huggingface/transformers

Pull Requests Commits

Fixing convert.

Narsil committed 2 years ago

93d5fcd2

Fixing BPE spm converter.

Narsil committed 2 years ago

164d4841

Automatically create/update tiny models (#22275)

ydshieh committed 2 years ago

Verified e8cc0255

Enable training Llama with model or pipeline parallelism (#22329)

kooshi committed 2 years ago

Verified a92e0ad2

Generate: add test for left-padding support (#22322)

gante committed 2 years ago

Verified 502fec77

Fix --bf16 option support for Neuron after PR #22300 (#22307)

jeffhataws committed 2 years ago

Verified ec9b18f6

Added type hints to TFDeiTModel (#22327)

Batese2001 committed 2 years ago

Verified aef488c5

Minor typo in pipeline FillMaskPipeline's documentation. (#22339)

SamuelLarkin committed 2 years ago

Verified 59b9351b

Fix various imports (#22281)

sgugger committed 2 years ago

Verified 506e7c63

Mention why one needs to specify max_steps in Trainer (#22333)

lhoestq committed 2 years ago

Verified 053c2153

Fixed gradient checkpoint bug for TimeSeriesTransformer (#22272)

pmollerus23 committed 2 years ago

Verified 5a9eb314

[`MBart`] Add `accelerate` support for MBart (#22309)

younesbelkada committed 2 years ago

Verified ff20f9cf

[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

Older