Update app.py

app.py

@@ -46,36 +46,34 @@ PREDICTOR_DIR = _prepare_predictor_dir()
 PREDICTOR = autogluon.tabular.TabularPredictor.load(PREDICTOR_DIR, require_py_version_match=False)
 
 def do_predict(flower_diameter_cm, petal_length_cm, petal_width_cm, petal_count, stem_height_cm, top_k):
-    row = {
-        FEATURE_COLS[0]: float(flower_diameter_cm),
-        FEATURE_COLS[1]: float(petal_length_cm),
-        FEATURE_COLS[2]: float(petal_width_cm),
-        FEATURE_COLS[3]: int(petal_count),
-        FEATURE_COLS[4]: float(stem_height_cm),
-    }
-    X = pandas.DataFrame([row], columns=FEATURE_COLS)
-
-    # Predicted label (string)
-    pred_series = PREDICTOR.predict(X)
-    pred_label = str(pred_series.iloc[0])
-
-    # Probabilities: dict[class] -> float
-    proba_dict = None
     try:
-        proba = PREDICTOR.predict_proba(X)
-        # AutoGluon can return Series for binary; normalize to DataFrame row
-        if isinstance(proba, pandas.Series):
-            proba = proba.to_frame().T
-        row0 = proba.iloc[0].sort_values(ascending=False)
-        if isinstance(top_k, (int, float)) and top_k > 0:
-            row0 = row0.head(int(top_k))
-        proba_dict = {str(cls): float(val) for cls, val in row0.items()}
-    except Exception:
-        # If proba not available, just put 1.0 on the predicted class
-        proba_dict = {pred_label: 1.0}
-
-    # Return TWO outputs: (predicted color string, dict of numeric probs)
-    return proba_dict
+        row = {
+            FEATURE_COLS[0]: float(flower_diameter_cm),
+            FEATURE_COLS[1]: float(petal_length_cm),
+            FEATURE_COLS[2]: float(petal_width_cm),
+            FEATURE_COLS[3]: int(petal_count),
+            FEATURE_COLS[4]: float(stem_height_cm),
+        }
+        X = pandas.DataFrame([row], columns=FEATURE_COLS)
+
+        pred_series = PREDICTOR.predict(X)
+        pred_label = str(pred_series.iloc[0])
+
+        try:
+            proba = PREDICTOR.predict_proba(X)
+            if isinstance(proba, pandas.Series):
+                proba = proba.to_frame().T
+            row0 = proba.iloc[0].sort_values(ascending=False)
+            if isinstance(top_k, (int, float)) and top_k > 0:
+                row0 = row0.head(int(top_k))
+            proba_dict = {str(cls): float(val) for cls, val in row0.items()}
+        except Exception:
+            proba_dict = {pred_label: 1.0}
+
+        return proba_dict, ""  # second output is debug text
+    except Exception as e:
+        import traceback
+        return {}, f"{e}\n\n{traceback.format_exc()}"
 
 # ----------------
 # Example records
@@ -107,12 +105,13 @@ with gradio.Blocks() as demo:
 
     # Separate outputs: Textbox for label, Label for probs (dict must be numeric)
     proba_pretty = gradio.Label(num_top_classes=10, label="Class probabilities")
-
+    debug_box    = gradio.Textbox(label="debug", visible=True)
+    
     inputs = [flower_diameter_cm, petal_length_cm, petal_width_cm, petal_count, stem_height_cm, top_k]
 
     # Trigger on any change
     for comp in inputs:
-        comp.change(fn=do_predict, inputs=inputs, outputs=[proba_pretty])
+        comp.change(fn=do_predict, inputs=inputs, outputs=[proba_pretty, debug_box])
 
     # Examples: only pass the first 5 inputs (excluding top_k) to match example rows
     gradio.Examples(