[
https://issues.apache.org/jira/browse/LUCENENET-288?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12779781#action_12779781
]
Michael Garski commented on LUCENENET-288:
------------------------------------------
A little more digging into this...
There are 4 tests in the class, TestCustomScoreByte, TestCustomScoreShort,
TestCustomScoreint, and TestCustomScoreFloat. TestCustomScoreByte currently
passes, but if a test case is added to test with the dboost parameter to 3.0,
it will fail. All of the tests pass with a dboost value of 2.0 or less, so
this could be due to differences in how the Explanation score and the scorer
score are being calculated (the reason for the +-0.00005f variance in the
Assert).
I'll be turning on the verbose output with FunctionTestSetup.DBG and comparing
the output between NUnit & Junit to verify that the root cause is a floating
point arithmetic issue.
> Function.TestCustomScoreQuery
> -----------------------------
>
> Key: LUCENENET-288
> URL: https://issues.apache.org/jira/browse/LUCENENET-288
> Project: Lucene.Net
> Issue Type: Bug
> Reporter: Michael Garski
>
> Three of the tests in TestCustomScoreQuery are not passing due to the
> comparison between the explanation score and the hit score do not fall within
> the tolerance of +-0.00005f. Altering the tolerance to 0.0005f causes all of
> the tests to pass, so there appears to be a subtle issue in how the scores
> are being calculated.
> A similar test exists in 2.4 that passes with the tolerance of 0.00005f,
> which to me rules out floating point calculation differences.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
