XGBoost二分类训练 (XGBoostClassifier)

Java 类名：com.alibaba.alink.pipeline.classification.XGBoostClassifier
Python 类名：XGBoostClassifier

功能介绍

XGBoost 组件是在开源社区的基础上进行包装，使功能和 PAI 更兼容，更易用。
XGBoost 算法在 Boosting 算法的基础上进行了扩展和升级，具有较好的易用性和鲁棒性，被广泛用在各种机器学习生产系统和竞赛领域。
当前支持分类，回归和排序。

参数说明

代码示例

以下代码仅用于示意，可能需要修改部分代码或者配置环境后才能正常运行！

Python 代码

df = pd.DataFrame([
    [0, 1, 1.1, 1.0],
    [1, -2, 0.9, 2.0],
    [0, 100, -0.01, 3.0],
    [1, -99, 0.1, 4.0],
    [0, 1, 1.1, 5.0],
    [1, -2, 0.9, 6.0]
])
batchSource = BatchOperator.fromDataframe(
    df, schemaStr='y int, x1 int, x2 double, x3 double'
)
streamSource = StreamOperator.fromDataframe(
    df, schemaStr='y int, x1 int, x2 double, x3 double'
)
xgboostClassifier = XGBoostClassifier() \
    .setNumRound(1)\
    .setPluginVersion('1.5.1')\
    .setLabelCol('y')\
    .setPredictionDetailCol('pred_detail')\
    .setPredictionCol('pred')
xgboostClassifier.fit(batchSource).transform(streamSource).print()
StreamOperator.execute()

Java 代码

import org.apache.flink.types.Row;
import com.alibaba.alink.operator.batch.BatchOperator;
import com.alibaba.alink.operator.batch.source.MemSourceBatchOp;
import com.alibaba.alink.operator.stream.StreamOperator;
import com.alibaba.alink.operator.stream.source.MemSourceStreamOp;
import com.alibaba.alink.pipeline.classification.XGBoostClassifier;
import org.junit.Test;
import java.util.Arrays;
import java.util.List;
public class XGBoostClassifierTest {
    @Test
    public void testXGBoostClassifier() throws Exception {
        List <Row> data = Arrays.asList(
            Row.of(0, 1, 1.1, 1.0),
            Row.of(1, -2, 0.9, 2.0),
            Row.of(0, 100, -0.01, 3.0),
            Row.of(1, -99, 0.1, 4.0),
            Row.of(0, 1, 1.1, 5.0),
            Row.of(1, -2, 0.9, 6.0)
        );
        BatchOperator <?> batchSource = new MemSourceBatchOp(data, "y int, x1 int, x2 double, x3 double");
        StreamOperator <?> streamSource = new MemSourceStreamOp(data, "y int, x1 int, x2 double, x3 double");
        XGBoostClassifier xgBoostClassifier = new XGBoostClassifier()
            .setNumRound(1)
            .setPluginVersion("1.5.1")
            .setLabelCol("y")
            .setPredictionDetailCol("pred_detail")
            .setPredictionCol("pred");
        xgBoostClassifier.fit(batchSource).transform(streamSource).print();
        StreamOperator.execute();
    }
}