Working through and exploring the code in this book is meant to be a learning exercise in how Reinforcement Learning (RL) algorithms work but also how difficult it can be to get them to work. It is because of this difficulty that so many open source RL frameworks seem to pop up every day. In this chapter, we will explore a couple of the more popular frameworks. We will start with why you would want to use a framework and then move on to exploring the more popular frameworks such as Dopamine, Keras-RL, TF-Agents, and RL Lib.
Here is a quick summary of the main topics we will cover in this chapter:
- Choosing a framework
- Introducing Google Dopamine
- Playing with Keras-RL
- Exploring RL Lib
- Using TF agents
We will use a combination of notebook environments on Google Colab and virtual environments depending on the complexity of the examples in this chapter. Jupyter Notebooks...
