[jboss-jira] [JBoss JIRA] (JBRULES-3405) Evaluate benchmark results per average ranking, not per average score

Fri Apr 20 14:20:18 EDT 2012

    [ https://issues.jboss.org/browse/JBRULES-3405?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12686181#comment-12686181 ] 

Lukáš Petrovický commented on JBRULES-3405:
-------------------------------------------

Actually, for what I am mentioning, implementing just the benchmark comparator won't be enough. We need to compare every benchmark against every other benchmark, while the comparators only allow us to compare two benchmarks.

What's necessary is to implement some sort of "comparator^2" that takes all the benchmarks into consideration. This will require some changes inside Planner.

> Evaluate benchmark results per average ranking, not per average score
> ---------------------------------------------------------------------
>
>                 Key: JBRULES-3405
>                 URL: https://issues.jboss.org/browse/JBRULES-3405
>             Project: Drools
>          Issue Type: Feature Request
>      Security Level: Public(Everyone can see) 
>          Components: drools-planner
>    Affects Versions: 5.4.0.Beta2
>            Reporter: Lukáš Petrovický
>            Assignee: Geoffrey De Smet
>              Labels: newbie_contributor
>
> Currently. when deciding which solver config is the winner in a particular benchmark, the one with the best average score is picked. I believe this to be the wrong approach.
> When averaging two numbers, one if which is significantly smaller than the other, the average isn't the best metric you could use. (10, 10, 10, 20 and 1000 give you 210 as the average - is that really the best metric available to describe the data set?)
> For this reason, I would like the winner-picking algorithm to work different:
> 1) For each input file in the benchmark, rank the solvers per their score. (Basically 1st to Nth place.)
> 2) Then make a median of all these "places" and the best-placed algorithm wins.
> This way, you don't compare the solver results themselves. You compare how the solvers did in relation to the other solvers - which is something I consider much more important.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.jboss.org/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira