The first assignment for the OMSCS machine learning class, in which I implement (steal) a ML library and muck around with it. Specifically, I steal scikit-learn (BSD License) since it handily implements all of the required functionality. Namely,

• Decision Trees
• Neural Networks
• Gradient Boosting
• Support Vector Machines
• k-Nearest Neighbors

I'm writing and testing all of this using Python 3.7.2, so you'll need to have that (other python versions >=3.4 may work, but no guarantees). The instructions for getting it vary by OS and by distribution within each OS. The general idea is to use the standard package manager for whatever you use. But this should help if you're feeling lost.

You should also get the code:

$ git clone https://github.com/astex/cs7641a1 .

There are a number of ways you can install the required libraries. I personally prefer to use a virtual environment so that the dependencies of my project stay isolated from any software needed elsewhere on the system. Some folks of a more scientific bent may prefer conda. But, the simplest way I can think of is this (run from the project directory):

$ pip install -r builddeps.txt
$ pip install -r deps.txt

Everything you need in order to generate the plots is included, so you can go ahead and run the scripts.

$ python -m src.california.main all; python -m src.imdb.main all;

This will take around a day to execute and will use an awful lot of memory and processing power. Alternatively, you can regenerate specific plots by passing a list of plot names instead of all. You can see the plot names in the applications' help texts:

$ python -m src.california.main -h
$ python -m src.imdb.main -h

This is a homework assignment, not production software. You may run it and look at it, but may not redistribute it or use it in any system for any reason.