Hitachi Vantara Pentaho Community Forums
Results 1 to 3 of 3

Thread: Help pleaase

  1. #1
    Join Date
    Oct 2015
    Posts
    11

    Default Help pleaase

    HELLO
    I'm using the PCA filter to a database with many instances , but it is taking too long preprocessing . How could speed up this process?

    Thanks!

  2. #2
    Join Date
    Aug 2006
    Posts
    1,741

    Default

    PCA uses matrix operations that have runtime that is effectively cubic in the number of attributes. How many attributes are in your data? The current nightly snapshot of trunk Weka now uses MTJ for linear algebra in LinearRegression and PCA. In pure Java mode this is slightly faster than the JAMA library that we used to use. If you are willing to check out and build Weka packages from SVN, then there is a package for each of the three major OSs that adds BLAS/LAPACK native libraries to the classpath. MTJ will then use these for additional speed instead of the pure Java version.

    Cheers,
    Mark.

  3. #3
    Join Date
    Oct 2015
    Posts
    11

    Default

    There are at the base 7130 attributes.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  
Privacy Policy | Legal Notices | Safe Harbor Privacy Policy

Copyright © 2005 - 2019 Hitachi Vantara Corporation. All Rights Reserved.