Mailing-List: contact issues-help@spark.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@spark.apache.org
Date: Wed, 30 Jul 2014 11:42:38 +0000 (UTC)
From: "Apache Spark (JIRA)" <jira@apache.org>
To: issues@spark.apache.org
Message-ID: <JIRA.12730765.1406720023532.74886.1406720558472@arcas>
In-Reply-To: <JIRA.12730765.1406720023532@arcas>
References: <JIRA.12730765.1406720023532@arcas>
Subject: [jira] [Commented] (SPARK-2748) Loss of precision for small
 arguments to Math.exp, Math.log
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


    [ https://issues.apache.org/jira/browse/SPARK-2748?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14079207#comment-14079207 ] 

Apache Spark commented on SPARK-2748:
-------------------------------------

User 'srowen' has created a pull request for this issue:
https://github.com/apache/spark/pull/1659

> Loss of precision for small arguments to Math.exp, Math.log
> -----------------------------------------------------------
>
>                 Key: SPARK-2748
>                 URL: https://issues.apache.org/jira/browse/SPARK-2748
>             Project: Spark
>          Issue Type: Bug
>          Components: GraphX, MLlib
>    Affects Versions: 1.0.1
>            Reporter: Sean Owen
>            Priority: Minor
>
> In a few places in MLlib, an expression of the form log(1.0 + p) is evaluated. When p is so small that 1.0 + p == 1.0, the result is 0.0. However the correct answer is very near p. This is why Math.log1p exists.
> Similarly for one instance of exp(m) - 1 in GraphX; there's a special Math.expm1 method.
> While the errors occur only for very small arguments, given their use in machine learning algorithms, this is entirely possible.
> Also, while we're here, naftaliharris discovered a case in Python where 1 - 1 / (1 + exp(margin)) is less accurate than exp(margin) / (1 + exp(margin)). I don't think there's a JIRA on that one, so maybe this can serve as an umbrella for all of these related issues.


--
This message was sent by Atlassian JIRA
(v6.2#6252)