-
Notifications
You must be signed in to change notification settings - Fork 28.5k
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.
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
Author
Label
Projects
Milestones
Assignee
Sort
Issues list
Docs: return type of
get_default_model_and_revision might be incorrectly documented?
#35981
by MarcoGorelli
was closed Feb 13, 2025
Deformable DETR custom kernel fails to compile with PyTorch 2.6
bug
Vision
#35976
by hassonofer
was closed Feb 4, 2025
4 tasks
Document Question Answering Pipeline fails due to array with an inhomogeneous shape
bug
#35948
by tillschander
was closed Jan 30, 2025
3 of 4 tasks
Learning rate logging off by one training step
bug
#35942
by phoerious
was closed Mar 26, 2025
2 of 4 tasks
4.48.1 breaks sliding window in eager attention for qwen2
bug
#35924
by wizyoung
was closed Feb 14, 2025
4 tasks
meta-llama/Llama-3.2-11B-Vision-Instruct, device_map = 'auto', padding ruins _prepare_4d_causal_attention_mask_with_cache_position
bug
#35918
by AndriiZelenko
was closed Mar 7, 2025
2 of 4 tasks
Attention can be None in ModernBertForSequenceClassification
#35917
by ashmikuz
was closed Mar 7, 2025
ForSequenceClassification models assume right-padding
bug
#35909
by Rocketknight1
was closed Feb 5, 2025
Transformers is incompatible with flash attention version 2.7.3
bug
#35899
by pavelgein
was closed Feb 6, 2025
2 of 4 tasks
Qwen2FlashAttention sliding windows are applied to wrong layers
bug
#35896
by hzhua
was closed Feb 14, 2025
Paliegemma Pad Token not Masked
bug
Multimodal
VLM
#35855
by TangsengT
was closed Feb 13, 2025
2 of 4 tasks
forward() got an unexpected keyword argument 'num_items_in_batch'
bug
#35838
by Bachstelze
was closed Feb 6, 2025
2 of 4 tasks
tokenizer_class:
LlamaTokenizerFast becomes LlamaTokenizer after load + immediate save
bug
#35832
by Qubitium
was closed Feb 13, 2025
2 of 4 tasks
ImportError: cannot import name 'NoneType' from 'types' on main in Python 3.9
bug
#35827
by harupy
was closed Feb 10, 2025
1 of 4 tasks
[Feature Request] Support register customize quantization method out-of-tree
Feature request
Request for a new feature
#35814
by ice-tong
was closed Feb 18, 2025
Handle python 3.11 for huggingface["video"] and dependencies
Feature request
Request for a new feature
#35803
by CalOmnie
was closed Jan 21, 2025
Auto-resume from checkpoint throws error if last checkpoint is incomplete
bug
#35782
by SilverSoldier
was closed Feb 6, 2025
2 of 4 tasks
LLaVA-OneVision image features and image tokens mismatch
bug
#35775
by sheryc
was closed Jan 24, 2025
2 of 4 tasks
transformers.image_transforms.normalize documents and checks for the wrong type for std and mean arguments
bug
#35772
by CalOmnie
was closed Jan 20, 2025
1 of 4 tasks
multi-gpu: test_model_parallel_beam_search tests fail with "RuntimeError: Expected all tensors to be on the same device"
#35762
by dvrogozh
was closed Feb 12, 2025
Giving dictionary to Audio Classification Pipeline overwrites dictionary data
bug
#35740
by wilke0818
was closed Jan 17, 2025
2 of 4 tasks
Audio-Classification Pipeline top_k Documentation mismatch and bug (possibly generalizes to any classification pipelines)
bug
#35736
by wilke0818
was closed Feb 5, 2025
2 of 4 tasks
Break point in Transformers V4.48.0 and python 3.9
bug
#35723
by prajeet26
was closed Jan 16, 2025
2 of 4 tasks
Training with Accelerate will crash if using TrainingArguments
bug
#35702
by tiandeyu-cs
was closed Jan 15, 2025
1 of 4 tasks
ProTip!
Mix and match filters to narrow down what you’re looking for.