Skip to content

Issues: huggingface/transformers

New issue

Have a question about this project?Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to ourterms of serviceand privacy statement.We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

1,002 Open 14,954 Closed
1,002 Open 14,954 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Usealt+click/returnto exclude labels
or⇧+click/returnfor logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

Llama model not generating mentioned "max_tokens" number of tokens as output bug
#34296 openedOct 21, 2024by vineel96
2 of 4 tasks
SiglipVisionEmbeddings doesn't cast pixel_values like CLIPVisionEmbeddings does bug
#34294 openedOct 21, 2024by fpgaminer
4 tasks
DinoV2 is incorrectly documented as a default patch size of 16 instead of 14 bug
#34292 openedOct 21, 2024by OFSkean
1 of 4 tasks
LLaVa with multiple image input throws error: Image features and image tokens do not match bug Multimodal Vision
#34284 openedOct 21, 2024by Shruthi42
4 tasks
1
cross attention mask is always zeros in mllama bug Multimodal Vision
#34280 openedOct 21, 2024by xgal
1 of 4 tasks
3
[Trainer][Eval] Why the model output for the first element in eval batch is skipped in logits? trainer
#34278 openedOct 21, 2024by konradkalita
2
bitnet support Feature request Request for a new feature Quantization
#34277 openedOct 21, 2024by Darshan2104
4
Limit number of parametes logged withMLflowCallback Feature request Request for a new feature trainer
#34276 openedOct 21, 2024by cecheta
1
1
GenerationConfig parameters are deleted from the website on 4.52.2 and main bug Documentation
#34273 openedOct 21, 2024by eyalmazuz
4 tasks
2
1
image_transforms preprocess quite slow when run large image with qwen2vl bug Performance Vision
#34272 openedOct 21, 2024by zhjunqin
4 tasks
2
Access to model outputs inside LogitProcessor Feature request Request for a new feature Generation
#34265 openedOct 20, 2024by AdityaMayukhSom
1
T5 models fail when loaded withtorch_dtype=torch.half bug Core: Modeling Internals of the library; Models. Usage General questions about the library
#34264 openedOct 19, 2024by Rohan138
2 of 4 tasks
2
New GA fix causes training loss multiple times higher across the board (5x to 10x higher) bug trainer
#34263 openedOct 19, 2024by JianbangZ
4 tasks
1
19
Finetuned LLAMA Model is working same old pretrained model after combining LORA weights with old model lora PEFT Quantization
#34262 openedOct 19, 2024by abhi201002
3
Mixtral manualhead_dim Feature request Request for a new feature Usage General questions about the library
#34261 openedOct 19, 2024by wavy-jung
1
Add support for Janus model from DeepSeek AI New model
#34249 openedOct 18, 2024by ighoshsubho
2 tasks done
RuntimeError: Failed to import transformers.pipelines because of the following error (look up to see its traceback): dlopen: cannot load any more object with static TLS bug Installation Usage General questions about the library
#34243 openedOct 18, 2024by liuzhao104
4 tasks
1
Add DDP token averaging for equivalent non-parallel training similar to #34191 Discussion Discussion on a topic (keep it focused or open a new issue though) Feature request Request for a new feature
#34242 openedOct 18, 2024by sbwww
1
How to output token by token use transformers? bug Discussion Discussion on a topic (keep it focused or open a new issue though)
#34241 openedOct 18, 2024by xuanzhangyang
4 tasks
1
Trainer.hyperparameter_searchkwargsparameter has an inexact definition if using Optuna Documentation
#34239 openedOct 18, 2024by GuillemGSubies
1
GGUF support for BERT architecture Feature request Request for a new feature
#34238 openedOct 18, 2024by Dimmension
1
LlamaRotaryEmbeddinginv_freqbuffer is left uninitialized byinit_empty_weights+load_checkpoint_and_dispatch bug
#34234 openedOct 18, 2024by ringohoffman
3 of 4 tasks
cache wrong code bug
#34232 openedOct 18, 2024by mdy666
4 tasks
10
Add support for HuBERT batch norm instead of weight norm in pos_conv_emb Feature request Request for a new feature
#34229 openedOct 17, 2024by gallilmaimon
1
ISSUES in #TRANSLATING.md
#34227 openedOct 17, 2024by anshumangahlot
ProTip! Typegion any issue or pull request to go back to the issue listing page.