Basic examples fixes

[ddrevicky]

What does this PR do?

• Fixes broken links to basic_examples models in pl_examples/README.md
• Fixes wrong paths to models in basic_examples README (e.g., mnist.py does not exist)
• Adds missing validation and test steps to basic_examples/autoencoder.py
• Uses DATASET_PATHS variable for storing downloaded datasets for examples in files where it was not used

Random question

When defining a validation/test step could the following code be used? Or does calling training_step undesirably modify the model weights + other internal stuff?

# Same for test_step
def validation_step(self, batch, batch_idx):
    loss = self.training_step(batch, batch_idx)
    self.log('valid_loss', loss, on_step=True)

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

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
• Check that target branch and milestone match!

Did you have fun?

Make sure you had fun coding 🙃

[codecov]

Codecov Report

Merging #5912 (a61ca1d) into master (fcfa7fa) will decrease coverage by 59%.
The diff coverage is n/a.

@@           Coverage Diff            @@
##           master   #5912     +/-   ##
========================================
- Coverage      90%     31%    -59%     
========================================
  Files         170     170             
  Lines       11784   11663    -121     
========================================
- Hits        10664    3661   -7003     
- Misses       1120    8002   +6882     

[Borda]

@ddrevicky thank you for sending your PR, just a minor issue coming from our side... we have swapped branches regarding upcoming feat 1.2, mind rebase on actual master? if needed see How to fix PR with mixed base and target branches?

[carmocca]

When defining a validation/test step could the following code be used? Or does calling training_step undesirably modify the model weights + other internal stuff?

Yes. That can be done with no issues

[tchaton]

LGTM !

[tchaton]

There is usually no need to log on step for validation :)