Chapter06 Windy Grid World

using ReinforcementLearning, ReinforcementLearningEnvironments
using RLIntro, RLIntro.WindyGridWorld

env = WindyGridWorldEnv()
ns = length(observation_space(env))
na = length(action_space(env))

agent = Agent(
    π=QBasedPolicy(
        learner=TDLearner(
            approximator=TabularQApproximator(;n_state=ns, n_action=na),
            optimizer=Descent(0.5)
        ),
        selector=EpsilonGreedySelector(0.1)
    ),
    buffer=episode_RTSA_buffer()
);

hook = StepsPerEpisode()
run(agent, env, StopAfterStep(8000);hook=hook)

using Plots
plot([i for (i, x) in enumerate(hook.steps) for _ in 1:x])