There are 5 methods you can try, namely policy iteration, value iteration, Monte Carlo, SARSA, and Q-learning, with algorithms.py file. I2C is implementation for [DeepMind's Paper] Imagination Augmented Agents for Deep Reinforcement Learning. Before running
pip install gym
pip install gym_sokoban
https://github.com/thanhthanhhp123/Sokoban_Env.git