向量缺失值填充训练 (VectorImputerTrainBatchOp)

Java 类名:com.alibaba.alink.operator.batch.dataproc.vector.VectorImputerTrainBatchOp

Python 类名:VectorImputerTrainBatchOp

功能介绍

训练Vecotor 缺失值填充模型的组件,输出模型。

填充策略包含最大值,最小值,均值和指定数值4种。

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
selectedCol 选中的列名 计算列对应的列名 String 所选列类型为 [DENSE_VECTOR, SPARSE_VECTOR, STRING, VECTOR]
fillValue 填充缺失值 自定义的填充值。当strategy为value时,读取fillValue的值 Double null
strategy 缺失值填充规则 缺失值填充的规则,支持mean,max,min或者value。选择value时,需要读取fillValue的值 String “MEAN”, “MIN”, “MAX”, “VALUE” “MEAN”

代码示例

Python 代码

from pyalink.alink import *

import pandas as pd

useLocalEnv(1)

df = pd.DataFrame([
    ["1:3,2:4,4:7", 1],
    ["1:3,2:NaN", 3],
    ["2:4,4:5", 4]
])

data = BatchOperator.fromDataframe(df, schemaStr="vec string, id bigint")
vecFill = VectorImputerTrainBatchOp().setSelectedCol("vec")
model = data.link(vecFill)
VectorImputerPredictBatchOp().setOutputCol("vec1").linkFrom(model, data).print()

Java 代码

import org.apache.flink.types.Row;

import com.alibaba.alink.operator.batch.BatchOperator;
import com.alibaba.alink.operator.batch.dataproc.vector.VectorImputerPredictBatchOp;
import com.alibaba.alink.operator.batch.dataproc.vector.VectorImputerTrainBatchOp;
import com.alibaba.alink.operator.batch.source.MemSourceBatchOp;
import org.junit.Test;

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

public class VectorImputerTrainBatchOpTest {
	@Test
	public void testVectorImputerTrainBatchOp() throws Exception {
		List <Row> df = Arrays.asList(
			Row.of("1:3,2:4,4:7", 1),
			Row.of("1:3,2:NaN", 3),
			Row.of("2:4,4:5", 4)
		);
		BatchOperator <?> data = new MemSourceBatchOp(df, "vec string, id int");
		BatchOperator <?> vecFill = new VectorImputerTrainBatchOp().setSelectedCol("vec");
		BatchOperator <?> model = data.link(vecFill);
		new VectorImputerPredictBatchOp().setOutputCol("vec1").linkFrom(model, data).print();
	}
}

运行结果

vec id vec1
1:3,2:4,4:7 1 1:3.0 2:4.0 4:7.0
1:3,2:NaN 3 1:3.0 2:4.0
2:4,4:5 4 2:4.0 4:5.0