huggingface/transformers

Pull Requests Commits

better error message

ydshieh committed 357 days ago

5d154537

ydshieh committed 358 days ago

78257cac

ydshieh committed 358 days ago

1212cb5e

ydshieh committed 358 days ago

53e70d9c

Skip `MobileNetV1ModelTest::test_batching_equivalence` for now (#35614)

ydshieh committed 1 year ago

Verified 1e3c6c1f

Fix flaky `test_beam_search_low_memory` (#35611)

ydshieh committed 1 year ago

Verified 04eae987

Let `EarlyStoppingCallback` not require `load_best_model_at_end` (#35101)

muellerzr committed 1 year ago

Verified b02828e4

Added error when sequence length is bigger than max_position_embeddings (#32156)

Taha1506 committed 1 year ago

Verified 0aaf124f

Use inherit tempdir makers for tests + fix failing DS tests (#35600)

muellerzr committed 1 year ago

Verified 1211e616

Fix flaky `test_custom_4d_attention_mask` (#35606)

ydshieh committed 1 year ago

Verified bbc00046

ArthurZucker committed 1 year ago

f63829c8

[WIP] Emu3: add model (#33770)

zucchini-nlp committed 1 year ago

Verified 52e1f87c

Fix flex_attention in training mode (#35605)

Cyrilvallez committed 1 year ago

Verified ccc0381d

Remove `benchmark.py` after #34275

ArthurZucker committed 1 year ago

a9bd1e62

Chat template: return vectorized output in processors (#34275)

zucchini-nlp committed 1 year ago

Verified e0646f3d

Add Moonshine (#34784)

eustlb committed 1 year ago

Verified 5f087d13

Skip `torchscript` tests if a cache object is in model's outputs (#35596)

ydshieh committed 1 year ago

Verified 6f127d3f

ModernBert: reuse GemmaRotaryEmbedding via modular + Integration tests (#35459)

tomaarsen committed 1 year ago

Verified 6b73ee89

Add flex_attn to diffllama (#35601)

muellerzr committed 1 year ago

Verified 8de7b1ba

ModernBERT bug fixes (#35404)

warner-benjamin committed 1 year ago

Verified 1e3ddcb2

add `_supports_flex_attn = True` for models that do support it (#35598)

ArthurZucker committed 1 year ago

Verified e97d7a5b

[doc] deepspeed universal checkpoint (#35015)

huyiwen committed 1 year ago

Verified c9c682d1

Refactor/fix Cohere2 (#35594)

Cyrilvallez committed 1 year ago

Verified 3a4ae6ea

[`tokenizers`] Ensure that add_prefix_space is propagated to backend_tokenizer.pre_tokenizer (#35593)

tomaarsen committed 1 year ago

Verified 32e0db8a

Fix modular edge case + modular sorting order (#35562)

Cyrilvallez committed 1 year ago

Verified 46276f9a

PR for Issue #22694: Fixed Training Evaluation table display for VSCode (#35557)

dedsec995 committed 1 year ago

Verified d3fe9fa3

Small fix rope kwargs (#35589)

molbap committed 1 year ago

Verified 395b114b

Fix flaky `SwitchTransformersModelTest::test_training_gradient` (#35587)

ydshieh committed 1 year ago

Verified 82dd6c14

`tokenizer` train from iterator without pre_tokenizers (#35396)

ArthurZucker committed 1 year ago

Verified eb4579cf

feat: add TP plan for granite (#35573)

kmehant committed 1 year ago

Verified 320512df

Older