new goalie

config/trainer_config.yaml

     time_horizon: 1000
     num_layers: 2
     self_play:
-        window: 50
+        window: 10
         play_against_latest_model_ratio: 0.5
         save_steps: 50000
         swap_steps: 25000
     time_horizon: 1000
     num_layers: 2
     self_play:
-        window: 50
+        window: 10
         play_against_latest_model_ratio: 0.5
         save_steps: 50000
         swap_steps: 100000