Save trained SparkML model to storage. Load model then transform new dataset.

SparkML_ModelSaver.py

########################################
## Title: Spark MLlib Model Saver
## Language: PySpark
## Author: Colby T. Ford, Ph.D.
########################################

## Write Model to Blob
lrcvModel.save("/mnt/trainedmodels/lr")
rfcvModel.save("/mnt/trainedmodels/rf")
dtcvModel.save("/mnt/trainedmodels/dt")
display(dbutils.fs.ls("/mnt/trainedmodels/"))

## Remove an Old Model Directory
dbutils.fs.rm("/mnt/trainedmodels/dt", True)

## Load Trained Model and Transform Dataset
# Score the data using the model
from pyspark.ml.tuning import CrossValidatorModel
lrcvModel = CrossValidatorModel.load("/mnt/trainedmodels/lr/")
output = lrcvModel.bestModel.transform(dataset)