[
https://issues.apache.org/jira/browse/LUCENE-7997?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Robert Muir updated LUCENE-7997:
--------------------------------
Attachment: LUCENE-7997_wip.patch
Updated patch, also enforcing that explain == score (exactly, no floating point
differences).
I cleaned up the BM25 explain to be transparent and reflect how the calculation
is done.
Most importantly, explanation is now broken out as {{scaling * df * tf}}, like
how we compute it, and described in
http://kak.tx0.org/Information-Retrieval/TFxIDF rather than displaying the
"re-arranged formula" with tf including the {{k1 + 1}} scaling factor. Maybe
its an improvement for debugging, too since it pulls out the independent
scaling factor, making it easier to see the specifics of term frequency
saturation and IDF across docs/terms?
> 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
> Attachments: LUCENE-7997_wip.patch, LUCENE-7997_wip.patch,
> LUCENE-7997_wip.patch
>
>
> 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: [email protected]
For additional commands, e-mail: [email protected]
