Solving the Taxi problem with SARSA

In this recipe, we will solve the Taxi environment with the SARSA algorithm and fine-tune the hyperparameters with the grid search algorithm.

We will start with our default set of hyperparameter values under the SARSA model. These are selected based on intuition and a number of trials. Moving on, we will come up with the best set of values.

How to do it...

We perform SARSA to solve the Taxi environment as follows:

1. Import PyTorch and the gym module, and create an instance of the Taxi environment:

>>> import torch
>>> import gym
>>> env = gym.make('Taxi-v2')

2. Then, start defining the epsilon-greedy behavior policy. We will reuse the gen_epsilon_greedy_policy...