[jira] [Commented] (SOLR-11597) Implement RankNet.

classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view
|

[jira] [Commented] (SOLR-11597) Implement RankNet.

JIRA jira@apache.org

    [ https://issues.apache.org/jira/browse/SOLR-11597?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16324712#comment-16324712 ]

Michael A. Alcorn commented on SOLR-11597:
------------------------------------------

[~cpoerschke] - yes, they are separate. RankNet is specifically a learning to rank model whereas the other architectures being discussed are more for modeling language. I describe how to train a RankNet model using Keras [here|https://github.com/airalcorn2/Solr-LTR#RankNet]. The weights can be exported from Keras in any format the user wants. The original Solr representation I suggested made it easy to export the weights from Keras since the weights are contained in matrices.

> Implement RankNet.
> ------------------
>
>                 Key: SOLR-11597
>                 URL: https://issues.apache.org/jira/browse/SOLR-11597
>             Project: Solr
>          Issue Type: New Feature
>      Security Level: Public(Default Security Level. Issues are Public)
>          Components: contrib - LTR
>            Reporter: Michael A. Alcorn
>
> Implement RankNet as described in [this tutorial|https://github.com/airalcorn2/Solr-LTR].



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

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]