Repository navigation
Fix flaky test_wins by seeding DDMC and GMM - #574
Merged
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
CI failures on #571 and #572, and an earlier run on the expand-kinome-pspls branch, all came from
test_cluster.py::test_wins. It fits an unseededDDMCand an unseededGaussianMixtureand requires their cluster centers to match with cosine similarity > 0.95. Across 30 seeds, 2 (about 7%) fail because one model lands in a poor local optimum. That also explains why push runs pass while PR runs on the same code fail.This PR sets
random_state=0on both models, which is a passing seed, so the test is deterministic. Only the test changes. Ruff lint and format are clean, andtest_cluster.pypasses locally (30 passed). No other failures turned up in the recent runs.