[FIX] threshold images right before ploting in GLM reports

[Remi-Gau]

• Closes [BUG] Glm reports does not apply image threshold at the right place #4192

Changes proposed in this pull request:

• pass threshold to _stat_map_to_svg

[github-actions]

👋 @Remi-Gau 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.

[Remi-Gau]

before:

after:

Also adding all before / after reports from the code snippet from #4192

report.zip

[Remi-Gau]

@bthirion

I feel we should have a "test" for this in the visual inspection suite of the GLM reporter.

https://github.com/nilearn/nilearn/blob/310f5ca659eb135b0c9489199b0c0cc9895780fa/nilearn/reporting/_visual_testing/_glm_reporter_visual_inspection_suite_.py

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Comparison is base (abb80ff) 91.85% compared to head (21dc748) 91.98%.
Report is 6 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #4258      +/-   ##
==========================================
+ Coverage   91.85%   91.98%   +0.12%     
==========================================
  Files         144      144              
  Lines       16419    16421       +2     
  Branches     3434     3435       +1     
==========================================
+ Hits        15082    15105      +23     
+ Misses        792      776      -16     
+ Partials      545      540       -5     

Flag	Coverage Δ
macos-latest_3.10_test_plotting	91.85% <100.00%> (?)
macos-latest_3.11_test_plotting	91.85% <100.00%> (+<0.01%)	⬆️
macos-latest_3.12_test_plotting	91.85% <100.00%> (?)
macos-latest_3.8_test_plotting	91.81% <100.00%> (?)
macos-latest_3.9_test_plotting	91.82% <100.00%> (?)
ubuntu-latest_3.10_test_plotting	91.85% <100.00%> (+<0.01%)	⬆️
ubuntu-latest_3.11_test_plotting	91.85% <100.00%> (?)
ubuntu-latest_3.12_test_plotting	91.85% <100.00%> (?)
ubuntu-latest_3.12_test_pre	91.85% <100.00%> (?)
ubuntu-latest_3.8_test_plot_min	91.56% <100.00%> (?)
ubuntu-latest_3.8_test_plotting	91.81% <100.00%> (?)
ubuntu-latest_3.9_test_plotting	91.82% <100.00%> (?)
windows-latest_3.10_test_plotting	91.83% <100.00%> (?)
windows-latest_3.11_test_plotting	91.83% <100.00%> (?)
windows-latest_3.12_test_plotting	91.83% <100.00%> (?)
windows-latest_3.8_test_plotting	91.79% <100.00%> (?)
windows-latest_3.9_test_plotting	91.80% <100.00%> (?)

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

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

[bthirion]

@bthirion

I feel we should have a "test" for this in the visual inspection suite of the GLM reporter.

https://github.com/nilearn/nilearn/blob/310f5ca659eb135b0c9489199b0c0cc9895780fa/nilearn/reporting/_visual_testing/_glm_reporter_visual_inspection_suite_.py

This means that we just need to look at the results from time to time ? Before releases ? Then we should add that to the dev doc.

[Remi-Gau]

This means that we just need to look at the results from time to time ? Before releases ? Then we should add that to the dev doc.

That would be the "generate once in a while / manual check".

Other alternative "generate automatically / manual check": would be the actually show what the reports look like as part of our doc: see this comment in another thread #2194 (comment)

In any case this would be for another PR.

[bthirion]

Actually you cannot do that: you need to get the thresholdedimage because you want to take into account the cluster_threshold that is provided. In the implementation you propose, this gets forgotten.

[Remi-Gau]

report 1 after:

report 2 after:

report_after.zip

[Remi-Gau]

unrelated but realized that the cut_coords was not passed to the mask plotting

[bthirion]

Seems to do the job properly, thx.

[Remi-Gau]

Will update the changelog to log the bug fix