6efe8ce53b30db7dd08dd727b23d6acb11796a8a,examples/grid_world.py,,,#,54

Before Change

    np.save("rQ1.npy", np.mean(rewardQ1, axis=0))
    np.save("rQ08.npy", np.mean(rewardQ08, axis=0))
    np.save("rDQ1.npy", np.mean(rewardDQ1, axis=0))
    np.save("rDQ08.npy", np.mean(rewardDQ08, axis=0))

After Change

    logger.Logger(3)

    names = {1:"1", 0.8:"08", QLearning:"Q", DoubleQLearning:"DQ"}
    for e in [1, .8]:
        for a in [QLearning, DoubleQLearning]:
            r = Parallel(n_jobs=-1)(
                delayed(experiment)(a, e) for _ in xrange(n_experiment))

Instances