diff --git a/reinforcement_learning/multi_agent_training.py b/reinforcement_learning/multi_agent_training.py
index 6a34939604b1d8d85fcbbd7f1c3693d8ad3ae48d..872f69503774bdaf6694e1d4241be2d8296a41fd 100755
--- a/reinforcement_learning/multi_agent_training.py
+++ b/reinforcement_learning/multi_agent_training.py
@@ -19,9 +19,9 @@ from flatland.utils.rendertools import RenderTool
 from torch.utils.tensorboard import SummaryWriter
 
 from reinforcement_learning.dddqn_policy import DDDQNPolicy
-from reinforcement_learning.ppo_agent import PPOPolicy
 from reinforcement_learning.deadlockavoidance_with_decision_agent import DeadLockAvoidanceWithDecisionAgent
 from reinforcement_learning.multi_decision_agent import MultiDecisionAgent
+from reinforcement_learning.ppo_agent import PPOPolicy
 from utils.agent_action_config import get_flatland_full_action_size, get_action_size, map_actions, map_action
 from utils.dead_lock_avoidance_agent import DeadLockAvoidanceAgent
 
@@ -189,7 +189,7 @@ def train_agent(train_params, train_env_params, eval_env_params, obs_params):
         policy = DDDQNPolicy(state_size, get_action_size(), train_params)
 
     # Load existing policy
-    if train_params.load_policy is not "":
+    if train_params.load_policy is not '':
         policy.load(train_params.load_policy)
 
     # Loads existing replay buffer