Mini project where I had to implement a solution of the Lunar Lander DQN problem !
The Lunar lander is a classic beginner problem of Deep-Q networks. Here the environement is simulated with Gym and the DQN is made with Tensorflow and Keras. You'll find here my code. It has comments in French.
Expect like 300 ~ 400 episodes for the Agent (the DQN) to be good.
Link to the Google Colab: https://colab.research.google.com/drive/1hV2j0GioumeIuJaHdn_3sol-7gZes0_3