Java 类名：com.alibaba.alink.operator.batch.source.LibSvmSourceBatchOp
Python 类名：LibSvmSourceBatchOp

功能介绍

读LibSVM文件。支持从本地、hdfs、oss、http等读取。

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
filePath 文件路径 文件路径 String ✓
startIndex 起始索引 起始索引 Integer 1

代码示例

Python 代码

  1. from pyalink.alink import *
  3. import pandas as pd
  5. useLocalEnv(1)
  7. df_data = pd.DataFrame([
  8.     ['1:2.0 2:1.0 4:0.5', 1.5],
  9.     ['1:2.0 2:1.0 4:0.5', 1.7],
  10.     ['1:2.0 2:1.0 4:0.5', 3.6]
  11. ])
  13. batch_data = BatchOperator.fromDataframe(df_data, schemaStr='f1 string, f2  double')
  15. filepath = '/tmp/abc.svm'
  17. sink = LibSvmSinkBatchOp().setFilePath(filepath).setLabelCol("f2").setVectorCol("f1").setOverwriteSink(True)
  18. batch_data = batch_data.link(sink)
  20. BatchOperator.execute()
  22. batch_data = LibSvmSourceBatchOp().setFilePath(filepath)
  23. batch_data.print()

Java 代码

  1. import org.apache.flink.types.Row;
  3. import com.alibaba.alink.operator.batch.BatchOperator;
  4. import com.alibaba.alink.operator.batch.sink.LibSvmSinkBatchOp;
  5. import com.alibaba.alink.operator.batch.source.LibSvmSourceBatchOp;
  6. import com.alibaba.alink.operator.batch.source.MemSourceBatchOp;
  7. import org.junit.Test;
  9. import java.util.Arrays;
  10. import java.util.List;
  12. public class LibSvmSourceBatchOpTest {
  13.     @Test
  14.     public void testLibSvmSourceBatchOp() throws Exception {
  15.         List <Row> df_data = Arrays.asList(
  16.             Row.of("1:2.0 2:1.0 4:0.5", 1.5),
  17.             Row.of("1:2.0 2:1.0 4:0.5", 1.7),
  18.             Row.of("1:2.0 2:1.0 4:0.5", 3.6)
  19.         );
  20.         BatchOperator <?> batch_data = new MemSourceBatchOp(df_data, "f1 string, f2  double");
  21.         String filepath = "/tmp/abc.svm";
  22.         BatchOperator <?> sink = new LibSvmSinkBatchOp().setFilePath(filepath).setLabelCol("f2").setVectorCol("f1")
  23.             .setOverwriteSink(true);
  24.         batch_data = batch_data.link(sink);
  25.         BatchOperator.execute();
  26.         batch_data = new LibSvmSourceBatchOp().setFilePath(filepath);
  27.         batch_data.print();
  28.     }
  29. }

运行结果

| label | features | | —- | —- |

| 1.7000 | 1:2.0 2:1.0 4:0.5 |

| 1.5000 | 1:2.0 2:1.0 4:0.5 |

| 3.6000 | 1:2.0 2:1.0 4:0.5 |