New arguments and argument handling to specify 0,1,2 demo count for D…

…emPref.
HARPLab · Jun 16, 2022 · ec181c0 · ec181c0
1 parent 16ad4ce
commit ec181c0
Show file tree

Hide file tree

Showing 4 changed files with 528 additions and 133 deletions.
diff --git a/inquire/environments/lunar_lander.py b/inquire/environments/lunar_lander.py
@@ -109,8 +109,8 @@ def generate_random_state(self, random_state):
     def generate_random_reward(self, random_state):
         """Randomly generate a weight vector for trajectory features."""
         # reward = np.array([-0.4, 0.4, -0.2, -0.7])
-        # reward = np.array([0.55, 0.55, 0.41, 0.48])
-        reward = np.random.uniform(-1,1,4)
+        reward = np.array([0.55, 0.55, 0.41, 0.48])
+        # reward = np.random.uniform(-1, 1, 4)
         return reward / np.linalg.norm(reward)
 
     def reset(self) -> np.ndarray: