Update app.py

app.py

@@ -1469,34 +1469,53 @@ def get_llm_client():
         _HF_CLIENT = None
         return None
 
-def build_context(summary_jenis: pd.DataFrame, agg_total: pd.DataFrame, wilayah: str, kew: str) -> str:
+def build_context(summary_jenis: pd.DataFrame, agg_total: pd.DataFrame, verif_total: pd.DataFrame, wilayah: str, kew: str) -> str:
     lines = []
     lines.append(f"Wilayah filter: {wilayah}")
     lines.append(f"Kewenangan: {kew}")
-    lines.append("Metode: Indeks dasar dihitung per entitas (Yeo-Johnson + MinMax nasional), lalu agregat per wilayah×jenis.")
-    lines.append("Penyesuaian berbasis kecukupan sampel (target 68%) dihitung PER JENIS (sekolah/umum/khusus).")
-    lines.append("Keseluruhan wilayah (FIX): rata-rata 3 jenis (sekolah+umum+khusus) ÷ 3 (missing=0, tetap ÷3).")
+    lines.append("Metode: Indeks dasar dihitung per entitas (Yeo-Johnson + MinMax nasional per indikator), lalu diagregasi per wilayah×jenis.")
+    lines.append("Penyesuaian berbasis kecukupan sampel minimum 68% memakai faktor = min(terkumpul/target68, 1.0).")
+    lines.append("Nilai keseluruhan (FIX): rata-rata 3 jenis (sekolah+umum+khusus) ÷ 3 (missing=0, tetap ÷3).")
 
     if summary_jenis is not None and not summary_jenis.empty:
         lines.append("\nRingkasan (jenis + keseluruhan):")
-        for _, r in summary_jenis.iterrows():
+        tmp = summary_jenis.copy()
+        for _, r in tmp.iterrows():
+            jenis = r.get("Jenis", "")
+            jw = int(pd.to_numeric(r.get("Jumlah_Wilayah", 0), errors="coerce") or 0)
+            tp = int(pd.to_numeric(r.get("Total_Perpus", 0), errors="coerce") or 0)
+            pop = int(pd.to_numeric(r.get("Pop_Total_Jenis", 0), errors="coerce") or 0)
+            tgt = int(pd.to_numeric(r.get("Target68_Total_Jenis", 0), errors="coerce") or 0)
+            terk = int(pd.to_numeric(r.get("Terkumpul_Jenis", 0), errors="coerce") or 0)
+            cov = float(pd.to_numeric(r.get("Coverage_Target68_Jenis_%", 0), errors="coerce") or 0.0)
+
+            dasar = float(pd.to_numeric(r.get("Indeks_Dasar_0_100", 0), errors="coerce") or 0.0)
+            final = float(pd.to_numeric(r.get("Indeks_Final_Disesuaikan_0_100", 0), errors="coerce") or 0.0)
+            adj = float(pd.to_numeric(r.get("Penyesuaian_Poin", 0), errors="coerce") or 0.0)
+
             lines.append(
-                f"- {r['Jenis']}: wilayah={int(r['Jumlah_Wilayah'])}, total_perpus={int(r['Total_Perpus'])}, "
-                f"pop={int(r['Pop_Total_Jenis'])}, target68={int(r['Target68_Total_Jenis'])}, terkumpul={int(r['Terkumpul_Jenis'])}, "
-                f"coverage={float(r['Coverage_Target68_Jenis_%']):.2f}%, "
-                f"dasar={float(r['Indeks_Dasar_0_100']):.2f}, final={float(r['Indeks_Final_Disesuaikan_0_100']):.2f}"
+                f"- {jenis}: wilayah={jw}, perpus={tp}, pop={pop}, target68={tgt}, terkumpul={terk}, coverage={cov:.2f}%, "
+                f"dasar={dasar:.2f}, final={final:.2f}, penyesuaian={adj:.2f}"
             )
 
     if agg_total is not None and not agg_total.empty:
         label_col = "Kab/Kota" if "Kab/Kota" in agg_total.columns else ("Provinsi" if "Provinsi" in agg_total.columns else None)
-        lines.append("\nTop 5 wilayah (Final tertinggi):")
-        top = agg_total.sort_values("Indeks_Final_Wilayah_0_100", ascending=False).head(5)
-        for _, r in top.iterrows():
-            wl = r.get(label_col, "(wilayah)") if label_col else "(wilayah)"
-            lines.append(f"- {wl}: Final={float(r['Indeks_Final_Wilayah_0_100']):.2f} | Dasar={float(r['Indeks_Dasar_Agregat_0_100']):.2f} | n_total={int(r.get('n_total', 0))}")
+        if label_col and "Indeks_Final_Wilayah_0_100" in agg_total.columns:
+            lines.append("\nTop 5 wilayah (Final disesuaikan tertinggi):")
+            top = agg_total.sort_values("Indeks_Final_Wilayah_0_100", ascending=False).head(5)
+            for _, r in top.iterrows():
+                wl = r.get(label_col, "(wilayah)")
+                fin = float(pd.to_numeric(r.get("Indeks_Final_Wilayah_0_100", 0), errors="coerce") or 0.0)
+                ntt = int(pd.to_numeric(r.get("n_total", 0), errors="coerce") or 0)
+                lines.append(f"- {wl}: Final={fin:.2f} | n_total={ntt}")
+
+    # verif_total optional
+    if verif_total is not None and not verif_total.empty:
+        pass
 
     return "\n".join(lines)
 
+
 def generate_llm_analysis(summary_jenis, agg_total, verif_total, wilayah, kew):
     ctx = build_context(summary_jenis, agg_total, verif_total, wilayah, kew)
     client = get_llm_client()