This is an implementation of Stacked Generalization, as explained in the paper Stacked Generalization, 1992, by D.H. Wolpert.
It extends Stacked Generalization by also offering Bucket of Models for the selection of the classifiers and parameters. The classifiers and parameters are choosing by using cross-validation on the training set.
There's an included file showing an exmaple usage.
- This algorithm is very computationally expensive, but as long as the parameters are not set too high, it should be feasible. (see the file
example_usage_stacked_gen.pyfor example parameters)
GNU GPL version 2 or later