README

===========
SuPyLearner
===========

An implementation of the SuperLearner algorithm in Python built on scikit-learn.  

Typical useage:

    #Import stuff
    from supylearner import *
    from sklearn import datasets, svm, linear_model, neighbors, svm
    import numpy as np

    #Generate a dataset.
    np.random.seed(100)
    X, y=datasets.make_friedman1(1000)

    ols=linear_model.LinearRegression()
    elnet=linear_model.ElasticNetCV(rho=.1)
    ridge=linear_model.RidgeCV()
    lars=linear_model.LarsCV()
    lasso=linear_model.LassoCV()
    nn=neighbors.KNeighborsRegressor()
    svm1=svm.SVR(scale_C=True) 
    svm2=svm.SVR(kernel='poly', scale_C=True)
    lib=[ols, elnet, ridge,lars, lasso, nn, svm1, svm2]
    libnames=["OLS", "ElasticNet", "Ridge", "LARS", "LASSO", "kNN", "SVM rbf", "SVM poly"]

    sl=SuperLearner(lib, libnames, loss="L2")

    sl.fit(X, y)

    sl.summarize()

    cv_superlearner(sl, X, y, K=5)