Building a Custom Environment for Deep Reinforcement Learning with OpenAI Gym and Python

This beginner course demonstrates how to build a custom reinforcement learning environment with OpenAI Gym and Python, including its __init__, step, and reset methods. It then trains and tests a DQN agent on the environment using Keras-RL.

Syllabus

- Start
- Cloning Baseline Reinforcement Learning Code
- Custom Environment Blueprint and Scenario
- Installing and Importing Dependencies
- Creating a Custom Environment with OpenAI Gym
- Coding the __init__ method for a OpenAI Environment
- Coding the step method for an OpenAI Environment
- Coding the reset method for an OpenAI Environment
- Testing a Custom OpenAI Environment
- Training a DQN Agent with Keras-RL
- Running a DQN Agent on a Custom Environment using Keras-RL