Update app.py

app.py

@@ -360,24 +360,17 @@ def suggest_one_per_band(synth: pd.DataFrame, sigma_mkt: float, universe_user: L
     for band in ["Low", "Medium", "High"]:
         lo, hi = _band_bounds(sigma_mkt, band)
         pool = synth[(synth["sigma_hist"] >= lo) & (synth["sigma_hist"] <= hi)].copy()
-        note = ""
         if pool.empty:
-            # Closest-available logic to keep band meaning intuitive
+            # choose reasonable substitutes quietly (no extra text)
             if band.lower() == "low":
                 pool = synth.nsmallest(50, "sigma_hist").copy()
-                note = " (closest available: min σ)"
             elif band.lower() == "high":
                 pool = synth.nlargest(50, "sigma_hist").copy()
-                note = " (closest available: max σ)"
             else:  # medium
                 tmp = synth.copy()
                 tmp["dist_med"] = (tmp["sigma_hist"] - sigma_mkt).abs()
                 pool = tmp.nsmallest(100, "dist_med").drop(columns=["dist_med"])
-                note = " (closest to market σ)"
         chosen = rerank_and_pick_one(pool, universe_user, band)
-        if not chosen.empty:
-            chosen = chosen.copy()
-            chosen["__note__"] = note
         out[band.lower()] = chosen
     return out
 
@@ -517,8 +510,7 @@ def compute(
     def _fmt(row: pd.Series) -> str:
         if row is None or row.empty:
             return "No pick available."
-        note = str(row.get("__note__", "") or "")
-        return f"CAPM E[r] {row['mu_capm']*100:.2f}%, σ(h) {row['sigma_hist']*100:.2f}%{note}"
+        return f"CAPM E[r] {row['mu_capm']*100:.2f}%, σ(h) {row['sigma_hist']*100:.2f}%"
 
     txt_low   = _fmt(picks.get("low", pd.Series(dtype=object)))
     txt_med   = _fmt(picks.get("medium", pd.Series(dtype=object)))
@@ -553,7 +545,7 @@ def compute(
         sugg_sigma_hist=chosen_sigma, sugg_mu_capm=chosen_mu
     )
 
-    # ---------- ONLY EXPLANATION ADDED BELOW (no calc changes) ----------
+    # ---------- summary text (cosmetic-only changes made) ----------
     info = "\n".join([
         "### Inputs",
         f"- Lookback years {years_lookback}",
@@ -562,7 +554,7 @@ def compute(
         f"- Market ERP {erp_ann:.2%}",
         f"- Market σ (hist) {sigma_mkt:.2%}",
         "",
-        "### Your portfolio (CAPM on CML; x=σ_hist, y=CAPM E[r])",
+        "### Your portfolio",
         f"- CAPM E[r] {mu_capm:.2%}",
         f"- σ (historical) {sigma_hist:.2%}",
         "",
@@ -603,7 +595,7 @@ with gr.Blocks(title="Efficient Portfolio Advisor") as demo:
                 search_btn = gr.Button("Search")
                 add_btn = gr.Button("Add selected to portfolio")
 
-            gr.Markdown("### Portfolio positions (enter $ amounts; negatives allowed)")
+            gr.Markdown("### Portfolio positions")
             table = gr.Dataframe(
                 headers=["ticker", "amount_usd"],
                 datatype=["str", "number"],
@@ -614,7 +606,7 @@ with gr.Blocks(title="Efficient Portfolio Advisor") as demo:
             horizon = gr.Number(label="Horizon in years (1–100)", value=HORIZON_YEARS, precision=0)
             lookback = gr.Slider(1, 15, value=DEFAULT_LOOKBACK_YEARS, step=1, label="Lookback years for betas & covariances")
 
-            gr.Markdown("### Suggestions (one per band)")
+            gr.Markdown("### Suggestions")
             with gr.Row():
                 btn_low = gr.Button("Show Low")
                 btn_med = gr.Button("Show Medium")