NielsRogge/transformers

Pull Requests Commits

[Whisper] Computing features on GPU in batch mode for whisper feature extractor. (#29900)

vaibhavagg303 committed 1 year ago

Verified 1ed93be4

doc: Correct spelling mistake (#30107)

caiyili committed 1 year ago

Verified 1fc34aa6

Fix whisper kwargs and generation config (#30018)

zucchini-nlp committed 1 year ago

Verified 76fa17c1

Fix auto tests (#30067)

ydshieh committed 1 year ago

Verified 9b5a6450

Add docstrings and types for MambaCache (#30023)

koayon committed 1 year ago

Verified d9fa13ce

Refactor daily CI workflow (#30012)

ydshieh committed 1 year ago

Verified b17b54d3

Fix `torch.fx` symbolic tracing for LLama (#30047)

michaelbenayoun committed 1 year ago

Verified 17cd7a9d

[test fetcher] Always include the directly related test files (#30050)

ydshieh committed 1 year ago

Verified 48795317

Update quantizer_bnb_4bit.py: In the ValueError string there should be "....you need to set `llm_int8_enable_fp32_cpu_offload=True`...." instead of "`load_in_8bit_fp32_cpu_offload=True`". (#30013)

miRx923 committed 1 year ago

Verified de11d0bd

[bnb] Fix offload test (#30039)

SunMarc committed 1 year ago

Verified 4207a407

[Trainer] Allow passing image processor (#29896)

NielsRogge committed 1 year ago

Verified 1ab71364

Fix mixtral ONNX Exporter Issue. (#29858)

AdamLouly committed 1 year ago

Verified d704c0b6

if output is tuple like facebook/hf-seamless-m4t-medium, waveform is … (#29722)

sywangyi committed 1 year ago

Verified 79d62b2d

skip `test_encode_decode_fast_slow_all_tokens` for now (#30044)

ydshieh committed 1 year ago

Verified 8b52fa6b

Add `whisper` to `IMPORTANT_MODELS` (#30046)

ydshieh committed 1 year ago

Verified 24d787ce

Refactor Cohere Model (#30027)

saurabhdash2512 committed 1 year ago

Verified 517a3e67

[`ProcessingIdefics`] Attention mask bug with padding (#29449)

byi8220 committed 1 year ago

Verified 75b76a5e

Add a converter from mamba_ssm -> huggingface mamba (#29705)

byi8220 committed 1 year ago

Verified 4e6c5eb0

Enable multi-device for efficientnet (#29989)

jla524 committed 1 year ago

Verified 03732dea

Make clearer about zero_init requirements (#29879)

muellerzr committed 1 year ago

Verified 863e2562

[`Main CIs`] Fix the red cis (#30022)

ArthurZucker committed 1 year ago

Verified 695d8233

Superpoint imports fix (#29898)

zucchini-nlp committed 1 year ago

Verified c10b5dd2

[docs] Fix audio file (#30006)

stevhliu committed 1 year ago

Verified 34bfe95a

Fix vipllava for generation (#29874)

zucchini-nlp committed 1 year ago

Verified cc75f1ac

Fix probability computation in `WhisperNoSpeechDetection` when recomputing scores (#29248)

cifkao committed 1 year ago

Verified 240e1062

Fix `kwargs` handling in `generate_with_fallback` (#29225)

cifkao committed 1 year ago

Verified bcd42c4a

Fix Qwen2Tokenizer (#29929)

jklj077 committed 1 year ago

Verified 851f253f

Fix Swinv2ForImageClassification NaN output (#29981)

Miguel Almeida committed 1 year ago

Verified 17b06e2c

Make EncodecModel.decode ONNX exportable (#29913)

fxmarty committed 1 year ago

Verified 81642d2b

Update `tests/utils/tiny_model_summary.json` (#29941)

ydshieh committed 1 year ago

Verified b44df05b

Newer Older