b559f5b97f4227c4f6f7daf4a725a1e0c71c47a2,agents/reinforce.py,REINFORCEContinuous,build_network_normal,#REINFORCEContinuous#,259

Before Change

            biases_initializer=tf.zeros_initializer())
        mu = tf.squeeze(mu, name="mu")

        sigma_L1 = tf.contrib.layers.fully_connected(
            inputs=self.states,
            num_outputs=self.config["n_hidden_units"],
            activation_fn=tf.tanh,
            weights_initializer=tf.truncated_normal_initializer(mean=0.0, stddev=0.02),
            biases_initializer=tf.zeros_initializer())
        sigma = tf.contrib.layers.fully_connected(
            inputs=sigma_L1,
            num_outputs=1,
            activation_fn=None,
            weights_initializer=tf.truncated_normal_initializer(mean=0.0, stddev=0.02),
            biases_initializer=tf.zeros_initializer())
        sigma = tf.squeeze(sigma)
        sigma = tf.nn.softplus(sigma) + 1e-5

        self.normal_dist = tf.contrib.distributions.Normal(mu, sigma)
        self.action = self.normal_dist.sample(1)
        self.action = tf.clip_by_value(self.action, self.env.action_space.low[0], self.env.action_space.high[0])

After Change

            weights_initializer=tf.truncated_normal_initializer(mean=0.0, stddev=0.02),
            biases_initializer=tf.zeros_initializer())

        mu, sigma = mu_sigma_layer(L1, 1)

        self.normal_dist = tf.contrib.distributions.Normal(mu, sigma)
        self.action = self.normal_dist.sample(1)

Instances