You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@spark.apache.org by "Angel Martinez Gonzalez (JIRA)" <ji...@apache.org> on 2015/05/14 18:41:01 UTC
[jira] [Commented] (SPARK-7615) MLLIB Word2Vec wordVectors divided
by Euclidean Norm equals to zero
[ https://issues.apache.org/jira/browse/SPARK-7615?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14543977#comment-14543977 ]
Angel Martinez Gonzalez commented on SPARK-7615:
------------------------------------------------
Hi, is any body working on this (and SPARK-7617, SPARK-7618)? I have been using Spark for some time, and would like get started contributing. This looks like a trivial issue (good fo a newbie).
Shall I provide a pull request for this? Would it make sense a pull request for the three related issues?
> MLLIB Word2Vec wordVectors divided by Euclidean Norm equals to zero
> --------------------------------------------------------------------
>
> Key: SPARK-7615
> URL: https://issues.apache.org/jira/browse/SPARK-7615
> Project: Spark
> Issue Type: Bug
> Components: MLlib
> Affects Versions: 1.3.1
> Reporter: Eric Li
> Priority: Minor
> Original Estimate: 1h
> Remaining Estimate: 1h
>
> In Word2VecModel, wordVecNorms may contains Euclidean Norm equals to zero. This will cause incorrect calculation for cosine distance. when you do cosineVec(ind) / wordVecNorms(ind). Cosine distance should be equal to 0 for norm = 0.
--
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