import joblib
from sklearn.ensemble import RandomForestClassifier
from sklearn.datasets import make_classification
from sklearn.metrics import accuracy_score

# Thứ tự câu hỏi (13)
FEATURE_ORDER = [str(i) for i in range(1, 14)]

# Sinh dữ liệu giả
X, y = make_classification(n_samples=100, n_features=13, n_informative=6, random_state=42)

# Huấn luyện
model = RandomForestClassifier()
model.fit(X, y)

# Accuracy
print(f"Accuracy: {accuracy_score(y, model.predict(X)):.2f}")

# Lưu mô hình
joblib.dump(model, "ml_model.pkl")
joblib.dump(["binh thuong", "lao_or_copd"], "label_encoder.pkl")