[teiid-issues] [JBoss JIRA] (TEIID-3398) Use Weka algorithm to caculate serializing, deserializing linear regression rate

[Kylin Soong (JIRA)]

     [ https://issues.jboss.org/browse/TEIID-3398?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Kylin Soong updated TEIID-3398:
-------------------------------
    Priority: Optional  (was: Major)


> Use Weka algorithm to caculate serializing, deserializing linear regression rate
> --------------------------------------------------------------------------------
>
>                 Key: TEIID-3398
>                 URL: https://issues.jboss.org/browse/TEIID-3398
>             Project: Teiid
>          Issue Type: Feature Request
>          Components: Sizing Application
>            Reporter: Kylin Soong
>            Assignee: Kylin Soong
>            Priority: Optional
>
> We have lots of data points {(size i, time i), i = 1, ..., n}, the goal is to find the equation of the straight line like
>     time = K * size
> The key is to find the linear regression rate K, current we do not use any algorithm, Weka(http://sourceforge.net/projects/weka/) have algorithm to fine the K, we may use the Weka to enhance Sizing application
> Doc link: http://www.cs.waikato.ac.nz/~ml/weka/documentation.html



--
This message was sent by Atlassian JIRA
(v7.2.3#72005)