diff --git a/demos/cartpole_single_agent.py b/demos/cartpole_single_agent.py
index 141d5a5..5c57229 100644
--- a/demos/cartpole_single_agent.py
+++ b/demos/cartpole_single_agent.py
@@ -21,7 +21,7 @@
                                       action_size = num_actions, 
                                       action_space = action_space, 
                                       policy_type = 'fcn',
-                                      layers = [[], [64, 32, 14]])
+                                      layers = [64, 32, 14])
 
 reinforce = REINFORCE(env, policy_network, artificial_truncation=512)
 
diff --git a/demos/dummy_single_agent_continuous.py b/demos/dummy_single_agent_continuous.py
index 1c80be7..f935bdd 100644
--- a/demos/dummy_single_agent_continuous.py
+++ b/demos/dummy_single_agent_continuous.py
@@ -1,5 +1,5 @@
 from spurl.algorithms.reinforce.continuous import REINFORCE
-from spurl.utils import * 
+from spurl.utils import save_model, load_model, save_environment_render, build_policy_network
 from spurl.core import train, test
 
 import tensorflow as tf
diff --git a/demos/dummy_single_agent_discrete.py b/demos/dummy_single_agent_discrete.py
index f5a0480..299f987 100644
--- a/demos/dummy_single_agent_discrete.py
+++ b/demos/dummy_single_agent_discrete.py
@@ -1,7 +1,7 @@
 from spurl.algorithms.reinforce.discrete import REINFORCE
 
 from spurl.core import train, test
-from spurl.utils import * 
+from spurl.utils import save_model, load_model, save_environment_render, build_policy_network 
 
 import tensorflow as tf
 import gymnasium as gym
diff --git a/demos/pendulum_single_agent.py b/demos/pendulum_single_agent.py
index 7c51fde..d0576b3 100644
--- a/demos/pendulum_single_agent.py
+++ b/demos/pendulum_single_agent.py
@@ -20,7 +20,7 @@
                                       action_size = action_size, 
                                       action_space = action_space,
                                       policy_type = 'fcn',
-                                      layers = [[], [128, 64, 32]],
+                                      layers = [128, 64, 32],
                                       activation_fn = 'tanh')
 
 # for linearly annealing scale