ALINK(二十八)：特征工程（七）特征组合与交叉（二）Cross特征预测/训练 (CrossFeaturePredictBatchOp)

Cross特征预测 (CrossFeaturePredictBatchOp)

Java 类名：com.alibaba.alink.operator.batch.feature.CrossFeaturePredictBatchOp

Python 类名：CrossFeaturePredictBatchOp

功能介绍

特征列组合算法能够将选定的离散列组合成单列的向量类型的数据。

参数说明

名称	中文名称	描述	类型	是否必须？	默认值
outputCol	输出结果列列名	输出结果列列名，必选	String	✓
numThreads	组件多线程线程个数	组件多线程线程个数	Integer		1
modelStreamFilePath	模型流的文件路径	模型流的文件路径	String		null
modelStreamScanInterval	扫描模型路径的时间间隔	描模型路径的时间间隔，单位秒	Integer		10
modelStreamStartTime	模型流的起始时间	模型流的起始时间。默认从当前时刻开始读。使用yyyy-mm-dd hh:mm:ss.fffffffff格式，详见Timestamp.valueOf(String s)	String		null

代码示例

Python 代码

from pyalink.alink import *
import pandas as pd
useLocalEnv(1)
df = pd.DataFrame([
["1.0", "1.0", 1.0, 1],
["1.0", "1.0", 0.0, 1],
["1.0", "0.0", 1.0, 1],
["1.0", "0.0", 1.0, 1],
["2.0", "3.0", None, 0],
["2.0", "3.0", 1.0, 0],
["0.0", "1.0", 2.0, 0],
["0.0", "1.0", 1.0, 0]])
data = BatchOperator.fromDataframe(df, schemaStr="f0 string, f1 string, f2 double, label bigint")
train = CrossFeatureTrainBatchOp().setSelectedCols(['f0','f1','f2']).linkFrom(data)
CrossFeaturePredictBatchOp().setOutputCol("cross").linkFrom(train, data).collectToDataFrame()

Java 代码

import org.apache.flink.types.Row;
import com.alibaba.alink.operator.batch.BatchOperator;
import com.alibaba.alink.operator.batch.feature.CrossFeaturePredictBatchOp;
import com.alibaba.alink.operator.batch.feature.CrossFeatureTrainBatchOp;
import com.alibaba.alink.operator.batch.source.MemSourceBatchOp;
import org.junit.Test;
import java.util.Arrays;
import java.util.List;
public class CrossFeaturePredictBatchOpTest {
  @Test
  public void testCrossFeaturePredictBatchOp() throws Exception {
    List <Row> df = Arrays.asList(
      Row.of("1.0", "1.0", 1.0, 1),
      Row.of("1.0", "1.0", 0.0, 1),
      Row.of("1.0", "0.0", 1.0, 1),
      Row.of("1.0", "0.0", 1.0, 1),
      Row.of("2.0", "3.0", null, 0),
      Row.of("2.0", "3.0", 1.0, 0),
      Row.of("0.0", "1.0", 2.0, 0)
    );
    BatchOperator <?> data = new MemSourceBatchOp(df, "f0 string, f1 string, f2 double, label int");
    BatchOperator <?> train = new CrossFeatureTrainBatchOp().setSelectedCols("f0", "f1", "f2").linkFrom(data);
    new CrossFeaturePredictBatchOp().setOutputCol("cross").linkFrom(train, data).print();
  }
}

运行结果

f0	f1	f2	label	cross
1.0	1.0	1.0000	1	$36$0:1.0
1.0	1.0	0.0000	1	$36$9:1.0
1.0	0.0	1.0000	1	$36$6:1.0
1.0	0.0	1.0000	1	$36$6:1.0
2.0	3.0	null	0	$36$22:1.0
2.0	3.0	1.0000	0	$36$4:1.0
0.0	1.0	2.0000	0	$36$29:1.0
0.0	1.0	1.0000	0	$36$2:1.0

Cross特征训练 (CrossFeatureTrainBatchOp)

Java 类名：com.alibaba.alink.operator.batch.feature.CrossFeatureTrainBatchOp

Python 类名：CrossFeatureTrainBatchOp