Update app.py

app.py

@@ -1,4 +1,4 @@
-# app_ucs.py — ST_GeoMech_UCS (Unified workflow like Tc)
+# app_ucs.py — ST_GeoMech_UCS (Unified workflow like Tc) — MAPE instead of MAE
 
 import io, json, os, base64, math
 from pathlib import Path
@@ -15,7 +15,7 @@ import matplotlib.pyplot as plt
 from matplotlib.ticker import FuncFormatter
 
 import plotly.graph_objects as go
-from sklearn.metrics import mean_squared_error, mean_absolute_error
+from sklearn.metrics import mean_squared_error  # MAE removed
 
 # =========================
 # Constants / Defaults
@@ -138,6 +138,18 @@ def pearson_r(y_true, y_pred) -> float:
     if np.all(a == a[0]) or np.all(p == p[0]): return float("nan")
     return float(np.corrcoef(a, p)[0, 1])
 
+def mape(y_true, y_pred, eps: float = 1e-8) -> float:
+    """
+    Mean Absolute Percentage Error in PERCENT.
+    Ignores rows where |y_true| < eps to avoid division blowups.
+    """
+    a = np.asarray(y_true, dtype=float)
+    p = np.asarray(y_pred, dtype=float)
+    denom = np.where(np.abs(a) < eps, np.nan, np.abs(a))
+    pct = np.abs(a - p) / denom * 100.0
+    val = np.nanmean(pct)
+    return float(val) if np.isfinite(val) else float("nan")
+
 @st.cache_resource(show_spinner=False)
 def load_model(model_path: str):
     return joblib.load(model_path)
@@ -321,7 +333,7 @@ def build_export_workbook(selected: list[str] | None = None) -> tuple[bytes|None
             sheets["Validation_Summary"] = _round_numeric(pd.DataFrame([sv])); order.append("Validation_Summary")
         oor_tbl = res.get("oor_tbl")
         if oor_tbl is not None and isinstance(oor_tbl, pd.DataFrame) and not oor_tbl.empty:
-            sheets["Validation_ORE"] = _round_numeric(oor_tbl.reset_index(drop=True)); order.append("Validation_ORE")
+            sheets["Validation_OOR"] = _round_numeric(oor_tbl.reset_index(drop=True)); order.append("Validation_OOR")
 
     # Prediction
     if ("Prediction" in (selected or _available_sections())) and "PredictOnly" in res:
@@ -406,7 +418,6 @@ def cross_plot_static(actual, pred):
     a = pd.Series(actual, dtype=float)
     p = pd.Series(pred, dtype=float)
 
-    # Use data-driven limits with a small pad (or keep your fixed 6000–10000 if you prefer)
     lo = float(min(a.min(), p.min()))
     hi = float(max(a.max(), p.max()))
     pad = 0.03 * (hi - lo if hi > lo else 1.0)
@@ -707,14 +718,14 @@ if st.session_state.app_step == "dev":
 
         st.session_state.results["Train"]=tr; st.session_state.results["Test"]=te
         st.session_state.results["m_train"]={
-            "R":   pearson_r(tr[TARGET], tr[PRED_COL]),
+            "R":    pearson_r(tr[TARGET], tr[PRED_COL]),
             "RMSE": rmse(tr[TARGET], tr[PRED_COL]),
-            "MAE":  mean_absolute_error(tr[TARGET], tr[PRED_COL])
+            "MAPE": mape(tr[TARGET], tr[PRED_COL])
         }
         st.session_state.results["m_test"]={
-            "R":   pearson_r(te[TARGET], te[PRED_COL]),
+            "R":    pearson_r(te[TARGET], te[PRED_COL]),
             "RMSE": rmse(te[TARGET], te[PRED_COL]),
-            "MAE":  mean_absolute_error(te[TARGET], te[PRED_COL])
+            "MAPE": mape(te[TARGET], te[PRED_COL])
         }
 
         tr_min = tr[FEATURES].min().to_dict(); tr_max = tr[FEATURES].max().to_dict()
@@ -723,12 +734,14 @@ if st.session_state.app_step == "dev":
 
     def _dev_block(df, m):
         c1,c2,c3 = st.columns(3)
-        c1.metric("R", f"{m['R']:.3f}"); c2.metric("RMSE", f"{m['RMSE']:.2f}"); c3.metric("MAE", f"{m['MAE']:.2f}")
+        c1.metric("R", f"{m['R']:.3f}")
+        c2.metric("RMSE", f"{m['RMSE']:.2f}")
+        c3.metric("MAPE (%)", f"{m['MAPE']:.2f}%")
         st.markdown("""
             <div style='text-align: left; font-size: 0.8em; color: #6b7280; margin-top: -16px; margin-bottom: 8px;'>
                 <strong>R:</strong> Pearson Correlation Coefficient<br>
                 <strong>RMSE:</strong> Root Mean Square Error<br>
-                <strong>MAE:</strong> Mean Absolute Error
+                <strong>MAPE:</strong> Mean Absolute Percentage Error (percent of actual; rows with near-zero actuals are ignored).
             </div>
         """, unsafe_allow_html=True)
         col_track, col_cross = st.columns([2, 3], gap="large")
@@ -785,9 +798,9 @@ if st.session_state.app_step == "validate":
                     lambda r:", ".join([c for c,v in r.items() if v]), axis=1
                 )
         st.session_state.results["m_val"]={
-            "R": pearson_r(df[TARGET], df[PRED_COL]),
+            "R":    pearson_r(df[TARGET], df[PRED_COL]),
             "RMSE": rmse(df[TARGET], df[PRED_COL]),
-            "MAE": mean_absolute_error(df[TARGET], df[PRED_COL])
+            "MAPE": mape(df[TARGET], df[PRED_COL])
         }
         st.session_state.results["sv_val"]={"n":len(df), "pred_min":float(df[PRED_COL].min()), "pred_max":float(df[PRED_COL].max()), "oor":oor_pct}
         st.session_state.results["oor_tbl"]=tbl
@@ -795,12 +808,14 @@ if st.session_state.app_step == "validate":
     if "Validate" in st.session_state.results:
         m = st.session_state.results["m_val"]
         c1,c2,c3 = st.columns(3)
-        c1.metric("R", f"{m['R']:.3f}"); c2.metric("RMSE", f"{m['RMSE']:.2f}"); c3.metric("MAE", f"{m['MAE']:.2f}")
+        c1.metric("R", f"{m['R']:.3f}")
+        c2.metric("RMSE", f"{m['RMSE']:.2f}")
+        c3.metric("MAPE (%)", f"{m['MAPE']:.2f}%")
         st.markdown("""
             <div style='text-align: left; font-size: 0.8em; color: #6b7280; margin-top: -16px; margin-bottom: 8px;'>
                 <strong>R:</strong> Pearson Correlation Coefficient<br>
                 <strong>RMSE:</strong> Root Mean Square Error<br>
-                <strong>MAE:</strong> Mean Absolute Error
+                <strong>MAPE:</strong> Mean Absolute Percentage Error (percent of actual; rows with near-zero actuals are ignored).
             </div>
         """, unsafe_allow_html=True)