如何在 Apache Spark Pipeline 中打印最佳模型参数？

Question

我正在使用 Apache Spark 的管道 API 来验证参数。我正在像这样构建 TrainValidationSplitModel :

Pipeline pipeline = ...
ParamMap[] paramGrid = ...

TrainValidationSplit trainValidationSplit = new TrainValidationSplit().setEstimator(pipeline).setEvaluator(new MulticlassClassificationEvaluator()).setEstimatorParamMaps(paramGrid).setTrainRatio(0.8);
TrainValidationSplitModel model = trainValidationSplit.fit(training);

我的问题是：如何提取和打印最佳训练模型的参数？

Answer 1

我终于做到了。 Spark 在训练后打印这些指标。我有 spark 的错误日志级别，所以我没有看到这个：

2015-10-21 12:57:33,828 [INFO  org.apache.spark.ml.tuning.TrainValidationSplit]
Train validation split metrics: WrappedArray(0.7141940371838821, 0.7358721053749735)

2015-10-21 12:57:33,831 [INFO  org.apache.spark.ml.tuning.TrainValidationSplit]
Best set of parameters:
{
    hashingTF_79cf758f5ab1-numFeatures: 2000000,
    nb_67d55ce4e1fc-smoothing: 1.0
}

2015-10-21 12:57:33,831 [INFO  org.apache.spark.ml.tuning.TrainValidationSplit]
Best train validation split metric: 0.7358721053749735.

现在我在 log4j.properties 文件中添加了 class TrainValidationSplit 的关卡信息：

log4j.logger.org.apache.spark.ml.tuning.TrainValidationSplit=INFO
log4j.additivity.org.apache.spark.ml.tuning.TrainValidationSplit=false

如何在 Apache Spark Pipeline 中打印最佳模型参数？

How to print best model params in Apache Spark Pipeline?

java

machine-learning

apache-spark

apache-spark-mllib