[Tutor] PCA on sparse matrices, tolerance of eigenvalues

[Alan Gauld]

"Jaidev Deshpande" <deshpande.jaidev at gmail.com> wrote

> I tried using the 'scipy.sparse.eigs' tool for performing principal
> component analysis on a matrix which is roughly 80% sparse.
>
> First of all, is that a good way to go about it?

Umm, I remember the term eigenvalue from University math.
Other than that the first sentence could be in Chinese! :-)

I suspect you might get a better resoponse on a scipy
or numpy mailing list or forum. Although there are a few
users of that here it sounds like your queries are likely
to be more scipy oriented than general Python.

There is a list of options here:

http://www.scipy.org/Mailing_Lists

> Second, the operation failed when the function failed to converge on
> accurate eigenvalues. I noticed the 'tol' attribute in the function, 
> but how
> does one define a reasonable tolerance and calculate it?

Sorry, I understand the sentence this time but again
its more a scipy question than a Python one.

Alan G.