Scale window rewards between 0 and 10 (Farama-Foundation#340)

* Scale window rewards between 0 and 10 * Fix scaling on dial-turn env
Usaywook · Apr 2, 2021 · 1ce1848 · 1ce1848
1 parent 48f6f68
commit 1ce1848
Show file tree

Hide file tree

Showing 3 changed files with 3 additions and 3 deletions.
diff --git a/metaworld/envs/mujoco/sawyer_xyz/v2/sawyer_dial_turn_v2.py b/metaworld/envs/mujoco/sawyer_xyz/v2/sawyer_dial_turn_v2.py
@@ -131,7 +131,7 @@ def compute_reward(self, action, obs):
         tcp_opened = 0
         object_grasped = reach
 
-        reward = reward_utils.hamacher_product(reach, in_place)
+        reward = 10 * reward_utils.hamacher_product(reach, in_place)
 
         return (reward,
                tcp_to_obj,

diff --git a/metaworld/envs/mujoco/sawyer_xyz/v2/sawyer_window_close_v2.py b/metaworld/envs/mujoco/sawyer_xyz/v2/sawyer_window_close_v2.py
@@ -140,7 +140,7 @@ def compute_reward(self, actions, obs):
         tcp_opened = 0
         object_grasped = reach
 
-        reward = reward_utils.hamacher_product(reach, in_place)
+        reward = 10 * reward_utils.hamacher_product(reach, in_place)
 
         return (reward,
                tcp_to_obj,

diff --git a/metaworld/envs/mujoco/sawyer_xyz/v2/sawyer_window_open_v2.py b/metaworld/envs/mujoco/sawyer_xyz/v2/sawyer_window_open_v2.py
@@ -130,7 +130,7 @@ def compute_reward(self, actions, obs):
         tcp_opened = 0
         object_grasped = reach
 
-        reward = reward_utils.hamacher_product(reach, in_place)
+        reward = 10 * reward_utils.hamacher_product(reach, in_place)
         return (reward,
                tcp_to_obj,
                tcp_opened,