PyTorch 1.x Reinforcement Learning Cookbook

How to do it...

We develop deep Q-learning using double DQNs as follows:

1. Import the necessary modules and create a Mountain Car environment:

 >>> import gym >>> import torch >>> from collections import deque >>> import random >>> import copy >>> from torch.autograd import Variable >>> env = gym.envs.make("MountainCar-v0")

2. To incorporate the target network in the experience replay phase, we first initialize it in the __init__ method of the DQN class:

>>> class DQN(): ...     def __init__(self, n_state, n_action,                      n_hidden=50, lr=0.05): ...         self.criterion = torch.nn.MSELoss() ...         self.model = torch.nn.Sequential( ...                         torch.nn.Linear(n_state, n_hidden), ...                         torch.nn.ReLU(), ...                         torch.nn.Linear(n_hidden, n_action) ...                 ) ... self.optimizer = ...