Java 类名:com.alibaba.alink.pipeline.dataproc.StandardScaler
Python 类名:StandardScaler
标准化是对数据进行按正态化处理的组件
训练过程计算数据的均值和标准差,生成标准化模型
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[] | ✓ | ||
modelFilePath | 模型的文件路径 | 模型的文件路径 | String | null | ||
outputCols | 输出结果列列名数组 | 输出结果列列名数组,可选,默认null | String[] | null | ||
overwriteSink | 是否覆写已有数据 | 是否覆写已有数据 | Boolean | false | ||
withMean | 是否使用均值 | 是否使用均值,默认使用 | Boolean | true | ||
withStd | 是否使用标准差 | 是否使用标准差,默认使用 | Boolean | true | ||
numThreads | 组件多线程线程个数 | 组件多线程线程个数 | Integer | 1 | ||
modelStreamFilePath | 模型流的文件路径 | 模型流的文件路径 | String | null | ||
modelStreamScanInterval | 扫描模型路径的时间间隔 | 描模型路径的时间间隔,单位秒 | Integer | 10 | ||
modelStreamStartTime | 模型流的起始时间 | 模型流的起始时间。默认从当前时刻开始读。使用yyyy-mm-dd hh:mm:ss.fffffffff格式,详见Timestamp.valueOf(String s) | String | null |
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') sinOp = StreamOperator.fromDataframe(df, schemaStr='col1 string, col2 double, col3 long') model = StandardScaler()\ .setSelectedCols(selectedColNames)\ .fit(inOp) model.transform(inOp).print() model.transform(sinOp).print() StreamOperator.execute()
import org.apache.flink.types.Row; import com.alibaba.alink.operator.batch.BatchOperator; import com.alibaba.alink.operator.batch.source.MemSourceBatchOp; import com.alibaba.alink.operator.stream.StreamOperator; import com.alibaba.alink.operator.stream.source.MemSourceStreamOp; import com.alibaba.alink.pipeline.dataproc.StandardScaler; import com.alibaba.alink.pipeline.dataproc.StandardScalerModel; import org.junit.Test; import java.util.Arrays; import java.util.List; public class StandardScalerTest { @Test public void testStandardScaler() 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"); StreamOperator <?> sinOp = new MemSourceStreamOp(df, "col1 string, col2 double, col3 int"); StandardScalerModel model = new StandardScaler() .setSelectedCols(selectedColNames) .fit(inOp); model.transform(inOp).print(); model.transform(sinOp).print(); StreamOperator.execute(); } }
col1 | col2 | col3 |
---|---|---|
a | -0.0784 | 1.4596 |
b | -0.2592 | -0.4814 |
c | 1.2270 | -0.6521 |
d | -1.6687 | 1.4596 |
a | -0.2028 | -0.6521 |
b | -0.2549 | -0.4814 |
c | 1.2371 | -0.6521 |
col1 | col2 | col3 |
c | 1.2371 | -0.6521 |
b | -0.2592 | -0.4814 |
c | 1.2270 | -0.6521 |
b | -0.2549 | -0.4814 |
a | -0.0784 | 1.4596 |
a | -0.2028 | -0.6521 |
d | -1.6687 | 1.4596 |