Skip to content

test: regenerate TestLigrec baselines after per-permutation seeding#1240

Merged
timtreis merged 1 commit into
mainfrom
fix/ligrec-baselines-seedseq
Jul 7, 2026
Merged

test: regenerate TestLigrec baselines after per-permutation seeding#1240
timtreis merged 1 commit into
mainfrom
fix/ligrec-baselines-seedseq

Conversation

@timtreis

@timtreis timtreis commented Jul 7, 2026

Copy link
Copy Markdown
Member

What

Regenerate the three TestLigrec plotting baselines.

Why

#1234 changed permutation tests to spawn per-permutation seeds from a numpy.random.SeedSequence (results now independent of n_jobs/backend, but different for a given seed). The ligrec dot-plot fixtures therefore select a different set of significant interactions, changing the figure dimensions. The stale baselines failed with:

ImageComparisonFailure: Image sizes do not match
  expected (166,237,3) actual (166,252,3)   # test_plot_pvalue_threshold
  expected (390,257,3) actual (390,193,3)   # test_plot_dendrogram_clusters
  expected (152,134,3) actual (166,193,3)   # test_plot_remove_nonsig_interactions

Reproduces on Linux and macOS (data change, not a rendering one), which is why it surfaced in daily integration testing (#1237/#1238/#1239) and on main CI starting at the #1234 commit.

How

Baselines pulled from the failing run's visual_test_results_Linux figure artifacts. The 3 tests pass locally against the new images.

Closes #1237, closes #1238, closes #1239.

…1234)

#1234 switched permutation tests to per-permutation seeds spawned from a
numpy.random.SeedSequence, which changes ligrec results for a given seed.
The TestLigrec dot-plot fixtures now yield a different set of significant
interactions, so the rendered figure dimensions changed and the stale
baselines failed with ImageComparisonFailure (size mismatch). Baselines
regenerated from CI figure artifacts to match the new results.
@codecov

codecov Bot commented Jul 7, 2026

Copy link
Copy Markdown

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 77.08%. Comparing base (005c905) to head (bf60f47).
⚠️ Report is 2 commits behind head on main.

Additional details and impacted files
@@            Coverage Diff             @@
##             main    #1240      +/-   ##
==========================================
+ Coverage   76.81%   77.08%   +0.26%     
==========================================
  Files          63       63              
  Lines        9269     9368      +99     
  Branches     1565     1576      +11     
==========================================
+ Hits         7120     7221     +101     
+ Misses       1547     1545       -2     
  Partials      602      602              

see 2 files with indirect coverage changes

🚀 New features to boost your workflow:
  • ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

@timtreis timtreis changed the title test: regenerate TestLigrec baselines after per-permutation seeding (#1234) test: regenerate TestLigrec baselines after per-permutation seeding Jul 7, 2026
@timtreis timtreis requested a review from selmanozleyen July 7, 2026 11:47
@timtreis timtreis merged commit ac08971 into main Jul 7, 2026
15 checks passed
@timtreis timtreis deleted the fix/ligrec-baselines-seedseq branch July 7, 2026 11:51
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

2 participants