huggingface/transformers

Pull Requests Commits

Typo fix in error message (#29535)

clefourrier committed 1 year ago

Verified 14536c33

fix image-to-text batch incorrect output issue (#29342)

sywangyi committed 1 year ago

Verified 8ee1d472

[tests] add the missing `require_sacremoses` decorator (#29504)

faaany committed 1 year ago

Verified 8e589c83

Generate: left-padding test, revisited (#29515)

gante committed 1 year ago

Verified bc764f42

Typo in mlx tensor support (#29509)

pcuenca committed 1 year ago

Verified 631fa7bf

Fix `VisionEncoderDecoder` Positional Arg (#29497)

nickthegroot committed 1 year ago

Verified b338a6c3

Set `inputs` as kwarg in `TextClassificationPipeline` (#29495)

alvarobartt committed 1 year ago

Verified ddf177ee

test_generation_config_is_loaded_with_model - fall back to pytorch model for now (#29521)

amyeroberts committed 1 year ago

Verified 4ed9ae62

Add support for metadata format MLX (#29335)

alexweberk committed 1 year ago

Verified 45c06510

Flava multimodal add attention mask (#29446)

zucchini-nlp committed 1 year ago

Verified 923733c2

fix: Avoid error when fsdp_config is missing xla_fsdp_v2 (#29480)

ashokponkumar committed 1 year ago

Verified 9288e759

Revert "Automatic safetensors conversion when lacking these files (#2… (#29507)

LysandreJik committed 1 year ago

Verified f6133d76

v4.39 deprecations 🧼 (#29492)

gante committed 1 year ago

Verified ffe60fdc

Enable BLIP for auto VQA (#29499)

regisss committed 1 year ago

Verified 979fccc9

Fix: Disable torch.autocast in RotaryEmbedding of Gemma and LLaMa for MPS device (#29439)

currybab committed 1 year ago

Verified d45f47ab

Substantially reduce memory usage in _update_causal_mask for large batches by using .expand instead of .repeat [needs tests+sanity check] (#29413)

nqgl committed 1 year ago

Verified 2a939f20

Fix `TextGenerationPipeline.__call__` docstring (#29491)

alvarobartt committed 1 year ago

Verified 965cf677

added the max_matching_ngram_size to GenerationConfig (#29131)

mosheber committed 1 year ago

Verified 19fb1e22

Generate: torch.compile-ready generation config preparation (#29443)

gante committed 1 year ago

Verified ddb4fda3

Fix test failure on DeepSpeed (#29444)

muellerzr committed 1 year ago

Verified 9322576e

Avoid dummy token in PLD to optimize performance (#29445)

ofirzaf committed 1 year ago

Verified 0a5b0516

Generate: get generation mode from the generation config instance 🧼 (#29441)

gante committed 1 year ago

Verified 700d48fb

Generate: add tests for caches with `pad_to_multiple_of` (#29462)

gante committed 1 year ago

Verified 41f7b7ae

Fix TrainingArguments regression with torch <2.0.0 for dataloader_prefetch_factor (#29447)

Matthew Hoffman committed 1 year ago

Verified 2890116a

[`docs`] Add starcoder2 docs (#29454)

younesbelkada committed 1 year ago

Verified b27aa206

[`Docs` / `Awq`] Add docs on exllamav2 + AWQ (#29474)

younesbelkada committed 1 year ago

Verified 2a002d07

[FIX] `offload_weight()` takes from 3 to 4 positional arguments but 5 were given (#29457)

faaany committed 1 year ago

Verified 00bf4427

🌐 [i18n-KO] Translated generation_strategies.md to Korean (#29086)

AI4Harmony committed 1 year ago

Verified 7b01579f

[i18n-zh] Translate add_new_pipeline.md into Chinese (#29432)

windsonsea committed 1 year ago

Verified 638c423c

Automatic safetensors conversion when lacking these files (#29390)

LysandreJik committed 1 year ago

Verified a69cbf4e

Newer Older