NielsRogge/transformers

Pull Requests Commits

Merge remote-tracking branch 'upstream/main' into fix_issue_29302

NielsRogge committed 2 years ago

8ec02408

Avoid using uncessary `get_values(MODEL_MAPPING)` (#29362)

ydshieh committed 2 years ago

Verified 44fe1a1c

Merge remote-tracking branch 'upstream/main' into fix_issue_29302

NielsRogge committed 2 years ago

f2962be1

FIX [`CI`] `require_read_token` in the llama FA2 test (#29361)

younesbelkada committed 2 years ago

Verified b647acdb

FIX [`CI`]: Fix failing tests for peft integration (#29330)

younesbelkada committed 2 years ago

Verified 8d8ac9c2

FIX [`CI` / `starcoder2`] Change starcoder2 path to correct one for slow tests (#29359)

younesbelkada committed 2 years ago

Verified 1aee9afd

NielsRogge committed 2 years ago

a50c3f1e

[i18n-zh] Sync source/zh/index.md (#29331)

windsonsea committed 2 years ago

Verified 2209b7af

Better SDPA unmasking implementation (#29318)

fxmarty committed 2 years ago

Verified 49204c1d

[CI] Quantization workflow (#29046)

SunMarc committed 2 years ago

Verified f54d82ca

check if position_ids exists before using it (#29306)

jiqing-feng committed 2 years ago

Verified 554e7ada

RoPE loses precision for Llama / Gemma + Gemma logits.float() (#29285)

danielhanchen committed 2 years ago

Verified d3a4b475

Idefics: generate fix (#29320)

gante committed 2 years ago

Verified 7628b3a0

Disable Mixtral `output_router_logits` during inference (#29249)

LeonardoEmili committed 2 years ago

Verified 2ce56d35

[`Llama ROPE`] Fix torch export but also slow downs in forward (#29198)

ArthurZucker committed 2 years ago

Verified 8a8a0a4a

[`T5 and Llama Tokenizer`] remove warning (#29346)

ArthurZucker committed 2 years ago

Verified 7c87f357

[`require_read_token`] fix typo (#29345)

ArthurZucker committed 2 years ago

Verified a5288852

Remove numpy usage from owlvit (#29326)

fxmarty committed 2 years ago

Verified e715c78c

FIX [`Gemma` / `CI`] Make sure our runners have access to the model (#29242)

younesbelkada committed 2 years ago

Verified ad00c482

simplify get_class_in_module and fix for paths containing a dot (#29262)

cebtenzzre committed 2 years ago

Verified bd5b9863

Starcoder2 model - bis (#29215)

RaymondLi0 committed 2 years ago

Verified 63caa370

[i18n-zh] Translate fsdp.md into Chinese (#29305)

windsonsea committed 2 years ago

Verified 83ab0115

Fix a few typos in `GenerationMixin`'s docstring (#29277)

sadra-barikbin committed 2 years ago

Verified 227cd54a

Token level timestamps for long-form generation in Whisper (#29148)

zucchini-nlp committed 2 years ago

Verified ddf7ac42

Add compatibility with skip_memory_metrics for mps device (#29264)

SunMarc committed 2 years ago

Verified 8a1faf28

Use torch 2.2 for deepspeed CI (#29246)

ydshieh committed 2 years ago

Verified 5c341d45

[tests] enable benchmark unit tests on XPU (#29284)

faaany committed 2 years ago

Verified 63a0c8f1

Fix `attn_implementation` documentation (#29295)

fxmarty committed 2 years ago

Verified 6d3b643e

Image Feature Extraction docs (#28973)

merveenoyan committed 2 years ago

Verified 83e366bf

Cleaner Cache `dtype` and `device` extraction for CUDA graph generation for quantizers compatibility (#29079)

BlackSamorez committed 2 years ago

Verified e3fc90ae

Older