lightgbm的GPU版本和CPU版本运行速度比较

[1]中比较了catboost的GPU速度和lightgbm之间速度

[2]中的一个例子，GPU版本运行得比CPU还慢

 

另外注意，lightgbm的贝叶斯优化其实是过拟合，不要因为迷恋cv的上升而花费大量精力在这上面

 

[1]中的lightgbm的参数：

lgb_params = {
        'bagging_fraction': 0.77,
        'bagging_freq': 2,
        'lambda_l1': 0.7,
        'lambda_l2': 2,
        'learning_rate': 0.01,
        'max_depth': 3,
        'min_data_in_leaf': 22,
        'min_gain_to_split': 0.07,
        'min_sum_hessian_in_leaf': 19,
        'num_leaves': 20,
        'feature_fraction': 1,
        'save_binary': True,
        'seed': 42,
        'feature_fraction_seed': 42,
        'bagging_seed': 42,
        'drop_seed': 42,
        'data_random_seed': 42,
        'objective': 'binary',
        'boosting_type': 'gbdt',
        'verbosity': -1,
        'metric': 'auc',
        'is_unbalance': True,
        'boost_from_average': 'false',
        'num_threads': 6,
        'device': 'gpu',
        'gpu_platform_id': 0,
        'gpu_device_id': 0
}

比较效果如下：

Reference:

[1]https://www.kaggle.com/raimonds1993/gbms-cpu-vs-gpu-400-features-augmentation

[2]https://github.com/FedericoRaimondi/myProjects/blob/master/Santander_Customer_Transaction_Prediction/PredictiveAnalysis_ModelTuning/PredictiveAnalysis_ModelTuning_GPU.ipynb