modify metaworld to include done - finish v1

Author: lzyang2000

cs285/agents/__pycache__/dqn_agent.cpython-36.pyc

@@ -1,4 +1,3 @@
-README.md
 setup.py
 cs285.egg-info/PKG-INFO
 cs285.egg-info/SOURCES.txt

cs285/data/hw3_ check_CartPole-v0_26-10-2020_01-25-41/events.out.tfevents.1603700741.Bennys-MacBook-Pro.local

@@ -5,10 +5,4 @@
     name='cs285',
     version='0.1.0',
     packages=['cs285'],
-)
-
-setup(
-    name='metaworld',
-    version='2',
-    packages=['metaworld'],
 )

cs285/data/hw3_ todo_CartPole-v0_26-10-2020_00-53-47/events.out.tfevents.1603698827.Bennys-MacBook-Pro.local

@@ -45,7 +45,10 @@ def step(self, action):
         ob = super().step(action)
         reward, reachDist, pressDist = self.compute_reward(action, ob)
         self.curr_path_length += 1
-
+        if self.curr_path_length == self.max_path_length:	
+            done = True	
+        else:	
+            done = False
         info = {
             'reachDist': reachDist,
             'goalDist': pressDist,
@@ -54,7 +57,7 @@ def step(self, action):
             'success': float(pressDist <= 0.04)
         }
 
-        return ob, reward, False, info
+        return ob, reward, done, info
 
     @property
     def _target_site_config(self):

cs285/data/hw3_ todo_CartPole-v0_26-10-2020_01-20-13/events.out.tfevents.1603700413.Bennys-MacBook-Pro.local

@@ -45,7 +45,10 @@ def step(self, action):
         ob = super().step(action)
         reward, reachDist, pressDist = self.compute_reward(action, ob)
         self.curr_path_length += 1
-
+        if self.curr_path_length == self.max_path_length:	
+            done = True	
+        else:	
+            done = False
         info = {
             'reachDist': reachDist,
             'goalDist': pressDist,
@@ -54,7 +57,7 @@ def step(self, action):
             'success': float(pressDist <= 0.04)
         }
 
-        return ob, reward, False, info
+        return ob, reward, done, info
 
     @property
     def _target_site_config(self):

cs285/data/hw3_check_pick-place-v1_26-10-2020_01-26-42/events.out.tfevents.1603700802.Bennys-MacBook-Pro.local

@@ -45,7 +45,10 @@ def step(self, action):
         ob = super().step(action)
         reward, reachDist, pressDist = self.compute_reward(action, ob)
         self.curr_path_length += 1
-
+        if self.curr_path_length == self.max_path_length:	
+            done = True	
+        else:	
+            done = False
         info = {
             'reachDist': reachDist,
             'goalDist': pressDist,
@@ -54,7 +57,7 @@ def step(self, action):
             'success': float(pressDist <= 0.04)
         }
 
-        return ob, reward, False, info
+        return ob, reward, done, info
 
     @property
     def _target_site_config(self):

cs285/data/hw3_check_pick-place-v1_26-10-2020_01-48-25/events.out.tfevents.1603702105.Bennys-MacBook-Pro.local

@@ -46,7 +46,10 @@ def step(self, action):
         ob = super().step(action)
         reward, reachDist, pullDist = self.compute_reward(action, ob)
         self.curr_path_length += 1
-
+        if self.curr_path_length == self.max_path_length:	
+            done = True	
+        else:	
+            done = False
         info = {
             'reachDist': reachDist,
             'goalDist': pullDist,
@@ -55,7 +58,7 @@ def step(self, action):
             'success': float(pullDist <= 0.05)
         }
 
-        return ob, reward, False, info
+        return ob, reward, done, info
 
     def _get_pos_objects(self):
         return self._get_site_pos('leverStart')

cs285/data/hw3_check_pick-place-v1_26-10-2020_01-48-41/events.out.tfevents.1603702121.Bennys-MacBook-Pro.local

@@ -55,7 +55,10 @@ def step(self, action):
         ob = super().step(action)
         reward, _, reachDist, pickRew, _, placingDist = self.compute_reward(action, ob)
         self.curr_path_length += 1
-
+        if self.curr_path_length == self.max_path_length:	
+            done = True	
+        else:	
+            done = False
         info = {
             'reachDist': reachDist,
             'pickRew': pickRew,
@@ -64,7 +67,7 @@ def step(self, action):
             'success': float(placingDist <= 0.07)
         }
 
-        return ob, reward, False, info
+        return ob, reward, done, info
 
     def _get_pos_objects(self):
         return self.get_body_com('peg')

cs285/data/hw3_check_pick-place-v1_26-10-2020_01-48-53/events.out.tfevents.1603702133.Bennys-MacBook-Pro.local

@@ -48,7 +48,10 @@ def step(self, action):
         ob = super().step(action)
         reward, _, reachDist, pickRew, _, placingDist = self.compute_reward(action, ob)
         self.curr_path_length += 1
-
+        if self.curr_path_length == self.max_path_length:	
+            done = True	
+        else:	
+            done = False
         info = {
             'reachDist': reachDist,
             'pickRew': pickRew,
@@ -57,7 +60,7 @@ def step(self, action):
             'success': float(placingDist <= 0.07)
         }
 
-        return ob, reward, False, info
+        return ob, reward, done, info
 
     def _get_pos_objects(self):
         return self._get_site_pos('pegEnd')

cs285/data/hw3_check_pick-place-v1_26-10-2020_01-49-16/events.out.tfevents.1603702156.Bennys-MacBook-Pro.local

@@ -50,7 +50,10 @@ def step(self, action):
         ob = super().step(action)
         reward, reachDist, pickRew, placingDist = self.compute_reward(action, ob)
         self.curr_path_length +=1
-
+        if self.curr_path_length == self.max_path_length:	
+            done = True	
+        else:	
+            done = False
         info = {
             'reachDist': reachDist,
             'goalDist': placingDist,
@@ -59,7 +62,7 @@ def step(self, action):
             'success': float(placingDist <= 0.08)
         }
 
-        return ob, reward, False, info
+        return ob, reward, done, info
 
     def _get_pos_objects(self):
         return self.data.get_geom_xpos('objGeom')

cs285/data/hw3_check_pick-place-v1_26-10-2020_01-54-13/events.out.tfevents.1603702453.Bennys-MacBook-Pro.local

@@ -47,7 +47,10 @@ def step(self, action):
         ob = super().step(action)
         reward, reachDist, pullDist = self.compute_reward(action, ob)
         self.curr_path_length += 1
-
+        if self.curr_path_length == self.max_path_length:	
+            done = True	
+        else:	
+            done = False
         info = {
             'reachDist': reachDist,
             'goalDist': pullDist,
@@ -56,7 +59,7 @@ def step(self, action):
             'success': float(pullDist <= 0.08)
         }
 
-        return ob, reward, False, info
+        return ob, reward, done, info
 
     def _get_pos_objects(self):
         return self.data.get_geom_xpos('objGeom')

cs285/data/hw3_check_pick-place-v1_26-10-2020_01-54-57/events.out.tfevents.1603702497.Bennys-MacBook-Pro.local

@@ -47,7 +47,10 @@ def step(self, action):
         ob = super().step(action)
         reward, reachDist, pullDist = self.compute_reward(action, ob)
         self.curr_path_length += 1
-
+        if self.curr_path_length == self.max_path_length:	
+            done = True	
+        else:	
+            done = False
         info = {
             'reachDist': reachDist,
             'goalDist': pullDist,
@@ -56,7 +59,7 @@ def step(self, action):
             'success': float(pullDist <= 0.07)
         }
 
-        return ob, reward, False, info
+        return ob, reward, done, info
 
     def _get_pos_objects(self):
         return self.data.get_geom_xpos('objGeom')

cs285/data/hw3_check_pick-place-v1_26-10-2020_01-55-33/events.out.tfevents.1603702533.Bennys-MacBook-Pro.local

@@ -47,7 +47,10 @@ def step(self, action):
         ob = super().step(action)
         reward, reachDist, pullDist = self.compute_reward(action, ob)
         self.curr_path_length += 1
-
+        if self.curr_path_length == self.max_path_length:	
+            done = True	
+        else:	
+            done = False
         info = {
             'reachDist': reachDist,
             'goalDist': pullDist,
@@ -56,7 +59,7 @@ def step(self, action):
             'success': float(pullDist <= 0.07)
         }
 
-        return ob, reward, False, info
+        return ob, reward, done, info
 
     def _get_pos_objects(self):
         return self.data.get_geom_xpos('objGeom')

cs285/data/hw3_check_pick-place-v1_26-10-2020_01-56-18/events.out.tfevents.1603702578.Bennys-MacBook-Pro.local

@@ -47,7 +47,10 @@ def step(self, action):
         ob = super().step(action)
         reward, reachDist, pullDist = self.compute_reward(action, ob)
         self.curr_path_length += 1
-
+        if self.curr_path_length == self.max_path_length:	
+            done = True	
+        else:	
+            done = False
         info = {
             'reachDist': reachDist,
             'goalDist': pullDist,
@@ -56,7 +59,7 @@ def step(self, action):
             'success': float(pullDist <= 0.08)
         }
 
-        return ob, reward, False, info
+        return ob, reward, done, info
 
     def _get_pos_objects(self):
         return self.data.get_geom_xpos('objGeom')

cs285/data/hw3_check_pick-place-v1_26-10-2020_02-05-12/events.out.tfevents.1603703112.Bennys-MacBook-Pro.local

@@ -46,7 +46,10 @@ def step(self, action):
         ob = super().step(action)
         reward, reachDist, pushDist = self.compute_reward(action, ob)
         self.curr_path_length += 1
-
+        if self.curr_path_length == self.max_path_length:	
+            done = True	
+        else:	
+            done = False
         info = {
             'reachDist': reachDist,
             'goalDist': pushDist,
@@ -55,7 +58,7 @@ def step(self, action):
             'success': float(pushDist <= 0.07)
         }
 
-        return ob, reward, False, info
+        return ob, reward, done, info
 
     def _get_pos_objects(self):
         return self.data.get_geom_xpos('objGeom')

cs285/data/hw3_check_pick-place-v1_26-10-2020_02-05-44/events.out.tfevents.1603703144.Bennys-MacBook-Pro.local

@@ -70,7 +70,10 @@ def step(self, action):
         ob = super().step(action)
         reward, _, reachDist, _, pushDist, pickRew, _, placingDist = self.compute_reward(action, ob)
         self.curr_path_length +=1
-
+        if self.curr_path_length == self.max_path_length:	
+            done = True	
+        else:	
+            done = False
         goal_dist = placingDist if self.task_type == 'pick_place' else pushDist
 
         if self.task_type == 'reach':
@@ -86,7 +89,7 @@ def step(self, action):
             'success': success
         }
 
-        return ob, reward, False, info
+        return ob, reward, done, info
 
     @property
     def _target_site_config(self):

cs285/data/hw3_check_pick-place-v1_26-10-2020_02-12-19/events.out.tfevents.1603703539.Bennys-MacBook-Pro.local

@@ -70,6 +70,10 @@ def step(self, action):
         ob = super().step(action)
         reward, _, reachDist, _, pushDist, pickRew, _, placingDist = self.compute_reward(action, ob)
         self.curr_path_length +=1
+        if self.curr_path_length == self.max_path_length:	
+            done = True	
+        else:	
+            done = False
         goal_dist = placingDist if self.task_type == 'pick_place' else pushDist
 
         if self.task_type == 'reach':
@@ -85,7 +89,7 @@ def step(self, action):
             'success': success
         }
 
-        return ob, reward, False, info
+        return ob, reward, done, info
 
     @property
     def _target_site_config(self):

cs285/data/hw3_check_pick-place-v1_26-10-2020_02-13-13/events.out.tfevents.1603703593.Bennys-MacBook-Pro.local

@@ -55,7 +55,10 @@ def step(self, action):
         ob = super().step(action)
         reward, _, reachDist, pickRew, _, placingDist = self.compute_reward(action, ob)
         self.curr_path_length += 1
-
+        if self.curr_path_length == self.max_path_length:	
+            done = True	
+        else:	
+            done = False
         info = {
             'reachDist': reachDist,
             'pickRew': pickRew,
@@ -64,7 +67,7 @@ def step(self, action):
             'success': float(placingDist <= 0.08)
         }
 
-        return ob, reward, False, info
+        return ob, reward, done, info
 
     def _get_pos_objects(self):
         return self.data.get_geom_xpos('objGeom')

cs285/data/hw3_check_pick-place-v1_26-10-2020_02-17-48/events.out.tfevents.1603703868.Bennys-MacBook-Pro.local

@@ -46,7 +46,10 @@ def step(self, action):
         ob = super().step(action)
         reward, reachDist, pushDist = self.compute_reward(action, ob)
         self.curr_path_length += 1
-
+        if self.curr_path_length == self.max_path_length:	
+            done = True	
+        else:	
+            done = False
         info = {
             'reachDist': reachDist,
             'goalDist': pushDist,
@@ -55,7 +58,7 @@ def step(self, action):
             'success': float(pushDist <= 0.07)
         }
 
-        return ob, reward, False, info
+        return ob, reward, done, info
 
     def _get_pos_objects(self):
         return self.data.get_geom_xpos('objGeom')

cs285/data/hw3_check_pick-place-v1_26-10-2020_02-19-34/events.out.tfevents.1603703974.Bennys-MacBook-Pro.local

@@ -52,7 +52,10 @@ def step(self, action):
         ob = super().step(action)
         reward, _, reachDist, pickRew, _, pullDist, _ = self.compute_reward(action, ob)
         self.curr_path_length += 1
-
+        if self.curr_path_length == self.max_path_length:	
+            done = True	
+        else:	
+            done = False
         info = {
             'reachDist': reachDist,
             'pickRew': pickRew,
@@ -61,7 +64,7 @@ def step(self, action):
             'success': float(pullDist <= 0.08 and reachDist <= 0.05)
         }
 
-        return ob, reward, False, info
+        return ob, reward, done, info
 
     def _get_pos_objects(self):
         return np.hstack((

cs285/data/hw3_check_pick-place-v1_26-10-2020_02-57-37/events.out.tfevents.1603706257.Bennys-MacBook-Pro.local

@@ -49,7 +49,10 @@ def step(self, action):
         ob = super().step(action)
         reward, _, reachDist, pickRew, _, pushDist = self.compute_reward(action, ob)
         self.curr_path_length += 1
-
+        if self.curr_path_length == self.max_path_length:	
+            done = True	
+        else:	
+            done = False
         info = {
             'reachDist': reachDist,
             'pickRew': pickRew,
@@ -58,7 +61,7 @@ def step(self, action):
             'success': float(pushDist <= 0.1 and reachDist <= 0.05)
         }
 
-        return ob, reward, False, info
+        return ob, reward, done, info
 
     @property
     def _target_site_config(self):