H2o MOJO model load issue in pyspark

Question

I am getting below error while trying to load MOJO using pyspark.

code: settings = H2OMOJOSettings(convertUnknownCategoricalLevelsToNa = True, convertInvalidNumbersToNa = True) model = H2OMOJOModel.createFromMojo(in_dir + '/sparkle_comm.zip', settings)

error: py4j.protocol.Py4JJavaError: An error occurred while calling z:py_sparkling.ml.models.H2OMOJOModel.createFromMojo. : java.io.IOException: MOJO version incompatibility - the model MOJO version (1.10) is higher than the current h2o version (1.00) supports. Please, use the older version of h2o to load MOJO model.

spark version : version 2.4.4 Python version : Python 3.5.2

using h2o_pysparkling_2.4-3.26.2-2.4.zip

Thanks

Answer 1

It looks like the MOJO was trained with a newer version of sparkling water and you're trying to load it into an older version which is incompatible. Loading MOJOs is version dependant. Double check what version of H2O you used to train the MOJO.