[
https://issues.apache.org/jira/browse/MADLIB-1037?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15658387#comment-15658387
]
Frank McQuillan commented on MADLIB-1037:
-----------------------------------------
Update reported today:
Have added regularization.
Accuracy looking pretty good:
* 84% accuracy on mnist on Postgres single node
* training takes 3 sec on 50,000 images
* testing size is ~23,000 images
Porting to Greenplum/HAWQ (merge function).
> Multi-class SVM with mini-batching
> -----------------------------------
>
> Key: MADLIB-1037
> URL: https://issues.apache.org/jira/browse/MADLIB-1037
> Project: Apache MADlib
> Issue Type: New Feature
> Components: Module: Support Vector Machines
> Reporter: Frank McQuillan
> Fix For: v1.10
>
>
> Prototype of structure svm trained with mini-batch
> Acceptance
> 1) train on mnist and achieve acceptable accuracy
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
