DeadLockAvoidanceAgent -> observation

reinforcement_learning/multi_agent_training.py

@@ -475,7 +475,7 @@ def eval_policy(env, tree_observation, policy, train_params, obs_params):
 if __name__ == "__main__":
     parser = ArgumentParser()
     parser.add_argument("-n", "--n_episodes", help="number of episodes to run", default=2500, type=int)
-    parser.add_argument("-t", "--training_env_config", help="training config id (eg 0 for Test_0)", default=0, type=int)
+    parser.add_argument("-t", "--training_env_config", help="training config id (eg 0 for Test_0)", default=2, type=int)
     parser.add_argument("-e", "--evaluation_env_config", help="evaluation config id (eg 0 for Test_0)", default=0,
                         type=int)
     parser.add_argument("--n_evaluation_episodes", help="number of evaluation episodes", default=5, type=int)

run.py

@@ -25,7 +25,7 @@ from reinforcement_learning.dddqn_policy import DDDQNPolicy
 VERBOSE = True
 
 # Checkpoint to use (remember to push it!)
-checkpoint = "./checkpoints/201103172118-0.pth"
+checkpoint = "./checkpoints/201103180606-1400.pth"
 
 # Use last action cache
 USE_ACTION_CACHE = True