Multiple objectives for one epoch

config.yaml

@@ -92,7 +92,7 @@ train:
   compile: True
   compile_mode: "reduce-overhead"
   float32_matmul_precision: "high"
-  total_timesteps: 1_000_000_000 # 100_000_000_000 for full games
+  total_timesteps: 250_000_000 # 1_000_000_000 # 100_000_000_000 for full games
   batch_size: 65536
   minibatch_size: 2048
   anneal_lr: False
@@ -140,7 +140,10 @@ train:
     EVENT_BEAT_MISTY: 90
     EVENT_GOT_HM01: 180
     EVENT_BEAT_ROUTE_9_TRAINER_0: 300
-  one_epoch: "EVENT_BEAT_CHAMPION_RIVAL"
+  one_epoch:
+    # - "EVENT_BEAT_CHAMPION_RIVAL"
+    - HM_03
+    - HM_04
 
 wrappers:
   empty:

pokemonred_puffer/cleanrl_puffer.py

@@ -691,7 +691,10 @@ def done_training(self):
             or (
                 self.config.one_epoch
                 and self.states
-                and any(self.config.one_epoch in key for key in self.states.keys())
+                and (
+                    len(set(self.config.one_epoch).intersection(self.states.keys()))
+                    == len(self.config.one_epoch)
+                )
             )
         )
 

sweep-config.yaml

@@ -5,6 +5,10 @@ env:
       max: 81920
       is_integer: True
       scale: 2500
+  map_id_scalefactor:
+    carbs.utils.LinearSpace:
+      min: 1.0
+      max: 10.0
 
 train:
   total_timesteps: