Fixing test flakiness #2242
Merged
Fixing test flakiness #2242
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
mergify
bot
requested review from
a team,
yeukfu and
ziyiwu9494
and removed request for
a team
|
/ok-to-test |
1 similar comment
|
hi, it seems the checks failed due to some config issue? |
|
@ziyiwu9494 can you PTAL? |
|
/ok-to-test |
krzentner
approved these changes
Mar 4, 2021
|
Hi all, thanks for approving the PR. Is there something blocking for merge? |
ziyiwu9494
approved these changes
Mar 15, 2021
ziyiwu9494
approved these changes
Mar 15, 2021
|
/ok-to-test |
haydenshively
approved these changes
Mar 19, 2021
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Outdated suggestions cannot be applied.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 test
test_update_envs_env_updateis flaky. It sometimes fails more than twice. The assertion on line 78assert np.var(mean_rewards) > 0fails. This PR addresses this issue.To find a solution, I collected samples of
np.var(mean_rewards)from several test executions and computed the tail distribution just to check how often can the value be 0.Based on the collected samples, it seems there is ~12% chance that the test will fail. I suggest to run this test 3 times, then the probability of failure will be <1%.
I think refactoring this test using the statistical evaluation might be a good way to reduce the flakiness of this test.
Do you guys think this makes sense? Please let me know if this looks good or if you have any other suggestions. Also, here I assume there are no bugs in the code under test.
Also, I am curious to know why do you check if
var > 0? Statistically, zero variance can happen sometimes. Is it better to have something likeassert np.mean(mean_rewards) > 0?