Commits

Branch selector

 v4.9.2

User selector

All users

Datepicker

All time

Commit History

Commits on Aug 9, 2021

• Patch release: v4.9.2

  LysandreJik

  committed
  41981a2

• Tpu tie weights (#13030)

  Show description for ec78422
  

  sgugger

  authored and

  LysandreJik

  committed
  ec78422

• Add to ONNX docs (#13048)

  Show description for bfd5354
  

  LysandreJik

  and

  sgugger

  committed
  bfd5354

• Add MBART to models exportable with ONNX (#13049)

  Show description for 226763a
  LysandreJik

  committed
  226763a

• Put smaller ALBERT model (#13028)

  LysandreJik

  committed
  f595ea3

• T5 with past ONNX export (#13014)

  Show description for a12fa50
  

  michaelbenayoun

  authored and

  LysandreJik

  committed
  a12fa50

• GPT-Neo ONNX export (#12911)

  Show description for 94b7db9
  

  michaelbenayoun

  authored and

  LysandreJik

  committed
  94b7db9

• Fix push_to_hub for TPUs (#12895)

  

  sgugger

  authored and

  LysandreJik

  committed
  2c255a2

• ONNX v2 raises an Exception when using PyTorch < 1.8.0 (#12933)

  Show description for ca272fc
  

  mfuntowicz

  authored and

  LysandreJik

  committed
  ca272fc

Commits on Jul 26, 2021

• Release: v4.9.1

  sgugger

  committed
  bff1c71

• Fix barrier for SM distributed (#12853)

  sgugger

  committed
  8ee16d8

• Add doc for v4.9.0

  sgugger

  committed
  6cab8b3

Commits on Jul 22, 2021

• Release: v4.9.0

  LysandreJik

  committed
  72aee83

• Fix type of max_seq_length arg in run_swag.py (#12832)

  mbforbes

  authored
  fcf8301

Commits on Jul 21, 2021

• [parallelism doc] document Deepspeed-Inference and parallelformers (#12836)

  Show description for 27a8c9e
  

  stas00

  and

  sgugger

  authored
  27a8c9e

• [Deepspeed] warmup_ratio docs (#12830)

  Show description for 807b6bd
  

  stas00

  and

  sgugger

  authored
  807b6bd

• Raise warning in HP search when hp is not in args (#12831)

  sgugger

  authored
  8c2384d

• [debug] DebugUnderflowOverflow doesn't work with DP (#12816)

  stas00

  authored
  cf0755a

• Add _CHECKPOINT_FOR_DOC to all models (#12811)

  Show description for ac3cb66
  

  LysandreJik

  and

  sgugger

  authored
  ac3cb66

• Add versioning system to fast tokenizer files (#12713)

  Show description for 786ced3
  

  sgugger

  and

  LysandreJik

  authored
  786ced3

• Refer warmup_ratio when setting warmup_num_steps. (#12818)

  Show description for 037bdf8
  tsuchm

  authored
  037bdf8

• fix convert_tokens_to_string calls (#11716)

  PhilipMay

  authored
  15d19ec

• Expose get_config() on ModelTesters (#12812)

  Show description for c3d9ac7
  LysandreJik

  authored
  c3d9ac7

Commits on Jul 20, 2021

• [trainer] sanity checks for save_steps=0|None and logging_steps=0 (#12796)

  Show description for cabcc75
  

  stas00

  and

  sgugger

  authored
  cabcc75

• Update README.md

  patrickvonplaten

  authored
  acdd78d

• add and fix examples (#12810)

  patil-suraj

  authored
  b5b4e54

• Update README.md

  patrickvonplaten

  authored
  31d0672

• [Longformer] Correct longformer docs (#12809)

  Show description for 2955d50
  

  patrickvonplaten

  and

  Patrick von Platen

  authored
  2955d50

• Update README.md

  Show description for 13fefdf
  patrickvonplaten

  authored
  13fefdf

• Flax MLM: Allow validation split when loading dataset from local file (#12689)

  Show description for 66197ad
  fgaim

  authored
  66197ad

• Fix Padded Batch Error 12282 (#12487)

  Show description for 6f8e367
  will-rice

  authored
  6f8e367

• add troubleshooting docs (#12791)

  stas00

  authored
  7fae535

Commits on Jul 19, 2021

• Enforce eval and save strategies are compatible when --load_best_model_at_end (#12786)

  Show description for 0118ef8
  sgugger

  authored
  0118ef8

• Longer timeout for slow tests (#12779)

  LysandreJik

  authored
  546dc24

• [ray] Fix datasets_modules ImportError with Ray Tune (#12749)

  Show description for cab3b86
  Yard1

  authored
  cab3b86

Pagination

PreviousNext