Getting error while using the H2o AutoML model for h2o stacking

user_19196718 0
Viewed 44

I tried using the 2 best models from AutoML and used one of them as Meta learner for stacking. Named the new model stack_test. Code that I used is:

stack_test = H2OStackedEnsembleEstimator(base_models=[model1_xg, model2_xg],
metalearner_algorithm=model1_xg)

stack_test.train(x=x, y=y, training_frame=h2o_train)
stack_test.model_performance(h2o_test).auc()

Error I am getting:

NameError: name 'stack_test' is not defined

What am I doing wrong here?

h2o python
1 Answers

I believe the issue is in the metalearner_algorithm specification - it should be one of "AUTO", "glm", "gbm", "drf", "deeplearning", or "xgboost". See the documentation for more details.

user_135666780
Related