You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@ignite.apache.org by "Anton Dmitriev (JIRA)" <ji...@apache.org> on 2018/06/05 08:13:00 UTC

[jira] [Updated] (IGNITE-8668) K-fold cross validation of models

     [ https://issues.apache.org/jira/browse/IGNITE-8668?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Anton Dmitriev updated IGNITE-8668:
-----------------------------------
    Description: 
Cross validation is a well knows approach that allows to avoid overfitting and therefore improve model quality. K-fold cross validation is based on splitting dataset on _k_ disjoint subsets and using _k-1_ of them as train subset and the remaining subset for test (with all possible combinations).

The goal of this task is to implement K-fold cross validation based on an ability to filter dataset added recently in IGNITE-8666.

> K-fold cross validation of models
> ---------------------------------
>
>                 Key: IGNITE-8668
>                 URL: https://issues.apache.org/jira/browse/IGNITE-8668
>             Project: Ignite
>          Issue Type: New Feature
>          Components: ml
>            Reporter: Yury Babak
>            Assignee: Anton Dmitriev
>            Priority: Major
>             Fix For: 2.6
>
>
> Cross validation is a well knows approach that allows to avoid overfitting and therefore improve model quality. K-fold cross validation is based on splitting dataset on _k_ disjoint subsets and using _k-1_ of them as train subset and the remaining subset for test (with all possible combinations).
> The goal of this task is to implement K-fold cross validation based on an ability to filter dataset added recently in IGNITE-8666.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)