This repository consists of some Python codes for implementing the Q-Learning algorithm from scratch and training the agent to learn a maze.
env_map_maker.py:
This code consists of a helper function for implementing any standard mazes that have a start point, a goal point, and an inner wall.
mdp_meta_data.py:
Here, this is a helper function in which we create the metadata, such as the MDP, that we need for defining the Environment class.
grid_world_general_env.py
This piece of code defines the Environment class.
plot_environment.py:
This is a helper class that consists of some methods to make an image of the maze and to visualize the agent moves within the maze.
agents.py:
Here, we implemented a Random Agent and Q-learning Agent classes.
train_QAgent.py:
By this code, we can train the Q-learning agent to find the best policy to navigate inside the maze.
test_QAgent.py:
This is for testing the trained agent performance.
visutils.py:
This includes some helper functions to visualize the agent performance in the training phase.
training_pipline.py:
Here, we call the objects we've defined and run the training and test phases.
RL_QLearningMaze_training_pipline.ipynb:
This is similir to training_pipline.py, but in the Google colab.
Tutorial_RL_QLearning_for_Maze.ipynb:
A tutorial fof the programming session of the course.