Commits

Branch selector

 main

User selector

All users

Datepicker

All time

Commit History

Commits on Feb 2, 2024

• Mark test_encoder_decoder_model_generate for vision_encoder_deocder as flaky (huggingface#28842)

  Show description for 3d2900e
  amyeroberts

  authored
  3d2900e

• Reduce GPU memory usage when using FSDP+PEFT (huggingface#28830)

  Show description for 80d5007
  pacman100

  authored
  80d5007

• Use -v for pytest on CircleCI (huggingface#28840)

  Show description for f497795
  ydshieh

  authored
  f497795

• fix / skip (for now) some tests before switch to torch 2.2 (huggingface#28838)

  Show description for a7cb92a
  ydshieh

  authored
  a7cb92a

• Fix issues caused by natten (huggingface#28834)

  Show description for 0e75aee
  ydshieh

  authored
  0e75aee

• Add missing None check for hf_quantizer (huggingface#28804)

  Show description for ec29d25
  

  jganitkevitch

  and

  younesbelkada

  authored
  ec29d25

• Explicitly check if token ID's are None in TFBertTokenizer constructor (huggingface#28824)

  Show description for 1efb21c
  skumar951

  authored
  1efb21c

• [Docs] Fix spelling and grammar mistakes (huggingface#28825)

  Show description for 721ee78
  khipp

  authored
  721ee78

• [docs] HfQuantizer (huggingface#28820)

  Show description for 2418c64
  stevhliu

  authored
  2418c64

Commits on Feb 1, 2024

• [docs] Backbone (huggingface#28739)

  Show description for abbffc4
  stevhliu

  authored
  abbffc4

• Add models from deit (huggingface#28302)

  Show description for 23ea674
  

  rajveer43

  and

  stevhliu

  authored
  23ea674

• [docs] fix some bugs about parameter description (huggingface#28806)

  Show description for d98591a
  

  zspo

  and

  p_spozzhang

  authored
  d98591a

• enable graident checkpointing in DetaObjectDetection and add tests in Swin/Donut_Swin (huggingface#28615)

  Show description for e19c12e
  SangbumChoi

  authored
  e19c12e

• Add tip on setting tokenizer attributes (huggingface#28764)

  Show description for 7bc6d76
  Rocketknight1

  authored
  7bc6d76

• Fix symbolic_trace with kv cache (huggingface#28724)

  Show description for 709dc43
  fxmarty

  authored
  709dc43

• Make is_torch_bf16_available_on_device more strict (huggingface#28796)

  Show description for eb8e7a0
  ydshieh

  authored
  eb8e7a0

• Adding [T5/MT5/UMT5]ForTokenClassification (huggingface#28443)

  Show description for 0d26abd
  hackyon

  authored
  0d26abd

Commits on Jan 31, 2024

• [docs] Correct the statement in the docstirng of compute_transition_scores in generation/utils.py (huggingface#28786)

  Ki-Seki

  authored
  7b2bd1f

• Split daily CI using 2 level matrix (huggingface#28773)

  Show description for 4735866
  ydshieh

  authored
  4735866

• Add artifact name in job step to maintain job / artifact correspondence (huggingface#28682)

  Show description for 95346e9
  ydshieh

  authored
  95346e9

• DeepSpeed: hardcode torch.arange dtype on float usage to avoid incorrect initialization (huggingface#28760)

  gante

  authored
  beb2a09

• Flax mistral (huggingface#26943)

  Show description for f7076cd
  kiansierra

  authored
  f7076cd

• Wrap Keras methods to support BatchEncoding (huggingface#28734)

  Show description for 7a49610
  Rocketknight1

  authored
  7a49610

• canonical repos moves (huggingface#28795)

  Show description for 721e2d9
  

  julien-c

  and

  LysandreJik

  authored
  721e2d9

• Resolve DeepSpeed cannot resume training with PeftModel (huggingface#28746)

  Show description for bebeeee
  lh0x00

  authored
  bebeeee

• [Whisper] Refactor forced_decoder_ids & prompt ids (huggingface#28687)

  Show description for 65a926e
  

  patrickvonplaten

  and

  sanchit-gandhi

  authored
  65a926e

• [HFQuantizer] Remove check_packages_compatibility logic (huggingface#28789)

  Show description for f9f1f2a
  younesbelkada

  authored
  f9f1f2a

• don't initialize the output embeddings if we're going to tie them to input embeddings (huggingface#28192)

  Show description for ae0c27a
  tom-p-reichel

  authored
  ae0c27a

• Prevent MLflow exception from disrupting training (huggingface#28779)

  Show description for a937425
  

  codiceSpaghetti

  and

  codiceSpaghetti

  authored
  a937425

• [bnb] Fix bnb slow tests (huggingface#28788)

  Show description for d703eaa
  younesbelkada

  authored
  d703eaa

Commits on Jan 30, 2024

• Pin Torch to <2.2.0 (huggingface#28785)

  Show description for 74c9cfe
  

  Rocketknight1

  and

  ydshieh

  authored
  74c9cfe

• Add tf_keras imports to prepare for Keras 3 (huggingface#28588)

  Show description for 415e9a0
  Rocketknight1

  authored
  415e9a0

• Task-specific pipeline init args (huggingface#28439)

  Show description for 1d489b3
  amyeroberts

  authored
  1d489b3

• [Backbone] Use load_backbone instead of AutoBackbone.from_config (huggingface#28661)

  Show description for 2fa1c80
  amyeroberts

  authored
  2fa1c80

• Further pin pytest version (in a temporary way) (huggingface#28780)

  Show description for c24c524
  ydshieh

  authored
  c24c524

Pagination

PreviousNext