Split value and policy networks

ml-agents/mlagents/trainers/ppo/models.py

         :param num_layers: Number of hidden linear layers.
         """
         hidden_streams = self.create_observation_streams(
-            1, h_size, num_layers, vis_encode_type
+            2, h_size, num_layers, vis_encode_type
-        hidden = hidden_streams[0]
 
         if self.use_recurrent:
             self.prev_action = tf.placeholder(
                 ],
                 axis=1,
             )
-            hidden = tf.concat([hidden, prev_action_oh], axis=1)
+            hidden_policy = tf.concat([hidden_streams[0], prev_action_oh], axis=1)
-            hidden, memory_out = self.create_recurrent_encoder(
-                hidden, self.memory_in, self.sequence_length
+            _half_point = int(self.m_size / 2)
+            hidden_policy, memory_policy_out = self.create_recurrent_encoder(
+                hidden_policy,
+                self.memory_in[:, :_half_point],
+                self.sequence_length,
+                name="lstm_policy",
+            )
+
+            hidden_value, memory_value_out = self.create_recurrent_encoder(
+                hidden_streams[1],
+                self.memory_in[:, _half_point:],
+                self.sequence_length,
+                name="lstm_value",
+            )
+            self.memory_out = tf.concat(
+                [memory_policy_out, memory_value_out], axis=1, name="recurrent_out"
-            self.memory_out = tf.identity(memory_out, name="recurrent_out")
+        else:
+            hidden_policy = hidden_streams[0]
+            hidden_value = hidden_streams[1]
-                    hidden,
+                    hidden_policy,
                     size,
                     activation=None,
                     use_bias=False,
         self.output = tf.identity(output)
         self.normalized_logits = tf.identity(normalized_logits, name="action")
 
-        self.create_value_heads(self.stream_names, hidden)
+        self.create_value_heads(self.stream_names, hidden_value)
 
         self.action_holder = tf.placeholder(
             shape=[None, len(policy_branches)], dtype=tf.int32, name="action_holder"