huggingface/transformers

Pull Requests Commits

SunMarc committed 32 days ago

2167b9a4

SunMarc committed 32 days ago

496ed953

fix: typehits for Causal LM models (#42885)

CandiedCode committed 32 days ago

Verified 4c64a8fb

Remove duplicated processor class from config (#42806)

zucchini-nlp committed 32 days ago

Verified b61da251

Add local kernel loading support to KernelConfig(). (#42800)

zheliuyu committed 32 days ago

Verified 24275124

Simplify tie weights logic (#42895)

Cyrilvallez committed 32 days ago

Verified 4d6516e2

fix FastSpeech2ConformerTokenizer crash in tokenize (#42888)

sywangyi committed 32 days ago

Verified 24b311ee

Added kernels from kernel hub for Bamba model (#41540)

romitjain committed 32 days ago

Verified 0f896619

Fix GraniteMoeHybrid in transformers v5 (#42872)

avihu111 committed 32 days ago

Verified 5d2f82b5

[`Ernie 4.5 Moe`] Fix routing, weights, and update expectations (#42653)

vasqu committed 32 days ago

Verified 4e7cecb2

Fix convert_tekken_tokenizer (#42592)

juliendenize committed 32 days ago

Verified 252afd89

Stop collecting all model parameters to save models when using DeepSpeed and LoRA (#41416)

Taise228 committed 32 days ago

Verified 89998bdd

[modular] Fix a weird renaming edge-case (#42844)

Cyrilvallez committed 32 days ago

Verified 8d526c23

[Devstral] Make sure FP8 conversion works correctly (#42715)

patrickvonplaten committed 32 days ago

Verified 7960b5ea

Add missing ModelOutput subclass return type hints (#41219)

tomaarsen committed 32 days ago

Verified 6c7c992f

Fix BLT training_ci overfit test (#42685)

preetam1407 committed 32 days ago

Verified 0f97c688

Add `.on_push_begin()` callback to Trainer and implement for `TrackioCallback` (#42850)

abidlabs committed 32 days ago

Verified 7f52a2a4

[docs] optimizations quickstart (#42538)

stevhliu committed 32 days ago

Verified 31de95ef

Simplify using custom resolution for sam3 and sam3_video inference (#42787)

yonigozlan committed 33 days ago

Verified 23394cc4

fix: Initialize ApertusMLP's xielu activation using `torch_dtype` (#42864)

wasertech committed 33 days ago

Verified 06378d40

Remove null values from fast image processors dict (#42780)

yonigozlan committed 33 days ago

Verified fc50bdc6

Enforce call to `post_init` and fix all of them (#42873)

Cyrilvallez committed 33 days ago

Verified c7aec088

[CB] Easy optimizations for continuous batching (#42839)

remi-or committed 33 days ago

Verified f3d5f255

AYou0207 committed 33 days ago

Verified 298d08dc

Remove tied weights from internal attribute if they are not tied (#42871)

Cyrilvallez committed 33 days ago

Verified a187b857

[docs] Improve contribution guidelines for Quantization (#42870)

MekkCyber committed 33 days ago

Verified 64c12fdf

Fixes 2 failing tests from AMD CI (#42777)

Sai-Suraj-27 committed 33 days ago

Verified f0d9cd1f

Fix speccht5_tts pipeline (#42830)

jiqing-feng committed 33 days ago

Verified 66623a1f

[Fix] Fix FA2 kernels ut (#42803)

YangKai0616 committed 33 days ago

Verified e17b1b85

Fix Gemma (#42847)

Cyrilvallez committed 36 days ago

Verified 40dc11cd

Older