时间序列插值 (LookupValueInTimeSeriesBatchOp)

Java 类名：com.alibaba.alink.operator.batch.timeseries.LookupValueInTimeSeriesBatchOp

Python 类名：LookupValueInTimeSeriesBatchOp

功能介绍

在时间序列中查找对应时间的值。

注意事项

• 时间序列列，是特殊的MTable类型，一列是时间，一列是值，参考运行结果的data列。
• 查找的时间在数据列中不存在时，用相邻时刻的值差值得到结果。

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	✓
timeCol	时间戳列(TimeStamp)	时间戳列(TimeStamp)	String	✓	所选列类型为 [TIMESTAMP]
timeSeriesCol	时间序列列	时间序列列，是特殊的MTable类型，一列是时间，一列是值	String	✓	所选列类型为 [M_TABLE, STRING]
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

from pyalink.alink import *

import pandas as pd

useLocalEnv(1)

import time, datetime
import numpy as np
import pandas as pd

data = pd.DataFrame([
			[1,  datetime.datetime.fromtimestamp(1), 10.0],
			[1,  datetime.datetime.fromtimestamp(2), 11.0],
			[1,  datetime.datetime.fromtimestamp(3), 12.0],
			[1,  datetime.datetime.fromtimestamp(4), 13.0],
			[1,  datetime.datetime.fromtimestamp(5), 14.0],
			[1,  datetime.datetime.fromtimestamp(6), 15.0],
			[1,  datetime.datetime.fromtimestamp(7), 16.0],
			[1,  datetime.datetime.fromtimestamp(8), 17.0],
			[1,  datetime.datetime.fromtimestamp(9), 18.0],
			[1,  datetime.datetime.fromtimestamp(10), 19.0]
])

source = dataframeToOperator(data, schemaStr='id int, ts timestamp, val double', op_type='batch')

source.link(
        GroupByBatchOp()
			.setGroupByPredicate("id")
			.setSelectClause("id, mtable_agg(ts, val) as data")
		).link(
            ShiftBatchOp()
					.setValueCol("data")
					.setShiftNum(7)
					.setPredictNum(12)
					.setPredictionCol("predict")
		).link(
            FlattenMTableBatchOp()
					.setReservedCols(["id", "predict"])
					.setSelectedCol("predict")
					.setSchemaStr("ts timestamp, val double")
        ).link(
            LookupValueInTimeSeriesBatchOp()
				.setTimeCol("ts")
				.setTimeSeriesCol("predict")
				.setOutputCol("out")
				.setReservedCols(["id","ts"])
        ).print()

Java 代码

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

import org.apache.flink.types.Row;

import com.alibaba.alink.common.MTable;
import com.alibaba.alink.operator.batch.source.MemSourceBatchOp;
import org.junit.Test;

import java.sql.Timestamp;
import java.util.Arrays;
import java.util.List;

public class LookupValueInTimeSeriesBatchOpTest {
	@Test
	public void test() throws Exception {

		List <Row> mTableData = Arrays.asList(
			Row.of(new Timestamp(1), 10.0),
			Row.of(new Timestamp(2), 11.0),
			Row.of(new Timestamp(3), 12.0),
			Row.of(new Timestamp(4), 13.0),
			Row.of(new Timestamp(5), 14.0),
			Row.of(new Timestamp(6), 15.0),
			Row.of(new Timestamp(7), 16.0),
			Row.of(new Timestamp(8), 17.0),
			Row.of(new Timestamp(9), 18.0),
			Row.of(new Timestamp(10), 19.0)
		);

		MTable mtable = new MTable(mTableData, "ts timestamp, val double");

		MemSourceBatchOp source = new MemSourceBatchOp(
			new Object[][] {
				{1, new Timestamp(5), mtable}
			},
			new String[] {"id", "ts", "data"});

		source
			.link(new LookupValueInTimeSeriesBatchOp()
				.setTimeCol("ts")
				.setTimeSeriesCol("data")
				.setOutputCol("out")
				.setReservedCols("id","ts")
			)
			.print();
	}
}

运行结果

id	ts	data	out
1	1970-01-01 08:00:00.005	MTable(10,2)(ts,val) 1970-01-01 08:00:00.001 | 10.000 1970-01-01 08:00:00.002 | 11.000 1970-01-01 08:00:00.003 | 12.000 1970-01-01 08:00:00.004 | 13.000 1970-01-01 08:00:00.005 | 14.000	14.0000