[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Principle components analysis on a large dataset
|
From: |
Søren Hauberg |
|
Subject: |
Re: Principle components analysis on a large dataset |
|
Date: |
Sat, 22 Aug 2009 08:54:40 +0200 |
lør, 22 08 2009 kl. 07:41 +0200, skrev Jaroslav Hajek:
> One option, if you're only interested in a few principal components,
> is to use eigs with an implicit matrix - you need to have Octave
> linked with ARPACK for that.
This is the approach I'd go for. Another option is to use iterative
implementations of PCA. Pretty good EM-style methods are out there. I
don't have code to share though :-(
Søren