MTable展开 (FlattenMTableBatchOp)

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, STRING]
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

import numpy as np
import pandas as pd
from pyalink.alink import *

df_data = pd.DataFrame([
      ["a1", "11L", 2.2],
      ["a1", "12L", 2.0],
      ["a2", "11L", 2.0],
      ["a2", "12L", 2.0],
      ["a3", "12L", 2.0],
      ["a3", "13L", 2.0],
      ["a4", "13L", 2.0],
      ["a4", "14L", 2.0],
      ["a5", "14L", 2.0],
      ["a5", "15L", 2.0],
      ["a6", "15L", 2.0],
      ["a6", "16L", 2.0]
])

input = BatchOperator.fromDataframe(df_data, schemaStr='id string, f0 string, f1 double')

zip = GroupByBatchOp()\
	.setGroupByPredicate("id")\
	.setSelectClause("id, mtable_agg(f0, f1) as m_table_col")

flatten = FlattenMTableBatchOp()\
	.setReservedCols(["id"])\
	.setSelectedCol("m_table_col")\
	.setSchemaStr('f0 string, f1 int')

zip.linkFrom(input).link(flatten).print()

Java 代码

package com.alibaba.alink.operator.batch.dataproc;

import org.apache.flink.types.Row;

import com.alibaba.alink.operator.batch.BatchOperator;
import com.alibaba.alink.operator.batch.sql.GroupByBatchOp;
import com.alibaba.alink.operator.batch.source.MemSourceBatchOp;
import com.alibaba.alink.testutil.AlinkTestBase;
import org.junit.Test;

import java.util.ArrayList;
import java.util.List;

/**
 * Test cases for gbdt.
 */

public class FlattenMTableTest extends AlinkTestBase {
	
	@Test
	public void test() throws Exception {
		List <Row> rows = new ArrayList <>();
		rows.add(Row.of("a1", "11L", 2.2));

		rows.add(Row.of("a1", "12L", 2.0));
		rows.add(Row.of("a2", "11L", 2.0));
		rows.add(Row.of("a2", "12L", 2.0));
		rows.add(Row.of("a3", "12L", 2.0));
		rows.add(Row.of("a3", "13L", 2.0));
		rows.add(Row.of("a4", "13L", 2.0));
		rows.add(Row.of("a4", "14L", 2.0));
		rows.add(Row.of("a5", "14L", 2.0));
		rows.add(Row.of("a5", "15L", 2.0));
		rows.add(Row.of("a6", "15L", 2.0));
		rows.add(Row.of("a6", "16L", 2.0));

		BatchOperator input = new MemSourceBatchOp(rows, "id string, f0 string, f1 double");

		GroupByBatchOp zip = new GroupByBatchOp()
			.setGroupByPredicate("id")
			.setSelectClause("id, mtable_agg(f0, f1) as m_table_col");

		FlattenMTableBatchOp flatten = new FlattenMTableBatchOp()
			.setReservedCols("id")
			.setSelectedCol("m_table_col")
			.setSchemaStr("f0 string, f1 int");

		zip.linkFrom(input).link(flatten).print();
	}
}

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