I have both historical sales figures and a live market feed ready, and I want to turn them into a production-ready house-price prediction system with strong, verifiable accuracy. The core dataset already captures location specifics, property size, and bedroom count. My preference is to start with Linear Regression for a transparent baseline, then move to a Random Forest ensemble to push performance. I’m open to your suggestions on feature engineering and validation strategies, but I do need side-by-side metrics that prove the uplift between the two models. What I need from you: • Well-documented Python code (Jupyter notebooks or scripts) that loads the data, cleans it, engineers features, trains both models, and outputs predictions. • A concise report or markdown read-me that explains preprocessing steps, hyper-parameters, and evaluation (e.g., MAE, RMSE, R²) on a held-out test set. • Exported, versioned model files ready for deployment, along with guidance on integrating the chosen model into a simple API or dashboard. • Reproducibility: requirements.txt or environment.yml so I can spin this up locally or on the cloud without hassle. If you can demonstrate previous work on property-valuation or similar tabular-data projects, that will help me decide quickly.