Java 类名：com.alibaba.alink.operator.batch.dataproc.ImputerPredictBatchOp
Python 类名：ImputerPredictBatchOp

功能介绍

数据缺失值填充处理，批式预测组件
运行时需要指定缺失值模型，由ImputerTrainBatchOp产生。缺失值填充的4种策略，即最大值、最小值、均值、指定数值，在生成缺失值模型时指定。

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
modelFilePath 模型的文件路径 模型的文件路径 String null
outputCols 输出结果列列名数组 输出结果列列名数组，可选，默认null 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_data = pd.DataFrame([
  8.             ["a", 10.0, 100],
  9.             ["b", -2.5, 9],
  10.             ["c", 100.2, 1],
  11.             ["d", -99.9, 100],
  12.             ["a", 1.4, 1],
  13.             ["b", -2.2, 9],
  14.             ["c", 100.9, 1],
  15.             [None, None, None]
  16. ])
  18. colnames = ["col1", "col2", "col3"]
  19. selectedColNames = ["col2", "col3"]
  21. inOp = BatchOperator.fromDataframe(df_data, schemaStr='col1 string, col2 double, col3 double')
  23. # train
  24. trainOp = ImputerTrainBatchOp()\
  25.            .setSelectedCols(selectedColNames)
  27. model = trainOp.linkFrom(inOp)
  29. # batch predict
  30. predictOp = ImputerPredictBatchOp()
  31. predictOp.linkFrom(model, inOp).print()
  33. # stream predict
  34. sinOp = StreamOperator.fromDataframe(df_data, schemaStr='col1 string, col2 double, col3 double')
  36. predictStreamOp = ImputerPredictStreamOp(model)
  37. predictStreamOp.linkFrom(sinOp).print()
  39. 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.dataproc.ImputerPredictBatchOp;
  5. import com.alibaba.alink.operator.batch.dataproc.ImputerTrainBatchOp;
  6. import com.alibaba.alink.operator.batch.source.MemSourceBatchOp;
  7. import com.alibaba.alink.operator.stream.StreamOperator;
  8. import com.alibaba.alink.operator.stream.dataproc.ImputerPredictStreamOp;
  9. import com.alibaba.alink.operator.stream.source.MemSourceStreamOp;
  10. import org.junit.Test;
  12. import java.util.Arrays;
  13. import java.util.List;
  15. public class ImputerPredictBatchOpTest {
  16.     @Test
  17.     public void testImputerPredictBatchOp() throws Exception {
  18.         List <Row> df_data = Arrays.asList(
  19.             Row.of("a", 10.0, 100),
  20.             Row.of("b", -2.5, 9),
  21.             Row.of("c", 100.2, 1),
  22.             Row.of("d", -99.9, 100),
  23.             Row.of("a", 1.4, 1),
  24.             Row.of("b", -2.2, 9),
  25.             Row.of("c", 100.9, 1),
  26.             Row.of(null, null, null)
  27.         );
  29.         String[] selectedColNames = new String[] {"col2", "col3"};
  30.         BatchOperator <?> inOp = new MemSourceBatchOp(df_data, "col1 string, col2 double, col3 int");
  31.         BatchOperator <?> trainOp = new ImputerTrainBatchOp()
  32.             .setSelectedCols(selectedColNames);
  33.         BatchOperator model = trainOp.linkFrom(inOp);
  34.         BatchOperator <?> predictOp = new ImputerPredictBatchOp();
  35.         predictOp.linkFrom(model, inOp).print();
  36.         StreamOperator <?> sinOp = new MemSourceStreamOp(df_data, "col1 string, col2 double, col3 int");
  37.         StreamOperator <?> predictStreamOp = new ImputerPredictStreamOp(model);
  38.         predictStreamOp.linkFrom(sinOp).print();
  39.         StreamOperator.execute();
  40.     }
  41. }

运行结果

| col1 | col2 | col3 | | —- | —- | —- |

| a | 10.000000 | 100 |

| b | -2.500000 | 9 |

| c | 100.200000 | 1 |

| d | -99.900000 | 100 |

| a | 1.400000 | 1 |

| b | -2.200000 | 9 |

| c | 100.900000 | 1 |

| null | 15.414286 | 31 |