[jira] [Commented] (LUCENE-7997) More sanity testing of similarities

["Robert Muir (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/LUCENE-7997?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16208515#comment-16208515 ] 

Robert Muir commented on LUCENE-7997:
-------------------------------------

Exactly we need a picky base sim test for this (like how BaseTokenStreamTestCase checks various requirements for analyzers). Currently these properties are "scattered" across various parts of the code/tests/issues: such as scores not being inf/NaN for some collectors, not being negative, monotonic tf, etc that maxscore requires. Sims that use certain statistics should fallback to other things when term frequencies are omitted, etc. It would be better to ensure we test all sims for all these things with direct tests. We should also try to test all norm values explicitly so that there aren't problems with super large documents and so on.



> More sanity testing of similarities
> -----------------------------------
>
>                 Key: LUCENE-7997
>                 URL: https://issues.apache.org/jira/browse/LUCENE-7997
>             Project: Lucene - Core
>          Issue Type: Task
>            Reporter: Adrien Grand
>            Priority: Minor
>
> LUCENE-7993 is a potential optimization that we could only apply if the similarity is an increasing functions of {{freq}} (all other things like DF and length being equal). This sounds like a very reasonable requirement for a similarity, so we should test it in the base similarity test case and maybe move broken similarities to sandbox?



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org