CommencerCommencer gratuitement

Computing state-values for a policy

Using the same deterministic environment MyGridWorld, now you need to evaluate the effectiveness of the policy you defined in the previous exercise. You'll do this by computing the state value function for each state under this policy.

The environment has been imported as env along with the necessary variables needed (terminal_state, num_states, policy, gamma).

Cet exercice fait partie du cours

Reinforcement Learning with Gymnasium in Python

Afficher le cours

Instructions

  • Complete the function compute_state_value() to compute the value for each state under the given policy.
  • Create a state_values dictionary where each key is the state, and each value is the state value.

Exercice interactif pratique

Essayez cet exercice en complétant cet exemple de code.

# Complete the function
def compute_state_value(state):
    if state == terminal_state:
        return ____
    action = ____
    _, next_state, reward, _ = env.unwrapped.P[state][action][0]
    return ____

# Compute all state values 
state_values = {____: ____ for ____ in range(____)}

print(state_values)
Modifier et exécuter le code