Exemple #1
0
def cartpole_get_action(theta, ob, rng=np.random):
    """
    :param theta: A matrix of size |A| * (|S|+1)
    :param ob: A vector of size |S|
    :return: An integer
    """
    return weighted_sample(compute_logits(theta, ob), rng=rng)
def cartpole_get_action(theta, ob, rng=np.random):
    """
    :param theta: A matrix of size |A| * (|S|+1)
    :param ob: A vector of size |S|
    :return: An integer
    """
    return weighted_sample(compute_logits(theta, ob), rng=rng)
Exemple #3
0
def cartpole_get_action(theta, ob, rng=np.random):
    ob_1 = include_bias(ob)
    logits = ob_1.dot(theta.T)
    return weighted_sample(logits, rng=rng)
def cartpole_get_action(theta, ob, rng=np.random):
    ob_1 = include_bias(ob)
    logits = ob_1.dot(theta.T)
    return weighted_sample(logits, rng=rng)