"Why do we fall, sir? Because it's Reinforcement learning."

Implemented AI player in Pong game, with Custom Environment!

How does it play?

-> After 140k episodes

-> After 220k episodes

-> After 320k episodes

Loss Graph

Installation

git clone https://github.com/a-b-h-a-y-s-h-i-n-d-e/Pong-Game-with-Reinforcement-Learning.git

pip install -r requirements.txt

Play against AI

python main.py 

Train from scratch?

-> Delete the trained_model.pth and buffer.pkl files first!

cd DQN

-> select num_of_episodes in train.py

python train.py