huggingface/transformers

Pull Requests Commits

[gptj] support older pytorch version (#22325)

stas00 committed 2 years ago

Verified 61f79b29

Really fix quality due to ruff release

sgugger committed 2 years ago

Verified 80e3b363

Fix quality due to ruff release

sgugger committed 2 years ago

ef28df05

[deepspeed zero3] need `generate(synced_gpus=True, ...)` (#22242)

stas00 committed 2 years ago

Verified 73fdc8c5

Fix PipelineTests skip conditions (#22320)

ydshieh committed 2 years ago

Verified 8b05ace0

Chunkable token classification pipeline (#21771)

luccailliau committed 2 years ago

Verified d62e7d88

docs: Resolve incorrect type typo in trainer methods (#22316)

tomaarsen committed 2 years ago

Verified f48d3314

Add Pix2Struct (#21400)

younesbelkada committed 2 years ago

Verified 0f68a7f4

Beef up Llama tests (#22314)

gante committed 2 years ago

Verified fd3eb3e3

Generate: Export TF generate with a TF tokenizer (#22310)

gante committed 2 years ago

Verified 12febc20

Enforce `max_memory` for device_map strategies (#22311)

sgugger committed 2 years ago

Verified 5fd4e3c8

Fixed bug to calculate correct xpath_sub_list in MarkupLMTokenizer (#22302)

silentghoul-spec committed 2 years ago

Verified 48bef3a7

Fix position embeddings for GPT-J and CodeGen (#22069)

njhill committed 2 years ago

Verified 4e94c6c0

fix: Allow only test_file in pytorch and flax summarization (#22293)

Connor Henderson committed 2 years ago

Verified 8e6c34b3

add low_cpu_mem_usage option in run_clm.py example which will benefit… (#22288)

sywangyi committed 2 years ago

Verified 4ccaf268

Enable traced model for text-generation task (#22265)

jiqing-feng committed 2 years ago

Verified 8472a224

Add MaskedImageModelingOutput (#22212)

alaradirik committed 2 years ago

Verified 0558914d

Final update of doctest (#22299)

ydshieh committed 2 years ago

Verified 0dcb46e7

[deepspeed] offload + non-cpuadam optimizer exception doc (#22044)

stas00 committed 2 years ago

Verified 89a0a9ea

Correct NATTEN function signatures and force new version (#22298)

alihassanijr committed 2 years ago

Verified 5990743f

Restore fp16 support on xla gpu device (#22300)

ymwangg committed 2 years ago

Verified d35f7296

Time to Say Goodbye, torch 1.7 and 1.8 (#22291)

ydshieh committed 2 years ago

Verified 67c2dbdb

Add translation perf_infer_gpu_one for it (#22296)

davidegazze committed 2 years ago

Verified 86c7931a

fix more doctests (#22292)

ydshieh committed 2 years ago

Verified d0b942d1

More doctests (#22268)

ydshieh committed 2 years ago

Verified 48327c57

Fix error in mixed precision training of `TFCvtModel` (#22267)

gcuder committed 2 years ago

Verified 5a2b77a6

replace_8bit_linear modules_to_not_convert default value fix (#22238)

Andrei Panferov committed 2 years ago

Verified 330d8b99

Update vision docstring bool masked pos (#22237)

amyeroberts committed 2 years ago

Verified c07a02a4

Example of pad_to_multiple_of for padding and truncation guide & docstring update (#22278)

MKhalusova committed 2 years ago

Verified 7bd86505

Move torch.compile() wrapping after DDP/FSDP wrapping to ensure correct graph breaks during training (#22279)

ani300 committed 2 years ago

Verified fb0a38b4

Older