Java 类名：com.alibaba.alink.operator.batch.regression.KerasSequentialRegressorPredictBatchOp
Python 类名：KerasSequentialRegressorPredictBatchOp

功能介绍

与 KerasSequential 回归训练组件对应的预测组件。

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
predictionCol 预测结果列名 预测结果列名 String ✓
inferBatchSize 推理数据批大小 推理数据批大小 Integer 256
reservedCols 算法保留列名 算法保留列 String[] null

代码示例

以下代码仅用于示意，可能需要修改部分代码或者配置环境后才能正常运行！

Python 代码

  1. source = CsvSourceBatchOp() \
  2.     .setFilePath("https://alink-release.oss-cn-beijing.aliyuncs.com/data-files/random_tensor.csv") \
  3.     .setSchemaStr("tensor string, label double")
  5. source = ToTensorBatchOp() \
  6.     .setSelectedCol("tensor") \
  7.     .setTensorDataType("DOUBLE") \
  8.     .setTensorShape([200, 3]) \
  9.     .linkFrom(source)
  11. trainBatchOp = KerasSequentialRegressorTrainBatchOp() \
  12.     .setTensorCol("tensor") \
  13.     .setLabelCol("label") \
  14.     .setLayers([
  15.         "Conv1D(256, 5, padding='same', activation='relu')",
  16.         "Conv1D(128, 5, padding='same', activation='relu')",
  17.         "Dropout(0.1)",
  18.         "MaxPooling1D(pool_size=8)",
  19.         "Conv1D(128, 5, padding='same', activation='relu')",
  20.         "Conv1D(128, 5, padding='same', activation='relu')",
  21.         "Flatten()"
  22.     ]) \
  23.     .setOptimizer("Adam()") \
  24.     .setNumEpochs(1) \
  25.     .linkFrom(source)
  27. predictBatchOp = KerasSequentialRegressorPredictBatchOp() \
  28.     .setPredictionCol("pred") \
  29.     .setReservedCols(["label"]) \
  30.     .linkFrom(trainBatchOp, source)
  31. predictBatchOp.lazyPrint(10)
  32. BatchOperator.execute()

Java 代码

  1. import com.alibaba.alink.operator.batch.BatchOperator;
  2. import com.alibaba.alink.operator.batch.dataproc.ToTensorBatchOp;
  3. import com.alibaba.alink.operator.batch.regression.KerasSequentialRegressorPredictBatchOp;
  4. import com.alibaba.alink.operator.batch.regression.KerasSequentialRegressorTrainBatchOp;
  5. import com.alibaba.alink.operator.batch.source.CsvSourceBatchOp;
  6. import org.junit.Test;
  8. public class KerasSequentialRegressorTrainBatchOpTest {
  10.     @Test
  11.     public void testKerasSequentialRegressorTrainBatchOp() throws Exception {
  12.         BatchOperator<?> source = new CsvSourceBatchOp()
  13.             .setFilePath("https://alink-release.oss-cn-beijing.aliyuncs.com/data-files/random_tensor.csv")
  14.             .setSchemaStr("tensor string, label double");
  16.         source = new ToTensorBatchOp()
  17.             .setSelectedCol("tensor")
  18.             .setTensorDataType("DOUBLE")
  19.             .setTensorShape(200, 3)
  20.             .linkFrom(source);
  22.         KerasSequentialRegressorTrainBatchOp trainBatchOp = new KerasSequentialRegressorTrainBatchOp()
  23.             .setTensorCol("tensor")
  24.             .setLabelCol("label")
  25.             .setLayers(new String[] {
  26.                 "Conv1D(256, 5, padding='same', activation='relu')",
  27.                 "Conv1D(128, 5, padding='same', activation='relu')",
  28.                 "Dropout(0.1)",
  29.                 "MaxPooling1D(pool_size=8)",
  30.                 "Conv1D(128, 5, padding='same', activation='relu')",
  31.                 "Conv1D(128, 5, padding='same', activation='relu')",
  32.                 "Flatten()"
  33.             })
  34.             .setOptimizer("Adam()")
  35.             .setNumEpochs(1)
  36.             .linkFrom(source);
  38.         KerasSequentialRegressorPredictBatchOp predictBatchOp = new KerasSequentialRegressorPredictBatchOp()
  39.             .setPredictionCol("pred")
  40.             .setReservedCols("label")
  41.             .linkFrom(trainBatchOp, source);
  42.         predictBatchOp.lazyPrint(10);
  43.         BatchOperator.execute();
  44.     }
  45. }

运行结果

label pred
1.0000 0.4822
0.0000 0.4826
0.0000 0.4752
0.0000 0.4702
1.0000 0.4907
1.0000 0.4992
0.0000 0.4866
1.0000 0.5045
0.0000 0.4994
1.0000 0.4837