🚀 Upload Flask + Docker backend files

SuperKart_Sales_Predictor.py

@@ -24,7 +24,7 @@ def predict_sales():
     """
     data = request.get_json()
 
-    # You must ensure these keys match your model’s training features
+    # Extract relevant features from the JSON data
     sample = {
         'Product_Weight': data['Product_Weight'],
         'Product_Allocated_Area': data['Product_Allocated_Area'],
@@ -39,11 +39,12 @@ def predict_sales():
 
     input_df = pd.DataFrame([sample])
 
-    # Ensure encoding matches model training (e.g., one-hot or label encoding)
-    # This assumes you used label encoding in preprocessing before model training
-
     # Predict
     predicted_sales = model.predict(input_df)[0]
+
+    # Calculate actual price
+    predicted_sales = np.exp(predicted_sales)
+
     predicted_sales = round(float(predicted_sales), 2)
 
     print("🚀 Flask app is starting...")