Commits · d1dc66d9718a29a4e00143c2b622dc177e00d3c0 · OpenDAS / Fairseq

"tests/python/vscode:/vscode.git/clone" did not exist on "c09d466082b4f7e5bc2f7509cb90bd200d9ead50"

06 Dec, 2018 1 commit

Fix arg formatting in preprocess.py and add fmt control for black formatting (#399) · 82a9f923

Myle Ott authored Dec 06, 2018

Summary:
Not switching to Black formatting just yet, but adding fmt: off directives in case we decide to later.
Pull Request resolved: https://github.com/pytorch/fairseq/pull/399

Differential Revision: D13364674

Pulled By: myleott

fbshipit-source-id: a20a11a18be3d583ee30eff770278fb4bd05b93c

82a9f923

15 Jun, 2018 5 commits
- remove unused verbose option & make arguments to averaging script nicer · a3e4c4c3
  alexeib authored May 23, 2018
  
  a3e4c4c3
- ability to checkpoint when reaching certain number of updates · fc312d28
  Alexei Baevski authored May 23, 2018
  
  fc312d28
- add support for averaging last n checkpoints · f6a5a54e
  Alexei Baevski authored May 04, 2018
  
  f6a5a54e
- Fix tests · 8afb7761
  Myle Ott authored Apr 24, 2018
  
  8afb7761
- Add FP16 support · 7ee1d284
  Myle Ott authored Apr 10, 2018
  
  7ee1d284
02 Apr, 2018 1 commit

Merge internal changes (#136) · d3795d6c

Myle Ott authored Apr 02, 2018

Changes:
- 7d19e36: Add `--sampling` flag to generate.py to sample instead of doing beam search
- c777340: Add `scripts/average_checkpoints.py` to average multiple checkpoints into a combined model
- 3ea882c: Add `--max-update` option to train.py to stop training after a given number of updates
- small bugfixes for distributed training, LSTM, inverse square root LR scheduler

d3795d6c