def main():
model = PolicyGradient(
gym.make('CartPole-v1'),
n_in=4,
hidden_net=lambda x: tf.layers.dense(x, 10, tf.nn.elu),
n_out=2)
model.learn()
model.play()
def main():
model = PolicyGradient()
model.learn(gym.make("CartPole-v0"))
model.play(gym.make("CartPole-v0"))
def main():
model = PolicyGradient(gym.make('CartPole-v1'),
n_in=4, n_hidden=[10], n_out=2)
model.learn()
model.play()
def main():
model = PolicyGradient(lambda x: tf.layers.dense(x, 4, tf.nn.relu))
model.learn(gym.make("CartPole-v0"))
model.play(gym.make("CartPole-v0"))