huggingface/transformers

Pull Requests Commits

Rocketknight1 committed 3 years ago

223937ab

Any tensor with rank > 3 probably has a variable dimension

Rocketknight1 committed 3 years ago

4f8984e9

EXTREMELY BREAKING changes to predict_step

Rocketknight1 committed 3 years ago

945a0c99

Force use_cache to be False in PyTorch (#15385)

ydshieh committed 3 years ago

Verified 6a5472a8

[GPTJ] fix docs (#15558)

patil-suraj committed 3 years ago

Verified 0acd84f7

electra is added to onnx supported model (#15084)

aaron-dunamu committed 3 years ago

Verified 87d08afb

FX tracing improvement (#14321)

michaelbenayoun committed 3 years ago

Verified 0fe17f37

Create a custom model guide (#15489)

stevhliu committed 3 years ago

Verified 552f8d30

Make TF Wav2Vec2 outputs the same as PT's version (#15530)

ydshieh committed 3 years ago

Verified ad1d3c4d

Fix TF T5/LED missing cross attn in retrun values (#15511)

ydshieh committed 3 years ago

Verified 131e2584

Remove Longformers from ONNX-supported models (#15273)

lewtun committed 3 years ago

Verified 6775b211

Wav2Vec2 models must either throw or deal with add_apater (#15409)

FremyCompany committed 3 years ago

Verified 7a1412e1

Add ASR CTC streaming example (#15309)

anton-l committed 3 years ago

Verified a459f7f9

[Trainer] Deeper length checks for IterableDatasetShard (#15539)

anton-l committed 3 years ago

Verified 75b13f82

Add ConvNeXT (#15277)

NielsRogge committed 3 years ago

Verified 84eec9e6

[torch_int_div] Correct true division in generation (#15498)

patrickvonplaten committed 3 years ago

Verified c47d2592

[ASR pipeline] correct asr pipeline for seq2seq models (#15541)

patrickvonplaten committed 3 years ago

Verified 5f1918a4

Revert "Handle PyTorch to Flax conversion of 1D convolutions (#15519)" (#15540)

patrickvonplaten committed 3 years ago

Verified e02bdce7

[deepspeed docs] DeepSpeed ZeRO Inference (#15486)

stas00 committed 3 years ago

Verified 8ce13306

Standardize semantic segmentation models outputs (#15469)

sgugger committed 3 years ago

Verified ac6aa10f

[deepspeed docs] Megatron-Deepspeed info (#15488)

stas00 committed 3 years ago

Verified 31be2f45

Fix TFRemBertEncoder all_hidden_states (#15510)

ydshieh committed 3 years ago

Verified bbe9c698

Handle PyTorch to Flax conversion of 1D convolutions (#15519)

sanchit-gandhi committed 3 years ago

Verified 854a0d52

use kwargs (#15509)

ydshieh committed 3 years ago

Verified 486260c6

Remove loss from some flax models docs & examples (#15492)

ydshieh committed 3 years ago

Verified 525dbbf8

[deepspeed docs] memory requirements (#15506)

stas00 committed 3 years ago

Verified 21dcaec5

[WIP] Add preprocess_logits_for_metrics Trainer param (#15473)

davidleonfdez committed 3 years ago

Verified f1a4c4ea

[deepspeed] fix a bug in a test (#15493)

stas00 committed 3 years ago

Verified 4f5faaf0

Add general vision docstrings (#15501)

NielsRogge committed 3 years ago

Verified 90166121

[Flax tests] Disable scheduled GPU tests (#15503)

patrickvonplaten committed 3 years ago

Verified e2b6e73f

Older