#42 run baselines in ci

torch_training/bla.py

@@ -162,12 +162,13 @@ def main(argv):
             # print(all_rewards,action)
             obs_original = next_obs.copy()
             for a in range(env.get_num_agents()):
-                data, distance, agent_data = split_tree(tree=np.array(next_obs[a]),
-                                                        current_depth=0)
-                data = norm_obs_clip(data)
-                distance = norm_obs_clip(distance)
-                agent_data = np.clip(agent_data, -1, 1)
-                next_obs[a] = np.concatenate((np.concatenate((data, distance)), agent_data))
+                a = 5
+                # data, distance, agent_data = split_tree(tree=np.array(next_obs[a]),
+                #                                         current_depth=0)
+                # data = norm_obs_clip(data)
+                # distance = norm_obs_clip(distance)
+                # agent_data = np.clip(agent_data, -1, 1)
+                # next_obs[a] = np.concatenate((np.concatenate((data, distance)), agent_data))
             time_obs.append(next_obs)
         #
         #     # Update replay buffer and train agent