statereg

curl_sac.py

@@ -71,9 +71,9 @@ class Actor(nn.Module):
 
     def forward(
         self, obs, compute_pi=True, compute_log_pi=True, detach_encoder=False
-    ):
+    ):  
         obs = self.encoder(obs, detach=detach_encoder)
-
+    
         mu, log_std = self.trunk(obs).chunk(2, dim=-1)
 
         # constrain log_std inside [log_std_min, log_std_max]

scripts/run.sh

@@ -1,7 +1,7 @@
 CUDA_VISIBLE_DEVICES=5 python train.py \
     --domain_name cartpole \
     --task_name swingup \
-    --encoder_type pixel \
+    --encoder_type identity \
     --action_repeat 8 \
     --save_tb --pre_transform_image_size 100 --image_size 84 \
     --work_dir ./tmp/test \

utils.py

@@ -203,6 +203,7 @@ class StateMask(gym.Wrapper):
     def __init__(self, env, pos_dim):
         gym.Wrapper.__init__(self, env)
         self.pos_dim = pos_dim
+        self._max_episode_steps = env._max_episode_steps
         self.observation_space = gym.spaces.Box(
             low=-np.inf,
             high=np.inf,