Hands-On Artificial Intelligence for Beginners

Running the network 

Now, let's get to the moment we've been waiting for! Let's import gym, NumPy, our deep-q network file, as well as a few handler functions: 

import cv2import sysfrom deepQ import deepQimport numpy as npimport gym

We'll define our agent class as Atari, and initialize the environment, network, and actions with the class:

class Atari:    def __init__(self):    self.env = gym.make('SpaceInvaders-v0')    self.env.reset()    self.actions = self.env.action_space.n    self.deepQ = deepQ(self.actions)    self.action0 = 0

Our Deep Q-network can't innately ingest the Atari games, so we have to write a bit of preprocessing code to handle the video input. We'll call this function preprocess and it will take in a single game observation: 

def preprocess(self,observation): ...