PyTorch 1.x Reinforcement Learning Cookbook

How to do it...

We solve the Mountain Car problem using DDQNs as follows:

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

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

2. Next, we define the DDQN model as follows:

>>> class DuelingModel(nn.Module): ...     def __init__(self, n_input, n_output, n_hidden): ...         super(DuelingModel, self).__init__() ...         self.adv1 = nn.Linear(n_input, n_hidden) ...         self.adv2 = nn.Linear(n_hidden, n_output) ...         self.val1 = nn.Linear(n_input, n_hidden) ...         self.val2 = nn.Linear(n_hidden, 1) ... ...     def forward(self, x): ... adv = nn.functional.relu(self.adv1(x)) ...