Sales revenue forecasting using linear regression to support inventory and regional planning at SuperKart. 🛒 SuperKart Sales Revenue Forecasting
This project involved developing and deploying a linear regression model to forecast sales revenue across SuperKart's retail outlets. The goal was to support data-driven planning for inventory management and regional sales strategies.
Project Highlights
Forecast Model: Built and deployed a linear regression model to predict sales revenue across retail locations.
Performance: Achieved an R² score of 0.91 and MAPE of 5.9%, enabling accurate demand forecasting.
Streamlit App: Created an interactive frontend hosted on Hugging Face Spaces, allowing real-time single and batch predictions.
Integration: Backend API built with Flask and model serialized using joblib for seamless deployment.
Business Impact: Improved procurement efficiency and regional planning across SuperKart’s multi-city network.
Tools & Technologies Machine Learning: Linear Regression, sklearn.metrics, model evaluation Deployment: Flask, jsonify, model serialization with joblib Frontend: Streamlit Packaging: Docker, Dockerfile Hosting: Hugging Face Spaces