Build your own model in RL module
How to build the model
To build the model you’ll need to inherit the Algorithm class from the RL module documentation and implement all of the methods
How to test if the model was implemented correctly
Model should be created in algorithms folder.
Model needs to be registered by AlgorithmManager
Test if the model works on Freecell or 2048 Game
Your model should work for every game, not only on specific games
Model example (DQN)
Because DQN implementation is long we recommend that you the implementation on GitHub.
The model is based on DQN tutorial made by PyTorch team.
Our implementation of DQN penalizes illegal moves