mahout-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "ASF GitHub Bot (JIRA)" <>
Subject [jira] [Commented] (MAHOUT-1493) Port Naive Bayes to the Spark DSL
Date Fri, 08 Aug 2014 22:00:13 GMT


ASF GitHub Bot commented on MAHOUT-1493:

Github user andrewpalumbo commented on the pull request:
    I made most of the changes from Dmitriy's comments.  I've done some (hackish) work here
just to get this in the right package, compiling and and testing.  Changed the Array[DrmLike]
to DrmLike for the sparse feature input (for now).  This is very basic and assumes that each
row correspnds to a unique label.  The only real engine specific DRM work right now is done
        val weightsPerFeature = observationsPerLabel.colSums
    I've added a Spark test suite with a test for a skeleton NB model. Tests pass here on
Spark.   I've also added an h2o test suite on my MAHOUT-1493-1500 branch with relativly minimal
effort (had to make a few dependency changes to the h20/pom.xml). h2o tests pass.
    Obviously there is still a lot of work to do here and it won't be ready to merge anytime
soon, so I'll leave this PR open for a little while in case anybody's interested and then
close it until i have some more work done on it as to not clog up the PR page. 

> Port Naive Bayes to the Spark DSL
> ---------------------------------
>                 Key: MAHOUT-1493
>                 URL:
>             Project: Mahout
>          Issue Type: Bug
>          Components: Classification
>            Reporter: Sebastian Schelter
>            Assignee: Sebastian Schelter
>             Fix For: 1.0
>         Attachments: MAHOUT-1493.patch, MAHOUT-1493.patch, MAHOUT-1493.patch, MAHOUT-1493.patch,
> Port our Naive Bayes implementation to the new spark dsl. Shouldn't require more than
a few lines of code.

This message was sent by Atlassian JIRA

View raw message