huggingface/transformers

Pull Requests Commits

Fix modular conversion CI

Rocketknight1 committed 1 year ago

6630d624

Fix a mistake in #36175 (#36179)

ydshieh committed 1 year ago

Verified 8fd4bc7d

Follow up to SpQR integration (#36176)

MekkCyber committed 1 year ago

Verified b1a2de07

Fix the key name for _load_rng_state under torch.cuda (#36138)

wizyoung committed 1 year ago

Verified 12962fe8

Make `check_repository_consistency` run faster by MP (#36175)

ydshieh committed 1 year ago

Verified bfe46c98

Optimize Qwen2VL vision model by precomputing cos/sin embeds before ViT blocks (#35837)

li-plus committed 1 year ago

Verified 5f0fd118

Use tqdm auto (#35726)

SmartManoj committed 1 year ago

Verified d72642bc

CI: avoid human error, automatically infer generative models (#33212)

gante committed 1 year ago

Verified 62c7ea02

add disable compile option (#36161)

ArthurZucker committed 1 year ago

Verified 06231fdf

fix training issues (#36158)

ArthurZucker committed 1 year ago

Verified 0ca72592

Efficient Inference Kernel for SpQR (#34976)

elvircrn committed 1 year ago

Verified 845b0a26

Bump transformers from 4.38.0 to 4.48.0 in /examples/research_projects/adversarial (#36168)

dependabot[bot] committed 1 year ago

Verified c5506f4f

Bump transformers from 4.38.0 to 4.48.0 in /examples/tensorflow/language-modeling-tpu (#36167)

dependabot[bot] committed 1 year ago

Verified d7c5d1b5

[generate] revert change in Aria: the maximum cache length must match `max_length` (#36120)

gante committed 1 year ago

Verified 636ee574

Fix : fix doc fp8 (#36173)

MekkCyber committed 1 year ago

Verified b41591d8

Fix red CI (#36174)

ArthurZucker committed 1 year ago

Verified b079dd1f

[Modular] skip modular checks based on diff (#36130)

gante committed 1 year ago

Verified d114a6f7

Remove loading custom kernel for RT-DETRv2 (#36098)

qubvel committed 1 year ago

Verified 6397916d

Adding FP8 Quantization to transformers (#36026)

MekkCyber committed 1 year ago

Verified efe72fe2

Helium documentation fixes (#36170)

LysandreJik committed 1 year ago

Verified c82319b4

Move `DataCollatorForMultipleChoice` from the docs to the package (#34763)

bauwenst committed 1 year ago

Verified 8f137b24

Fix PretrainedTokenizerFast check => Fix PretrainedTokenizerFast Save (#35835)

CL-ModelCloud committed 1 year ago

Verified 35c15505

docs: fix return type annotation of `get_default_model_revision` (#35982)

MarcoGorelli committed 1 year ago

Verified 3c912c90

qwen2.5vl: fix bugs when using flash2+bf16 or num_return_sequences>1 (#36083)

gewenbin0992 committed 1 year ago

Verified 6a1ab634

Fix tests for vision models (#35654)

qubvel committed 1 year ago

Verified d4198628

Replace deprecated update_repo_visibility (#35970)

Wauplin committed 1 year ago

Verified e60ae0d0

Fix Gemma2 dtype issue when storing weights in float16 precision (#35398)

Nerogar committed 1 year ago

Verified 9065cf0d

Add reminder config to issue template and print DS version in env (#35156)

Ben-Schneider-code committed 1 year ago

Verified 08ab1abf

Fix PaliGemma Pad Token Masking During Training #35855 (#35859)

sambhavnoobcoder committed 1 year ago

Verified 950cfb0b

Mllama fsdp (#36000)

blbadger committed 1 year ago

Verified 1614d196

Older