[
https://issues.apache.org/jira/browse/LUCENE-5258?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13789369#comment-13789369
]
Robert Muir commented on LUCENE-5258:
-------------------------------------
{quote}
I build up a more principled test using caliper and now see the Fast.* methods
coming in about 30% faster. Dot method is slower.
{quote}
Hi Ted: thank you for spending the time to do the caliper benchmarks!
{quote}
It occurs to me that you don't need the final inverse trig function if you only
want to rank distances.
{quote}
I absolutely agree. I do think its handy to get the real distance back in the
TopFieldDocs from this function, and I think honestly this function is often
way-overkill especially if you combine with many other scoring factors?
I know Uwe has some stuff here like optimized flat-earth formulas, I hope he
adds it in the future.
> add distance function to expressions/
> -------------------------------------
>
> Key: LUCENE-5258
> URL: https://issues.apache.org/jira/browse/LUCENE-5258
> Project: Lucene - Core
> Issue Type: New Feature
> Components: modules/other
> Reporter: Robert Muir
> Fix For: 5.0, 4.6
>
> Attachments: LUCENE-5258.patch
>
>
> Adding this static function makes it really easy to incorporate distance with
> the score or other signals in arbitrary ways, e.g. score / (1 +
> log(distance)) or whatever.
--
This message was sent by Atlassian JIRA
(v6.1#6144)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
