Refactor pl_bolts.datamodules and pl_bolts.datasets

[akihironitta]

What does this PR do?

Fixes #337.

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 🙃

[pep8speaks]

Hello @akihironitta! Thanks for updating this PR.

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

Comment last updated at 2020-12-14 08:10:05 UTC

[codecov]

Codecov Report

Merging #338 (fc579ea) into master (7c2e651) will decrease coverage by 0.31%.
The diff coverage is 89.28%.

@@            Coverage Diff             @@
##           master     #338      +/-   ##
==========================================
- Coverage   81.09%   80.77%   -0.32%     
==========================================
  Files         100      101       +1     
  Lines        5722     5706      -16     
==========================================
- Hits         4640     4609      -31     
- Misses       1082     1097      +15     

Flag	Coverage Δ
cpu	25.20% <85.71%> (-0.04%)	⬇️
pytest	25.20% <85.71%> (-0.04%)	⬇️
unittests	80.05% <75.00%> (-0.41%)	⬇️

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

Impacted Files	Coverage Δ
pl_bolts/datasets/kitti_dataset.py	34.61% <50.00%> (-0.68%)	⬇️
pl_bolts/datasets/ssl_amdim_datasets.py	74.32% <50.00%> (-0.68%)	⬇️
pl_bolts/datasets/imagenet_dataset.py	20.11% <66.66%> (+0.70%)	⬆️
pl_bolts/datamodules/__init__.py	100.00% <100.00%> (+22.64%)	⬆️
pl_bolts/datamodules/experience_source.py	95.93% <100.00%> (+0.06%)	⬆️
pl_bolts/datasets/cifar10_dataset.py	96.77% <100.00%> (-1.10%)	⬇️
pl_bolts/datasets/concat_dataset.py	41.66% <100.00%> (ø)
pl_bolts/datasets/dummy_dataset.py	100.00% <100.00%> (ø)
pl_bolts/datasets/mnist_dataset.py	57.14% <100.00%> (+14.28%)	⬆️
pl_bolts/utils/__init__.py	100.00% <100.00%> (ø)
... and 10 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 7c2e651...fc579ea. Read the comment docs.

[akihironitta]

Do we want to keep warn()? Without torchvision and PIL installed, import pl_bolts gives so many warnings and is very noisy... Maybe handle this in another PR?

noisy warnings here...

python -c "import pl_bolts"
/home/nitta/Dropbox/home/src/projects/pytorch-lightning-bolts/pl_bolts/models/mnist_module.py:14: UserWarning: You want to use `torchvision` which is not installed yet, install it with `pip install torchvision`.
  warn('You want to use `torchvision` which is not installed yet,'  # pragma: no-cover
/home/nitta/Dropbox/home/src/projects/pytorch-lightning-bolts/pl_bolts/callbacks/variational.py:10: UserWarning: You want to use `torchvision` which is not installed yet, install it with `pip install torchvision`.
  warn('You want to use `torchvision` which is not installed yet,'  # pragma: no-cover
/home/nitta/Dropbox/home/src/projects/pytorch-lightning-bolts/pl_bolts/callbacks/vision/image_generation.py:9: UserWarning: You want to use `torchvision` which is not installed yet, install it with `pip install torchvision`.
  warn('You want to use `torchvision` which is not installed yet,'  # pragma: no-cover
/home/nitta/Dropbox/home/src/projects/pytorch-lightning-bolts/pl_bolts/datamodules/binary_mnist_datamodule.py:12: UserWarning: You want to use `torchvision` which is not installed yet, install it with `pip install torchvision`.
  warn('You want to use `torchvision` which is not installed yet,'  # pragma: no-cover
/home/nitta/Dropbox/home/src/projects/pytorch-lightning-bolts/pl_bolts/datasets/cifar10_dataset.py:13: UserWarning: You want to use `Pillow` which is not installed yet, install it with `pip install Pillow`.
  warn('You want to use `Pillow` which is not installed yet,'  # pragma: no-cover
/home/nitta/Dropbox/home/src/projects/pytorch-lightning-bolts/pl_bolts/transforms/dataset_normalizations.py:5: UserWarning: You want to use `torchvision` which is not installed yet, install it with `pip install torchvision`.
  warn('You want to use `torchvision` which is not installed yet,'  # pragma: no-cover
/home/nitta/Dropbox/home/src/projects/pytorch-lightning-bolts/pl_bolts/datamodules/cifar10_datamodule.py:17: UserWarning: You want to use `torchvision` which is not installed yet, install it with `pip install torchvision`.
  warn('You want to use `torchvision` which is not installed yet,'  # pragma: no-cover
/home/nitta/Dropbox/home/src/projects/pytorch-lightning-bolts/pl_bolts/datamodules/fashion_mnist_datamodule.py:11: UserWarning: You want to use `torchvision` which is not installed yet, install it with `pip install torchvision`.
  warn('You want to use `torchvision` which is not installed yet,'  # pragma: no-cover
/home/nitta/Dropbox/home/src/projects/pytorch-lightning-bolts/pl_bolts/datamodules/imagenet_datamodule.py:13: UserWarning: You want to use `torchvision` which is not installed yet, install it with `pip install torchvision`.
  warn('You want to use `torchvision` which is not installed yet,'  # pragma: no-cover
/home/nitta/Dropbox/home/src/projects/pytorch-lightning-bolts/pl_bolts/datamodules/imagenet_datamodule.py:22: UserWarning: You want to use `torchvision` which is not installed yet, install it with `pip install torchvision`.
  warn('You want to use `torchvision` which is not installed yet,'  # pragma: no-cover
/home/nitta/Dropbox/home/src/projects/pytorch-lightning-bolts/pl_bolts/datamodules/mnist_datamodule.py:11: UserWarning: You want to use `torchvision` which is not installed yet, install it with `pip install torchvision`.
  warn('You want to use `torchvision` which is not installed yet,'  # pragma: no-cover
/home/nitta/Dropbox/home/src/projects/pytorch-lightning-bolts/pl_bolts/datamodules/stl10_datamodule.py:16: UserWarning: You want to use `torchvision` which is not installed yet, install it with `pip install torchvision`.
  warn('You want to use `torchvision` which is not installed yet,'  # pragma: no-cover
/home/nitta/Dropbox/home/src/projects/pytorch-lightning-bolts/pl_bolts/datamodules/cityscapes_datamodule.py:11: UserWarning: You want to use `torchvision` which is not installed yet, install it with `pip install torchvision`.
  warn('You want to use `torchvision` which is not installed yet,'  # pragma: no-cover

[Borda]

@akihironitta I think we can add global var to the package which would "count" if the user already warned about the one error... >> warn user just one per load/usage Bolts

Alternatively, add a custom warning function which would keep the count inside

WARNINGS = {}

def warn_missing_lib(pkg_name: str, extra_text: str = None)
  global WARNINGS
  if pkg_name not in WARNINGS:
    # do the warning
  else:
    WARNINGS[pkg_name] = WARNINGS.get(pkg_name, 0) + 1

lets see #341

[annikabrundyn]

Please look at some of the comments left :)

[akihironitta]

@Borda @annikabrundyn Once this PR is merged, I'll quickly submit PRs to update try: ... except ModuleNotFoundError: ... in other submouldes as well!

[Borda]

nice, I will update #448 accordingly...