Grid World with Q-Learning

We created a Smart agent looking for its path to the BIG reward. We used QLearning to train the agent.

System requirements

• Python3
• Numpy

Run the program

To Start the training, type the below command:

python3 run.py <number_of_episodes>

NOTE:

The argument <number_of_episodes> is Optional
The default value of <number_of_episodes> is 50