Skip to content
  1. Mar 15, 2019
    • Myle Ott's avatar
      0.6.1 -> 0.6.2 (#577) · e6422528
      Myle Ott authored 
      Summary:
Changelog:
- 998ba4f: Add language models from Baevski & Auli (2018)
- 4294c4f6: Add mixture of experts code from Shen et al. (2019)
- 00493490: Add example for multilingual training
- 48d9afbe: Speed improvements, including fused operators from apex
- 44d27e64: Add Tensorboard support
- d17fa851: Add Adadelta optimizer
- 9e1c880f: Add `FairseqEncoderModel`
- b65c579b: Add `FairseqTask.inference_step` to modularize generate.py
- 2ad1178e: Add back `--curriculum`
- Misc bug fixes and other features

Pull Request resolved: https://github.com/pytorch/fairseq/pull/577

Differential Revision: D14481233

Pulled By: myleott

fbshipit-source-id: 4ff8625ef1c0b24273fc65df7c5658e3c932e8b7
      v0.6.2
      e6422528
  3. Mar 14, 2019
  5. Mar 13, 2019
  7. Mar 12, 2019
    • Dmytro Okhonko's avatar
      Handle 3+ dimensional input in sequence_generator + nits · 860010e9
      Dmytro Okhonko authored 
      Summary: sequence_generator assumes that model input is 2d tensor of longs. But it can be something like 3d tensor of floats and we should be able to handle this as long as first dimension is batch size followed by source lengths.

Reviewed By: myleott

Differential Revision: D14420044

fbshipit-source-id: bf8b1e42ad1873f7b803c1a377b0af21648db015
      860010e9
    • Dmytro Okhonko's avatar
      Adadelta optimizer · d17fa851
      Dmytro Okhonko authored 
      Summary: Adding Adadelta optimizer to fairseq as wrapper around torch.optim.Adadelta

Reviewed By: myleott

Differential Revision: D14418635

fbshipit-source-id: 6bf5ec008e905a4a2cbf7415e9492f5eea3ff07f
      d17fa851
    • Dmytro Okhonko's avatar
      FairseqEncoderModel · 9e1c880f
      Dmytro Okhonko authored 
      Summary: Base class for encoder-only models. Some models doesn't have decoder part.

Reviewed By: myleott

Differential Revision: D14413406

fbshipit-source-id: f36473b91dcf3c835fd6d50e2eb6002afa75f11a
      9e1c880f
  9. Mar 11, 2019
    • Matt Le's avatar
      Create fairseq_cli_lib · 7fc9a3be
      Matt Le authored 
      Summary: This allows one to call fairseq_cli functions from within python without dispatching to bash.

Reviewed By: myleott

Differential Revision: D14404719

fbshipit-source-id: 044eb652045bb15fc40e72ecbaf6fb10df9f8c61
      7fc9a3be
    • Jose Fonollosa's avatar
      Add missing parentheses in regex expression (#567) · fef4e002
      Jose Fonollosa authored 
      Summary:
The regex pattern without parentheses is not correct. The checkpoints are not sorted in descending order
Pull Request resolved: https://github.com/pytorch/fairseq/pull/567

Differential Revision: D14404380

Pulled By: myleott

fbshipit-source-id: 98cd0cfa8c92b78a03ffbb94840bc0f7a118eca1
      fef4e002
  11. Mar 04, 2019
  13. Mar 02, 2019
  15. Mar 01, 2019
  17. Feb 28, 2019
  19. Feb 26, 2019
  21. Feb 24, 2019
  23. Feb 23, 2019
  25. Feb 22, 2019
  27. Feb 19, 2019
    • Ruty Rinott's avatar
      moving masking logic to collate · 08e866f9
      Ruty Rinott authored 
      Summary: Move masking logic to data_utils

Reviewed By: kartikayk, jingfeidu

Differential Revision: D14098403

fbshipit-source-id: c7b7e811ab48b9c5a12662dc1e2f2ed694724176
      08e866f9
  29. Feb 16, 2019
  31. Feb 12, 2019
  33. Feb 09, 2019
    • Myle Ott's avatar
      Add fairseq to PyPI (#495) · fbd4cef9
      Myle Ott authored 
      Summary:
- fairseq can now be installed via pip: `pip install fairseq`
- command-line tools are globally accessible: `fairseq-preprocess`, `fairseq-train`, `fairseq-generate`, etc.
Pull Request resolved: https://github.com/pytorch/fairseq/pull/495

Differential Revision: D14017761

Pulled By: myleott

fbshipit-source-id: 10c9f6634a3056074eac2f33324b4f1f404d4235
      v0.6.1
      fbd4cef9
  35. Feb 07, 2019
    • Ruty Rinott's avatar
      stitch preprocessing pipeline · cea0e4b9
      Ruty Rinott authored 
      Summary:
1. add call to binarization to complete preprocessing pipeline
2. add ability to specify task to select the dictionary, and add a bert task
3. Get rid of function calls that are no longer needed after moving functions from fairseq here

Reviewed By: jingfeidu

Differential Revision: D13977842

fbshipit-source-id: ec9bbb4e98e62e12c20ba68bb52b8bcc94aee91d
      cea0e4b9
  37. Feb 06, 2019
  39. Feb 05, 2019
  41. Feb 01, 2019
    • Davide Caroselli's avatar
      Support custom Dictionary implementations in 'preprocess.py' (#448) · bbb4120b
      Davide Caroselli authored 
      Summary:
The `preprocess.py` script has been refactored in order to:

1. Use the `options` module for command line arguments  parsing. This will give to `preprocess.py` the ability to load custom modules with `--user-dir` flag (already implemented to all other binaries)
2. Dictionary loading and building code has moved to Task implementation. This allows custom Dictionary classes to be used during the data generation step.
Pull Request resolved: https://github.com/pytorch/fairseq/pull/448

Differential Revision: D13674819

Pulled By: myleott

fbshipit-source-id: b40648a98ed6c08284577e5ec25876e018d8c822
      bbb4120b
  43. Jan 30, 2019
    • Myle Ott's avatar
      Do distributed init after data loading · ec6f8ef9
      Myle Ott authored 
      Summary:
FACEBOOK

This switches back to torch.multiprocessing.spawn, instead of directly calling fb_train.par using a subprocess.Process. This has the advantage that exceptions are propagated properly. It also moves the distributed_init part to happen after data loading, which gets around the timeout issue.

The downside of this approach is that it's not so easy to pipe stdout to multiple places, which was nice when using the sweep.py scripts. I'm still working on a fix for that.

Reviewed By: rutyrinott, ngoyal2707

Differential Revision: D13873224

fbshipit-source-id: 08d593233b8d23590c01c723363630a79804a8b0
      ec6f8ef9
    • Myle Ott's avatar
      Add --input option to interactive.py to support reading from file · 3dce7c9f
      Myle Ott authored 
      Summary: Pull Request resolved: https://github.com/pytorch/fairseq/pull/484

Differential Revision: D13880636

Pulled By: myleott

fbshipit-source-id: 984b2e1c3b281c28243102eb971ea45ec891d94e
      3dce7c9f
    • Myle Ott's avatar
      Merge internal changes (#483) · 42be3ebd
      Myle Ott authored 
      Summary:
Changelog:
- `4889802`: can now remove detokenize sentencepiece output with `--remove-bpe=sentencepiece` (fixes #331). Also added `--sacrebleu` for computing detokenized BLEU.
- `0d76427`: fix assertion error when training language model with dataset containing empty sentences
- minor bug and style fixes
Pull Request resolved: https://github.com/pytorch/fairseq/pull/483

Differential Revision: D13867899

Pulled By: myleott

fbshipit-source-id: 25c940b847fe270262ac8f5ac838407b3977fdda
      42be3ebd