Fix HistogramStandardization example

[vedal]

Fixes #{issue_number}.

Description
Currently the example isn't loading the landmarks from path. This PR fixes that.

Checklist

• [v] I have read the CONTRIBUTING docs and have a developer setup (especially important are pre-commitand pytest)
• [v] Non-breaking change (would not break existing functionality)
• Breaking change (would cause existing functionality to change)
• Tests added or modified to cover the changes
• Integration tests passed locally by running pytest
• [v] In-line docstrings updated
• Documentation updated, tested running make html inside the docs/ folder
• [v] This pull request is ready to be reviewed
• If the PR is ready and there are multiple commits, I have squashed them and force-pushed

[fepegar]

Hi, @vedal. Please read the contributing guide as requested in the PR template.

According to the documentation, this transform supports paths to NumPy arrays, so this change is not needed.

[vedal]

Hi @fepegar, thanks for reviewing this. I think particularly the change to add a torch.load makes the example more readable, since the variable name "t1_landmarks" suggests its not a path (such as "t1_landmarks_path" used previously).
As such, the change might not be necessary but at least increase readiblity.

Also, its unusual to store .npy with torch.save, as done if running the example once.

[fepegar]

Thanks for clarifying, that makes sense. I'll take a look later.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 86.59%. Comparing base (aa4d609) to head (2af647f).
Report is 102 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1022      +/-   ##
==========================================
+ Coverage   86.47%   86.59%   +0.11%     
==========================================
  Files          91       91              
  Lines        5774     5787      +13     
==========================================
+ Hits         4993     5011      +18     
+ Misses        781      776       -5     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[fepegar]

I think the idea is that the list of landmarks is saved an an npy file, and pth would be used for a pickled dict. So I think the change I would do is replace torch.save(t1_landmarks, t1_landmarks_path) with np.save(t1_landmarks_path, t1_landmarks). What do you think?

[fepegar]

@all-contributors please add @vedal for doc

[allcontributors]

@fepegar

I've put up a pull request to add @vedal! 🎉