OpenAI-GYM-Examples

Implementation of the basic examples from OpenAI GYM environment for Reainforcement Learning

Dependencies (it is recomended to have conda environment installed and sourced with base python ML packages eg. numpy:

conda install tensorflow

pip3 install --upgrade gym

Structure:

├── Cart Pole example
│   ├── CartPole  -- core module for basic policy and Policy Gradients
│   ├── show_rewards  -- module for plotting accumulated rewards over games
│   ├── model_trained  -- folder with saved neural network weights
├── MsPacman
│   ├── MsPacman -- Deep Q-Learning MsPacman game with tensorflow
├── Q-Learning
│   ├── q_learning  -- core module for simple q-learning example (board game)
│   ├── show_exploration  -- module for live exploration visualization

Q-Learning

Cart Pole

MsPacman