27 lines
689 B
Python
27 lines
689 B
Python
"""
|
|
Demo for using xgboost with sklearn
|
|
===================================
|
|
"""
|
|
import multiprocessing
|
|
|
|
from sklearn.datasets import fetch_california_housing
|
|
from sklearn.model_selection import GridSearchCV
|
|
|
|
import xgboost as xgb
|
|
|
|
if __name__ == "__main__":
|
|
print("Parallel Parameter optimization")
|
|
X, y = fetch_california_housing(return_X_y=True)
|
|
xgb_model = xgb.XGBRegressor(
|
|
n_jobs=multiprocessing.cpu_count() // 2, tree_method="hist"
|
|
)
|
|
clf = GridSearchCV(
|
|
xgb_model,
|
|
{"max_depth": [2, 4, 6], "n_estimators": [50, 100, 200]},
|
|
verbose=1,
|
|
n_jobs=2,
|
|
)
|
|
clf.fit(X, y)
|
|
print(clf.best_score_)
|
|
print(clf.best_params_)
|