[FIX/ENH] Improve generate_report method to work with fit_transform for the maskers

[ymzayek]

• Deals with most of [ENH] generate_report method should also work after fit_transform was called #3709

Changes proposed in this pull request:

• Pass image data to fit() in fit_transform()
• Use better colormap or instead allow users to pass colormap
• Improve docstrings to explain image data being passed to reporter
• Store 1 time point for reporting using 4D images (will handle in a separate PR)

[github-actions]

👋 @ymzayek Thanks for creating a PR!

Until this PR is ready for review, you can include the [WIP] tag in its title, or leave it as a github draft.

Please make sure it is compliant with our contributing guidelines. In particular, be sure it checks the boxes listed below.

• PR has an interpretable title.
• PR links to Github issue with mention Closes #XXXX (see our documentation on PR structure)
• Code is PEP8-compliant (see our documentation on coding style)
• Changelog or what's new entry in doc/changes/latest.rst (see our documentation on PR structure)

For new features:

• There is at least one unit test per new function / class (see our documentation on testing)
• The new feature is demoed in at least one relevant example.

For bug fixes:

• There is at least one test that would fail under the original bug conditions.

We will review it as quick as possible, feel free to ping us with questions if needed.

[ymzayek]

Now

data = masker.fit_transform(img)
report = masker.generate_report()

works the same as:

masker.fit(img)
report = masker.generate_report()

to generate:

I changed the colormap but maybe there are suggestions for something better??

To fully reproduce:

from nilearn.maskers import NiftiLabelsMasker
from nilearn import datasets

atlas = datasets.fetch_atlas_harvard_oxford("cort-maxprob-thr25-2mm")
atlas_filename, labels = atlas.maps, atlas.labels

dataset = datasets.fetch_development_fmri(n_subjects=1)
fmri_filename = dataset.func[0]
#%%
masker = NiftiLabelsMasker(labels_img=atlas_filename, standardize=True)
masker.fit(fmri_filename)

#%%
masker = NiftiLabelsMasker(labels_img=atlas_filename, standardize=True)
masker.fit_transform(fmri_filename)

[codecov]

Codecov Report

Merging #3897 (7f6e11c) into main (457c79e) will increase coverage by 0.00%.
Report is 1 commits behind head on main.
The diff coverage is 100.00%.

@@           Coverage Diff           @@
##             main    #3897   +/-   ##
=======================================
  Coverage   91.76%   91.77%           
=======================================
  Files         134      134           
  Lines       15747    15750    +3     
  Branches     3283     3283           
=======================================
+ Hits        14451    14454    +3     
  Misses        752      752           
  Partials      544      544           

Flag	Coverage Δ
macos-latest_3.10	91.68% <100.00%> (?)
macos-latest_3.11	91.68% <100.00%> (?)
macos-latest_3.8	91.64% <100.00%> (?)
macos-latest_3.9	91.64% <100.00%> (?)
ubuntu-latest_3.10	91.68% <100.00%> (+<0.01%)	⬆️
ubuntu-latest_3.11	91.68% <100.00%> (+<0.01%)	⬆️
ubuntu-latest_3.8	91.64% <100.00%> (?)
ubuntu-latest_3.9	91.64% <100.00%> (+<0.01%)	⬆️
windows-latest_3.10	91.62% <100.00%> (?)
windows-latest_3.11	91.62% <100.00%> (+<0.01%)	⬆️
windows-latest_3.8	91.59% <100.00%> (+<0.01%)	⬆️
windows-latest_3.9	91.59% <100.00%> (?)

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

Files Changed	Coverage Δ
nilearn/maskers/nifti_labels_masker.py	90.00% <100.00%> (+0.05%)	⬆️
nilearn/maskers/nifti_maps_masker.py	93.92% <100.00%> (+0.03%)	⬆️
nilearn/maskers/nifti_masker.py	86.89% <100.00%> (+0.09%)	⬆️
nilearn/maskers/nifti_spheres_masker.py	93.15% <100.00%> (ø)

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[ymzayek]

Storing one time point for the report seems a bit more involved than I thought given the failures here: https://github.com/nilearn/nilearn/actions/runs/5787425257 so I will handle it in a separate follow-up PR

[bthirion]

The red background is not great for the atlas display. I'll try to get a neutral one.

[bthirion]

LGTM otherwise

[ymzayek]

I changed it back to the default colormap but now allowing the user to pass a custom colormap as a **kwarg that is passed to plot_img.

For reference, the default colormap looks like this in my example (same as in the example gallery)

But I guess it is suboptimal for the example shown in the issue #3709

[bthirion]

LGTM, thx.

[Remi-Gau]

Good with me!