On 04/24/2014 08:28 PM, Robert McGibbon wrote:
> Thanks for the quick response! Does something like this look appropriate?
>
> X = np.random.randn(1000, 50) # some 50-dimensional data
> p = Pipeline([ ('nystroem', Nystroem(n_components=100)), ('pca',
> PCA(n_components=2)) ])
> p.fit(X)
> components = p.named_steps['pca'].components_
>
Yes.
Though I'm not sure looking at the components will tell you much.
The coefficients correspond to the 100 randomly sampled points in the
Nystroem estimation.
You can do
X_kpca = p.fit_transform(X)
and then scatter X_kpca, which might be a better way to look at what is
happening (might ;)
------------------------------------------------------------------------------
Start Your Social Network Today - Download eXo Platform
Build your Enterprise Intranet with eXo Platform Software
Java Based Open Source Intranet - Social, Extensible, Cloud Ready
Get Started Now And Turn Your Intranet Into A Collaboration Platform
http://p.sf.net/sfu/ExoPlatform
_______________________________________________
Scikit-learn-general mailing list
Scikit-learn-general@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
