[pymvpa] MappedClassifer

[Jacob Bollinger]

Hello all,

on line 1279 of mvpa2/clfs/meta.py it reads:
class MappedClassifier(ProxyClassifier):
    """`ProxyClassifier` which uses some mapper prior training/testing.





    `MaskMapper` can be used just a subset of features to


    train/classify.


    Having such classifier we can easily create a set of classifiers


    for BoostedClassifier, where each classifier operates on some set


    of features, e.g. set of best spheres from SearchLight, set of


    ROIs selected elsewhere. It would be different from simply


    applying whole mask over the dataset, since here initial decision


    is made by each classifier and then later on they vote for the


    final decision across the set of classifiers.


    """

Has anyone accomplished this?
In other words:
1) Data: 50 features, each with 10000 observations
2) Train several clfs, each with its own optimized feature selection
method; because each classifier will prefer a distinct set of features,
overall feature coverage after feature selection will be better if many
classifiers are combined.
3) Combine function
4) According to the literature, the combined classifiers (even if many of
them are weak) should lead to a dramatic improvement in classification
accuracy.
5) Right now I can get ~65% accuracy on a three-fold cross validation with
SMLR(), but am hoping to get near 80% with a combined approach.
6) It does not have to be the one stated in meta.py, but this seems like a
good lead.

Any help is much appreciated!
Thanks!
*
*
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.alioth.debian.org/pipermail/pkg-exppsy-pymvpa/attachments/20130626/4f0aa502/attachment.html>