[
https://issues.apache.org/jira/browse/OPENNLP-338?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13440888#comment-13440888
]
Hyosup Shim commented on OPENNLP-338:
-------------------------------------
I'm still working on it. In changing internal parameters, the best F1 mesasure
I could get was around 0.30. (precision =: 0.20, recall =: 0.75)
It seems that minimizing loglikelihood doesn't make improvement in classified
result.
But I can't tell which part of code is buggy yet.
> Add L-BFGS parameter estimation training to maxent
> --------------------------------------------------
>
> Key: OPENNLP-338
> URL: https://issues.apache.org/jira/browse/OPENNLP-338
> Project: OpenNLP
> Issue Type: New Feature
> Components: Maxent
> Reporter: Joern Kottmann
> Assignee: Hyosup Shim
> Fix For: tools-1.5.3
>
> Attachments: nl-per.testa, nl-per.testb, nl-per.train,
> patch20120814-lbfgs.txt, patch20120821-lbfgs, patch-lbfgs.txt
>
>
> Add support for the L-BFGS algorithm to train a maxent classifier.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
