On Fri, Jul 29, 2011 at 10:39 AM, <[email protected]> wrote: > Dear Greg, > > > I have added the code how I generate "my own confusion matrix" to the Wiki. > In my understanding, my function uses the predictions from the out-of-bag > prediction. But I guess that I have overlooked some nasty detail.
You call: pred,conf=cmp.ClassifyExample(pts[i]) This uses the full composite model to make each prediction, it doesn't do the same out-of-bag prediction done by ScreenComposite.ShowVoteResults It is possible to get both the out-of-bag confusion matrix as a python object (it's part of the return tuple from ScreenComposite.ShowVoteResults) and the breakdown of the out-of-bag predictions by point (not quite as straightforward, but possible). What exactly are you trying to do? > Cheers & Thanks, > Paul > > > P.S.: When comparing the results with a PipelinePilot-based Bayesian > catagorization model (ECFP_4 & standard settings), I'm surprised to see > that the PipelinePilot model is significantly better. I thought that the > MorganFingerprints are comparable to the ECFPs and would have assumed that > the model quality is in a similar range. It's probably not the fingerprints, but the model-building approach that makes the difference here. You can test this if you want in knime using the RDKit morgan fingerprints with the naive bayes fingerprint learner they added in version 2.4. Best Regards, -greg ------------------------------------------------------------------------------ Got Input? Slashdot Needs You. Take our quick survey online. Come on, we don't ask for help often. Plus, you'll get a chance to win $100 to spend on ThinkGeek. http://p.sf.net/sfu/slashdot-survey _______________________________________________ Rdkit-discuss mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/rdkit-discuss
