You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@spark.apache.org by "Piotr Suszyński (JIRA)" <ji...@apache.org> on 2016/04/15 16:18:25 UTC
[jira] [Updated] (SPARK-14661) Trim PCAModel by required explained
variance
[ https://issues.apache.org/jira/browse/SPARK-14661?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Piotr Suszyński updated SPARK-14661:
------------------------------------
Description: PCA is often used to trim the number of features by using the minimal variance retained by those features. It would be very convenient to have a method in PCAModel that would return a trimmed model, especially given that the PCAModel constructor is private so doing this somewhere in the user code is not possible. (was: PCA is often used to trim the number of features by using the minimal variance retained by those features. It would be very convenient to have a method in PCAModel that would return a trimmed model, especially given that the PCAModel constructor is private so doing this somewhere in the user code is not possible.
I already have almost ready code here: https://github.com/psuszyns/spark/commit/072e9b50c29047700c6482e2c718969d9f6d9753)
> Trim PCAModel by required explained variance
> --------------------------------------------
>
> Key: SPARK-14661
> URL: https://issues.apache.org/jira/browse/SPARK-14661
> Project: Spark
> Issue Type: Improvement
> Components: MLlib
> Affects Versions: 2.0.0
> Reporter: Piotr Suszyński
> Priority: Trivial
>
> PCA is often used to trim the number of features by using the minimal variance retained by those features. It would be very convenient to have a method in PCAModel that would return a trimmed model, especially given that the PCAModel constructor is private so doing this somewhere in the user code is not possible.
--
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