Reinforcement-Learning-Deep-Q-Learning

This repo contains implementation of Solving a Landing problem of LunarLander (OpenAI) game by implementing Deep Q-Learning using TensorFlow and Python.

OpenAI’s Lunar Lander problem, an 8-dimensional state space and 4-dimensional action space problem. The goal was to create an agent that can guide a space vehicle to land autonomously in the environment without crashing.

Installing required libraries.

pip3 install -r Requirements.txt

Running the files

Main file: Environment_setting_DQN.py is where we set the environment we want the agent to be trained in. Also here we decide the parameters that we want to set before training the agent.

Set train_dqn = 1 to train the deep q network.

Other files

Train_MyRLAgent_DQN.py is called to start training the agent

QLearning_Agent_DQN.py is where the Agent is created

My_DQN.py is where the deep learning network is structured.

Plots

Plots of rewards, averaged rewards , steps and epsilo verses the number of episoded are created to show whether the agent is correctly getting trained.