Update app.py

app.py

@@ -8,27 +8,27 @@ from datetime import datetime
 
 from utils import generate_features, pick_top15
 
-# In-memory history of predictions (max 12 rows for the month)
+# In-memory history (up to 12 predictions)
 prediction_history = []
 
 
 def train_and_predict(file_obj):
-    # --- Load dataset correctly ---
+    # --- Load dataset ---
     df = pd.read_csv(file_obj.name, header=0)
-    df = df.iloc[:, :8]  # Keep first 8 cols: draw_no, draw_date, n1..n6
+    df = df.iloc[:, :8]  # draw_no, draw_date, n1..n6
     df.columns = ["draw_no", "draw_date", "n1", "n2", "n3", "n4", "n5", "n6"]
 
-    # Convert numbers to integers
+    # Convert numbers to int
     for col in ["n1", "n2", "n3", "n4", "n5", "n6"]:
         df[col] = pd.to_numeric(df[col], errors="coerce")
 
     debug_log = []
-    debug_log.append(f"✅ Loaded dataset with {len(df)} draws")
+    debug_log.append(f"✅ Loaded {len(df)} draws")
     debug_log.append(f"First draw: {df['draw_date'].iloc[0]}, Last draw: {df['draw_date'].iloc[-1]}")
 
     # --- Generate features ---
     features, labels = generate_features(df)
-    debug_log.append(f"Generated {len(features)} rows, Label distribution: {np.bincount(labels)}")
+    debug_log.append(f"Features: {len(features)} rows | Label distribution: {np.bincount(labels)}")
 
     top15 = None
     auc = None
@@ -44,6 +44,7 @@ def train_and_predict(file_obj):
             model = GradientBoostingClassifier(n_estimators=200, max_depth=3, random_state=42)
             model.fit(X_train, y_train)
 
+            # AUC check
             if len(np.unique(y_test)) > 1:
                 auc = roc_auc_score(y_test, model.predict_proba(X_test)[:, 1])
             else:
@@ -52,10 +53,14 @@ def train_and_predict(file_obj):
             # Score all numbers 1–50
             all_numbers = pd.DataFrame({"number": range(1, 51)})
             all_features, _ = generate_features(df, candidate_numbers=all_numbers["number"].tolist())
-            scores = model.predict_proba(all_features)[:, 1]
-            all_numbers["score"] = scores
+            probs = model.predict_proba(all_features)[:, 1]
 
-            top15 = pick_top15(all_numbers)
+            # Aggregate mean probability per number
+            all_features["prob"] = probs
+            scored = all_features.groupby("number")["prob"].mean().reset_index()
+            scored.columns = ["number", "score"]
+
+            top15 = pick_top15(scored)
             debug_log.append(f"🎯 ML Top 15: {top15}")
             debug_log.append(f"Model AUC: {auc:.3f}")
 
@@ -63,10 +68,10 @@ def train_and_predict(file_obj):
             debug_log.append(f"⚠️ ML failed: {str(e)}")
             used_fallback = True
     else:
-        debug_log.append("⚠️ Only one class found — using fallback")
+        debug_log.append("⚠️ Only one class found — fallback mode")
         used_fallback = True
 
-    # --- Fallback: frequency-based ---
+    # --- Fallback (frequency-based) ---
     if used_fallback or top15 is None:
         nums = df[["n1", "n2", "n3", "n4", "n5", "n6"]].values.flatten()
         freq = pd.Series(nums).value_counts().reset_index()
@@ -74,28 +79,28 @@ def train_and_predict(file_obj):
         top15 = sorted(freq.head(15)["number"].tolist())
         debug_log.append(f"👉 Fallback Top 15: {top15}")
 
-    # --- Record in prediction history ---
-    today = datetime.now().strftime("%Y-%m-%d %H:%M")
+    # --- Record prediction history ---
+    today = datetime.now().strftime("%Y-%m-%d")
     prediction_history.append({"date": today, "numbers": top15})
-    # Keep last 12 only
     if len(prediction_history) > 12:
         prediction_history.pop(0)
 
-    # Build table view
-    history_df = pd.DataFrame(prediction_history)
-    history_df["numbers"] = history_df["numbers"].apply(lambda x: " ".join(str(n).zfill(2) for n in x))
+    # Format history log
+    history_lines = []
+    for row in prediction_history:
+        history_lines.append(f"{row['date']} Top 15: {row['numbers']}")
 
-    table_view = history_df.to_string(index=False)
+    log_output = "\n".join(debug_log) + "\n\n📊 Prediction History (last 12 runs):\n" + "\n".join(history_lines)
 
-    return "\n".join(debug_log) + "\n\n📊 Prediction History (last 12):\n" + table_view
+    return log_output
 
 
 demo = gr.Interface(
     fn=train_and_predict,
     inputs=gr.File(file_types=[".txt", ".csv"], label="Upload Toto650.txt"),
-    outputs=gr.Textbox(label="Training, Prediction, and History Log", lines=25),
+    outputs=gr.Textbox(label="Training, Prediction & History Log", lines=25),
     title="Sure Win Club - Star Toto 6/50 Predictor",
-    description="Upload Toto650.txt after each draw. System trains fresh and shows Top 15 hot numbers + running history (up to 12 rows)."
+    description="Upload Toto650.txt after each draw. System trains fresh and shows Top 15 hot numbers + running history (up to 12 runs)."
 )
 
 if __name__ == "__main__":
@@ -106,3 +111,4 @@ if __name__ == "__main__":
 
 
 
+