- Youtube link : https://www.youtube.com/watch?v=0CNezkeuNr4&t=1s
- Requirments :
- Make sure you've installed a library "pygame". This game is made using the library.
- "Tensorflow" must be installed.
- Explaination for the attached files :
- bird_game.py : the game environment file made using the library "pygame"
- bird_agent.py : contains game-proceeding-part. So-called, agent file.
- DDDQN.py :
- How to run the program/files? / How to use it?
- Open 'bird_agent.py' and scroll down to '#Execute' section.
- Type 'train' or 'replay' next to "train_replay_option =".
- While training, the trained models will be stored in the "model_saved" folder.
- If you'd like to change some of the hyperparameters, you can.