hi andreas and alex,
the basic idea is to do a transform of feature space for the training set
and apply it to the test set. since the cv routines do the split of the set
internally, there is no mechanism that i could find to do that.
alex: regarding the pipeline, doesn't it require fit and predict
operations. will it accept fit and transform?
cheers,
satra
On Wed, Nov 9, 2011 at 11:17 AM, Alexandre Gramfort <
[email protected]> wrote:
> hi satra,
>
> wouldn't a pipeline solve your pb?
>
> see eg. :
>
>
> http://scikit-learn.org/stable/auto_examples/feature_selection_pipeline.html#example-feature-selection-pipeline-py
>
> Alex
>
> On Wed, Nov 9, 2011 at 10:55 AM, Andreas Müller
> <[email protected]> wrote:
> > Hi Satra.
> > Can you explain the use-case a bit more?
> > I don't understand it from the code, sorry.
> > Cheers,
> > Andy
> >
> > On 11/09/2011 04:51 PM, Satrajit Ghosh wrote:
> >
> > i often find myself doing the following for cross-validation. i.e.
> > estimating the transform from the training set. would this be useful as a
> > parameter on cross_val_score, gridsearchcv, etc.,. if so i'll send a pr.
> > ----
> > class NoTransform():
> > def fit(self, X):
> > return self
> > def transform(self, X):
> > return X
> > def doCV(clf, X, y, Tx, cvf):
> > result = []
> > for train, test in cvf:
> > if Tx is None:
> > T = NoTransform()
> > else:
> > T = Tx()
> > result.append((y1[test],
> > clf.fit(T.fit(X[train]).transform(X[train]),
> > y[train]).predict(T.transform(X[test]))
> > ))
> > return result
> > ----
> > cheers,
> >
> > satra
> >
> >
> >
> ------------------------------------------------------------------------------
> > RSA(R) Conference 2012
> > Save $700 by Nov 18
> > Register now
> > http://p.sf.net/sfu/rsa-sfdev2dev1
> >
> > _______________________________________________
> > Scikit-learn-general mailing list
> > [email protected]
> > https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
> >
> >
> >
> ------------------------------------------------------------------------------
> > RSA(R) Conference 2012
> > Save $700 by Nov 18
> > Register now
> > http://p.sf.net/sfu/rsa-sfdev2dev1
> > _______________________________________________
> > Scikit-learn-general mailing list
> > [email protected]
> > https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
> >
> >
>
>
> ------------------------------------------------------------------------------
> RSA(R) Conference 2012
> Save $700 by Nov 18
> Register now
> http://p.sf.net/sfu/rsa-sfdev2dev1
> _______________________________________________
> Scikit-learn-general mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
>
------------------------------------------------------------------------------
RSA(R) Conference 2012
Save $700 by Nov 18
Register now
http://p.sf.net/sfu/rsa-sfdev2dev1
_______________________________________________
Scikit-learn-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
