Commits

Branch selector

 main

User selector

All users

All time

Commit History

Commits on May 24, 2023

• Minor awesome-transformers.md fixes (#23453)

  Show description for 28aa438
  pagarsky

  authored
  28aa438

  Copy full SHA for 28aa438

• Better TF docstring types (#23477)

  Show description for f8b2574
  Rocketknight1

  authored
  f8b2574

  Copy full SHA for f8b2574

• fix gptj could not jit.trace in GPU (#23317)

  Show description for 767e6b5
  sywangyi

  authored
  767e6b5

  Copy full SHA for 767e6b5

• fix: use bool instead of uint8/byte in Deberta/DebertaV2/SEW-D to make it compatible with TensorRT (#23683)

  Show description for b4698b7
  

  uchuhimo

  and

  yuxianq

  authored
  b4698b7

  Copy full SHA for b4698b7

• Export to ONNX doc refocused on using optimum, added tflite (#23434)

  Show description for 2eaaf17
  

  MKhalusova

  and

  regisss

  authored
  2eaaf17

  Copy full SHA for 2eaaf17

• Paged Optimizer + Lion Optimizer for Trainer (#23217)

  Show description for 796162c
  

  TimDettmers

  and

  younesbelkada

  authored
  796162c

  Copy full SHA for 796162c

• 4-bit QLoRA via bitsandbytes (4-bit base model + LoRA) (#23479)

  Show description for 9d73b92
  

  3 peopleauthored
  9d73b92

  Copy full SHA for 9d73b92

• add GPTJ/bloom/llama/opt into model list and enhance the jit support (#23291)

  Show description for 33687a3
  sywangyi

  authored
  33687a3

  Copy full SHA for 33687a3

Commits on May 23, 2023

• Fix some docs what layerdrop does (#23691)

  Show description for 003a0cf
  

  zspo

  and

  sgugger

  authored
  003a0cf

  Copy full SHA for 003a0cf

• fix: load_best_model_at_end error when load_in_8bit is True (#23443)

  Show description for 357f281
  dkqkxx

  authored
  357f281

  Copy full SHA for 357f281

• Skip `TFCvtModelTest::test_keras_fit_mixed_precision` for now (#23699)

  Show description for de5f86e
  ydshieh

  authored
  de5f86e

  Copy full SHA for de5f86e

• is_batched fix for remaining 2-D numpy arrays (#23309)

  Show description for 3d57404
  LWprogramming

  authored
  3d57404

  Copy full SHA for 3d57404

• [`Blip`] Fix blip doctest (#23698)

  Show description for 6b7d6f8
  younesbelkada

  authored
  6b7d6f8

  Copy full SHA for 6b7d6f8

• TF version compatibility fixes (#23663)

  Show description for 876d9a3
  Rocketknight1

  authored
  876d9a3

  Copy full SHA for 876d9a3

• [`SAM`] Fixes pipeline and adds a dummy pipeline test (#23684)

  Show description for 42baa58
  younesbelkada

  authored
  42baa58

  Copy full SHA for 42baa58

• Fix a `BridgeTower` test (#23694)

  Show description for 71a5ed3
  ydshieh

  authored
  71a5ed3

  Copy full SHA for 71a5ed3

• 🌐 [i18n-KO] Translated `tasks/monocular_depth_estimation.mdx` to Korean (#23621)

  Show description for 1fe1e3c
  

  6 peopleauthored
  1fe1e3c

  Copy full SHA for 1fe1e3c

• Making `safetensors` a core dependency. (#23254)

  Show description for 9e8d706
  Narsil

  authored
  9e8d706

  Copy full SHA for 9e8d706

• Fix PyTorch SAM tests (#23682)

  Show description for abf691a
  ydshieh

  authored
  abf691a

  Copy full SHA for abf691a

• Fix typo in a parameter name for open llama model (#23637)

  Show description for b687af0
  aaalexlit

  authored
  b687af0

  Copy full SHA for b687af0

• Add PerSAM [bis] (#23659)

  Show description for 527ab89
  NielsRogge

  authored
  527ab89

  Copy full SHA for 527ab89

• Bump requests from 2.22.0 to 2.31.0 in /examples/research_projects/lxmert (#23668)

  Show description for aa30cd4
  dependabot[bot]

  authored
  aa30cd4

  Copy full SHA for aa30cd4

• Bump requests from 2.22.0 to 2.31.0 in /examples/research_projects/visual_bert (#23670)

  Show description for 9bf72ae
  dependabot[bot]

  authored
  9bf72ae

  Copy full SHA for 9bf72ae

• Bump requests from 2.27.1 to 2.31.0 in /examples/research_projects/decision_transformer (#23673)

  Show description for ecc05f8
  dependabot[bot]

  authored
  ecc05f8

  Copy full SHA for ecc05f8

• small fix to remove unused eos in processor when it's not used. (#23408)

  Narsil

  authored
  e30ceae

  Copy full SHA for e30ceae

Commits on May 22, 2023

• [image-to-text pipeline] Add conditional text support + GIT (#23362)

  Show description for 2f424d7
  NielsRogge

  authored
  2f424d7

  Copy full SHA for 2f424d7

• Update workflow files (#23658)

  Show description for e69feab
  ydshieh

  authored
  e69feab

  Copy full SHA for e69feab

• Update all no_trainer with skip_first_batches (#23664)

  muellerzr

  authored
  b191d7d

  Copy full SHA for b191d7d

• Fix SAM tests and use smaller checkpoints (#23656)

  Show description for 26a0681
  Rocketknight1

  authored
  26a0681

  Copy full SHA for 26a0681

• changing the requirements to a cpu torch version that works (#23483)

  sshahrokhi

  authored
  6f72e71

  Copy full SHA for 6f72e71

• Fix wav2vec2 is_batched check to include 2-D numpy arrays (#23223)

  Show description for 5de2a6d
  

  LWprogramming

  and

  sanchit-gandhi

  authored
  5de2a6d

  Copy full SHA for 5de2a6d

• Bugfix: LLaMA layer norm incorrectly changes input type and consumers lots of memory (#23535)

  Show description for 4ddd9de
  

  TimDettmers

  and

  younesbelkada

  authored
  4ddd9de

  Copy full SHA for 4ddd9de

• Muellerzr fix deepspeed (#23657)

  Show description for fe34486
  muellerzr

  authored
  fe34486

  Copy full SHA for fe34486

• Fix accelerate logger bug (#23650)

  Show description for 7bbdfd7
  

  younesbelkada

  and

  muellerzr

  authored
  7bbdfd7

  Copy full SHA for 7bbdfd7

• Fix tensor device while attention_mask is not None (#23538)

  Show description for 29294b0
  zspo

  authored
  29294b0

  Copy full SHA for 29294b0