You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@hama.apache.org by "Yexi Jiang (JIRA)" <ji...@apache.org> on 2013/08/07 05:40:49 UTC

[jira] [Updated] (HAMA-791) Fix the problem that MultilayerPerceptron fails to learn a good hypothesis sometimes

     [ https://issues.apache.org/jira/browse/HAMA-791?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Yexi Jiang updated HAMA-791:
----------------------------

    Attachment: HAMA-791.patch
    
> Fix the problem that MultilayerPerceptron fails to learn a good hypothesis sometimes
> ------------------------------------------------------------------------------------
>
>                 Key: HAMA-791
>                 URL: https://issues.apache.org/jira/browse/HAMA-791
>             Project: Hama
>          Issue Type: Bug
>            Reporter: Yexi Jiang
>            Assignee: Yexi Jiang
>             Fix For: 0.6.3
>
>         Attachments: HAMA-791.patch
>
>
> Fix the problem that MultilayerPerceptron fails to learn a good hypothesis sometimes. If the weights of MLP initialized to very bad values, the model might fail the test.
> Since the new model developed in [HAMA-770|https://issues.apache.org/jira/browse/HAMA-770] will replace the current old model, this problem will be simply fixed by increase the training iteration.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira