Advanced Deep Learning with Python

Behavioral cloning with PyTorch

In this section, we'll implement a behavioral cloning example with PyTorch 1.3.1. To help us with this task, we'll use OpenAI Gym (https://gym.openai.com/), which is an open source toolkit for the development and comparison of reinforcement learning algorithms. It allows us to teach agents to undertake various tasks, such as walking or playing games such as Pong, Pinball, some other Atari games, and even Doom.

We can install it with pip:

pip install gym[box2d]

In this example, we'll use the CarRacing-v0 OpenAI Gym environment, as shown in the following screenshot:

In the CarRacing-v0 environment, the agent is ...