.

metastable_baselines/ppo/policies.py

@@ -88,7 +88,7 @@ class ActorCriticPolicy(BasePolicy):
         activation_fn: Type[nn.Module] = nn.Tanh,
         ortho_init: bool = True,
         use_sde: bool = False,
-        log_std_init: float = 0.0,
+        std_init: float = 1.0,
         full_std: bool = True,
         sde_net_arch: Optional[List[int]] = None,
         use_expln: bool = False,
@@ -135,7 +135,7 @@ class ActorCriticPolicy(BasePolicy):
         self.features_dim = self.features_extractor.features_dim
 
         self.normalize_images = normalize_images
-        self.log_std_init = log_std_init
+        self.log_std_init = math.log(std_init)
         # Keyword arguments for gSDE distribution
         if dist_kwargs == None:
             dist_kwargs = {}

metastable_baselines/sac/policies.py

@@ -265,7 +265,7 @@ class SACPolicy(BasePolicy):
         net_arch: Optional[Union[List[int], Dict[str, List[int]]]] = None,
         activation_fn: Type[nn.Module] = nn.ReLU,
         use_sde: bool = False,
-        log_std_init: float = -3,
+        std_init: float = 0.05,
         sde_net_arch: Optional[List[int]] = None,
         use_expln: bool = False,
         clip_mean: float = 2.0,
@@ -314,7 +314,7 @@ class SACPolicy(BasePolicy):
 
         sde_kwargs = {
             "use_sde": use_sde,
-            "log_std_init": log_std_init,
+            "log_std_init": math.log(std_init),
             "use_expln": use_expln,
             "clip_mean": clip_mean,
         }