I have a project where I train on big data, but I predict on one sample. I want to train the model using PySpark and MLlib. Then transfer the trained model to Python out of