This is a RL project based on Snake!
For a quick run just run: python3 agent.py and see the base DQN algorithm training process.
Presentation is in .pdf file
Project Strucrure:
SNAKE:
Environment:
game.py
Model:
model.py
ddqn_model.py
Agent:
agent.py
ddqn_agent.py (actually also training process handlers)
Helper and data management:
plotCSV.py
helper.py