Pull requests: huggingface/transformers
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
enable unit tests to run on third-party devcies other than CUDA and CPU.
#25327
opened Aug 5, 2023 by
statelesshz
•
Draft
Adding more information in help parser on train_file and validation_file
#25324
opened Aug 4, 2023 by
pphuc25
Loading 1�7
add docstring examples to Encoder repetition penalty logits processor
#25317
opened Aug 4, 2023 by
rajveer43
Loading 1�7
5 tasks
Fixed "Dynamic" issue in LlamaDynamicNTKScalingRotaryEmbedding
#25308
opened Aug 4, 2023 by
LetianLee
Loading 1�7
2 of 5 tasks
MaskFormer, Mask2Former - replace einsum for tracing
#25297
opened Aug 3, 2023 by
amyeroberts
Loading 1�7
1 of 5 tasks
Fix Llama's attention map handling for left padding which causes numerical instability and performance drops
#25284
opened Aug 3, 2023 by
Randolph-zeng
Loading 1�7
[
Docs / BetterTransformer ] Added more details about flash attention + SDPA
#25265
opened Aug 2, 2023 by
younesbelkada
Loading 1�7
Allow
trust_remote_code in example scripts
#25248
opened Aug 1, 2023 by
Jackmin801
Loading 1�7
1 of 5 tasks
WIP In assisted decoding, pass model_kwargs to model's forward call (fix prepare_input_for_generation in all models)
#25242
opened Aug 1, 2023 by
sinking-point
Loading 1�7
4 of 5 tasks
Save tokenizer and model config when training with FSDP
#25198
opened Jul 31, 2023 by
J38
Loading 1�7
Previous Next
ProTip!
Updated in the last three days: updated:>2023-08-03.

