add fire clean up docstrings in create policies (#4391)

Author: andrewcoh

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

@@ -216,9 +216,10 @@ def create_tf_policy(
         create_graph: bool = False,
     ) -> TFPolicy:
         """
-        Creates a PPO policy to trainers list of policies.
+        Creates a policy with a Tensorflow backend and PPO hyperparameters
+        :param parsed_behavior_id:
         :param behavior_spec: specifications for policy construction
-        :param create_graph: whether to create the graph when policy is constructed
+        :param create_graph: whether to create the Tensorflow graph on construction
         :return policy
         """
         policy = TFPolicy(
@@ -234,9 +235,9 @@ def create_torch_policy(
         self, parsed_behavior_id: BehaviorIdentifiers, behavior_spec: BehaviorSpec
     ) -> TorchPolicy:
         """
-        Creates a PPO policy to trainers list of policies.
+        Creates a policy with a PyTorch backend and PPO hyperparameters
         :param parsed_behavior_id:
-        :param brain_parameters: specifications for policy construction
+        :param behavior_spec: specifications for policy construction
         :return policy
         """
         policy = TorchPolicy(

ml-agents/mlagents/trainers/sac/trainer.py

@@ -233,6 +233,13 @@ def create_tf_policy(
         behavior_spec: BehaviorSpec,
         create_graph: bool = False,
     ) -> TFPolicy:
+        """
+        Creates a policy with a Tensorflow backend and SAC hyperparameters
+        :param parsed_behavior_id:
+        :param behavior_spec: specifications for policy construction
+        :param create_graph: whether to create the Tensorflow graph on construction
+        :return policy
+        """
         policy = TFPolicy(
             self.seed,
             behavior_spec,
@@ -248,7 +255,7 @@ def create_torch_policy(
         self, parsed_behavior_id: BehaviorIdentifiers, behavior_spec: BehaviorSpec
     ) -> TorchPolicy:
         """
-        Creates a PPO policy to trainers list of policies.
+        Creates a policy with a PyTorch backend and SAC hyperparameters
         :param parsed_behavior_id:
         :param behavior_spec: specifications for policy construction
         :return policy