[jira] Commented: (MAHOUT-24) Skeletal LWLR implementation

["Alexander Hans (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/MAHOUT-24?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12666101#action_12666101 ] 

Alexander Hans commented on MAHOUT-24:
--------------------------------------

> Alexander did you further pursue this issue?

Sorry for being so quiet on this one, the last couple of months have been somewhat stressful for me day job-wise, so I didn't find any time to work on this issue. I think things will become easier in a few weeks.

Ted, thanks for pointing out other methods for inverting a matrix. For a single LWLR prediction this has to be done only once by the reducer, so I think it would be ok to implement whatever method works for a first version and change it later to something better once the whole algorithm is functional.

> Skeletal LWLR implementation
> ----------------------------
>
>                 Key: MAHOUT-24
>                 URL: https://issues.apache.org/jira/browse/MAHOUT-24
>             Project: Mahout
>          Issue Type: New Feature
>         Environment: n/a
>            Reporter: Samee Zahur
>         Attachments: LWLR.patch.tar.bz2
>
>
> This is a very skeletal but functional implementation for LWLR. It outputs n lines where n is the number of dimensions. ith line = sum(x[i]*x[ind]) where ind is the index of independant variable. So the actual gradient = 2nd line/1st line for the classical 2D.
> Contains a single small test case for demonstration.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.