made a bit of progress

test.py

@@ -58,7 +58,7 @@ env_config = {
     
     # ===== STANDARD ARGUMENTS ======
     # kwargs that are used by every Scenario class (i.e. defined in BaseEnvironment)
-    'agent_composition': {"BasicMobileAgent": 20,"TradingAgent":5},        # Number of non-planner agents (must be > 1)
+    'agent_composition': {"BasicMobileAgent": 5,"TradingAgent":10},        # Number of non-planner agents (must be > 1)
     'world_size': [5, 5], # [Height, Width] of the env world
     'episode_length': 256, # Number of timesteps per episode
     'isoelastic_eta':0.001,
@@ -186,7 +186,7 @@ def printReplay(econ,agentid):
 econ=foundation.make_env_instance(**env_config)
 market=econ.get_component("ContinuousDoubleAuction")
 action=market.get_n_actions("TradingAgent")
-baseEconWrapper=BaseEconWrapper(econ)
+baseEconWrapper=BaseEconWrapper(econ,False)
 baseEconWrapper.run()
 time.sleep(0.5)
 mobileRecieverEconWrapper=RecieverEconWrapper(base_econ=baseEconWrapper,agent_classname="BasicMobileAgent")
@@ -229,6 +229,7 @@ exporter=MongoExporter(CONNECTION_STRING,"econ",econ)
 
 while True:
     print("### EVAL ###")
+    baseEconWrapper.force_reset()
     obs_basic=stackenv_basic.reset()
     obs_trade=stackenv_traid.reset()
     done=False
@@ -254,7 +255,7 @@ while True:
    # trades=market.get_dense_log()
     build=craft.get_dense_log()
     met=econ.previous_episode_metrics
-    printReplay(econ,21)
+    printReplay(econ,5)
    #printMarket(trades)
   #printBuilds(builds=build)
     print("social/productivity: {}".format(met["social/productivity"]))