Add Type Annotations to pl_bolts/utils/*

[hassiahk]

What does this PR do?

Part of #434. This PR adds type annotations to pl_bolts.utils.*

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

• Is this pull request ready for review? (if not, please submit in draft mode)

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 🙃

[codecov]

Codecov Report

Merging #455 (4510e1f) into master (a6a395f) will decrease coverage by 55.84%.
The diff coverage is 66.66%.

@@             Coverage Diff             @@
##           master     #455       +/-   ##
===========================================
- Coverage   80.91%   25.07%   -55.85%     
===========================================
  Files         101      101               
  Lines        5676     5671        -5     
===========================================
- Hits         4593     1422     -3171     
- Misses       1083     4249     +3166     

Flag	Coverage Δ
cpu	25.07% <66.66%> (+0.12%)	⬆️
pytest	25.07% <66.66%> (+0.12%)	⬆️
unittests	?

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
pl_bolts/utils/arguments.py	20.77% <41.66%> (-75.38%)	⬇️
pl_bolts/utils/semi_supervised.py	21.87% <50.00%> (-74.90%)	⬇️
pl_bolts/utils/__init__.py	100.00% <100.00%> (ø)
pl_bolts/utils/pretrained_weights.py	100.00% <100.00%> (ø)
pl_bolts/utils/self_supervised.py	100.00% <100.00%> (ø)
pl_bolts/utils/shaping.py	36.36% <100.00%> (+6.36%)	⬆️
pl_bolts/utils/warnings.py	100.00% <100.00%> (ø)
pl_bolts/models/rl/common/agents.py	0.00% <0.00%> (-100.00%)	⬇️
pl_bolts/models/rl/dueling_dqn_model.py	0.00% <0.00%> (-100.00%)	⬇️
...l_bolts/models/rl/vanilla_policy_gradient_model.py	0.00% <0.00%> (-96.37%)	⬇️
... and 63 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update a6a395f...4510e1f. Read the comment docs.

[hassiahk]

@akihironitta, need help with the below two errors. I am not sure how to fix them.

• Cannot access "__init__" directly
• Module has no attribute "_empty"

[akihironitta]

@hassiahk Thank you for your contribution! I left some comments on the changes. Could you have a look? As for your comment below, I'll look into it later...

@akihironitta, need help with the below two errors. I am not sure how to fix them.

• Cannot access "__init__" directly
• Module has no attribute "_empty"

[hassiahk]

@akihironitta, I took care of your requested changes but I also had to change Y -> labels in balance_classes while annotating it as Sequence[int] or List[int] because mypy is throwing below errors:

• Argument 1 to "len" has incompatible type "int"; expected "Sized" -> mypy is basically saying that you did not take care of Y being None.
• Value of type "int" is not indexable -> Not sure of this.

Another workaround would be to annotate Y as plain Sequence or List. Let me know your thoughts.

[akihironitta]

@hassiahk Thank you very much for your work!

For Module has no attribute "_empty", let's replace inspect._empty with inspect.Parameter.empty because both should be equivalent which is defined in inspect. The same issue was also discussed in allenai/allennlp#1191 (comment).

For Cannot access "__init__" directly, I couldn't find any existing solution, so shall we just skip the line with # type: ignore...?

[hassiahk]

@akihironitta, made requested changes. Kindly review it once again. Annotated X as Union[Tensor, np.ndarray] and Y as Union[Tensor, np.ndarray, Sequence[int]].

[akihironitta]

I'm still concerned about if the annotated types include all the possibility, but I think these are fine for now.

[hassiahk]

I will use this balance_classes and see if we can use other types as well.

[akihironitta]

@hassiahk Thank you for your work! LGTM :]

[akihironitta]

Since these variables are the output of the function, I think they should match with the return type of the function balance_classes():

) -> Tuple[np.ndarray, np.ndarray]:

but in order to do that, you might need to change/define other variables, too. We can tighten these types later in another PR, but it'd be nice to do it in this PR. Do you think you can improve typing here?

[hassiahk]

Sure, I will see what I can do. :)