You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@systemml.apache.org by "Deron Eriksson (JIRA)" <ji...@apache.org> on 2016/02/04 01:55:39 UTC

[jira] [Commented] (SYSTEMML-468) Create Contributing to SystemML document

    [ https://issues.apache.org/jira/browse/SYSTEMML-468?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15131480#comment-15131480 ] 

Deron Eriksson commented on SYSTEMML-468:
-----------------------------------------

Initial version of Contributing doc merged by [PR63 | https://github.com/apache/incubator-systemml/pull/63]. Decide whether adding additional info should be added or if the should be resolved for now. Separate issues could be created for additional info if needed.

> Create Contributing to SystemML document
> ----------------------------------------
>
>                 Key: SYSTEMML-468
>                 URL: https://issues.apache.org/jira/browse/SYSTEMML-468
>             Project: SystemML
>          Issue Type: Task
>          Components: Documentation
>            Reporter: Deron Eriksson
>            Assignee: Deron Eriksson
>
> Create a Contributing to SystemML document for developers and data scientists.
> Describe our typical git workflow for contributing to the project (fork on github, clone to local, checkout a branch and do work on branch, push to github branch, file pull request, pull request reviewed, push additional changes to github branch/pull request, pull request accepted and integrated into project).
> Describe basic IDE setup for working on SystemML code (move Eclipse debug configuration info from Beginner's Guide to here).
> Describe creation of a test class.
> Describe running new code locally.
> Describe running new code on Hadoop.
> Describe running new code on Spark.
> Describe running new code via MLContext.
> Any other pointers that would help a new developer or data scientist contribute to SystemML could go here. For instance, a data scientist might want to contribute a new algorithm to SystemML. However, this algorithm probably needs a test class, so the contribution may require more than DML. The test class could be written by the data scientist, or perhaps a request could be made for a developer to handle the test class. It's good to spell out this kind of process to help grow our community.



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