Mailing-List: contact issues-help@spark.apache.org; run by ezmlm
Precedence: bulk
Date: Wed, 6 Apr 2016 23:48:25 +0000 (UTC)
From: "Joseph K. Bradley (JIRA)" <jira@apache.org>
To: issues@spark.apache.org
Message-ID: <JIRA.12924758.1451431044000.157279.1459986505759@Atlassian.JIRA>
In-Reply-To: <JIRA.12924758.1451431044000@Atlassian.JIRA>
References: <JIRA.12924758.1451431044000@Atlassian.JIRA>
 <JIRA.12924758.1451431044778@arcas>
Subject: [jira] [Comment Edited] (SPARK-12566) GLM model family, link
 function support in SparkR:::glm
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


    [ https://issues.apache.org/jira/browse/SPARK-12566?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15229368#comment-15229368 ] 

Joseph K. Bradley edited comment on SPARK-12566 at 4/6/16 11:47 PM:
--------------------------------------------------------------------

Here's my preferred design.  I prefer to abstract the implementation (solver) from the API (model) as much as possible.
* R glm calls Scala GLM, using solver = auto by default
* Scala GLM has solver = auto by default.  Auto should mean "best effort"
** With few features (< 4K or so),
*** For family = gaussian and link = identity, use normal equations.
*** For others, use IRLS.
** With many features, use LBFGS if possible (for family, link).  Otherwise, throw an exception.
* Scala LinearRegression, LogisticRegression call GLM.  I.e., they uses normal equations, IRLS when possible.

What do yall think?

(update: now catching up on PR comments...)


was (Author: josephkb):
Here's my preferred design.  I prefer to abstract the implementation (solver) from the API (model) as much as possible.
* R glm calls Scala GLM, using solver = auto by default
* Scala GLM has solver = auto by default.  Auto should mean "best effort"
** With few features (< 4K or so),
*** For family = gaussian and link = identity, use normal equations.
*** For others, use IRLS.
** With many features, use LBFGS if possible (for family, link).  Otherwise, throw an exception.
* Scala LinearRegression, LogisticRegression call GLM.  I.e., they uses normal equations, IRLS when possible.

What do yall think?

> GLM model family, link function support in SparkR:::glm
> -------------------------------------------------------
>
>                 Key: SPARK-12566
>                 URL: https://issues.apache.org/jira/browse/SPARK-12566
>             Project: Spark
>          Issue Type: New Feature
>          Components: ML, SparkR
>            Reporter: Joseph K. Bradley
>            Assignee: yuhao yang
>            Priority: Critical
>
> This JIRA is for extending the support of MLlib's Generalized Linear Models (GLMs) to more model families and link functions in SparkR. After SPARK-12811, we should be able to wrap GeneralizedLinearRegression in SparkR with support of popular families and link functions.


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

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org