Following an example in the documentation here:
when the model is trained and saved via h2o.saveModel and then the h2o cluster is shutdown and initialized again, then loading the model throws this error 'java.lang.NullPointerException'. Tryied various ways to save, also other examples and the issue is still there.
What is the proper way to load rulefit models, when a new instance of h2o.init is run for future predictions? Loading and predicting works while still in the same session.