urgency,2,stale/low urgency,1,0,9016,,,,0,deeb38fa8d1cb4b85d03ac268eba264e7e869883029edc217f17a51576e47c27
thread:issue:9024,issue #9024: cmake error:Could NOT find IDEEP (missing: /usr/local/include),stale/low urgency,2,stale/low urgency,1,0,9024,,,,0,49bb81fe6dd9fd4bb7ef8307b6e871c9e4b7ddee75f1b3d7e701b82952efc555
thread:issue:90256,issue #90256: LibTorch static build from source missing libshm.so,stale/low urgency,2,stale/low urgency,1,0,90256,,,,0,d75a4502d0299e8d88f8738217539ed51dd69d968b7fce3da00730fb0902f6af
thread:issue:90261,issue #90261: Why torch.mode return different value between CPU and GPU,stale/low urgency,2,stale/low urgency,1,0,90261,,,,0,6f2c746252309067fcf7138f9655f4abf310fc23b03c575308de982d8e9a89d0
thread:issue:9028,"issue #9028: [caffe2]build problem, can not find caffe2_pybind11_state_hip",stale/low urgency,2,stale/low urgency,1,0,9028,,,,0,bd3fa4c6acdff6f11ee71dbbccc2f45980b2e87f21a89fee51aef301a483ce99
thread:issue:90284,issue #90284: Public API definition is not compatible with `torch.testing`,stale/low urgency,2,stale/low urgency,1,0,90284,,,,0,518aaf6c909c38287cbbf35601b9392d1919ff95bdb155f77d9e53cee7b109ac
thread:issue:90288,issue #90288: quantization observers: can we relax the default epsilon value?,stale/low urgency,2,stale/low urgency,1,0,90288,,,,0,9dd3e67b91ba6641b77aac93dfd56f3703791b8e2baedae4d876e254666ae8da
thread:issue:90289,issue #90289: quantization qconfig: can we set per-channel quant as default for qnnpack?,stale/low urgency,2,stale/low urgency,1,0,90289,,,,0,01f87bb7527b09db180ae77e6856b1290844300717dee93f6375b5e91793652f
thread:issue:9029,issue #9029: DataWorkersTest::testRNNInput timeout,stale/low urgency,2,stale/low urgency,1,0,9029,,,,0,4fad23f29db16b2ac7b474b2af61add7da1813b3683a0359989f38dec752dc63
thread:issue:9032,issue #9032: Lint check for non-Unicode characters in diffs / Unicode characters without coding,stale/low urgency,2,stale/low urgency,1,0,9032,,,,0,aab78939a654a52a039fab0646d90810d081866b3945c4eb51c641a2d335e448
thread:issue:90394,issue #90394: FX graph mode quant: backendconfig configuration missing for torch.nn.GRU,stale/low urgency,2,stale/low urgency,1,0,90394,,,,0,ef8bd6c4fb1a8292fcda7488adabc01475788d9f315998400534e0e92023f8df
thread:issue:90395,"issue #90395: Cannot add target-level dependencies to non-existent target ""gloo_cuda"".",stale/low urgency,2,stale/low urgency,1,0,90395,,,,0,2c6cf2c0665f783ab5274f68c3e67077c16b17a0846041c01816f1f005f864fd
thread:issue:90398,issue #90398: IValue(c10::List<IValue>) constructor is confusing and undocumented,stale/low urgency,2,stale/low urgency,1,0,90398,,,,0,4fa0f54e12083ab258bb1d86a204e2c9cb3e07857bdb599bf55659b883dbcf9c
thread:issue:90412,issue #90412: Importing numpy makes Tensor min max crash,stale/low urgency,2,stale/low urgency,1,0,90412,,,,0,bd8baf3df334078508bcc9c17973a5a14fda9cf9fac85c63aa16d4099afbcc97
thread:issue:90424,issue #90424: A few functions in fbgemm_utils.cpp are defined in global namespace,stale/low urgency,2,stale/low urgency,1,0,90424,,,,0,5964b4b6453c086306067915951899caedd2ef7d8ee4b58dc9ff3d7a74e7c30e
thread:issue:90447,issue #90447: Different behavior for complex numbers operations with numpy,stale/low urgency,2,stale/low urgency,1,0,90447,,,,0,911706b91b685259b14ebcc33048a79ae63e386484a866f1a27742388162c0d7
thread:issue:9046,issue #9046: [feature request] freeze() for nn.Module,stale/low urgency,2,stale/low urgency,1,0,9046,,,,0,8dd07f8220942ed9412f22e203a41030fdfb808e029cf7fbeb84c38316fad070
thread:issue:90481,issue #90481: Internal assert when ctx.saved_tensors fails when saving results of an intermediate view tensor with torch.utils.checkpoint and use_reentrant=False,stale/low urgency,2,stale/low urgency,1,0,90481,,,,0,a363e7c659e268fac6ce3a39d8ab25550be279e2dbf1ff8962f21ee8c781e215
thread:issue:90485,issue #90485: Tensor indexing and slicing documentation should explicitly state that indexing follows numpy semantics and link to the numpy indexing documentation.,stale/low urgency,2,stale/low urgency,1,0,90485,,,,0,301fbdd51812aab896ab41f7a31127a2fff672ee7ea79f1d68c1fd351cc16348
thread:issue:90526,issue #90526: valgrind failure `Conditional jump or move depends on uninitialised value(s)`,stale/low urgency,2,stale/low urgency,1,0,90526,,,,0,6f66e99a2261b112972f1e7b3d26c9169652a451f0c505fe91f88f223577c32f
thread:issue:90536,issue #90536: Illegal hardware instruction following Real Time Inference on Raspberry Pi 4 tutorial,stale/low urgency,2,stale/low urgency,1,0,90536,,,,0,eb95326ba48c5522f1e8934b59941cef70251ac13413121f77c076e3dcce4c38
thread:issue:90551,issue #90551: Could not run 'aten::as_strided' with arguments from the 'Metal' backend.,stale/low urgency,2,stale/low urgency,1,0,90551,,,,0,d937ee443a7c0bf32179ffed25df40d4e653a2ca9cd15743bd86876eb8a59b32
thread:issue:90553,issue #90553: Embedding dynamic quantization is not documented and hard to use,stale/low urgency,2,stale/low urgency,1,0,90553,,,,0,e7d2688605e5d1dcdbdb10eeafc3f921d5acb9a0a7b54a589130006024cd4f91
thread:issue:90555,issue #90555: [RFC] Add torch.backends.tbb.is_available(),stale/low urgency,2,stale/low urgency,1,0,90555,,,,0,470c848fd698f05190129b9f44472375d44057599360036ef2156771f7899ea5
thread:issue:90585,issue #90585: [threaded pg] All threads share one Random Number Generator,stale/low urgency,2,stale/low urgency,1,0,90585,,,,0,2119663277304c69cf2ee0e97cf480f26136d634385b3b9c1811d68cea7f7079
thread:issue:90607,"issue #90607: Export to ONNX of `as_strided()` hard codes stride in the graph, although it should be dynamic",stale/low urgency,2,stale/low urgency,1,0,90607,,,,0,f7de6ed234f7761eeb1496ff33a491c249441ae2422407e5d6490b3dfcc84f39
thread:issue:90613,issue #90613: `torch.inverse` multi-threading RuntimeError: lazy wrapper should be called at most once,stale/low urgency,2,stale/low urgency,1,0,90613,,,,0,f3465e5c3bfa7def56e4ea3c819209ae70af91b2a9f2587bf7777ebe6c4f1e4e
thread:issue:90633,issue #90633: torch.min document not up to date,stale/low urgency,2,stale/low urgency,1,0,90633,,,,0,732a7d1be22342cc6a5b2bd65d7cbd902442bc63339b1c1857ba6d7bde464168
thread:issue:9068,issue #9068: Deployment for ios for 1.0,stale/low urgency,2,stale/low urgency,1,0,9068,,,,0,1fa118030634132db1d29ee241dec9a1db87bfb2c038adf2adf111b2d33ac3b8
thread:issue:90695,issue #90695: `torch.empty` produces incorrect tensors with `layout=sparse_csr|sparse_csc` on the CPU,stale/low urgency,2,stale/low urgency,1,0,90695,,,,0,29ea9b28d7e4805be83a6ad3b014b22ec7285c38e4b52f7a0cf38e18dc903c56
thread:issue:90708,issue #90708: PixelShuffle/Unshuffle Channels Last Support on GPU,stale/low urgency,2,stale/low urgency,1,0,90708,,,,0,ba7375656890b2fbb346a4e9501f62e78ef84ef848b92f8d76f7460208aa2e0a
thread:issue:90742,issue #90742: Adopt full_backward_pre_hook in DDP,stale/low urgency,2,stale/low urgency,1,0,90742,,,,0,de858fbf744744e2e476cb289e158f64385c18703ec1c5cfe125a5ede3c6d088
thread:issue:90751,issue #90751: pytorch prune in libtorch,stale/low urgency,2,stale/low urgency,1,0,90751,,,,0,f70450cae0bc0d395af2e6c28843086cb7e83bd8e5860542fefc36a7bd7f4a17
thread:issue:90775,"issue #90775: Extend ""torch.utils.cpp_extension.load"" for both lib64 and **lib**",stale/low urgency,2,stale/low urgency,1,0,90775,,,,0,9e7bc7a72935ce230cc7e104a8d1f42e3d219a85fca4628018400ca212816ddb
thread:issue:90793,issue #90793: nn.CrossEntropyLoss error out when the sample size is large,stale/low urgency,2,stale/low urgency,1,0,90793,,,,0,8c89cafb4f3b52c791d6484c84efb1eabc7d415d06f745f16aa26efd246d982d
thread:issue:90820,issue #90820: Support `divmod` for tensors,stale/low urgency,2,stale/low urgency,1,0,90820,,,,0,cfade17a3b1be2982a368c2d11fb0ad6f16878a9a8aaaee090e11f63d345a031
thread:issue:90827,issue #90827: Support for Pylint,stale/low urgency,2,stale/low urgency,1,0,90827,,,,0,5c422d5637b13c3b4959073dd6a613d6700bc778d8b29f4ac7cdb3353593156d
thread:issue:90832,issue #90832: Inference Mode docs,stale/low urgency,2,stale/low urgency,1,0,90832,,,,0,03642d349a54ae5ca10ba3e927647c6b54e751c68772053e56fd297335e32c82
thread:issue:90842,issue #90842: nn.MultiheadAttention softmax inconsistent in training mode,stale/low urgency,2,stale/low urgency,1,0,90842,,,,0,8ac11bdc2e935e3cc5043b18ee92d5f12dbdc7aa3fb91fcb97cde2debc3bb439
thread:issue:90866,issue #90866: Support arbitrary masks for _nested_tensor_from_mask in nn.TransformerEncoder,stale/low urgency,2,stale/low urgency,1,0,90866,,,,0,9673772cdbe592670a8f58cee94f076bbc1caac8aedc5fb31a27789c5f22d8f2
thread:issue:90895,issue #90895: ModuleNotFoundError: No module named 'torch._C._distributed_c10d'; 'torch._C' is not a package,stale/low urgency,2,stale/low urgency,1,0,90895,,,,0,5f89220e39f0b63488965a8e49c144cc10f0437ef0c427181780955c7ed1c3bb
thread:issue:90905,issue #90905: Retrieve Tensor from Tensor.data_ptr(),stale/low urgency,2,stale/low urgency,1,0,90905,,,,0,908edf67156e6d6147622f65e89d4506641d862875e39c1fa9fa47d8b0f7799a
thread:issue:90920,issue #90920: Support for Transformer Models on Android with Vulkan Backend,stale/low urgency,2,stale/low urgency,1,0,90920,,,,0,e85f5ba6926626d464a3129fded1b9baa32d7394a443a14a1e0fa8bac76e1f0b
thread:issue:9098,issue #9098: [caffe2] how to substract the mean value ?,stale/low urgency,2,stale/low urgency,1,0,9098,,,,0,c0f4749eadd0c3c6b8b21c540452847387921f5c96294f250f2cbafa0320cc54
thread:issue:90992,issue #90992: as_strided_scatter : INTERNAL_ASSERT_FAILED for requires_grad=True and non-config input,stale/low urgency,2,stale/low urgency,1,0,90992,,,,0,b64d6b34c971567ed2cdac02818d61dcaaf5e2d981ba3ec350ffbffacab50096
thread:issue:90998,"issue #90998: Crash in `index_select` with singleton `self`, non-singleton `index`",stale/low urgency,2,stale/low urgency,1,0,90998,,,,0,0e126a7ac657df21d2e74b535638fe92a9dc1ff2db338c8a5b36df3074ecd7e4
thread:issue:90999,issue #90999: overflow (?) on cuda tensor after matrix multiplication,stale/low urgency,2,stale/low urgency,1,0,90999,,,,0,d3f1eae6cc1d4ecd880af60eb1404c4b25f80cd1315db47c0fec9393f8053f05
thread:issue:91039,issue #91039: Simple deleting from the sys cache fails on reimport,stale/low urgency,2,stale/low urgency,1,0,91039,,,,0,61a49cf22e4f982748331f8b172d194d8cd44ec6eabc5e00c20e76184b7ce9d7
thread:issue:91182,issue #91182: onednn(mkldnn) backend support for quantized operators,stale/low urgency,2,stale/low urgency,1,0,91182,,,,0,3d534ce169426a991232b96075a05aaf15ffdb906f67140630cc48a6c78c8ad1
thread:issue:91199,issue #91199: functorch.functionalize doesn't work with torch.autograd.grad,stale/low urgency,2,stale/low urgency,1,0,91199,,,,0,3aed4b3538b81282f9f1ad67f2f4cb032206cf32521807fd5ec9f4190f222696
thread:issue:91210,issue #91210: Odd/hand-wavy mathematical notation for Conv2D,stale/low urgency,2,stale/low urgency,1,0,91210,,,,0,bb6b06b3cdd8ca253a748b2363310f570e169dd8d2ca1cf7ae32aacbcee64285
thread:issue:91244,issue #91244: first class dims leak memeory,stale/low urgency,2,stale/low urgency,1,0,91244,,,,0,fd1a8b23c53a55f2623540114cb40af690e58cd6ef6727bdc337afc6e72b65c8
thread:issue:91251,issue #91251: Potential bug found with pybind11 dec_ref while gil released,stale/low urgency,2,stale/low urgency,1,0,91251,,,,0,ca298c0e543fcee919f6b6cdd77dd1d69a99f3b2be54f221ed61bf6ae319de29
thread:issue:91252,issue #91252: Open file leak when dataloader is using persistent_workers and pin_memory AND you create multiple dataloaders.,stale/low urgency,2,stale/low urgency,1,0,91252,,,,0,4c68ea41f071581e9fa62e4a0ce8349eee4c6cb278f3c42605a3b7cb35c195e1
thread:issue:91263,issue #91263: Make quant_min/quant_max required for observer/fake_quant,stale/low urgency,2,stale/low urgency,1,0,91263,,,,0,3bcef263d0044125c269f125c28e4caa70ab24c33b7f5f1bcbc4c692516c8fc9
thread:issue:91280,issue #91280: custom Function that supports functorch jvp doesn't work with in-place,stale/low urgency,2,stale/low urgency,1,0,91280,,,,0,a4d987e838b6f2bc7c9ab8dfefc483bb41e8acc1d07a2b5de2ed8eb251e0c317
thread:issue:91325,issue #91325: Support for saving multiple storages/tensors that view same data as different dtypes,stale/low urgency,2,stale/low urgency,1,0,91325,,,,0,c870f27667667de8aa5f1fe31a9e7527c15b5378c8a927851a9bfcfbb43b2432
thread:issue:9133,issue #9133: [Caffe2] Error running net train when running resnet50,stale/low urgency,2,stale/low urgency,1,0,9133,,,,0,f7523880879ec0a20fd7d2f52c0118f79ca08507e2bec662dcfc9485d59fefc3
thread:issue:91334,issue #91334: Pytorch 1.13 conda package with cuda requires too many unneccessary packages,stale/low urgency,2,stale/low urgency,1,0,91334,,,,0,52b9e4f87a1cf8abbfba022243ee46e77663e04c4352b1da97f5cb070da6fa80
thread:issue:91353,issue #91353: Perf reduction due to munmap with dataloader pinning thread ?,stale/low urgency,2,stale/low urgency,1,0,91353,,,,0,ba43ba107d81e702a5f65fefcd9fb7f6506d92634c5c1c231565b6131b5835b3
thread:issue:91368,"issue #91368: PyTorch memory leak reference cycle in for loop, Mac M1",stale/low urgency,2,stale/low urgency,1,0,91368,,,,0,e2d8604d5f61f85a8762c607cd83656cb581848bfccb86fd46e6de14176304bf
thread:issue:91375,issue #91375: Internal Assert failed,stale/low urgency,2,stale/low urgency,1,0,91375,,,,0,d3d8fe8c51357da29696e1a751009ba01be57fce4dcfccaf7d71a7985231a5a3
thread:issue:91395,issue #91395: PyObject preservation and resurrection for `StorageImpl`,stale/low urgency,2,stale/low urgency,1,0,91395,,,,0,913a4037a1dbaf8b8d3f11b899f195d6eb8c862c8e0a60581238f420f220dbb3
thread:issue:91406,issue #91406: Pytorch clang-tidy header-filter is still broken,stale/low urgency,2,stale/low urgency,1,0,91406,,,,0,284a71e0a5dc2be1d1ccf5667236685fffbcbae26ab092b1fd203aef4000f09c
thread:issue:91417,issue #91417: [bazel] error: use of undeclared identifier 'cudaGraphDebugDotPrint',stale/low urgency,2,stale/low urgency,1,0,91417,,,,0,ab15a2c866c15c5b6875f7a5a47c48b73ad9a117c544a104f33835f0ede3a26b
thread:issue:91435,issue #91435: Batch_first attribute in quantizable multiheadattention,stale/low urgency,2,stale/low urgency,1,0,91435,,,,0,e4cfd1031c0095f4a58df6056bb012c732c8874208802ca9f73b6143d777fe3f
thread:issue:91437,issue #91437: using Tensor subclass between vmap layers,stale/low urgency,2,stale/low urgency,1,0,91437,,,,0,f49ac2437e8c55ed0926f396bbb3fce296707687d06516fb6d8b7ae9c24e9464
thread:issue:91447,"issue #91447: Degenerate ranges are allowed in NumPy, but not in PyTorch.",stale/low urgency,2,stale/low urgency,1,0,91447,,,,0,b76a36d4e17764b245ded26f897f6c9d7aa7e13277d2b6535b7a337eb0f6cdd5
thread:issue:9150,issue #9150: [caffe2] Drop connections,stale/low urgency,2,stale/low urgency,1,0,9150,,,,0,46f2858ccde60f67a6484e3353ce1cee2012571efe31eb17f4a3d3869caaac81
thread:issue:91509,issue #91509: [functorch] make batch norm docs point to UX limitations,stale/low urgency,2,stale/low urgency,1,0,91509,,,,0,e92e524e542b40e86e7189e9685cce8a390b5de72a58147dd183e2289791d311
thread:issue:91536,issue #91536: The speed of matrix inversion is relatively slow for many small matrices,stale/low urgency,2,stale/low urgency,1,0,91536,,,,0,d52c1f41f2313df031b6dc83f771b53bc0c9377716c85ca26b6a04f167eb0eb1
thread:issue:91552,issue #91552: Overflow when running torch.nn.AdaptiveMaxPool2d,stale/low urgency,2,stale/low urgency,1,0,91552,,,,0,16714d80e11399862c9d38d66e8c6490ebf3ed01adb8675cce9ddc79849e4c5d
thread:issue:91553,issue #91553: Segmentation fault when running torch.nn.AdaptiveMaxPool2d,stale/low urgency,2,stale/low urgency,1,0,91553,,,,0,fcb55e6001b4435e33684b4f653436fd3bb8d25000f144841181cdf0e2f65575
thread:issue:91556,issue #91556: Overflow when running torch.nn.AdaptiveMaxPool3d on torch 1.12.0 and 1.13.1,stale/low urgency,2,stale/low urgency,1,0,91556,,,,0,47f10b524649d83f220cd35442d98b018dc1aaa33897f2e13e957915a47763d6
thread:issue:91557,issue #91557: Segmentation fault when running torch.nn.AdaptiveMaxPool3d,stale/low urgency,2,stale/low urgency,1,0,91557,,,,0,1fc2f816aa0369a73b508147774c893b7586acfed4484d0518e6309f3a2634a8
thread:issue:91566,issue #91566: Build Error: OpenMP library could not be found. Proceeding might lead to highly sub-optimal performance.,stale/low urgency,2,stale/low urgency,1,0,91566,,,,0,f5bff21805e6f3954aba0e9a1e8f5dd16859e0100860e6284bb8bdc887f184fb
thread:issue:91574,issue #91574: Add BlockWise Distribution Support to the torch.distributions Package,stale/low urgency,2,stale/low urgency,1,0,91574,,,,0,a48b57029247b6af74812c59f5e486f17a3cb996872711bfd3d60ec0efce4ee7
thread:issue:91582,issue #91582: Softmax function slows down for data with large range,stale/low urgency,2,stale/low urgency,1,0,91582,,,,0,8827c1fb45e2ec19b20b4f6f5f4f9ea7d1b699650447a297fb75e5d43e11e998
thread:issue:91593,issue #91593: Division by zero error when running torch.nn.functional.lp_pool1d,stale/low urgency,2,stale/low urgency,1,0,91593,,,,0,632cdcb8eccba2d6b77ab38c84cc61c4119b457a9f595a45c3f684c869558764
thread:issue:91599,"issue #91599: `mul(CSC, CSC)` fails with layout mismatch between the inputs and the output.",stale/low urgency,2,stale/low urgency,1,0,91599,,,,0,5a0ab26d94fc0cdad25ced1a0aed77f6cba92d3aac7735cc9f2a449af1950ffe
thread:issue:91604,issue #91604: wrong assert message,stale/low urgency,2,stale/low urgency,1,0,91604,,,,0,a360e3743fbca484ab2a05b849d3bbf17a4377fca84e8ecc3f83fdb1f7462433
thread:issue:91608,issue #91608: SSL: CERTIFICATE_VERIFY_FAILED while trying to download pretrained model within a company that transforms SSL certificates for security purposes,stale/low urgency,2,stale/low urgency,1,0,91608,,,,0,746e8fd9dcd06700216106c898112f0ffa6deb8b339dc49782a26a745ed05d75
thread:issue:91615,"issue #91615: sparse.mm(coo, dense) produces wrong results on T4/V100 GPUs",stale/low urgency,2,stale/low urgency,1,0,91615,,,,0,786384e6957d4ec7458e8ca90730a667403c03811756c8589ee7f0a571efa16b
thread:issue:91618,issue #91618: There is no developer documentation about getting started with MPS native debugging,stale/low urgency,2,stale/low urgency,1,0,91618,,,,0,de9e75db5126eb63f34f0e0cd00a090fa2afc850fe287378c973d07ef8a0764c
thread:issue:91624,issue #91624: trainer,stale/low urgency,2,stale/low urgency,1,0,91624,,,,0,0a583757dfa792c6fcde133542685d3c9cd8c248de8fd2fe266dc40059d7aff5
thread:issue:91630,issue #91630: Periodic ROCM distribtued jobs are broken,stale/low urgency,2,stale/low urgency,1,0,91630,,,,0,4c07e09f19efe68b8c98558bee5467a3b1418c9ca9a54c8b9a514a0605e676cb
thread:issue:91653,issue #91653: Stochastic Illegal Memory Access error mid-epoch on AWS p4d instances,stale/low urgency,2,stale/low urgency,1,0,91653,,,,0,170b54e3bf64888397dd4a0d21b79fa268323e55cf690d2b6b2a97087100f2fd
thread:issue:9166,issue #9166: nvcc fatal : A single input file is required for a non-link phase when an outputfile is specified,stale/low urgency,2,stale/low urgency,1,0,9166,,,,0,41156e536c377b5e4353b3597c1820cde8b90e3ab7f87f41b382cae68e2d4bcb
thread:issue:91678,issue #91678: Wrong in building torch from source,stale/low urgency,2,stale/low urgency,1,0,91678,,,,0,68ab1bc13d94bb36f7df9fc71ad06bf882ec4644c716a820932c9dc9196e2a21
thread:issue:91697,issue #91697: Build from source fails: undefined reference to caffe2::DeviceQuery,stale/low urgency,2,stale/low urgency,1,0,91697,,,,0,f310e2164ec562ddc104781b922dd8049ec4a4d6456487affc1c2d18990acbf2
thread:issue:9171,issue #9171: Mismatch in behaviour of WeightedRandomSampler and other samplers,stale/low urgency,2,stale/low urgency,1,0,9171,,,,0,32ead051b95fa1963f094553fe1f867fb3560608dc87567d775aa78e911a0254
thread:issue:91716,issue #91716: linspace (and arange) behaves differently on GPU and CPU,stale/low urgency,2,stale/low urgency,1,0,91716,,,,0,b998eb60dbc28fa199187f4cc0284d317d3f39209e15f5b83ff1db01268bf10e
thread:issue:91737,issue #91737: [MPS] Improve the performance of torch.linear(),stale/low urgency,2,stale/low urgency,1,0,91737,,,,0,8cebe5e01dec354f34632beef7118f76b793a676d7790f54f593852af2b5a4ba
thread:issue:9174,issue #9174: [Caffe2] build_ios.sh => 'is only available on iOS 11 or newer',stale/low urgency,2,stale/low urgency,1,0,9174,,,,0,4bafe06c163dcea287d1d5c04a3e2deed488762c1a125eaa3c02753bdbcc2554
thread:issue:91751,issue #91751: [bazel] replace //c10:headers dependency by //c10 dependency,stale/low urgency,2,stale/low urgency,1,0,91751,,,,0,0e61ff49860551102c69296e1f339ebcfe37b8271dd075f2a82b9ebd03f69e2a
thread:issue:91759,"issue #91759: Error while building pytorch from source on windows - Ninja Build Stopped, Subcommand Failed",stale/low urgency,2,stale/low urgency,1,0,91759,,,,0,ebd67510f7d7922135f55d61105e30c950e4f97c1679b677d6fd8aeb6036eec7
thread:issue:91760,issue #91760: Inconsistent rank among torch.distributed primitives,stale/low urgency,2,stale/low urgency,1,0,91760,,,,0,bd19651de3fbdcb3d60990670537103525e4f1607abaa9d93b212561e6e97a19
thread:issue:91809,issue #91809: Better API for `torch.cov` (and `Tensor.cov`),stale/low urgency,2,stale/low urgency,1,0,91809,,,,0,5ea9fc8ec945be30fc6d9b8a4006e64506e679357070fc4ecafa10b48559ca00
thread:issue:91810,issue #91810: [Bug/functorch] Cannot use `tensor.detach().numpy()` for `GradTrackingTensor`: Cannot access data pointer of Tensor that doesn't have storage,stale/low urgency,2,stale/low urgency,1,0,91810,,,,0,0845904f997b5cf8811fccc60914f624aa675460ef290a8b6deb7c005a267abf
thread:issue:91842,issue #91842: Adding a page for subfolder/subfile overview/descriptions in the developer wiki,stale/low urgency,2,stale/low urgency,1,0,91842,,,,0,d21a033792af73c7f919f6de957928de9ec46815a93433b3c6786cf284f724d8
thread:issue:91855,issue #91855: torchrun default value of command line options,stale/low urgency,2,stale/low urgency,1,0,91855,,,,0,aeebb4d6f2c5025276a92a450dee7ee7a3e396ae8552c025b9347955acdc3616
thread:issue:91856,issue #91856: torchrun --help is too slow,stale/low urgency,2,stale/low urgency,1,0,91856,,,,0,4282c8b7a66b4a93133e4d1bdf86cd144b82611a56560c4ea90931d92d28174f
thread:issue:9186,issue #9186: [pytorch] Make dtype second positional argument of tensor factory methods,stale/low urgency,2,stale/low urgency,1,0,9186,,,,0,b28717fb7a012ee8bd2979b9f0f9b69c9495bf0a7701dd2b128198db80bec9a6
thread:issue:91863,issue #91863: torch.Categorical samples indexes with 0 probability when given logits as argument,stale/low urgency,2,stale/low urgency,1,0,91863,,,,0,5eb1d19b8ae78af3d97133d1db2226f02648ecfa4ccf8cbf100b1851e144bae9
thread:issue:91889,issue #91889: ToTensor deadlock in subprocess,stale/low urgency,2,stale/low urgency,1,0,91889,,,,0,83bc994a41ecaa7c2507fde3b17f287b81ab27ad73e4a7bd70585686360cdfd9
thread:issue:9193,issue #9193: [Feature Request] Additional torch.nn.LSTM functionality,stale/low urgency,2,stale/low urgency,1,0,9193,,,,0,42ce946b92502fa1301f346da8dcbf1bb343098ccc75a96d129e2a0bc4bb8875
thread:issue:91936,issue #91936: Fails to build on ppc64le with clang,stale/low urgency,2,stale/low urgency,1,0,91936,,,,0,02416edc9b47dfb3b1fe20ea442efc5324d4d39b1dfeb132ea0b8d12a7af225c
thread:issue:91940,issue #91940: torchrun elastic always “address already in use” error,stale/low urgency,2,stale/low urgency,1,0,91940,,,,0,30eefc1d22257caefddab622cfde4a23fe5b26ff5cc059edfa0325d61ef0996f
thread:issue:91943,issue #91943: elastic job failed when scale down,stale/low urgency,2,stale/low urgency,1,0,91943,,,,0,75e5e01be4b753121d7c2a99c0596908a2f6ecf1d14b2f7407a4f3218f819c13
thread:issue:91950,issue #91950: RuntimeError: philox_cuda_state for an unexpected CUDA generator used during capture,stale/low urgency,2,stale/low urgency,1,0,91950,,,,0,13db6dc7b8a9d7df831a596cc13cb4e55c09b5b8434789ce1b2b6a26d7196398
thread:issue:91958,issue #91958: CUDA error `CUBLAS_STATUS_NOT_INITIALIZED`,stale/low urgency,2,stale/low urgency,1,0,91958,,,,0,02dbb430325b9b3c7f0427d576a597c52984eb9ea89fe8b038d827f05fa5bf40
thread:issue:91996,issue #91996: autograd.functional.jacobian : Imaginary part is lost for functions with real input and complex output.,stale/low urgency,2,stale/low urgency,1,0,91996,,,,0,f6117e36e49c366979a6bae2b92b2a0c7d41fa0a3ec0b08ecbec1fdf67157384
thread:issue:9207,issue #9207: Where is the include and lib path for caffe2?,stale/low urgency,2,stale/low urgency,1,0,9207,,,,0,2738309e50b421464962b54245f363e201233bc7bdae3481092d2ea72628d67e
thread:issue:92072,issue #92072: Implement forward AD with grid_sampler_2d,stale/low urgency,2,stale/low urgency,1,0,92072,,,,0,7bf487994054fef18e24b6c7518ae1812f75c9319017218b47acf168c480da60
thread:issue:92075,"issue #92075: Hessian produces wrong results, but works if I add a perturbation",stale/low urgency,2,stale/low urgency,1,0,92075,,,,0,cd1ba1e8cf6086195a86824b7b571a65716fb14d74c18eac53b4f4658409510d
thread:issue:9208,issue #9208: [Caffe2] compiling error with gcc-6,stale/low urgency,2,stale/low urgency,1,0,9208,,,,0,376304f0669b8dedeb61664df5849eaa1b4104a505bfb6b30eacc5174017eede
thread:issue:92086,issue #92086: Error while building pytorch mobile binaries from source,stale/low urgency,2,stale/low urgency,1,0,92086,,,,0,46b1b109c84c1312eb23eb310bc6c75916b4838753a9de93cb74cbab58e13ac9
thread:issue:92128,issue #92128: Forward arguments are not updated in DDP,stale/low urgency,2,stale/low urgency,1,0,92128,,,,0,7dac1db1e5fa66f2169d512549f1fbb0d448306c715fcce0263d48035363d16f
thread:issue:92132,issue #92132: Torchrun seems to have problem with virtual environment,stale/low urgency,2,stale/low urgency,1,0,92132,,,,0,5c5fe916ec637bd39c3bd18c3c80e8ee5d58cb468af1a3b98b05f025951137ee
thread:issue:92134,issue #92134: [Bug][Dataloader] unable to mmap 2048 bytes from file <filename not specified>: Cannot allocate memory (12),stale/low urgency,2,stale/low urgency,1,0,92134,,,,0,1c378d1de857da668e763fb41e4cee13a1918c179061e0485d0f696485e1b03d
thread:issue:92148,issue #92148: Occassional OverflowError with mps running yolov7,stale/low urgency,2,stale/low urgency,1,0,92148,,,,0,b45e87c89a00d038e983ba577bbc8237bfa3c4d4eaff33e0dae9efae67b4cb89
thread:issue:92173,"issue #92173: ""multi device"" tests get skipped in standard CI",stale/low urgency,2,stale/low urgency,1,0,92173,,,,0,219b11be6e29cec6212a9010e872790b4eee97fec2d0da342bb4e33427b47d8d
thread:issue:9218,issue #9218: [Caffe2] Cannot load caffe2.python. Error: libcaffe2.so: cannot open shared object file: No such file or directory,stale/low urgency,2,stale/low urgency,1,0,9218,,,,0,f71e80eb26c523ad3efb851331c58bd1cb7c4a9c7a41a9223c4544586457d444
thread:issue:92217,"issue #92217: false INTERNAL ASSERT FAILED at ""../c10/cuda/CUDAGraphsC10Utils.h"":73, please report a bug to PyTorch. Unknown CUDA graph CaptureStatus32680",stale/low urgency,2,stale/low urgency,1,0,92217,,,,0,e18713812715c178ed382b954880cd58dc6ed1ed5b10b9e680d0b2b486e36014
thread:issue:92223,issue #92223: Improve make_fx tracing speed,stale/low urgency,2,stale/low urgency,1,0,92223,,,,0,393d46c97d3f3ad59be8915a68b6c96136878c603a112cb2155db1fbdc7b020b
thread:issue:92251,"issue #92251: `torch.load(..., map_location=""cuda:0"")` allocates memory on both CPU and GPU",stale/low urgency,2,stale/low urgency,1,0,92251,,,,0,513abec584e7969ae27c2d64d0512440d1f734727dcbadcb859b7769c000412c
thread:issue:92252,"issue #92252: `model.to(""cuda:0"")` does not release all CPU memory",stale/low urgency,2,stale/low urgency,1,0,92252,,,,0,defa48a269c7440e7e0ced660daa64feeeee1b93268cd6c479b368c043e3b351
thread:issue:92259,issue #92259: [discussion] Fused MLPs,stale/low urgency,2,stale/low urgency,1,0,92259,,,,0,9ecba5359ef24879de0a98a97d7e24d6e069e08dfdab1d1163b1b84482bc8700
thread:issue:92260,issue #92260: distributions.Beta returning incorrect results at 0 and 1,stale/low urgency,2,stale/low urgency,1,0,92260,,,,0,51bc8dd7f68bb2c9df8c3c112688f297643262eb5183237484e03f77adab7a2f
thread:issue:92273,issue #92273: Always install cpu version automatically,stale/low urgency,2,stale/low urgency,1,0,92273,,,,0,0da1678756960d25cf29dbf1bb4148afcfc99c39825c47498722588c0da61855
thread:issue:92285,issue #92285: InstanceNorm operator support for Vulkan devices,stale/low urgency,2,stale/low urgency,1,0,92285,,,,0,67e35ec9777d33cdd6991ea961967c301010246ac7cc4faaf0bcdd9d65a19e60
thread:issue:92292,"issue #92292: ""Get Started"" tells us to use the anaconda installer for PyTorch 3.x - but this should be python 3.x",stale/low urgency,2,stale/low urgency,1,0,92292,,,,0,aeb454d231a98a3188cf6a41ea3c2104333fb83989512837590c2d13dfc1aaaf
thread:issue:92310,issue #92310: [Releng] Windows AMI needs to be pinned for release,stale/low urgency,2,stale/low urgency,1,0,92310,,,,0,4e1a0e08a06daaec1eb45c3bc2c868977a0e39368303c117cf6537f986ed96e6
thread:issue:92331,issue #92331: backward(inputs= does not need to execute grad_fn of the inputs,stale/low urgency,2,stale/low urgency,1,0,92331,,,,0,1b05a7d19df0f4febab7a254398a0ef2d4a36797e89234358b2efee742e66a93
thread:issue:92350,issue #92350: TypeError: no implementation found for 'torch._ops.aten.max.default' on types that implement __torch_dispatch__: [<class 'torch.masked.maskedtensor.core.MaskedTensor'>],stale/low urgency,2,stale/low urgency,1,0,92350,,,,0,b2c966beb6c2f9efcabb04976e6ee15b43fa7b435642f40d11cacf0580023d53
thread:issue:92398,issue #92398: [Inductor] support complex dtypes,stale/low urgency,2,stale/low urgency,1,0,92398,,,,0,0666de3c7b8d3e624cdecb071209203052dab3286bdbc2fd702b7279077a7e97
thread:issue:9250,issue #9250: [feature request] SSIM-based cost function as part of the standard set of loss functions,stale/low urgency,2,stale/low urgency,1,0,9250,,,,0,56d1542a3ff14b321514d2ea21ea4a48400e5e3f06289e72d0c3bcb72e069c89
thread:issue:92542,issue #92542: Pytorch AMP performance issue.,stale/low urgency,2,stale/low urgency,1,0,92542,,,,0,01e15d7ed5f37bca76aa0cdcebffc249f4886233fb6ad7997a4fffde0c6505f5
thread:issue:92594,issue #92594: INTERNAL ASSERT FAILED when mixed dtypes for `addcmul_`,stale/low urgency,2,stale/low urgency,1,0,92594,,,,0,674a375c2f74f352b43bee9d52d247b966486345c8b6a7a1cad9cefa71e91a68
thread:issue:92683,issue #92683: Torchscript troubles with complex values. RuntimeError: isInt() INTERNAL ASSERT FAILED,stale/low urgency,2,stale/low urgency,1,0,92683,,,,0,0468c98b1d9ac0f7bd62131283182b646892e4788bfee4fb9c96772b0908ad3c
thread:issue:92701,issue #92701: set_default_device/torch.device has performance impact for non-factory functions,stale/low urgency,2,stale/low urgency,1,0,92701,,,,0,1785a2ce364ee13cdb795a0bba674cc21073cc2ee685a8e8bd032cd724afe4e1
thread:issue:92778,issue #92778: Process get killed when running torch.normal,stale/low urgency,2,stale/low urgency,1,0,92778,,,,0,96044bdb91ab65af9ac9354d0e6cd0a92d143f0e397cd2a0eaedd30a39edaf1e
thread:issue:92781,issue #92781: Floating point exception when running torch.nn.AdaptiveMaxPool3d,stale/low urgency,2,stale/low urgency,1,0,92781,,,,0,8e8940fe43e4bb698b3dd4754a3ed4d6685e3229a7107936ac01aedde1582df1
thread:issue:92811,issue #92811: [complex] Jacobian of a non-holomorphic complex valued function,stale/low urgency,2,stale/low urgency,1,0,92811,,,,0,d397c5e53df18897c0d0ce74a3d5c23155e7f0fc351eabf53700cda1bc957199
thread:issue:92818,issue #92818: Segfault when running torch.atan2,stale/low urgency,2,stale/low urgency,1,0,92818,,,,0,d6711a1608006e74f02f3b8f13d0fbdc4c207fdc0ab02b5beeaadb7865974855
thread:issue:9282,issue #9282: [gradcheck] warn about the case that mulitple inputs share storage,stale/low urgency,2,stale/low urgency,1,0,9282,,,,0,23d9d0c468a6e9e88fde75ac4fc6306c442792eeb48bb93c01e106db291902c0
thread:issue:92828,issue #92828: segfault when running torch.igamma,stale/low urgency,2,stale/low urgency,1,0,92828,,,,0,119b2008216be4e79d6c0e95a3d29d495b525ba335479c891ebcddf97b90ab6d
thread:issue:92835,issue #92835: Double free when running torch.linalg.ldl_solve,stale/low urgency,2,stale/low urgency,1,0,92835,,,,0,c935cf7b9e7c47d9da64f9641e4c352bd77f20b72f71e9a7b41af530419573b4
thread:issue:92838,issue #92838: Enable OnDemand for Open Source CI,stale/low urgency,2,stale/low urgency,1,0,92838,,,,0,128ffda9c06a5b1152fa469e3fcaa2193864dd44d59cfc45895e2a41c7c37f05
thread:issue:9286,issue #9286: Accumulate into accreal instead of real for CPU loss functions,stale/low urgency,2,stale/low urgency,1,0,9286,,,,0,fa38df64598e7d388495b9a433b19d8dbcafb8861a425805e619976a03ddb0b5
thread:issue:92907,issue #92907: Unable to find an engine to execute when using pip to install but not with conda,stale/low urgency,2,stale/low urgency,1,0,92907,,,,0,f9a776b97ec2578195ccf415f0b1730998460eb0289c007cf0de832cdc6b2899
thread:issue:92909,issue #92909: [RFC] XLA Lazy Backend Support In DistributedTensor API,stale/low urgency,2,stale/low urgency,1,0,92909,,,,0,da94f037a421c8b73c05533015e77439c61d088f0acc8728e15fac06acebf078
thread:issue:92912,issue #92912: functorch.functionalize doesn't error out with logcumsumexp.out,stale/low urgency,2,stale/low urgency,1,0,92912,,,,0,a12aa87b0d8dba325c6ab62051e7ed50a6ad96abc3ffabe894626837ef262f00
thread:issue:92920,issue #92920: Add Stride Argument For Constructors,stale/low urgency,2,stale/low urgency,1,0,92920,,,,0,074870c6924aae09bd91eaac29d65e4f63919d350be722f297469788ea739d04
thread:issue:92967,issue #92967: Tracking issue for segfaults and floating point exceptions on 1.12.0,stale/low urgency,2,stale/low urgency,1,0,92967,,,,0,6aa2f701bb8d8a5f58bc07f13b69be0916dbd89f79dd4b9b0b28fae5b448cc6f
thread:issue:92990,issue #92990: Feature request: access to variable,stale/low urgency,2,stale/low urgency,1,0,92990,,,,0,7299e451ef9c31e708b847b3de40c7c6379651b7f01200f6ef3c4116c23613f3
thread:issue:93017,issue #93017: numpy v1.24 does not work with `writer.add_histogram`,stale/low urgency,2,stale/low urgency,1,0,93017,,,,0,a2f1c9d25c704a93b78eb864a52d5c82f13742b09a6b0305109de32cfda40b99
thread:issue:93121,issue #93121: [RFC] Flop counters in PyTorch,stale/low urgency,2,stale/low urgency,1,0,93121,,,,0,83d90f46bc3a48fe830e0de2501d4375e1ce0d2ea5a372ed06aeecab1477ca17
thread:issue:93188,issue #93188: Why is AvgPool2D taking longer than Conv2D for the same input?,stale/low urgency,2,stale/low urgency,1,0,93188,,,,0,0858e4ff452375f8d7164671b055a2a30e5f4f659d89fe1ac6eeb3004dc16692
thread:issue:93197,issue #93197: Add Support for RockChip NPUs (RKNN(2)),stale/low urgency,2,stale/low urgency,1,0,93197,,,,0,6ead6cafb18314d597a87069a57d654a9de7bcc15bdd1c87dd0b07e6c04c85e7
thread:issue:93206,issue #93206: Build from Source Issues on MacOS Ventura 13.2,stale/low urgency,2,stale/low urgency,1,0,93206,,,,0,ef6580232e58b804f1c6742accd6407d55f4e670616b490676c6981f33084f30
thread:issue:93227,issue #93227: squeezenet1_1 fails accuracy with AMP (but not on CI and dashboard); minifier does not work (when not using cuDNN?),stale/low urgency,2,stale/low urgency,1,0,93227,,,,0,4e87426fde643168780fed7cd0c756fc02b7dc38bc4418c3daf6dfd7bf5ee1fa
thread:issue:93231,"issue #93231: Minifier produces minifier script that doesn't fail accuracy on Background_Matting (dynamic shapes, inductor, inference)",stale/low urgency,2,stale/low urgency,1,0,93231,,,,0,5c6330a0a4b6a6454124149c39ceaebeda7bf3f6516c7ad8a21262a23c65c454
thread:issue:93240,issue #93240: USE_CUDNN=1 doesn't force cmake to fail if cudnn is not found,stale/low urgency,2,stale/low urgency,1,0,93240,,,,0,1cd9a2749a8870d8b69094673ed117b584aacfcbeceedcb8ba2082d7c6b8f7f4
thread:issue:93275,issue #93275: Bug in torch.linalg.svd,stale/low urgency,2,stale/low urgency,1,0,93275,,,,0,2472b82554e2ad708cc9db2129db3545cd090e1c4c8fadbb7ae1ca10e5dfd2fc
thread:issue:93317,issue #93317: Minifier doesn't save/load functorch config,stale/low urgency,2,stale/low urgency,1,0,93317,,,,0,c526741301e08294b31c7dbb38cff7ffe0d20876b7b4ce7ef52a743e0a02c57d
thread:issue:93319,issue #93319: Minifier should not use pickle to save state into minifier launcher,stale/low urgency,2,stale/low urgency,1,0,93319,,,,0,5a2fb34f39a9ea622c149f6b8173d403ec1d51e5fb2967c998a7d45980dba8f6
thread:issue:93347,"issue #93347: when I want to use a new backend, how to deal with the op with 'device' argument?",stale/low urgency,2,stale/low urgency,1,0,93347,,,,0,7ba0b5f3e5041a8253d59244d12f41124f7c2e609b385334665c401e05e83c51
thread:issue:93362,issue #93362: Make torch.testing functions overrideable with torch_function?,stale/low urgency,2,stale/low urgency,1,0,93362,,,,0,3ba564b5bc7ddac438015d0b6a791204f146cdb71e3a3ea14c759cc84d55013e
thread:issue:934,issue #934: dataloader parallels over elements vs over batches,stale/low urgency,2,stale/low urgency,1,0,934,,,,0,472f9f6a9cdefdad18d56def534378dd759dfbf946401008d25e7e9ee90850f8
thread:issue:93444,issue #93444: [Inductor] [CPU] Maxpooling is not vectorized in shufflenet_v2_x1_0,stale/low urgency,2,stale/low urgency,1,0,93444,,,,0,66f67345f5e2279d8fb95af842c5dbeade198d2cd273a7cba2f5de4e5cc17152
thread:issue:93446,issue #93446: [Inductor] [CPU] Vectorization not supporting python pass-in scalar double in speech_transformer,stale/low urgency,2,stale/low urgency,1,0,93446,,,,0,bef9274786a34216be622e579e5d8df694a71fedf7b95e903ff5a0398bbffb8e
thread:issue:93454,issue #93454: MMDet 3.x cannot run successfully in inductor mode,stale/low urgency,2,stale/low urgency,1,0,93454,,,,0,6c2217475655c7a6c6da917e92d8bd3c45c58944a1402945f8c342e28b80504e
thread:issue:93455,issue #93455: Performance regression on interpolation in Kornia,stale/low urgency,2,stale/low urgency,1,0,93455,,,,0,6e5e1263837579be262002017aad7bac9fb85c21a9a1d848a42a45e2adef7e7e
thread:issue:93456,issue #93456: Way to run accuracy minifier on only one particular subgraph,stale/low urgency,2,stale/low urgency,1,0,93456,,,,0,a843eab7b37838a4504fd7c89f3431f1859789fa90e9f45156f9b686dccf6f84
thread:issue:93458,issue #93458: minified code can not produce fp64_ref result,stale/low urgency,2,stale/low urgency,1,0,93458,,,,0,f803364d72275034c89fd3cbc9cf317f44c877c8672228483de1f6a24c9df61b
thread:issue:93463,issue #93463: Traceable tensor subclasses cannot actually be used with AOTAutograd,stale/low urgency,2,stale/low urgency,1,0,93463,,,,0,9ce7c55d9723101881779b55aca70d4c959523a17d1c2716e2d9da57aefb90b0
thread:issue:93464,issue #93464: wav2vec2 model: error trying to do inference,stale/low urgency,2,stale/low urgency,1,0,93464,,,,0,63441b3adeb1bb6e6414672b858ff42ea0b684c6d6ed8af08c1a84b97582590f
thread:issue:93465,issue #93465: Graph breaks with HuggingFace Stable Diffusion,stale/low urgency,2,stale/low urgency,1,0,93465,,,,0,d3c8961618b2c76950bc3d55e902797d353766bf78896cbd49aecfad8d4aaa57
thread:issue:93477,issue #93477: [Inductor] [CPU] optimize thread parallel and loop collapse,stale/low urgency,2,stale/low urgency,1,0,93477,,,,0,99541c7bca50cbf09885c9b398daca2b24f7e474597d30438b44f93b904fed40
thread:issue:93480,issue #93480: Umbrella issue for only populate real_value_cache in export test suite fails,stale/low urgency,2,stale/low urgency,1,0,93480,,,,0,2271d9b3917d4ce8c3bbdc68f5ecd2345fc4a21980a72c3e7ead1111ed86a8b2
thread:issue:93482,issue #93482: [inductor] Add more matmul configurations to `TORCHINDUCTOR_MAX_AUTOTUNE=1` mode,stale/low urgency,2,stale/low urgency,1,0,93482,,,,0,194671aaf8334f6ffff508e73b7a8ea926f6033d25e2a006117199f6d4801a85
thread:issue:93486,issue #93486: A Simple Function Causing Graph Break,stale/low urgency,2,stale/low urgency,1,0,93486,,,,0,1bddb8e96aaa2fed55e934a6a61f70552f76ab19ffc7e471851bda8fce379457
thread:issue:93488,"issue #93488: If minifier test fails, stderr/stdout of subprocess calls is not printed",stale/low urgency,2,stale/low urgency,1,0,93488,,,,0,361f179fcf0db648d8cc84675f781167e59d78add3dee06fd71f544caaa36bc8
thread:issue:93511,issue #93511: support setattr of arbitrary user provided types in tracing,stale/low urgency,2,stale/low urgency,1,0,93511,,,,0,213d7279f9db6650c7ca3283fd53c0f96d20e03708c6ef6997a6c2693215452f
thread:issue:93523,issue #93523: [functorch] [vmap] [SymInt][fake tensor],stale/low urgency,2,stale/low urgency,1,0,93523,,,,0,5952494056e287549826e13dd47b71c0013d1cc4e1048e038e723e3fbbb516b7
thread:issue:93579,issue #93579: torchdynamo.export doesn't work with data-dependent control flow,stale/low urgency,2,stale/low urgency,1,0,93579,,,,0,2b0782ab91a6c29bdf1996f8f25af8093386e58914a5a0ca10f3c1a38bd351f5
thread:issue:93580,issue #93580: AssertionError: Unknown expression s2,stale/low urgency,2,stale/low urgency,1,0,93580,,,,0,429ae04ba5c6f506565ea8c5127078f1ee8ab176ef8e0f8ef9170bf574d4defe
thread:issue:93592,issue #93592: Turning on minifier causes bug to go away (on DebertaForMaskedLM),stale/low urgency,2,stale/low urgency,1,0,93592,,,,0,ed0af951cb011f27aaa3e0b49e3704b7a62e716c2ae68e0e4ff4ee2ab594a5a5
thread:issue:93596,issue #93596: Minifier doesn't work on DebertaForQuestionAnswering,stale/low urgency,2,stale/low urgency,1,0,93596,,,,0,422ea5be61797d371db670836b912cca717db0b9f85d1f3ada93590cfcca0d2b
thread:issue:93602,issue #93602: Diffuser pipeline device attribute broken when using optimized model,stale/low urgency,2,stale/low urgency,1,0,93602,,,,0,8536708c759951315d19e69e1a7f5e13c093b7927bab6b3456c1939ea2fe29a0
thread:issue:93604,issue #93604: TorchBench - moco - RuntimeError: Tensors must be CUDA and dense,stale/low urgency,2,stale/low urgency,1,0,93604,,,,0,e4718c7ed7c74de195d8a01805a5dd3d9dcc59894e8c0feb389bb16fb679057b
thread:issue:93609,issue #93609: Don't store example_value on FX node meta,stale/low urgency,2,stale/low urgency,1,0,93609,,,,0,1a6fa536ffac7ee9b7f3de6ae77b10cfceb7019cbbb9dc0d2f82b06f8de0faaf
thread:issue:93639,issue #93639: Explore TorchInductor optimization pass to reorder kernel bodies,stale/low urgency,2,stale/low urgency,1,0,93639,,,,0,97869de59f44557f5310ca732d4ba4a8d756b10e12ed847f4965e76eaf28331b
thread:issue:93716,issue #93716: End-to-End AMP training with GradScaler,stale/low urgency,2,stale/low urgency,1,0,93716,,,,0,c18aa45ebc84f256fdbdf119155879a330860a99ea1c46fd58b05e01ab7aeb8a
thread:issue:93719,issue #93719: Use bytecode rewriting to offer a __future__ feature,stale/low urgency,2,stale/low urgency,1,0,93719,,,,0,bc8c286ed5f2e2b8f0fb13ac590da08dfd24e4d09cc6973bffa3229fdc6d568d
thread:issue:93737,issue #93737: Improve handling of generators,stale/low urgency,2,stale/low urgency,1,0,93737,,,,0,50cd76e3101adf2cd2eb9a493a8a6e64587b8608f2dd9dcfdd7419bcb1c39861
thread:issue:93743,issue #93743: Support data-dependent control flow operators,stale/low urgency,2,stale/low urgency,1,0,93743,,,,0,f86d9eb6ab6a168ac45367ac35d37f681548a3936448b15c4527fef911882b2e
thread:issue:9375,issue #9375: [Caffe2] Not handle attribute 'spatial' of ONNX operator 'BatchNormalization',stale/low urgency,2,stale/low urgency,1,0,9375,,,,0,7865bf48805f1a0aba3cf198118117caa4360bb51db7a9e5409dfdbf8534cd8a
thread:issue:9381,issue #9381: [doc] many losses still mention size_average in formula,stale/low urgency,2,stale/low urgency,1,0,9381,,,,0,ec077c84d9241b62cf27598894ed675a31a2932478c87992aedca57da759463d
thread:issue:93884,issue #93884: Importing tensorflow (2.12) before torch (2.0) hangs at import torch,stale/low urgency,2,stale/low urgency,1,0,93884,,,,0,600aa11ab97d7af4cd14a9b8dc7d82a61dac5a5765f8bccbb69b973393f42fe9
thread:issue:93923,issue #93923: Dynamo uses CONSTANT_MATCH guards for string inputs,stale/low urgency,2,stale/low urgency,1,0,93923,,,,0,6643450ef0773f725e402cbdd372bd4311798d60d9acfcca86e9c01acf7def5a
thread:issue:93944,issue #93944: error: no member named 'residual_with_sum_zero_point' in 'ideep::attr_t,stale/low urgency,2,stale/low urgency,1,0,93944,,,,0,eee4856eb0982221f94b374024739e03f2577cd780c0b5654ee36c5ad5ed6b50
thread:issue:93947,issue #93947: [RFC] Support Huge Model Init Without mallocs for Compile/Distributed Use Cases,stale/low urgency,2,stale/low urgency,1,0,93947,,,,0,c3f5e27173cca3540846fc03b56f4e3f3a33699ff91935aa9ed3402484a6be75
thread:issue:93982,issue #93982: `linalg.lstsq` fails the gradient computation in forward mode,stale/low urgency,2,stale/low urgency,1,0,93982,,,,0,0e06f2e572331f58dcf195bbe0231f65bff41dceb6c1ec2b5e3114ed0f9d12b4
thread:issue:94021,issue #94021: Set AVX2 is minimum supported instruction set for Linux X86,stale/low urgency,2,stale/low urgency,1,0,94021,,,,0,1ff1f180c939121109abff3c253618e3924f328c0a439d99bf0e6686b35f3657
thread:issue:9407,issue #9407: [feature request] Make `torch.gather` broadcastable.,stale/low urgency,2,stale/low urgency,1,0,9407,,,,0,315d1c4118bef347db0b4adb8396e837f3c6b78ed39b5cb5028ebb3db40816a9
thread:issue:94085,issue #94085: `unfold` fails in forward mode when unfolding a scalar tensor,stale/low urgency,2,stale/low urgency,1,0,94085,,,,0,5dc61406ba881bcd49e49dcf40880689615c847f2283076937a9d526d7fb59ae
thread:issue:94111,issue #94111: `svd` triggers INTERNAL ASSERT FAILED when computing jacobian in forward mode,stale/low urgency,2,stale/low urgency,1,0,94111,,,,0,2c4f7f734bece08584983e167b27409376c0631f728d5201da19086a34d9c3e6
thread:issue:94115,issue #94115: `cat` fails the gradient computation in forward mode with empty tensors when used with legacy vmap,stale/low urgency,2,stale/low urgency,1,0,94115,,,,0,e9df51b46548cb3fc89bc8c51dd2d31ebfbfb7798f36de73054d66672102604f
thread:issue:94131,issue #94131: faster WeightedRandomSampler implementation based on alias method,stale/low urgency,2,stale/low urgency,1,0,94131,,,,0,9d0eae09d209fa07c33bb2c21e8b1b2f65b20824c65de16f1690132948995953
thread:issue:94132,issue #94132: Asking for a LAZYMODULEMIXIN warning,stale/low urgency,2,stale/low urgency,1,0,94132,,,,0,f4f0e409443c8f6867a400151304032f91e725d1b6e0f7c8674cfe73b1bb345d
thread:issue:94161,issue #94161: JIT: Dropout fails codegen on the third forward passes,stale/low urgency,2,stale/low urgency,1,0,94161,,,,0,97c4b9c2bf904f8aeb53ece4c3fb397c1bbedc9c0d01625f2c012fcc31a77138
thread:issue:94164,issue #94164: Pytorch 2.0: Detection models from torchvision don't work with onnx and tensorrt backends,stale/low urgency,2,stale/low urgency,1,0,94164,,,,0,ef4b344988ebf87f258049d864d4d481fd821a666f278dd74f472e26c72ce087
thread:issue:94167,"issue #94167: Faster `pad_sequence` and `tensor_split` function with CUDA kernel, are they possible?",stale/low urgency,2,stale/low urgency,1,0,94167,,,,0,f7e9d684d076c65b6f0e02d5e19860af6871fdb95573137bdf377ffd8ca4cc26
thread:issue:9417,"issue #9417: [feature request] allow forward_pre_hook to ""preprocess"" input",stale/low urgency,2,stale/low urgency,1,0,9417,,,,0,65e90e4b6d72a7349d482628f477a5ca5a03a2db5a9effdf9af0bf5e6a4b2deb
thread:issue:94183,issue #94183: Addition of batch CSR tensors produces incorrect and invalid CSR tensor,stale/low urgency,2,stale/low urgency,1,0,94183,,,,0,b7a2f4d1360c40ee15005e0899c588157d7909a1271d65190ac3bb033e5e8876
thread:issue:94185,issue #94185: Addition of CSC/BSR/BSC tensors raises RuntimeError exceptions,stale/low urgency,2,stale/low urgency,1,0,94185,,,,0,6848e4ebbe5bc2875d4eaa6aa4ebda0d98a263e2b5fa04b47b540f48d4ad5d98
thread:issue:94186,issue #94186: Addition of hybrid CSR tensors produces incorrect and invalid CSR tensor,stale/low urgency,2,stale/low urgency,1,0,94186,,,,0,93d665d5628d36bf4d229aacade6daf80857fb26c6c1be6bba038911f8082f56
thread:issue:94208,issue #94208: `zeros_like` + `fill_` makes the gradient computation in forward mode fail,stale/low urgency,2,stale/low urgency,1,0,94208,,,,0,8893534f16a24d89ab578758e1a54995a88ff83574547419c2026470d18f6414
thread:issue:94238,issue #94238: `PyTorchFileWriter` should drop the GIL while writing files,stale/low urgency,2,stale/low urgency,1,0,94238,,,,0,03a285fab8e075387ae2eabf2223bbcd6c06a2ab5e989f7c27969188a3090dd3
thread:issue:94304,"issue #94304: RuntimeError: p.block != nullptr && p.block->ptr != nullptr INTERNAL ASSERT FAILED at ""../c10/cuda/CUDACachingAllocator.cpp"":1275, please report a bug to PyTorch.",stale/low urgency,2,stale/low urgency,1,0,94304,,,,0,9917df7e7aa5a90e159f45a1e42e6f6d13c2f5fb6dff805c9bb938ef2415321d
thread:issue:94322,issue #94322: RFC: Enabling AVX512 dispatch for compute-intensive ATen ops,stale/low urgency,2,stale/low urgency,1,0,94322,,,,0,24a30b2710aae1fb768e86486df1409e506cf66c94996f5c45e24e5ed4e3a7e7
thread:issue:9433,issue #9433: Build error: cannot find -lonnxifi_loader,stale/low urgency,2,stale/low urgency,1,0,9433,,,,0,0e51c8104e6aa3595a3535a8f9025fab9ae2bbbeb2c1b5c625d88e7eda57d899
thread:issue:94333,issue #94333: cpu log1p for bfloat16 gives wrong result.,stale/low urgency,2,stale/low urgency,1,0,94333,,,,0,86922dd83aa819817c90ac0a8c74f985bab83a202e604aa2e0c04926eb73fac4
thread:issue:94371,"issue #94371: QAT + torch.autocast does not work with default settings, missing fused fake_quant support for half",stale/low urgency,2,stale/low urgency,1,0,94371,,,,0,52469aa0af2950356572f9e4e0ce948b05be6b627a0b207b17f3effad481b42a
thread:issue:94374,issue #94374: [fx] const_fold.split_const_subgraphs leads to UserWarning,stale/low urgency,2,stale/low urgency,1,0,94374,,,,0,636a4b10730dd457d410116058ea4a7a6f1a5a34bc1f57405bcb9d7d1e4c1184
thread:issue:94388,issue #94388: Inconsistent results when using torch.Tensor.bernoulli with float instead of Tensor probabilities,stale/low urgency,2,stale/low urgency,1,0,94388,,,,0,36789c6f7f9ef8efbd06c8bdd46d78a0c8669459e1874bea676158846fe8df2d
thread:issue:94395,issue #94395: `func.jacrev()` should be implemented as `func.jacfwd().mT.contiguous()`,stale/low urgency,2,stale/low urgency,1,0,94395,,,,0,de91cf006cacb42f41af77c9424d0de42a92a1d24b0ae325e9e84ea87eaa8f66
thread:issue:94397,issue #94397: jacfwd and jacrev are fundamentally broken for complex inputs,stale/low urgency,2,stale/low urgency,1,0,94397,,,,0,5407ac23569a8b6bcff8b22ec65beef17144a111abcfe84ac63b545207915573
thread:issue:94441,issue #94441: `get_debug_state` a script function causes INTERNAL ASSERT FAILED,stale/low urgency,2,stale/low urgency,1,0,94441,,,,0,27fd501bd12f9af2d5f5105b5b1a9abab03f1a40b2618c010f923e0dce68c008
thread:issue:94443,issue #94443: A better error msg for `cuda.jiterator` when input is on `cpu`,stale/low urgency,2,stale/low urgency,1,0,94443,,,,0,cdc79fe22318b226eceaacf3019d256ec67f82315795cf95980fc009f4a6845b
thread:issue:94451,"issue #94451: `jacrev` raise ""Cannot access storage of TensorWrapper"" error when computing the grad of `storage`",stale/low urgency,2,stale/low urgency,1,0,94451,,,,0,67b4d9e3fb274c67ff34553887bd2c17508b2fe24bf096d520b79affa903bbc7
thread:issue:94454,issue #94454: Document and promise reproducibility torch.randn / torch.rand / torch.randint family behavior on CPU devices,stale/low urgency,2,stale/low urgency,1,0,94454,,,,0,b85a0d2660bcbb38a96431813bf195eb430e36b75a362b717231afba6e7f5e2b
thread:issue:94457,issue #94457: LSTM on CPU is significantly slower on PyTorch compared to other frameworks,stale/low urgency,2,stale/low urgency,1,0,94457,,,,0,34dd4673758ee065be888f044d509f9eb1ea852bf0f135d64fd1251e83b7ccf6
thread:issue:94471,issue #94471: [Functionalization] `index_reduce_` op tests with functionalization enabled,stale/low urgency,2,stale/low urgency,1,0,94471,,,,0,152602935316fd2e6cd58c8ba98bdf8a5a3d069fce301e2b5cddae6e993526c7
thread:issue:94474,issue #94474: pybind11 SymNode binding is a footgun py::cast,stale/low urgency,2,stale/low urgency,1,0,94474,,,,0,d7e715e80ad37d3d8f7a012b0e9124f9693a8aea83931e0c883e0ba511cf9f4f
thread:issue:9449,issue #9449: PredictorTest.SimpleBatchSizedMapInput intermittently hangs,stale/low urgency,2,stale/low urgency,1,0,9449,,,,0,7126e880e78a1ccda9d85db7390f83d1587f06ba12ed6037ceac127a103e7a39
thread:issue:94504,issue #94504: [mypy] skipping mypy for a few torch/fx and torch/_subclass files,stale/low urgency,2,stale/low urgency,1,0,94504,,,,0,bef0ae8476af76ced148138f87459082b951d54d6c7dbdfc35ee64d8dda44c80
thread:issue:94590,issue #94590: A Segment Fault can be triggered in torch.affine_grid_generator,stale/low urgency,2,stale/low urgency,1,0,94590,,,,0,75a092c9ef810559f9833fcf4663cd9f3a8b1b9fd7d660335db8b0c527cd0b6d
thread:issue:94591,issue #94591: Issue with `upsample_nearest2d` decomposition,stale/low urgency,2,stale/low urgency,1,0,94591,,,,0,0bb3e4c83c257f695f65e2798a5a7fd7f422878152ce733d321ff6ecf8b1f70a
thread:issue:94593,issue #94593: Memory corruptions can be triggered in torch._remove_batch_dim,stale/low urgency,2,stale/low urgency,1,0,94593,,,,0,599e7ffddd9af53a37e52eede6bbf8d9dc8038976bb8e2f1a14c6f2a32a6c4aa
thread:issue:94614,issue #94614: Jetson CI needs Updates,stale/low urgency,2,stale/low urgency,1,0,94614,,,,0,fb4b7f2dcdb2aa70ba6b0201be3fef440bb79ae2d9248d487d1de2b40e371fad
thread:issue:94654,issue #94654: Missing FX documents for some modules,stale/low urgency,2,stale/low urgency,1,0,94654,,,,0,9a42be9cc8af35f5375e7a534315ba28920f7d9fb73ae6332f7a681e9d82bd26
thread:issue:94675,"issue #94675: `UnsupportedOperatorError`, `OnnxExporterError` and `SymbolicValueError` related to MultiheadAttention export to onnx with torch.jit.script",stale/low urgency,2,stale/low urgency,1,0,94675,,,,0,99b2278847e2cc7e659ca6712b9cd7f647ea34de749df1b7140840e95dedeec4
thread:issue:94704,issue #94704: `where` triggers INTERNAL ASSERT FAILED when `out` is a long tensor due to mixed types,stale/low urgency,2,stale/low urgency,1,0,94704,,,,0,bc48b8763aae3f07576129e52c48d97231c79e15e83554b48a42bcad921e6f71
thread:issue:94705,issue #94705: Split getitem OpInfo into dynamic and non-dynamic inputs,stale/low urgency,2,stale/low urgency,1,0,94705,,,,0,7db144f8560163e8109c901b01ad01859e25f5210a689eda03ea1c62617ed252
thread:issue:94739,issue #94739: Update PyTorch's default C standard to C17 from C11,stale/low urgency,2,stale/low urgency,1,0,94739,,,,0,637243957ffbabb4101699f42b422d66e2e4799fcec8e2b866ee9b866ce77d7e
thread:issue:94801,issue #94801: Compiling PyTorch from Source on Xavier,stale/low urgency,2,stale/low urgency,1,0,94801,,,,0,b37506585768ebddcd408e3ffc6210c82e3144bbe0c2b2730dd8e88d70b6b9fe
thread:issue:94804,issue #94804: Build Error: no matching function for call to ‘dnnl::graph::stream::stream(<brace-enclosed initializer list>)’,stale/low urgency,2,stale/low urgency,1,0,94804,,,,0,3f3c918fe62afb72fbb3539c63ae4da95609585d52f5660046ed203c2887ed7e
thread:issue:94806,issue #94806: ImportError: cannot import name 'Backend' from 'torch._C._distributed_c10d' (unknown location),stale/low urgency,2,stale/low urgency,1,0,94806,,,,0,716f11d95b9866fd69745b12200baecb632e229cd9f0da16cb0b389de120c786
thread:issue:94816,issue #94816: grid_sample with relative grid,stale/low urgency,2,stale/low urgency,1,0,94816,,,,0,10aec76db6f3cc4340537f1c7868bdcdb5d7987efcaaa8d9edc4302ffb56c3b8
thread:issue:94827,issue #94827: Linking error with Libtorch,stale/low urgency,2,stale/low urgency,1,0,94827,,,,0,37f6744bc5123864701e632864065e12095f729d231a320bad1332f3e4b2ecdc
thread:issue:94909,issue #94909: [numpy] mean & nanmean should support int dtypes,stale/low urgency,2,stale/low urgency,1,0,94909,,,,0,81d3028bb240f35a5a00b1cf15320e3e359335835640edec3713a6c27124e90d
thread:issue:94912,issue #94912: PR #88607 breaks build for POWER9 CPU,stale/low urgency,2,stale/low urgency,1,0,94912,,,,0,1d7955066ce2d41e01094f2d72e46400eb4072a0a59468752beb23772d63ed15
thread:issue:94966,issue #94966: CSR matrix add_ error with RuntimeError: CUDA error: kernel launch failure when calling cusparseXcsrgeam2Nnz,stale/low urgency,2,stale/low urgency,1,0,94966,,,,0,96f0d370b579a02826bc21be93041e0396220075aac1583886d33902cb7d3c1d
thread:issue:94990,issue #94990: interactions between views + autograd.Function + AOTAutograd causes memory leak,stale/low urgency,2,stale/low urgency,1,0,94990,,,,0,10330473630425a454ef2c02cc9850aa6af343b8c7c6e3b6f41675c18099e923
thread:issue:95021,issue #95021: Custom operations in inductor,stale/low urgency,2,stale/low urgency,1,0,95021,,,,0,fd6241949c03acd5901768af3cc58ae3840595b4464d9c8abf388be99517e12f
thread:issue:95058,issue #95058: Upsampling ResBlock GPU memory spike,stale/low urgency,2,stale/low urgency,1,0,95058,,,,0,8af518a04dc20f901e08acadc915ed6eebab98759c234946712f62f08feaa6c1
thread:issue:95077,issue #95077: Implement a `torch.cuda.visible_device_indexes` function.,stale/low urgency,2,stale/low urgency,1,0,95077,,,,0,85a6398df17819d01258139e7c190fef744afa063d3e4fa0dcf8db030f2df5e4
thread:issue:95121,issue #95121: cuda 12 support request.,stale/low urgency,2,stale/low urgency,1,0,95121,,,,0,3e449ff8f23ab9abe79d3b32f73a7c95cbe5513b3607c7f6c2e716877876a0a3
thread:issue:95122,issue #95122: CosineAnnealingWarmRestarts but restarts are becoming more frequent,stale/low urgency,2,stale/low urgency,1,0,95122,,,,0,85d80a8732b743a02caa4fd8ff79f4cc7a7d459103760847e2e20c904715935d
thread:issue:95129,issue #95129: Default value of `validate_args` is set to `True` when passed as `None` in `Multinomial`,stale/low urgency,2,stale/low urgency,1,0,95129,,,,0,9c1cc17c2766e61553ead759f75b5e4c2f08cd660a8271f8f88724c6cb6b6bea
thread:issue:95132,issue #95132: Differentiate with regard a subset of the input,stale/low urgency,2,stale/low urgency,1,0,95132,,,,0,c9661d973343b5cf57086fcdfc8168879f09db73c8b04a02820dc00d8507d936
thread:issue:95135,issue #95135: Add local version identifier to wheel file names,stale/low urgency,2,stale/low urgency,1,0,95135,,,,0,b331bfe17b93e14f3cdb7105d4c2923e6d175e1ed5eb925e6b32fde7dd685cf1
thread:issue:95146,issue #95146: Whether to consider native support for intel gpu？,stale/low urgency,2,stale/low urgency,1,0,95146,,,,0,f46be708d00200c35dd89236fbdeeec6957bded85ac6d0f54e5f4e2be1ab4120
thread:issue:95169,issue #95169: COO @ COO tries to allocate way too much memory on CUDA,stale/low urgency,2,stale/low urgency,1,0,95169,,,,0,1e384fd4343c7d8f5f56c8470d0b98572e80709fce3aefc7cb8a33613abb2fb8
thread:issue:95207,"issue #95207: new backend privateuseone with ""to"" op",stale/low urgency,2,stale/low urgency,1,0,95207,,,,0,bb2847919b4e054ba807182840510a944a3a6d8a7c17ef90a422dc51d2c33fcf
thread:issue:95210,issue #95210: Add parallel attention layers and Multi-Query Attention (MQA) from PaLM to the fast path for transformers,stale/low urgency,2,stale/low urgency,1,0,95210,,,,0,2abc87b171ce2149bf219b99d1bc877ea822b73ff950d258b78d494cf05805c0
thread:issue:95225,issue #95225: Remove conda virtualenv from the docker image,stale/low urgency,2,stale/low urgency,1,0,95225,,,,0,495e2a2e4634dd1b2c1c7814e2c3384ea56aaed748a01705ac655db59bad7c89
thread:issue:95237,issue #95237: test_foreach failing cuda memory leak check,stale/low urgency,2,stale/low urgency,1,0,95237,,,,0,592fda0916e5374deefa69ad648f517434902a4e93fecde0068a3bd29993557e
thread:issue:95244,issue #95244: Make this ridiculously long error message more user friendly,stale/low urgency,2,stale/low urgency,1,0,95244,,,,0,00bf312f1dc8a799f72fe9b320039ee529cbb24523471c0289f1937f53180df1
thread:issue:95276,issue #95276: tabulate is used by `torch.fx.graph_module.GraphModule.print_tabular` but is not installed when installing pytorch,stale/low urgency,2,stale/low urgency,1,0,95276,,,,0,0e5f463efcefb03870f31cbae54ebd7009284eeb126d272b86e57cc7380e25df
thread:issue:95290,issue #95290: Continuous dropout layer,stale/low urgency,2,stale/low urgency,1,0,95290,,,,0,50b615a503fd3c4df86d131590183c64a5e6da0f0ccfa26ba020cf213625331f
thread:issue:95304,issue #95304: `torch.cartesian_prod` returns inconsistent dimensions with only one input,stale/low urgency,2,stale/low urgency,1,0,95304,,,,0,0594166954b9f75b5dfb0fad7bb8c9cb175fda3ceee88cb8d61bcc10626c1880
thread:issue:95309,issue #95309: `torch.distributed.Store` triggers INTERNAL ASSER FAILED when seting,stale/low urgency,2,stale/low urgency,1,0,95309,,,,0,8743b0caa645b8809abb0589f20590ea6e1ddfba87c58c4ee7746d141e12aa59
thread:issue:9531,issue #9531: [caffe2]train on GPU and test on cpu failed,stale/low urgency,2,stale/low urgency,1,0,9531,,,,0,fe3beda511b4067c0509cc4fe827d42c0a7fdae0d1ec174b5c3e1ebc5a76fac2
thread:issue:95369,issue #95369: build failed when strictly following the guidelines,stale/low urgency,2,stale/low urgency,1,0,95369,,,,0,3d837a0d2da25b1635ce7b688bd486da96db109a798d53a44ced444b1a72e011
thread:issue:95403,issue #95403: test_ddp_apply_optim_in_backward in distributed_test.py fails for gloo backend,stale/low urgency,2,stale/low urgency,1,0,95403,,,,0,886dad64679266e09bb02b2037686af90eec7f0dd8e114b7d7dae9ad871b9595
thread:issue:95423,issue #95423: PT2 Computes Multi Device Backward in a Single Thread,stale/low urgency,2,stale/low urgency,1,0,95423,,,,0,cb34bcdeef3e31c82d0015a4b3e5bd97cd0d3ba01af6de0a099011d8ea980937
thread:issue:95460,issue #95460: torch.profiler.tensorboard_trace_handler Generates an incorrect JSON file,stale/low urgency,2,stale/low urgency,1,0,95460,,,,0,9af72ef2ff7223a24cfe14a8c78747f542943766482a123cac9baa1d622fde33
thread:issue:95462,issue #95462: Extend docs - Fixing out of memory with python garbage collection,stale/low urgency,2,stale/low urgency,1,0,95462,,,,0,46508a153a5f4f5011f651dc870b4432f007c48dddd42b50647b8c0f1aa6c197
thread:issue:95463,issue #95463: `add/add_` for CSC: errors when trying to access non-existent `crow_indices`.,stale/low urgency,2,stale/low urgency,1,0,95463,,,,0,df2916c48128ee4b47bed100f65aad735fe1d9cc749b465e8512d6f8aaf3f911
thread:issue:95485,issue #95485: Enrich shape operations with nested tensors,stale/low urgency,2,stale/low urgency,1,0,95485,,,,0,5da8c8ca7a3df76162e8670362849ae82fe37aa2da0626ff883ae781bfbf0fc0
thread:issue:95548,"issue #95548: torch.distributions.kumaraswamy.Kumaraswamy generates samples outside its support (0,1)",stale/low urgency,2,stale/low urgency,1,0,95548,,,,0,f231593b5c10a3def76298251dc44787f962483856d8f5d4fb3583b09c1be314
thread:issue:95562,issue #95562: test_layer_norm_backward and test_layer_norm_backward_5d run OOM in slow gradcheck,stale/low urgency,2,stale/low urgency,1,0,95562,,,,0,f1f0c5a408dcfb6722cc18997296452dcfb597b4504979156197c164685c73d9
thread:issue:95572,issue #95572: Support datatype argument for torch.distributed.all_gather() (And the whole distributed module),stale/low urgency,2,stale/low urgency,1,0,95572,,,,0,4dcb16a8a481c6134edeee77db609ea315ed2c7bca74fe3fa5ef5a994c0f309c
thread:issue:9559,issue #9559: [feature request] Rename `Subset` -> `Resample` to reflect wider use,stale/low urgency,2,stale/low urgency,1,0,9559,,,,0,df165160ef46e2ecc1e168fb7b4ce6b18f045a581432816a8a1af6691f217ed0
thread:issue:9560,issue #9560: Incorrect term in _LRScheduler.,stale/low urgency,2,stale/low urgency,1,0,9560,,,,0,6e4fe916108b6c0bf3192ff7d189a6c81d2876b713383d2d5be1ff3d87c9ac1e
thread:issue:95622,"issue #95622: Testing InvokeAI 2.3.1.post1, using mps, with PyTorch nightly dev20230226 yields RuntimeError cross-device copies are not allowed!)",stale/low urgency,2,stale/low urgency,1,0,95622,,,,0,e8d8c3f2c417f714f894fa1e7402da35591f8d4777947dabc0e321b6cdabf54a
thread:issue:9563,issue #9563: batch_sampler/test_worker_seed intermittently fails with address already in use on OS X,stale/low urgency,2,stale/low urgency,1,0,9563,,,,0,86eb09a962b03ecc05152d32b788fd85bb44b875ac9c836e7b1d53c6b1d716c8
thread:issue:95645,issue #95645: Create a new Docker image with all inductor benchmarks and pre-trained models downloaded,stale/low urgency,2,stale/low urgency,1,0,95645,,,,0,91234a5b8fba4f425c6f6bdefc9d7153851589a8fad87295f77509df01ddc30f
thread:issue:95648,issue #95648: torch needs to SHOW that it support sm_89 even if functionally the same as sm_86,stale/low urgency,2,stale/low urgency,1,0,95648,,,,0,ba21f26c1afedd7bb5222836d01664be3187bffba063ae4f805517577897ac7f
thread:issue:95696,issue #95696: Can only import torch after Tensorflow accessed its gpu device,stale/low urgency,2,stale/low urgency,1,0,95696,,,,0,154735eede6b34aa7e023a7343df2a54a3fbae07846cf1551c99e3b0a0479bcf
thread:issue:9570,issue #9570: [feature request] More options for Fractional Max Pooling,stale/low urgency,2,stale/low urgency,1,0,9570,,,,0,cd6f5286c110d1275d59d42f716789e753989f7c9cfed267673b7e9d1def42f8
thread:issue:95708,issue #95708: Dynamo + MacOS: fatal error: 'omp.h' file not found,stale/low urgency,2,stale/low urgency,1,0,95708,,,,0,06ae679da8b6299c4f07d3c05c13d83e4be46d60211d2c2aa057ce608319f3a0
thread:issue:95711,issue #95711: `copy.deepcopy` does not copy gradients of nn.Parameter,stale/low urgency,2,stale/low urgency,1,0,95711,,,,0,be6596fb98b932b12f219d2ff7e1d9760c348c3a716d364c8e137c7cf74bac7f
thread:issue:95718,issue #95718: functorch.compile.memory_efficient_fusion errors with: RuntimeError: forward() Expected a value of type 'Tensor (inferred)' for argument 'primals_356' but instead found type 'int'.,stale/low urgency,2,stale/low urgency,1,0,95718,,,,0,a251260b74b4cc8189616cd94c26805f692c94b547daa0e654cd971e5ccb7077
thread:issue:95768,issue #95768: Inconsistent behaviour of torch.all(),stale/low urgency,2,stale/low urgency,1,0,95768,,,,0,304bcb4ea7462ec7e58dc25a60dc81a866eede76b86a0eec0df47673a8420140
thread:issue:95779,issue #95779: SymInt'ify _gather_sparse_backward,stale/low urgency,2,stale/low urgency,1,0,95779,,,,0,26b951378fc355aa59403f09f635fdc8476190b81600e03a4521806cdeb30cb8
thread:issue:95785,issue #95785: Fully quantized model (`torch.quantization.convert`) produces incorrect output compared to analytical solution,stale/low urgency,2,stale/low urgency,1,0,95785,,,,0,7e7723fbd6a75f0f2950ee5110cea01b37b9145ddc568a82107a4827659246f5
thread:issue:95786,issue #95786: Static asserts on accessor templates,stale/low urgency,2,stale/low urgency,1,0,95786,,,,0,a1509989b04f48c76c0c95030acd84632cbdd751252edace3329bf13e7dd2fc3
thread:issue:95815,issue #95815: Let Nested Tensor Metadata be cached on GPU,stale/low urgency,2,stale/low urgency,1,0,95815,,,,0,58904fc7a432d9d2414b5d182d18b774919ecdd7f9583ab223769fc7e33b4caa
thread:issue:95872,issue #95872: corrupted size vs prev size error,stale/low urgency,2,stale/low urgency,1,0,95872,,,,0,1dc53ea0ded16bbad39b91ce341b4b09ad2027694168a18752164a117c317b2d
thread:issue:95882,issue #95882: Proposal: accept ellipsis as index in `torch.index_put()` to enable broadcasting,stale/low urgency,2,stale/low urgency,1,0,95882,,,,0,d2e3d8369e5cdc24fb2c0b528ece3336301f57a5c01ecb4356d68fd49aab97ef
thread:issue:95917,issue #95917: torchvision Caltech101 collate_fn error,stale/low urgency,2,stale/low urgency,1,0,95917,,,,0,9bdab679e0419a39247931bd7e0dd3f9182161fca16aaa229bd9a203273ddd42
thread:issue:95921,issue #95921: ROCm distributed flaky on test_distributed_spawn,stale/low urgency,2,stale/low urgency,1,0,95921,,,,0,1132376bc524fc383b4b8ddd378f472ccf788c5767b572e54b0f61200e2d677f
thread:issue:95934,issue #95934: arange bug,stale/low urgency,2,stale/low urgency,1,0,95934,,,,0,5f32b73ca22f9c690f31f23c7d08542b75aaeec78419698ae1ce98421c95b7dc
thread:issue:95945,issue #95945: CPU time performance is unstable,stale/low urgency,2,stale/low urgency,1,0,95945,,,,0,5d172c85694522abb0660b2e19b6169c54ba0cdfec0cf0537c55e1b532243b86
thread:issue:95946,"issue #95946: Build from source, Undefined symbol: c10::detail::maybe_wrap_dim_slow(long long, long long, bool)",stale/low urgency,2,stale/low urgency,1,0,95946,,,,0,4f459b880d83483cbb50d1507796759e803e5af8434bea45964004444888fbe5
thread:issue:95953,issue #95953: The torch.sparse document's typo error,stale/low urgency,2,stale/low urgency,1,0,95953,,,,0,afe992bb8acdbd342e4d7d3e452912e656dfff7be2efac774498a147740a9cf6
thread:issue:95956,issue #95956: [vulkan] missing aten::reflection_pad1d.out operator,stale/low urgency,2,stale/low urgency,1,0,95956,,,,0,0252546cb016c4f9c36219acce732c669be95e635b9dea780303d9a1fe5a4dc9
thread:issue:96073,issue #96073: Pytorch 2.0 installation tutorial does not work under Macbook,stale/low urgency,2,stale/low urgency,1,0,96073,,,,0,a64a6b3f099a23cebdc4ce2dd97f096cfc8f4f6f9e7be4f0926a963c9b6eb41e
thread:issue:96111,issue #96111: Static size boolean masking,stale/low urgency,2,stale/low urgency,1,0,96111,,,,0,e598c9803a8c27d43e7faea75f8e56107a7bb070f3ed93e9c34991ff7e2d4084
thread:issue:96140,issue #96140: mkldnn matmul kernel may be slower than openblas kernel for very small tensor shapes,stale/low urgency,2,stale/low urgency,1,0,96140,,,,0,299f14fea2324a98bdc3676564df2a024f3527ad4a6080bd05b42ad483c8ded1
thread:issue:96187,issue #96187: `gradgradcheck` does not work with sparse inputs.,stale/low urgency,2,stale/low urgency,1,0,96187,,,,0,230ce9b296a1dc133cae7c2982434a0a58633b9b78d3232967586553b7de07e0
thread:issue:96225,issue #96225: [MPS] F.conv1d and F.conv2d produce incorrect gradients when minibatch >= 2^16,stale/low urgency,2,stale/low urgency,1,0,96225,,,,0,f0c19e25b4179c03a3d608202f1ecda8636294a3d3422a458eab0c42a1806910
thread:issue:96236,issue #96236: nn.interpolate scale_factor floors output size with floating,stale/low urgency,2,stale/low urgency,1,0,96236,,,,0,44eeb829924845dd03514f9ff5ab54fb1d523f6a79fec648e1e8329e6c67137e
thread:issue:9624,issue #9624: Request: Pycuda interoperability,stale/low urgency,2,stale/low urgency,1,0,9624,,,,0,2256c7f3274db95e6e6cbb21a28608f5c7080970eb0a0eca2f0bc65d1ad684e2
thread:issue:96265,issue #96265: RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation,stale/low urgency,2,stale/low urgency,1,0,96265,,,,0,19a16149adac33df1209b31ac163d291602c1c6f6050e851a862350053998dd6
thread:issue:96305,"issue #96305: Reuse autograd.grad graph for rapid, repeated gradient calculation",stale/low urgency,2,stale/low urgency,1,0,96305,,,,0,c45036b7459a200a1e74490a829843ee44097462d597fee20c27c2c2b7628bfc
thread:issue:96318,issue #96318: Support managed memory backed dlpack with torch.from_dlpack,stale/low urgency,2,stale/low urgency,1,0,96318,,,,0,b34fc80a0f9855281d972e915c8f013e2938381fe52d37a090e8cf64043fda23
thread:issue:96337,issue #96337: pytreeify decorators,stale/low urgency,2,stale/low urgency,1,0,96337,,,,0,ab582591d759fad97d1cc7e3f85229c2419588b758133d0d25b21a00766f43d1
thread:issue:96386,"issue #96386: [export] ""strict subset of traced input/output"" error when huggingface `ModelOutput` is returned",stale/low urgency,2,stale/low urgency,1,0,96386,,,,0,2d97b088be281c68fc49a4d57a21bb675731a2c252cd27fb1bf1c89ea7e153e8
thread:issue:96396,issue #96396: [dynamo] add hook to modify instructions before/after instructions be generated,stale/low urgency,2,stale/low urgency,1,0,96396,,,,0,a1bccbdbdb25aa830e97a786d7cd1d0c592fc249a96b7362aedabab94fd6511c
thread:issue:96428,issue #96428: Compressed sparse constructor allows mixed `int32/int64` indices which leads to dtype promotion/demotion in conversions.,stale/low urgency,2,stale/low urgency,1,0,96428,,,,0,5e4ba3bef9db142171f79df182e80761c70e3d7437311c061548090d66e00ee1
thread:issue:96447,issue #96447: Information about CPU in `collect_env` is too verbose,stale/low urgency,2,stale/low urgency,1,0,96447,,,,0,52decf3f0bd5fc2dbae27c673df188f17607b24da9a1407910ac38b3c4925a24
thread:issue:96494,issue #96494: The sign of torch.distributions.transforms.PowerTransform seems to be incorrect,stale/low urgency,2,stale/low urgency,1,0,96494,,,,0,d9ff47c5c9d20cb959d2c6823c3c8cc956151a8b165162612254295803052df7
thread:issue:96518,issue #96518: fft should ignore dims with shape 1,stale/low urgency,2,stale/low urgency,1,0,96518,,,,0,b6c20656962668c60b7f9234474b2d2db064c72d84177ef23d912c33f4c878f4
thread:issue:96523,"issue #96523: MPS Backend Doc, model = YourFavoriteNet() not defined",stale/low urgency,2,stale/low urgency,1,0,96523,,,,0,ebe3af417a3436245098d243d221f86a69ce69424e2317a540e9227091e77dd2
thread:issue:96585,issue #96585: Proposal: Disable GC in test suite; GC after every test case,stale/low urgency,2,stale/low urgency,1,0,96585,,,,0,11b33338e48592b3034edd0665dfe2df9cf7c569780e34a86680c3983f28a882
thread:issue:96595,issue #96595: Why doesn't PyTorch install the REAL nvidia cuDNN pip package?,stale/low urgency,2,stale/low urgency,1,0,96595,,,,0,42317735ec12cadffa0dca8d14d6436972ac44e7152adfc19c723ea2774fd497
thread:issue:96608,issue #96608: No matching distribution found for torch==1.13.1+cu117,stale/low urgency,2,stale/low urgency,1,0,96608,,,,0,55074ef53881c1c221d5f972341c05614f1f68215a54120cd05875c18c12a121
thread:issue:96613,issue #96613: The output of torch.histc is incorrect on both CPU and CUDA,stale/low urgency,2,stale/low urgency,1,0,96613,,,,0,5aa8f0020a9077404b742b71203b384672987320dcb70547eec5dbac75866895
thread:issue:96615,issue #96615: Tensor Permutation Along Given Axis,stale/low urgency,2,stale/low urgency,1,0,96615,,,,0,8d5fc375f99b65561f421cc681007846dbb75a8fd372041f731b4077ab6efeba
thread:issue:96629,issue #96629: Dataloader should kill & restart workers when timeout is hit,stale/low urgency,2,stale/low urgency,1,0,96629,,,,0,50f360400bf9f96ad29491c1f4535e4160a11b373da1731ebb7cf75e3fd7732f
thread:issue:96633,issue #96633: Request for adding Warning/Error feature when dropout set to 1.0 in Transformer layer,stale/low urgency,2,stale/low urgency,1,0,96633,,,,0,5cecbf78a4672e4ff87242cf6e5c5f016ecb0d06ee6cbcae67101a7d83a804c4
thread:issue:96637,issue #96637: Not allow force merge when lint fails and not because of broken trunk,stale/low urgency,2,stale/low urgency,1,0,96637,,,,0,e45a4143e7e66f667a94b26b80db56c30e7101d3de7dfc474d265b1c9785668d
thread:issue:96643,issue #96643: PyTorch SGEMV is using 1 single core on AMD CPUs (very slow),stale/low urgency,2,stale/low urgency,1,0,96643,,,,0,950796bcf922c7e0d2f7f47d8b7681d03e3f4286ff9de8438870cb016be9a138
thread:issue:96704,issue #96704: `logical_xx` operations trigger INTERNAL ASSERT FAIL when `input` is complex tensor on cuda and `other` is on cpu,stale/low urgency,2,stale/low urgency,1,0,96704,,,,0,7ed7b9c8365833649378611dbf4391e27719aed99e5aad2e822453819879bc3e
thread:issue:96738,issue #96738: Many padding Module fail memory_format tests,stale/low urgency,2,stale/low urgency,1,0,96738,,,,0,b249acc35c6984cb7860262b5d8f2e72682a2f80bcafbae87e51ac0a12f3916c
thread:issue:96743,issue #96743: Pruning under channels_last format,stale/low urgency,2,stale/low urgency,1,0,96743,,,,0,8143f0db359a645d872e3a2001a113230d075cb9be8f8a3d6ee47fc3e3aad6ca
thread:issue:9676,"issue #9676: [caffe2]running problem, No module named caffe2_pybind11_state_hip",stale/low urgency,2,stale/low urgency,1,0,9676,,,,0,39bf2966b43f5d4c43dca933dc450547f6a46de2685ee2fd41d51e2406b72912
thread:issue:96769,issue #96769: `sparse.mm` triggers INTERNAL ASSERT FAILED when backwarding,stale/low urgency,2,stale/low urgency,1,0,96769,,,,0,51422b172f6a372484ec588d6f5a74b9e0840903f55db1982a8ada0368bdd362
thread:issue:9678,issue #9678: Remove BatchNorm layers once the training is completed.,stale/low urgency,2,stale/low urgency,1,0,9678,,,,0,bae238edf0947cdb93cc24c6528fdb7dc10e6d44437b6703b05ab0af9a62c2b6
thread:issue:96789,issue #96789: `cumprod` triggers INTERNAL ASSERT FAILED when `out` is a tensor on cuda but input is on cpu,stale/low urgency,2,stale/low urgency,1,0,96789,,,,0,d1e6e72f27fa2434d778e8bfabbbe07c2e3b8977e034d313b4a4d7e74fad2171
thread:issue:9681,issue #9681: [feature request] Support for 0-length sequences in packed_sequences,stale/low urgency,2,stale/low urgency,1,0,9681,,,,0,748b73fb2c65c733034642dc5895cfba4c0e5babdffda337a6f40ab790931f5b
thread:issue:96855,issue #96855: Performance Drop for linalg_ldl_factor and ldl_solve,stale/low urgency,2,stale/low urgency,1,0,96855,,,,0,fd3fe6740b91dc1ec7275282a2ff7bafe87a8bd5fff63059f520e7385553f3e0
thread:issue:96883,issue #96883: Expected scalar type Half but found Float when running nn.MultiheadAttention with AMP,stale/low urgency,2,stale/low urgency,1,0,96883,,,,0,948851066c724ab73a25d59d0b5c3af20df335cdd4d9ad51713d82089759718a
thread:issue:96908,issue #96908: Doing inplace on a inplace view of tensor that retains_grad triggers internal assert,stale/low urgency,2,stale/low urgency,1,0,96908,,,,0,c0b64a220dd24f4151b32b8c04ecaf576406bbec5d5c8aa7fd7e14b4b31834c6
thread:issue:96982,issue #96982: [mps] conv1d outputs zeros,stale/low urgency,2,stale/low urgency,1,0,96982,,,,0,c6914e7d972663dbfc216be4279be605323ae42d077e06408a266edee17057f3
thread:issue:9699,issue #9699: [Build error] libcudnn.so: error adding symbols: File in wrong format,stale/low urgency,2,stale/low urgency,1,0,9699,,,,0,e67746a697cc8c029d1940747a18413b52c26f477a040e4d189c6fd2785c0399
thread:issue:97004,issue #97004: Small learning rate with `capturable=True` causes Adam optimizer to blow up model parameters.,stale/low urgency,2,stale/low urgency,1,0,97004,,,,0,74722ab10a0ed67244837e8df7364dd2fa8160ec241d862f3d0b05e8c6357063
thread:issue:9701,issue #9701: [request] speed-up multidim slicing backward,stale/low urgency,2,stale/low urgency,1,0,9701,,,,0,855f4a8af3f0fcd9c3d810a04edd70318635fbc433984ce8ae3a1e8122976a51
thread:issue:97014,issue #97014: deprecate integral and boolean dtype support torch.logit and torch.special.logit,stale/low urgency,2,stale/low urgency,1,0,97014,,,,0,fff7846504a2fec6eaba8554d3e884f8eb283cb498205cbbe2d6d04194f05272
thread:issue:97026,issue #97026: How to get list of all valid devices?,stale/low urgency,2,stale/low urgency,1,0,97026,,,,0,709b50cc6a492734a2a4f2a13d30e4a76c56a42a6f96ff4497418883429596e5
thread:issue:97030,issue #97030: DDP static graph fails for static model,stale/low urgency,2,stale/low urgency,1,0,97030,,,,0,51b8148968b8beee706dbfec7d5c1978b751941e7fbec4571f5bbfebb989ddff
thread:issue:9704,"issue #9704: A newer protobuf should be used, so that cmake install will work when building for mobile",stale/low urgency,2,stale/low urgency,1,0,9704,,,,0,bc86c26863c83c79dcfbc0aeca93c41d9e56afaa25555246b0676c11f42f7822
thread:issue:9708,"issue #9708: [caffe2]mpirun multi-node multi GPU in Distributed mode ,run resnet50_trainer.py get RuntimeError",stale/low urgency,2,stale/low urgency,1,0,9708,,,,0,7a9edde0304be32d415c7c08d82a6e2fdb5ddf7371b886f0d6bb3351e6417e27
thread:issue:97087,issue #97087: A Segment Fault can be triggered in torch.adjoint,stale/low urgency,2,stale/low urgency,1,0,97087,,,,0,18ecce84e392da29aae3310754402d61ef4296d99bc3deafc407d9d98b508b4e
thread:issue:97088,issue #97088: A Segment Fault can be triggered in torch.embedding,stale/low urgency,2,stale/low urgency,1,0,97088,,,,0,ead13ff1cdf6fe0201163a11af57d44055d8a96b9896dc00ce7b7ea2fb12d42e
thread:issue:97106,issue #97106: slow torch import on macos,stale/low urgency,2,stale/low urgency,1,0,97106,,,,0,07405dc82b0c300efe4661a618ad7ed2925656a0d4218711e42be9caea1fde3e
thread:issue:97111,issue #97111: TransformerEncoder truncates output when some token positions are masked by `src_key_padding_mask` across batch,stale/low urgency,2,stale/low urgency,1,0,97111,,,,0,5f7e163bf9101833c18ce594a5c98da9ff8f8f489fd527f51220e11ac68bde78
thread:issue:97114,issue #97114: [Feature Proposal: New Distributed Training Algorithms] LSGD and EASGD,stale/low urgency,2,stale/low urgency,1,0,97114,,,,0,d4d28e06b9b9935b89139041c8e5a1c426016a10b60325abe11d2004d4ef1a45
thread:issue:97118,issue #97118: nn.Conv function to compute conv formula,stale/low urgency,2,stale/low urgency,1,0,97118,,,,0,76a010465699d46ac54434f2597170eb72ef74a7e2158bee3ccfa813da1d80aa
thread:issue:97128,"issue #97128: MultiHeadAttention, fast path broken with `bias=False` or uneven number of heads",stale/low urgency,2,stale/low urgency,1,0,97128,,,,0,a15b9763f6102beee846ab61ced9e9c03a5c71c7c8eadbd122c8cad2f6b00653
thread:issue:97155,issue #97155: Custom recurrent network takes very long to compile for long sequences,stale/low urgency,2,stale/low urgency,1,0,97155,,,,0,f4c7db54ea5f05bf53e652f754ba6448208f80a416bb05bc68d0881c1f3a6ad3
thread:issue:97156,issue #97156: Implement `torch.distributions.Poisson.cdf()`,stale/low urgency,2,stale/low urgency,1,0,97156,,,,0,79041e8c91b91db3a34734ee7117af7e25c9ede73a3ac35f05aba1100f23ef2f
thread:issue:97188,issue #97188: Building LibTorch on Ubuntu with Mac M1,stale/low urgency,2,stale/low urgency,1,0,97188,,,,0,58fa374a7177d66550f2b34d2d3753dcf9ac575f7f6244348ead1a71712bdb74
thread:issue:97189,issue #97189: torch.randint range for torch.int64 dtype seems wrong,stale/low urgency,2,stale/low urgency,1,0,97189,,,,0,7b3bf9d64e55751e8b0482c6e1b5c46631a5f2140c97ac21a206f0989652f569
thread:issue:97236,issue #97236: Non-deterministic results when training a model on GPU with MPS backend,stale/low urgency,2,stale/low urgency,1,0,97236,,,,0,355c892c7e4f64ab839904b93a7e815e170294aacb6c1a55706087b8032dd47a
thread:issue:97268,issue #97268: Desync debugger encounters traceMap error,stale/low urgency,2,stale/low urgency,1,0,97268,,,,0,bc6a32ab7d895356edfd9fc46fde986a355f30ea9c3fe3d682a890f733ad43a8
thread:issue:97283,issue #97283: `jacfwd` fails when computing the gradient for `channels_last` tensor,stale/low urgency,2,stale/low urgency,1,0,97283,,,,0,ca8e31ed1f8533e5a760a95e2a9c71b4e0a76cb2541e95e9ab24669b4ed87982
thread:issue:97286,issue #97286: `jacrev` and `jacfwd` raise an error that `Sparse CSR tensors do not have strides`,stale/low urgency,2,stale/low urgency,1,0,97286,,,,0,5312baddaf14dcb4273ee9f39d3d76cfa1efcc41937f1e961835a2990ce5aac9
thread:issue:97329,"issue #97329: torch.ops.aten.pow(2.0, 3) return unexpected value with complex type",stale/low urgency,2,stale/low urgency,1,0,97329,,,,0,fc94659be0f57b764cbbb2c8927c20cdabd4fbc14f332bf66cfd9b13a174774a
thread:issue:97395,issue #97395: Expanded weights tests broken,stale/low urgency,2,stale/low urgency,1,0,97395,,,,0,c363764368579565c95c64f7175261329fa1944730c8b948320cbd9be3d85b7a
thread:issue:97397,issue #97397: Import fails when both `USE_TENSORPIPE=OFF` and `USE_DISTRIBUTED=ON`.,stale/low urgency,2,stale/low urgency,1,0,97397,,,,0,2668bde2af62adcac0ae325d16211f5f28758341223879cc43c1abc80b67d004
thread:issue:97408,issue #97408: Further memcopy improvement at FX body level,stale/low urgency,2,stale/low urgency,1,0,97408,,,,0,33fd5a93cee9f8a38e9ae463ad15b3b466d05d09671e04a717845ac7aa9bf52e
thread:issue:97421,"issue #97421: After the release of pytorch 2.0.0, the compilation of ACLs is problematic.",stale/low urgency,2,stale/low urgency,1,0,97421,,,,0,97d71da5dd28274c75652659e9403058beb9cd078140ede2d0cbcdcfe3261ec6
thread:issue:97432,issue #97432: suspicious memory leak when increase DataLoader's prefetch_factor and enable pin_memory,stale/low urgency,2,stale/low urgency,1,0,97432,,,,0,79dafeeecd6f34c3937a9a930258f8d2dc8959bc05a907415d8860502730eb3e
thread:issue:97484,issue #97484: [DTensor] Add a unittest to cover default PG condition for DeviceMesh,stale/low urgency,2,stale/low urgency,1,0,97484,,,,0,4096874fcafa7d17fb233210f8d92862ea4eaab354335d2473ae3b3395ef686a
thread:issue:97498,issue #97498: Function Registry for extending collate_fn,stale/low urgency,2,stale/low urgency,1,0,97498,,,,0,82b269c64b12f4e8abada480afa7bc7fb0dcac841cc98192559d57a08185ee3d
thread:issue:97503,issue #97503: torch.cppExtension won't work with wsl2,stale/low urgency,2,stale/low urgency,1,0,97503,,,,0,be8a5894c7c30069757eec574546398f58b31646fd67ea00049d192d4de79425
thread:issue:97504,"issue #97504: Burn benchmark suites into CI docker image. Not only this saves test time, but also it will get rid of occasional model installation failures. (@weiwangmeta )",stale/low urgency,2,stale/low urgency,1,0,97504,,,,0,2b5186f16201e0d35df5200277caa231f2f1e8347bc0628e8a191a0082135e45
thread:issue:9753,issue #9753: dataloader stuck at sched_yield =0,stale/low urgency,2,stale/low urgency,1,0,9753,,,,0,c0801db3ea3b42d12662ca59307b70f134fb55190218317b813dd4068bfa16f4
thread:issue:97539,issue #97539: InfoNCE loss for contrastive learning,stale/low urgency,2,stale/low urgency,1,0,97539,,,,0,35772697cdc2a4260b11f98fce5d2082377166bf9865338ab32a9a99da86cf92
thread:issue:97595,issue #97595: can get_submodule be called within a ScriptFunction ?,stale/low urgency,2,stale/low urgency,1,0,97595,,,,0,e2c0aa56baf0283734b0133d43e8097ed72158bad4c50296babe538e5a03f77d
thread:issue:97597,issue #97597: Insufficient MPS Documentation,stale/low urgency,2,stale/low urgency,1,0,97597,,,,0,dabd4e44f18782a2fee3408c1b44eb21d6fdbb43e9132fbaeda401d21db04382
thread:issue:9761,issue #9761: [caffe2] Questions about conv_transpose && ConvTranspose,stale/low urgency,2,stale/low urgency,1,0,9761,,,,0,5ceedafef38e33871ea1e2b0f81655cc1624076c83813c3a6916db704711e05d
thread:issue:97631,issue #97631: torch.zeros_like on a zero-sized BSR/BSC tensor results invalid tensor,stale/low urgency,2,stale/low urgency,1,0,97631,,,,0,f47a39bb76e13ed9e4065b76204ecb0d4a694186d2837d7d8ff4e3ca1b9c59e3
thread:issue:97638,issue #97638: Regression in jit for f-strings with new lines,stale/low urgency,2,stale/low urgency,1,0,97638,,,,0,6a08c961178281e61321bc83c8dd26842c9bf3271758fd4c33ab78d408dbdae2
thread:issue:97653,issue #97653: transposed 2d copy bfloat16 support,stale/low urgency,2,stale/low urgency,1,0,97653,,,,0,39ca171b9d9da572aa12bed41266adadaeadfb49393cbeffe76c860df40faae4
thread:issue:97656,issue #97656: [bug] Internal assert failed when using pyro,stale/low urgency,2,stale/low urgency,1,0,97656,,,,0,8c1523d952f1919665f295fa5affb71b745364c27a1594926452abdb26742a10
thread:issue:97659,issue #97659: How do I get the original object wrapped by the torch.fx.proxy class？,stale/low urgency,2,stale/low urgency,1,0,97659,,,,0,1d0d75d8abfc5f4f814bfdcdc4ff3a3eab1491729595c16ccd78b041ad32dd20
thread:issue:97670,issue #97670: [FSDP] test model.eval() + keep_low_precision_grads,stale/low urgency,2,stale/low urgency,1,0,97670,,,,0,e3c2ccdf782ec8abb31b18ad48439f99ca87c89142a7678de2ac58b960447b53
thread:issue:97676,issue #97676: Pytorch 2 compile + fsdp + transformers crash,stale/low urgency,2,stale/low urgency,1,0,97676,,,,0,39844fc91e3118a260ae3f46e469bf0dbc546bcf886d1c456d35be1040a579b7
thread:issue:97757,issue #97757: oneDNN 3.0+ support,stale/low urgency,2,stale/low urgency,1,0,97757,,,,0,75959a4102cc9a5049b26355ce076fd90d714f6b83c621ea6650a11c256d0bdb
thread:issue:97760,issue #97760: Test Failure: TestUnaryUfuncsCPU.test_reference_numerics_normal_cos_cpu_float32 on s390x,stale/low urgency,2,stale/low urgency,1,0,97760,,,,0,420d674b1a07d16a0590e55d94e02c95da8590518f7aa700939aaca949a0c399
thread:issue:97761,issue #97761: torch.testing.assert_close: allow check to fail on part on the input,stale/low urgency,2,stale/low urgency,1,0,97761,,,,0,650c1cdd4713cd0d26f01ecf225955930e70d72ba7a56a212c723aac3c5ef635
thread:issue:97802,issue #97802: Some c++ library docstrings incorrectly linked/repeated,stale/low urgency,2,stale/low urgency,1,0,97802,,,,0,67991c42d23483dd61d4eb87907cda1edc7a84ab6fbc060c72990f491610ea3d
thread:issue:97823,issue #97823: Using `param in param_list` can trigger `non-singleton dimension` error?,stale/low urgency,2,stale/low urgency,1,0,97823,,,,0,b98b9a04e91d9fed048d00b64b1981791f1312f7e0713775f0853d419a774c52
thread:issue:97827,issue #97827: Memory leak when saving an input tensor returned as-is if mark_dirty and running with dual tensors,stale/low urgency,2,stale/low urgency,1,0,97827,,,,0,79944f97520764e00e6f4ecad97e4ac5a6c3db824909589fc34e43fb459c7d33
thread:issue:97849,issue #97849: TORCH_LIBRARIES variable leads to undefined reference function error in compiling while using libtorch in c++,stale/low urgency,2,stale/low urgency,1,0,97849,,,,0,35acd00c1a0c3fea5beccbd069fb64fd637f1b8ed52892485ff2d8b2054f8a44
thread:issue:97852,issue #97852: Functionalize crashes on train_step GraphModule,stale/low urgency,2,stale/low urgency,1,0,97852,,,,0,4a73e4601e457a5a089e9b24c5fede48cd701a00a81afab575aa6c20321ea087
thread:issue:97856,issue #97856: make tensor data const correct,stale/low urgency,2,stale/low urgency,1,0,97856,,,,0,4444ef37f040cfd3692d1684a3e4e70f6dcef49a79906a99be828296973d0c64
thread:issue:97902,issue #97902: Multiple model init using OpenMP in c++ does not speed up,stale/low urgency,2,stale/low urgency,1,0,97902,,,,0,31d16e47f25d95abad110c87228ce20c0e5ab5d43d0d0c158db9f563f1dc2d5b
thread:issue:97915,issue #97915: Investigate Lazy{*}Norm{*}d modules no batch dim support,stale/low urgency,2,stale/low urgency,1,0,97915,,,,0,6bca5d2f70c00e05d663c07914daa655c63bc095ed126708f0c711a5f59f018c
thread:issue:97961,issue #97961: [CI/Infra] Record keeping: runner shutdown spike,stale/low urgency,2,stale/low urgency,1,0,97961,,,,0,440f40d58d47248891181013f9bd4140cab0c353f60032968c84f4daabfac836
thread:issue:97966,issue #97966: torch.randn signature is missing generator,stale/low urgency,2,stale/low urgency,1,0,97966,,,,0,a427946e54726bfa08101d3691793d56d3707decc9dbaf5d32d702973960f52b
thread:issue:97991,issue #97991: functional collective should respect the whole mesh,stale/low urgency,2,stale/low urgency,1,0,97991,,,,0,82a053515d63c0e0c569bdc684311e3d8076b16e675dd13255cfbf9c43110374
thread:issue:98002,issue #98002: DataLoader with collate_fn that returns tensors in GPU memory raises warnings when deleted,stale/low urgency,2,stale/low urgency,1,0,98002,,,,0,f6420babbee4e58b0233997b76a7cef2553917cb0cb747af441e00567e2a08ab
thread:issue:98004,issue #98004: matmul with CSR matrix in inference mode throws an exception,stale/low urgency,2,stale/low urgency,1,0,98004,,,,0,168f1da2ad9a45b9950dc5fac014eb74aadfbb8f314afdbb6d7bc1a68e9e41c5
thread:issue:98006,issue #98006: Create a plan on removing conda dependency from CI/CD,stale/low urgency,2,stale/low urgency,1,0,98006,,,,0,fde2116d6548dfc58f00d9bc3387e877defb5c498c0169c531aa817e57ee5c12
thread:issue:98070,issue #98070: Request custom backend device memory Allocator.,stale/low urgency,2,stale/low urgency,1,0,98070,,,,0,c59301ce1faefff0b3c23a93ed344d8b64d4ca36ac57e71d23425fd99c10d70f
thread:issue:98075,issue #98075: Unexpected results with torch.nn.functional.layer_norm,stale/low urgency,2,stale/low urgency,1,0,98075,,,,0,913cf54f544db23dc5ebcfc2bbb6671469cde991b6f482628834c07e8402d0f7
thread:issue:98115,issue #98115: Request to cherrypick a fix into v1.13.1 (v1.8 has a CVE),stale/low urgency,2,stale/low urgency,1,0,98115,,,,0,98b9d44c009d279fc968ec66f2abcb95d96bb698c8d0634836d9cb9a5f7fcc45
thread:issue:98164,issue #98164: forward AD implimentation : _scaled_dot_product_efficient_attention,stale/low urgency,2,stale/low urgency,1,0,98164,,,,0,bab1c3c911ddd6a60452abefbf3769e7d4c5efce9543b98181eb3a1bedf2f9a3
thread:issue:98187,issue #98187: how can i load seperate pytorch_model.bin?,stale/low urgency,2,stale/low urgency,1,0,98187,,,,0,6b926f15a5f804b42fa9c7da0d12f4f966027fb027f58cb80a0e24dfa9eba7a7
thread:issue:98193,"issue #98193: RuntimeError: CUDA error: an illegal memory access was encountered, torch/cuda/streams.py"", line 94, in synchronize",stale/low urgency,2,stale/low urgency,1,0,98193,,,,0,295b397f45407d9c4832f87e402648cb8f9489497071bf0d38e9f935e78ff5ba
thread:issue:98203,"issue #98203: AssertionError: was expecting embedding dimension of 22, but got 1320",stale/low urgency,2,stale/low urgency,1,0,98203,,,,0,1f87d344e3ac6b0e3a9ad3c8d4b970c2cab772697eb91e740095ecc6818a2b55
thread:issue:98208,issue #98208: Add a deterministic version of reflection_pad2d_backward_cuda,stale/low urgency,2,stale/low urgency,1,0,98208,,,,0,9791d1e12275aa024c18d0c4d7c61addfc25a1e32a67c656fa765ced952fe526
thread:issue:98260,issue #98260: Broken mypy check in test_type_hints.py::TestTypeHints::test_doc_examples,stale/low urgency,2,stale/low urgency,1,0,98260,,,,0,a728d14921ef59be6220c4be6c1967c4213580ab9a8be645bd0bf093bae8f7c3
thread:issue:98286,"issue #98286: When I use the DDP model, I use a custom loss function, when the batch size changes during training, the process will be stuck.",stale/low urgency,2,stale/low urgency,1,0,98286,,,,0,2016cbb4a556051d942e1324b764322641022cce3ed0ec5a24f19aad72dc7f6e
thread:issue:98291,issue #98291: Make BetterTransformer implementation non-blocking,stale/low urgency,2,stale/low urgency,1,0,98291,,,,0,21efb5d82e9ddf45240230b795d035458f9c4bcf1ca5bfb9e705b21d323852d6
thread:issue:98338,issue #98338: [functorch] [vmap] tests fail when `_set_vmap_fallback_enabled(False)`.,stale/low urgency,2,stale/low urgency,1,0,98338,,,,0,9be60b90cbe7a86911005a78070d52d5c37980eee093acace1d17a04731b3dbd
thread:issue:98361,issue #98361: pip doesn't install the right version of pytorch when torchtext is involved,stale/low urgency,2,stale/low urgency,1,0,98361,,,,0,4bc6fd80c42fd1486bb4fadaf2c3d89614925b64cb044ab270cbcee820497483
thread:issue:98409,issue #98409: `torch.Tensor.layout` is not documented,stale/low urgency,2,stale/low urgency,1,0,98409,,,,0,d32a8796a38d68884f61b76863c8fb169a18fc005014622c9c5f12f632a4c2b7
thread:issue:98456,issue #98456: README could use link to governance,stale/low urgency,2,stale/low urgency,1,0,98456,,,,0,e144d39a70dc10f6bbe0add99c707c6cec284ea0067512f17e78eb68921065ec
thread:issue:98459,issue #98459: Fail to pass test HAVE_XXX_REGEX while building pytorch,stale/low urgency,2,stale/low urgency,1,0,98459,,,,0,b1c8699e09187f72de95f948fb15cd311f8439aca32d625db31bb96fc6c2e984
thread:issue:98481,issue #98481: Bring CudaPluggableAllocator to feature parity with the Native Allocator,stale/low urgency,2,stale/low urgency,1,0,98481,,,,0,ef7f77e26f15e93a6a238e197231458c4f5750c8bc615bf47683039f9a654ee1
thread:issue:98484,issue #98484: Improvements to FSDP debugability,stale/low urgency,2,stale/low urgency,1,0,98484,,,,0,e7759a8a7c6a321f241984746e5ed33d52919ed57f12971057203e26671d7038
thread:issue:98495,issue #98495: Strided to batch BSR/BSC conversion fails when the number of zeros per block varies while the number of blocks per patch is constant,stale/low urgency,2,stale/low urgency,1,0,98495,,,,0,f8a475a11e4fd09a203d8b1482672c64c8ed0177853c3115a44de12e2a1193a4
thread:issue:98498,"issue #98498: Higher order derivatives not working when setting compute device to `torch.device(""mps"")`",stale/low urgency,2,stale/low urgency,1,0,98498,,,,0,4327997da03df8e249976b1a245ad2d5d9d1d20b5bfacde64f2b2db044675b02
thread:issue:98499,issue #98499: `torch.nn.utils.rnn.unpad_sequence` modifies arguments in-place,stale/low urgency,2,stale/low urgency,1,0,98499,,,,0,3c2bd42e88eeb0de185a47edc2bbcc61ac608ee17cd6b7dcbeacce9f43d6fc67
thread:issue:98515,issue #98515: torch.cond should work with expressions involving SymInt,stale/low urgency,2,stale/low urgency,1,0,98515,,,,0,2cad64cb8027c1159fa07a8e56814ab963188a3cfeeb27744a5682d25291edc2
thread:issue:98557,issue #98557: torch.jit.script codegen warning with cuda and vmap,stale/low urgency,2,stale/low urgency,1,0,98557,,,,0,fc755853ad2c7f1af688686c65907ce1c61e1540449b4a384b8ddb9bb9c62061
thread:issue:98566,issue #98566: `F.interpolate` and `F.grid_sample` - documentation error and bug,stale/low urgency,2,stale/low urgency,1,0,98566,,,,0,7c3712043276feec66532571a2706d7b8b0ec59cfb64e00ea77131064d668329
thread:issue:98587,issue #98587: Cannot use AT_CUDA_DRIVER_CHECK from user code,stale/low urgency,2,stale/low urgency,1,0,98587,,,,0,6dbd9f23547f1e9ac8f7b83e49962fe6b4224ba7ced4e9f3d13133f24395ed58
thread:issue:98600,issue #98600: Wrong illustration in README.md,stale/low urgency,2,stale/low urgency,1,0,98600,,,,0,5a4d48a02978f950aad7f7ca10b1415dee5f880a281bfcdc7c43f2f739882325
thread:issue:98617,issue #98617: Add test/distributed/test_c10d_mpi.py,stale/low urgency,2,stale/low urgency,1,0,98617,,,,0,89af6cdcd4eaf50e4b9eb2ed0620f51c9067ed4bcf99c9f0910136fe6a5b3600
thread:issue:98668,issue #98668: Cannot use `checkpoint_sequential` with `torch.compile`,stale/low urgency,2,stale/low urgency,1,0,98668,,,,0,d43aad7dce3b56b742cc7226fadebceb1fff7af70778fede3e04824c6cb41cad
thread:issue:98673,issue #98673: [ux] Non-blocking tensor constructors,stale/low urgency,2,stale/low urgency,1,0,98673,,,,0,5235b9322ae2de5eeb1da3b936b88cc3793198865dfac447035b3b2ea78a130e
thread:issue:98724,issue #98724: Conflict between ``torch.func`` transformations and ``torch.jit.trace``,stale/low urgency,2,stale/low urgency,1,0,98724,,,,0,1b9f5b041b39b7adc2ef757cf785b381b3312e94f0fac46f81f7b4601abd2fce
thread:issue:98727,issue #98727: Pytorch member variable not working after converting to onnx format,stale/low urgency,2,stale/low urgency,1,0,98727,,,,0,8877b28678c3b02bc0fa0e3226621f7de6c96504c48e895bebaf5616846e2095
thread:issue:9873,"issue #9873: Pytorch is slow when only using CPU, and cannot utilize multicore of CPU",stale/low urgency,2,stale/low urgency,1,0,9873,,,,0,9b4092fe52170774a22f8fe9f2890fee54d1cafa7c768a4d2e806d050336def6
thread:issue:98792,issue #98792: DataLoader doesn't accept non-cpu device for loading.,stale/low urgency,2,stale/low urgency,1,0,98792,,,,0,3916024cca591e0c94e6cc3b13f3de456c6ac4127e10a6de16e3d4c209e844e1
thread:issue:98805,issue #98805: Discrepancy of supported Python versions between Get Started page and index of pre-built binaries for PIP installation,stale/low urgency,2,stale/low urgency,1,0,98805,,,,0,fe42ae279a1bd08bcbbd16dfbcdaeaf964676575ae1057ca271f4235a43cab2c
thread:issue:98808,issue #98808: [FSDP] move up the first all gather,stale/low urgency,2,stale/low urgency,1,0,98808,,,,0,fba2cc9813b65d48a0618f70277d5caeb5ed8b2c35a88e099718747c4cca9f02
thread:issue:9881,issue #9881: cannot open caffe2.lib in VS2015 debug mode,stale/low urgency,2,stale/low urgency,1,0,9881,,,,0,fbeb7cee22215166f62eb0323b7a03d741f99ff44067d72a769a817e28e2503c
thread:issue:98814,issue #98814: autocast does not work properly on embedding module,stale/low urgency,2,stale/low urgency,1,0,98814,,,,0,313c6d985aa4561b790410ba4c56b9a081682d5cecfbf5a08f82c8ee4a070e8d
thread:issue:9882,issue #9882: Unintuitive reduction of mini-batch loss for NLLLoss,stale/low urgency,2,stale/low urgency,1,0,9882,,,,0,4c84833f56fc0d923eca0c68c0453b40ad558c81dcb98f0da14a589070a65efa
thread:issue:9883,issue #9883: NetTest.OperatorWithExecutorHelper intermittently hangs,stale/low urgency,2,stale/low urgency,1,0,9883,,,,0,b0e439f32b164172eaf8bb1db41041eaee4012135f5ae56b5e2bedbf8ee16d63
thread:issue:98860,issue #98860: Sharded Grad Scaler Issue Tracker,stale/low urgency,2,stale/low urgency,1,0,98860,,,,0,de8b42579324f1a3eb1058ef712c8570f060ad3a3408cf29d4c22dce3f9aaf3d
thread:issue:98863,issue #98863: Problem with instalation torch2 on a100+cu12.1,stale/low urgency,2,stale/low urgency,1,0,98863,,,,0,28907b43a0997320f7485fc14cb55c1f6e64d67279b427addb6737aad19aa6d6
thread:issue:98871,issue #98871: Document the user-facing API for the component-level logging system,stale/low urgency,2,stale/low urgency,1,0,98871,,,,0,fbfe35fcff21ecade5bfaf39b3e3421178a2c080a356e94842ca70343e150ec3
thread:issue:98872,issue #98872: Expand component configurable logging system to C++,stale/low urgency,2,stale/low urgency,1,0,98872,,,,0,41a6dde21fa1652683d5ca2b47cfc3f190b37496222b55e82dc05b778a234f58
thread:issue:98904,issue #98904: Test failure: TestCommonCPU.test_python_ref__refs_abs_cpu_complex32,stale/low urgency,2,stale/low urgency,1,0,98904,,,,0,382b54cae6d2a3835fe9b26a131f6ffecb53468b0cb0b7723b3f4c1a9643bdb8
thread:issue:98907,issue #98907: Move template code to header,stale/low urgency,2,stale/low urgency,1,0,98907,,,,0,458544a1e8136c22b6557162e20fe6220f63fa9c6e30c9e123e952eba9d244cf
thread:issue:98917,issue #98917: Change module to module_ in torch/csrc/api/include/torch/python.h,stale/low urgency,2,stale/low urgency,1,0,98917,,,,0,0a075a44c3c13352feacc1cb160775abb59520e51d54402b3001b4da89217bd0
thread:issue:98924,issue #98924: Integrate open device privateuse1 customized method registration,stale/low urgency,2,stale/low urgency,1,0,98924,,,,0,e1479f0babcfc0abe344ce4976f173ed78e0ec5d2a4bf1e742440f71b2aedf74
thread:issue:98925,issue #98925: Request for deterministic support for reflection_pad2d_backward_cuda,stale/low urgency,2,stale/low urgency,1,0,98925,,,,0,e31fb05d8133497ec49e66d000c015247a94727680b0b4748270113ab3e93d76
thread:issue:98929,issue #98929: torch.sparse_csr_tensor() stops gradients,stale/low urgency,2,stale/low urgency,1,0,98929,,,,0,0551775cd38b9303506dc5a70b262f80a9a76f2555f88e3e56eb85158be7cca3
thread:issue:9893,issue #9893: interaction with FindCUDA causes spurious re-cmakes,stale/low urgency,2,stale/low urgency,1,0,9893,,,,0,e0a69cb9a39d35e47d23ed7b5f1a7fee2883f5dfca0e74d5c2761186c68a9d45
thread:issue:98939,"issue #98939: torch.dist with minus norm returns tensor(0.), while with -inf can return result",stale/low urgency,2,stale/low urgency,1,0,98939,,,,0,43d04cb6208a599629b48de9d34c69229068851ad4138f2b52bf61cca98bcd9e
thread:issue:98942,issue #98942: [torch.fx] Upgrade on node info,stale/low urgency,2,stale/low urgency,1,0,98942,,,,0,3f96135359060e844896441cd7e25c38de871adb8bd26ff4e955d99dc3aaba4f
thread:issue:98948,issue #98948: behaviour of `torch.tensor()` changes after editing `Tensor.__getitem__`,stale/low urgency,2,stale/low urgency,1,0,98948,,,,0,7995ae74f9544a4678a37041329439c803a927a6752c3803e56868dcc212d802
thread:issue:98977,issue #98977: [BE] Dedup the functorch skipOps mechanism and the common_method_invocations one,stale/low urgency,2,stale/low urgency,1,0,98977,,,,0,91976dce8cf31330f138bb1cab1fd65e20905e0ba1631575e7d2746c0d29a9c0
thread:issue:9898,issue #9898: [Feature Request] Checkpoint manager,stale/low urgency,2,stale/low urgency,1,0,9898,,,,0,3b87275145137075700ece8815230fbd33660b9f1cfdcdbe77c6958a42761c87
thread:issue:99012,"issue #99012: [BUG]Float32 attention mask not working with torch.autocast(""cpu"")",stale/low urgency,2,stale/low urgency,1,0,99012,,,,0,ee82c97bdcb57273f924c4c14f204b8e100a7b8673397543be7d2997fbb1834b
thread:issue:99023,issue #99023: Dtype changes while going from FX graph -> Torchscript,stale/low urgency,2,stale/low urgency,1,0,99023,,,,0,1a34e1ddb321eccf1a059df1dde4750e4ba2633a3f4b1d08bef56f4c72368963
thread:issue:99025,issue #99025: Is there a way to get the full call stack of pytorch from python to C/C++?,stale/low urgency,2,stale/low urgency,1,0,99025,,,,0,456b09d246d382b4124a6159046a8fe83fce211c4d8e2afd95ab204391a139eb
thread:issue:99035,issue #99035: Issue on building from source: Remove -mfpu=neon option on MacOS with Apple silicon,stale/low urgency,2,stale/low urgency,1,0,99035,,,,0,d3b1904882844d93cdc6905de5a54940f68562da7db799a7c303d8728671e2e6
thread:issue:9912,issue #9912: Stop ifdef'ing out scatter/gather (comm) in libtorch,stale/low urgency,2,stale/low urgency,1,0,9912,,,,0,728d316757f9674bdacf1d1922a8d873258400afb5361a46b4052f6d4337a3a5
thread:issue:99138,issue #99138: torch.cuda.is_available() return False,stale/low urgency,2,stale/low urgency,1,0,99138,,,,0,d8a13894ec34bff562bbed4cccd8d961e78676fea3d5b2f07000ba7c25a46bdf
thread:issue:99143,issue #99143: No documentation to show how to implement aten::view for custom backend,stale/low urgency,2,stale/low urgency,1,0,99143,,,,0,e0894c26e18f51239f7fbffb5eb6c084533f5821fc6590c729e4f5b30fd45e2b
thread:issue:99147,issue #99147: `torch.sparse.sum` backward fails when reducing over dense dimensions.,stale/low urgency,2,stale/low urgency,1,0,99147,,,,0,c82a557ceefdd1a18835e943b2160a05e8d629dc929f97fb55c59b95a6335bb7
thread:issue:99160,"issue #99160: Run ChatRWKV on MBP(intel CPU)+eGPU[rx6800 16G], returna a very big number -9223372036854775808, looks like overflow",stale/low urgency,2,stale/low urgency,1,0,99160,,,,0,9d50e6df3769fa530fa495e5f254a95fe4ef130a8750b9cb0120f374aa7788f3
thread:issue:9921,issue #9921: A serious problem when installing caffe2. Can anyone help me?,stale/low urgency,2,stale/low urgency,1,0,9921,,,,0,11e4e842f78e65870fa64038473d1de2bb6699dd08dea444dc96cd2dc0da4a2a
thread:issue:99225,issue #99225: Libtorch consumes too much memory as 16225,stale/low urgency,2,stale/low urgency,1,0,99225,,,,0,40a0abcf8bb462e50469a0982a837f21aa1750c92e16240158941f4434c36a65
thread:issue:99248,issue #99248: tools PYTHONPATH trick in run_test.py does not work reliably,stale/low urgency,2,stale/low urgency,1,0,99248,,,,0,86ad6261cf6264f56377e24a884dfe47bb5836c5e2f7efc17905daada3143849
thread:issue:99287,issue #99287: Add `TORCH_ASSERT_ONLY_METHOD_OPERATORS` to functorch codebase,stale/low urgency,2,stale/low urgency,1,0,99287,,,,0,cd994eb9432745d719cb507251cc9cdc2b14db605cf878743736050c6f62633a
thread:issue:99297,issue #99297: Slicing and indexing support negative steps,stale/low urgency,2,stale/low urgency,1,0,99297,,,,0,f2eef25cccc0c501e65cd47220935cdc710a1e3fbe61f92d5e7e3f35a9387d73
thread:issue:99299,issue #99299: There has implmenet bug in LTC IrBuilder's MakeSizeMul method.,stale/low urgency,2,stale/low urgency,1,0,99299,,,,0,3746b8660d516d12357988fb83bcfe97cf0d064f3d54825ad86bc979f47b2ebb
thread:issue:99304,issue #99304: Need TransformerEncoder to output attention map,stale/low urgency,2,stale/low urgency,1,0,99304,,,,0,55458c1fbadbf898956527851b8377a42aa3466e229fcbd9be11eb2dca739337
thread:issue:99310,issue #99310: Torch func Documentation for trees,stale/low urgency,2,stale/low urgency,1,0,99310,,,,0,754a49f0a74679962cb19e8804606f9ab27ebb3c8cfd8e66da457a6a5104ca2d
thread:issue:99311,issue #99311: Functorch pytrees with custom iterables,stale/low urgency,2,stale/low urgency,1,0,99311,,,,0,f46dd4c097dded1a429da5efd68a90cd1a2acb375bd0b9b5eec4f78b721c785c
thread:issue:99316,issue #99316: torch.linalg.lstsq doc arguments error,stale/low urgency,2,stale/low urgency,1,0,99316,,,,0,8e65240042dd4433b032df46921f511720acb72d7a22b34870f36f34eeef773b
thread:issue:99359,issue #99359: [c++17] Replace lock_guard with scoped_lock,stale/low urgency,2,stale/low urgency,1,0,99359,,,,0,5cd4166073ad864a3ee2b5ab1bc65a7763b859b589e4db7987b20b6bc957414b
thread:issue:99372,issue #99372: PyTorch 2.0.0 encountered CUDA error: an illegal memory access was encountered,stale/low urgency,2,stale/low urgency,1,0,99372,,,,0,c38cd7fbf7628b8377358de86f9a4267c36a0b1ac2cbfeaface04e8bac3bf7b7
thread:issue:99407,issue #99407: Torch.fx.symbolic_trace removes some of the keys from module state_dict,stale/low urgency,2,stale/low urgency,1,0,99407,,,,0,079ba52919d9b60f103bb2dbdce50c4642dd6378377387b2bab078ec0e6f7ac5
thread:issue:99414,issue #99414: The meta implementation of `index_put` does not do any check,stale/low urgency,2,stale/low urgency,1,0,99414,,,,0,1fecb1d01b035514ed66473e0e6e93ff7acb028168647088e004d1c08f2a9497
thread:issue:99432,issue #99432: [DTensor] parallelize_module failed with nn.Transformer and the PairwiseParallel plan,stale/low urgency,2,stale/low urgency,1,0,99432,,,,0,d082ba0ed7caf00dbba262e4b5c19d7cc05296a2a9c6e2ea5651781e5515c0e6
thread:issue:9945,issue #9945: WERROR=1 doesn't work with FULL_CAFFE2,stale/low urgency,2,stale/low urgency,1,0,9945,,,,0,67f16fb42039b8fb5b8260718a67608515d5f809d7cecc9583132f0083b375d0
thread:issue:9950,issue #9950: Stop passing inplace/out arguments as (non-const) Tensor& to functions,stale/low urgency,2,stale/low urgency,1,0,9950,,,,0,564c7322a4e5e917406faa81982c5e4c908dc1ee3fe976e6bcac38f293f97f66
thread:issue:99515,issue #99515: Support polyphase channelizer,stale/low urgency,2,stale/low urgency,1,0,99515,,,,0,70cb56fc789ad73b3fb3b4f53e1c9cedd952c002badc0fd90b425925b77221c2
thread:issue:9952,issue #9952: [feature request] Add COCOB Optimizer,stale/low urgency,2,stale/low urgency,1,0,9952,,,,0,fbf7896b17cb9f0a3655505867577f4d7eef6d780d3cae50405723ce8e03a41e
thread:issue:99544,issue #99544: Cross compile Pytorch for ARM in Bazel,stale/low urgency,2,stale/low urgency,1,0,99544,,,,0,d602007672e7a6cf1a8eb26c8c0d8d144f794a9f9628c4c26a80b1ff9b905a6c
thread:issue:99556,issue #99556: torch.func.jacrev fails if model contains full_backward_hook,stale/low urgency,2,stale/low urgency,1,0,99556,,,,0,d3b0115bc4f94655e2c62e7805f64abd77cfb63503e16fcf5e2302f1a7a4c0d9
thread:issue:99558,issue #99558: Batching rule not implemented for aten::narrow.Tensor,stale/low urgency,2,stale/low urgency,1,0,99558,,,,0,8ba9429d290b769cca5a9debd3072fa87bcb932387115c13e54f0140def7a622
thread:issue:99584,issue #99584: Training Faster R-CNN model with COCO dataset has been consistently unsuccessful.,stale/low urgency,2,stale/low urgency,1,0,99584,,,,0,f3c8eb0bb31475098550f00d63f999e6c95911b236d79f406f5881df0bb63135
thread:issue:99630,issue #99630: Parameterisation of MultivariateNormal distribution using Cholesky decomposition of precision matrix,stale/low urgency,2,stale/low urgency,1,0,99630,,,,0,2111702b87b3e3c8a0a1ec39b2719dcb372a25068632239fd5a03994487012d7
thread:issue:99637,issue #99637: Torch hangs at import if tensorflow is imported first,stale/low urgency,2,stale/low urgency,1,0,99637,,,,0,a0c81ec65f4ca9e019d0b81502b7030e815e2b2a8fe9cdb2f2a3a7aba1e265cc
thread:issue:99681,issue #99681: `torch.ops.aten.empty` is not discoverable from `dir(torch.ops.aten)` until explicitly calling getattr,stale/low urgency,2,stale/low urgency,1,0,99681,,,,0,a55e2eba143234054266be8ed7984e37ac797583170956d4bc9a5c195b662677
thread:issue:99684,issue #99684: In torchelastic support running worker rank 0 on agent rank 0 consistently,stale/low urgency,2,stale/low urgency,1,0,99684,,,,0,a896a55a108ed5bcacbc5cedecedca9db3c4cda2cb8bd9ded15ae9254c9d9641
thread:issue:99690,"issue #99690: gpu training work well, but cpu training not work",stale/low urgency,2,stale/low urgency,1,0,99690,,,,0,777c0979610b05285febf10947eaeacde6387e047dd40512f59ebf85347f8f0d
thread:issue:99710,issue #99710: pca_lowrank and svd_lowrank broken under automatic mixed precision.,stale/low urgency,2,stale/low urgency,1,0,99710,,,,0,cc1dd41ea0e389922573488ef72ea33f3e654a739f9ca35c5407872cdd28bb54
thread:issue:9975,issue #9975: python caffe2/python/operator_test/activation_ops_test.py Segmentation fault (core dumped),stale/low urgency,2,stale/low urgency,1,0,9975,,,,0,3630c02ce029d89f51a3f55d65535ead4c9685223cf0be47b21f81abff66904d
thread:issue:99770,issue #99770: Deformable Convolution export to onnx,stale/low urgency,2,stale/low urgency,1,0,99770,,,,0,ee431a17cc2a2fd0642e0a7efbf755c4c1efffee35a97033a1af3edd53eaf9b1
thread:issue:99781,issue #99781: 2.0.0+cu118 package missing proper libnvrtc-builtins.so.11.8,stale/low urgency,2,stale/low urgency,1,0,99781,,,,0,00816c7eaa58cf4796a754832b4022bb2102ce4e2ae106bdbc682aecfcefa44a
thread:issue:99790,issue #99790: Cannot compile torch 1.10 in CentOS 7.3,stale/low urgency,2,stale/low urgency,1,0,99790,,,,0,1d2cde9980afc0abae30b6c3e7e071f047fec9dc8cd4b763de1fbe1b60428097
thread:issue:99797,issue #99797: Automatic broadcasting for sparse csr tensors,stale/low urgency,2,stale/low urgency,1,0,99797,,,,0,405a810463cfcf95e68632f42e4ccb0a7b5417e8b89d0f8f4adce00cfbb8a853
thread:issue:99806,issue #99806: `cat` gradgrad tests failing,stale/low urgency,2,stale/low urgency,1,0,99806,,,,0,f14b53cfb0ff4f066f4fff44ab71e708d053f07ae2ead48280f083d63ce56d04
thread:issue:99807,issue #99807: AOTAutograd/Inductor file system cache,stale/low urgency,2,stale/low urgency,1,0,99807,,,,0,50d5cb4e49ea40b49eedf8431ee5cb39ed83ba545640836b46d30ac21ea4cb06
thread:issue:99812,issue #99812: cuda.is_available() error,stale/low urgency,2,stale/low urgency,1,0,99812,,,,0,6552506cc320d15c273e8b739c1bd7c20fe6141add539460d362cbd64e698cb8
thread:issue:99836,issue #99836: NTK notebook calculates wrong object - wrong output dimensions,stale/low urgency,2,stale/low urgency,1,0,99836,,,,0,c965ad5e03938eb0719dda6b8bb8d3bf3922b5f275a73d6f39f8fd5c79d36e46
thread:issue:99999,issue #99999: Runtime Error,stale/low urgency,2,stale/low urgency,1,0,99999,,,,0,5a8afbd14ed100f82d3243c5a1e0555fb1e815cda917daef902668aaae278434