绝对值最大化训练 (MaxAbsScalerTrainBatchOp)

Java 类名:com.alibaba.alink.operator.batch.dataproc.MaxAbsScalerTrainBatchOp

Python 类名:MaxAbsScalerTrainBatchOp

功能介绍

  • 绝对值最大标准化是对数据按照最大值和最小值进行标准化的组件, 将数据归一到-1和1之间。
  • 使用绝对值最大标准化预测组件使用生成的模型,转换输入的数据

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
selectedCols 选择的列名 计算列对应的列名列表 String[] 所选列类型为 [BIGDECIMAL, BIGINTEGER, BYTE, DOUBLE, FLOAT, INTEGER, LONG, SHORT]

代码示例

Python 代码

from pyalink.alink import *

import pandas as pd

useLocalEnv(1)

df = pd.DataFrame([
            ["a", 10.0, 100],
            ["b", -2.5, 9],
            ["c", 100.2, 1],
            ["d", -99.9, 100],
            ["a", 1.4, 1],
            ["b", -2.2, 9],
            ["c", 100.9, 1]
])
             
colnames = ["col1", "col2", "col3"]
selectedColNames = ["col2", "col3"]


inOp = BatchOperator.fromDataframe(df, schemaStr='col1 string, col2 double, col3 long')
         

# train
trainOp = MaxAbsScalerTrainBatchOp()\
           .setSelectedCols(selectedColNames)

trainOp.linkFrom(inOp)

# batch predict
predictOp = MaxAbsScalerPredictBatchOp()
predictOp.linkFrom(trainOp, inOp).print()

Java 代码

import org.apache.flink.types.Row;

import com.alibaba.alink.operator.batch.BatchOperator;
import com.alibaba.alink.operator.batch.dataproc.MaxAbsScalerPredictBatchOp;
import com.alibaba.alink.operator.batch.dataproc.MaxAbsScalerTrainBatchOp;
import com.alibaba.alink.operator.batch.source.MemSourceBatchOp;
import org.junit.Test;

import java.util.Arrays;
import java.util.List;

public class MaxAbsScalerTrainBatchOpTest {
	@Test
	public void testMaxAbsScalerTrainBatchOp() throws Exception {
		List <Row> df = Arrays.asList(
			Row.of("a", 10.0, 100),
			Row.of("b", -2.5, 9),
			Row.of("c", 100.2, 1),
			Row.of("d", -99.9, 100),
			Row.of("a", 1.4, 1),
			Row.of("b", -2.2, 9),
			Row.of("c", 100.9, 1)
		);

		String[] selectedColNames = new String[] {"col2", "col3"};
		BatchOperator <?> inOp = new MemSourceBatchOp(df, "col1 string, col2 double, col3 int");
		BatchOperator <?> trainOp = new MaxAbsScalerTrainBatchOp()
			.setSelectedCols(selectedColNames);
		trainOp.linkFrom(inOp);
		BatchOperator <?> predictOp = new MaxAbsScalerPredictBatchOp();
		predictOp.linkFrom(trainOp, inOp).print();
	}
}

运行结果

col1 col2 col3
a 0.0991 1.0000
b -0.0248 0.0900
c 0.9931 0.0100
d -0.9901 1.0000
a 0.0139 0.0100
b -0.0218 0.0900
c 1.0000 0.0100