NielsRogge/transformers

Pull Requests Commits

Merge branch 'main' into improve_hiera_docs

NielsRogge committed 2 years ago

Verified 638d00db

[mistral] Support passing `head_dim` through config (and do not require `head_dim * num_heads == hidden_size`) (#32050)

xenova committed 2 years ago

Verified 4c040aba

Fix code snippet

NielsRogge committed 2 years ago

a8aa4b13

Bump scikit-learn from 1.1.2 to 1.5.0 in /examples/research_projects/codeparrot/examples (#32052)

dependabot[bot] committed 2 years ago

Verified c50e0551

Bump scikit-learn from 1.0.2 to 1.5.0 in /examples/research_projects/decision_transformer (#31458)

dependabot[bot] committed 2 years ago

Verified c25dde1f

Chameleon: minor fixes after shipping (#32037)

zucchini-nlp committed 2 years ago

Verified 673d30b8

unpin `numpy<2.0` (#32018)

ydshieh committed 2 years ago

Verified 765732e9

NielsRogge committed 2 years ago

714a989d

NielsRogge committed 2 years ago

4f955619

Add `sdpa` and FA2 for CLIP (#31940)

qubvel committed 2 years ago

Verified 1c37e8c1

Add language to word timestamps for Whisper (#31572)

algo-robin committed 2 years ago

Verified b31d5950

Pass missing arguments to `SeamlessM4Tv2ConformerEncoderLayer.forward()` when gradient checkpointing is enabled (#31945)

anferico committed 2 years ago

Verified cb23d1b2

doc: fix broken BEiT and DiNAT model links on Backbone page (#32029)

dvrogozh committed 2 years ago

Verified bc36c26f

Fix typo in classification function selection logic to improve code consistency (#32031)

moses committed 2 years ago

Verified 63be8e6f

Fixed `log messages` that are resulting in TypeError due to too many arguments (#32017)

Sai-Suraj-27 committed 2 years ago

Verified 72fb02c4

Fix tests skip (#32012)

qubvel committed 2 years ago

Verified 691586b0

Chameleon: add model (#31534)

zucchini-nlp committed 2 years ago

Verified 24cfcc21

SpeechEncoderDecoder doesn't support param buffer assignments (#32009)

muellerzr committed 2 years ago

Verified 4037a2b5

Fix if else and *actually* enable superfast init (#32007)

muellerzr committed 2 years ago

Verified 6f40a213

Fix gather when collecting 'num_input_tokens_seen' (#31974)

CodeCreator committed 2 years ago

Verified e3917064

Bug report update -- round 2 (#32006)

gante committed 2 years ago

Verified c22efa61

fix: Fixed incorrect dictionary assignment in `src/transformers/__init__.py` (#31993)

Sai-Suraj-27 committed 2 years ago

Verified 88e0813d

add flash-attn deterministic option to flash-attn>=2.4.1 (#31961)

junrae6454 committed 2 years ago

Verified 036d3de2

Bug report update (#31983)

gante committed 2 years ago

Verified 89eec5cf

Tests: remove cuda versions when the result is the same 🧹🧹 (#31955)

gante committed 2 years ago

Verified 999981da

Fix bad test about slower init (#32002)

muellerzr committed 2 years ago

Verified 693cb828

[tests] fix deepspeed zero3 config for `test_stage3_nvme_offload` (#31881)

faaany committed 2 years ago

Verified 25e5e3fa

Speedup model init on CPU (by 10x+ for llama-3-8B as one example) (#31771)

muellerzr committed 2 years ago

Verified e0dfd7bc

Cambricon MLUs support SDPA and flash_attn (#31102)

huismiling committed 2 years ago

Verified 03a3becc

Fix the incorrect permutation of gguf (#31788)

PenutChen committed 2 years ago

Verified ac946aac

Older