Lunar Lander-V2 using DQN and Actor Critic Methods

Setting up project environment:

• Create new enviroment using anaconda or miniconda

conda create -n env_name pthon=3.6

• Install dependencies from requirements.txt

pip install -r requirements.txt

After installing dependencies, DQN and Actor Critic implemented in respective folders.

DQN:

• DQN algorithm is implemented in DQN.py
• For training and testing the DQN, use the dqn_run.ipynb
• DQN testing results was stored in DQN results folder
• DQN trained model was stored in DQN model folder

Actor Critic:

• Actor critic algorithm is implemented in main.py
• For training and testing the Actor Critic use the actor_critic_run.ipynb folder
• Actor Critic testing results was stored in actor critic results folder.
• Actor critic model is stored in Actor Critic