[
https://issues.apache.org/jira/browse/MADLIB-1374?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Frank McQuillan updated MADLIB-1374:
------------------------------------
Fix Version/s: (was: v1.17)
v2.0
> DL: Split madlib_keras.sql_in
> -----------------------------
>
> Key: MADLIB-1374
> URL: https://issues.apache.org/jira/browse/MADLIB-1374
> Project: Apache MADlib
> Issue Type: Task
> Components: Deep Learning
> Reporter: Ekta Khanna
> Priority: Major
> Fix For: v2.0
>
>
> Currently, src/ports/postgres/modules/deep_learning/madlib_keras.sql_in file
> for deep_learning module is getting really big and hard to manage.
> Proposal:
> 1. Separate deep_learning/madlib_keras.sql_in into multiple sql files like
> fit, evalute, predict etc.
> For reference, we can look at the following modules:
> 1. src/ports/postgres/modules/recursive_partitioning - Split into Decision
> Tree and Random Forest
> 1. src/ports/postgres/modules/regress
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
