On Sun, Mar 24, 2013 at 11:41 AM, Mathieu Blondel <math...@mblondel.org> wrote:
> If you use a recent version of scikit-learn, you should be able to do this: > ovr = OneVsRestClassifier(SGDClassifier(loss="log")) > ovr.fit(X_train, y_train) > prob = ovr.predict_proba(X_test) Ah, right, you want to use partial_fit so my solution doesn't work. We need to add partial_fit support to OneVsRestClassifier or multiclass support in SGDClassifier.predict_proba (the latter should be simpler since it's just a matter of normalizing the probabilities). Mathieu ------------------------------------------------------------------------------ Everyone hates slow websites. So do we. Make your web apps faster with AppDynamics Download AppDynamics Lite for free today: http://p.sf.net/sfu/appdyn_d2d_mar _______________________________________________ Scikit-learn-general mailing list Scikit-learn-general@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
