Discussion #489: add CopyAffine transform

[albansteff]

Fixes #489 .

Description
Add a CopyAffine transform with a target_key parameter to copy the affine from this key to all LabelMap objects of a given torchio subject.

Checklist

• I have read the CONTRIBUTING docs and have a developer setup (especially important are pre-commitand pytest)
• 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
• In-line docstrings updated
• Documentation updated, tested running make html inside the docs/ folder
• 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

[romainVala]

merci alban

[fepegar]

Thanks for your PR! I added a comment. Could you please update the corresponding rst file in the documentation?

[codecov]

Codecov Report

Merging #584 (73cc955) into master (213fff3) will decrease coverage by 0.04%.
The diff coverage is 82.00%.

@@            Coverage Diff             @@
##           master     #584      +/-   ##
==========================================
- Coverage   96.83%   96.79%   -0.05%     
==========================================
  Files         122      124       +2     
  Lines        6358     6401      +43     
==========================================
+ Hits         6157     6196      +39     
- Misses        201      205       +4     

Impacted Files	Coverage Δ
print_system.py	0.00% <0.00%> (ø)
tests/transforms/test_transforms.py	100.00% <ø> (ø)
torchio/data/image.py	89.93% <ø> (ø)
torchio/transforms/data_parser.py	92.92% <ø> (ø)
tests/transforms/preprocessing/test_copy_affine.py	100.00% <100.00%> (ø)
torchio/__init__.py	100.00% <100.00%> (ø)
torchio/reference.py	100.00% <100.00%> (ø)
torchio/transforms/__init__.py	100.00% <100.00%> (ø)
torchio/transforms/preprocessing/__init__.py	100.00% <100.00%> (ø)
...io/transforms/preprocessing/spatial/copy_affine.py	100.00% <100.00%> (ø)
... and 2 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 3bccf13...73cc955. Read the comment docs.

[romainVala]

Hi, I am not sure to understand what the error is,
should we change things ... ?

[fepegar]

Hi, I am not sure to understand what the error is,
should we change things ... ?

I think the error is happening because of a Sphinx problem that was fixed in 6e61447, before this branch was created. Don't worry about it. I'll take a look at this PR tonight or tomorrow.

[fepegar]

Thanks for your contribution, @Albans98!

@all-contributors please add @Albans98 for code

[allcontributors]

@fepegar

I've put up a pull request to add @Albans98! 🎉

[albansteff]

Thanks for your time and your reviews !

[romainVala]

@Albans98 just find out, there is one line missing at the end of the init

        self.args_names = ['target']

[fepegar]

That's my fault, I removed it thinking it wasn't necessary. It's needed to reproduce the transform, right?

[romainVala]

may be @Albans98 can confirm, I don't know what the exact problem was,
In our use case, we do save the transform history (to be able to reproduce later)

but anyway It make sens to have it for all the transform (preprocessing and augmentation)
(I quickly check, and there may be some other missing like OneHot)

[albansteff]

Hello, yes the problem is when we try to save the history of all applied transforms, we get an error 'TypeError: init() missing 1 required positional argument: 'target''.

Adding the target in the list of args makes it possible to trace back to the 'target' value.

[fepegar]

Ok, let me fix that.

[fepegar]

Done. Thanks for pointing that out!

[albansteff]

Thank you for fixing this !