Upload folder using huggingface_hub

app.py

@@ -45,12 +45,12 @@ def predict_product_sales():
     # Convert the extracted data into a Pandas DataFrame
     input_data = pd.DataFrame([sample])
 
-    # Make prediction (get log_price)
-    predicted_log_price = model.predict(input_data)[0]
-    print(f"Predicted log price: {predicted_log_price}")
+    # Make prediction (get Product_Store_Sales_Total)
+    predicted_Product_Store_Sales_Total = model.predict(input_data)[0]
+    print(f"Predicted Product_Store_Sales_Total: {predicted_Product_Store_Sales_Total}")
 
     # Calculate actual price
-    predicted_price = np.exp(predicted_log_price)
+    predicted_price = np.exp(predicted_Product_Store_Sales_Total)
 
     # Convert predicted_price to Python float
     predicted_price = round(float(predicted_price), 2)
@@ -77,10 +77,10 @@ def predict_product_sale_price_batch():
     input_data = pd.read_csv(file)
 
     # Make predictions for all product sale in the stores in the DataFrame (get log_prices)
-    predicted_log_prices = model.predict(input_data).tolist()
+    predicted_Product_Store_Sales_Total = model.predict(input_data).tolist()
 
     # Calculate actual prices
-    predicted_prices = [round(float(np.exp(log_price)), 2) for log_price in predicted_log_prices]
+    predicted_prices = [round(float(np.exp(total_sale_price)), 2) for total_sale_price in predicted_Product_Store_Sales_Total]
 
     # Create a dictionary of predictions with product IDs as keys
     product_ids = input_data['id'].tolist()  # Assuming 'id' is the product ID column