[ 
https://issues.apache.org/jira/browse/MADLIB-1135?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Frank McQuillan updated MADLIB-1135:
------------------------------------
    Description: 
Follow on from https://issues.apache.org/jira/browse/MADLIB-413 and 
https://issues.apache.org/jira/browse/MADLIB-1134

Story

As a MADlib developer, I want to get 3nd phase implementation of NN going with 
training and prediction functions, so that I can have a more advanced and 
performant version of NN

Features to add in order of priority:

* dynamic structs
* other algos (e.g., resilient backpropagation)
* multiple loss functions

  was:
Follow on from https://issues.apache.org/jira/browse/MADLIB-413 and 
https://issues.apache.org/jira/browse/MADLIB-1134

Story

As a MADlib developer, I want to get 3nd phase implementation of NN going with 
training and prediction functions, so that I can have a more advanced and 
performant version of NN

Features to add in order of priority:

* mini-batching
* momentum
* dynamic structs
* other algos (e.g., resilient backpropagation)
* multiple loss functions


> Neural Networks - MLP - Phase 3
> -------------------------------
>
>                 Key: MADLIB-1135
>                 URL: https://issues.apache.org/jira/browse/MADLIB-1135
>             Project: Apache MADlib
>          Issue Type: Improvement
>          Components: Module: Neural Networks
>            Reporter: Frank McQuillan
>            Priority: Major
>             Fix For: v2.0
>
>
> Follow on from https://issues.apache.org/jira/browse/MADLIB-413 and 
> https://issues.apache.org/jira/browse/MADLIB-1134
> Story
> As a MADlib developer, I want to get 3nd phase implementation of NN going 
> with training and prediction functions, so that I can have a more advanced 
> and performant version of NN
> Features to add in order of priority:
> * dynamic structs
> * other algos (e.g., resilient backpropagation)
> * multiple loss functions



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

Reply via email to