Merge pull request #1045 from Unity-Technologies/develop-unityagents-nan-reward

Nan Rewards converted to 0 and throwing a warning

python/unityagents/environment.py

 import glob
 import io
 import logging
+import math
 import numpy as np
 import os
 import subprocess
             else:
                 [x.memories.extend([0] * (memory_size - len(x.memories))) for x in agent_info_list]
                 memory = np.array([x.memories for x in agent_info_list])
+            if any([math.isnan(x.reward) for x in agent_info_list]):
+                logger.warning("An agent had a NaN reward for brain "+b)
-                reward=[x.reward for x in agent_info_list],
+                reward=[x.reward if not math.isnan(x.reward) else 0 for x in agent_info_list],
                 agents=[x.id for x in agent_info_list],
                 local_done=[x.done for x in agent_info_list],
                 vector_action=np.array([x.stored_vector_actions for x in agent_info_list]),