Summary:
Pull Request resolved: https://github.com/fairinternal/fairseq-py/pull/1487
Here's the code for CosineLRScheduler that I used as a reference: 577e4fa78a/fairseq/optim/lr_scheduler/cosine_lr_schedul
In the reference:
- `warmup_init_lr` defaults to `args.lr[0]`
- `warmup_end_lr` defaults to `args.max_lr`
- `min_lr` defaults to `args.lr[0]` (note that there's also a `args.min_lr` option defined in the global fairseq config, but this is unused by the cosine scheduler)
- `max_lr` is a required option
This diff removes `max_lr` and replaces it with `lr[0]` to be more
consistent with other LR schedulers. We then add an explicit `min_lr`
option to the Config.
Test Plan: Imported from OSS
Reviewed By: alexeib
Differential Revision: D25342180
Pulled By: myleott
fbshipit-source-id: 61281666e68839da8efc4714c2ce8c49dc4c8e6e
Summary:
# Before submitting
- [ ] Was this discussed/approved via a Github issue? (no need for typos, doc improvements)
- [x] Did you read the [contributor guideline](https://github.com/pytorch/fairseq/blob/master/CONTRIBUTING.md)?
- [x] Did you make sure to update the docs?
- [ ] Did you write any new necessary tests?
## What does this PR do?
Update WikiText-103 url
## PR review
Anyone in the community is free to review the PR once the tests have passed.
If we didn't discuss your PR in Github issues there's a high chance it will not be merged.
## Did you have fun?
Make sure you had fun coding �
Pull Request resolved: https://github.com/pytorch/fairseq/pull/2871
Reviewed By: myleott
Differential Revision: D24835953
Pulled By: alexeib
fbshipit-source-id: 890e911d528c04de0dc056e55866afb46a2bd87f
Summary:
now that we are moving to using dataclasses to define fairseq configuration, having aliases for options is no longer practical. this pr removes "max-sentences" argument while keeping its alias "batch-size", which is more appropriate
Pull Request resolved: https://github.com/fairinternal/fairseq-py/pull/1333
Reviewed By: shruti-bh
Differential Revision: D24121305
Pulled By: alexeib
fbshipit-source-id: 34343cea54c8f2c8b059c38ef9f29b66e76df9fb
Summary:
Very minor update, just changing the url. The old one is 404.
Pull Request resolved: https://github.com/pytorch/fairseq/pull/1847
Reviewed By: ngoyal2707
Differential Revision: D20490970
Pulled By: myleott
fbshipit-source-id: 9c2fbe97e3dd450094a95db9f934945e87637ae1
Summary:
Adding a backslash in the convolutional language model training usage.
Pull Request resolved: https://github.com/pytorch/fairseq/pull/941
Differential Revision: D16581388
Pulled By: myleott
fbshipit-source-id: 7e2e05ecf13e86cb844dc5200d49f560c63b12ff
Summary:
Hi,
currently, the link to the language model readme is broken on the `examples/language_model/transformer_lm` page.
This PR fixes the link :)
Pull Request resolved: https://github.com/pytorch/fairseq/pull/600
Differential Revision: D14680985
Pulled By: myleott
fbshipit-source-id: 62291efbf4ece2af54fae45c408c2759863f9847
A Task defines the data format, stores shared state (e.g., dictionaries) and provides helpers for building the model/criterion and calculating the loss.
Changes:
- Add TranslationTask and LanguageModelingTask. New tasks can be registered with @register_task decorator.
- Add EpochBatchIterator to encapsulate batching and saving/restoring dataloader position
- Remove LEFT_PAD_* constants and make them configurable per task
