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

功能介绍

该组件将 MTable 展开成 Table。

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
schemaStr Schema Schema。格式为”colname coltype[, colname2, coltype2[, …]]”，例如”f0 string, f1 bigint, f2 double” String ✓
selectedCol 选中的列名 计算列对应的列名 String ✓ 所选列类型为 [M_TABLE]
handleInvalidMethod 处理无效值的方法 处理无效值的方法，可取 error, skip String “ERROR”, “SKIP” “ERROR”
reservedCols 算法保留列名 算法保留列 String[] null

代码示例

Python 代码

  1. import numpy as np
  2. import pandas as pd
  3. from pyalink.alink import *
  5. df_data = pd.DataFrame([
  6.       ["a1", "11L", 2.2],
  7.       ["a1", "12L", 2.0],
  8.       ["a2", "11L", 2.0],
  9.       ["a2", "12L", 2.0],
  10.       ["a3", "12L", 2.0],
  11.       ["a3", "13L", 2.0],
  12.       ["a4", "13L", 2.0],
  13.       ["a4", "14L", 2.0],
  14.       ["a5", "14L", 2.0],
  15.       ["a5", "15L", 2.0],
  16.       ["a6", "15L", 2.0],
  17.       ["a6", "16L", 2.0]
  18. ])
  20. input = BatchOperator.fromDataframe(df_data, schemaStr='id string, f0 string, f1 double')
  22. zip = GroupByBatchOp()\
  23.     .setGroupByPredicate("id")\
  24.     .setSelectClause("id, mtable_agg(f0, f1) as m_table_col")
  26. flatten = FlattenMTableBatchOp()\
  27.     .setReservedCols(["id"])\
  28.     .setSelectedCol("m_table_col")\
  29.     .setSchemaStr('f0 string, f1 int')
  31. zip.linkFrom(input).link(flatten).print()

Java 代码

  1. package com.alibaba.alink.operator.batch.dataproc;
  3. import org.apache.flink.types.Row;
  5. import com.alibaba.alink.operator.batch.BatchOperator;
  6. import com.alibaba.alink.operator.batch.sql.GroupByBatchOp;
  7. import com.alibaba.alink.operator.batch.source.MemSourceBatchOp;
  8. import com.alibaba.alink.testutil.AlinkTestBase;
  9. import org.junit.Test;
  11. import java.util.ArrayList;
  12. import java.util.List;
  14. /**
  15.  * Test cases for gbdt.
  16.  */
  18. public class FlattenMTableTest extends AlinkTestBase {
  20.     @Test
  21.     public void test() throws Exception {
  22.         List <Row> rows = new ArrayList <>();
  23.         rows.add(Row.of("a1", "11L", 2.2));
  25.         rows.add(Row.of("a1", "12L", 2.0));
  26.         rows.add(Row.of("a2", "11L", 2.0));
  27.         rows.add(Row.of("a2", "12L", 2.0));
  28.         rows.add(Row.of("a3", "12L", 2.0));
  29.         rows.add(Row.of("a3", "13L", 2.0));
  30.         rows.add(Row.of("a4", "13L", 2.0));
  31.         rows.add(Row.of("a4", "14L", 2.0));
  32.         rows.add(Row.of("a5", "14L", 2.0));
  33.         rows.add(Row.of("a5", "15L", 2.0));
  34.         rows.add(Row.of("a6", "15L", 2.0));
  35.         rows.add(Row.of("a6", "16L", 2.0));
  37.         BatchOperator input = new MemSourceBatchOp(rows, "id string, f0 string, f1 double");
  39.         GroupByBatchOp zip = new GroupByBatchOp()
  40.             .setGroupByPredicate("id")
  41.             .setSelectClause("id, mtable_agg(f0, f1) as m_table_col");
  43.         FlattenMTableBatchOp flatten = new FlattenMTableBatchOp()
  44.             .setReservedCols("id")
  45.             .setSelectedCol("m_table_col")
  46.             .setSchemaStr("f0 string, f1 int");
  48.         zip.linkFrom(input).link(flatten).print();
  49.     }
  50. }

运行结果

| id | f0 | f1 | | —- | —- | —- |

| a2 | 11L | 2 |

| a2 | 12L | 2 |

| a4 | 13L | 2 |

| a4 | 14L | 2 |

| a5 | 14L | 2 |

| a5 | 15L | 2 |

| a1 | 11L | 2 |

| a1 | 12L | 2 |

| a3 | 12L | 2 |

| a3 | 13L | 2 |

| a6 | 15L | 2 |

| a6 | 16L | 2 |