cloud run

config/trainer_config.yaml

     time_horizon: 1000
     num_layers: 3
     self_play:
-        window: 30
-        play_against_current_self_ratio: 0.5
-        save_steps: 50000
-        swap_steps: 50000
+        window: 100
+        play_against_current_self_ratio: 0.2
+        save_steps: 100000
+        swap_steps: 100000
 
 CrawlerStatic:
     normalize: true