miftanurfarid / machinelearningpracticalsessions

My Machine Learning practical sessions done at the INSA of Rouen & the University of Rouen in 2013 and 2014.

Machine Learning Practical Sessions

This repository gather almost all Machine Learning practical sessions I attended during my studies at the INSA of Rouen & the University of Rouen in 2013 and 2014.

For most of theses sessions, the point was to implement Machine Learning algorithms to get a sense of how they work.

Also, having been written in French schools, most of the comments are in French, although I sometimes wrote them in English. Sorry for non-French speakers. However, code and charts should be understandable.

Organization

Each practical session consist of a folder containing the sources (that you should be able to run) and a report of the results as a PDF file.

Each session is numbered x.y, where x correspond to the number of the course, and y correspond to the number of the practical session of the course. I numbered the courses approximately chronologically.

About the code

The code is either in Matlab (for the most part) or in Python (for a few ones).

Matlab

For each session, the main file that should be run is usually prefixed by an underscore.

For Matlab, the following libraries might be needed:

• CVX (very often)
• SimpleMKL (very often for monqp.m)
• PROPACK (rarely)
• PRTools (very rarely)
• SOM Toolbox (very rarely)

Python

For Python you will need the scipy environment, pickle, pykalman and yahmm.

The code is available as a Python file or a IPython Notebook.