-
Notifications
You must be signed in to change notification settings - Fork 27.5k
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
FA2 broken for Cohere2 if Optional
Mask
is not passed in forward
bug
#35547
opened Jan 7, 2025 by
Qubitium
2 of 4 tasks
AttributeError: 'Config' object has no attribute '_get_non_default_generation_parameters'
bug
#35543
opened Jan 7, 2025 by
XYZ-916
4 tasks
OverflowError: out of range integral type conversion attempted
bug
#35540
opened Jan 7, 2025 by
test3211234
4 tasks
is possible convert transforms tokenizers in sentence piece .model?
#35538
opened Jan 6, 2025 by
Caio-lima-santos
Mask2FormerImageProcessor support overlapping features
bug
#35536
opened Jan 6, 2025 by
mherzog01
2 of 4 tasks
RagTokenizer Missing patch_token_id, patch_token, and encode Functionality
Feature request
Request for a new feature
#35532
opened Jan 6, 2025 by
hanshengzhu0001
Trainer: update Request for a new feature
state.num_input_tokens_seen
to use num_items_in_batch
Feature request
#35529
opened Jan 6, 2025 by
avishaiElmakies
Trainer: Use second last checkpoint if last checkpoint loading fails
Feature request
Request for a new feature
#35525
opened Jan 6, 2025 by
SilverSoldier
Warning 'The attention mask is not set'
bug
#35524
opened Jan 6, 2025 by
AlessandroSpallina
2 of 4 tasks
How about adding a combined step and epoch feature to save_strategy?
Feature request
Request for a new feature
#35523
opened Jan 6, 2025 by
jp1924
Very slow to load deep seekv3 int4 model and device_map="auto" "sequential" bug
bug
#35522
opened Jan 6, 2025 by
wenhuach21
Batch size deprecation warning issued even when it is not used
bug
#35518
opened Jan 6, 2025 by
quintenroets
2 of 4 tasks
Perhaps your features (
videos
in this case) have excessive nesting (inputs type list
where type int
is expected).
bug
#35512
opened Jan 5, 2025 by
kunling-cxk
2 of 4 tasks
When gradient checkpointing is enabled, flash_attn_kwargs cannot be passed into the decoder_layer
bug
#35509
opened Jan 4, 2025 by
efsotr
4 tasks
Reload Transformers imports
Feature request
Request for a new feature
#35508
opened Jan 4, 2025 by
KareemMusleh
Memory Access out of bounds in mra/cuda_kernel.cu::index_max_cuda_kernel()
bug
#35507
opened Jan 4, 2025 by
dingfen
4 tasks
Wav2Vec2BertForSequenceClassification. return_attention_mask work wrong
bug
#35495
opened Jan 3, 2025 by
HERIUN
2 of 4 tasks
Loss.. should be specified as either training loss or validation loss
bug
#35494
opened Jan 3, 2025 by
FlogramMatt
4 tasks
Instructions to raise PR for addition of shared library files(.so) and .cpp files
Feature request
Request for a new feature
#35492
opened Jan 2, 2025 by
vineel96
Possible bug when using cosine lr scheduler with gradient accumulation
bug
#35484
opened Jan 2, 2025 by
hanyin88
4 tasks
Previous Next
ProTip!
no:milestone will show everything without a milestone.