Add pympler

ml-agents/mlagents/trainers/trainer_controller.py

 from mlagents.trainers.agent_processor import AgentManager
 from mlagents.tf_utils.globals import get_rank
 from mlagents import torch_utils
+from pympler import muppy, summary
 
 
 class TrainerController:
                 n_steps = self.advance(env_manager)
                 for _ in range(n_steps):
                     self.reset_env_if_ready(env_manager)
+            all_objects = muppy.get_objects()
+            sum1 = summary.summarize(all_objects)
+            # Prints out a summary of the large objects
+            summary.print_(sum1)
             # Stop advancing trainers
             self.join_threads()
         except (

ml-agents/setup.py

         # We don't actually need six, but tensorflow does, and pip seems
         # to get confused and install the wrong version.
         "six>=1.12.0",
+        "pympler>=0.8",
     ],
     python_requires=">=3.6.1",
     entry_points={