scikit-learn Compatability

To use LFSpy as part of an sklearn pipeline:

from LFS import LocalFeatureSelection
from sklearn.pipeline import Pipeline

lfs = LocalFeatureSelection()
pipeline = Pipeline([('lfs', lfs)])
pipeline.fit(training_data, training_labels)
predicted_labels = pipeline.predict(testing_data)
total_error, class_error = pipeline.score(testing_data, testing_labels)