NielsRogge/transformers

Pull Requests Commits

🚨🚨🚨 Update min version of accelerate to 0.26.0 (#32627)

SunMarc committed 1 year ago

Verified fd06ad54

Allow-head-dim (#32857)

ArthurZucker committed 1 year ago

Verified 13e645bb

Add tip to clarify tool calling (#32883)

Rocketknight1 committed 1 year ago

Verified 85345bb4

Docs: Fixed `whisper-large-v2` model link in docs (#32871)

Sai-Suraj-27 committed 1 year ago

Verified 37204848

Fix: Mamba2 generation mismatch between input_ids and inputs_embeds (#32694)

vasqu committed 1 year ago

Verified 61d89c19

Mamba / FalconMamba: Fix mamba left padding (#32677)

younesbelkada committed 1 year ago

Verified 93e538ae

Fix incorrect vocab size retrieval in GGUF config (#32551)

Isotr0py committed 1 year ago

Verified 59e8f191

RT-DETR parameterized batchnorm freezing (#32631)

AlanBlanchet committed 1 year ago

Verified 5f6c080b

Support save/load ckpt for XLA FSDP (#32311)

yitongh committed 1 year ago

Verified 8a4857c0

Add __repr__ for Conv1D (#32425)

AaronZLT committed 1 year ago

Verified f1b720ed

[tests] make `test_sdpa_can_compile_dynamic` device-agnostic (#32519)

faaany committed 1 year ago

Verified e55b33ce

support torch-speech (#32537)

itazap committed 1 year ago

Verified 54b77036

Add Descript-Audio-Codec model (#31494)

kamilakesbi committed 1 year ago

Verified 8260cb31

Add Flax Dinov2 (#31960)

MHRDYN7 committed 1 year ago

Verified 843e5e20

generate: missing `to` in DoLa body, causing exceptions in multi-gpu generation (#32856)

gante committed 1 year ago

Verified 52cb4034

Make beam_constraints.Constraint.advance() docstring more accurate (#32674)

alex-calderwood committed 1 year ago

Verified 6806d335

Reduce the error log when using core models that need their weights renamed, and provide a step forward (#32656)

muellerzr committed 1 year ago

Verified 8ec028ad

fix multi-gpu with static cache (#32543)

SunMarc committed 1 year ago

Verified 1c36db69

Revert PR 32299, flag users when Zero-3 was missed (#32851)

muellerzr committed 1 year ago

Verified 0b066bed

improve _get_is_as_tensor_fns (#32596)

zrr1999 committed 1 year ago

Verified f20d0e81

Fix AutoConfig and AutoModel support for Llava-Next-Video (#32844)

TKONIY committed 1 year ago

Verified a27182b7

Cache: use `batch_size` instead of `max_batch_size` (#32657)

gante committed 1 year ago

Verified cf32ee17

[tests] make test_sdpa_equivalence device-agnostic (#32520)

faaany committed 1 year ago

Verified 8f9fa3b0

Generate: unify `LogitsWarper` and `LogitsProcessor` (#32626)

gante committed 1 year ago

Verified 70d5df61

Use head_dim if in config for RoPE (#32495)

suiyoubi committed 1 year ago

Verified 5fd7ca7b

add back the position ids (#32554)

ArthurZucker committed 1 year ago

Verified c2155235

VLMs: small clean-up for cache class (#32417)

zucchini-nlp committed 1 year ago

Verified f3c8b180

fix: update doc link for runhouse in README.md (#32664)

muddlebee committed 1 year ago

Verified d6751d91

fix: Corrected ` falcon-mamba-7b` model checkpoint name (#32837)

Sai-Suraj-27 committed 1 year ago

Verified ab7e893d

reopen: llava-next fails to consider padding_side during Training (#32679)

jp1924 committed 1 year ago

Verified e8401273

Newer Older