This is the implementation of training a game using Q-learning. The game is made using pygame.

The main code is in dqn_game.py

It is not pre-trained. As the code is run the game starts training the agent, and it gets better and better as the score increases.