Add run_model_family_balanced.py

scripts/run_model_family_balanced.py

@@ -0,0 +1,40 @@
+#!/usr/bin/env python3
+"""Run model-family ablation with balanced config selection."""
+import os, sys, yaml, json, time
+from datetime import datetime
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+from scripts.run_synthetic import build_model_family_grid, run_single_config
+from src.utils import save_jsonl, ensure_dir
+
+with open('config/synthetic_grid.yaml') as f:
+    cfg = yaml.safe_load(f)
+
+configs = build_model_family_grid(cfg)
+
+# Select balanced: 8 per family
+pg = [c for c in configs if c['model_family']=='poisson_gamma'][:8]
+gg = [c for c in configs if c['model_family']=='gaussian_gaussian'][:8]
+gm = [c for c in configs if c['model_family']=='gaussian_gamma_map'][:8]
+selected = pg + gg + gm
+
+print(f"Running {len(selected)} balanced configs: {len(pg)} PG, {len(gg)} GG, {len(gm)} GM")
+
+output_dir = ensure_dir('results/raw')
+timestamp = datetime.now().strftime("%Y%m%d_%H%M%S")
+output_file = os.path.join(output_dir, f'model_family_balanced_{timestamp}.jsonl')
+
+all_records = []
+for idx, config in enumerate(selected):
+    print(f"\n>>> Config {idx+1}/{len(selected)} ({config['model_family']})")
+    try:
+        records = run_single_config(config, mode='model_family')
+        all_records.extend(records)
+        save_jsonl(records, output_file)
+        print(f"  Saved {len(records)} records (total: {len(all_records)})")
+    except Exception as e:
+        print(f"  ERROR: {e}")
+        import traceback
+        traceback.print_exc()
+
+print(f"\nCompleted. Total: {len(all_records)} records in {output_file}")