You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@spark.apache.org by "Manoj Kumar (JIRA)" <ji...@apache.org> on 2015/04/22 16:56:59 UTC
[jira] [Comment Edited] (SPARK-7045) Word2Vec: avoid intermediate
representation when creating model
[ https://issues.apache.org/jira/browse/SPARK-7045?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14507199#comment-14507199 ]
Manoj Kumar edited comment on SPARK-7045 at 4/22/15 2:56 PM:
-------------------------------------------------------------
I did try the method you suggested, but it does not work. Simple code to reproduce
{code}
class a(f: Int) {
def tmp(c: Float) = c.asInstanceOf[Int]
def this(c: Float) = this(tmp(c))
}
{code}
This fails with an error: not found: value tmp
was (Author: mechcoder):
I did try the method you suggested, but it does not work. Simple code to reproduce
{code}
class a(f: Int) {
def tmp(c: Float) = c.asInstanceOf[Int]
def this(c: Float) = this(c)
}
{code}
This fails with an error called "constructor's definition must precede calling constructor's definition."
> Word2Vec: avoid intermediate representation when creating model
> ---------------------------------------------------------------
>
> Key: SPARK-7045
> URL: https://issues.apache.org/jira/browse/SPARK-7045
> Project: Spark
> Issue Type: Improvement
> Components: MLlib
> Affects Versions: 1.4.0
> Reporter: Joseph K. Bradley
> Priority: Minor
>
> Word2VecModel now stores the word vectors as a single, flat array; Word2Vec does as well. However, when Word2Vec creates the model, it builds an intermediate representation. We should skip that intermediate representation.
> However, it will be nice to create a public constructor for Word2VecModel which takes that intermediate representation (a Map from String words to their Vectors), since it's a user-friendly representation.
--
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