906ba34449e0f19b7f3b674d5e005a1903d0642e,examples/mountain_car.py,,experiment,#Any#,20

Before Change

    n_tilings = 10
    tiles_dim = 10
    offset = 1. / (n_tilings * tiles_dim)
    for i in xrange(10):
        x_min = mdp.info.observation_space.low - (n_tilings - 1 - i) * offset
        x_max = mdp.info.observation_space.high + i * offset
        x_range = [[x, y] for x, y in zip(x_min, x_max)]
        tilings.append(Tiles(x_range, [n_tilings, n_tilings]))
    features = Features(tilings=tilings)

    approximator_params = dict(input_shape=(features.size,),
                               output_shape=(mdp.info.action_space.n,),

After Change

    // Agent
    learning_rate = Parameter(alpha)
    tilings = Tiles.generate(10, [10, 10],
                             mdp.info.observation_space.low,
                             mdp.info.observation_space.high)
    features = Features(tilings=tilings)

    approximator_params = dict(input_shape=(features.size,),
                               output_shape=(mdp.info.action_space.n,),

Instances