Commits

Breadcrumbs

History for

onv4.4.2

User selector

All users

Datepicker

All time

Commit History

Commits on Mar 16, 2021

• Release v4.4.0

  LysandreJik

  committed
  c988db5

  Copy full SHA for c988db5

Commits on Mar 15, 2021

• independent training / eval with local files (#10710)

  Show description for 87d685b
  riklopfer

  authored
  87d685b

  Copy full SHA for 87d685b

• Add minimum version check in examples (#10724)

  Show description for 4c379da
  sgugger

  authored
  4c379da

  Copy full SHA for 4c379da

• zero-shot pipeline multi_class -> multi_label (#10727)

  joeddav

  authored
  966ba08

  Copy full SHA for 966ba08

• split seq2seq script into summarization & translation (#10611)

  Show description for 6f84099
  

  5 peopleauthored
  6f84099

  Copy full SHA for 6f84099

Commits on Mar 12, 2021

• AdamW is now supported by default (#9624)

  stas00

  authored
  4c32f9f

  Copy full SHA for 4c32f9f

Commits on Mar 11, 2021

• Specify minimum version for sacrebleu (#10662)

  LysandreJik

  authored
  9fbb4cd

  Copy full SHA for 9fbb4cd

• Update README.md (#10647)

  Show description for 27d9e05
  Arvid-pku

  authored
  27d9e05

  Copy full SHA for 27d9e05

Commits on Mar 10, 2021

• Add new GLUE example with no Trainer. (#10555)

  Show description for efb5c0a
  sgugger

  authored
  efb5c0a

  Copy full SHA for efb5c0a

• Fixes an issue in `text-classification` where MNLI eval/test datasets are not being preprocessed. (#10621)

  Show description for 6f52fce
  

  Allen Wang

  authored
  6f52fce

  Copy full SHA for 6f52fce

Commits on Mar 9, 2021

• Fairscale FSDP fix model save (#10596)

  Show description for 0d909f6
  sgugger

  authored
  0d909f6

  Copy full SHA for 0d909f6

Commits on Mar 8, 2021

• [examples tests on multigpu] resolving require_torch_non_multi_gpu_but_fix_me (#10561)

  Show description for f284089
  stas00

  authored
  f284089

  Copy full SHA for f284089

• Added max_sample_ arguments (#10551)

  Show description for dfd16af
  bhadreshpsavani

  authored
  dfd16af

  Copy full SHA for dfd16af

• [examples tests] various fixes (#10584)

  Show description for 917f104
  stas00

  authored
  917f104

  Copy full SHA for 917f104

• fix nltk lookup (#10585)

  stas00

  authored
  e6ce636

  Copy full SHA for e6ce636

Commits on Mar 6, 2021

• offline mode for firewalled envs (#10407)

  Show description for 88a951e
  

  stas00

  and

  sgugger

  authored
  88a951e

  Copy full SHA for 88a951e

Commits on Mar 5, 2021

• fix run seq2seq (#10547)

  patrickvonplaten

  authored
  395ffcd

  Copy full SHA for 395ffcd

Commits on Mar 4, 2021

• Not always consider a local model a checkpoint in run_glue (#10517)

  sgugger

  authored
  a5bd40b

  Copy full SHA for a5bd40b

• Revert "Not always consider a local model a checkpoint in run_glue"

  Show description for 745ea78
  sgugger

  committed
  745ea78

  Copy full SHA for 745ea78

• Not always consider a local model a checkpoint in run_glue

  sgugger

  committed
  f366061

  Copy full SHA for f366061

Commits on Mar 1, 2021

• Add Fine-Tuning for Wav2Vec2 (#10145)

  Show description for 0234de8
  

  patrickvonplaten

  and

  LysandreJik

  authored
  0234de8

  Copy full SHA for 0234de8

Commits on Feb 27, 2021

• updated logging and saving metrics (#10436)

  Show description for aca6288
  bhadreshpsavani

  authored
  aca6288

  Copy full SHA for aca6288

• [run_seq2seq.py] restore functionality: saving to test_generations.txt (#10428)

  Show description for f52a158
  stas00

  authored
  f52a158

  Copy full SHA for f52a158

• [examples] better model example (#10427)

  Show description for ee04b69
  stas00

  authored
  ee04b69

  Copy full SHA for ee04b69

Commits on Feb 25, 2021

• Fix run_glue evaluation when model has a label correspondence (#10401)

  sgugger

  authored
  17b6e0d

  Copy full SHA for 17b6e0d

• Add support for ZeRO-2/3 and ZeRO-offload in fairscale (#10354)

  Show description for 9d14be5
  

  sgugger

  and

  stas00

  authored
  9d14be5

  Copy full SHA for 9d14be5

• [PretrainedFeatureExtractor] + Wav2Vec2FeatureExtractor, Wav2Vec2Processor, Wav2Vec2Tokenizer (#10324)

  Show description for cb38ffc
  patrickvonplaten

  authored
  cb38ffc

  Copy full SHA for cb38ffc

Commits on Feb 24, 2021

• [Trainer/Deepspeed] handle get_last_lr() before first step() (#10362)

  Show description for 3437d12
  stas00

  authored
  3437d12

  Copy full SHA for 3437d12

Commits on Feb 23, 2021

• Fix broken examples/seq2seq/README.md markdown (#10344)

  Wikidepia

  authored
  23e87c2

  Copy full SHA for 23e87c2

Commits on Feb 22, 2021

• [trainer] add Trainer methods for metrics logging and saving (#10266)

  Show description for 622a8c5
  

  stas00

  and

  sgugger

  authored
  622a8c5

  Copy full SHA for 622a8c5

• [Trainer] implement gradient_accumulation_steps support in DeepSpeed integration (#10310)

  Show description for eab0afc
  stas00

  authored
  eab0afc

  Copy full SHA for eab0afc

• defensive programming + expand/correct README (#10295)

  stas00

  authored
  f991dae

  Copy full SHA for f991dae

Commits on Feb 19, 2021

• Move the TF NER example (#10276)

  jplu

  authored
  536aee9

  Copy full SHA for 536aee9

• Zero shot distillation script cuda patch (#10284)

  joeddav

  authored
  cbadb52

  Copy full SHA for cbadb52

Commits on Feb 18, 2021

• Script for distilling zero-shot classifier to more efficient student (#10244)

  Show description for c6fe175
  

  joeddav

  and

  LysandreJik

  authored
  c6fe175

  Copy full SHA for c6fe175

Pagination

Previous

Next