Submission of my first coursework in Machine Learning class at King's College London.
This coursework is to manually implement a classifier without using any open-source library. The training data is in the good-moves.txt. The trained classifier will then help Pacman to survive the maze. Obviously, the result will disastrous. The good news is the classifier works.
The classifier is implemented in the "classifierAgents.py". It is a Support Vector Machine classifier.
To run Pacman with the classifier as the control, type in into the terminal:
python pacman.py --pacman ClassifierAgent
This coursework is based on the PacmanAI from UC Berkeley: http://ai.berkeley.edu/project_overview.html