metadata
language: uz
license: mit
tags:
- xgboost
- regression
- car-price-prediction
- automotive
- uzbekistan
metrics:
- r2: 0.9007
- mae: 682.02
🚗 XGBoost Avtomobil Narxini Bashorat Qilish (Uzbekistan Market)
Ushbu model O'zbekistondagi mashhur avtomobil modellari va ularning bozor narxlari asosida o'qitilgan. Model yili, masofasi va transmissiyasiga qarab avtomobil narxini ($) bashorat qiladi.
📊 Model Natijalari
- Aniqlik (R2 Score): 0.9007
- O'rtacha xatolik (MAE): $682.02
🛠 Modelni Ishlatish (Usage Guide)
Modelni yuklash va bashorat qilish uchun quyidagi Python kodidan foydalaning:
import xgboost as xgb
import pandas as pd
from huggingface_hub import hf_hub_download
# 1. Modelni yuklab olish
model_path = hf_hub_download(repo_id="Mehriddin1997/xgboost_car_model", filename="xgboost_car_model.json")
# 2. Modelni yuklash
model = xgb.XGBRegressor()
model.load_model(model_path)
# 3. Mashina modelini aniqlash (Dictionary)
car_models = {
1: "Captiva", 2: "Cobalt", 3: "Damas", 4: "Epica", 5: "Equinox",
6: "Gentra", 7: "Labo", 8: "Lacetti", 9: "Malibu", 10: "Matiz",
11: "Nexia", 12: "Onix", 13: "Orlando", 14: "Spark", 15: "Tracker"
}
def predict_car(year, mileage, transmission, model_id):
# Model o'qitilgan ustun nomlari bilan bir xil dataframe yaratamiz
data = pd.DataFrame([[year, mileage, transmission, model_id]],
columns=['year', 'yurgan_masofasi', 'transmission', 'model'])
price = model.predict(data)[0]
return price
# Test: 2022-yil, 35,000 km, Avtomat (1), Cobalt (2)
res = predict_car(2022, 35000, 1, 2)
print(f"Bashorat qilingan narx: ${res:,.2f}")