We need to evaluate the agent to see how well it has learned to play the game. We just trained our agent to navigate through the world to reach the target. In this recipe, we will evaluate the trained Malmo agent.
Evaluating a Malmo agent
Getting ready
As a prerequisite, we will need to persist the agent policies and reload them back during evaluation.
The final policy (policy to make movements in Malmo space) used by the agent after training can be saved as shown here:
DQNPolicy<MalmoBox> pol = dql.getPolicy();
pol.save("cliffwalk_pixel.policy");
dql refers to the DQN model. We retrieve the final policies and store them as a DQNPolicy. A DQN policy provides actions that have the highest Q-value estimated...
