Applies classification from a Gradient Boosted Trees model that is provided in PMML format. Note that it is currently not possible to load models that were learned on a bit-, byte- or double-vector column and then written to PMML because PMML does not support vector columns. The implementation follows the algorithms described in "Greedy Function Approximation: A Gradient Boosting Machine" by Jerome H. Friedman (1999)". For more information you can also take a look at this .