[
https://issues.apache.org/jira/browse/MAHOUT-529?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12979453#action_12979453
]
Sean Owen commented on MAHOUT-529:
----------------------------------
I'm trolling through updating some issues that might be able to move forward.
The patch works and all but do you think it's suitable for addition at this
point or you want to work more on it?
It could be useful to explain how this relates to the other code already there.
I am not an expert but it seems to overlap a fair bit? For instance what is
DefaultGradientLinear doing vs existing DefaultGradient?
> Implement LinearRegression
> --------------------------
>
> Key: MAHOUT-529
> URL: https://issues.apache.org/jira/browse/MAHOUT-529
> Project: Mahout
> Issue Type: New Feature
> Reporter: Frank Wang
> Fix For: 0.5
>
> Attachments: MAHOUT-529.patch
>
>
> The implementation would an OnlineLearner that would be to similar to
> AbstractOnlineLogisticRegression for logistic regression.
> The learning will proceed by stochastic gradient descent.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
