Upload 8 files

apartment_price_model.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:476146d55b19877ff5a612807dc75bd5c2ceb18af22b5ca30f640e29644a5e34
+size 16501313

app.py

@@ -0,0 +1,76 @@
+import joblib
+import pandas as pd
+import numpy as np
+import gradio as gr
+import geopy.distance
+from sklearn.ensemble import RandomForestRegressor
+from sklearn.model_selection import train_test_split
+
+# ------------------------------
+# Load Data
+# ------------------------------
+DATA_PATH = "apartments_data_enriched_lat_lon_combined.csv"
+df = pd.read_csv(DATA_PATH)
+
+# Handle missing coordinate columns
+lat_col = "latitude" if "latitude" in df.columns else "lat"
+lon_col = "longitude" if "longitude" in df.columns else "lon"
+if lat_col not in df.columns or lon_col not in df.columns:
+    raise KeyError("Latitude and Longitude columns are missing!")
+
+# Define public transport stations
+public_transit_stations = [
+    {"name": "Hauptbahnhof", "lat": 47.378177, "lon": 8.540192},
+    {"name": "Bahnhof Stadelhofen", "lat": 47.366321, "lon": 8.548008},
+    {"name": "Hardbrücke", "lat": 47.385118, "lon": 8.517220},
+    {"name": "Enge", "lat": 47.364751, "lon": 8.531601}
+]
+
+# Function to compute distance to nearest station
+def distance_to_nearest_station(lat, lon):
+    min_distance = np.inf
+    for station in public_transit_stations:
+        dist = geopy.distance.geodesic((lat, lon), (station["lat"], station["lon"]))
+        min_distance = min(min_distance, dist.km)
+    return min_distance
+
+# Compute new feature
+df["distance_to_transit"] = df.apply(lambda row: distance_to_nearest_station(row[lat_col], row[lon_col]), axis=1)
+
+# ------------------------------
+# Train Model
+# ------------------------------
+features = ["rooms", "area", "pop_dens", "tax_income", "distance_to_transit"]
+X = df[features]
+y = df["price"]
+X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
+
+model = RandomForestRegressor(n_estimators=100, random_state=42)
+model.fit(X_train, y_train)
+
+# Save model
+joblib.dump(model, "apartment_price_model.pkl")
+
+# ------------------------------
+# Gradio Web Interface
+# ------------------------------
+def predict_price(rooms, area, pop_dens, tax_income, distance_to_transit):
+    model = joblib.load("apartment_price_model.pkl")
+    input_data = pd.DataFrame([[rooms, area, pop_dens, tax_income, distance_to_transit]], columns=features)
+    prediction = model.predict(input_data)[0]
+    return f"Geschätzter Preis: {prediction:.2f} CHF"
+
+app = gr.Interface(
+    fn=predict_price,
+    inputs=[
+        gr.Number(label="Anzahl Zimmer"),
+        gr.Number(label="Fläche (m²)"),
+        gr.Number(label="Bevölkerungsdichte"),
+        gr.Number(label="Steuerbares Einkommen"),
+        gr.Number(label="Distanz zur nächsten ÖV-Station (km)")
+    ],
+    outputs="text"
+)
+
+if __name__ == "__main__":
+    app.launch()

data_with_large_residuals.csv

@@ -0,0 +1,15 @@
+rooms,area,pop,pop_dens,frg_pct,emp,tax_income,room_per_m2,luxurious,temporary,furnished,area_cat_ecoded,zurich_city,recidual (error),price,predicted_price,description_raw,bfs_name,postalcode,town
+2.0,54,29944,10008.0,32.4584678868,491193.0,85446,27.0,0,0,0,2.0,1,902.5900000000001,4853,3950.41,«2 Bedroom Apartment Senior Balcony»,Zürich,8004, Zürich
+2.0,49,50950,5792.0,32.4584678868,491193.0,85446,24.5,0,0,0,0.0,1,981.3466666666668,4335,3353.653333333333,«2 Bedroom Apartment Junior Terrace»,Zürich,8003, Zürich
+3.0,76,50950,5792.0,32.4584678868,491193.0,85446,25.33,0,0,0,2.0,1,923.5333333333338,1970,2893.5333333333338,«3- Zimmerwohnung im Kreis 3 direkt beim Goldbrunnenplatz»,Zürich,8003, Zürich
+2.5,65,41411,4512.0,32.4584678868,491193.0,85446,26.0,0,0,0,2.0,1,887.0349999999999,5900,5012.965,«Serviced 1-bedroom with great view and luxurious interior design»,Zürich,8037, Zürich
+2.5,75,41411,4512.0,32.4584678868,491193.0,85446,30.0,0,0,0,2.0,1,1016.3000000000002,6900,5883.7,"«""Serviced 1-bedroom apartment: spectacular view & luxurious interior»",Zürich,8037, Zürich
+3.5,90,41411,4512.0,32.4584678868,491193.0,85446,25.71,0,1,0,2.0,1,941.585,1750,2691.585,«BEFRISTETES WOHNEN BIS 31. MÄRZ 2024»,Zürich,8049, Zürich
+2.5,70,78801,5736.0,32.4584678868,491193.0,85446,28.0,0,0,0,2.0,1,1590.6400000000003,5900,4309.36,«Spacious apartment with spectacular view and luxururious design»,Zürich,8057, Zürich
+3.5,126,37639,3254.0,32.4584678868,491193.0,85446,36.0,0,0,0,1.0,1,1282.1999999999998,6575,5292.8,«Live Above - Exklusive Mietwohnung über dem FIFA World Football Museum»,Zürich,8002, Zürich
+2.5,70,39647,2574.0,32.4584678868,491193.0,85446,28.0,0,0,0,2.0,1,811.4266666666672,5900,5088.573333333333,"«""Wohnoase im Englisch Viertel - Serviced Apartment in Jugendstil-Villa»",Zürich,8032, Zürich
+4.5,137,39647,2574.0,32.4584678868,491193.0,85446,30.44,0,0,0,1.0,1,1573.3366666666666,875,2448.3366666666666,«HISTORISCH MIT PRIVATER SONNENTERRASSE IM ZENTRUM»,Zürich,8001, Zürich
+5.5,200,113173,1662.5973262818,24.535003932,73339.0,70966,36.36,0,0,0,1.0,0,899.6033333333335,5600,4700.3966666666665,"«RUHIG UND SONNIG, MIT WUNDERSCHÖNER AUSSICHT»",Winterthur,8404, Winterthur
+3.5,130,18139,3298.0,29.0313688737,6582.0,113194,37.14,0,0,0,1.0,0,1256.3100000000004,6490,5233.69,«Weitblick im neu renovierten 2-Schlafzimmer-Serviced Apartment»,Thalwil,8800, Thalwil
+3.5,100,13067,1664.5859872612,26.0580087243,5764.0,171602,28.57,0,0,0,2.0,0,812.1949999999997,5550,4737.805,«3 1/2-Zimmer-Wohnung mit Seesicht»,Zollikon,8702, Zollikon
+4.5,101,35007,1228.7469287469,23.0096837775,17237.0,79839,22.44,0,0,0,1.0,0,861.7101666666663,3210,2348.2898333333337,«Helle und moderne Maisonette-Wohnung mit Galerie»,Uster,8610, Uster

requirement.txt

@@ -0,0 +1,8 @@
+scikit-learn
+geopy
+pandas
+numpy
+scikit-learn
+geopy
+gradio
+joblib