Small FastAPI backend that evaluates an apartment's asking price using recent comparable sales and long-term trends.
Data source (for now): a local CSV at data/transactions.csv (synthetic/dev).
- Filters comparable sales in the same city + neighborhood, exact rooms, and ±8% size tolerance.
- Computes a fair price per sqm from the last 2 years and classifies asking price as cheap / fair / expensive with a ±4% band.
- Returns recent KPIs (avg/min/max), long-term bucketed means (~2-year windows across ~10 years), and sales volume of the last 5 years.
- Python 3.10+ (tested on macOS)
- CSV at
data/transactions.csv(already provided)
python3 -m pip install -r requirements.txt