About

This project implements a production-ready demand forecasting system using historical Walmart retail sales data.
The workflow performs data cleaning and weekly sales aggregation to prepare time-series data.
Lag-based feature engineering is applied to capture temporal demand patterns.
A regression-based predictive model is trained to forecast future demand values.
The dataset is split into training and testing sets for proper evaluation.
Model performance is validated using Mean Absolute Error (MAE).
Root Mean Squared Error (RMSE) is also used to measure prediction accuracy.
The system demonstrates reliable forecasting on unseen data.
The complete pipeline is deployed as a scheduled workflow in Zerve.
This solution helps businesses make data-driven inventory and planning decisions.