StackedEnsemble: Support saving & loading binary models

Description

We should add support for saving and loading binary Stacked Ensemble models.

Currently, you can save a model using the standard `h2o.saveModel(ensemble, path = "my_ensemble")` functionality, but there are some errors when you load it back in and try to make predictions (after stopping and restarting the H2O cluster).

This works:

But if you kill the cluster, load the ensemble and try to make a prediction on a frame, you get something like this:

Related (since we may just make a generic `h2o.saveModels()` function that works on Grid, AutoML and Stacked Ensembles):

Assignee

Navdeep

Fix versions

Reporter

Erin LeDell

Support ticket URL

None

Labels

None

Affected Spark version

None

Customer Request Type

None

Task progress

None

CustomerVisible

No

Components

Priority

Major
Configure