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 our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

1,004 Open 14,956 Closed
1,004 Open 14,956 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for 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 3.2, inference error with "repetition_penalty" in generation_config bug
#34304 opened Oct 22, 2024 by ruian1
3 of 4 tasks
1
Pickled custom model with '.' and trust_remote_code=True and set_start_method("spawn") raises ModuleNotFoundError bug
#34303 opened Oct 22, 2024 by rsamf
2 of 4 tasks
Predicted depth map incorrectly rendered as an image bug
#34300 opened Oct 21, 2024 by pkubiak
2 of 4 tasks
Llama model not generating mentioned "max_tokens" number of tokens as output bug
#34296 opened Oct 21, 2024 by vineel96
2 of 4 tasks
1
SiglipVisionEmbeddings doesn't cast pixel_values like CLIPVisionEmbeddings does bug
#34294 opened Oct 21, 2024 by fpgaminer
4 tasks
1
DinoV2 is incorrectly documented as a default patch size of 16 instead of 14 bug
#34292 opened Oct 21, 2024 by OFSkean
1 of 4 tasks
LLaVa with multiple image input throws error: Image features and image tokens do not match bug Multimodal Vision
#34284 opened Oct 21, 2024 by Shruthi42
4 tasks
1
cross attention mask is always zeros in mllama bug Multimodal Vision
#34280 opened Oct 21, 2024 by xgal
1 of 4 tasks
4
[Trainer][Eval] Why the model output for the first element in eval batch is skipped in logits? trainer
#34278 opened Oct 21, 2024 by konradkalita
2
bitnet support Feature request Request for a new feature Quantization
#34277 opened Oct 21, 2024 by Darshan2104
4
Limit number of parametes logged with MLflowCallback Feature request Request for a new feature trainer
#34276 opened Oct 21, 2024 by cecheta
1
1
GenerationConfig parameters are deleted from the website on 4.52.2 and main bug Documentation
#34273 opened Oct 21, 2024 by eyalmazuz
4 tasks
2
1
image_transforms preprocess quite slow when run large image with qwen2vl bug Performance Vision
#34272 opened Oct 21, 2024 by zhjunqin
4 tasks
2
Access to model outputs inside LogitProcessor Feature request Request for a new feature Generation
#34265 opened Oct 20, 2024 by AdityaMayukhSom
1
T5 models fail when loaded with torch_dtype=torch.half bug Core: Modeling Internals of the library; Models. Usage General questions about the library
#34264 opened Oct 19, 2024 by 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 opened Oct 19, 2024 by JianbangZ
4 tasks
1
21
Finetuned LLAMA Model is working same old pretrained model after combining LORA weights with old model lora PEFT Quantization
#34262 opened Oct 19, 2024 by abhi201002
3
Mixtral manual head_dim Feature request Request for a new feature Usage General questions about the library
#34261 opened Oct 19, 2024 by wavy-jung
1
Add support for Janus model from DeepSeek AI New model
#34249 opened Oct 18, 2024 by 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 opened Oct 18, 2024 by 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 opened Oct 18, 2024 by sbwww
2
How to output token by token use transformers? bug Discussion Discussion on a topic (keep it focused or open a new issue though)
#34241 opened Oct 18, 2024 by xuanzhangyang
4 tasks
1
Trainer.hyperparameter_search kwargs parameter has an inexact definition if using Optuna Documentation
#34239 opened Oct 18, 2024 by GuillemGSubies
1
GGUF support for BERT architecture Feature request Request for a new feature
#34238 opened Oct 18, 2024 by Dimmension
1
LlamaRotaryEmbedding inv_freq buffer is left uninitialized by init_empty_weights + load_checkpoint_and_dispatch bug
#34234 opened Oct 18, 2024 by ringohoffman
3 of 4 tasks
ProTip! Add no:assignee to see everything that’s not assigned.