Milksets: Machine Learning Datasets for Python

Fork me on GitHub

This packages contains some U.C.I. datasets in a common Pythonic interface, which makes them easy to use with milk.

It is used as:

from milksets import wine
features,labels = wine.load()

features will be a 2d-numpy.ndarray of features and labels will be a 1d-numpy.ndarray of labels starting at 0 through N-1 (independently of how the labels were coded in the original data).

Where can I get it?

Github as usual. Alternatively the python packages index also contains official releases,the latest of which can be obtained by

easy_install milksets


pip install milksets

if you use these tools.