[jira] [Closed] (MADLIB-1356) DL: Fix keras session creation logic for postgres

["Frank McQuillan (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/MADLIB-1356?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Frank McQuillan closed MADLIB-1356.
-----------------------------------
    Resolution: Fixed

https://github.com/apache/madlib/pull/407

> DL: Fix keras session creation logic for postgres
> -------------------------------------------------
>
>                 Key: MADLIB-1356
>                 URL: https://issues.apache.org/jira/browse/MADLIB-1356
>             Project: Apache MADlib
>          Issue Type: Improvement
>          Components: Deep Learning
>            Reporter: Nikhil
>            Priority: Major
>             Fix For: v1.16
>
>
> Currently for postgres, we create a keras session in the fit function and keep it open for all the iterations and then close it at the end. 
> We did a few experiments and found out that if we create 1 keras session per iteration , it performs slightly better.
>  We also found a bug in postgres deep learning madlib UDFs
>     If gpus are available on the host but the user passed in 0 for the gpus_per_host param, we still end up using gpu memory. This is because in the madlib_keras.fit UDF we call model.get_weights() to initialize the model if transfer learning and warm start are not applicable, and this function ends up using gpu if there are any available.



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