[Metrics] R2Score

[SkafteNicki]

What does this PR do?

Adds R2Score metric similar to sklearns (https://scikit-learn.org/stable/modules/generated/sklearn.metrics.r2_score.html)

Before submitting

• Was this discussed/approved via a Github issue? (no need for typos and docs improvements)
• 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? Otherwise, we ask you to create a separate PR for every change.
• Did you make sure to update the documentation with your changes?
• Did you write any new necessary tests?
• Did you verify new and existing tests pass locally with your changes?
• If you made a notable change (that affects users), did you update the CHANGELOG?

PR review

Anyone in the community is free to review the PR once the tests have passed.
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; Bugfixes should be including in bug-fix release milestones (m.f.X) and features should be included in (m.X.b) releases.

Did you have fun?

Make sure you had fun coding 🙃

[pep8speaks]

Hello @SkafteNicki! Thanks for updating this PR.

There are currently no PEP 8 issues detected in this Pull Request. Cheers! 🍻

Comment last updated at 2020-12-31 14:28:20 UTC

[rohitgr7]

NICE!!
how about adding adjusted r2 too? or maybe extending it with additional arguments required for adjusted r2. Won't be so trivial I guess.

[teddykoker]

LGTM :)

[codecov]

Codecov Report

Merging #5241 (0218f5d) into release/1.2-dev (73e06fd) will increase coverage by 0%.
The diff coverage is 94%.

@@               Coverage Diff               @@
##           release/1.2-dev   #5241   +/-   ##
===============================================
  Coverage               93%     93%           
===============================================
  Files                  144     146    +2     
  Lines                10146   10211   +65     
===============================================
+ Hits                  9425    9486   +61     
- Misses                 721     725    +4     

[gianscarpe]

Just a suggestion, let me know :)

[SkafteNicki]

@rohitgr7 good point about adjusted r2 score.
I have added an adjusted arg that corresponds to the number of independent regressors that the user needs to provide. If provided the metric returns the adjusted score.

[gianscarpe]

lgtm! :)

[rohitgr7]

Suggested change

	of the following strings (default is `'uniform_average'`.):
	of the following strings (default is ``'uniform_average'``.):

[rohitgr7]

should defaults be removed from the docs??

[bibinwils]

https://forums.pytorchlightning.ai/t/to-find-r2score-of-my-model/764 could you please, anyone help in this?
I have a UNet model. I’m trying for a regression model since, in my output, I have different floating values for each pixel. In order to check the r2score, I tried to put the below code in the ‘model class’, training_step, validation_step, and test_step.

r2 = r2score(logits, y)
self.log('r2:',r2)

But it’s giving the following error

ValueError: Expected both prediction and target to be 1D or 2D tensors, but recevied tensors with dimension torch.Size([50, 1, 32, 32])
How can I check my model fit?