Ahhh, that is a good point Joel, I hadn't thought about that. I think you
may be right, check out the updated gist
<http://nbviewer.ipython.org/gist/choldgraf/6a7be7866f2a3a3d3f98>.
I also wrote in some objgraph code to print the backrefs before and after
the "fit" method is called using Ridge. Here's a comparison
<http://imgur.com/0ndE03S> of the reference graph before (left) and after
(right) calling fit. Here's an example of the code that I used:
objgraph.show_backrefs([X], filename='./temp1.png')
> out = super(Ridge, self).fit(X, y, sample_weight=sample_weight)
> objgraph.show_backrefs([X], filename='./temp2.png')
One thing to note is that there seems to be another reference to the arrays
you pass to Ridge.fit, though I don't have a ton of experience using
objgraph so again, maybe this is expected?
Oh, and @alex, with respect to RidgeCV, there are a couple reasons: 1.
there are some autocorrelations in my data so I can't do LOO cross
validation. 2. I want to use one of the non-default solvers for Ridge
(lsqr) and I can't figure out how to set this with RidgeCV, and 3. the way
my code is structured it just makes it easier for me to pass a
non-cv-specific model and combine it with GridSearch, and I'm lazy :)
Chris
--
_____________________________________
PhD Candidate in Neuroscience | UC Berkeley <http://hwni.org/>
Editor and Web Master | Berkeley Science Review
<http://sciencereview.berkeley.edu/>
_____________________________________
------------------------------------------------------------------------------
Meet PCI DSS 3.0 Compliance Requirements with EventLog Analyzer
Achieve PCI DSS 3.0 Compliant Status with Out-of-the-box PCI DSS Reports
Are you Audit-Ready for PCI DSS 3.0 Compliance? Download White paper
Comply to PCI DSS 3.0 Requirement 10 and 11.5 with EventLog Analyzer
http://p.sf.net/sfu/Zoho
_______________________________________________
Scikit-learn-general mailing list
Scikit-learn-general@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
