[feature] Add StochasticWeightAveraging (SWA) callback

[tchaton]

What does this PR do?

This PR adds SWA Callback to Lightning.
https://pytorch.org/blog/pytorch-1.6-now-includes-stochastic-weight-averaging/

Fixes #1894

TODO:

• Docs issue: https://github.com/PyTorchLightning/pytorch-lightning/pull/5640/checks?check_run_id=1839752310 edit: not related to SWA but CI.

Before submitting

• Was this discussed/approved via a GitHub issue? (not for typos and docs)
• Did you read the contributor guideline, Pull Request section?
• Did you make sure your PR does only one thing, instead of bundling different changes together?
• Did you make sure to update the documentation with your changes? (if necessary)
• Did you write any new necessary tests? (not for typos and docs)
• Did you verify new and existing tests pass locally with your changes?
• Did you update the CHANGELOG? (not for typos, docs, test updates, or internal minor changes/refactorings)

PR review

• Is this pull request ready for review? (if not, please submit in draft mode)
• Check that all items from Before submitting are resolved
• Make sure the title is self-explanatory and the description concisely explains the PR
• Add labels and milestones (and optionally projects) to the PR so it can be classified
• Check that target branch and milestone match!

[pep8speaks]

Hello @tchaton! Thanks for updating this PR.

• In the file pytorch_lightning/callbacks/swa.py:

Line 100:13: W503 line break before binary operator

Comment last updated at 2021-02-10 23:45:21 UTC

[SkafteNicki]

LGTM :]

[Borda]

can we add typing?