url
stringlengths
62
66
repository_url
stringclasses
1 value
labels_url
stringlengths
76
80
comments_url
stringlengths
71
75
events_url
stringlengths
69
73
html_url
stringlengths
50
56
id
int64
377M
2.15B
node_id
stringlengths
18
32
number
int64
1
29.2k
title
stringlengths
1
487
user
dict
labels
list
state
stringclasses
2 values
locked
bool
2 classes
assignee
dict
assignees
list
comments
list
created_at
int64
1.54k
1.71k
updated_at
int64
1.54k
1.71k
closed_at
int64
1.54k
1.71k
βŒ€
author_association
stringclasses
4 values
active_lock_reason
stringclasses
2 values
body
stringlengths
0
234k
βŒ€
reactions
dict
timeline_url
stringlengths
71
75
state_reason
stringclasses
3 values
draft
bool
2 classes
pull_request
dict
https://api.github.com/repos/huggingface/transformers/issues/27645
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27645/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27645/comments
https://api.github.com/repos/huggingface/transformers/issues/27645/events
https://github.com/huggingface/transformers/pull/27645
2,005,811,418
PR_kwDOCUB6oc5gGj3r
27,645
[`DocString`] Support a revision in the docstring `add_code_sample_docstrings` to facilitate integrations
{ "login": "ArthurZucker", "id": 48595927, "node_id": "MDQ6VXNlcjQ4NTk1OTI3", "avatar_url": "https://avatars.githubusercontent.com/u/48595927?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ArthurZucker", "html_url": "https://github.com/ArthurZucker", "followers_url": "https://api.github.c...
[]
closed
false
null
[]
[ "_The documentation is not available anymore as the PR was closed or merged._", "Thanks for the review but a PR in draft mode means it's a draft πŸ˜‰ ", "oh no :sweat: missed the draft status somehow", "No worries πŸ€— " ]
1,700
1,700
1,700
COLLABORATOR
null
# What does this PR do? When PRs on the hub are not merged yet + sometimes for safety reasons we can pin a revision in the docstring code samples added. Seems like the CI is enough to test, dummy testing the build with a flax t5 model πŸ˜‰ Example: https://moon-ci-docs.huggingface.co/docs/transformers/pr_27645/e...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27645/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27645/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27645", "html_url": "https://github.com/huggingface/transformers/pull/27645", "diff_url": "https://github.com/huggingface/transformers/pull/27645.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27645.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27644
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27644/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27644/comments
https://api.github.com/repos/huggingface/transformers/issues/27644/events
https://github.com/huggingface/transformers/issues/27644
2,005,627,044
I_kwDOCUB6oc53i3Ck
27,644
Crash when running `examples/flax/question-answering`
{ "login": "DwarKapex", "id": 11195921, "node_id": "MDQ6VXNlcjExMTk1OTIx", "avatar_url": "https://avatars.githubusercontent.com/u/11195921?v=4", "gravatar_id": "", "url": "https://api.github.com/users/DwarKapex", "html_url": "https://github.com/DwarKapex", "followers_url": "https://api.github.com/users/...
[]
closed
false
null
[]
[ "Meet same issue when JAX v0.4.20." ]
1,700
1,702
1,702
NONE
null
### System Info - `transformers` version: 4.36.0.dev0 - Platform: Linux-5.15.0-88-generic-x86_64-with-glibc2.35 - Python version: 3.10.12 - Huggingface_hub version: 0.19.4 - Safetensors version: 0.4.0 - Accelerate version: not installed - Accelerate config: not found - PyTorch version (GPU?): not installed (NA)...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27644/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27644/timeline
completed
null
null
https://api.github.com/repos/huggingface/transformers/issues/27643
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27643/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27643/comments
https://api.github.com/repos/huggingface/transformers/issues/27643/events
https://github.com/huggingface/transformers/pull/27643
2,005,474,441
PR_kwDOCUB6oc5gFa96
27,643
Simplify the implementation of jitter noise in moe models
{ "login": "jiangwangyi", "id": 39762734, "node_id": "MDQ6VXNlcjM5NzYyNzM0", "avatar_url": "https://avatars.githubusercontent.com/u/39762734?v=4", "gravatar_id": "", "url": "https://api.github.com/users/jiangwangyi", "html_url": "https://github.com/jiangwangyi", "followers_url": "https://api.github.com/...
[]
closed
false
null
[]
[]
1,700
1,700
1,700
CONTRIBUTOR
null
# What does this PR do? <!-- Congratulations! You've made it this far! You're not quite done yet though. Once merged, your PR is going to appear in the release notes with the title you set, so make sure it's a great title that fully reflects the extent of your awesome contribution. Then, please replace this w...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27643/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27643/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27643", "html_url": "https://github.com/huggingface/transformers/pull/27643", "diff_url": "https://github.com/huggingface/transformers/pull/27643.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27643.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27642
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27642/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27642/comments
https://api.github.com/repos/huggingface/transformers/issues/27642/events
https://github.com/huggingface/transformers/issues/27642
2,005,472,954
I_kwDOCUB6oc53iRa6
27,642
difference in tokenization - bert-base-multilingual-cased vs uncased
{ "login": "dsplog", "id": 5105387, "node_id": "MDQ6VXNlcjUxMDUzODc=", "avatar_url": "https://avatars.githubusercontent.com/u/5105387?v=4", "gravatar_id": "", "url": "https://api.github.com/users/dsplog", "html_url": "https://github.com/dsplog", "followers_url": "https://api.github.com/users/dsplog/foll...
[]
closed
false
null
[]
[ "Hey πŸ€— \r\n1. If you want to know the actual tokenization you should be using:\r\n```python\r\n>>> text = 'hello my dear'\r\n>>> \r\n>>> from transformers import BertTokenizer\r\n>>> bert_c_tok = BertTokenizer.from_pretrained('bert-base-multilingual-cased')\r\n>>> bert_c_tok.tokenize(text)\r\n['hell', '##o', 'my',...
1,700
1,700
1,700
NONE
null
### System Info - `transformers` version: 4.33.3 - Platform: Linux-5.15.0-88-generic-x86_64-with-glibc2.17 - Python version: 3.8.18 - Huggingface_hub version: 0.17.3 - Safetensors version: 0.4.0 - Accelerate version: 0.23.0 - Accelerate config: not found - PyTorch version (GPU?): 2.1.0+cu121 (True) - Tensorfl...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27642/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27642/timeline
completed
null
null
https://api.github.com/repos/huggingface/transformers/issues/27641
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27641/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27641/comments
https://api.github.com/repos/huggingface/transformers/issues/27641/events
https://github.com/huggingface/transformers/pull/27641
2,005,118,209
PR_kwDOCUB6oc5gEONj
27,641
[docs] Quantization
{ "login": "stevhliu", "id": 59462357, "node_id": "MDQ6VXNlcjU5NDYyMzU3", "avatar_url": "https://avatars.githubusercontent.com/u/59462357?v=4", "gravatar_id": "", "url": "https://api.github.com/users/stevhliu", "html_url": "https://github.com/stevhliu", "followers_url": "https://api.github.com/users/ste...
[]
closed
false
null
[]
[ "_The documentation is not available anymore as the PR was closed or merged._", "I just saw this today @stevhliu, but I think that we table of content is not displayed correctly. It would be great to fix this since the doc is quite big. Thanks ! Feel free to merge the PR after that ! \r\n\r\n<img width=\"1287\" a...
1,700
1,701
1,701
MEMBER
null
This PR resolves #27575 to clean up the Quantization API docs and create a new section for the existing content in the Performance and scalability section for more visibility.
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27641/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27641/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27641", "html_url": "https://github.com/huggingface/transformers/pull/27641", "diff_url": "https://github.com/huggingface/transformers/pull/27641.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27641.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27640
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27640/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27640/comments
https://api.github.com/repos/huggingface/transformers/issues/27640/events
https://github.com/huggingface/transformers/issues/27640
2,004,924,110
I_kwDOCUB6oc53gLbO
27,640
Allow passing 2D attention mask
{ "login": "UniverseFly", "id": 46997596, "node_id": "MDQ6VXNlcjQ2OTk3NTk2", "avatar_url": "https://avatars.githubusercontent.com/u/46997596?v=4", "gravatar_id": "", "url": "https://api.github.com/users/UniverseFly", "html_url": "https://github.com/UniverseFly", "followers_url": "https://api.github.com/...
[ { "id": 2648621985, "node_id": "MDU6TGFiZWwyNjQ4NjIxOTg1", "url": "https://api.github.com/repos/huggingface/transformers/labels/Feature%20request", "name": "Feature request", "color": "FBCA04", "default": false, "description": "Request for a new feature" } ]
open
false
null
[]
[ "Hey, the model's forward already supports passing a 2d attention mask, it is just expended to 4d because that is the format required by the attention implementation. \r\nWould you mind elaborating on what you cannot currently do? (Might be related to #27539?) ", "> Hey, the model's forward already supports passi...
1,700
1,701
null
NONE
null
### Feature request Allow passing a 2D attention mask in `model.forward`. ### Motivation With this feature, it would be much easier to avoid cross-context contamination during pretraining and supervised finetuning when packing the sequences together for more efficient training. Here is an example usecase discusse...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27640/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27640/timeline
null
null
null
https://api.github.com/repos/huggingface/transformers/issues/27639
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27639/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27639/comments
https://api.github.com/repos/huggingface/transformers/issues/27639/events
https://github.com/huggingface/transformers/pull/27639
2,004,901,114
PR_kwDOCUB6oc5gDeqc
27,639
Add `TimedTextStreamer` for measuring per-token latency using `generate()`
{ "login": "danielkorat", "id": 32893314, "node_id": "MDQ6VXNlcjMyODkzMzE0", "avatar_url": "https://avatars.githubusercontent.com/u/32893314?v=4", "gravatar_id": "", "url": "https://api.github.com/users/danielkorat", "html_url": "https://github.com/danielkorat", "followers_url": "https://api.github.com/...
[]
closed
false
null
[]
[ "Hi @danielkorat πŸ‘‹ Thank you for opening the PR!\r\n\r\nI haven't seen other requests to measure per-token time, so I'm assuming a limited number of users would benefit from it. Moreover, I have in my list of TODOs for `generate()` to refactor how streaming works: instead of printing/emitting tokens, `generate()` ...
1,700
1,704
1,704
CONTRIBUTOR
null
# What does this PR do? Adds a `TimedTextStreamer` for measuring per-token latency using `generate()` . The idea behind this implementation is a non-intrusive method to measure per-token generation latency, which does not modify code in `genration/utils.py` making it compatible with future versions of `utils.py` as...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27639/reactions", "total_count": 3, "+1": 0, "-1": 0, "laugh": 0, "hooray": 3, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27639/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27639", "html_url": "https://github.com/huggingface/transformers/pull/27639", "diff_url": "https://github.com/huggingface/transformers/pull/27639.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27639.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27638
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27638/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27638/comments
https://api.github.com/repos/huggingface/transformers/issues/27638/events
https://github.com/huggingface/transformers/pull/27638
2,004,795,258
PR_kwDOCUB6oc5gDHWH
27,638
translate internal folder files to chinese
{ "login": "jiaqiw09", "id": 60021713, "node_id": "MDQ6VXNlcjYwMDIxNzEz", "avatar_url": "https://avatars.githubusercontent.com/u/60021713?v=4", "gravatar_id": "", "url": "https://api.github.com/users/jiaqiw09", "html_url": "https://github.com/jiaqiw09", "followers_url": "https://api.github.com/users/jia...
[]
closed
false
null
[]
[ "@stevhliu\r\n\r\nHi, here is another pr. I will fix merge conflic later.\r\n\r\nAnd for file `trainer_utils.md`, I think the subtile \"Distributed Evaluation\" is repeated. Would you mind providing a useful subtitle? I can update both en file and zh file at this pr.", "@stevhliu \r\n\r\nHi, thansk for your comm...
1,700
1,701
1,701
CONTRIBUTOR
null
# What does this PR do? Part of #26803 ## Before submitting - [x] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case). - [x] Did you read the [contributor guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-reques...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27638/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27638/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27638", "html_url": "https://github.com/huggingface/transformers/pull/27638", "diff_url": "https://github.com/huggingface/transformers/pull/27638.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27638.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27637
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27637/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27637/comments
https://api.github.com/repos/huggingface/transformers/issues/27637/events
https://github.com/huggingface/transformers/pull/27637
2,004,763,823
PR_kwDOCUB6oc5gDAdF
27,637
Added test cases for rembert refering to albert and reformer test_tok…
{ "login": "nileshkokane01", "id": 8201108, "node_id": "MDQ6VXNlcjgyMDExMDg=", "avatar_url": "https://avatars.githubusercontent.com/u/8201108?v=4", "gravatar_id": "", "url": "https://api.github.com/users/nileshkokane01", "html_url": "https://github.com/nileshkokane01", "followers_url": "https://api.gith...
[]
closed
false
null
[]
[ "I do not know why tests_tf, tests_torch, run_tests and tests_jax is failing. \r\nCan anyone let me know? I can work on it to resolve.", "> make sure the mask_token is initialized the same way as an added token for both the fast and the slow tokenizers\r\n\r\nYou are right. [This](https://github.com/huggingface/...
1,700
1,707
1,701
CONTRIBUTOR
null
# What does this PR do? It addresses #16627 to create test cases for RemBert. <!-- Congratulations! You've made it this far! You're not quite done yet though. Once merged, your PR is going to appear in the release notes with the title you set, so make sure it's a great title that fully reflects the extent o...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27637/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27637/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27637", "html_url": "https://github.com/huggingface/transformers/pull/27637", "diff_url": "https://github.com/huggingface/transformers/pull/27637.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27637.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27636
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27636/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27636/comments
https://api.github.com/repos/huggingface/transformers/issues/27636/events
https://github.com/huggingface/transformers/pull/27636
2,004,741,340
PR_kwDOCUB6oc5gC7li
27,636
Reflect RoCm support in the documentation
{ "login": "fxmarty", "id": 9808326, "node_id": "MDQ6VXNlcjk4MDgzMjY=", "avatar_url": "https://avatars.githubusercontent.com/u/9808326?v=4", "gravatar_id": "", "url": "https://api.github.com/users/fxmarty", "html_url": "https://github.com/fxmarty", "followers_url": "https://api.github.com/users/fxmarty/...
[]
closed
false
null
[]
[ "We may want to update https://huggingface.co/docs/transformers/perf_hardware as well, explaining `rocm-smi --showtopoweight` and `rocm-smi --shownodesbw` output.", "@LysandreJik @ArthurZucker @amyeroberts WDYT?", "_The documentation is not available anymore as the PR was closed or merged._", "Thanks @ArthurZ...
1,700
1,700
1,700
COLLABORATOR
null
As per title. We will need https://github.com/huggingface/optimum/pull/1546 to be merged first, and an Optimum release.
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27636/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27636/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27636", "html_url": "https://github.com/huggingface/transformers/pull/27636", "diff_url": "https://github.com/huggingface/transformers/pull/27636.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27636.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27635
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27635/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27635/comments
https://api.github.com/repos/huggingface/transformers/issues/27635/events
https://github.com/huggingface/transformers/pull/27635
2,004,708,527
PR_kwDOCUB6oc5gC0Zv
27,635
Explicitely specify `use_cache=True` in Flash Attention tests
{ "login": "fxmarty", "id": 9808326, "node_id": "MDQ6VXNlcjk4MDgzMjY=", "avatar_url": "https://avatars.githubusercontent.com/u/9808326?v=4", "gravatar_id": "", "url": "https://api.github.com/users/fxmarty", "html_url": "https://github.com/fxmarty", "followers_url": "https://api.github.com/users/fxmarty/...
[]
closed
false
null
[]
[ "_The documentation is not available anymore as the PR was closed or merged._" ]
1,700
1,700
1,700
COLLABORATOR
null
Fixes https://github.com/huggingface/transformers/pull/27625#issuecomment-1821029483 making things more readable
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27635/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27635/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27635", "html_url": "https://github.com/huggingface/transformers/pull/27635", "diff_url": "https://github.com/huggingface/transformers/pull/27635.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27635.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27634
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27634/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27634/comments
https://api.github.com/repos/huggingface/transformers/issues/27634/events
https://github.com/huggingface/transformers/pull/27634
2,004,683,236
PR_kwDOCUB6oc5gCuxs
27,634
Update chat template warnings/guides
{ "login": "Rocketknight1", "id": 12866554, "node_id": "MDQ6VXNlcjEyODY2NTU0", "avatar_url": "https://avatars.githubusercontent.com/u/12866554?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Rocketknight1", "html_url": "https://github.com/Rocketknight1", "followers_url": "https://api.githu...
[]
closed
false
null
[]
[ "_The documentation is not available anymore as the PR was closed or merged._" ]
1,700
1,701
1,701
MEMBER
null
The default ChatML template doesn't add a BOS token, but some models expect it. This PR makes a small update to the docs and warning messages to make users more aware of the potential issue here.
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27634/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27634/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27634", "html_url": "https://github.com/huggingface/transformers/pull/27634", "diff_url": "https://github.com/huggingface/transformers/pull/27634.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27634.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27633
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27633/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27633/comments
https://api.github.com/repos/huggingface/transformers/issues/27633/events
https://github.com/huggingface/transformers/pull/27633
2,004,576,314
PR_kwDOCUB6oc5gCWoL
27,633
Add deepspeed test to amd scheduled CI
{ "login": "echarlaix", "id": 80481427, "node_id": "MDQ6VXNlcjgwNDgxNDI3", "avatar_url": "https://avatars.githubusercontent.com/u/80481427?v=4", "gravatar_id": "", "url": "https://api.github.com/users/echarlaix", "html_url": "https://github.com/echarlaix", "followers_url": "https://api.github.com/users/...
[]
closed
false
null
[]
[ "The docs for this PR live [here](https://moon-ci-docs.huggingface.co/docs/transformers/pr_27633). All of your documentation changes will be reflected on that endpoint.", "There is a PR merged before this one\r\n\r\nhttps://github.com/huggingface/transformers/pull/27743\r\n\r\nDon't hesitate if you need my help o...
1,700
1,702
1,702
COLLABORATOR
null
null
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27633/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27633/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27633", "html_url": "https://github.com/huggingface/transformers/pull/27633", "diff_url": "https://github.com/huggingface/transformers/pull/27633.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27633.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27632
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27632/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27632/comments
https://api.github.com/repos/huggingface/transformers/issues/27632/events
https://github.com/huggingface/transformers/issues/27632
2,004,313,660
I_kwDOCUB6oc53d2Y8
27,632
Deployment in notebook(Kaggle) Transformers can't Integration DeepSpeed
{ "login": "ohand007", "id": 66542185, "node_id": "MDQ6VXNlcjY2NTQyMTg1", "avatar_url": "https://avatars.githubusercontent.com/u/66542185?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ohand007", "html_url": "https://github.com/ohand007", "followers_url": "https://api.github.com/users/oha...
[]
closed
false
{ "login": "pacman100", "id": 13534540, "node_id": "MDQ6VXNlcjEzNTM0NTQw", "avatar_url": "https://avatars.githubusercontent.com/u/13534540?v=4", "gravatar_id": "", "url": "https://api.github.com/users/pacman100", "html_url": "https://github.com/pacman100", "followers_url": "https://api.github.com/users/...
[ { "login": "pacman100", "id": 13534540, "node_id": "MDQ6VXNlcjEzNTM0NTQw", "avatar_url": "https://avatars.githubusercontent.com/u/13534540?v=4", "gravatar_id": "", "url": "https://api.github.com/users/pacman100", "html_url": "https://github.com/pacman100", "followers_url": "https://a...
[ "cc @pacman100 and @muellerzr ", "99% sure you can't use DeepSpeed in a notebook. @pacman100 ", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow th...
1,700
1,706
1,706
NONE
null
### System Info transformers==4.35.0 deepspeed==0.12.2 pytorch==2.0.0 ### Who can help? _No response_ ### Information - [X] The official example scripts - [x] My own modified scripts ### Tasks - [x] An officially supported task in the `examples` folder (such as GLUE/SQuAD, ...) - [ ] My own tas...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27632/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27632/timeline
completed
null
null
https://api.github.com/repos/huggingface/transformers/issues/27631
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27631/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27631/comments
https://api.github.com/repos/huggingface/transformers/issues/27631/events
https://github.com/huggingface/transformers/issues/27631
2,004,233,116
I_kwDOCUB6oc53diuc
27,631
Add the learning rate (in exponential representation, like "9.197948717948718e-08") to the model training output
{ "login": "blademoon", "id": 17560478, "node_id": "MDQ6VXNlcjE3NTYwNDc4", "avatar_url": "https://avatars.githubusercontent.com/u/17560478?v=4", "gravatar_id": "", "url": "https://api.github.com/users/blademoon", "html_url": "https://github.com/blademoon", "followers_url": "https://api.github.com/users/...
[ { "id": 2648621985, "node_id": "MDU6TGFiZWwyNjQ4NjIxOTg1", "url": "https://api.github.com/repos/huggingface/transformers/labels/Feature%20request", "name": "Feature request", "color": "FBCA04", "default": false, "description": "Request for a new feature" } ]
open
false
{ "login": "muellerzr", "id": 7831895, "node_id": "MDQ6VXNlcjc4MzE4OTU=", "avatar_url": "https://avatars.githubusercontent.com/u/7831895?v=4", "gravatar_id": "", "url": "https://api.github.com/users/muellerzr", "html_url": "https://github.com/muellerzr", "followers_url": "https://api.github.com/users/mu...
[ { "login": "muellerzr", "id": 7831895, "node_id": "MDQ6VXNlcjc4MzE4OTU=", "avatar_url": "https://avatars.githubusercontent.com/u/7831895?v=4", "gravatar_id": "", "url": "https://api.github.com/users/muellerzr", "html_url": "https://github.com/muellerzr", "followers_url": "https://api...
[ "Hey! I think the way you log it is nice, just needs the log to be in scientific format! ", "@ArthurZucker I already try convert `float` to scientific notation by using:\r\n```python\r\n...\r\n logs[\"learning_rate\"] = \"{:.2e}\".format(self._get_learning_rate())\r\n...\r\n```\r\n\r\nIt dosent work. Tensorboa...
1,700
1,702
null
NONE
null
### Feature request Good day. Can you add current learning rate to output of training process. Now, if we train model with Trainer we get those output (without leraning rate): <img width="302" alt="image" src="https://github.com/huggingface/transformers/assets/17560478/239c3ce1-c7dc-4d79-9664-52ac4e3ea605"> ...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27631/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27631/timeline
null
null
null
https://api.github.com/repos/huggingface/transformers/issues/27630
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27630/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27630/comments
https://api.github.com/repos/huggingface/transformers/issues/27630/events
https://github.com/huggingface/transformers/issues/27630
2,004,224,309
I_kwDOCUB6oc53dgk1
27,630
Error while trying to resume from checkpoint
{ "login": "sstoia", "id": 129397487, "node_id": "U_kgDOB7Zy7w", "avatar_url": "https://avatars.githubusercontent.com/u/129397487?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sstoia", "html_url": "https://github.com/sstoia", "followers_url": "https://api.github.com/users/sstoia/follower...
[]
open
false
{ "login": "muellerzr", "id": 7831895, "node_id": "MDQ6VXNlcjc4MzE4OTU=", "avatar_url": "https://avatars.githubusercontent.com/u/7831895?v=4", "gravatar_id": "", "url": "https://api.github.com/users/muellerzr", "html_url": "https://github.com/muellerzr", "followers_url": "https://api.github.com/users/mu...
[ { "login": "muellerzr", "id": 7831895, "node_id": "MDQ6VXNlcjc4MzE4OTU=", "avatar_url": "https://avatars.githubusercontent.com/u/7831895?v=4", "gravatar_id": "", "url": "https://api.github.com/users/muellerzr", "html_url": "https://github.com/muellerzr", "followers_url": "https://api...
[ "cc @muellerzr ", "gently pinging @muellerzr or @SunMarc if he has time look into it! " ]
1,700
1,707
null
NONE
null
### System Info - `transformers` version: 4.28.1 - Platform: Linux-3.10.0-1160.95.1.el7.x86_64-x86_64-with-glibc2.17 - Python version: 3.9.16 - Huggingface_hub version: 0.19.3 - Safetensors version: 0.4.0 - PyTorch version (GPU?): 1.13.1 (False) - Tensorflow version (GPU?): not installed (NA) - Flax version (CP...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27630/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27630/timeline
null
null
null
https://api.github.com/repos/huggingface/transformers/issues/27629
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27629/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27629/comments
https://api.github.com/repos/huggingface/transformers/issues/27629/events
https://github.com/huggingface/transformers/issues/27629
2,004,135,688
I_kwDOCUB6oc53dK8I
27,629
Whisper: Language detection not working after setting a language once
{ "login": "brunjo", "id": 1618488, "node_id": "MDQ6VXNlcjE2MTg0ODg=", "avatar_url": "https://avatars.githubusercontent.com/u/1618488?v=4", "gravatar_id": "", "url": "https://api.github.com/users/brunjo", "html_url": "https://github.com/brunjo", "followers_url": "https://api.github.com/users/brunjo/foll...
[]
closed
false
null
[]
[ "cc @sanchit-gandhi and @ylacombe ", "I've found a workaround: Running `del pipe.model.generation_config.language` before calling the `pipe` method resets the language parameter and fixes the issue." ]
1,700
1,701
1,701
NONE
null
### System Info - `transformers` version: 4.36.0.dev0 - Platform: Linux-6.2.0-1018-gcp-x86_64-with-glibc2.31 - Python version: 3.11.6 - Huggingface_hub version: 0.19.4 - Safetensors version: 0.4.0 - Accelerate version: 0.24.1 - Accelerate config: not found - PyTorch version (GPU?): 2.1.1+cu121 (True) - Tens...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27629/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27629/timeline
completed
null
null
https://api.github.com/repos/huggingface/transformers/issues/27628
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27628/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27628/comments
https://api.github.com/repos/huggingface/transformers/issues/27628/events
https://github.com/huggingface/transformers/pull/27628
2,004,094,694
PR_kwDOCUB6oc5gAsd3
27,628
update Openai API call method
{ "login": "Strive-for-excellence", "id": 26090323, "node_id": "MDQ6VXNlcjI2MDkwMzIz", "avatar_url": "https://avatars.githubusercontent.com/u/26090323?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Strive-for-excellence", "html_url": "https://github.com/Strive-for-excellence", "followers_...
[]
closed
false
null
[]
[ "This is also same problem in the AzureOpenAiAgent.\r\n```\r\nclass AzureOpenAiAgent(Agent):\r\n.....\r\n def _chat_generate(self, prompt, stop):\r\n result = openai.ChatCompletion.create(\r\n engine=self.deployment_id,\r\n messages=[{\"role\": \"user\", \"content\": prompt}],\r\n ...
1,700
1,700
1,700
CONTRIBUTOR
null
# What does this PR do? Fixes #27623 ## Before submitting - [ ] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case). - [ ] Did you read the [contributor guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-reques...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27628/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27628/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27628", "html_url": "https://github.com/huggingface/transformers/pull/27628", "diff_url": "https://github.com/huggingface/transformers/pull/27628.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27628.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27627
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27627/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27627/comments
https://api.github.com/repos/huggingface/transformers/issues/27627/events
https://github.com/huggingface/transformers/issues/27627
2,004,024,924
I_kwDOCUB6oc53cv5c
27,627
Unable to register own custom tokenizer
{ "login": "Ranitbag007", "id": 133197492, "node_id": "U_kgDOB_ButA", "avatar_url": "https://avatars.githubusercontent.com/u/133197492?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Ranitbag007", "html_url": "https://github.com/Ranitbag007", "followers_url": "https://api.github.com/users/...
[ { "id": 1843244711, "node_id": "MDU6TGFiZWwxODQzMjQ0NzEx", "url": "https://api.github.com/repos/huggingface/transformers/labels/New%20model", "name": "New model", "color": "fbca04", "default": false, "description": "" } ]
open
false
null
[]
[ "Hey πŸ€— thanks for opening an issue! We try to keep the github issues for bugs/feature requests.\r\nCould you ask your question on the [forum](https://discuss.huggingface.co/) instead? I'm sure the community will be of help!\r\n\r\nThanks!" ]
1,700
1,700
null
NONE
null
### Model description @sgugger I have tried to register my own tokenization model based on sentencepiece using CustomAITokenizer.register_for_auto_class("AutoTokenizer") . But I am falied to do so. ### Open source status - [X] The model implementation is available - [ ] The model weights are available ### Provide ...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27627/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27627/timeline
null
null
null
https://api.github.com/repos/huggingface/transformers/issues/27626
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27626/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27626/comments
https://api.github.com/repos/huggingface/transformers/issues/27626/events
https://github.com/huggingface/transformers/issues/27626
2,003,985,273
I_kwDOCUB6oc53cmN5
27,626
8-Bit LlamaModel returns different outputs for the same input when in different batches
{ "login": "maximek3", "id": 37376714, "node_id": "MDQ6VXNlcjM3Mzc2NzE0", "avatar_url": "https://avatars.githubusercontent.com/u/37376714?v=4", "gravatar_id": "", "url": "https://api.github.com/users/maximek3", "html_url": "https://github.com/maximek3", "followers_url": "https://api.github.com/users/max...
[]
closed
false
null
[]
[ "Hi @maximek3 \r\nI think those small instabilities are expected, can you test out with `torch.allclose()` with different values of `rtol` and `atol` ? cc @SunMarc @Titus-von-Koeller in case I am missing anything", "Hi @younesbelkada, thanks for your quick response.\r\n\r\n`torch.allclose()` returns `False` even...
1,700
1,703
1,703
NONE
null
### System Info - `transformers` version: 4.36.0.dev0 - Platform: Linux-4.18.0-372.58.1.el8_6.x86_64-x86_64-with-glibc2.28 - Python version: 3.10.6 - Huggingface_hub version: 0.19.4 - Safetensors version: 0.4.0 - Accelerate version: 0.21.0 - Accelerate config: not found - PyTorch version (GPU?): 2.1.1+cu12...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27626/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27626/timeline
completed
null
null
https://api.github.com/repos/huggingface/transformers/issues/27625
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27625/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27625/comments
https://api.github.com/repos/huggingface/transformers/issues/27625/events
https://github.com/huggingface/transformers/pull/27625
2,003,909,504
PR_kwDOCUB6oc5gADzy
27,625
Fix flash attention bugs with Mistral and Falcon
{ "login": "fxmarty", "id": 9808326, "node_id": "MDQ6VXNlcjk4MDgzMjY=", "avatar_url": "https://avatars.githubusercontent.com/u/9808326?v=4", "gravatar_id": "", "url": "https://api.github.com/users/fxmarty", "html_url": "https://github.com/fxmarty", "followers_url": "https://api.github.com/users/fxmarty/...
[]
closed
false
null
[]
[ "_The documentation is not available anymore as the PR was closed or merged._", "Hi @amyeroberts thank you for the review! I added by mistake `test_flash_attn_2_generate_use_cache` in `test_modeling_llama.py` while it was meant to be added in `test_modeling_mistral.py`, hence the confusion, apology!", "@fxmarty...
1,700
1,700
1,700
COLLABORATOR
null
This PR fixes some important bugs in the Mistral and Falcon integration. https://github.com/huggingface/transformers/pull/26933 broke flash attention for Falcon due to the modification of the layout The following tests were not passing: ``` FAILED tests/models/mistral/test_modeling_mistral.py::MistralModelTest:...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27625/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27625/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27625", "html_url": "https://github.com/huggingface/transformers/pull/27625", "diff_url": "https://github.com/huggingface/transformers/pull/27625.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27625.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27624
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27624/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27624/comments
https://api.github.com/repos/huggingface/transformers/issues/27624/events
https://github.com/huggingface/transformers/pull/27624
2,003,882,145
PR_kwDOCUB6oc5f_9zL
27,624
Fix `max_steps` documentation regarding the end-of-training condition
{ "login": "qgallouedec", "id": 45557362, "node_id": "MDQ6VXNlcjQ1NTU3MzYy", "avatar_url": "https://avatars.githubusercontent.com/u/45557362?v=4", "gravatar_id": "", "url": "https://api.github.com/users/qgallouedec", "html_url": "https://github.com/qgallouedec", "followers_url": "https://api.github.com/...
[]
closed
false
null
[]
[]
1,700
1,701
1,700
CONTRIBUTOR
null
# What does this PR do? Fixes #26635 ## Before submitting - [x] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case). - [ ] Did you read the [contributor guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-request...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27624/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27624/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27624", "html_url": "https://github.com/huggingface/transformers/pull/27624", "diff_url": "https://github.com/huggingface/transformers/pull/27624.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27624.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27623
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27623/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27623/comments
https://api.github.com/repos/huggingface/transformers/issues/27623/events
https://github.com/huggingface/transformers/issues/27623
2,003,818,000
I_kwDOCUB6oc53b9YQ
27,623
Issue with transformers==4.29.0 and openai==1.3.4 Integration
{ "login": "Strive-for-excellence", "id": 26090323, "node_id": "MDQ6VXNlcjI2MDkwMzIz", "avatar_url": "https://avatars.githubusercontent.com/u/26090323?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Strive-for-excellence", "html_url": "https://github.com/Strive-for-excellence", "followers_...
[]
closed
false
null
[]
[ "Hey! Thanks for reporting πŸ€— would you like to open a PR for a fix? " ]
1,700
1,700
1,700
CONTRIBUTOR
null
### System Info transformers==4.29.0 openai==1.3.4 When running the code provided in the version openai==1.3.4, which can be found at the following link: https://huggingface.co/docs/transformers/transformers_agents, an error occurs: ``` APIRemovedInV1 Traceback (most recent call last...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27623/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27623/timeline
completed
null
null
https://api.github.com/repos/huggingface/transformers/issues/27622
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27622/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27622/comments
https://api.github.com/repos/huggingface/transformers/issues/27622/events
https://github.com/huggingface/transformers/pull/27622
2,003,687,080
PR_kwDOCUB6oc5f_TUu
27,622
[Fuyu and Persimmon] Add FA2 and fused kernels
{ "login": "jzhang38", "id": 42993249, "node_id": "MDQ6VXNlcjQyOTkzMjQ5", "avatar_url": "https://avatars.githubusercontent.com/u/42993249?v=4", "gravatar_id": "", "url": "https://api.github.com/users/jzhang38", "html_url": "https://github.com/jzhang38", "followers_url": "https://api.github.com/users/jzh...
[]
closed
false
null
[]
[ "@jzhang38 \r\n\r\nI'm close to finishing the FA2 implementation for Persimmon -- see #27052. Also am developing a separate package for loading cuda kernels dynamically that would allow for incorporating fused kernels (including FA's fused ops / layers). ", "This issue has been automatically marked as stale beca...
1,700
1,703
1,703
NONE
null
# What does this PR do? - [ ] Add FA2 to Fuyu and Persimmon(just like it has been done to Llama through use_flash_attention_2=True) - [ ] Add other [fused kernels from the FA2 repo](https://github.com/Dao-AILab/flash-attention/blob/2c3baba4a63c4007c8a132c5380edc9430f88a22/training/README.md?plain=1#L66) - [ ] Add a ...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27622/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27622/timeline
null
true
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27622", "html_url": "https://github.com/huggingface/transformers/pull/27622", "diff_url": "https://github.com/huggingface/transformers/pull/27622.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27622.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27621
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27621/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27621/comments
https://api.github.com/repos/huggingface/transformers/issues/27621/events
https://github.com/huggingface/transformers/pull/27621
2,003,683,699
PR_kwDOCUB6oc5f_Slj
27,621
[ConvNext] Improve backbone
{ "login": "NielsRogge", "id": 48327001, "node_id": "MDQ6VXNlcjQ4MzI3MDAx", "avatar_url": "https://avatars.githubusercontent.com/u/48327001?v=4", "gravatar_id": "", "url": "https://api.github.com/users/NielsRogge", "html_url": "https://github.com/NielsRogge", "followers_url": "https://api.github.com/use...
[]
closed
false
null
[]
[]
1,700
1,700
1,700
CONTRIBUTOR
null
# What does this PR do? This PR makes sure that ConvNextBackbone is implemented as other backbones in the library, meaning no hardcoded `return_dict=True`.
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27621/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27621/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27621", "html_url": "https://github.com/huggingface/transformers/pull/27621", "diff_url": "https://github.com/huggingface/transformers/pull/27621.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27621.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27620
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27620/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27620/comments
https://api.github.com/repos/huggingface/transformers/issues/27620/events
https://github.com/huggingface/transformers/issues/27620
2,003,557,732
I_kwDOCUB6oc53a91k
27,620
How can check hugging face pretrained model has metadata attribute?
{ "login": "JinSeoung-Oh", "id": 78573459, "node_id": "MDQ6VXNlcjc4NTczNDU5", "avatar_url": "https://avatars.githubusercontent.com/u/78573459?v=4", "gravatar_id": "", "url": "https://api.github.com/users/JinSeoung-Oh", "html_url": "https://github.com/JinSeoung-Oh", "followers_url": "https://api.github.c...
[]
closed
false
null
[]
[ "Hey! πŸ€— thanks for the kind words!\r\nCould you share the full traceback and how you called `transformers`. \r\nI am not sure about the integration with Llama-index and I am this unfamiliar with the Metadata attribute", "It's likely this is something outside `transformers`, and need to open an issue in those lib...
1,700
1,700
1,700
NONE
null
Hi, at first, I want to say thanks for you guys work and contribution. My question is how can I check hugging face model has metadata attribute. Theses day, I'm building GraphRAG with Llama-index and NubularDB for korean Baseline for this work has already been completed. But korean LLM performance is not good, so...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27620/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27620/timeline
completed
null
null
https://api.github.com/repos/huggingface/transformers/issues/27619
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27619/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27619/comments
https://api.github.com/repos/huggingface/transformers/issues/27619/events
https://github.com/huggingface/transformers/issues/27619
2,003,408,505
I_kwDOCUB6oc53aZZ5
27,619
runtime error clip_vision_bert TypeError: __init__() got an unexpected keyword argument '_do_init'
{ "login": "guanhdrmq", "id": 81207745, "node_id": "MDQ6VXNlcjgxMjA3NzQ1", "avatar_url": "https://avatars.githubusercontent.com/u/81207745?v=4", "gravatar_id": "", "url": "https://api.github.com/users/guanhdrmq", "html_url": "https://github.com/guanhdrmq", "followers_url": "https://api.github.com/users/...
[]
closed
false
null
[]
[ "Hey! Can you update the version of transformers you are using ? \r\nHey! Seems like an error with `multilingual_vqa` calling from pretrained. Would recommend you to upgrade to a more recent version but also isolate this to transformers or `multilingual_vqa` and post the issue there! ", "Hi! @ArthurZucker. Thanks...
1,700
1,704
1,704
NONE
null
### System Info transformer 4.25.1 python 3.9 torch 1.13 windows 10 or ubuntu 20 ### Who can help? @sanchit-gandhi ### Information - [X] The official example scripts - [ ] My own modified scripts ### Tasks - [X] An officially supported task in the `examples` folder (such as GLUE/SQuAD, ...) - ...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27619/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27619/timeline
completed
null
null
https://api.github.com/repos/huggingface/transformers/issues/27618
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27618/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27618/comments
https://api.github.com/repos/huggingface/transformers/issues/27618/events
https://github.com/huggingface/transformers/issues/27618
2,003,365,582
I_kwDOCUB6oc53aO7O
27,618
I get 【KeyError: 'label'】 when I run transformers/examples/pytorch/text-classification/run_glue.py
{ "login": "WeiChunyu-star", "id": 54472778, "node_id": "MDQ6VXNlcjU0NDcyNzc4", "avatar_url": "https://avatars.githubusercontent.com/u/54472778?v=4", "gravatar_id": "", "url": "https://api.github.com/users/WeiChunyu-star", "html_url": "https://github.com/WeiChunyu-star", "followers_url": "https://api.gi...
[]
closed
false
null
[]
[]
1,700
1,700
1,700
NONE
null
### System Info transformers: 4.36.0.dev0 accelerate : 0.24.1 platform: x86+V100 ### Who can help? _No response_ ### Information - [x] The official example scripts - [ ] My own modified scripts ### Tasks - [x] An officially supported task in the `examples` folder (such as GLUE/SQuAD, ...) - [ ] My own task or ...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27618/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27618/timeline
completed
null
null
https://api.github.com/repos/huggingface/transformers/issues/27617
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27617/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27617/comments
https://api.github.com/repos/huggingface/transformers/issues/27617/events
https://github.com/huggingface/transformers/pull/27617
2,003,364,444
PR_kwDOCUB6oc5f-NnG
27,617
remove the deprecated method `init_git_repo`
{ "login": "statelesshz", "id": 28150734, "node_id": "MDQ6VXNlcjI4MTUwNzM0", "avatar_url": "https://avatars.githubusercontent.com/u/28150734?v=4", "gravatar_id": "", "url": "https://api.github.com/users/statelesshz", "html_url": "https://github.com/statelesshz", "followers_url": "https://api.github.com/...
[]
closed
false
null
[]
[]
1,700
1,700
1,700
CONTRIBUTOR
null
# What does this PR do? As docstring says `init_git_repo is deprecated and will be removed in v4.34.0 of Transformers.` ## Before submitting - [ ] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case). - [x] Did you read the [contributor guideline](https://github.com/...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27617/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27617/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27617", "html_url": "https://github.com/huggingface/transformers/pull/27617", "diff_url": "https://github.com/huggingface/transformers/pull/27617.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27617.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27616
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27616/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27616/comments
https://api.github.com/repos/huggingface/transformers/issues/27616/events
https://github.com/huggingface/transformers/pull/27616
2,003,268,459
PR_kwDOCUB6oc5f95nY
27,616
Code refactor for nested conditional statements
{ "login": "YeonwooSung", "id": 30489717, "node_id": "MDQ6VXNlcjMwNDg5NzE3", "avatar_url": "https://avatars.githubusercontent.com/u/30489717?v=4", "gravatar_id": "", "url": "https://api.github.com/users/YeonwooSung", "html_url": "https://github.com/YeonwooSung", "followers_url": "https://api.github.com/...
[]
closed
false
null
[]
[ "> Hey! Would recommend you to install `ruff==1.5` to make sure some of the quality check pass. Not against refactoring as long as the CI are all green!\r\n\r\nThanks. Will definitely try this!", "The docs for this PR live [here](https://moon-ci-docs.huggingface.co/docs/transformers/pr_27616). All of your documen...
1,700
1,700
1,700
CONTRIBUTOR
null
# What does this PR do? Simple code refactoring for "src/transformers/hf_argparser.py" and "src/transformers/image_utils.py". ## Before submitting - [ ] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case). - [v] Did you read the [contributor guideline](https://github....
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27616/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27616/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27616", "html_url": "https://github.com/huggingface/transformers/pull/27616", "diff_url": "https://github.com/huggingface/transformers/pull/27616.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27616.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27615
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27615/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27615/comments
https://api.github.com/repos/huggingface/transformers/issues/27615/events
https://github.com/huggingface/transformers/issues/27615
2,003,244,755
I_kwDOCUB6oc53ZxbT
27,615
How to get the number of trainable parameters for a hf model
{ "login": "mathmax12", "id": 32367611, "node_id": "MDQ6VXNlcjMyMzY3NjEx", "avatar_url": "https://avatars.githubusercontent.com/u/32367611?v=4", "gravatar_id": "", "url": "https://api.github.com/users/mathmax12", "html_url": "https://github.com/mathmax12", "followers_url": "https://api.github.com/users/...
[]
closed
false
null
[]
[ "Hey, I would use something like this:\r\n```python \r\ndef count_trainable_parameters(model):\r\n model_parameters = filter(lambda p: p.requires_grad, model.parameters())\r\n params = sum([np.prod(p.size()) for p in model_parameters])\r\n return params\r\n```\r\n\r\nnote that we try to keep the github is...
1,700
1,700
1,700
NONE
null
### Feature request ' peft_parameters = LoraConfig( lora_alpha=16, lora_dropout=0.1, r=8, bias="none", task_type="CAUSAL_LM" ) train_params = TrainingArguments( output_dir="./results_modified", num_train_epochs=1, per_device_train_batch_size=4, gradient_accumulation_step...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27615/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27615/timeline
completed
null
null
https://api.github.com/repos/huggingface/transformers/issues/27614
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27614/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27614/comments
https://api.github.com/repos/huggingface/transformers/issues/27614/events
https://github.com/huggingface/transformers/issues/27614
2,003,237,457
I_kwDOCUB6oc53ZvpR
27,614
Nougat-small VisionEncoderDecoderModel failed when max_new_tokens > 3584, Index out of range in self
{ "login": "c3-YuelingWu", "id": 103061109, "node_id": "U_kgDOBiSWdQ", "avatar_url": "https://avatars.githubusercontent.com/u/103061109?v=4", "gravatar_id": "", "url": "https://api.github.com/users/c3-YuelingWu", "html_url": "https://github.com/c3-YuelingWu", "followers_url": "https://api.github.com/use...
[ { "id": 3081136536, "node_id": "MDU6TGFiZWwzMDgxMTM2NTM2", "url": "https://api.github.com/repos/huggingface/transformers/labels/Good%20Difficult%20Issue", "name": "Good Difficult Issue", "color": "684CC7", "default": false, "description": "" } ]
open
false
null
[]
[ "cc @molbap and @NielsRogge who ported this model! ", "Hi @c3-YuelingWu ! In `nougat-base` tokenizer, `model_max_length` and `max_length` are both at 4096. In `nougat-small`, they are at 3584. However it's indeed anormal that nougat-small is generating 3584 tokens when nougat-base is generating only 6 with the sa...
1,700
1,707
null
NONE
null
### System Info # System Info - transformers == 4.35.0 - torch == 2.0.1+cu117 - Error shows on both cpu and cuda. # Task Use nougat to parse text from large documents. # Issue NougatTokenizerFast. model_max_length = 3584 . Setting max_new_tokens=3585 (or larger) in model.generate works well on most pages, b...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27614/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27614/timeline
reopened
null
null
https://api.github.com/repos/huggingface/transformers/issues/27613
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27613/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27613/comments
https://api.github.com/repos/huggingface/transformers/issues/27613/events
https://github.com/huggingface/transformers/issues/27613
2,002,837,466
I_kwDOCUB6oc53YN_a
27,613
RuntimeError "Some tensors share memory" occurred when saving checkpoints during LoRA
{ "login": "Yuta555", "id": 59324565, "node_id": "MDQ6VXNlcjU5MzI0NTY1", "avatar_url": "https://avatars.githubusercontent.com/u/59324565?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Yuta555", "html_url": "https://github.com/Yuta555", "followers_url": "https://api.github.com/users/Yuta55...
[]
closed
false
null
[]
[ "cc @LysandreJik, should we do what we did similar for accelerate where we just warned instead? ", "What we did in accelerate: https://github.com/huggingface/accelerate/pull/2136 cc @SunMarc ", "cc @Narsil also", "I've also hit this, and it seems like a backwards incompatible change from `4.33.2`. I apprecia...
1,700
1,707
1,707
NONE
null
### System Info - `transformers` version: 4.35.0 - Platform: Linux-5.10.0-26-cloud-amd64-x86_64-with-glibc2.31 - Python version: 3.10.12 - Huggingface_hub version: 0.17.3 - Safetensors version: 0.4.0 - Accelerate version: 0.24.1 - Accelerate config: - compute_environment: LOCAL_MACHINE - distributed_...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27613/reactions", "total_count": 1, "+1": 1, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27613/timeline
completed
null
null
https://api.github.com/repos/huggingface/transformers/issues/27612
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27612/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27612/comments
https://api.github.com/repos/huggingface/transformers/issues/27612/events
https://github.com/huggingface/transformers/pull/27612
2,002,694,649
PR_kwDOCUB6oc5f7764
27,612
Generate: Update docs regarding reusing `past_key_values` in `generate`
{ "login": "gante", "id": 12240844, "node_id": "MDQ6VXNlcjEyMjQwODQ0", "avatar_url": "https://avatars.githubusercontent.com/u/12240844?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gante", "html_url": "https://github.com/gante", "followers_url": "https://api.github.com/users/gante/follow...
[]
closed
false
null
[]
[ "_The documentation is not available anymore as the PR was closed or merged._", "The right-handed TOC is now fully functional!\r\n\r\n<img width=\"239\" alt=\"Screenshot 2023-11-20 at 18 31 56\" src=\"https://github.com/huggingface/transformers/assets/12240844/cb7da28e-adf2-4bc6-a71c-09e8343b984d\">\r\n" ]
1,700
1,700
1,700
MEMBER
null
# What does this PR do? Due to a [recent PR](https://github.com/huggingface/transformers/pull/25086), which enables `generate` to return `past_key_values`, we can now reuse `past_key_values` to speed up multi-round conversations with LLMs. This was already described in the [LLM optimization docs](https://huggingface...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27612/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27612/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27612", "html_url": "https://github.com/huggingface/transformers/pull/27612", "diff_url": "https://github.com/huggingface/transformers/pull/27612.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27612.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27611
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27611/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27611/comments
https://api.github.com/repos/huggingface/transformers/issues/27611/events
https://github.com/huggingface/transformers/pull/27611
2,002,675,861
PR_kwDOCUB6oc5f734U
27,611
Flash Attention 2 support for RoCm
{ "login": "fxmarty", "id": 9808326, "node_id": "MDQ6VXNlcjk4MDgzMjY=", "avatar_url": "https://avatars.githubusercontent.com/u/9808326?v=4", "gravatar_id": "", "url": "https://api.github.com/users/fxmarty", "html_url": "https://github.com/fxmarty", "followers_url": "https://api.github.com/users/fxmarty/...
[]
closed
false
null
[]
[ "The docs for this PR live [here](https://moon-ci-docs.huggingface.co/docs/transformers/pr_27611). All of your documentation changes will be reflected on that endpoint.", "@LysandreJik @ArthurZucker @amyeroberts @younesbelkada WDYT?", "Reviewing now\r\n" ]
1,700
1,701
1,701
COLLABORATOR
null
As per title. https://github.com/ROCmSoftwarePlatform/flash-attention has been bumped to `2.0.4` recently, and we expect it to be bumped to `2.1` soon. Meanwhile, this PR adds suport for FA2 on RoCm devices. It can be simplified once FA2 RoCm version is bumped to `2.1`. Tests that are passing on A100 are also pas...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27611/reactions", "total_count": 1, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 1 }
https://api.github.com/repos/huggingface/transformers/issues/27611/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27611", "html_url": "https://github.com/huggingface/transformers/pull/27611", "diff_url": "https://github.com/huggingface/transformers/pull/27611.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27611.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27610
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27610/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27610/comments
https://api.github.com/repos/huggingface/transformers/issues/27610/events
https://github.com/huggingface/transformers/pull/27610
2,002,672,467
PR_kwDOCUB6oc5f73Ii
27,610
[`core` / `gradient_checkpointing`] add support for old GC method
{ "login": "younesbelkada", "id": 49240599, "node_id": "MDQ6VXNlcjQ5MjQwNTk5", "avatar_url": "https://avatars.githubusercontent.com/u/49240599?v=4", "gravatar_id": "", "url": "https://api.github.com/users/younesbelkada", "html_url": "https://github.com/younesbelkada", "followers_url": "https://api.githu...
[]
closed
false
null
[]
[ "As discussed offline, let's just display a warning for now, merging! " ]
1,700
1,700
1,700
CONTRIBUTOR
null
# What does this PR do? Fixes https://github.com/huggingface/transformers/issues/27596 Models on the Hub still rely on the previous `_set_gradient_checkpointing` logic that has been removed and refactored in https://github.com/huggingface/transformers/pull/27073 While models with code on the Hub maintainers s...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27610/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27610/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27610", "html_url": "https://github.com/huggingface/transformers/pull/27610", "diff_url": "https://github.com/huggingface/transformers/pull/27610.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27610.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27609
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27609/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27609/comments
https://api.github.com/repos/huggingface/transformers/issues/27609/events
https://github.com/huggingface/transformers/issues/27609
2,002,526,716
I_kwDOCUB6oc53XCH8
27,609
Extend Chat Template Tokenization for Training/Finetuning
{ "login": "siddk", "id": 2498509, "node_id": "MDQ6VXNlcjI0OTg1MDk=", "avatar_url": "https://avatars.githubusercontent.com/u/2498509?v=4", "gravatar_id": "", "url": "https://api.github.com/users/siddk", "html_url": "https://github.com/siddk", "followers_url": "https://api.github.com/users/siddk/follower...
[ { "id": 2648621985, "node_id": "MDU6TGFiZWwyNjQ4NjIxOTg1", "url": "https://api.github.com/repos/huggingface/transformers/labels/Feature%20request", "name": "Feature request", "color": "FBCA04", "default": false, "description": "Request for a new feature" } ]
open
false
null
[]
[ "FYI @Rocketknight1 \r\n\r\nThis would also need a support for chat templates in tokenizers IMO", "Hey @siddk, this definitely seems like a good suggestion, and mirrors suggestions I got from e.g. @philschmid!\r\n\r\nThe first step is relatively easy - we could just check the first element of the input to figure ...
1,700
1,706
null
CONTRIBUTOR
null
### Feature request Extend `tokenizer.apply_chat_template` with functionality for training/finetuning, returning `attention_masks` and (optional) `labels` (for ignoring "System" and "User" messages during loss computation). I think this requires the following steps: - Adding support for taking in a batch of conve...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27609/reactions", "total_count": 2, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 2 }
https://api.github.com/repos/huggingface/transformers/issues/27609/timeline
null
null
null
https://api.github.com/repos/huggingface/transformers/issues/27608
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27608/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27608/comments
https://api.github.com/repos/huggingface/transformers/issues/27608/events
https://github.com/huggingface/transformers/pull/27608
2,002,486,268
PR_kwDOCUB6oc5f7Ob1
27,608
dvclive callback: warn instead of fail when logging non-scalars
{ "login": "dberenbaum", "id": 2308172, "node_id": "MDQ6VXNlcjIzMDgxNzI=", "avatar_url": "https://avatars.githubusercontent.com/u/2308172?v=4", "gravatar_id": "", "url": "https://api.github.com/users/dberenbaum", "html_url": "https://github.com/dberenbaum", "followers_url": "https://api.github.com/users...
[]
closed
false
null
[]
[ "@muellerz This makes the tests pass, but I'm not sure if it's intended that the test here logs the learning rate as a list rather than as a scalar (which will fail under several of the existing loggers, but only with a warning like in this PR). \r\n\r\nhttps://github.com/huggingface/transformers/blob/e4280d650c579...
1,700
1,700
1,700
CONTRIBUTOR
null
# What does this PR do? Fixes https://github.com/huggingface/transformers/pull/27352#issuecomment-1819131456. This will warn instead of fail when trying to log non-scalars as metrics. ## Before submitting - [ ] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case). - [x...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27608/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27608/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27608", "html_url": "https://github.com/huggingface/transformers/pull/27608", "diff_url": "https://github.com/huggingface/transformers/pull/27608.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27608.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27607
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27607/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27607/comments
https://api.github.com/repos/huggingface/transformers/issues/27607/events
https://github.com/huggingface/transformers/pull/27607
2,002,449,277
PR_kwDOCUB6oc5f7GZ3
27,607
Deprecate `TransfoXL`
{ "login": "ydshieh", "id": 2521628, "node_id": "MDQ6VXNlcjI1MjE2Mjg=", "avatar_url": "https://avatars.githubusercontent.com/u/2521628?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ydshieh", "html_url": "https://github.com/ydshieh", "followers_url": "https://api.github.com/users/ydshieh/...
[]
closed
false
{ "login": "LysandreJik", "id": 30755778, "node_id": "MDQ6VXNlcjMwNzU1Nzc4", "avatar_url": "https://avatars.githubusercontent.com/u/30755778?v=4", "gravatar_id": "", "url": "https://api.github.com/users/LysandreJik", "html_url": "https://github.com/LysandreJik", "followers_url": "https://api.github.com/...
[ { "login": "LysandreJik", "id": 30755778, "node_id": "MDQ6VXNlcjMwNzU1Nzc4", "avatar_url": "https://avatars.githubusercontent.com/u/30755778?v=4", "gravatar_id": "", "url": "https://api.github.com/users/LysandreJik", "html_url": "https://github.com/LysandreJik", "followers_url": "htt...
[ "_The documentation is not available anymore as the PR was closed or merged._" ]
1,700
1,700
1,700
COLLABORATOR
null
# What does this PR do? Deprecate `TransfoXL` as discussed.
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27607/reactions", "total_count": 1, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 1 }
https://api.github.com/repos/huggingface/transformers/issues/27607/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27607", "html_url": "https://github.com/huggingface/transformers/pull/27607", "diff_url": "https://github.com/huggingface/transformers/pull/27607.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27607.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27606
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27606/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27606/comments
https://api.github.com/repos/huggingface/transformers/issues/27606/events
https://github.com/huggingface/transformers/pull/27606
2,002,404,263
PR_kwDOCUB6oc5f68is
27,606
Align backbone stage selection with out_indices & out_features
{ "login": "amyeroberts", "id": 22614925, "node_id": "MDQ6VXNlcjIyNjE0OTI1", "avatar_url": "https://avatars.githubusercontent.com/u/22614925?v=4", "gravatar_id": "", "url": "https://api.github.com/users/amyeroberts", "html_url": "https://github.com/amyeroberts", "followers_url": "https://api.github.com/...
[]
closed
false
null
[]
[ "The docs for this PR live [here](https://moon-ci-docs.huggingface.co/docs/transformers/pr_27606). All of your documentation changes will be reflected on that endpoint.", "> Hey! Could you elaborate on the motivation behind the out of order or duplicated (seems to be a choice rather than a bug fix for me no?)\r\n...
1,700
1,703
1,703
COLLABORATOR
null
# What does this PR do? This PR adds a set of input verification for the `out_features` and `out_indices` arguments for backbones, making sure that any accepted values align with the returned model outputs. ## More details `out_features` and `out_indices` are used to specify which blocks' attentions are retur...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27606/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27606/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27606", "html_url": "https://github.com/huggingface/transformers/pull/27606", "diff_url": "https://github.com/huggingface/transformers/pull/27606.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27606.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27605
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27605/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27605/comments
https://api.github.com/repos/huggingface/transformers/issues/27605/events
https://github.com/huggingface/transformers/issues/27605
2,002,369,380
I_kwDOCUB6oc53Wbtk
27,605
Wav2Vec2ForCTC architecture models load incorrectly with torch 2.1 and later
{ "login": "eindenbom", "id": 45334274, "node_id": "MDQ6VXNlcjQ1MzM0Mjc0", "avatar_url": "https://avatars.githubusercontent.com/u/45334274?v=4", "gravatar_id": "", "url": "https://api.github.com/users/eindenbom", "html_url": "https://github.com/eindenbom", "followers_url": "https://api.github.com/users/...
[]
closed
false
null
[]
[ "rip. Not sure if there's a way to fix this without updating transformers. Have you filed a bug there?", "This is transformers repo and PR breaking backward compatibility has landed here. Where should I file a bug then?\r\n\r\nThe major problem is that transformers itself lack ANY provision for backward compatibi...
1,700
1,704
1,704
NONE
null
### System Info - `transformers` version: 4.35.2 - Platform: Windows-10-10.0.22621-SP0 - Python version: 3.8.0 - Huggingface_hub version: 0.19.4 - Safetensors version: 0.4.0 - Accelerate version: not installed - Accelerate config: not found - PyTorch version (GPU?): 2.1.1 (False) - Tensorflow version (GPU?): 2...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27605/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27605/timeline
completed
null
null
https://api.github.com/repos/huggingface/transformers/issues/27604
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27604/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27604/comments
https://api.github.com/repos/huggingface/transformers/issues/27604/events
https://github.com/huggingface/transformers/issues/27604
2,002,335,062
I_kwDOCUB6oc53WTVW
27,604
Whisper language not returned when return_timestamps="word" and return_language=True
{ "login": "Oscaarjs", "id": 37636054, "node_id": "MDQ6VXNlcjM3NjM2MDU0", "avatar_url": "https://avatars.githubusercontent.com/u/37636054?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Oscaarjs", "html_url": "https://github.com/Oscaarjs", "followers_url": "https://api.github.com/users/Osc...
[]
closed
false
null
[]
[ "This more a feature request than a bug, and is probably related to the fact that the language is not detected for the small chunk but for the full 30sec chunk of audio. + repeating the language does not seem to make sense for me, but we should also de-limit the start and end of detected language, which would bloat...
1,700
1,703
1,703
NONE
null
### System Info - `transformers` version: 4.35.2 - Platform: Linux-5.15.120+-x86_64-with-glibc2.35 - Python version: 3.10.12 - Huggingface_hub version: 0.19.3 - Safetensors version: 0.4.0 - Accelerate version: 0.24.1 - Accelerate config: not found - PyTorch version (GPU?): 2.1.0+cu118 (False) - Tensorflow ver...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27604/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27604/timeline
completed
null
null
https://api.github.com/repos/huggingface/transformers/issues/27603
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27603/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27603/comments
https://api.github.com/repos/huggingface/transformers/issues/27603/events
https://github.com/huggingface/transformers/pull/27603
2,002,299,664
PR_kwDOCUB6oc5f6lrR
27,603
Report to none for reduce_lr_on_plateu
{ "login": "muellerzr", "id": 7831895, "node_id": "MDQ6VXNlcjc4MzE4OTU=", "avatar_url": "https://avatars.githubusercontent.com/u/7831895?v=4", "gravatar_id": "", "url": "https://api.github.com/users/muellerzr", "html_url": "https://github.com/muellerzr", "followers_url": "https://api.github.com/users/mu...
[]
closed
false
null
[]
[ "_The documentation is not available anymore as the PR was closed or merged._" ]
1,700
1,700
1,700
CONTRIBUTOR
null
# What does this PR do? Modifies test to explicitly use `report_to="none"` rather than the default "all", as if trackers are installed it can lead to issues (https://github.com/huggingface/transformers/pull/27352#issuecomment-1819131456) and is a current failing test in Accelerate. ## Before submitting - [ ] T...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27603/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27603/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27603", "html_url": "https://github.com/huggingface/transformers/pull/27603", "diff_url": "https://github.com/huggingface/transformers/pull/27603.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27603.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27602
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27602/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27602/comments
https://api.github.com/repos/huggingface/transformers/issues/27602/events
https://github.com/huggingface/transformers/issues/27602
2,002,280,958
I_kwDOCUB6oc53WGH-
27,602
Converting facebook-encodec to onnx fails with KeyError: 'encodec'
{ "login": "kalradivyanshu", "id": 12642750, "node_id": "MDQ6VXNlcjEyNjQyNzUw", "avatar_url": "https://avatars.githubusercontent.com/u/12642750?v=4", "gravatar_id": "", "url": "https://api.github.com/users/kalradivyanshu", "html_url": "https://github.com/kalradivyanshu", "followers_url": "https://api.gi...
[ { "id": 2392046359, "node_id": "MDU6TGFiZWwyMzkyMDQ2MzU5", "url": "https://api.github.com/repos/huggingface/transformers/labels/Good%20Second%20Issue", "name": "Good Second Issue", "color": "dd935a", "default": false, "description": "Issues that are more difficult to do than \"Good First...
closed
false
null
[]
[ "Hey! Thanks for this, I'll add this as a feature request as the error I am getting is the following:\r\n```python \r\nFramework not requested. Using torch to export to ONNX.\r\n/Users/arthurzucker/.pyenv/versions/py310/lib/python3.10/site-packages/torch/nn/utils/weight_norm.py:30: UserWarning: torch.nn.utils.weigh...
1,700
1,704
1,704
NONE
null
### System Info - `transformers` version: 4.36.0.dev0 - Platform: Linux-5.15.133.1-microsoft-standard-WSL2-x86_64-with-glibc2.29 - Python version: 3.8.10 - Huggingface_hub version: 0.19.4 - Safetensors version: 0.4.0 - Accelerate version: 0.24.1 - Accelerate config: not found - PyTorch version (GPU?): 2.0....
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27602/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27602/timeline
completed
null
null
https://api.github.com/repos/huggingface/transformers/issues/27601
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27601/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27601/comments
https://api.github.com/repos/huggingface/transformers/issues/27601/events
https://github.com/huggingface/transformers/issues/27601
2,002,273,073
I_kwDOCUB6oc53WEMx
27,601
Getting equivalent results between Transformer's resize and tf.image.resize
{ "login": "NielsRogge", "id": 48327001, "node_id": "MDQ6VXNlcjQ4MzI3MDAx", "avatar_url": "https://avatars.githubusercontent.com/u/48327001?v=4", "gravatar_id": "", "url": "https://api.github.com/users/NielsRogge", "html_url": "https://github.com/NielsRogge", "followers_url": "https://api.github.com/use...
[ { "id": 2648621985, "node_id": "MDU6TGFiZWwyNjQ4NjIxOTg1", "url": "https://api.github.com/repos/huggingface/transformers/labels/Feature%20request", "name": "Feature request", "color": "FBCA04", "default": false, "description": "Request for a new feature" } ]
open
false
null
[]
[ "cc @amyeroberts " ]
1,700
1,700
null
CONTRIBUTOR
null
### Feature request For the SigLIP model (#26522), I'd like to get equivalent results between tf.image.resize and the resize method available in Transformers. Here's what I tried: ``` from PIL import Image import requests import tensorflow as tf import numpy as np def resize(image, size, method="bili...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27601/reactions", "total_count": 1, "+1": 1, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27601/timeline
null
null
null
https://api.github.com/repos/huggingface/transformers/issues/27600
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27600/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27600/comments
https://api.github.com/repos/huggingface/transformers/issues/27600/events
https://github.com/huggingface/transformers/issues/27600
2,002,192,230
I_kwDOCUB6oc53Vwdm
27,600
How to get input sentence embedding from Llama or Llama2?
{ "login": "waterluck", "id": 111731547, "node_id": "U_kgDOBqjjWw", "avatar_url": "https://avatars.githubusercontent.com/u/111731547?v=4", "gravatar_id": "", "url": "https://api.github.com/users/waterluck", "html_url": "https://github.com/waterluck", "followers_url": "https://api.github.com/users/waterl...
[]
closed
false
null
[]
[ "Hey @waterluck πŸ‘‹ \r\n\r\nFollowing our [issues guidelines](https://github.com/huggingface/transformers/blob/main/ISSUES.md), we reserve GitHub issues for bugs in the repository and/or feature requests. For any other matters, we'd like to invite you to use our [forum](https://discuss.huggingface.co/) or our [disco...
1,700
1,700
1,700
NONE
null
I'm trying to get the sentence embedding that I input, I checked some common practice to do it, but I'm not sure I'm doing the it right. Who may be help? @gante Thanks if you can be help. my code is as below: ``` model = LlamaForCausalLM.from_pretrained( args.pretrained_name_or_path, torch_dtype=torch...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27600/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27600/timeline
completed
null
null
https://api.github.com/repos/huggingface/transformers/issues/27599
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27599/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27599/comments
https://api.github.com/repos/huggingface/transformers/issues/27599/events
https://github.com/huggingface/transformers/pull/27599
2,001,912,637
PR_kwDOCUB6oc5f5Quc
27,599
Enable safetensors conversion from PyTorch to other frameworks without the torch requirement
{ "login": "LysandreJik", "id": 30755778, "node_id": "MDQ6VXNlcjMwNzU1Nzc4", "avatar_url": "https://avatars.githubusercontent.com/u/30755778?v=4", "gravatar_id": "", "url": "https://api.github.com/users/LysandreJik", "html_url": "https://github.com/LysandreJik", "followers_url": "https://api.github.com/...
[]
closed
false
null
[]
[ "The docs for this PR live [here](https://moon-ci-docs.huggingface.co/docs/transformers/pr_27599). All of your documentation changes will be reflected on that endpoint.", "Should be good for a second look @sanchit-gandhi if you have time for it.", "This issue has been automatically marked as stale because it ha...
1,700
1,706
1,706
MEMBER
null
This removes the need to have `torch` installed to proceed to a safetensors (serialized from PyTorch) checkpoint converted into Flax. This therefore, enables loading any model in Flax that has a safetensors file on the Hub.
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27599/reactions", "total_count": 1, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 1, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27599/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27599", "html_url": "https://github.com/huggingface/transformers/pull/27599", "diff_url": "https://github.com/huggingface/transformers/pull/27599.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27599.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27598
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27598/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27598/comments
https://api.github.com/repos/huggingface/transformers/issues/27598/events
https://github.com/huggingface/transformers/issues/27598
2,001,588,573
I_kwDOCUB6oc53TdFd
27,598
Hyperparameter search error with Ray tune
{ "login": "Shamik-07", "id": 39588365, "node_id": "MDQ6VXNlcjM5NTg4MzY1", "avatar_url": "https://avatars.githubusercontent.com/u/39588365?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Shamik-07", "html_url": "https://github.com/Shamik-07", "followers_url": "https://api.github.com/users/...
[]
closed
false
null
[]
[ "Hello, this PR https://github.com/huggingface/transformers/pull/26499 might fix this issue. Could you please try it out and let us know?", "I tried running the notebook with the PR, however, i found a different error now:\r\n```py\r\n2023-11-20 16:02:53,411\tINFO worker.py:1673 -- Started a local Ray instance.\r...
1,700
1,701
1,701
NONE
null
### System Info Hello Someone, The version of Ray is 2.8.0 and the version of Transformers is 4.35.2 I am trying to run the hyperparameter search for this notebook with ray tune [notebooks/examples/text_classification.ipynb at main Β· huggingface/notebooks (github.com)](https://github.com/huggingface/notebooks/bl...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27598/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27598/timeline
completed
null
null
https://api.github.com/repos/huggingface/transformers/issues/27597
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27597/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27597/comments
https://api.github.com/repos/huggingface/transformers/issues/27597/events
https://github.com/huggingface/transformers/pull/27597
2,001,579,142
PR_kwDOCUB6oc5f4J9Y
27,597
Paged Attention Based the Latest Cache Design
{ "login": "liangan1", "id": 46986936, "node_id": "MDQ6VXNlcjQ2OTg2OTM2", "avatar_url": "https://avatars.githubusercontent.com/u/46986936?v=4", "gravatar_id": "", "url": "https://api.github.com/users/liangan1", "html_url": "https://github.com/liangan1", "followers_url": "https://api.github.com/users/lia...
[ { "id": 2796628563, "node_id": "MDU6TGFiZWwyNzk2NjI4NTYz", "url": "https://api.github.com/repos/huggingface/transformers/labels/WIP", "name": "WIP", "color": "234C99", "default": false, "description": "Label your PR/Issue with WIP for some long outstanding Issues/PRs that are work in pro...
open
false
null
[]
[ "This should wait until the cache refactor is finished cc @gante: https://github.com/huggingface/transformers/pull/27407", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that iss...
1,700
1,708
null
NONE
null
# What does this PR do? Based on the latest cache design on [#PR26681](https://github.com/huggingface/transformers/pull/26681), This PR implements the Paged Attention KV cache which is proposed by this [paper](https://arxiv.org/pdf/2309.06180.pdf). Fixes # ([issue#27303](https://github.com/huggingface/transforme...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27597/reactions", "total_count": 1, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 1, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27597/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27597", "html_url": "https://github.com/huggingface/transformers/pull/27597", "diff_url": "https://github.com/huggingface/transformers/pull/27597.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27597.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27596
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27596/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27596/comments
https://api.github.com/repos/huggingface/transformers/issues/27596/events
https://github.com/huggingface/transformers/issues/27596
2,001,267,031
I_kwDOCUB6oc53SOlX
27,596
An error occurred when using the model.gradient_checkpointing_enable() feature.
{ "login": "CaC033", "id": 34328295, "node_id": "MDQ6VXNlcjM0MzI4Mjk1", "avatar_url": "https://avatars.githubusercontent.com/u/34328295?v=4", "gravatar_id": "", "url": "https://api.github.com/users/CaC033", "html_url": "https://github.com/CaC033", "followers_url": "https://api.github.com/users/CaC033/fo...
[]
closed
false
null
[]
[ "the same issues with transformers version: 4.35.2 when load baichuan2-13b-chat model .", "cc @younesbelkada who worked on this recently πŸ˜‰ ", "Hi @CaC033 @rangehow \r\n\r\nhttps://github.com/huggingface/transformers/pull/27610 should fix the issue. However, note that with respect to the new refactor of gradien...
1,700
1,700
1,700
NONE
null
### System Info - `transformers` version: 4.35.2 - Platform: Linux-4.19.91-014.kangaroo.alios7.x86_64-x86_64-with-glibc2.29 - Python version: 3.8.10 - Huggingface_hub version: 0.19.4 - Safetensors version: 0.4.0 - Accelerate version: 0.24.1 - Accelerate config: not found - PyTorch version (GPU?): 1.14.0a0+41...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27596/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27596/timeline
completed
null
null
https://api.github.com/repos/huggingface/transformers/issues/27595
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27595/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27595/comments
https://api.github.com/repos/huggingface/transformers/issues/27595/events
https://github.com/huggingface/transformers/pull/27595
2,001,184,462
PR_kwDOCUB6oc5f20VU
27,595
Fixed passing scheduler-specific kwargs via TrainingArguments lr_scheduler_kwargs
{ "login": "CharbelAD", "id": 45701489, "node_id": "MDQ6VXNlcjQ1NzAxNDg5", "avatar_url": "https://avatars.githubusercontent.com/u/45701489?v=4", "gravatar_id": "", "url": "https://api.github.com/users/CharbelAD", "html_url": "https://github.com/CharbelAD", "followers_url": "https://api.github.com/users/...
[]
closed
false
null
[]
[ "> Thanks LGTM! Would mind integrating the small snippet to the testing suite? πŸ˜‰\r\n\r\nSure thing! However, I notice that the `reduce_lr_on_plateau` scheduler is not set up to accept kwargs (see line 362 in `src/transformers/optimization.py`).\r\n\r\nShould I change that and allow arguments to be passed to this s...
1,700
1,701
1,701
CONTRIBUTOR
null
# What does this PR do? This PR fixes the `lr_scheduler_kwargs` argument of TrainingArguments being passed incorrectly to `get_scheduler()` in `src/transformers/optimization.py` which raises a TypeError. Minimum code to reproduce the issue (code snippets taken from `test_trainer.py`): ``` import numpy as np ...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27595/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27595/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27595", "html_url": "https://github.com/huggingface/transformers/pull/27595", "diff_url": "https://github.com/huggingface/transformers/pull/27595.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27595.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27594
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27594/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27594/comments
https://api.github.com/repos/huggingface/transformers/issues/27594/events
https://github.com/huggingface/transformers/issues/27594
2,001,021,392
I_kwDOCUB6oc53RSnQ
27,594
prompt_ids are included in distil-whisper response when using ASR pipeline()
{ "login": "benniekiss", "id": 63211101, "node_id": "MDQ6VXNlcjYzMjExMTAx", "avatar_url": "https://avatars.githubusercontent.com/u/63211101?v=4", "gravatar_id": "", "url": "https://api.github.com/users/benniekiss", "html_url": "https://github.com/benniekiss", "followers_url": "https://api.github.com/use...
[]
open
false
null
[]
[ "I believe this part of the code, [tokenization_whisper.py#L853](https://github.com/huggingface/transformers/blob/dc68a39c8111217683bf49a4912d0c9018bab33d/src/transformers/models/whisper/tokenization_whisper.py#L853), would have to be changed to include an argument containing the prompt and an argument to remove th...
1,700
1,706
null
NONE
null
### System Info - `transformers` version: 4.36.0.dev0 - Platform: macOS-14.1.1-arm64-arm-64bit - Python version: 3.11.6 - Huggingface_hub version: 0.19.4 - Safetensors version: 0.4.0 - Accelerate version: 0.24.1 - Accelerate config: not found - PyTorch version (GPU?): 2.1.0 (False) - Tensorflow version (GPU...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27594/reactions", "total_count": 1, "+1": 1, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27594/timeline
null
null
null
https://api.github.com/repos/huggingface/transformers/issues/27593
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27593/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27593/comments
https://api.github.com/repos/huggingface/transformers/issues/27593/events
https://github.com/huggingface/transformers/pull/27593
2,000,988,045
PR_kwDOCUB6oc5f2K3v
27,593
[JAX] Replace uses of jax.devices("cpu") with jax.local_devices(backend="cpu")
{ "login": "hvaara", "id": 1535968, "node_id": "MDQ6VXNlcjE1MzU5Njg=", "avatar_url": "https://avatars.githubusercontent.com/u/1535968?v=4", "gravatar_id": "", "url": "https://api.github.com/users/hvaara", "html_url": "https://github.com/hvaara", "followers_url": "https://api.github.com/users/hvaara/foll...
[]
closed
false
null
[]
[ "/cc @sanchit-gandhi ", "The docs for this PR live [here](https://moon-ci-docs.huggingface.co/docs/transformers/pr_27593). All of your documentation changes will be reflected on that endpoint." ]
1,700
1,701
1,701
CONTRIBUTOR
null
An upcoming change to JAX will include non-local (addressable) CPU devices in jax.devices() when JAX is used multicontroller-style, where there are multiple Python processes. This change preserves the current behavior by replacing uses of jax.devices("cpu"), which previously only returned local devices, with jax.loc...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27593/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27593/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27593", "html_url": "https://github.com/huggingface/transformers/pull/27593", "diff_url": "https://github.com/huggingface/transformers/pull/27593.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27593.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27592
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27592/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27592/comments
https://api.github.com/repos/huggingface/transformers/issues/27592/events
https://github.com/huggingface/transformers/issues/27592
2,000,981,685
I_kwDOCUB6oc53RI61
27,592
How to always use initial prompt in Whisper?
{ "login": "GanymedeNil", "id": 9687786, "node_id": "MDQ6VXNlcjk2ODc3ODY=", "avatar_url": "https://avatars.githubusercontent.com/u/9687786?v=4", "gravatar_id": "", "url": "https://api.github.com/users/GanymedeNil", "html_url": "https://github.com/GanymedeNil", "followers_url": "https://api.github.com/us...
[]
closed
false
null
[]
[ "Hey πŸ€— thanks for opening an issue! We try to keep the github issues for bugs/feature requests.\r\nCould you ask your question on the [forum](https://discuss.huggingface.co/) instead? I'm sure the community will be of help!\r\n\r\nThanks!", "> Hey πŸ€— thanks for opening an issue! We try to keep the github issues...
1,700
1,700
1,700
NONE
null
I checked this PR (#22496 ) but still can't figure out how to always use the initial prompt. is it possible to provide a use case?
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27592/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27592/timeline
completed
null
null
https://api.github.com/repos/huggingface/transformers/issues/27591
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27591/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27591/comments
https://api.github.com/repos/huggingface/transformers/issues/27591/events
https://github.com/huggingface/transformers/issues/27591
2,000,910,575
I_kwDOCUB6oc53Q3jv
27,591
Seq2Seq trainer cannot generate tokens with deepspeed and Zero3
{ "login": "Lingy12", "id": 54443474, "node_id": "MDQ6VXNlcjU0NDQzNDc0", "avatar_url": "https://avatars.githubusercontent.com/u/54443474?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Lingy12", "html_url": "https://github.com/Lingy12", "followers_url": "https://api.github.com/users/Lingy1...
[]
closed
false
null
[]
[ "I tried to upgrade the transformers and deepspeed version, I think the training loop stuck when evaluate is triggered. ", "Hi @Lingy12 πŸ‘‹ \r\n\r\nI see that the exception is `TypeError: 'NoneType' object is not subscriptable`. Can you confirm:\r\n1. The type of `past_key_values` before the exception?\r\n2. If `p...
1,700
1,707
1,703
NONE
null
### System Info - `transformers` version: 4.31.0 - Platform: Linux-5.4.0-54-generic-x86_64-with-glibc2.27 - Python version: 3.10.11 - Huggingface_hub version: 0.19.4 - Safetensors version: 0.4.0 - Accelerate version: 0.23.0 - Accelerate config: not found - PyTorch version (GPU?): 2.1.1+cu118 (True) - Tensor...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27591/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27591/timeline
completed
null
null
https://api.github.com/repos/huggingface/transformers/issues/27590
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27590/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27590/comments
https://api.github.com/repos/huggingface/transformers/issues/27590/events
https://github.com/huggingface/transformers/pull/27590
2,000,890,567
PR_kwDOCUB6oc5f13CZ
27,590
Add `convert_hf_to_openai.py` script to Whisper documentation resources
{ "login": "zuazo", "id": 1878434, "node_id": "MDQ6VXNlcjE4Nzg0MzQ=", "avatar_url": "https://avatars.githubusercontent.com/u/1878434?v=4", "gravatar_id": "", "url": "https://api.github.com/users/zuazo", "html_url": "https://github.com/zuazo", "followers_url": "https://api.github.com/users/zuazo/follower...
[]
closed
false
null
[]
[]
1,700
1,700
1,700
CONTRIBUTOR
null
# What does this PR do? This PR updates the Whisper model documentation by adding a link to the [`convert_hf_to_openai.py`](https://github.com/zuazo-forks/transformers/blob/convert_hf_to_openai/src/transformers/models/whisper/convert_hf_to_openai.py) script in the Resources section. The script converts Whisper model...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27590/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27590/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27590", "html_url": "https://github.com/huggingface/transformers/pull/27590", "diff_url": "https://github.com/huggingface/transformers/pull/27590.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27590.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27589
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27589/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27589/comments
https://api.github.com/repos/huggingface/transformers/issues/27589/events
https://github.com/huggingface/transformers/pull/27589
2,000,871,677
PR_kwDOCUB6oc5f1zT9
27,589
Fix idx2sym not loaded from pretrained vocab file in Transformer XL
{ "login": "jtang98", "id": 44188317, "node_id": "MDQ6VXNlcjQ0MTg4MzE3", "avatar_url": "https://avatars.githubusercontent.com/u/44188317?v=4", "gravatar_id": "", "url": "https://api.github.com/users/jtang98", "html_url": "https://github.com/jtang98", "followers_url": "https://api.github.com/users/jtang9...
[]
closed
false
null
[]
[]
1,700
1,700
1,700
CONTRIBUTOR
null
# What does this PR do? Fixes [27584](https://github.com/huggingface/transformers/issues/27584). When loading vocab file from a pretrained tokenizer for Transformer XL, although the pickled vocabulary file contains a idx2sym key, it isn't loaded, because it is discarded as the empty list already exists as an attrib...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27589/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27589/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27589", "html_url": "https://github.com/huggingface/transformers/pull/27589", "diff_url": "https://github.com/huggingface/transformers/pull/27589.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27589.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27588
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27588/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27588/comments
https://api.github.com/repos/huggingface/transformers/issues/27588/events
https://github.com/huggingface/transformers/pull/27588
2,000,786,315
PR_kwDOCUB6oc5f1ij6
27,588
translation main-class files to chinese
{ "login": "jiaqiw09", "id": 60021713, "node_id": "MDQ6VXNlcjYwMDIxNzEz", "avatar_url": "https://avatars.githubusercontent.com/u/60021713?v=4", "gravatar_id": "", "url": "https://api.github.com/users/jiaqiw09", "html_url": "https://github.com/jiaqiw09", "followers_url": "https://api.github.com/users/jia...
[]
closed
false
null
[]
[ "@stevhliu @statelesshz \r\n\r\nHi, here is remaining translation work of main-class folder.\r\n\r\nFor these translation, I just keep many title or sub-title in origin format as they are just class attributes.\r\n\r\nYou can double check of these files.\r\n\r\nBest", "The docs for this PR live [here](https://moo...
1,700
1,701
1,701
CONTRIBUTOR
null
# What does this PR do? Part of #26803 ## Before submitting - [x] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case). - [x] Did you read the [contributor guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-reques...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27588/reactions", "total_count": 1, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 1, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27588/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27588", "html_url": "https://github.com/huggingface/transformers/pull/27588", "diff_url": "https://github.com/huggingface/transformers/pull/27588.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27588.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27587
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27587/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27587/comments
https://api.github.com/repos/huggingface/transformers/issues/27587/events
https://github.com/huggingface/transformers/pull/27587
2,000,771,170
PR_kwDOCUB6oc5f1fil
27,587
Fuyu Multi-image interleaved processor
{ "login": "cliangyu", "id": 45140242, "node_id": "MDQ6VXNlcjQ1MTQwMjQy", "avatar_url": "https://avatars.githubusercontent.com/u/45140242?v=4", "gravatar_id": "", "url": "https://api.github.com/users/cliangyu", "html_url": "https://github.com/cliangyu", "followers_url": "https://api.github.com/users/cli...
[]
open
false
null
[]
[ "Hey! Feel free to ping @molbap once this is ready for review and cis are green or me if you need help with the CIs! ", "Hi Seungyoun,\r\n\r\nThanks for working on this. I suggest integrating the changes with mine\r\nfirst.\r\n\r\nThank you!\r\n\r\nBest regards,\r\nLiangyu Chen\r\nMMLab, School of Computer Scienc...
1,700
1,706
null
NONE
null
# What does this PR do? <!-- Congratulations! You've made it this far! You're not quite done yet though. Once merged, your PR is going to appear in the release notes with the title you set, so make sure it's a great title that fully reflects the extent of your awesome contribution. Then, please replace this w...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27587/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27587/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27587", "html_url": "https://github.com/huggingface/transformers/pull/27587", "diff_url": "https://github.com/huggingface/transformers/pull/27587.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27587.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27586
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27586/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27586/comments
https://api.github.com/repos/huggingface/transformers/issues/27586/events
https://github.com/huggingface/transformers/issues/27586
2,000,734,164
I_kwDOCUB6oc53QMfU
27,586
Outputs are not consistent when using DeBERTa for inference
{ "login": "polarispw", "id": 78252964, "node_id": "MDQ6VXNlcjc4MjUyOTY0", "avatar_url": "https://avatars.githubusercontent.com/u/78252964?v=4", "gravatar_id": "", "url": "https://api.github.com/users/polarispw", "html_url": "https://github.com/polarispw", "followers_url": "https://api.github.com/users/...
[]
closed
false
null
[]
[ "Hi, the mismatch seems to be related to the classification layer, namely `cls.predictions.transform.dense.weight`, whose weights are randomly initialized at each loading from the pretrained model. I assume it is expected since it is a pretrained model, hence this layer should depend on fine-tuning.", "#22790 It ...
1,700
1,703
1,703
NONE
null
### System Info - `transformers` version: 4.34.1 - Platform: Windows-10-10.0.22621-SP0 - Python version: 3.8.18 - Huggingface_hub version: 0.17.3 - Safetensors version: 0.4.0 - Accelerate version: 0.24.0 - Accelerate config: not found - PyTorch version (GPU?): 2.1.0+cu121 (True) ### Who can help? @...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27586/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27586/timeline
completed
null
null
https://api.github.com/repos/huggingface/transformers/issues/27585
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27585/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27585/comments
https://api.github.com/repos/huggingface/transformers/issues/27585/events
https://github.com/huggingface/transformers/pull/27585
2,000,719,255
PR_kwDOCUB6oc5f1VN8
27,585
update d_kv'annotation in mt5'configuration
{ "login": "callanwu", "id": 63695429, "node_id": "MDQ6VXNlcjYzNjk1NDI5", "avatar_url": "https://avatars.githubusercontent.com/u/63695429?v=4", "gravatar_id": "", "url": "https://api.github.com/users/callanwu", "html_url": "https://github.com/callanwu", "followers_url": "https://api.github.com/users/cal...
[]
closed
false
null
[]
[ "Thank you for your contribution! Please make sure that the quality checks pass - here's the [contributing guide](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md). It looks like you need to run `make style`.", "> Thank you for your contribution! Please make sure that the quality checks pass ...
1,700
1,700
1,700
CONTRIBUTOR
null
# What does this PR do? <!-- Congratulations! You've made it this far! You're not quite done yet though. Once merged, your PR is going to appear in the release notes with the title you set, so make sure it's a great title that fully reflects the extent of your awesome contribution. Then, please replace this w...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27585/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27585/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27585", "html_url": "https://github.com/huggingface/transformers/pull/27585", "diff_url": "https://github.com/huggingface/transformers/pull/27585.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27585.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27584
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27584/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27584/comments
https://api.github.com/repos/huggingface/transformers/issues/27584/events
https://github.com/huggingface/transformers/issues/27584
2,000,666,059
I_kwDOCUB6oc53P73L
27,584
tokenizer.decode throwing an error for TransfoXLTokenizer
{ "login": "dsplog", "id": 5105387, "node_id": "MDQ6VXNlcjUxMDUzODc=", "avatar_url": "https://avatars.githubusercontent.com/u/5105387?v=4", "gravatar_id": "", "url": "https://api.github.com/users/dsplog", "html_url": "https://github.com/dsplog", "followers_url": "https://api.github.com/users/dsplog/foll...
[]
closed
false
null
[]
[ "Thanks for reporting πŸ˜‰ @jtang98 fixed this πŸ”₯ " ]
1,700
1,700
1,700
NONE
null
### System Info - `transformers` version: 4.34.0 - Platform: Linux-5.15.0-88-generic-x86_64-with-glibc2.17 - Python version: 3.8.18 - Huggingface_hub version: 0.17.3 - Safetensors version: 0.4.0 - Accelerate version: 0.23.0 - Accelerate config: not found - PyTorch version (GPU?): 2.1.0+cu121 (True) - Tensorfl...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27584/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27584/timeline
completed
null
null
https://api.github.com/repos/huggingface/transformers/issues/27583
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27583/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27583/comments
https://api.github.com/repos/huggingface/transformers/issues/27583/events
https://github.com/huggingface/transformers/issues/27583
2,000,660,651
I_kwDOCUB6oc53P6ir
27,583
Whisper is not learning a new tokenizer, even when i make test and train dataset the same
{ "login": "P-Sood", "id": 55671093, "node_id": "MDQ6VXNlcjU1NjcxMDkz", "avatar_url": "https://avatars.githubusercontent.com/u/55671093?v=4", "gravatar_id": "", "url": "https://api.github.com/users/P-Sood", "html_url": "https://github.com/P-Sood", "followers_url": "https://api.github.com/users/P-Sood/fo...
[]
closed
false
null
[]
[ "Hey πŸ€— thanks a lot for opening an issue and using transformers! \r\n\r\nWe try to keep the github issues for bugs/feature requests.\r\nCould you ask your question on the [forum](https://discuss.huggingface.co/) instead? I'm sure the community will be of help!\r\n\r\nOtherwise you should follow the tutorial ressou...
1,700
1,704
1,704
NONE
null
### System Info - `transformers` version: 4.35.2 - Platform: Linux-5.15.120+-x86_64-with-glibc2.35 - Python version: 3.10.12 - Huggingface_hub version: 0.19.3 - Safetensors version: 0.4.0 - Accelerate version: 0.24.1 - Accelerate config: not found - PyTorch version (GPU?): 2.1.0+cu118 (False) - Tensorflow v...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27583/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27583/timeline
completed
null
null
https://api.github.com/repos/huggingface/transformers/issues/27582
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27582/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27582/comments
https://api.github.com/repos/huggingface/transformers/issues/27582/events
https://github.com/huggingface/transformers/pull/27582
2,000,574,051
PR_kwDOCUB6oc5f05RS
27,582
Add options to load CLIP Transformer models
{ "login": "amitkumarj441", "id": 14039450, "node_id": "MDQ6VXNlcjE0MDM5NDUw", "avatar_url": "https://avatars.githubusercontent.com/u/14039450?v=4", "gravatar_id": "", "url": "https://api.github.com/users/amitkumarj441", "html_url": "https://github.com/amitkumarj441", "followers_url": "https://api.githu...
[]
closed
false
null
[]
[ "> Hey! Could you share the motivations behind this? πŸ€—\r\n\r\n@ArthurZucker I needed to have an independent feature extractor using CLIP text model for generating (query) embeddings, though, [CLIPTextModelWithProjection](https://github.com/huggingface/transformers/blob/50726f9ea7afc6113da617f8f4ca1ab264a5e28a/src/...
1,700
1,704
1,704
NONE
null
# What does this PR do? <!-- Congratulations! You've made it this far! You're not quite done yet though. Once merged, your PR is going to appear in the release notes with the title you set, so make sure it's a great title that fully reflects the extent of your awesome contribution. Then, please replace this...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27582/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27582/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27582", "html_url": "https://github.com/huggingface/transformers/pull/27582", "diff_url": "https://github.com/huggingface/transformers/pull/27582.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27582.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27581
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27581/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27581/comments
https://api.github.com/repos/huggingface/transformers/issues/27581/events
https://github.com/huggingface/transformers/pull/27581
2,000,536,606
PR_kwDOCUB6oc5f0xsJ
27,581
[Time series] Add patchtst
{ "login": "kashif", "id": 8100, "node_id": "MDQ6VXNlcjgxMDA=", "avatar_url": "https://avatars.githubusercontent.com/u/8100?v=4", "gravatar_id": "", "url": "https://api.github.com/users/kashif", "html_url": "https://github.com/kashif", "followers_url": "https://api.github.com/users/kashif/followers", ...
[]
closed
false
null
[]
[ "Hi @amyeroberts , can you review the latest commits? We have made all the changes based on your requests. There is a deadline from our side, so we would appreciate your help to make it faster.", "_The documentation is not available anymore as the PR was closed or merged._", "Hi @amyeroberts, we have resolved a...
1,700
1,701
1,701
CONTRIBUTOR
null
# What does this PR do? Adds the patchtst model re-opened from closed #25927
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27581/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27581/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27581", "html_url": "https://github.com/huggingface/transformers/pull/27581", "diff_url": "https://github.com/huggingface/transformers/pull/27581.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27581.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27580
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27580/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27580/comments
https://api.github.com/repos/huggingface/transformers/issues/27580/events
https://github.com/huggingface/transformers/issues/27580
2,000,397,110
I_kwDOCUB6oc53O6M2
27,580
ImportError: tokenizers>=0.11.1,!=0.11.3,<0.14 is required for a normal functioning of this module, but found tokenizers==0.15.0. Try: pip install transformers -U or pip install -e '.[dev]' if you're working with git main Press any key to continue . . .
{ "login": "NiuDaVinci", "id": 39637556, "node_id": "MDQ6VXNlcjM5NjM3NTU2", "avatar_url": "https://avatars.githubusercontent.com/u/39637556?v=4", "gravatar_id": "", "url": "https://api.github.com/users/NiuDaVinci", "html_url": "https://github.com/NiuDaVinci", "followers_url": "https://api.github.com/use...
[]
closed
false
null
[]
[ "Hey! Sorry I'm not really sure I can help you with this. Do you have a code snippet of what you were using? (Other wise this should fix it: `pip install --upgrade transformers`", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be address...
1,700
1,703
1,703
NONE
null
### System Info [DxDiag.txt](https://github.com/huggingface/transformers/files/13400236/DxDiag.txt) ### Who can help? _No response_ ### Information - [ ] The official example scripts - [ ] My own modified scripts ### Tasks - [ ] An officially supported task in the `examples` folder (such as GLUE/SQuAD, .....
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27580/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27580/timeline
completed
null
null
https://api.github.com/repos/huggingface/transformers/issues/27579
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27579/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27579/comments
https://api.github.com/repos/huggingface/transformers/issues/27579/events
https://github.com/huggingface/transformers/pull/27579
2,000,385,964
PR_kwDOCUB6oc5f0UOZ
27,579
Fix broken distilbert url
{ "login": "osanseviero", "id": 7246357, "node_id": "MDQ6VXNlcjcyNDYzNTc=", "avatar_url": "https://avatars.githubusercontent.com/u/7246357?v=4", "gravatar_id": "", "url": "https://api.github.com/users/osanseviero", "html_url": "https://github.com/osanseviero", "followers_url": "https://api.github.com/us...
[]
closed
false
null
[]
[]
1,700
1,700
1,700
MEMBER
null
null
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27579/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27579/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27579", "html_url": "https://github.com/huggingface/transformers/pull/27579", "diff_url": "https://github.com/huggingface/transformers/pull/27579.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27579.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27578
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27578/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27578/comments
https://api.github.com/repos/huggingface/transformers/issues/27578/events
https://github.com/huggingface/transformers/issues/27578
2,000,349,721
I_kwDOCUB6oc53OuoZ
27,578
"Attempted to access the data pointer on an invalid python storage" when saving model in TPU mode (Kaggle)
{ "login": "Zaphat", "id": 93189118, "node_id": "U_kgDOBY3z_g", "avatar_url": "https://avatars.githubusercontent.com/u/93189118?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Zaphat", "html_url": "https://github.com/Zaphat", "followers_url": "https://api.github.com/users/Zaphat/followers"...
[]
closed
false
null
[]
[ "I know you provided a link to the notebook but a minimal reproducer would still be welcomed here! πŸ€— \r\n\r\ncc @LysandreJik this might be related to the latest changes? Do you want to have a look? ", "Would be eager to hear your thoughts on it @Narsil ", "Hellow\r\nI'm facing the same issue in Kaggle TPUs too...
1,700
1,704
1,704
NONE
null
### System Info It keeps happening whenever I try to use TPU mode to fine-tune BERT model for sentiment analysis. Everything works fine in GPU mode. I even tried to downgrade/upgrade TensorFlow & safetensors, but it didn't work either. Can you give me any suggestion? Link to that notebook: https://www.kaggle.com/co...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27578/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27578/timeline
completed
null
null
https://api.github.com/repos/huggingface/transformers/issues/27577
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27577/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27577/comments
https://api.github.com/repos/huggingface/transformers/issues/27577/events
https://github.com/huggingface/transformers/issues/27577
2,000,297,662
I_kwDOCUB6oc53Oh6-
27,577
Co-Pilot Store in Huggingface
{ "login": "vikas94", "id": 20750960, "node_id": "MDQ6VXNlcjIwNzUwOTYw", "avatar_url": "https://avatars.githubusercontent.com/u/20750960?v=4", "gravatar_id": "", "url": "https://api.github.com/users/vikas94", "html_url": "https://github.com/vikas94", "followers_url": "https://api.github.com/users/vikas9...
[ { "id": 2648621985, "node_id": "MDU6TGFiZWwyNjQ4NjIxOTg1", "url": "https://api.github.com/repos/huggingface/transformers/labels/Feature%20request", "name": "Feature request", "color": "FBCA04", "default": false, "description": "Request for a new feature" } ]
open
false
null
[]
[ "Would love to see what the community can do! πŸ€— " ]
1,700
1,700
null
NONE
null
### Feature request I think its high time we have one open source Co-pilot store to compete with the GPT Store, where we can setup, train, and add our custom documents, tools to any open source custom model and have a option to share it with the world or keep it private with the users. ### Motivation The motivation ...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27577/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27577/timeline
null
null
null
https://api.github.com/repos/huggingface/transformers/issues/27576
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27576/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27576/comments
https://api.github.com/repos/huggingface/transformers/issues/27576/events
https://github.com/huggingface/transformers/pull/27576
2,000,272,459
PR_kwDOCUB6oc5fz9GG
27,576
[WIP] Generate: implement prefilling
{ "login": "tom-p-reichel", "id": 43631024, "node_id": "MDQ6VXNlcjQzNjMxMDI0", "avatar_url": "https://avatars.githubusercontent.com/u/43631024?v=4", "gravatar_id": "", "url": "https://api.github.com/users/tom-p-reichel", "html_url": "https://github.com/tom-p-reichel", "followers_url": "https://api.githu...
[]
closed
false
null
[]
[ "Current issues:\r\n- I forgot to handle an edge case where 1 or 0 length `input_ids` are passed in, fixable by refusing to prefill those sequences.\r\n- Contrastive search broke because it works with past_key_values and expects to expand them itself.\r\n- By passing in caches, I think I have changed the output siz...
1,700
1,704
1,704
CONTRIBUTOR
null
# What does this PR do? <!-- Congratulations! You've made it this far! You're not quite done yet though. Once merged, your PR is going to appear in the release notes with the title you set, so make sure it's a great title that fully reflects the extent of your awesome contribution. Then, please replace this w...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27576/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27576/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27576", "html_url": "https://github.com/huggingface/transformers/pull/27576", "diff_url": "https://github.com/huggingface/transformers/pull/27576.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27576.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27575
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27575/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27575/comments
https://api.github.com/repos/huggingface/transformers/issues/27575/events
https://github.com/huggingface/transformers/issues/27575
2,000,120,586
I_kwDOCUB6oc53N2sK
27,575
[docs] Quantization
{ "login": "stevhliu", "id": 59462357, "node_id": "MDQ6VXNlcjU5NDYyMzU3", "avatar_url": "https://avatars.githubusercontent.com/u/59462357?v=4", "gravatar_id": "", "url": "https://api.github.com/users/stevhliu", "html_url": "https://github.com/stevhliu", "followers_url": "https://api.github.com/users/ste...
[]
closed
false
null
[]
[ "Sounds good to me πŸ˜‰ ", "Thanks for the proposal @stevhliu ! That sounds great !" ]
1,700
1,701
1,701
MEMBER
null
I was going through the [Quantization](https://huggingface.co/docs/transformers/main/en/main_classes/quantization) docs and noticed there is a ton of cool content in here that I wouldn't have necessarily discovered if I wasn't reviewing another PR that was translating this page. The current API page feels cluttered...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27575/reactions", "total_count": 2, "+1": 2, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27575/timeline
completed
null
null
https://api.github.com/repos/huggingface/transformers/issues/27574
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27574/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27574/comments
https://api.github.com/repos/huggingface/transformers/issues/27574/events
https://github.com/huggingface/transformers/pull/27574
2,000,021,396
PR_kwDOCUB6oc5fzHGq
27,574
Adding leaky relu in dict ACT2CLS
{ "login": "rafaelpadilla", "id": 31217453, "node_id": "MDQ6VXNlcjMxMjE3NDUz", "avatar_url": "https://avatars.githubusercontent.com/u/31217453?v=4", "gravatar_id": "", "url": "https://api.github.com/users/rafaelpadilla", "html_url": "https://github.com/rafaelpadilla", "followers_url": "https://api.githu...
[]
closed
false
null
[]
[]
1,700
1,700
1,700
CONTRIBUTOR
null
# What does this PR do? Simple thing.... it includes LeakyReLU in the `ACT2CLS`. Noticed that the RTDetr official implementation supports "leaky_relu" ([here](https://github.com/lyuwenyu/RT-DETR/blob/3330eca679a7d7cce16bbb10509099174a2f40bf/rtdetr_pytorch/src/nn/backbone/common.py#L70)), which is not being mapped i...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27574/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27574/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27574", "html_url": "https://github.com/huggingface/transformers/pull/27574", "diff_url": "https://github.com/huggingface/transformers/pull/27574.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27574.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27573
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27573/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27573/comments
https://api.github.com/repos/huggingface/transformers/issues/27573/events
https://github.com/huggingface/transformers/issues/27573
1,999,917,988
I_kwDOCUB6oc53NFOk
27,573
Support mix/complex FSDP wrap policy in Trainer
{ "login": "lchu-ibm", "id": 20955448, "node_id": "MDQ6VXNlcjIwOTU1NDQ4", "avatar_url": "https://avatars.githubusercontent.com/u/20955448?v=4", "gravatar_id": "", "url": "https://api.github.com/users/lchu-ibm", "html_url": "https://github.com/lchu-ibm", "followers_url": "https://api.github.com/users/lch...
[]
closed
false
null
[]
[ "Figured we can leverage `trainer.accelerator.state.fsdp_plugin.auto_wrap_policy()` to override the policy. closing this for now." ]
1,700
1,700
1,700
NONE
null
### Feature request Support complex FSDP wrapping policy in Trainer api. ### Motivation The current [implementation](https://github.com/huggingface/transformers/blob/2fc33ebead50383f7707b17f0e2a178d86347d10/src/transformers/trainer.py#L1387-L1404) seems to assume the wrap policy is either size based or transfo...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27573/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27573/timeline
completed
null
null
https://api.github.com/repos/huggingface/transformers/issues/27572
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27572/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27572/comments
https://api.github.com/repos/huggingface/transformers/issues/27572/events
https://github.com/huggingface/transformers/issues/27572
1,999,792,298
I_kwDOCUB6oc53Mmiq
27,572
Oneformer throws exception for when training for instance segmentation
{ "login": "nickponline", "id": 590151, "node_id": "MDQ6VXNlcjU5MDE1MQ==", "avatar_url": "https://avatars.githubusercontent.com/u/590151?v=4", "gravatar_id": "", "url": "https://api.github.com/users/nickponline", "html_url": "https://github.com/nickponline", "followers_url": "https://api.github.com/user...
[]
closed
false
{ "login": "amyeroberts", "id": 22614925, "node_id": "MDQ6VXNlcjIyNjE0OTI1", "avatar_url": "https://avatars.githubusercontent.com/u/22614925?v=4", "gravatar_id": "", "url": "https://api.github.com/users/amyeroberts", "html_url": "https://github.com/amyeroberts", "followers_url": "https://api.github.com/...
[ { "login": "amyeroberts", "id": 22614925, "node_id": "MDQ6VXNlcjIyNjE0OTI1", "avatar_url": "https://avatars.githubusercontent.com/u/22614925?v=4", "gravatar_id": "", "url": "https://api.github.com/users/amyeroberts", "html_url": "https://github.com/amyeroberts", "followers_url": "htt...
[ "@amyeroberts is the issue potentially: `preprocessor.image_processor.num_text = model.config.num_queries - model.config.text_encoder_n_ctx` line? Does that not work for instance segmentation?\r\n\r\nAdditionally here: https://github.com/NielsRogge/Transformers-Tutorials/issues/370", "It's still an issue, forward...
1,700
1,707
1,707
NONE
null
### System Info Copy-and-paste the text below in your GitHub issue and FILL OUT the two last points. - `transformers` version: 4.35.0.dev0 - Platform: macOS-13.4-arm64-arm-64bit - Python version: 3.10.6 - Huggingface_hub version: 0.16.4 - Safetensors version: 0.3.2 - Accelerate version: not installed - Ac...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27572/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27572/timeline
completed
null
null
https://api.github.com/repos/huggingface/transformers/issues/27571
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27571/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27571/comments
https://api.github.com/repos/huggingface/transformers/issues/27571/events
https://github.com/huggingface/transformers/pull/27571
1,999,768,071
PR_kwDOCUB6oc5fyPCM
27,571
Make using safetensors files automated.
{ "login": "Narsil", "id": 204321, "node_id": "MDQ6VXNlcjIwNDMyMQ==", "avatar_url": "https://avatars.githubusercontent.com/u/204321?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Narsil", "html_url": "https://github.com/Narsil", "followers_url": "https://api.github.com/users/Narsil/follow...
[]
closed
false
null
[]
[ "The docs for this PR live [here](https://moon-ci-docs.huggingface.co/docs/transformers/pr_27571). All of your documentation changes will be reflected on that endpoint.", "It's starting to look good! I've tried it with [Intel/neural-chat-7b-v3-1](https://huggingface.co/Intel/neural-chat-7b-v3-1), a 7b model, and ...
1,700
1,701
1,701
CONTRIBUTOR
null
If `use_safetensors=True` is used, and it doesn't exist: - Don't crash just yet - Lookup for an open PR containing it. - If yes, use that instead - If not, touch the space to convert, wait for conversion to be finished and the PR to be opened - Use that new PR - Profit. # What does this PR do? <!-- Congratulation...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27571/reactions", "total_count": 2, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 2, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27571/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27571", "html_url": "https://github.com/huggingface/transformers/pull/27571", "diff_url": "https://github.com/huggingface/transformers/pull/27571.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27571.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27570
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27570/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27570/comments
https://api.github.com/repos/huggingface/transformers/issues/27570/events
https://github.com/huggingface/transformers/pull/27570
1,999,735,132
PR_kwDOCUB6oc5fyHxV
27,570
Fix torch.fx import issue for torch 1.12
{ "login": "amyeroberts", "id": 22614925, "node_id": "MDQ6VXNlcjIyNjE0OTI1", "avatar_url": "https://avatars.githubusercontent.com/u/22614925?v=4", "gravatar_id": "", "url": "https://api.github.com/users/amyeroberts", "html_url": "https://github.com/amyeroberts", "followers_url": "https://api.github.com/...
[]
closed
false
null
[]
[ "_The documentation is not available anymore as the PR was closed or merged._" ]
1,700
1,700
1,700
COLLABORATOR
null
# What does this PR do? Directly imports `torch.fx` before calling `torch.fx.wrap`. In later versions of pytorch this works, but for torch 1.12 it breaks. Fixes #27534
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27570/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27570/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27570", "html_url": "https://github.com/huggingface/transformers/pull/27570", "diff_url": "https://github.com/huggingface/transformers/pull/27570.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27570.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27569
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27569/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27569/comments
https://api.github.com/repos/huggingface/transformers/issues/27569/events
https://github.com/huggingface/transformers/pull/27569
1,999,707,737
PR_kwDOCUB6oc5fyBzH
27,569
Broken links fixed related to datasets docs
{ "login": "VpkPrasanna", "id": 30804112, "node_id": "MDQ6VXNlcjMwODA0MTEy", "avatar_url": "https://avatars.githubusercontent.com/u/30804112?v=4", "gravatar_id": "", "url": "https://api.github.com/users/VpkPrasanna", "html_url": "https://github.com/VpkPrasanna", "followers_url": "https://api.github.com/...
[]
closed
false
null
[]
[ "The docs for this PR live [here](https://moon-ci-docs.huggingface.co/docs/transformers/pr_27569). All of your documentation changes will be reflected on that endpoint." ]
1,700
1,700
1,700
CONTRIBUTOR
null
# What does this PR do? <!-- Congratulations! You've made it this far! You're not quite done yet though. Once merged, your PR is going to appear in the release notes with the title you set, so make sure it's a great title that fully reflects the extent of your awesome contribution. Then, please replace this w...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27569/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27569/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27569", "html_url": "https://github.com/huggingface/transformers/pull/27569", "diff_url": "https://github.com/huggingface/transformers/pull/27569.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27569.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27568
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27568/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27568/comments
https://api.github.com/repos/huggingface/transformers/issues/27568/events
https://github.com/huggingface/transformers/pull/27568
1,999,695,654
PR_kwDOCUB6oc5fx_JH
27,568
Allow `resume_from_checkpoint` to handle `auto_find_batch_size`
{ "login": "muellerzr", "id": 7831895, "node_id": "MDQ6VXNlcjc4MzE4OTU=", "avatar_url": "https://avatars.githubusercontent.com/u/7831895?v=4", "gravatar_id": "", "url": "https://api.github.com/users/muellerzr", "html_url": "https://github.com/muellerzr", "followers_url": "https://api.github.com/users/mu...
[]
closed
false
null
[]
[ "The docs for this PR live [here](https://moon-ci-docs.huggingface.co/docs/transformers/pr_27568). All of your documentation changes will be reflected on that endpoint.", "@ArthurZucker agreed that it's a bit overkill. Would it be better to create a new file instead (something like `training_metadata.json`) inste...
1,700
1,702
1,702
CONTRIBUTOR
null
# What does this PR do? This PR adds the training batch size as part of the `TrainerState`. We do this because the `TrainerState` can be loaded in on `resume_from_checkpoint`, and so if a user has set `auto_find_batch_size` to be `True`, we can keep what that batch size was in there and load it back in if it were sa...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27568/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27568/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27568", "html_url": "https://github.com/huggingface/transformers/pull/27568", "diff_url": "https://github.com/huggingface/transformers/pull/27568.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27568.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27567
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27567/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27567/comments
https://api.github.com/repos/huggingface/transformers/issues/27567/events
https://github.com/huggingface/transformers/pull/27567
1,999,683,113
PR_kwDOCUB6oc5fx8Y5
27,567
Add patchtst
{ "login": "namctin", "id": 8682412, "node_id": "MDQ6VXNlcjg2ODI0MTI=", "avatar_url": "https://avatars.githubusercontent.com/u/8682412?v=4", "gravatar_id": "", "url": "https://api.github.com/users/namctin", "html_url": "https://github.com/namctin", "followers_url": "https://api.github.com/users/namctin/...
[]
closed
false
null
[]
[]
1,700
1,700
1,700
CONTRIBUTOR
null
# What does this PR do? <!-- Congratulations! You've made it this far! You're not quite done yet though. Once merged, your PR is going to appear in the release notes with the title you set, so make sure it's a great title that fully reflects the extent of your awesome contribution. Then, please replace this w...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27567/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27567/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27567", "html_url": "https://github.com/huggingface/transformers/pull/27567", "diff_url": "https://github.com/huggingface/transformers/pull/27567.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27567.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27566
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27566/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27566/comments
https://api.github.com/repos/huggingface/transformers/issues/27566/events
https://github.com/huggingface/transformers/issues/27566
1,999,638,147
I_kwDOCUB6oc53MA6D
27,566
Tokenizer loading: This breaks quite a few things in a lot of places
{ "login": "DevasiaThomas", "id": 14965729, "node_id": "MDQ6VXNlcjE0OTY1NzI5", "avatar_url": "https://avatars.githubusercontent.com/u/14965729?v=4", "gravatar_id": "", "url": "https://api.github.com/users/DevasiaThomas", "html_url": "https://github.com/DevasiaThomas", "followers_url": "https://api.githu...
[]
closed
false
null
[]
[ "Hey! πŸ€— Thanks for reporting this! \r\nIs there anyway you share the errors you were getting. \r\nthe goal behind `init_kwargs[key] = added_tokens_map.get(init_kwargs[key], init_kwargs[key])` is to make sure that the `special_tokens` passed as initi_kwargs are replace with their `AddedToken` version if it exists. ...
1,700
1,707
1,707
NONE
null
https://github.com/huggingface/transformers/blame/638d49983f36af910934b38771b4e55c835c1774/src/transformers/tokenization_utils_base.py#L2253 I was trying to use [text-generation-webui](https://github.com/oobabooga/text-generation-webui/issues/4370) and all tokenizer loads, using quite a few loaders that depend on tr...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27566/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27566/timeline
completed
null
null
https://api.github.com/repos/huggingface/transformers/issues/27565
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27565/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27565/comments
https://api.github.com/repos/huggingface/transformers/issues/27565/events
https://github.com/huggingface/transformers/issues/27565
1,999,495,513
I_kwDOCUB6oc53LeFZ
27,565
T5 Model on seq2seq task encountered a run time error.
{ "login": "Leonezz", "id": 33564074, "node_id": "MDQ6VXNlcjMzNTY0MDc0", "avatar_url": "https://avatars.githubusercontent.com/u/33564074?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Leonezz", "html_url": "https://github.com/Leonezz", "followers_url": "https://api.github.com/users/Leonez...
[]
closed
false
null
[]
[ "Hey! \r\nI am not entirely sure which code should run the evaluation, you did not provide a full reproducer. \r\nDoes the model use `self.prune_heads`? is `has_relative_attention_bias` properly set? \r\nCould you share a full reproducer with a usable checkpoint and make sure you are properly.\r\nI am also not real...
1,700
1,704
1,704
NONE
null
### System Info - `transformers` version: 4.32.1 - Platform: Linux-5.15.90.1-microsoft-standard-WSL2-x86_64-with-glibc2.35 - Python version: 3.11.5 - Huggingface_hub version: 0.17.3 - Safetensors version: 0.4.0 - Accelerate version: 0.24.1 - Accelerate config: not found - PyTorch version (GPU?): 2.1.0 (Tru...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27565/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27565/timeline
completed
null
null
https://api.github.com/repos/huggingface/transformers/issues/27564
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27564/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27564/comments
https://api.github.com/repos/huggingface/transformers/issues/27564/events
https://github.com/huggingface/transformers/pull/27564
1,999,446,946
PR_kwDOCUB6oc5fxJmw
27,564
Harmonize HF environment variables + other cleaning
{ "login": "Wauplin", "id": 11801849, "node_id": "MDQ6VXNlcjExODAxODQ5", "avatar_url": "https://avatars.githubusercontent.com/u/11801849?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Wauplin", "html_url": "https://github.com/Wauplin", "followers_url": "https://api.github.com/users/Waupli...
[]
closed
false
null
[]
[ "Thank you for the explanation. Just one more nit question, in https://github.com/huggingface/huggingface_hub/pull/1786, there is a table of the re-mapped names, like `HUGGINGFACE_HUB_CACHE` -> `HF_CACHE`. I am wondering why we still have the usage of `constants.HUGGINGFACE_HUB_CACHE`.", "Thanks for the reviews @...
1,700
1,700
1,700
CONTRIBUTOR
null
This PR aims at harmonizing environment variables in the HF ecosystem (following up on https://github.com/huggingface/huggingface_hub/pull/1786). I also took the time to review some integration of `huggingface_hub` into `transformers` (trying to break nothing). The goal is simply to have less duplication in the codebas...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27564/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27564/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27564", "html_url": "https://github.com/huggingface/transformers/pull/27564", "diff_url": "https://github.com/huggingface/transformers/pull/27564.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27564.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27563
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27563/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27563/comments
https://api.github.com/repos/huggingface/transformers/issues/27563/events
https://github.com/huggingface/transformers/pull/27563
1,999,231,248
PR_kwDOCUB6oc5fwbmt
27,563
[WIP] Testing safetensors==0.4.1rc1
{ "login": "Narsil", "id": 204321, "node_id": "MDQ6VXNlcjIwNDMyMQ==", "avatar_url": "https://avatars.githubusercontent.com/u/204321?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Narsil", "html_url": "https://github.com/Narsil", "followers_url": "https://api.github.com/users/Narsil/follow...
[]
closed
false
null
[]
[ "_The documentation is not available anymore as the PR was closed or merged._" ]
1,700
1,700
1,700
CONTRIBUTOR
null
# What does this PR do? <!-- Congratulations! You've made it this far! You're not quite done yet though. Once merged, your PR is going to appear in the release notes with the title you set, so make sure it's a great title that fully reflects the extent of your awesome contribution. Then, please replace this with a d...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27563/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27563/timeline
null
true
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27563", "html_url": "https://github.com/huggingface/transformers/pull/27563", "diff_url": "https://github.com/huggingface/transformers/pull/27563.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27563.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27562
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27562/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27562/comments
https://api.github.com/repos/huggingface/transformers/issues/27562/events
https://github.com/huggingface/transformers/pull/27562
1,999,202,624
PR_kwDOCUB6oc5fwVQE
27,562
Translate configuration.md to chinese
{ "login": "statelesshz", "id": 28150734, "node_id": "MDQ6VXNlcjI4MTUwNzM0", "avatar_url": "https://avatars.githubusercontent.com/u/28150734?v=4", "gravatar_id": "", "url": "https://api.github.com/users/statelesshz", "html_url": "https://github.com/statelesshz", "followers_url": "https://api.github.com/...
[]
closed
false
null
[]
[]
1,700
1,703
1,700
CONTRIBUTOR
null
# What does this PR do? Part of https://github.com/huggingface/transformers/issues/26803 ## Before submitting - [x] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case). - [x] Did you read the [contributor guideline](https://github.com/huggingface/transformers/blob/mai...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27562/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27562/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27562", "html_url": "https://github.com/huggingface/transformers/pull/27562", "diff_url": "https://github.com/huggingface/transformers/pull/27562.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27562.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27561
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27561/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27561/comments
https://api.github.com/repos/huggingface/transformers/issues/27561/events
https://github.com/huggingface/transformers/pull/27561
1,999,171,234
PR_kwDOCUB6oc5fwOe8
27,561
Fix tracing dinov2
{ "login": "amyeroberts", "id": 22614925, "node_id": "MDQ6VXNlcjIyNjE0OTI1", "avatar_url": "https://avatars.githubusercontent.com/u/22614925?v=4", "gravatar_id": "", "url": "https://api.github.com/users/amyeroberts", "html_url": "https://github.com/amyeroberts", "followers_url": "https://api.github.com/...
[]
closed
false
null
[]
[ "_The documentation is not available anymore as the PR was closed or merged._", "@ArthurZucker - nice idea, I'll add it! " ]
1,700
1,700
1,700
COLLABORATOR
null
# What does this PR do? Tracing currently fails for DinoV2 due to an issue when calling torch's `torch._C._nn._upsample_bicubic` function through [nn.functional.interpolate](https://github.com/huggingface/transformers/blob/5330b83bc5637b8e7eafe095c22ef19e21baff2d/src/transformers/models/dinov2/modeling_dinov2.py#L10...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27561/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27561/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27561", "html_url": "https://github.com/huggingface/transformers/pull/27561", "diff_url": "https://github.com/huggingface/transformers/pull/27561.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27561.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27560
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27560/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27560/comments
https://api.github.com/repos/huggingface/transformers/issues/27560/events
https://github.com/huggingface/transformers/pull/27560
1,998,982,277
PR_kwDOCUB6oc5fvkhY
27,560
fixed broken link
{ "login": "VpkPrasanna", "id": 30804112, "node_id": "MDQ6VXNlcjMwODA0MTEy", "avatar_url": "https://avatars.githubusercontent.com/u/30804112?v=4", "gravatar_id": "", "url": "https://api.github.com/users/VpkPrasanna", "html_url": "https://github.com/VpkPrasanna", "followers_url": "https://api.github.com/...
[]
closed
false
null
[]
[]
1,700
1,700
1,700
CONTRIBUTOR
null
# What does this PR do? Fixed the broken link for flatten method of dataset process <!-- Congratulations! You've made it this far! You're not quite done yet though. Once merged, your PR is going to appear in the release notes with the title you set, so make sure it's a great title that fully reflects the exte...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27560/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27560/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27560", "html_url": "https://github.com/huggingface/transformers/pull/27560", "diff_url": "https://github.com/huggingface/transformers/pull/27560.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27560.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27559
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27559/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27559/comments
https://api.github.com/repos/huggingface/transformers/issues/27559/events
https://github.com/huggingface/transformers/issues/27559
1,998,897,540
I_kwDOCUB6oc53JMGE
27,559
Visualbert VQA model inference lower accuracy in validation
{ "login": "guanhdrmq", "id": 81207745, "node_id": "MDQ6VXNlcjgxMjA3NzQ1", "avatar_url": "https://avatars.githubusercontent.com/u/81207745?v=4", "gravatar_id": "", "url": "https://api.github.com/users/guanhdrmq", "html_url": "https://github.com/guanhdrmq", "followers_url": "https://api.github.com/users/...
[]
closed
false
null
[]
[ "Hi @guanhdrmq, thanks for raising an issue! \r\n\r\nThis is a question best placed in our [forums](https://discuss.huggingface.co/). We try to reserve the github issues for feature requests and bug reports.", "Hi @amyeroberts Thanks for letting me know. I reckon the probelem already posts in hugginface.\r\nJust ...
1,700
1,703
1,703
NONE
null
null
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27559/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27559/timeline
completed
null
null
https://api.github.com/repos/huggingface/transformers/issues/27558
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27558/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27558/comments
https://api.github.com/repos/huggingface/transformers/issues/27558/events
https://github.com/huggingface/transformers/pull/27558
1,998,837,109
PR_kwDOCUB6oc5fvD0N
27,558
Generate: update compute transition scores doctest
{ "login": "gante", "id": 12240844, "node_id": "MDQ6VXNlcjEyMjQwODQ0", "avatar_url": "https://avatars.githubusercontent.com/u/12240844?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gante", "html_url": "https://github.com/gante", "followers_url": "https://api.github.com/users/gante/follow...
[]
closed
false
null
[]
[]
1,700
1,700
1,700
MEMBER
null
# What does this PR do? #27351 Corrected a detail in the beam score computation: when applying the length penalty to the generated response, which is active by default, the prompt length should not be included in the penalty computation. This PR corrects the doctest accordingly.
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27558/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27558/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27558", "html_url": "https://github.com/huggingface/transformers/pull/27558", "diff_url": "https://github.com/huggingface/transformers/pull/27558.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27558.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27557
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27557/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27557/comments
https://api.github.com/repos/huggingface/transformers/issues/27557/events
https://github.com/huggingface/transformers/pull/27557
1,998,781,732
PR_kwDOCUB6oc5fu3v1
27,557
Context Free Grammar Constrained Decoding (ebnf interface, compatible with llama-cpp)
{ "login": "Saibo-creator", "id": 53392976, "node_id": "MDQ6VXNlcjUzMzkyOTc2", "avatar_url": "https://avatars.githubusercontent.com/u/53392976?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Saibo-creator", "html_url": "https://github.com/Saibo-creator", "followers_url": "https://api.githu...
[]
open
false
null
[]
[ "I think it is a great idea to be compatible with llama.cpp! ", "Hey @Saibo-creator,\r\n\r\nI tried running `grammar_utils.py` on `json.gbnf`, but I get the following error:\r\n\r\n```bash\r\nTraceback (most recent call last):\r\n File \"/home/user/grammar.py\", line 657, in <module>\r\n state = parse_ebnf(in...
1,700
1,706
null
CONTRIBUTOR
null
# What does this PR do? This PR adds a new feature (Context Free Grammar Constrained Decoding) to the library. There is already one PR(WIP) for this feature( #26520 ), but this one has a different motivation and implementation. This implementation is inspired by and adapted from https://github.com/Shopify/torc...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27557/reactions", "total_count": 6, "+1": 2, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 4, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27557/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27557", "html_url": "https://github.com/huggingface/transformers/pull/27557", "diff_url": "https://github.com/huggingface/transformers/pull/27557.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27557.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27556
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27556/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27556/comments
https://api.github.com/repos/huggingface/transformers/issues/27556/events
https://github.com/huggingface/transformers/issues/27556
1,998,742,385
I_kwDOCUB6oc53ImNx
27,556
[i18n-JP] Translating `en/model_doc` docs to Japanese
{ "login": "rajveer43", "id": 64583161, "node_id": "MDQ6VXNlcjY0NTgzMTYx", "avatar_url": "https://avatars.githubusercontent.com/u/64583161?v=4", "gravatar_id": "", "url": "https://api.github.com/users/rajveer43", "html_url": "https://github.com/rajveer43", "followers_url": "https://api.github.com/users/...
[ { "id": 2796628563, "node_id": "MDU6TGFiZWwyNzk2NjI4NTYz", "url": "https://api.github.com/repos/huggingface/transformers/labels/WIP", "name": "WIP", "color": "234C99", "default": false, "description": "Label your PR/Issue with WIP for some long outstanding Issues/PRs that are work in pro...
closed
false
null
[]
[]
1,700
1,701
1,701
CONTRIBUTOR
null
<!-- Note: Please search to see if an issue already exists for the language you are trying to translate. --> Hi! Let's bring the documentation to all the Japanese-speaking community 🌐 Who would want to translate? Please follow the πŸ€— [TRANSLATING guide](https://github.com/huggingface/transformers/blob/main...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27556/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27556/timeline
completed
null
null
https://api.github.com/repos/huggingface/transformers/issues/27555
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27555/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27555/comments
https://api.github.com/repos/huggingface/transformers/issues/27555/events
https://github.com/huggingface/transformers/pull/27555
1,998,666,597
PR_kwDOCUB6oc5fue-n
27,555
Fix AMD CI not showing GPU
{ "login": "ydshieh", "id": 2521628, "node_id": "MDQ6VXNlcjI1MjE2Mjg=", "avatar_url": "https://avatars.githubusercontent.com/u/2521628?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ydshieh", "html_url": "https://github.com/ydshieh", "followers_url": "https://api.github.com/users/ydshieh/...
[]
closed
false
null
[]
[ "After merging are we able to reenable the tests from #27541 ? ", "> After merging are we able to reenable the tests from #27541 ?\r\n\r\nNo, these 2 are irrelevant. #27541 is about `AMD docker image can't be built` while this one is `some issue at testing time (even if we build the image manually)`" ]
1,700
1,700
1,700
COLLABORATOR
null
# What does this PR do? For AMD CI jobs, like [this one](https://github.com/huggingface/transformers/actions/runs/6879141724/job/18711232327), in the `Environment` section, sometimes it shows `Number of GPUs available: 0`. After investigation from the infra team (thanks to Guillaume), they told me the changes in ...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27555/reactions", "total_count": 1, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 1, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27555/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27555", "html_url": "https://github.com/huggingface/transformers/pull/27555", "diff_url": "https://github.com/huggingface/transformers/pull/27555.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27555.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27554
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27554/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27554/comments
https://api.github.com/repos/huggingface/transformers/issues/27554/events
https://github.com/huggingface/transformers/issues/27554
1,998,580,180
I_kwDOCUB6oc53H-nU
27,554
Bug of check_imports
{ "login": "wqh17101", "id": 26429138, "node_id": "MDQ6VXNlcjI2NDI5MTM4", "avatar_url": "https://avatars.githubusercontent.com/u/26429138?v=4", "gravatar_id": "", "url": "https://api.github.com/users/wqh17101", "html_url": "https://github.com/wqh17101", "followers_url": "https://api.github.com/users/wqh...
[]
closed
false
null
[]
[ "```python\r\nimport re\r\ncontent=\"\"\"\r\ntry:\r\n from .configuration_baichuan import BaichuanConfig\r\n from .generation_utils import build_chat_input, TextIterStreamer\r\nexcept:\r\n from configuration_baichuan import BaichuanConfig\r\n from generation_utils import build_chat_input, TextIterStream...
1,700
1,703
1,703
NONE
null
### System Info - `transformers` version: 4.30.2 - Platform: Linux-4.15.0-213-generic-x86_64-with-glibc2.27 - Python version: 3.9.18 - Huggingface_hub version: 0.17.3 - Safetensors version: 0.4.0 - PyTorch version (GPU?): 2.0.1+cpu (False) - Tensorflow version (GPU?): not installed (NA) - Flax version (CPU?/GPU...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27554/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27554/timeline
completed
null
null
https://api.github.com/repos/huggingface/transformers/issues/27553
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27553/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27553/comments
https://api.github.com/repos/huggingface/transformers/issues/27553/events
https://github.com/huggingface/transformers/pull/27553
1,998,556,602
PR_kwDOCUB6oc5fuG_6
27,553
Skip some fuyu tests
{ "login": "ydshieh", "id": 2521628, "node_id": "MDQ6VXNlcjI1MjE2Mjg=", "avatar_url": "https://avatars.githubusercontent.com/u/2521628?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ydshieh", "html_url": "https://github.com/ydshieh", "followers_url": "https://api.github.com/users/ydshieh/...
[]
closed
false
null
[]
[ "The code assistant model will soon learn thanks == approve πŸ˜„ ", "@ydshieh I hope not! Often I say thanks for the work so far + requests for change πŸ˜† " ]
1,700
1,700
1,700
COLLABORATOR
null
# What does this PR do? The old story of offload tests for tiny models ...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27553/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27553/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27553", "html_url": "https://github.com/huggingface/transformers/pull/27553", "diff_url": "https://github.com/huggingface/transformers/pull/27553.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27553.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27552
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27552/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27552/comments
https://api.github.com/repos/huggingface/transformers/issues/27552/events
https://github.com/huggingface/transformers/issues/27552
1,998,282,371
I_kwDOCUB6oc53G16D
27,552
Flash Attention 2 for audio/musicgen
{ "login": "yoinked-h", "id": 63889420, "node_id": "MDQ6VXNlcjYzODg5NDIw", "avatar_url": "https://avatars.githubusercontent.com/u/63889420?v=4", "gravatar_id": "", "url": "https://api.github.com/users/yoinked-h", "html_url": "https://github.com/yoinked-h", "followers_url": "https://api.github.com/users/...
[ { "id": 2392046359, "node_id": "MDU6TGFiZWwyMzkyMDQ2MzU5", "url": "https://api.github.com/repos/huggingface/transformers/labels/Good%20Second%20Issue", "name": "Good Second Issue", "color": "dd935a", "default": false, "description": "Issues that are more difficult to do than \"Good First...
open
false
null
[]
[ "cc @sanchit-gandhi @ylacombe ", "Would indeed be cool having FA2 support for MusicGen! Since MusicGen copies the attention mechanism from Bart, you can copy the changes from this PR: https://github.com/huggingface/transformers/pull/27203\r\n\r\nWould you like to have a go at this integration @yoinked-h? It shoul...
1,700
1,702
null
CONTRIBUTOR
null
### Feature request Add flash attention 2 to musicgen models (and or audio gen) ### Motivation Musicgen is a *really* large model, and its very hard to run on consumer GPUs, adding flash attention could make the model much lighter and possibly faster. ### Your contribution I can test on both WSL and windows 11 on ...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27552/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27552/timeline
null
null
null
https://api.github.com/repos/huggingface/transformers/issues/27551
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27551/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27551/comments
https://api.github.com/repos/huggingface/transformers/issues/27551/events
https://github.com/huggingface/transformers/issues/27551
1,998,152,353
I_kwDOCUB6oc53GWKh
27,551
Make style failed
{ "login": "jiqing-feng", "id": 107918818, "node_id": "U_kgDOBm614g", "avatar_url": "https://avatars.githubusercontent.com/u/107918818?v=4", "gravatar_id": "", "url": "https://api.github.com/users/jiqing-feng", "html_url": "https://github.com/jiqing-feng", "followers_url": "https://api.github.com/users/...
[]
closed
false
null
[]
[ "Hi @jiqing-feng, thanks for raising this issue! \r\n\r\nA recent PR #27144 updated our formatting logic to use ruff formatter instead of black. Could you try updating ruff and run again? `pip install -U ruff` ", "Had the same issue & updating ruff solved this for me", "This issue has been automatically marked ...
1,700
1,704
1,704
CONTRIBUTOR
null
### System Info No GPU/TPU found, falling back to CPU. (Set TF_CPP_MIN_LOG_LEVEL=0 and rerun for more info.) Copy-and-paste the text below in your GitHub issue and FILL OUT the two last points. - `transformers` version: 4.36.0.dev0 - Platform: Linux-5.16.0-rc8-intel-next-01534-g53cb5f883cf7-x86_64-with-glibc2.2...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27551/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27551/timeline
completed
null
null
https://api.github.com/repos/huggingface/transformers/issues/27550
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27550/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27550/comments
https://api.github.com/repos/huggingface/transformers/issues/27550/events
https://github.com/huggingface/transformers/issues/27550
1,998,104,294
I_kwDOCUB6oc53GKbm
27,550
Allow independent control of logging and progress bars across Trainer methods
{ "login": "harrisonfloam", "id": 130672912, "node_id": "U_kgDOB8npEA", "avatar_url": "https://avatars.githubusercontent.com/u/130672912?v=4", "gravatar_id": "", "url": "https://api.github.com/users/harrisonfloam", "html_url": "https://github.com/harrisonfloam", "followers_url": "https://api.github.com/...
[]
closed
false
null
[]
[ "cc @muellerzr @pacman100 ", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transforme...
1,700
1,703
1,703
NONE
null
### Feature request Goal: Finer control of logging and progress tracking output for Trainer methods. Specifically, I would like to suppress print logging and progress bars during in-training evaluation, but maintain the main training loop progress bar. A good solution would allow independent control of print loggin...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27550/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27550/timeline
completed
null
null
https://api.github.com/repos/huggingface/transformers/issues/27549
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27549/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27549/comments
https://api.github.com/repos/huggingface/transformers/issues/27549/events
https://github.com/huggingface/transformers/pull/27549
1,997,950,743
PR_kwDOCUB6oc5fsFtS
27,549
Adding bbox transformatinos
{ "login": "rafaelpadilla", "id": 31217453, "node_id": "MDQ6VXNlcjMxMjE3NDUz", "avatar_url": "https://avatars.githubusercontent.com/u/31217453?v=4", "gravatar_id": "", "url": "https://api.github.com/users/rafaelpadilla", "html_url": "https://github.com/rafaelpadilla", "followers_url": "https://api.githu...
[]
closed
false
null
[]
[ "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md) a...
1,700
1,703
1,703
CONTRIBUTOR
null
# What does this PR do? This PR implements bounding box transformations among different formats: `XYXY` <-> `XYWH` <-> `XCYCWH` <-> `RELATIVE_XYWH` <-> `RELATIVE_XCYCWH` Object detection models can output boxes in different formats, requesting manual transformations while porting these models. The idea to have a...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27549/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27549/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27549", "html_url": "https://github.com/huggingface/transformers/pull/27549", "diff_url": "https://github.com/huggingface/transformers/pull/27549.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27549.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27548
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27548/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27548/comments
https://api.github.com/repos/huggingface/transformers/issues/27548/events
https://github.com/huggingface/transformers/pull/27548
1,997,924,525
PR_kwDOCUB6oc5fr_81
27,548
Fix mistral generate for long prompt / response
{ "login": "lorabit110", "id": 108375850, "node_id": "U_kgDOBnWvKg", "avatar_url": "https://avatars.githubusercontent.com/u/108375850?v=4", "gravatar_id": "", "url": "https://api.github.com/users/lorabit110", "html_url": "https://github.com/lorabit110", "followers_url": "https://api.github.com/users/lor...
[]
closed
false
null
[]
[ "cc @gante @ArthurZucker ", "Hey @lorabit110 πŸ‘‹ \r\n\r\nI'm not sure if I follow the need for a fix here :) On my end, the test case you added passes:\r\n```py\r\nfrom transformers import AutoModelForCausalLM\r\nimport torch\r\n\r\nEXPECTED_OUTPUT_TOKEN_IDS = [306, 338]\r\n\r\ninput_ids = [1] + [306, 338] * 2048\...
1,700
1,701
1,701
CONTRIBUTOR
null
# What does this PR do? Fix the below issue: When use mistral model to generate texts, if prompt + max_tokens > 4095 and use_cache=True, you would get the below error. ValueError: past key much have a shape of (`batch_size, num_heads, self.config.sliding_window-1, head_dim`), got torch.Size([1, 8, 2989, 128]). ...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27548/reactions", "total_count": 2, "+1": 1, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 1 }
https://api.github.com/repos/huggingface/transformers/issues/27548/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27548", "html_url": "https://github.com/huggingface/transformers/pull/27548", "diff_url": "https://github.com/huggingface/transformers/pull/27548.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27548.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27547
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27547/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27547/comments
https://api.github.com/repos/huggingface/transformers/issues/27547/events
https://github.com/huggingface/transformers/pull/27547
1,997,663,495
PR_kwDOCUB6oc5frGA-
27,547
CLVP Fixes
{ "login": "susnato", "id": 56069179, "node_id": "MDQ6VXNlcjU2MDY5MTc5", "avatar_url": "https://avatars.githubusercontent.com/u/56069179?v=4", "gravatar_id": "", "url": "https://api.github.com/users/susnato", "html_url": "https://github.com/susnato", "followers_url": "https://api.github.com/users/susnat...
[]
closed
false
null
[]
[ "Hi @amyeroberts, @ArthurZucker, I have worked on the comments and added the logic to add `bos` and `eos` tokens according to this [comment](https://github.com/huggingface/transformers/pull/27547#discussion_r1399547566). \r\n\r\nPlease let me know if you are Ok with this change.", "Hi @ArthurZucker, I have pushe...
1,700
1,701
1,701
CONTRIBUTOR
null
# What does this PR do? <!-- Congratulations! You've made it this far! You're not quite done yet though. Once merged, your PR is going to appear in the release notes with the title you set, so make sure it's a great title that fully reflects the extent of your awesome contribution. Then, please replace this w...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27547/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27547/timeline
null
false
{ "url": "https://api.github.com/repos/huggingface/transformers/pulls/27547", "html_url": "https://github.com/huggingface/transformers/pull/27547", "diff_url": "https://github.com/huggingface/transformers/pull/27547.diff", "patch_url": "https://github.com/huggingface/transformers/pull/27547.patch", "merged_at...
https://api.github.com/repos/huggingface/transformers/issues/27546
https://api.github.com/repos/huggingface/transformers
https://api.github.com/repos/huggingface/transformers/issues/27546/labels{/name}
https://api.github.com/repos/huggingface/transformers/issues/27546/comments
https://api.github.com/repos/huggingface/transformers/issues/27546/events
https://github.com/huggingface/transformers/issues/27546
1,997,647,246
I_kwDOCUB6oc53Ea2O
27,546
Error when passing past_key_values back to generate()
{ "login": "dto", "id": 22819, "node_id": "MDQ6VXNlcjIyODE5", "avatar_url": "https://avatars.githubusercontent.com/u/22819?v=4", "gravatar_id": "", "url": "https://api.github.com/users/dto", "html_url": "https://github.com/dto", "followers_url": "https://api.github.com/users/dto/followers", "following...
[]
closed
false
null
[]
[ "cc @gante ", "Hi @dto πŸ‘‹ Thank you for raising this issue!\r\n\r\n`generate` expects the full input prompt to be passed, even if `past_key_values` is passed too. In this case, the input prompt is the full multi-round conversation. In your script, if you replace the `input_ids` input by `torch.cat((marvin_respons...
1,700
1,703
1,703
NONE
null
### System Info - `transformers` version: 4.36.0.dev0 - Platform: Linux-5.15.0-88-generic-x86_64-with-glibc2.35 - Python version: 3.10.13 - Huggingface_hub version: 0.19.3 - Safetensors version: 0.4.0 - Accelerate version: 0.24.0 - Accelerate config: not found - PyTorch version (GPU?): 2.1.0+rocm5.6 (True) - ...
{ "url": "https://api.github.com/repos/huggingface/transformers/issues/27546/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }
https://api.github.com/repos/huggingface/transformers/issues/27546/timeline
completed
null
null