Java 类名：com.alibaba.alink.operator.batch.feature.FeatureHasherBatchOp
Python 类名：FeatureHasherBatchOp

功能介绍

将多个特征组合成一个特征向量。

参数说明

名称 中文名称 描述 类型 是否必须？ 取值范围 默认值
outputCol 输出结果列列名 输出结果列列名，必选 String ✓
selectedCols 选择的列名 计算列对应的列名列表 String[] ✓
categoricalCols 离散特征列名 离散特征列名 String[]
numFeatures 向量维度 生成向量长度 Integer 262144
reservedCols 算法保留列名 算法保留列 String[] null
numThreads 组件多线程线程个数 组件多线程线程个数 Integer 1

代码示例

Python 代码

  1. from pyalink.alink import *
  3. import pandas as pd
  5. useLocalEnv(1)
  7. df = pd.DataFrame([
  8.     [1.1, True, "2", "A"],
  9.     [1.1, False, "2", "B"],
  10.     [1.1, True, "1", "B"],
  11.     [2.2, True, "1", "A"]
  12. ])
  14. inOp1 = BatchOperator.fromDataframe(df, schemaStr='double double, bool boolean, number int, str string')
  15. inOp2 = StreamOperator.fromDataframe(df, schemaStr='double double, bool boolean, number int, str string')
  17. hasher = FeatureHasherBatchOp().setSelectedCols(["double", "bool", "number", "str"]).setOutputCol("output").setNumFeatures(200)
  18. hasher.linkFrom(inOp1).print()
  20. hasher = FeatureHasherStreamOp().setSelectedCols(["double", "bool", "number", "str"]).setOutputCol("output").setNumFeatures(200)
  21. hasher.linkFrom(inOp2).print()
  23. StreamOperator.execute()

Java 代码

  1. import org.apache.flink.types.Row;
  3. import com.alibaba.alink.operator.batch.BatchOperator;
  4. import com.alibaba.alink.operator.batch.feature.FeatureHasherBatchOp;
  5. import com.alibaba.alink.operator.batch.source.MemSourceBatchOp;
  6. import com.alibaba.alink.operator.stream.StreamOperator;
  7. import com.alibaba.alink.operator.stream.feature.FeatureHasherStreamOp;
  8. import com.alibaba.alink.operator.stream.source.MemSourceStreamOp;
  9. import org.junit.Test;
  11. import java.util.Arrays;
  12. import java.util.List;
  14. public class FeatureHasherBatchOpTest {
  15.     @Test
  16.     public void testFeatureHasherBatchOp() throws Exception {
  17.         List <Row> df = Arrays.asList(
  18.             Row.of(1.1, true, 2, "A"),
  19.             Row.of(1.1, false, 2, "B"),
  20.             Row.of(1.1, true, 1, "B"),
  21.             Row.of(2.2, true, 1, "A")
  22.         );
  23.         BatchOperator <?> inOp1 = new MemSourceBatchOp(df, "double double, bool boolean, number int, str string");
  24.         StreamOperator <?> inOp2 = new MemSourceStreamOp(df, "double double, bool boolean, number int, str string");
  25.         BatchOperator <?> hasher = new FeatureHasherBatchOp().setSelectedCols("double", "bool", "number", "str")
  26.             .setOutputCol("output").setNumFeatures(200);
  27.         hasher.linkFrom(inOp1).print();
  28.         StreamOperator <?> hasher2 = new FeatureHasherStreamOp().setSelectedCols("double", "bool", "number", "str")
  29.             .setOutputCol("output").setNumFeatures(200);
  30.         hasher2.linkFrom(inOp2).print();
  31.         StreamOperator.execute();
  32.     }
  33. }

运行结果

输出数据

| double | bool | number | str | output | | —- | —- | —- | —- | —- |

| 1.1000 | true | 2 | A | $200$13:2.0 38:1.1 45:1.0 195:1.0 |

| 1.1000 | false | 2 | B | $200$13:2.0 30:1.0 38:1.1 76:1.0 |

| 1.1000 | true | 1 | B | $200$13:1.0 38:1.1 76:1.0 195:1.0 |

| 2.2000 | true | 1 | A | $200$13:1.0 38:2.2 45:1.0 195:1.0 |