diff --git a/.gitattributes b/.gitattributes index 87f5e49386c830aaa79b1be74ca34d05005323be..d163988ef4348e7b8fab2924b3f0ec06a6b7733b 100644 --- a/.gitattributes +++ b/.gitattributes @@ -90,3 +90,4 @@ tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/_inductor/_ tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/_inductor/__pycache__/codecache.cpython-311.pyc filter=lfs diff=lfs merge=lfs -text tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/_export/serde/__pycache__/serialize.cpython-311.pyc filter=lfs diff=lfs merge=lfs -text tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/_inductor/codegen/__pycache__/triton.cpython-311.pyc filter=lfs diff=lfs merge=lfs -text +tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/_inductor/codegen/__pycache__/cpp_wrapper_cpu.cpython-311.pyc filter=lfs diff=lfs merge=lfs -text diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/_inductor/codegen/__pycache__/cpp_wrapper_cpu.cpython-311.pyc b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/_inductor/codegen/__pycache__/cpp_wrapper_cpu.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..e9e12563f1f68d98a24ce612e9286f046eb8846b --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/_inductor/codegen/__pycache__/cpp_wrapper_cpu.cpython-311.pyc @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d824fb0b23e695740f6d8d3bdd3333dd7acde0422267f3d8a6bbda8e4c5c89dd +size 102551 diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_convolution_double_backward_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_convolution_double_backward_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..832718498aa78cfa4194a440d03d87f4095dcf3a --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_convolution_double_backward_compositeimplicitautograd_dispatch.h @@ -0,0 +1,24 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeimplicitautograd { + +TORCH_API ::std::tuple _convolution_double_backward(const c10::optional & ggI, const c10::optional & ggW, const c10::optional & ggb, const at::Tensor & gO, const at::Tensor & weight, const at::Tensor & self, at::IntArrayRef stride, at::IntArrayRef padding, at::IntArrayRef dilation, bool transposed, at::IntArrayRef output_padding, int64_t groups, ::std::array output_mask); +TORCH_API ::std::tuple _convolution_double_backward_symint(const c10::optional & ggI, const c10::optional & ggW, const c10::optional & ggb, const at::Tensor & gO, const at::Tensor & weight, const at::Tensor & self, c10::SymIntArrayRef stride, c10::SymIntArrayRef padding, c10::SymIntArrayRef dilation, bool transposed, c10::SymIntArrayRef output_padding, c10::SymInt groups, ::std::array output_mask); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_cudnn_rnn_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_cudnn_rnn_native.h new file mode 100644 index 0000000000000000000000000000000000000000..8082f6e7802f1240986089aad819445110880b0c --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_cudnn_rnn_native.h @@ -0,0 +1,22 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API ::std::tuple _cudnn_rnn_out_symint(const at::Tensor & input, at::TensorList weight, int64_t weight_stride0, const c10::optional & weight_buf, const at::Tensor & hx, const c10::optional & cx, int64_t mode, c10::SymInt hidden_size, c10::SymInt proj_size, int64_t num_layers, bool batch_first, double dropout, bool train, bool bidirectional, c10::SymIntArrayRef batch_sizes, const c10::optional & dropout_state, at::Tensor & out0, at::Tensor & out1, at::Tensor & out2, at::Tensor & out3, at::Tensor & out4); +TORCH_API ::std::tuple _cudnn_rnn(const at::Tensor & input, at::TensorList weight, int64_t weight_stride0, const c10::optional & weight_buf, const at::Tensor & hx, const c10::optional & cx, int64_t mode, int64_t hidden_size, int64_t proj_size, int64_t num_layers, bool batch_first, double dropout, bool train, bool bidirectional, at::IntArrayRef batch_sizes, const c10::optional & dropout_state); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_fake_quantize_learnable_per_channel_affine_backward_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_fake_quantize_learnable_per_channel_affine_backward_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..34d22bbeaba66cbcf922a82099ba7b280783eb4e --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_fake_quantize_learnable_per_channel_affine_backward_cuda_dispatch.h @@ -0,0 +1,23 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cuda { + +TORCH_API ::std::tuple _fake_quantize_learnable_per_channel_affine_backward(const at::Tensor & grad, const at::Tensor & self, const at::Tensor & scale, const at::Tensor & zero_point, int64_t axis, int64_t quant_min, int64_t quant_max, double grad_factor=1.0); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_fft_r2c_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_fft_r2c_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..6038230d73ce50265f1b16c36576095804735542 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_fft_r2c_cuda_dispatch.h @@ -0,0 +1,25 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cuda { + +TORCH_API at::Tensor _fft_r2c(const at::Tensor & self, at::IntArrayRef dim, int64_t normalization, bool onesided); +TORCH_API at::Tensor & _fft_r2c_out(at::Tensor & out, const at::Tensor & self, at::IntArrayRef dim, int64_t normalization, bool onesided); +TORCH_API at::Tensor & _fft_r2c_outf(const at::Tensor & self, at::IntArrayRef dim, int64_t normalization, bool onesided, at::Tensor & out); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_copy_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_copy_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..d2d8c7edadcdf38420437602e7bd6d83df5f6968 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_copy_cuda_dispatch.h @@ -0,0 +1,23 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cuda { + +TORCH_API void _foreach_copy_(at::TensorList self, at::TensorList src, bool non_blocking=false); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_copy_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_copy_native.h new file mode 100644 index 0000000000000000000000000000000000000000..1977274b1218721597cb2bac2aa52c7876381384 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_copy_native.h @@ -0,0 +1,24 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API ::std::vector _foreach_copy(at::TensorList self, at::TensorList src, bool non_blocking=false); +TORCH_API void _foreach_copy_out(at::TensorList self, at::TensorList src, bool non_blocking, at::TensorList out); +TORCH_API void foreach_tensor_copy_list_kernel_slow_(at::TensorList self, at::TensorList src, bool non_blocking=false); +TORCH_API void foreach_tensor_copy_list_kernel_cuda_(at::TensorList self, at::TensorList src, bool non_blocking=false); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_log_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_log_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..721482794ec6a38430cf440d75b8f52597b4113b --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_log_cuda_dispatch.h @@ -0,0 +1,24 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cuda { + +TORCH_API ::std::vector _foreach_log(at::TensorList self); +TORCH_API void _foreach_log_(at::TensorList self); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_fused_sdp_choice.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_fused_sdp_choice.h new file mode 100644 index 0000000000000000000000000000000000000000..56a9fc94b935eb83e0f76d9302155bb404cc6069 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_fused_sdp_choice.h @@ -0,0 +1,30 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::_fused_sdp_choice(Tensor query, Tensor key, Tensor value, Tensor? attn_mask=None, float dropout_p=0.0, bool is_causal=False, *, float? scale=None) -> int +inline int64_t _fused_sdp_choice(const at::Tensor & query, const at::Tensor & key, const at::Tensor & value, const c10::optional & attn_mask={}, double dropout_p=0.0, bool is_causal=false, c10::optional scale=c10::nullopt) { + return at::_ops::_fused_sdp_choice::call(query, key, value, attn_mask, dropout_p, is_causal, scale); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_masked_scale_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_masked_scale_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..278750621cc24b58b9c7fe9cebb0e84e16d5f086 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_masked_scale_ops.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { +namespace _ops { + + +struct TORCH_API _masked_scale { + using schema = at::Tensor (const at::Tensor &, const at::Tensor &, double); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_masked_scale") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_masked_scale(Tensor self, Tensor mask, float scale) -> Tensor") + static at::Tensor call(const at::Tensor & self, const at::Tensor & mask, double scale); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & mask, double scale); +}; + +struct TORCH_API _masked_scale_out { + using schema = at::Tensor & (const at::Tensor &, const at::Tensor &, double, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_masked_scale") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_masked_scale.out(Tensor self, Tensor mask, float scale, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, const at::Tensor & mask, double scale, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & mask, double scale, at::Tensor & out); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_nested_get_jagged_dummy_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_nested_get_jagged_dummy_native.h new file mode 100644 index 0000000000000000000000000000000000000000..5127ff1f0312f675412976861009a598591df696 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_nested_get_jagged_dummy_native.h @@ -0,0 +1,20 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_nested_tensor_softmax_with_shape_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_nested_tensor_softmax_with_shape_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..32a10fc8abeb4a3086df473dfb462e5701c4fc74 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_nested_tensor_softmax_with_shape_ops.h @@ -0,0 +1,28 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { +namespace _ops { + + +struct TORCH_API _nested_tensor_softmax_with_shape { + using schema = at::Tensor (const at::Tensor &, const at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_nested_tensor_softmax_with_shape") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_nested_tensor_softmax_with_shape(Tensor self, Tensor query) -> Tensor") + static at::Tensor call(const at::Tensor & self, const at::Tensor & query); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & query); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_nnz_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_nnz_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..f9b84a95717d5b206289f0309e8fb79040d07490 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_nnz_ops.h @@ -0,0 +1,28 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { +namespace _ops { + + +struct TORCH_API _nnz { + using schema = int64_t (const at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_nnz") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_nnz(Tensor self) -> int") + static int64_t call(const at::Tensor & self); + static int64_t redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_shape_as_tensor_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_shape_as_tensor_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..076f0fc1a63d87c504205f467a02a94c684b458e --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_shape_as_tensor_compositeimplicitautograd_dispatch.h @@ -0,0 +1,23 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeimplicitautograd { + +TORCH_API at::Tensor _shape_as_tensor(const at::Tensor & self); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_sparse_coo_tensor_with_dims_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_sparse_coo_tensor_with_dims_native.h new file mode 100644 index 0000000000000000000000000000000000000000..bb65aceaafe86fcbfa2fbef0034244c04dade13c --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_sparse_coo_tensor_with_dims_native.h @@ -0,0 +1,22 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor & _sparse_coo_tensor_with_dims_out(int64_t sparse_dim, int64_t dense_dim, at::IntArrayRef size, at::Tensor & out); +TORCH_API at::Tensor new_with_dims_sparse(int64_t sparse_dim, int64_t dense_dim, at::IntArrayRef size, c10::optional dtype={}, c10::optional layout={}, c10::optional device={}, c10::optional pin_memory={}); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_thnn_differentiable_lstm_cell_backward.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_thnn_differentiable_lstm_cell_backward.h new file mode 100644 index 0000000000000000000000000000000000000000..1431dd27f1c1978b540e8828ab56d423304ee31c --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_thnn_differentiable_lstm_cell_backward.h @@ -0,0 +1,30 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::_thnn_differentiable_lstm_cell_backward(Tensor? grad_hy, Tensor? grad_cy, Tensor input_gates, Tensor hidden_gates, Tensor? input_bias, Tensor? hidden_bias, Tensor cx, Tensor cy) -> (Tensor, Tensor, Tensor, Tensor, Tensor) +inline ::std::tuple _thnn_differentiable_lstm_cell_backward(const c10::optional & grad_hy, const c10::optional & grad_cy, const at::Tensor & input_gates, const at::Tensor & hidden_gates, const c10::optional & input_bias, const c10::optional & hidden_bias, const at::Tensor & cx, const at::Tensor & cy) { + return at::_ops::_thnn_differentiable_lstm_cell_backward::call(grad_hy, grad_cy, input_gates, hidden_gates, input_bias, hidden_bias, cx, cy); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_upsample_nearest_exact3d_meta.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_upsample_nearest_exact3d_meta.h new file mode 100644 index 0000000000000000000000000000000000000000..c247b3f12c14fccb9c39aaa845fa45ed392c07cb --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_upsample_nearest_exact3d_meta.h @@ -0,0 +1,27 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeMetaFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +namespace at { +namespace meta { + +struct TORCH_API structured__upsample_nearest_exact3d : public at::impl::MetaBase { + + + void meta(const at::Tensor & self, at::ArrayRef output_size, c10::optional scales_d, c10::optional scales_h, c10::optional scales_w); +}; + +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/adaptive_avg_pool2d_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/adaptive_avg_pool2d_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..c9a4fe2b177167fc3517555c5993275e4b4d6bb6 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/adaptive_avg_pool2d_cpu_dispatch.h @@ -0,0 +1,26 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cpu { + +TORCH_API at::Tensor & adaptive_avg_pool2d_out(at::Tensor & out, const at::Tensor & self, at::IntArrayRef output_size); +TORCH_API at::Tensor & adaptive_avg_pool2d_outf(const at::Tensor & self, at::IntArrayRef output_size, at::Tensor & out); +TORCH_API at::Tensor & adaptive_avg_pool2d_symint_out(at::Tensor & out, const at::Tensor & self, c10::SymIntArrayRef output_size); +TORCH_API at::Tensor & adaptive_avg_pool2d_symint_outf(const at::Tensor & self, c10::SymIntArrayRef output_size, at::Tensor & out); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/add_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/add_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..ce96b40c3dc3ece19c59c1be2ea5ebaed063678e --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/add_cuda_dispatch.h @@ -0,0 +1,26 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cuda { + +TORCH_API at::Tensor add(const at::Tensor & self, const at::Tensor & other, const at::Scalar & alpha=1); +TORCH_API at::Tensor & add_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & other, const at::Scalar & alpha=1); +TORCH_API at::Tensor & add_outf(const at::Tensor & self, const at::Tensor & other, const at::Scalar & alpha, at::Tensor & out); +TORCH_API at::Tensor & add_(at::Tensor & self, const at::Tensor & other, const at::Scalar & alpha=1); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/avg_pool3d_backward_meta_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/avg_pool3d_backward_meta_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..11e6c6cfd0b677c52071327c45e4da463a98f7bf --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/avg_pool3d_backward_meta_dispatch.h @@ -0,0 +1,25 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace meta { + +TORCH_API at::Tensor avg_pool3d_backward(const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef stride, at::IntArrayRef padding, bool ceil_mode, bool count_include_pad, c10::optional divisor_override); +TORCH_API at::Tensor & avg_pool3d_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef stride, at::IntArrayRef padding, bool ceil_mode, bool count_include_pad, c10::optional divisor_override); +TORCH_API at::Tensor & avg_pool3d_backward_outf(const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef stride, at::IntArrayRef padding, bool ceil_mode, bool count_include_pad, c10::optional divisor_override, at::Tensor & grad_input); + +} // namespace meta +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/bucketize_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/bucketize_native.h new file mode 100644 index 0000000000000000000000000000000000000000..96b681c12465973bdc4a604093627a1616f375ae --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/bucketize_native.h @@ -0,0 +1,27 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor bucketize_cpu(const at::Tensor & self, const at::Tensor & boundaries, bool out_int32=false, bool right=false); +TORCH_API at::Tensor & bucketize_out_cpu(const at::Tensor & self, const at::Tensor & boundaries, bool out_int32, bool right, at::Tensor & out); +TORCH_API at::Tensor bucketize_cuda(const at::Tensor & self, const at::Tensor & boundaries, bool out_int32=false, bool right=false); +TORCH_API at::Tensor & bucketize_out_cuda(const at::Tensor & self, const at::Tensor & boundaries, bool out_int32, bool right, at::Tensor & out); +TORCH_API at::Tensor & bucketize_Scalar_out(const at::Scalar & self, const at::Tensor & boundaries, bool out_int32, bool right, at::Tensor & out); +TORCH_API at::Tensor bucketize_cpu(const at::Scalar & self, const at::Tensor & boundaries, bool out_int32=false, bool right=false); +TORCH_API at::Tensor bucketize_cuda(const at::Scalar & self, const at::Tensor & boundaries, bool out_int32=false, bool right=false); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/column_stack_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/column_stack_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..1ec62b2605fd86e6ce7080d14a1a37545080708d --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/column_stack_ops.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { +namespace _ops { + + +struct TORCH_API column_stack { + using schema = at::Tensor (at::TensorList); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::column_stack") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "column_stack(Tensor[] tensors) -> Tensor") + static at::Tensor call(at::TensorList tensors); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, at::TensorList tensors); +}; + +struct TORCH_API column_stack_out { + using schema = at::Tensor & (at::TensorList, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::column_stack") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "column_stack.out(Tensor[] tensors, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(at::TensorList tensors, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, at::TensorList tensors, at::Tensor & out); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/copysign_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/copysign_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..22162d211ab101ca8358940bff0e40b88dc4303d --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/copysign_compositeexplicitautograd_dispatch.h @@ -0,0 +1,26 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeexplicitautograd { + +TORCH_API at::Tensor copysign(const at::Tensor & self, const at::Scalar & other); +TORCH_API at::Tensor & copysign_out(at::Tensor & out, const at::Tensor & self, const at::Scalar & other); +TORCH_API at::Tensor & copysign_outf(const at::Tensor & self, const at::Scalar & other, at::Tensor & out); +TORCH_API at::Tensor & copysign_(at::Tensor & self, const at::Scalar & other); + +} // namespace compositeexplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cudnn_convolution_relu.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cudnn_convolution_relu.h new file mode 100644 index 0000000000000000000000000000000000000000..874f8b146d77e2ff2cf921680ef3ee115708924d --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cudnn_convolution_relu.h @@ -0,0 +1,91 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::cudnn_convolution_relu(Tensor self, Tensor weight, Tensor? bias, SymInt[] stride, SymInt[] padding, SymInt[] dilation, SymInt groups) -> Tensor +inline at::Tensor cudnn_convolution_relu(const at::Tensor & self, const at::Tensor & weight, const c10::optional & bias, at::IntArrayRef stride, at::IntArrayRef padding, at::IntArrayRef dilation, int64_t groups) { + return at::_ops::cudnn_convolution_relu::call(self, weight, bias, c10::fromIntArrayRefSlow(stride), c10::fromIntArrayRefSlow(padding), c10::fromIntArrayRefSlow(dilation), groups); +} +namespace symint { + template ::value>> + at::Tensor cudnn_convolution_relu(const at::Tensor & self, const at::Tensor & weight, const c10::optional & bias, at::IntArrayRef stride, at::IntArrayRef padding, at::IntArrayRef dilation, int64_t groups) { + return at::_ops::cudnn_convolution_relu::call(self, weight, bias, c10::fromIntArrayRefSlow(stride), c10::fromIntArrayRefSlow(padding), c10::fromIntArrayRefSlow(dilation), groups); + } +} + +// aten::cudnn_convolution_relu(Tensor self, Tensor weight, Tensor? bias, SymInt[] stride, SymInt[] padding, SymInt[] dilation, SymInt groups) -> Tensor +inline at::Tensor cudnn_convolution_relu_symint(const at::Tensor & self, const at::Tensor & weight, const c10::optional & bias, c10::SymIntArrayRef stride, c10::SymIntArrayRef padding, c10::SymIntArrayRef dilation, c10::SymInt groups) { + return at::_ops::cudnn_convolution_relu::call(self, weight, bias, stride, padding, dilation, groups); +} +namespace symint { + template ::value>> + at::Tensor cudnn_convolution_relu(const at::Tensor & self, const at::Tensor & weight, const c10::optional & bias, c10::SymIntArrayRef stride, c10::SymIntArrayRef padding, c10::SymIntArrayRef dilation, c10::SymInt groups) { + return at::_ops::cudnn_convolution_relu::call(self, weight, bias, stride, padding, dilation, groups); + } +} + +// aten::cudnn_convolution_relu.out(Tensor self, Tensor weight, Tensor? bias, SymInt[] stride, SymInt[] padding, SymInt[] dilation, SymInt groups, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & cudnn_convolution_relu_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & weight, const c10::optional & bias, at::IntArrayRef stride, at::IntArrayRef padding, at::IntArrayRef dilation, int64_t groups) { + return at::_ops::cudnn_convolution_relu_out::call(self, weight, bias, c10::fromIntArrayRefSlow(stride), c10::fromIntArrayRefSlow(padding), c10::fromIntArrayRefSlow(dilation), groups, out); +} +namespace symint { + template ::value>> + at::Tensor & cudnn_convolution_relu_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & weight, const c10::optional & bias, at::IntArrayRef stride, at::IntArrayRef padding, at::IntArrayRef dilation, int64_t groups) { + return at::_ops::cudnn_convolution_relu_out::call(self, weight, bias, c10::fromIntArrayRefSlow(stride), c10::fromIntArrayRefSlow(padding), c10::fromIntArrayRefSlow(dilation), groups, out); + } +} + +// aten::cudnn_convolution_relu.out(Tensor self, Tensor weight, Tensor? bias, SymInt[] stride, SymInt[] padding, SymInt[] dilation, SymInt groups, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & cudnn_convolution_relu_outf(const at::Tensor & self, const at::Tensor & weight, const c10::optional & bias, at::IntArrayRef stride, at::IntArrayRef padding, at::IntArrayRef dilation, int64_t groups, at::Tensor & out) { + return at::_ops::cudnn_convolution_relu_out::call(self, weight, bias, c10::fromIntArrayRefSlow(stride), c10::fromIntArrayRefSlow(padding), c10::fromIntArrayRefSlow(dilation), groups, out); +} +namespace symint { + template ::value>> + at::Tensor & cudnn_convolution_relu_outf(const at::Tensor & self, const at::Tensor & weight, const c10::optional & bias, at::IntArrayRef stride, at::IntArrayRef padding, at::IntArrayRef dilation, int64_t groups, at::Tensor & out) { + return at::_ops::cudnn_convolution_relu_out::call(self, weight, bias, c10::fromIntArrayRefSlow(stride), c10::fromIntArrayRefSlow(padding), c10::fromIntArrayRefSlow(dilation), groups, out); + } +} + +// aten::cudnn_convolution_relu.out(Tensor self, Tensor weight, Tensor? bias, SymInt[] stride, SymInt[] padding, SymInt[] dilation, SymInt groups, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & cudnn_convolution_relu_symint_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & weight, const c10::optional & bias, c10::SymIntArrayRef stride, c10::SymIntArrayRef padding, c10::SymIntArrayRef dilation, c10::SymInt groups) { + return at::_ops::cudnn_convolution_relu_out::call(self, weight, bias, stride, padding, dilation, groups, out); +} +namespace symint { + template ::value>> + at::Tensor & cudnn_convolution_relu_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & weight, const c10::optional & bias, c10::SymIntArrayRef stride, c10::SymIntArrayRef padding, c10::SymIntArrayRef dilation, c10::SymInt groups) { + return at::_ops::cudnn_convolution_relu_out::call(self, weight, bias, stride, padding, dilation, groups, out); + } +} + +// aten::cudnn_convolution_relu.out(Tensor self, Tensor weight, Tensor? bias, SymInt[] stride, SymInt[] padding, SymInt[] dilation, SymInt groups, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & cudnn_convolution_relu_symint_outf(const at::Tensor & self, const at::Tensor & weight, const c10::optional & bias, c10::SymIntArrayRef stride, c10::SymIntArrayRef padding, c10::SymIntArrayRef dilation, c10::SymInt groups, at::Tensor & out) { + return at::_ops::cudnn_convolution_relu_out::call(self, weight, bias, stride, padding, dilation, groups, out); +} +namespace symint { + template ::value>> + at::Tensor & cudnn_convolution_relu_outf(const at::Tensor & self, const at::Tensor & weight, const c10::optional & bias, c10::SymIntArrayRef stride, c10::SymIntArrayRef padding, c10::SymIntArrayRef dilation, c10::SymInt groups, at::Tensor & out) { + return at::_ops::cudnn_convolution_relu_out::call(self, weight, bias, stride, padding, dilation, groups, out); + } +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/elu_compositeexplicitautogradnonfunctional_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/elu_compositeexplicitautogradnonfunctional_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..fc01c74e6acb77b8e705ddc3cc05e348bd5bb87c --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/elu_compositeexplicitautogradnonfunctional_dispatch.h @@ -0,0 +1,24 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeexplicitautogradnonfunctional { + +TORCH_API at::Tensor elu(const at::Tensor & self, const at::Scalar & alpha=1, const at::Scalar & scale=1, const at::Scalar & input_scale=1); +TORCH_API at::Tensor & elu_(at::Tensor & self, const at::Scalar & alpha=1, const at::Scalar & scale=1, const at::Scalar & input_scale=1); + +} // namespace compositeexplicitautogradnonfunctional +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/embedding_sparse_backward_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/embedding_sparse_backward_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..ce2ea61127e9199526ff5c41eb2a8a81e3b3a743 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/embedding_sparse_backward_compositeimplicitautograd_dispatch.h @@ -0,0 +1,23 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeimplicitautograd { + +TORCH_API at::Tensor embedding_sparse_backward(const at::Tensor & grad, const at::Tensor & indices, int64_t num_weights, int64_t padding_idx, bool scale_grad_by_freq); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/expand_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/expand_native.h new file mode 100644 index 0000000000000000000000000000000000000000..48202a3295a2eb9d09b5d1b6b68106b4c92e3b5e --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/expand_native.h @@ -0,0 +1,21 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor expand(const at::Tensor & self, at::IntArrayRef size, bool implicit=false); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fake_quantize_per_channel_affine_cachemask.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fake_quantize_per_channel_affine_cachemask.h new file mode 100644 index 0000000000000000000000000000000000000000..6c90b77dbc81b1da4d69f1d222783c655f366078 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fake_quantize_per_channel_affine_cachemask.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::fake_quantize_per_channel_affine_cachemask(Tensor self, Tensor scale, Tensor zero_point, int axis, int quant_min, int quant_max) -> (Tensor output, Tensor mask) +inline ::std::tuple fake_quantize_per_channel_affine_cachemask(const at::Tensor & self, const at::Tensor & scale, const at::Tensor & zero_point, int64_t axis, int64_t quant_min, int64_t quant_max) { + return at::_ops::fake_quantize_per_channel_affine_cachemask::call(self, scale, zero_point, axis, quant_min, quant_max); +} + +// aten::fake_quantize_per_channel_affine_cachemask.out(Tensor self, Tensor scale, Tensor zero_point, int axis, int quant_min, int quant_max, *, Tensor(a!) out0, Tensor(b!) out1) -> (Tensor(a!), Tensor(b!)) +inline ::std::tuple fake_quantize_per_channel_affine_cachemask_out(at::Tensor & out0, at::Tensor & out1, const at::Tensor & self, const at::Tensor & scale, const at::Tensor & zero_point, int64_t axis, int64_t quant_min, int64_t quant_max) { + return at::_ops::fake_quantize_per_channel_affine_cachemask_out::call(self, scale, zero_point, axis, quant_min, quant_max, out0, out1); +} +// aten::fake_quantize_per_channel_affine_cachemask.out(Tensor self, Tensor scale, Tensor zero_point, int axis, int quant_min, int quant_max, *, Tensor(a!) out0, Tensor(b!) out1) -> (Tensor(a!), Tensor(b!)) +inline ::std::tuple fake_quantize_per_channel_affine_cachemask_outf(const at::Tensor & self, const at::Tensor & scale, const at::Tensor & zero_point, int64_t axis, int64_t quant_min, int64_t quant_max, at::Tensor & out0, at::Tensor & out1) { + return at::_ops::fake_quantize_per_channel_affine_cachemask_out::call(self, scale, zero_point, axis, quant_min, quant_max, out0, out1); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/gelu_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/gelu_native.h new file mode 100644 index 0000000000000000000000000000000000000000..f078c216ad85dc390a31335d536298132766791e --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/gelu_native.h @@ -0,0 +1,32 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +namespace at { +namespace native { +struct TORCH_API structured_gelu_out_cpu : public at::meta::structured_gelu { +void impl(const at::Tensor & self, c10::string_view approximate, const at::Tensor & out); +}; +struct TORCH_API structured_gelu_out_cuda : public at::meta::structured_gelu { +void impl(const at::Tensor & self, c10::string_view approximate, const at::Tensor & out); +}; +TORCH_API at::Tensor NestedTensor_gelu(const at::Tensor & self, c10::string_view approximate="none"); +TORCH_API at::Tensor & NestedTensor_gelu_(at::Tensor & self, c10::string_view approximate="none"); +TORCH_API at::Tensor mkldnn_gelu(const at::Tensor & self, c10::string_view approximate="none"); +TORCH_API at::Tensor gelu_quantized_cpu(const at::Tensor & self, c10::string_view approximate="none"); +TORCH_API at::Tensor & gelu_quantized_cpu_(at::Tensor & self, c10::string_view approximate="none"); +TORCH_API at::Tensor gelu_quantized_cuda(const at::Tensor & self, c10::string_view approximate="none"); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/ger_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/ger_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..1329ffbb662862ee4e443e5403305dafdc8a6ea6 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/ger_compositeimplicitautograd_dispatch.h @@ -0,0 +1,25 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeimplicitautograd { + +TORCH_API at::Tensor ger(const at::Tensor & self, const at::Tensor & vec2); +TORCH_API at::Tensor & ger_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & vec2); +TORCH_API at::Tensor & ger_outf(const at::Tensor & self, const at::Tensor & vec2, at::Tensor & out); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardshrink.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardshrink.h new file mode 100644 index 0000000000000000000000000000000000000000..8a4c0618ef0f7e98dbb88fdf2cf6219f938a1c1a --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardshrink.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::hardshrink.out(Tensor self, Scalar lambd=0.5, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & hardshrink_out(at::Tensor & out, const at::Tensor & self, const at::Scalar & lambd=0.5) { + return at::_ops::hardshrink_out::call(self, lambd, out); +} +// aten::hardshrink.out(Tensor self, Scalar lambd=0.5, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & hardshrink_outf(const at::Tensor & self, const at::Scalar & lambd, at::Tensor & out) { + return at::_ops::hardshrink_out::call(self, lambd, out); +} + +// aten::hardshrink(Tensor self, Scalar lambd=0.5) -> Tensor +inline at::Tensor hardshrink(const at::Tensor & self, const at::Scalar & lambd=0.5) { + return at::_ops::hardshrink::call(self, lambd); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardswish_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardswish_native.h new file mode 100644 index 0000000000000000000000000000000000000000..7a138b3fe2c2ed0a5f02878a5608fa94225a2287 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardswish_native.h @@ -0,0 +1,23 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor hardswish(const at::Tensor & self); +TORCH_API at::Tensor & hardswish_out(const at::Tensor & self, at::Tensor & out); +TORCH_API at::Tensor & hardswish_(at::Tensor & self); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/huber_loss_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/huber_loss_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..320b0b3887a4698ce6e5eb634f4f5a0cc9c78130 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/huber_loss_ops.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { +namespace _ops { + + +struct TORCH_API huber_loss_out { + using schema = at::Tensor & (const at::Tensor &, const at::Tensor &, int64_t, double, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::huber_loss") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "huber_loss.out(Tensor self, Tensor target, int reduction=Mean, float delta=1.0, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, const at::Tensor & target, int64_t reduction, double delta, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & target, int64_t reduction, double delta, at::Tensor & out); +}; + +struct TORCH_API huber_loss { + using schema = at::Tensor (const at::Tensor &, const at::Tensor &, int64_t, double); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::huber_loss") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "huber_loss(Tensor self, Tensor target, int reduction=Mean, float delta=1.0) -> Tensor") + static at::Tensor call(const at::Tensor & self, const at::Tensor & target, int64_t reduction, double delta); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & target, int64_t reduction, double delta); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/i0.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/i0.h new file mode 100644 index 0000000000000000000000000000000000000000..2b64824d76ebf22dc922f7c20809a5d55051edae --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/i0.h @@ -0,0 +1,44 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::i0(Tensor self) -> Tensor +inline at::Tensor i0(const at::Tensor & self) { + return at::_ops::i0::call(self); +} + +// aten::i0_(Tensor(a!) self) -> Tensor(a!) +inline at::Tensor & i0_(at::Tensor & self) { + return at::_ops::i0_::call(self); +} + +// aten::i0.out(Tensor self, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & i0_out(at::Tensor & out, const at::Tensor & self) { + return at::_ops::i0_out::call(self, out); +} +// aten::i0.out(Tensor self, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & i0_outf(const at::Tensor & self, at::Tensor & out) { + return at::_ops::i0_out::call(self, out); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/indices_copy_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/indices_copy_native.h new file mode 100644 index 0000000000000000000000000000000000000000..513f05fdfdf6bed10c27c86a0c7603c79a7f3e3b --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/indices_copy_native.h @@ -0,0 +1,22 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor & indices_copy_out(const at::Tensor & self, at::Tensor & out); +TORCH_API at::Tensor indices_copy(const at::Tensor & self); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/lcm_compositeexplicitautogradnonfunctional_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/lcm_compositeexplicitautogradnonfunctional_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..6b4f61bade0489e636484d8c4e6498528984f45a --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/lcm_compositeexplicitautogradnonfunctional_dispatch.h @@ -0,0 +1,24 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeexplicitautogradnonfunctional { + +TORCH_API at::Tensor lcm(const at::Tensor & self, const at::Tensor & other); +TORCH_API at::Tensor & lcm_(at::Tensor & self, const at::Tensor & other); + +} // namespace compositeexplicitautogradnonfunctional +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_lu_solve_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_lu_solve_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..0cbd5eeeda1c426ab17da1833e0eaa768604858a --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_lu_solve_ops.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { +namespace _ops { + + +struct TORCH_API linalg_lu_solve { + using schema = at::Tensor (const at::Tensor &, const at::Tensor &, const at::Tensor &, bool, bool); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::linalg_lu_solve") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "linalg_lu_solve(Tensor LU, Tensor pivots, Tensor B, *, bool left=True, bool adjoint=False) -> Tensor") + static at::Tensor call(const at::Tensor & LU, const at::Tensor & pivots, const at::Tensor & B, bool left, bool adjoint); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & LU, const at::Tensor & pivots, const at::Tensor & B, bool left, bool adjoint); +}; + +struct TORCH_API linalg_lu_solve_out { + using schema = at::Tensor & (const at::Tensor &, const at::Tensor &, const at::Tensor &, bool, bool, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::linalg_lu_solve") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "linalg_lu_solve.out(Tensor LU, Tensor pivots, Tensor B, *, bool left=True, bool adjoint=False, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & LU, const at::Tensor & pivots, const at::Tensor & B, bool left, bool adjoint, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & LU, const at::Tensor & pivots, const at::Tensor & B, bool left, bool adjoint, at::Tensor & out); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_solve_triangular.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_solve_triangular.h new file mode 100644 index 0000000000000000000000000000000000000000..6c9c57d1534f1a4020703f74b94b335316e21e89 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_solve_triangular.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::linalg_solve_triangular.out(Tensor self, Tensor B, *, bool upper, bool left=True, bool unitriangular=False, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & linalg_solve_triangular_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & B, bool upper, bool left=true, bool unitriangular=false) { + return at::_ops::linalg_solve_triangular_out::call(self, B, upper, left, unitriangular, out); +} +// aten::linalg_solve_triangular.out(Tensor self, Tensor B, *, bool upper, bool left=True, bool unitriangular=False, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & linalg_solve_triangular_outf(const at::Tensor & self, const at::Tensor & B, bool upper, bool left, bool unitriangular, at::Tensor & out) { + return at::_ops::linalg_solve_triangular_out::call(self, B, upper, left, unitriangular, out); +} + +// aten::linalg_solve_triangular(Tensor self, Tensor B, *, bool upper, bool left=True, bool unitriangular=False) -> Tensor +inline at::Tensor linalg_solve_triangular(const at::Tensor & self, const at::Tensor & B, bool upper, bool left=true, bool unitriangular=false) { + return at::_ops::linalg_solve_triangular::call(self, B, upper, left, unitriangular); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_vecdot.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_vecdot.h new file mode 100644 index 0000000000000000000000000000000000000000..ce313cdcd03695d4a3c912ca89646c300995a3da --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_vecdot.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::linalg_vecdot(Tensor x, Tensor y, *, int dim=-1) -> Tensor +inline at::Tensor linalg_vecdot(const at::Tensor & x, const at::Tensor & y, int64_t dim=-1) { + return at::_ops::linalg_vecdot::call(x, y, dim); +} + +// aten::linalg_vecdot.out(Tensor x, Tensor y, *, int dim=-1, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & linalg_vecdot_out(at::Tensor & out, const at::Tensor & x, const at::Tensor & y, int64_t dim=-1) { + return at::_ops::linalg_vecdot_out::call(x, y, dim, out); +} +// aten::linalg_vecdot.out(Tensor x, Tensor y, *, int dim=-1, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & linalg_vecdot_outf(const at::Tensor & x, const at::Tensor & y, int64_t dim, at::Tensor & out) { + return at::_ops::linalg_vecdot_out::call(x, y, dim, out); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/log_meta_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/log_meta_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..a10faffd3437dcb3625c07aec7289e30fdd34a06 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/log_meta_dispatch.h @@ -0,0 +1,26 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace meta { + +TORCH_API at::Tensor log(const at::Tensor & self); +TORCH_API at::Tensor & log_out(at::Tensor & out, const at::Tensor & self); +TORCH_API at::Tensor & log_outf(const at::Tensor & self, at::Tensor & out); +TORCH_API at::Tensor & log_(at::Tensor & self); + +} // namespace meta +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/log_normal_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/log_normal_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..7f058396ee17549b3d40c9b2323e16fc7f7651a4 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/log_normal_cpu_dispatch.h @@ -0,0 +1,23 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cpu { + +TORCH_API at::Tensor & log_normal_(at::Tensor & self, double mean=1, double std=2, c10::optional generator=c10::nullopt); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logical_xor_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logical_xor_native.h new file mode 100644 index 0000000000000000000000000000000000000000..6f29c969c6691dac334b74e332fd6b509f1210fb --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logical_xor_native.h @@ -0,0 +1,23 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor logical_xor(const at::Tensor & self, const at::Tensor & other); +TORCH_API at::Tensor & logical_xor_(at::Tensor & self, const at::Tensor & other); +TORCH_API at::Tensor & logical_xor_out(const at::Tensor & self, const at::Tensor & other, at::Tensor & out); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logit_meta_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logit_meta_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..bd30f1815b66f275c817ce71a9375f6d87d80c7c --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logit_meta_dispatch.h @@ -0,0 +1,23 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace meta { + +TORCH_API at::Tensor & logit_(at::Tensor & self, c10::optional eps=c10::nullopt); + +} // namespace meta +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/lstm_cell_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/lstm_cell_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..940e3ca7dbfb15a15c4d6634111c33c06f367c92 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/lstm_cell_compositeimplicitautograd_dispatch.h @@ -0,0 +1,23 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeimplicitautograd { + +TORCH_API ::std::tuple lstm_cell(const at::Tensor & input, at::TensorList hx, const at::Tensor & w_ih, const at::Tensor & w_hh, const c10::optional & b_ih={}, const c10::optional & b_hh={}); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mH.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mH.h new file mode 100644 index 0000000000000000000000000000000000000000..a5f142b402135e9b421e210c4713ff3253f7ef70 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mH.h @@ -0,0 +1,26 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/max_unpool2d_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/max_unpool2d_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..9b201d8b06bfb94efa8d1e10f00fe79fbd30649a --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/max_unpool2d_ops.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { +namespace _ops { + + +struct TORCH_API max_unpool2d_out { + using schema = at::Tensor & (const at::Tensor &, const at::Tensor &, c10::SymIntArrayRef, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::max_unpool2d") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "max_unpool2d.out(Tensor self, Tensor indices, SymInt[2] output_size, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, const at::Tensor & indices, c10::SymIntArrayRef output_size, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & indices, c10::SymIntArrayRef output_size, at::Tensor & out); +}; + +struct TORCH_API max_unpool2d { + using schema = at::Tensor (const at::Tensor &, const at::Tensor &, c10::SymIntArrayRef); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::max_unpool2d") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "max_unpool2d(Tensor self, Tensor indices, SymInt[2] output_size) -> Tensor") + static at::Tensor call(const at::Tensor & self, const at::Tensor & indices, c10::SymIntArrayRef output_size); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & indices, c10::SymIntArrayRef output_size); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/min_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/min_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..769b7133a98b9e6ee4aa0bca9354e036f5fbc9d3 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/min_cpu_dispatch.h @@ -0,0 +1,28 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cpu { + +TORCH_API ::std::tuple min(const at::Tensor & self, int64_t dim, bool keepdim=false); +TORCH_API ::std::tuple min_out(at::Tensor & min, at::Tensor & min_indices, const at::Tensor & self, int64_t dim, bool keepdim=false); +TORCH_API ::std::tuple min_outf(const at::Tensor & self, int64_t dim, bool keepdim, at::Tensor & min, at::Tensor & min_indices); +TORCH_API at::Tensor min(const at::Tensor & self); +TORCH_API at::Tensor & min_out(at::Tensor & out, const at::Tensor & self); +TORCH_API at::Tensor & min_outf(const at::Tensor & self, at::Tensor & out); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/miopen_convolution_relu_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/miopen_convolution_relu_native.h new file mode 100644 index 0000000000000000000000000000000000000000..a9ab0aea6df2d409d5f5b04a84443a341a782e02 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/miopen_convolution_relu_native.h @@ -0,0 +1,21 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor miopen_convolution_relu(const at::Tensor & self, const at::Tensor & weight, const c10::optional & bias, at::IntArrayRef stride, at::IntArrayRef padding, at::IntArrayRef dilation, int64_t groups); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mish_backward_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mish_backward_native.h new file mode 100644 index 0000000000000000000000000000000000000000..1e5e927d0409f8ac90a1a5430b01ebd2576d3620 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mish_backward_native.h @@ -0,0 +1,22 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor math_mish_backward(const at::Tensor & grad_output, const at::Tensor & self); +TORCH_API at::Tensor mish_backward(const at::Tensor & grad_output, const at::Tensor & self); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/nextafter_meta_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/nextafter_meta_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..203d93e11f7c705952dadabd3b0c7c32979a5fd5 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/nextafter_meta_dispatch.h @@ -0,0 +1,26 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace meta { + +TORCH_API at::Tensor nextafter(const at::Tensor & self, const at::Tensor & other); +TORCH_API at::Tensor & nextafter_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & other); +TORCH_API at::Tensor & nextafter_outf(const at::Tensor & self, const at::Tensor & other, at::Tensor & out); +TORCH_API at::Tensor & nextafter_(at::Tensor & self, const at::Tensor & other); + +} // namespace meta +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/normal_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/normal_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..9047c6838522f22bff4f5fa1dc45247132554a05 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/normal_ops.h @@ -0,0 +1,138 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { +namespace _ops { + + +struct TORCH_API normal_ { + using schema = at::Tensor & (at::Tensor &, double, double, c10::optional); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::normal_") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "normal_(Tensor(a!) self, float mean=0, float std=1, *, Generator? generator=None) -> Tensor(a!)") + static at::Tensor & call(at::Tensor & self, double mean, double std, c10::optional generator); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, at::Tensor & self, double mean, double std, c10::optional generator); +}; + +struct TORCH_API normal_functional { + using schema = at::Tensor (const at::Tensor &, double, double, c10::optional); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::normal_functional") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "normal_functional(Tensor self, float mean=0, float std=1, *, Generator? generator=None) -> Tensor") + static at::Tensor call(const at::Tensor & self, double mean, double std, c10::optional generator); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, double mean, double std, c10::optional generator); +}; + +struct TORCH_API normal_Tensor_float_out { + using schema = at::Tensor & (const at::Tensor &, double, c10::optional, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::normal") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Tensor_float_out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "normal.Tensor_float_out(Tensor mean, float std=1, *, Generator? generator=None, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & mean, double std, c10::optional generator, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & mean, double std, c10::optional generator, at::Tensor & out); +}; + +struct TORCH_API normal_Tensor_float { + using schema = at::Tensor (const at::Tensor &, double, c10::optional); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::normal") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Tensor_float") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "normal.Tensor_float(Tensor mean, float std=1, *, Generator? generator=None) -> Tensor") + static at::Tensor call(const at::Tensor & mean, double std, c10::optional generator); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & mean, double std, c10::optional generator); +}; + +struct TORCH_API normal_float_Tensor_out { + using schema = at::Tensor & (double, const at::Tensor &, c10::optional, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::normal") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "float_Tensor_out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "normal.float_Tensor_out(float mean, Tensor std, *, Generator? generator=None, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(double mean, const at::Tensor & std, c10::optional generator, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, double mean, const at::Tensor & std, c10::optional generator, at::Tensor & out); +}; + +struct TORCH_API normal_float_Tensor { + using schema = at::Tensor (double, const at::Tensor &, c10::optional); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::normal") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "float_Tensor") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "normal.float_Tensor(float mean, Tensor std, *, Generator? generator=None) -> Tensor") + static at::Tensor call(double mean, const at::Tensor & std, c10::optional generator); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, double mean, const at::Tensor & std, c10::optional generator); +}; + +struct TORCH_API normal_Tensor_Tensor_out { + using schema = at::Tensor & (const at::Tensor &, const at::Tensor &, c10::optional, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::normal") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Tensor_Tensor_out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "normal.Tensor_Tensor_out(Tensor mean, Tensor std, *, Generator? generator=None, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & mean, const at::Tensor & std, c10::optional generator, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & mean, const at::Tensor & std, c10::optional generator, at::Tensor & out); +}; + +struct TORCH_API normal_Tensor_Tensor { + using schema = at::Tensor (const at::Tensor &, const at::Tensor &, c10::optional); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::normal") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Tensor_Tensor") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "normal.Tensor_Tensor(Tensor mean, Tensor std, *, Generator? generator=None) -> Tensor") + static at::Tensor call(const at::Tensor & mean, const at::Tensor & std, c10::optional generator); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & mean, const at::Tensor & std, c10::optional generator); +}; + +struct TORCH_API normal_float_float { + using schema = at::Tensor (double, double, c10::SymIntArrayRef, c10::optional, c10::optional, c10::optional, c10::optional, c10::optional); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::normal") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "float_float") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "normal.float_float(float mean, float std, SymInt[] size, *, Generator? generator=None, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor") + static at::Tensor call(double mean, double std, c10::SymIntArrayRef size, c10::optional generator, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, double mean, double std, c10::SymIntArrayRef size, c10::optional generator, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory); +}; + +struct TORCH_API normal_float_float_out { + using schema = at::Tensor & (double, double, c10::SymIntArrayRef, c10::optional, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::normal") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "float_float_out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "normal.float_float_out(float mean, float std, SymInt[] size, *, Generator? generator=None, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(double mean, double std, c10::SymIntArrayRef size, c10::optional generator, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, double mean, double std, c10::SymIntArrayRef size, c10::optional generator, at::Tensor & out); +}; + +struct TORCH_API normal_out { + using schema = at::Tensor & (const at::Tensor &, double, double, c10::optional, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::normal") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "normal.out(Tensor self, float mean=0, float std=1, *, Generator? generator=None, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, double mean, double std, c10::optional generator, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, double mean, double std, c10::optional generator, at::Tensor & out); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/pdist_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/pdist_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..74c7d4f6cd8c10cf3532824661bfb25000cb783b --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/pdist_ops.h @@ -0,0 +1,28 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { +namespace _ops { + + +struct TORCH_API pdist { + using schema = at::Tensor (const at::Tensor &, double); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::pdist") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "pdist(Tensor self, float p=2) -> Tensor") + static at::Tensor call(const at::Tensor & self, double p); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, double p); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/quantize_per_tensor_dynamic_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/quantize_per_tensor_dynamic_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..33d91236c7d078fb42c95b2a328596fe1c0fe5b4 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/quantize_per_tensor_dynamic_cuda_dispatch.h @@ -0,0 +1,23 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cuda { + +TORCH_API at::Tensor quantize_per_tensor_dynamic(const at::Tensor & self, at::ScalarType dtype, bool reduce_range); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/rad2deg_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/rad2deg_native.h new file mode 100644 index 0000000000000000000000000000000000000000..99a7f0ab43bc1105c9eb6a61bda1fcd6eff979e3 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/rad2deg_native.h @@ -0,0 +1,29 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor rad2deg(const at::Tensor & self); +TORCH_API at::Tensor & rad2deg_out(const at::Tensor & self, at::Tensor & out); +TORCH_API at::Tensor & rad2deg_(at::Tensor & self); +TORCH_API at::Tensor rad2deg_sparse(const at::Tensor & self); +TORCH_API at::Tensor & rad2deg_sparse_out(const at::Tensor & self, at::Tensor & out); +TORCH_API at::Tensor & rad2deg_sparse_(at::Tensor & self); +TORCH_API at::Tensor rad2deg_sparse_csr(const at::Tensor & self); +TORCH_API at::Tensor & rad2deg_sparse_csr_out(const at::Tensor & self, at::Tensor & out); +TORCH_API at::Tensor & rad2deg_sparse_csr_(at::Tensor & self); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/randint_like.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/randint_like.h new file mode 100644 index 0000000000000000000000000000000000000000..4d9b50d01f3c9d7948f1e57b5037990c786c5050 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/randint_like.h @@ -0,0 +1,201 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::randint_like(Tensor self, SymInt high, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None, MemoryFormat? memory_format=None) -> Tensor +inline at::Tensor randint_like(const at::Tensor & self, int64_t high, at::TensorOptions options={}, c10::optional memory_format=c10::nullopt) { + return at::_ops::randint_like::call(self, high, c10::optTypeMetaToScalarType(options.dtype_opt()), options.layout_opt(), options.device_opt(), options.pinned_memory_opt(), c10::impl::check_tensor_options_and_extract_memory_format(options, memory_format)); +} +namespace symint { + template ::value>> + at::Tensor randint_like(const at::Tensor & self, int64_t high, at::TensorOptions options={}, c10::optional memory_format=c10::nullopt) { + return at::_ops::randint_like::call(self, high, c10::optTypeMetaToScalarType(options.dtype_opt()), options.layout_opt(), options.device_opt(), options.pinned_memory_opt(), c10::impl::check_tensor_options_and_extract_memory_format(options, memory_format)); + } +} + +// aten::randint_like(Tensor self, SymInt high, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None, MemoryFormat? memory_format=None) -> Tensor +inline at::Tensor randint_like(const at::Tensor & self, int64_t high, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory, c10::optional memory_format) { + return at::_ops::randint_like::call(self, high, dtype, layout, device, pin_memory, memory_format); +} +namespace symint { + template ::value>> + at::Tensor randint_like(const at::Tensor & self, int64_t high, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory, c10::optional memory_format) { + return at::_ops::randint_like::call(self, high, dtype, layout, device, pin_memory, memory_format); + } +} + +// aten::randint_like(Tensor self, SymInt high, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None, MemoryFormat? memory_format=None) -> Tensor +inline at::Tensor randint_like_symint(const at::Tensor & self, c10::SymInt high, at::TensorOptions options={}, c10::optional memory_format=c10::nullopt) { + return at::_ops::randint_like::call(self, high, c10::optTypeMetaToScalarType(options.dtype_opt()), options.layout_opt(), options.device_opt(), options.pinned_memory_opt(), c10::impl::check_tensor_options_and_extract_memory_format(options, memory_format)); +} +namespace symint { + template ::value>> + at::Tensor randint_like(const at::Tensor & self, c10::SymInt high, at::TensorOptions options={}, c10::optional memory_format=c10::nullopt) { + return at::_ops::randint_like::call(self, high, c10::optTypeMetaToScalarType(options.dtype_opt()), options.layout_opt(), options.device_opt(), options.pinned_memory_opt(), c10::impl::check_tensor_options_and_extract_memory_format(options, memory_format)); + } +} + +// aten::randint_like(Tensor self, SymInt high, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None, MemoryFormat? memory_format=None) -> Tensor +inline at::Tensor randint_like_symint(const at::Tensor & self, c10::SymInt high, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory, c10::optional memory_format) { + return at::_ops::randint_like::call(self, high, dtype, layout, device, pin_memory, memory_format); +} +namespace symint { + template ::value>> + at::Tensor randint_like(const at::Tensor & self, c10::SymInt high, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory, c10::optional memory_format) { + return at::_ops::randint_like::call(self, high, dtype, layout, device, pin_memory, memory_format); + } +} + +// aten::randint_like.low_dtype(Tensor self, SymInt low, SymInt high, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None, MemoryFormat? memory_format=None) -> Tensor +inline at::Tensor randint_like(const at::Tensor & self, int64_t low, int64_t high, at::TensorOptions options={}, c10::optional memory_format=c10::nullopt) { + return at::_ops::randint_like_low_dtype::call(self, low, high, c10::optTypeMetaToScalarType(options.dtype_opt()), options.layout_opt(), options.device_opt(), options.pinned_memory_opt(), c10::impl::check_tensor_options_and_extract_memory_format(options, memory_format)); +} +namespace symint { + template ::value>> + at::Tensor randint_like(const at::Tensor & self, int64_t low, int64_t high, at::TensorOptions options={}, c10::optional memory_format=c10::nullopt) { + return at::_ops::randint_like_low_dtype::call(self, low, high, c10::optTypeMetaToScalarType(options.dtype_opt()), options.layout_opt(), options.device_opt(), options.pinned_memory_opt(), c10::impl::check_tensor_options_and_extract_memory_format(options, memory_format)); + } +} + +// aten::randint_like.low_dtype(Tensor self, SymInt low, SymInt high, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None, MemoryFormat? memory_format=None) -> Tensor +inline at::Tensor randint_like(const at::Tensor & self, int64_t low, int64_t high, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory, c10::optional memory_format) { + return at::_ops::randint_like_low_dtype::call(self, low, high, dtype, layout, device, pin_memory, memory_format); +} +namespace symint { + template ::value>> + at::Tensor randint_like(const at::Tensor & self, int64_t low, int64_t high, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory, c10::optional memory_format) { + return at::_ops::randint_like_low_dtype::call(self, low, high, dtype, layout, device, pin_memory, memory_format); + } +} + +// aten::randint_like.low_dtype(Tensor self, SymInt low, SymInt high, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None, MemoryFormat? memory_format=None) -> Tensor +inline at::Tensor randint_like_symint(const at::Tensor & self, c10::SymInt low, c10::SymInt high, at::TensorOptions options={}, c10::optional memory_format=c10::nullopt) { + return at::_ops::randint_like_low_dtype::call(self, low, high, c10::optTypeMetaToScalarType(options.dtype_opt()), options.layout_opt(), options.device_opt(), options.pinned_memory_opt(), c10::impl::check_tensor_options_and_extract_memory_format(options, memory_format)); +} +namespace symint { + template ::value>> + at::Tensor randint_like(const at::Tensor & self, c10::SymInt low, c10::SymInt high, at::TensorOptions options={}, c10::optional memory_format=c10::nullopt) { + return at::_ops::randint_like_low_dtype::call(self, low, high, c10::optTypeMetaToScalarType(options.dtype_opt()), options.layout_opt(), options.device_opt(), options.pinned_memory_opt(), c10::impl::check_tensor_options_and_extract_memory_format(options, memory_format)); + } +} + +// aten::randint_like.low_dtype(Tensor self, SymInt low, SymInt high, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None, MemoryFormat? memory_format=None) -> Tensor +inline at::Tensor randint_like_symint(const at::Tensor & self, c10::SymInt low, c10::SymInt high, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory, c10::optional memory_format) { + return at::_ops::randint_like_low_dtype::call(self, low, high, dtype, layout, device, pin_memory, memory_format); +} +namespace symint { + template ::value>> + at::Tensor randint_like(const at::Tensor & self, c10::SymInt low, c10::SymInt high, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory, c10::optional memory_format) { + return at::_ops::randint_like_low_dtype::call(self, low, high, dtype, layout, device, pin_memory, memory_format); + } +} + +// aten::randint_like.out(Tensor self, SymInt high, *, MemoryFormat? memory_format=None, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & randint_like_out(at::Tensor & out, const at::Tensor & self, int64_t high, c10::optional memory_format=c10::nullopt) { + return at::_ops::randint_like_out::call(self, high, memory_format, out); +} +namespace symint { + template ::value>> + at::Tensor & randint_like_out(at::Tensor & out, const at::Tensor & self, int64_t high, c10::optional memory_format=c10::nullopt) { + return at::_ops::randint_like_out::call(self, high, memory_format, out); + } +} + +// aten::randint_like.out(Tensor self, SymInt high, *, MemoryFormat? memory_format=None, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & randint_like_outf(const at::Tensor & self, int64_t high, c10::optional memory_format, at::Tensor & out) { + return at::_ops::randint_like_out::call(self, high, memory_format, out); +} +namespace symint { + template ::value>> + at::Tensor & randint_like_outf(const at::Tensor & self, int64_t high, c10::optional memory_format, at::Tensor & out) { + return at::_ops::randint_like_out::call(self, high, memory_format, out); + } +} + +// aten::randint_like.out(Tensor self, SymInt high, *, MemoryFormat? memory_format=None, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & randint_like_symint_out(at::Tensor & out, const at::Tensor & self, c10::SymInt high, c10::optional memory_format=c10::nullopt) { + return at::_ops::randint_like_out::call(self, high, memory_format, out); +} +namespace symint { + template ::value>> + at::Tensor & randint_like_out(at::Tensor & out, const at::Tensor & self, c10::SymInt high, c10::optional memory_format=c10::nullopt) { + return at::_ops::randint_like_out::call(self, high, memory_format, out); + } +} + +// aten::randint_like.out(Tensor self, SymInt high, *, MemoryFormat? memory_format=None, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & randint_like_symint_outf(const at::Tensor & self, c10::SymInt high, c10::optional memory_format, at::Tensor & out) { + return at::_ops::randint_like_out::call(self, high, memory_format, out); +} +namespace symint { + template ::value>> + at::Tensor & randint_like_outf(const at::Tensor & self, c10::SymInt high, c10::optional memory_format, at::Tensor & out) { + return at::_ops::randint_like_out::call(self, high, memory_format, out); + } +} + +// aten::randint_like.low_dtype_out(Tensor self, SymInt low, SymInt high, *, MemoryFormat? memory_format=None, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & randint_like_out(at::Tensor & out, const at::Tensor & self, int64_t low, int64_t high, c10::optional memory_format=c10::nullopt) { + return at::_ops::randint_like_low_dtype_out::call(self, low, high, memory_format, out); +} +namespace symint { + template ::value>> + at::Tensor & randint_like_out(at::Tensor & out, const at::Tensor & self, int64_t low, int64_t high, c10::optional memory_format=c10::nullopt) { + return at::_ops::randint_like_low_dtype_out::call(self, low, high, memory_format, out); + } +} + +// aten::randint_like.low_dtype_out(Tensor self, SymInt low, SymInt high, *, MemoryFormat? memory_format=None, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & randint_like_outf(const at::Tensor & self, int64_t low, int64_t high, c10::optional memory_format, at::Tensor & out) { + return at::_ops::randint_like_low_dtype_out::call(self, low, high, memory_format, out); +} +namespace symint { + template ::value>> + at::Tensor & randint_like_outf(const at::Tensor & self, int64_t low, int64_t high, c10::optional memory_format, at::Tensor & out) { + return at::_ops::randint_like_low_dtype_out::call(self, low, high, memory_format, out); + } +} + +// aten::randint_like.low_dtype_out(Tensor self, SymInt low, SymInt high, *, MemoryFormat? memory_format=None, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & randint_like_symint_out(at::Tensor & out, const at::Tensor & self, c10::SymInt low, c10::SymInt high, c10::optional memory_format=c10::nullopt) { + return at::_ops::randint_like_low_dtype_out::call(self, low, high, memory_format, out); +} +namespace symint { + template ::value>> + at::Tensor & randint_like_out(at::Tensor & out, const at::Tensor & self, c10::SymInt low, c10::SymInt high, c10::optional memory_format=c10::nullopt) { + return at::_ops::randint_like_low_dtype_out::call(self, low, high, memory_format, out); + } +} + +// aten::randint_like.low_dtype_out(Tensor self, SymInt low, SymInt high, *, MemoryFormat? memory_format=None, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & randint_like_symint_outf(const at::Tensor & self, c10::SymInt low, c10::SymInt high, c10::optional memory_format, at::Tensor & out) { + return at::_ops::randint_like_low_dtype_out::call(self, low, high, memory_format, out); +} +namespace symint { + template ::value>> + at::Tensor & randint_like_outf(const at::Tensor & self, c10::SymInt low, c10::SymInt high, c10::optional memory_format, at::Tensor & out) { + return at::_ops::randint_like_low_dtype_out::call(self, low, high, memory_format, out); + } +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/randn_like_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/randn_like_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..79f48d087e742a5b74064bf507689da86f482dd5 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/randn_like_compositeexplicitautograd_dispatch.h @@ -0,0 +1,26 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeexplicitautograd { + +TORCH_API at::Tensor randn_like(const at::Tensor & self, at::TensorOptions options={}, c10::optional memory_format=c10::nullopt); +TORCH_API at::Tensor randn_like(const at::Tensor & self, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory, c10::optional memory_format); +TORCH_API at::Tensor & randn_like_out(at::Tensor & out, const at::Tensor & self, c10::optional memory_format=c10::nullopt); +TORCH_API at::Tensor & randn_like_outf(const at::Tensor & self, c10::optional memory_format, at::Tensor & out); + +} // namespace compositeexplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/replication_pad2d_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/replication_pad2d_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..0c1f60f5abe016b699355e08f4c1f142c660f156 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/replication_pad2d_ops.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { +namespace _ops { + + +struct TORCH_API replication_pad2d_out { + using schema = at::Tensor & (const at::Tensor &, c10::SymIntArrayRef, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::replication_pad2d") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "replication_pad2d.out(Tensor self, SymInt[4] padding, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, c10::SymIntArrayRef padding, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, c10::SymIntArrayRef padding, at::Tensor & out); +}; + +struct TORCH_API replication_pad2d { + using schema = at::Tensor (const at::Tensor &, c10::SymIntArrayRef); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::replication_pad2d") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "replication_pad2d(Tensor self, SymInt[4] padding) -> Tensor") + static at::Tensor call(const at::Tensor & self, c10::SymIntArrayRef padding); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, c10::SymIntArrayRef padding); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/rnn_tanh_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/rnn_tanh_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..e8f5b9d34d31b2bdf33b1a94c4c104df6107d0ce --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/rnn_tanh_compositeimplicitautograd_dispatch.h @@ -0,0 +1,24 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeimplicitautograd { + +TORCH_API ::std::tuple rnn_tanh(const at::Tensor & input, const at::Tensor & hx, at::TensorList params, bool has_biases, int64_t num_layers, double dropout, bool train, bool bidirectional, bool batch_first); +TORCH_API ::std::tuple rnn_tanh(const at::Tensor & data, const at::Tensor & batch_sizes, const at::Tensor & hx, at::TensorList params, bool has_biases, int64_t num_layers, double dropout, bool train, bool bidirectional); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/scatter_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/scatter_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..79f90e3b3af556eeb406f4e019d51c55f867f665 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/scatter_ops.h @@ -0,0 +1,171 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { +namespace _ops { + + +struct TORCH_API scatter_src { + using schema = at::Tensor (const at::Tensor &, int64_t, const at::Tensor &, const at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::scatter") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "src") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "scatter.src(Tensor self, int dim, Tensor index, Tensor src) -> Tensor") + static at::Tensor call(const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & src); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & src); +}; + +struct TORCH_API scatter__src { + using schema = at::Tensor & (at::Tensor &, int64_t, const at::Tensor &, const at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::scatter_") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "src") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "scatter_.src(Tensor(a!) self, int dim, Tensor index, Tensor src) -> Tensor(a!)") + static at::Tensor & call(at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & src); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & src); +}; + +struct TORCH_API scatter_src_out { + using schema = at::Tensor & (const at::Tensor &, int64_t, const at::Tensor &, const at::Tensor &, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::scatter") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "src_out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "scatter.src_out(Tensor self, int dim, Tensor index, Tensor src, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & src, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & src, at::Tensor & out); +}; + +struct TORCH_API scatter_value { + using schema = at::Tensor (const at::Tensor &, int64_t, const at::Tensor &, const at::Scalar &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::scatter") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "value") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "scatter.value(Tensor self, int dim, Tensor index, Scalar value) -> Tensor") + static at::Tensor call(const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Scalar & value); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Scalar & value); +}; + +struct TORCH_API scatter__value { + using schema = at::Tensor & (at::Tensor &, int64_t, const at::Tensor &, const at::Scalar &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::scatter_") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "value") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "scatter_.value(Tensor(a!) self, int dim, Tensor index, Scalar value) -> Tensor(a!)") + static at::Tensor & call(at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Scalar & value); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Scalar & value); +}; + +struct TORCH_API scatter_value_out { + using schema = at::Tensor & (const at::Tensor &, int64_t, const at::Tensor &, const at::Scalar &, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::scatter") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "value_out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "scatter.value_out(Tensor self, int dim, Tensor index, Scalar value, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Scalar & value, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Scalar & value, at::Tensor & out); +}; + +struct TORCH_API scatter_reduce { + using schema = at::Tensor (const at::Tensor &, int64_t, const at::Tensor &, const at::Tensor &, c10::string_view); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::scatter") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "reduce") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "scatter.reduce(Tensor self, int dim, Tensor index, Tensor src, *, str reduce) -> Tensor") + static at::Tensor call(const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & src, c10::string_view reduce); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & src, c10::string_view reduce); +}; + +struct TORCH_API scatter__reduce { + using schema = at::Tensor & (at::Tensor &, int64_t, const at::Tensor &, const at::Tensor &, c10::string_view); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::scatter_") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "reduce") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "scatter_.reduce(Tensor(a!) self, int dim, Tensor index, Tensor src, *, str reduce) -> Tensor(a!)") + static at::Tensor & call(at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & src, c10::string_view reduce); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & src, c10::string_view reduce); +}; + +struct TORCH_API scatter_reduce_out { + using schema = at::Tensor & (const at::Tensor &, int64_t, const at::Tensor &, const at::Tensor &, c10::string_view, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::scatter") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "reduce_out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "scatter.reduce_out(Tensor self, int dim, Tensor index, Tensor src, *, str reduce, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & src, c10::string_view reduce, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & src, c10::string_view reduce, at::Tensor & out); +}; + +struct TORCH_API scatter_value_reduce { + using schema = at::Tensor (const at::Tensor &, int64_t, const at::Tensor &, const at::Scalar &, c10::string_view); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::scatter") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "value_reduce") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "scatter.value_reduce(Tensor self, int dim, Tensor index, Scalar value, *, str reduce) -> Tensor") + static at::Tensor call(const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Scalar & value, c10::string_view reduce); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Scalar & value, c10::string_view reduce); +}; + +struct TORCH_API scatter__value_reduce { + using schema = at::Tensor & (at::Tensor &, int64_t, const at::Tensor &, const at::Scalar &, c10::string_view); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::scatter_") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "value_reduce") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "scatter_.value_reduce(Tensor(a!) self, int dim, Tensor index, Scalar value, *, str reduce) -> Tensor(a!)") + static at::Tensor & call(at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Scalar & value, c10::string_view reduce); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Scalar & value, c10::string_view reduce); +}; + +struct TORCH_API scatter_value_reduce_out { + using schema = at::Tensor & (const at::Tensor &, int64_t, const at::Tensor &, const at::Scalar &, c10::string_view, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::scatter") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "value_reduce_out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "scatter.value_reduce_out(Tensor self, int dim, Tensor index, Scalar value, *, str reduce, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Scalar & value, c10::string_view reduce, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Scalar & value, c10::string_view reduce, at::Tensor & out); +}; + +struct TORCH_API scatter_dimname_src { + using schema = at::Tensor (const at::Tensor &, at::Dimname, const at::Tensor &, const at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::scatter") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "dimname_src") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "scatter.dimname_src(Tensor self, Dimname dim, Tensor index, Tensor src) -> Tensor") + static at::Tensor call(const at::Tensor & self, at::Dimname dim, const at::Tensor & index, const at::Tensor & src); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, at::Dimname dim, const at::Tensor & index, const at::Tensor & src); +}; + +struct TORCH_API scatter_dimname_value { + using schema = at::Tensor (const at::Tensor &, at::Dimname, const at::Tensor &, const at::Scalar &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::scatter") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "dimname_value") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "scatter.dimname_value(Tensor self, Dimname dim, Tensor index, Scalar value) -> Tensor") + static at::Tensor call(const at::Tensor & self, at::Dimname dim, const at::Tensor & index, const at::Scalar & value); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, at::Dimname dim, const at::Tensor & index, const at::Scalar & value); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/sigmoid_backward_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/sigmoid_backward_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..533eaada66d07272a635356789f1b4fbdb980ffb --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/sigmoid_backward_cpu_dispatch.h @@ -0,0 +1,25 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cpu { + +TORCH_API at::Tensor sigmoid_backward(const at::Tensor & grad_output, const at::Tensor & output); +TORCH_API at::Tensor & sigmoid_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, const at::Tensor & output); +TORCH_API at::Tensor & sigmoid_backward_outf(const at::Tensor & grad_output, const at::Tensor & output, at::Tensor & grad_input); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/softshrink_backward_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/softshrink_backward_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..72baaa7c7470f189b3dbf0fc731771d31991da1e --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/softshrink_backward_ops.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { +namespace _ops { + + +struct TORCH_API softshrink_backward_grad_input { + using schema = at::Tensor & (const at::Tensor &, const at::Tensor &, const at::Scalar &, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::softshrink_backward") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "grad_input") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "softshrink_backward.grad_input(Tensor grad_output, Tensor self, Scalar lambd, *, Tensor(a!) grad_input) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & grad_output, const at::Tensor & self, const at::Scalar & lambd, at::Tensor & grad_input); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad_output, const at::Tensor & self, const at::Scalar & lambd, at::Tensor & grad_input); +}; + +struct TORCH_API softshrink_backward { + using schema = at::Tensor (const at::Tensor &, const at::Tensor &, const at::Scalar &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::softshrink_backward") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "softshrink_backward(Tensor grad_output, Tensor self, Scalar lambd) -> Tensor") + static at::Tensor call(const at::Tensor & grad_output, const at::Tensor & self, const at::Scalar & lambd); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad_output, const at::Tensor & self, const at::Scalar & lambd); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/sparse_resize_and_clear.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/sparse_resize_and_clear.h new file mode 100644 index 0000000000000000000000000000000000000000..8d4b5cc3fcb8ebc2a8dbec8755c3d1b4e09fbc8d --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/sparse_resize_and_clear.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::sparse_resize_and_clear.out(Tensor self, int[] size, int sparse_dim, int dense_dim, *, Tensor(a!) out) -> Tensor(a!) +inline const at::Tensor & sparse_resize_and_clear_out(const at::Tensor & out, const at::Tensor & self, at::IntArrayRef size, int64_t sparse_dim, int64_t dense_dim) { + return at::_ops::sparse_resize_and_clear_out::call(self, size, sparse_dim, dense_dim, out); +} +// aten::sparse_resize_and_clear.out(Tensor self, int[] size, int sparse_dim, int dense_dim, *, Tensor(a!) out) -> Tensor(a!) +inline const at::Tensor & sparse_resize_and_clear_outf(const at::Tensor & self, at::IntArrayRef size, int64_t sparse_dim, int64_t dense_dim, const at::Tensor & out) { + return at::_ops::sparse_resize_and_clear_out::call(self, size, sparse_dim, dense_dim, out); +} + +// aten::sparse_resize_and_clear(Tensor self, int[] size, int sparse_dim, int dense_dim) -> Tensor +inline at::Tensor sparse_resize_and_clear(const at::Tensor & self, at::IntArrayRef size, int64_t sparse_dim, int64_t dense_dim) { + return at::_ops::sparse_resize_and_clear::call(self, size, sparse_dim, dense_dim); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/sparse_sampled_addmm_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/sparse_sampled_addmm_native.h new file mode 100644 index 0000000000000000000000000000000000000000..7018e305788353f31b5c55ed3dae79544f4d8f21 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/sparse_sampled_addmm_native.h @@ -0,0 +1,24 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor sparse_sampled_addmm_sparse_csr_cpu(const at::Tensor & self, const at::Tensor & mat1, const at::Tensor & mat2, const at::Scalar & beta=1, const at::Scalar & alpha=1); +TORCH_API at::Tensor & sparse_sampled_addmm_out_sparse_csr_cpu(const at::Tensor & self, const at::Tensor & mat1, const at::Tensor & mat2, const at::Scalar & beta, const at::Scalar & alpha, at::Tensor & out); +TORCH_API at::Tensor sparse_sampled_addmm_sparse_csr_cuda(const at::Tensor & self, const at::Tensor & mat1, const at::Tensor & mat2, const at::Scalar & beta=1, const at::Scalar & alpha=1); +TORCH_API at::Tensor & sparse_sampled_addmm_out_sparse_csr_cuda(const at::Tensor & self, const at::Tensor & mat1, const at::Tensor & mat2, const at::Scalar & beta, const at::Scalar & alpha, at::Tensor & out); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_multigammaln_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_multigammaln_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..95a3e938b0e40a8af67bb616661168578eca12a3 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_multigammaln_compositeimplicitautograd_dispatch.h @@ -0,0 +1,25 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeimplicitautograd { + +TORCH_API at::Tensor special_multigammaln(const at::Tensor & self, int64_t p); +TORCH_API at::Tensor & special_multigammaln_out(at::Tensor & out, const at::Tensor & self, int64_t p); +TORCH_API at::Tensor & special_multigammaln_outf(const at::Tensor & self, int64_t p, at::Tensor & out); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_round_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_round_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..4a4e46e0ef203916d9aaa357d3cd7f6b7bc560f2 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_round_ops.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { +namespace _ops { + + +struct TORCH_API special_round { + using schema = at::Tensor (const at::Tensor &, int64_t); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::special_round") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "special_round(Tensor self, *, int decimals=0) -> Tensor") + static at::Tensor call(const at::Tensor & self, int64_t decimals); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, int64_t decimals); +}; + +struct TORCH_API special_round_out { + using schema = at::Tensor & (const at::Tensor &, int64_t, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::special_round") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "special_round.out(Tensor self, *, int decimals=0, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, int64_t decimals, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, int64_t decimals, at::Tensor & out); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_xlogy_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_xlogy_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..eaa200dfd3c36729ee1a3d9e58b4009e88c081bc --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_xlogy_ops.h @@ -0,0 +1,83 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { +namespace _ops { + + +struct TORCH_API special_xlogy { + using schema = at::Tensor (const at::Tensor &, const at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::special_xlogy") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "special_xlogy(Tensor self, Tensor other) -> Tensor") + static at::Tensor call(const at::Tensor & self, const at::Tensor & other); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & other); +}; + +struct TORCH_API special_xlogy_self_scalar { + using schema = at::Tensor (const at::Scalar &, const at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::special_xlogy") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "self_scalar") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "special_xlogy.self_scalar(Scalar self, Tensor other) -> Tensor") + static at::Tensor call(const at::Scalar & self, const at::Tensor & other); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Scalar & self, const at::Tensor & other); +}; + +struct TORCH_API special_xlogy_other_scalar { + using schema = at::Tensor (const at::Tensor &, const at::Scalar &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::special_xlogy") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "other_scalar") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "special_xlogy.other_scalar(Tensor self, Scalar other) -> Tensor") + static at::Tensor call(const at::Tensor & self, const at::Scalar & other); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Scalar & other); +}; + +struct TORCH_API special_xlogy_out { + using schema = at::Tensor & (const at::Tensor &, const at::Tensor &, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::special_xlogy") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "special_xlogy.out(Tensor self, Tensor other, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, const at::Tensor & other, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & other, at::Tensor & out); +}; + +struct TORCH_API special_xlogy_self_scalar_out { + using schema = at::Tensor & (const at::Scalar &, const at::Tensor &, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::special_xlogy") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "self_scalar_out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "special_xlogy.self_scalar_out(Scalar self, Tensor other, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Scalar & self, const at::Tensor & other, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Scalar & self, const at::Tensor & other, at::Tensor & out); +}; + +struct TORCH_API special_xlogy_other_scalar_out { + using schema = at::Tensor & (const at::Tensor &, const at::Scalar &, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::special_xlogy") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "other_scalar_out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "special_xlogy.other_scalar_out(Tensor self, Scalar other, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, const at::Scalar & other, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Scalar & other, at::Tensor & out); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_zeta_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_zeta_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..5a62b97cd920bcad1149a2decf3f84a3970485ff --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_zeta_ops.h @@ -0,0 +1,83 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { +namespace _ops { + + +struct TORCH_API special_zeta { + using schema = at::Tensor (const at::Tensor &, const at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::special_zeta") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "special_zeta(Tensor self, Tensor other) -> Tensor") + static at::Tensor call(const at::Tensor & self, const at::Tensor & other); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & other); +}; + +struct TORCH_API special_zeta_self_scalar { + using schema = at::Tensor (const at::Scalar &, const at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::special_zeta") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "self_scalar") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "special_zeta.self_scalar(Scalar self, Tensor other) -> Tensor") + static at::Tensor call(const at::Scalar & self, const at::Tensor & other); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Scalar & self, const at::Tensor & other); +}; + +struct TORCH_API special_zeta_other_scalar { + using schema = at::Tensor (const at::Tensor &, const at::Scalar &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::special_zeta") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "other_scalar") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "special_zeta.other_scalar(Tensor self, Scalar other) -> Tensor") + static at::Tensor call(const at::Tensor & self, const at::Scalar & other); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Scalar & other); +}; + +struct TORCH_API special_zeta_out { + using schema = at::Tensor & (const at::Tensor &, const at::Tensor &, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::special_zeta") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "special_zeta.out(Tensor self, Tensor other, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, const at::Tensor & other, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & other, at::Tensor & out); +}; + +struct TORCH_API special_zeta_self_scalar_out { + using schema = at::Tensor & (const at::Scalar &, const at::Tensor &, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::special_zeta") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "self_scalar_out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "special_zeta.self_scalar_out(Scalar self, Tensor other, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Scalar & self, const at::Tensor & other, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Scalar & self, const at::Tensor & other, at::Tensor & out); +}; + +struct TORCH_API special_zeta_other_scalar_out { + using schema = at::Tensor & (const at::Tensor &, const at::Scalar &, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::special_zeta") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "other_scalar_out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "special_zeta.other_scalar_out(Tensor self, Scalar other, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, const at::Scalar & other, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Scalar & other, at::Tensor & out); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/split_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/split_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..bf0450777a795e56925f4d0a3df3c21072055353 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/split_ops.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { +namespace _ops { + + +struct TORCH_API split_Tensor { + using schema = ::std::vector (const at::Tensor &, c10::SymInt, int64_t); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::split") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Tensor") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "split.Tensor(Tensor(a -> *) self, SymInt split_size, int dim=0) -> Tensor(a)[]") + static ::std::vector call(const at::Tensor & self, c10::SymInt split_size, int64_t dim); + static ::std::vector redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, c10::SymInt split_size, int64_t dim); +}; + +struct TORCH_API split_sizes { + using schema = ::std::vector (const at::Tensor &, c10::SymIntArrayRef, int64_t); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::split") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "sizes") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "split.sizes(Tensor(a -> *) self, SymInt[] split_size, int dim=0) -> Tensor(a)[]") + static ::std::vector call(const at::Tensor & self, c10::SymIntArrayRef split_size, int64_t dim); + static ::std::vector redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, c10::SymIntArrayRef split_size, int64_t dim); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/sum_to_size_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/sum_to_size_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..1bea5d36509a11964e2bd283f4dc3607f63161a2 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/sum_to_size_ops.h @@ -0,0 +1,28 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { +namespace _ops { + + +struct TORCH_API sum_to_size { + using schema = at::Tensor (const at::Tensor &, c10::SymIntArrayRef); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::sum_to_size") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "sum_to_size(Tensor self, SymInt[] size) -> Tensor") + static at::Tensor call(const at::Tensor & self, c10::SymIntArrayRef size); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, c10::SymIntArrayRef size); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/sym_size_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/sym_size_native.h new file mode 100644 index 0000000000000000000000000000000000000000..a2f1a04db6211ec4b45aa7101ad3f5e306c565f5 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/sym_size_native.h @@ -0,0 +1,21 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API c10::SymInt sym_size(const at::Tensor & self, int64_t dim); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/sym_stride_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/sym_stride_native.h new file mode 100644 index 0000000000000000000000000000000000000000..b99f10a97326177f3bdcf0cee807d97491f1845f --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/sym_stride_native.h @@ -0,0 +1,21 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API c10::SymInt sym_stride(const at::Tensor & self, int64_t dim); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_linear1d_backward.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_linear1d_backward.h new file mode 100644 index 0000000000000000000000000000000000000000..97ed9255d742b6dd9e5d6c0486d8648411187dcd --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_linear1d_backward.h @@ -0,0 +1,91 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::upsample_linear1d_backward.grad_input(Tensor grad_output, SymInt[1] output_size, SymInt[3] input_size, bool align_corners, float? scales=None, *, Tensor(a!) grad_input) -> Tensor(a!) +inline at::Tensor & upsample_linear1d_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, bool align_corners, c10::optional scales=c10::nullopt) { + return at::_ops::upsample_linear1d_backward_grad_input::call(grad_output, c10::fromIntArrayRefSlow(output_size), c10::fromIntArrayRefSlow(input_size), align_corners, scales, grad_input); +} +namespace symint { + template ::value>> + at::Tensor & upsample_linear1d_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, bool align_corners, c10::optional scales=c10::nullopt) { + return at::_ops::upsample_linear1d_backward_grad_input::call(grad_output, c10::fromIntArrayRefSlow(output_size), c10::fromIntArrayRefSlow(input_size), align_corners, scales, grad_input); + } +} + +// aten::upsample_linear1d_backward.grad_input(Tensor grad_output, SymInt[1] output_size, SymInt[3] input_size, bool align_corners, float? scales=None, *, Tensor(a!) grad_input) -> Tensor(a!) +inline at::Tensor & upsample_linear1d_backward_outf(const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, bool align_corners, c10::optional scales, at::Tensor & grad_input) { + return at::_ops::upsample_linear1d_backward_grad_input::call(grad_output, c10::fromIntArrayRefSlow(output_size), c10::fromIntArrayRefSlow(input_size), align_corners, scales, grad_input); +} +namespace symint { + template ::value>> + at::Tensor & upsample_linear1d_backward_outf(const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, bool align_corners, c10::optional scales, at::Tensor & grad_input) { + return at::_ops::upsample_linear1d_backward_grad_input::call(grad_output, c10::fromIntArrayRefSlow(output_size), c10::fromIntArrayRefSlow(input_size), align_corners, scales, grad_input); + } +} + +// aten::upsample_linear1d_backward.grad_input(Tensor grad_output, SymInt[1] output_size, SymInt[3] input_size, bool align_corners, float? scales=None, *, Tensor(a!) grad_input) -> Tensor(a!) +inline at::Tensor & upsample_linear1d_backward_symint_out(at::Tensor & grad_input, const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, bool align_corners, c10::optional scales=c10::nullopt) { + return at::_ops::upsample_linear1d_backward_grad_input::call(grad_output, output_size, input_size, align_corners, scales, grad_input); +} +namespace symint { + template ::value>> + at::Tensor & upsample_linear1d_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, bool align_corners, c10::optional scales=c10::nullopt) { + return at::_ops::upsample_linear1d_backward_grad_input::call(grad_output, output_size, input_size, align_corners, scales, grad_input); + } +} + +// aten::upsample_linear1d_backward.grad_input(Tensor grad_output, SymInt[1] output_size, SymInt[3] input_size, bool align_corners, float? scales=None, *, Tensor(a!) grad_input) -> Tensor(a!) +inline at::Tensor & upsample_linear1d_backward_symint_outf(const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, bool align_corners, c10::optional scales, at::Tensor & grad_input) { + return at::_ops::upsample_linear1d_backward_grad_input::call(grad_output, output_size, input_size, align_corners, scales, grad_input); +} +namespace symint { + template ::value>> + at::Tensor & upsample_linear1d_backward_outf(const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, bool align_corners, c10::optional scales, at::Tensor & grad_input) { + return at::_ops::upsample_linear1d_backward_grad_input::call(grad_output, output_size, input_size, align_corners, scales, grad_input); + } +} + +// aten::upsample_linear1d_backward(Tensor grad_output, SymInt[1] output_size, SymInt[3] input_size, bool align_corners, float? scales=None) -> Tensor +inline at::Tensor upsample_linear1d_backward(const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, bool align_corners, c10::optional scales=c10::nullopt) { + return at::_ops::upsample_linear1d_backward::call(grad_output, c10::fromIntArrayRefSlow(output_size), c10::fromIntArrayRefSlow(input_size), align_corners, scales); +} +namespace symint { + template ::value>> + at::Tensor upsample_linear1d_backward(const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, bool align_corners, c10::optional scales=c10::nullopt) { + return at::_ops::upsample_linear1d_backward::call(grad_output, c10::fromIntArrayRefSlow(output_size), c10::fromIntArrayRefSlow(input_size), align_corners, scales); + } +} + +// aten::upsample_linear1d_backward(Tensor grad_output, SymInt[1] output_size, SymInt[3] input_size, bool align_corners, float? scales=None) -> Tensor +inline at::Tensor upsample_linear1d_backward_symint(const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, bool align_corners, c10::optional scales=c10::nullopt) { + return at::_ops::upsample_linear1d_backward::call(grad_output, output_size, input_size, align_corners, scales); +} +namespace symint { + template ::value>> + at::Tensor upsample_linear1d_backward(const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, bool align_corners, c10::optional scales=c10::nullopt) { + return at::_ops::upsample_linear1d_backward::call(grad_output, output_size, input_size, align_corners, scales); + } +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_nearest3d_backward_meta_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_nearest3d_backward_meta_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..2e260aaf393dbde5eb16acebd676da1255cc9ca1 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_nearest3d_backward_meta_dispatch.h @@ -0,0 +1,28 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace meta { + +TORCH_API at::Tensor upsample_nearest3d_backward(const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, c10::optional scales_d=c10::nullopt, c10::optional scales_h=c10::nullopt, c10::optional scales_w=c10::nullopt); +TORCH_API at::Tensor upsample_nearest3d_backward_symint(const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, c10::optional scales_d=c10::nullopt, c10::optional scales_h=c10::nullopt, c10::optional scales_w=c10::nullopt); +TORCH_API at::Tensor & upsample_nearest3d_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, c10::optional scales_d=c10::nullopt, c10::optional scales_h=c10::nullopt, c10::optional scales_w=c10::nullopt); +TORCH_API at::Tensor & upsample_nearest3d_backward_outf(const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, c10::optional scales_d, c10::optional scales_h, c10::optional scales_w, at::Tensor & grad_input); +TORCH_API at::Tensor & upsample_nearest3d_backward_symint_out(at::Tensor & grad_input, const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, c10::optional scales_d=c10::nullopt, c10::optional scales_h=c10::nullopt, c10::optional scales_w=c10::nullopt); +TORCH_API at::Tensor & upsample_nearest3d_backward_symint_outf(const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, c10::optional scales_d, c10::optional scales_h, c10::optional scales_w, at::Tensor & grad_input); + +} // namespace meta +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/var_mean_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/var_mean_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..0e10693ae81c5da1664544dec59b2b307c498745 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/var_mean_cuda_dispatch.h @@ -0,0 +1,23 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cuda { + +TORCH_API ::std::tuple var_mean(const at::Tensor & self, at::OptionalIntArrayRef dim=c10::nullopt, const c10::optional & correction=c10::nullopt, bool keepdim=false); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/xlogy_meta.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/xlogy_meta.h new file mode 100644 index 0000000000000000000000000000000000000000..1634701593160305d66923be4374989a798bc003 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/xlogy_meta.h @@ -0,0 +1,27 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeMetaFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +namespace at { +namespace meta { + +struct TORCH_API structured_xlogy_Tensor : public TensorIteratorBase { + + + void meta(const at::Tensor & self, const at::Tensor & other); +}; + +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/zeros.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/zeros.h new file mode 100644 index 0000000000000000000000000000000000000000..b3b849648328c2a7e7e128738056ade58db75a73 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/zeros.h @@ -0,0 +1,131 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::zeros.names(int[] size, *, Dimname[]? names, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor +inline at::Tensor zeros(at::IntArrayRef size, c10::optional names, at::TensorOptions options={}) { + return at::_ops::zeros_names::call(size, names, c10::optTypeMetaToScalarType(options.dtype_opt()), options.layout_opt(), options.device_opt(), options.pinned_memory_opt()); +} +// aten::zeros.names(int[] size, *, Dimname[]? names, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor +inline at::Tensor zeros(at::IntArrayRef size, c10::optional names, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory) { + return at::_ops::zeros_names::call(size, names, dtype, layout, device, pin_memory); +} + +// aten::zeros(SymInt[] size, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor +inline at::Tensor zeros(at::IntArrayRef size, at::TensorOptions options={}) { + return at::_ops::zeros::call(c10::fromIntArrayRefSlow(size), c10::optTypeMetaToScalarType(options.dtype_opt()), options.layout_opt(), options.device_opt(), options.pinned_memory_opt()); +} +namespace symint { + template ::value>> + at::Tensor zeros(at::IntArrayRef size, at::TensorOptions options={}) { + return at::_ops::zeros::call(c10::fromIntArrayRefSlow(size), c10::optTypeMetaToScalarType(options.dtype_opt()), options.layout_opt(), options.device_opt(), options.pinned_memory_opt()); + } +} + +// aten::zeros(SymInt[] size, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor +inline at::Tensor zeros(at::IntArrayRef size, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory) { + return at::_ops::zeros::call(c10::fromIntArrayRefSlow(size), dtype, layout, device, pin_memory); +} +namespace symint { + template ::value>> + at::Tensor zeros(at::IntArrayRef size, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory) { + return at::_ops::zeros::call(c10::fromIntArrayRefSlow(size), dtype, layout, device, pin_memory); + } +} + +// aten::zeros(SymInt[] size, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor +inline at::Tensor zeros_symint(c10::SymIntArrayRef size, at::TensorOptions options={}) { + return at::_ops::zeros::call(size, c10::optTypeMetaToScalarType(options.dtype_opt()), options.layout_opt(), options.device_opt(), options.pinned_memory_opt()); +} +namespace symint { + template ::value>> + at::Tensor zeros(c10::SymIntArrayRef size, at::TensorOptions options={}) { + return at::_ops::zeros::call(size, c10::optTypeMetaToScalarType(options.dtype_opt()), options.layout_opt(), options.device_opt(), options.pinned_memory_opt()); + } +} + +// aten::zeros(SymInt[] size, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor +inline at::Tensor zeros_symint(c10::SymIntArrayRef size, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory) { + return at::_ops::zeros::call(size, dtype, layout, device, pin_memory); +} +namespace symint { + template ::value>> + at::Tensor zeros(c10::SymIntArrayRef size, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory) { + return at::_ops::zeros::call(size, dtype, layout, device, pin_memory); + } +} + +// aten::zeros.out(SymInt[] size, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & zeros_out(at::Tensor & out, at::IntArrayRef size) { + return at::_ops::zeros_out::call(c10::fromIntArrayRefSlow(size), out); +} +namespace symint { + template ::value>> + at::Tensor & zeros_out(at::Tensor & out, at::IntArrayRef size) { + return at::_ops::zeros_out::call(c10::fromIntArrayRefSlow(size), out); + } +} + +// aten::zeros.out(SymInt[] size, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & zeros_outf(at::IntArrayRef size, at::Tensor & out) { + return at::_ops::zeros_out::call(c10::fromIntArrayRefSlow(size), out); +} +namespace symint { + template ::value>> + at::Tensor & zeros_outf(at::IntArrayRef size, at::Tensor & out) { + return at::_ops::zeros_out::call(c10::fromIntArrayRefSlow(size), out); + } +} + +// aten::zeros.out(SymInt[] size, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & zeros_symint_out(at::Tensor & out, c10::SymIntArrayRef size) { + return at::_ops::zeros_out::call(size, out); +} +namespace symint { + template ::value>> + at::Tensor & zeros_out(at::Tensor & out, c10::SymIntArrayRef size) { + return at::_ops::zeros_out::call(size, out); + } +} + +// aten::zeros.out(SymInt[] size, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & zeros_symint_outf(c10::SymIntArrayRef size, at::Tensor & out) { + return at::_ops::zeros_out::call(size, out); +} +namespace symint { + template ::value>> + at::Tensor & zeros_outf(c10::SymIntArrayRef size, at::Tensor & out) { + return at::_ops::zeros_out::call(size, out); + } +} + +// aten::zeros.names_out(int[] size, *, Dimname[]? names, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & zeros_out(at::Tensor & out, at::IntArrayRef size, c10::optional names) { + return at::_ops::zeros_names_out::call(size, names, out); +} +// aten::zeros.names_out(int[] size, *, Dimname[]? names, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & zeros_outf(at::IntArrayRef size, c10::optional names, at::Tensor & out) { + return at::_ops::zeros_names_out::call(size, names, out); +} + +}