Deprecate stochastic_weight_avg from the Trainer constructor

[ananthsub]

What does this PR do?

Fixes #8971

Does your PR introduce any breaking changes? If yes, please list them.

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 list all the breaking changes introduced by this pull request?
• Did you update the CHANGELOG? (not for typos, docs, test updates, or internal minor changes/refactorings)

PR review

Anyone in the community is welcome to review the PR.
Before you start reviewing make sure you have read Review guidelines. In short, see the following bullet-list:

• 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

Did you have fun?

Make sure you had fun coding 🙃

[codecov]

Codecov Report

Merging #8989 (bbbd171) into master (a4bc0ac) will decrease coverage by 5%.
The diff coverage is 100%.

❗ Current head bbbd171 differs from pull request most recent head 9c77178. Consider uploading reports for the commit 9c77178 to get more accurate results

@@           Coverage Diff            @@
##           master   #8989     +/-   ##
========================================
- Coverage      93%     88%     -5%     
========================================
  Files         178     175      -3     
  Lines       15317   14666    -651     
========================================
- Hits        14211   12889   -1322     
- Misses       1106    1777    +671     

[tchaton]

@williamFalcon @edenlightning Do you validate this depreciation ?

[ananthsub]

@williamFalcon @edenlightning Do you validate this depreciation ?

@tchaton isn't this what the leads group is for? I'd appreciate input from @Borda @ethanwharris @SeanNaren @carmocca @kaushikb11 either here or on #8971 where I've listed the motivation & pitch

[carmocca]

Any deprecations should also remove the associated docs for the deprecated feature. Following/copying a docs snippet to find out it's deprecated is a bad user experience.

For this case, the following would need to be removed:

https://github.com/PyTorchLightning/pytorch-lightning/blame/master/docs/source/advanced/training_tricks.rst#L54-L74

Part of the text there could be merged with the docstring for the callback

[carmocca]

isn't this what the leads group is for?

The leads are meant to push the project's direction and explore new areas of improvement (in addition to the basic contributing work).

To green-light all these deprecations, we need a final decision by our beloved BDFL. Let's continue this discussion in #9006

[tchaton]

@williamFalcon @edenlightning Do you validate this depreciation ?

@tchaton isn't this what the leads group is for? I'd appreciate input from @Borda @ethanwharris @SeanNaren @carmocca @kaushikb11 either here or on #8971 where I've listed the motivation & pitch

This decision was a strong push and final decision from Will. More importantly, I believe @edenlightning which represents the entire community should be present for strong Trainer depreciation.

Personally, I do agree with this change.

Best,
T.C

[Borda]

@ananthsub mind resolve conflicts? :]

[daniellepintz]

Looks great! @ananthsub can we resolve conflicts and merge?