Python Machine Learning Cookbook - Second Edition

How to do it...

Let's see how we can develop an AI-based dynamic modeling system:

1. We will use the dqn_cartpole.py file that is already provided for you as a reference. Let's start by importing the libraries:

import numpy as npimport gymfrom keras.models import Sequentialfrom keras.layers import Dense, Activation, Flattenfrom keras.optimizers import Adamfrom rl.agents.dqn import DQNAgentfrom rl.policy import BoltzmannQPolicyfrom rl.memory import SequentialMemory

2. Now, we will define and load the environment:

ENV_NAME = 'CartPole-v0'env = gym.make(ENV_NAME)

3. To set the seed value, the NumPy library's random.seed() function is used, as follows:

np.random.seed(123)env.seed(123)

4. Now, we will extract the actions that are available to the ...