[jira] [Commented] (FLINK-1901) Create sample operator for Dataset

["Till Rohrmann (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/FLINK-1901?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14499396#comment-14499396 ] 

Till Rohrmann commented on FLINK-1901:
--------------------------------------

A sampling operator for non-iterations can be realized as a mapPartition operator. This would also allow us to do sampling without replacement. 

The harder problem is to do sampling within an iteration. Usually, one would sample from a static code path which is only executed once. Maybe we can include the sampling operator on the dynamic code path to trigger it every iteration.

> Create sample operator for Dataset
> ----------------------------------
>
>                 Key: FLINK-1901
>                 URL: https://issues.apache.org/jira/browse/FLINK-1901
>             Project: Flink
>          Issue Type: Improvement
>          Components: Core
>            Reporter: Theodore Vasiloudis
>
> In order to be able to implement Stochastic Gradient Descent and a number of other machine learning algorithms we need to have a way to take a random sample from a Dataset.
> We need to be able to sample with or without replacement from the Dataset, choose the relative size of the sample, and set a seed for reproducibility.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)