monte python 0.1.0

Roland Memisevic — Fri, 09 May 2008 21:45:47 -0000

Monte (python) is a small machine learning library written in pure Python. The focus is on gradient based learning, in particular on the construction of complex models from many smaller components.

Monte in addition contains some simple implementations (often only a few lines long) of a variety of standard learning algorithms, such as kmeans, k-nearest neighbors, logistic regression, the Lasso, and others. Because of its simplicity and its focus on pure Python, Monte code has been claimed to be a useful aid for learning the basics of numerical- and ML-programming in Python, or to make the transition from the more common, commercial numerical languages.

The gradient based learning part of Monte is inspired mainly by the gradient based learning library available for the Lush language.

Most of the code in Monte depends on Python (>=2.4), Numpy (>=1.0) and Matplotlib (>=0.87).

mloss.org monte python

monte python 0.1.0