[Numpy-discussion] Partial least squares

Gael Varoquaux gael.varoquaux at normalesup.org
Thu Mar 24 18:32:27 EDT 2011


On Thu, Mar 24, 2011 at 01:57:46PM -0700, Nadav Horesh wrote:
> The code doc refer to an excellent reference [Wegelin et al. 2000], so
> no real problem here.  If the reference is critical I would suggest on
> of the following:
> 1. Put a link to the document.

Do you have one? I'd be happy to add one.

> 2. Is it possible to copy the paper to the project site?  As the paper
> becomes old (on the internet time scale), it becomes volatile, and may
> vanish while the project is still alive.

I don't really want to do this: it is not the scikit's learn goal to
maintain a library of documents. However, if you want to give us an
executive summary of the document (respecting the copyright, of course,
so copy and paste should be avoided). I'd be very happy to integrate it
to the documentation. Once again, I am not doing this myself right now
because I am not an expert of PLS, and I don't trust myself to do it
right. I would have to understand the usecases a bit more, which I
feasible, but takes time.

Your feedback is much appreciated,

Gaël



More information about the NumPy-Discussion mailing list