A project submitted in partial fulfilment of the requirements for the M.Sc. in Advanced Computing Technologies (Data Analytics specialisation) at Birkbeck College, University of London. Summer Term 2021.
This project implements a Fuzzy Inference System trained through Genetic Programming that can perform two separate tasks:
-
Classification using the FuzzyClassifier class, modelled on the scikit-learn classifiers API.
-
Reinforcement learning on OpenAI Gym environments with the GymRunner class.
The prerequisites for using the package are given in requirements.txt and requirements-dev.txt. It has been developed on python 3.9 and should work with 3.7 or 3.8, but this is untested.
Instructions for using the package are in user_guide.md and there are several demo scripts for both classification and for RL.