feat(training): add opt-in teamplay evaluation

training/evaluate.py

@@ -33,6 +33,7 @@ def run_half(
     seed: int,
     grounded_a: bool = False,
     grounded_b: bool = False,
+    team_size: int = 1,
 ) -> dict:
     cmd = [
         godot_bin,
@@ -47,6 +48,10 @@ def run_half(
         f"--speedup={speedup}",
         f"--env_seed={seed}",
     ]
+    if team_size not in (1, 2):
+        raise ValueError("team_size must be 1 or 2")
+    if team_size == 2:
+        cmd.append("--eval_team_size=2")
     # Must match how each model was actually trained (see AIShipController's
     # allow_vertical/allow_pitch_roll) — a grounded pre-generation-4 model
     # never got a reward gradient on these axes, so leaving them unmasked here
@@ -72,6 +77,7 @@ def evaluate_pair(
     seed: int,
     grounded_a: bool = False,
     grounded_b: bool = False,
+    team_size: int = 1,
 ) -> dict:
     """Replay one seeded state sequence with the models on opposite sides."""
     if episodes < 2 or episodes % 2 != 0:
@@ -81,10 +87,12 @@ def evaluate_pair(
     first = run_half(
         godot_bin, model_a, model_b, episodes_per_side, speedup, seed,
         grounded_a=grounded_a, grounded_b=grounded_b,
+        team_size=team_size,
     )
     second = run_half(
         godot_bin, model_b, model_a, episodes_per_side, speedup, seed,
         grounded_a=grounded_b, grounded_b=grounded_a,
+        team_size=team_size,
     )
 
     a_team_0 = {
@@ -133,6 +141,7 @@ def main():
         help="Path to the Godot binary (or set GODOT_BIN)",
     )
     parser.add_argument("--speedup", type=int, default=16)
+    parser.add_argument("--team-size", type=int, choices=(1, 2), default=1)
     parser.add_argument("--seed", type=int, default=1, help="Seed for the paired starting-state sequence")
     parser.add_argument("--history", default=str(TRAINING_DIR / "eval_history.json"))
     parser.add_argument(
@@ -149,6 +158,7 @@ def main():
         record = evaluate_pair(
             args.godot_bin, model_a, model_b, args.episodes, args.speedup, args.seed,
             grounded_a=args.grounded_a, grounded_b=args.grounded_b,
+            team_size=args.team_size,
         )
     except ValueError as error:
         parser.error(str(error))