python-3.x - 如何将python xgboost模型转换为pmml？

Question

如何将 python xgboost 模型转换为 pmml ？

reg = XGBRegressor(learning_rate=0.1, n_estimators=30, max_depth=4, min_child_weight=4, gamma=0.1,
                       subsample=0.9, colsample_bytree=0.8, objective='binary:logistic', reg_alpha=1,
                       scale_pos_weight=1, seed=27)
param_test = [{
        'max_depth': [i for i in range(1, 3)],
        'gamma': [i / 10.0 for i in range(0, 10)],
        'n_estimators': [i for i in range(2, 14, 2)],
}]
gsearch = GridSearchCV(reg, param_grid=param_test, scoring='neg_mean_squared_error', n_jobs=4, iid=False, cv=5)
gsearch.fit(x_train, y_train)
best_model = gsearch.best_estimator_

Answer 1

请参阅 SkLear2PMML 包:https://github.com/jpmml/sklearn2pmml

首先，定义一个新的 pmml 管道，并将您的 XGBRegressor 插入其中。然后，使用 GridSearchCV 拟合 pmml 管道学习者。最后，导出 GridSearchCV.best_estimator_ - 这应该是优化的 pmml 管道 - 使用 sklearn2pmml.sklearn2pmml 转换为 PMML 数据格式函数调用:

pmml_pipeline = PMMLPipeline([
  ("regressor", XGBRegressor())
])
tuner = GridSearchCV(pmml_pipeline, ...)
tuner.fit(X, y)
sklearn2pmml(tuner.best_estimator_, "xgbregressor-pipeline.pmml")

另请参阅以下演示文稿的幻灯片 #26: https://www.slideshare.net/VilluRuusmann/converting-scikitlearn-to-pmml