Chapter 3, Setting Up Your First Environment with OpenAI Gym

1. You can do this by cloning the source instead of installing the package from pip. Further instructions are available in the Gym documentation.
2. The term state is commonly used in the terminology of solving Markov decision processes, and the term observation is often used when describing RL environment state spaces. Both terms are equivalent in this context.
3. Calling env.reset() resets the environment's state and returns the environment's current observation or state variable.
4. The task will end when the done variable is set to True or the reward is set to 20, depending on your implementation. Both conditions indicate that the task has been solved.
5. Setting env.s will manually change the state of the environment. This is bad practice when implementing an RL strategy; the state should not be set manually when solving...