Search code examples
h2oensemble-learning

NullPointerException error while building h2oensemble model

I am trying to build an ensemble of 3 models viz.....Deeplearning, RandomForest and Gradient Boosting. I have passed the models ids as a list to the ensemble function, but i get the following error:

java.lang.NullPointerException

 java.lang.NullPointerException
     at hex.StackedEnsembleModel.checkAndInheritModelProperties(StackedEnsembleModel.java:258)
     at hex.ensemble.StackedEnsemble$StackedEnsembleDriver.computeImpl(StackedEnsemble.java:116)
     at hex.ModelBuilder$Driver.compute2(ModelBuilder.java:169)
     at water.H2O$H2OCountedCompleter.compute(H2O.java:1241)
     at jsr166y.CountedCompleter.exec(CountedCompleter.java:468)
     at jsr166y.ForkJoinTask.doExec(ForkJoinTask.java:263)
     at jsr166y.ForkJoinPool$WorkQueue.runTask(ForkJoinPool.java:974)
     at jsr166y.ForkJoinPool.runWorker(ForkJoinPool.java:1477)
     at jsr166y.ForkJoinWorkerThread.run(ForkJoinWorkerThread.java:104)

 Error: java.lang.NullPointerException

Here is my argument to the ensemble model:

my_ensemble <- h2o.stackedEnsemble(x=2:length(names(train)),y=1,
              training_frame = train,validation_frame = valid,
              base_models = list(ann1@model_id,rf1@model_id,
               gbm1@model_id),model_id = "my_ensemble_1")

Kindly advise as to where i have went wrong.

Note: I am trying to predict on a multinomial classification.

Solution

  • H2O Stacked Ensemble does not yet support multinomial classification -- only regression and binary classification. This is noted in the Stacked Ensemble documentation. That's why it's failing.