TST: stats: bump test tolerance in TestMultivariateNormal.test_broadcasting #7712
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
josef-pkt
reviewed
Aug 7, 2017
| @@ -216,7 +216,7 @@ def test_broadcasting(self): | |||
| # Repeat for cdf | |||
| actual = multivariate_normal.cdf(X[i, j], mean, cov) | |||
| desired = multivariate_normal.cdf(X, mean, cov)[i, j] | |||
There was a problem hiding this comment.
small optimization
AFAICS desired arrays only need to be computed once outside of the loop.
As a broadcasting test it might only need some "nice" x, in case there are values that are difficult to compute or have tiny cdf values.
There was a problem hiding this comment.
I'm not sure if such nice x exists (and won't be doing the work to find it for this PR) --- the computation appears to use a randomized algorithm.
…asting The multivariate computations use a randomized algorithm, so the outputs vary. The test with the previous tolerance appears to fail on all platforms if iterated several times -- and also failed on windows on the first round. Bump the tolerance to a value where the failures become less likely. Also move comparison values out from the loop.
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Outdated suggestions cannot be applied.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.
The multivariate computations use a randomized algorithm, so the outputs
vary. The test with the previous tolerance appears to fail on all
platforms if iterated several times -- and also failed on windows on the
first round. Bump the tolerance to a value where the failures become
less likely.
This test has failed sporadically on Travis and Appveyor several times now,
and should be fixed asap.
@asnelt: ping