mahout-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Sean Owen (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (MAHOUT-487) Issues with memory use and inconsistent or state-influenced results when using CBayesAlgorithm
Date Wed, 04 May 2011 12:17:03 GMT

     [ https://issues.apache.org/jira/browse/MAHOUT-487?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Sean Owen updated MAHOUT-487:
-----------------------------

    Affects Version/s: 0.5
                       0.4
        Fix Version/s:     (was: 0.5)
                       0.6

No movement in a month, coming up on our deadline for 0.5 code freeze -- 0.6?

> Issues with memory use and inconsistent or state-influenced results when using CBayesAlgorithm
> ----------------------------------------------------------------------------------------------
>
>                 Key: MAHOUT-487
>                 URL: https://issues.apache.org/jira/browse/MAHOUT-487
>             Project: Mahout
>          Issue Type: Bug
>          Components: Classification
>    Affects Versions: 0.3, 0.4, 0.5
>            Reporter: Drew Farris
>            Assignee: Robin Anil
>            Priority: Minor
>             Fix For: 0.6
>
>
> Came across this digging through the mailing list archives for something else, probably
worth tracking as an issue.
> {quote}
> During classification, every word still unknown is added to 
> featureDictionary. This leads to the excessive growth if lots of texts 
> with unknown words are to be classified. The inconsistency is caused by 
> using a "vocabCount" that is not reset after each classification. 
> Indeed, featureDictionary.size() is used for "vocabCount", which 
> increases every time new unknown words are discovered.
> {quote}
> See: http://www.lucidimagination.com/search/document/7dabe3efec8d136d/issues_with_memory_use_and_inconsistent_or_state_influenced_results_when_using_cbayesalgorit#8853165db260bf75
> Alternately per Robin:
> {quote}
> We can remove the addition features to the
> dictionary altogether. Will yield better performance, and lock down the
> model. Will require a bit more modification
> {quote}

--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira

Mime
View raw message