Skip to content

GitLab

Switch branch/tag
  1. 25 Apr, 2019 1 commit
  3. 22 Apr, 2019 1 commit
  5. 17 Apr, 2019 3 commits
  7. 16 Apr, 2019 1 commit
  9. 15 Apr, 2019 2 commits
  11. 12 Apr, 2019 1 commit
  13. 10 Apr, 2019 4 commits
  15. 09 Apr, 2019 1 commit
  17. 05 Apr, 2019 3 commits
  19. 04 Apr, 2019 1 commit
    • Jay Mahadeokar's avatar
      aligned training task and CE related changes · 3658fa32
      Jay Mahadeokar authored 
      Summary:
This diff adds:

1. Aligned training task specifically for doing cross entropy criterion training using prod data and prod like models
2. Few changes to correctly register the task and criterions.
3. Changes to trainer code for propogating accuracy metrics which we care about for training.

Couple of things are hacky right now:
- The reporting is not modular (this needs to be thought about in general for fairseq).

- The get dummy batch could be specific to task instead of specific for dataset.

Reviewed By: myleott

Differential Revision: D14670482

fbshipit-source-id: dc077247b2ae9d26a8e842a386ec5faa5771e836
      3658fa32
  21. 03 Apr, 2019 2 commits
  23. 02 Apr, 2019 1 commit
  25. 29 Mar, 2019 2 commits
    • Myle Ott's avatar
      Add utils.deprecation_warning · a78ad1ac
      Myle Ott authored 
      Summary: Pull Request resolved: https://github.com/pytorch/fairseq/pull/607

Differential Revision: D14681031

Pulled By: myleott

fbshipit-source-id: 466ee526a30543218e2b7138fb651db866ae5ab3
      a78ad1ac
    • Felix Wu's avatar
      Fixing a bug of DynamicConv in the unfolding mode (#593) · 34c9ebf0
      Felix Wu authored 
      Summary:
The unfold1d.py has the same name as the function `unfold1d` function, which will cause an error when using DynamicConv1dTBC with `unfold=True`.
This doesn't affect the NMT models which don't use the unfolding mode though.

I rename `unfold1d.py` as `unfold.py` to fix this bug.

Originally we would get `TypeError` when running this code:
```
import torch
from fairseq.modules import LightweightConv1dTBC, DynamicConv1dTBC

x = torch.rand(4, 10, 8)
m = LightweightConv1dTBC(8, 4, 3)
o = m(x, unfold=True)

m = DynamicConv1dTBC(8, 4, 3)
o = m(x, unfold=True)
```
Pull Request resolved: https://github.com/pytorch/fairseq/pull/593

Differential Revision: D14597117

Pulled By: myleott

fbshipit-source-id: 59752fd7ff62c53a4aba8b56b83155291e5f5792
      34c9ebf0
  27. 26 Mar, 2019 1 commit
  29. 15 Mar, 2019 1 commit
    • 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
      e6422528
  31. 14 Mar, 2019 1 commit
  33. 13 Mar, 2019 1 commit
  35. 12 Mar, 2019 3 commits
    • 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
  37. 04 Mar, 2019 2 commits
  39. 02 Mar, 2019 1 commit
  41. 01 Mar, 2019 2 commits
    • James King's avatar
      Fixed the issue that no space in string converted from tensor · 88bf8b56
      James King authored 
      Summary: Pull Request resolved: https://github.com/pytorch/fairseq/pull/548

Differential Revision: D14286021

Pulled By: myleott

fbshipit-source-id: 7c725304185e63787220371a812ec860e178872c
      88bf8b56
    • Kartikay Khandelwal's avatar
      Refactor BERTDataset to the more general MaskedLMDataset · 92a6c548
      Kartikay Khandelwal authored 
      Summary: The current BERTDataset has a lot of components needed for generic MaskedLM training but is too restrictive in terms of the assumptions it makes - two blocks being masked, the special tokens used for the sentence embedding as well as the separator etc. In this diff I refactor this dataset and at the same time add make some of the parameters including the probabilities associated with masking configurable.

Reviewed By: rutyrinott

Differential Revision: D14222467

fbshipit-source-id: e9f78788dfe7f56646ba09c62967c4c0bd30aed8
      92a6c548
  43. 28 Feb, 2019 2 commits
  45. 26 Feb, 2019 3 commits