Project · Shipped
Laptop Price Predictor
Streamlit regression application that predicts laptop prices from hardware features and exposes data exploration alongside the prediction workflow.
Data and model
The dataset contains roughly 1,300 laptops with brand, type, memory, storage, processor, graphics, screen, resolution, operating system, weight, touch, and IPS features.
Feature engineering covers storage combinations, storage types, pixel density, and categorical encoding. A scikit-learn pipeline combines OneHotEncoder with ensemble tree regressors such as Random Forest and Gradient Boosting. The recorded R² is approximately 0.86.
Product
The Streamlit interface supports interactive prediction from selected specifications, data exploration, similar-laptop context, and a multi-page workflow deployed on Render.