Fix test bug where size-1 sample may be used (with top_k=2) #46
Conversation
This file contains 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
This temporarily modifies test_predict so it produces the problem in every test run, in order to verify that the immediately forthcoming fix is effective. Because this commit decreases the general usefulness of the test, it needs to be undone/reverted after that has been verified.
This puts the upper bound for the randomly determined sample size back to 50. The immediately preceding commits verified that the bug can be fixed by a change in the lower bound, but we don't always want to use a lower bound of 2! It may be best to choose the sample size non-randomly, perhaps testing multiple sample sizes per run using parametrization or subtests. But that is more extensive than the small change needed for bazingagin#45, and thus beyond the scope here.
|
I think this situation will happen whenever the number of samples is smaller than |
# for free
to join this conversation on GitHub.
Already have an account?
# 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Outdated suggestions cannot be applied.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.
Closes #45
This changes the range of possible sample sizes in
TestKnnClassifier.test_predictto have a lower bound of2instead of1, so that the test, which always uses atop_kof2, no longer occasionally fails.The change is actually very simple, but I did it in a few commits to verify and show that a lower bound of
1was the problem. The sample size is being selected randomly, so it would otherwise not be immediately clear that this change really fixes the bug.