SuperKart Sales Forecasting Model
This model predicts Product_Store_Sales_Total using product and store attributes.
Features
- Numeric features: ['Product_Weight', 'Product_Allocated_Area', 'Product_MRP']
- Categorical features: ['Product_Sugar_Content', 'Product_Type', 'Store_Establishment_Year', 'Store_Size', 'Store_Location_City_Type', 'Store_Type']
Final Selected Model
Random Forest Base
Model Selection Reason
The Random Forest Base model was selected because it gave the best test performance among the evaluated models, with lower RMSE and higher R-squared on unseen test data compared with the tuned alternatives.
Intended Use
This model supports sales forecasting for inventory planning, supply-chain procurement, and store-level sales planning.