Update evaluate for new policy

NeuralMMO · Sep 12, 2023 · 1222980 · 1222980
1 parent 4888c58
commit 1222980
Showing 1 changed file with 1 addition and 1 deletion.
diff --git a/evaluate.py b/evaluate.py
@@ -62,7 +62,7 @@ def save_replays(policy_store_dir, save_dir):
     from reinforcement_learning import policy  # import your policy
     def make_policy(envs):
         learner_policy = policy.Baseline(
-            envs,
+            envs._driver_env,
             input_size=args.input_size,
             hidden_size=args.hidden_size,
             task_size=args.task_size