Hotfix for #331 (#342)

vwxyzjn · Jan 5, 2023 · 94a44b5 · 94a44b5 · vercel · Jan 5, 2023
1 parent a90742a
commit 94a44b5
Show file tree

Hide file tree

Showing 2 changed files with 12 additions and 2 deletions.
diff --git a/cleanrl/rpo_continuous_action.py b/cleanrl/rpo_continuous_action.py
@@ -135,12 +135,12 @@ def get_action_and_value(self, x, action=None):
         probs = Normal(action_mean, action_std)
         if action is None:
             action = probs.sample()
-        else: # new to RPO
+        else:  # new to RPO
             # sample again to add stochasticity to the policy
             z = torch.FloatTensor(action_mean.shape).uniform_(-self.rpo_alpha, self.rpo_alpha)
             action_mean = action_mean + z
             probs = Normal(action_mean, action_std)
-        
+
         return action, probs.log_prob(action).sum(1), probs.entropy().sum(1), self.critic(x)
 
 

diff --git a/tests/test_mujoco.py b/tests/test_mujoco.py
@@ -15,3 +15,13 @@ def test_mujoco():
         shell=True,
         check=True,
     )
+    subprocess.run(
+        "python cleanrl/rpo_continuous_action.py --env-id Hopper-v4 --num-envs 1 --num-steps 64 --total-timesteps 128",
+        shell=True,
+        check=True,
+    )
+    subprocess.run(
+        "python cleanrl/rpo_continuous_action.py --env-id dm_control/cartpole-balance-v0 --num-envs 1 --num-steps 64 --total-timesteps 128",
+        shell=True,
+        check=True,
+    )