sasctl.utils.pyml2ds package#

Submodules#

Translate a gradient boosting model and write SAS scoring code to file.

Supported models are: xgboost, lightgbm and pmml gradient boosting.

param in_file:: Pickled object to translate. String is assumed to be a path to a picked file, file-like is assumed to be an open file handle to a pickle object, and bytes is assumed to be the raw pickled bytes.
type in_file:: str or bytes or file-like
param out_var_name:: Output variable name.
type out_var_name:: str (optional)
returns:: str – A SAS Data Step program implementing the model.

Examples

Generate SAS code from an XGBoost model.

>>> from xgboost.sklearn import XGBRegressor
>>> xgb = XGBRegressor()
>>> xgb.fit(X, y)
>>> pkl = pickle.dumps(xgb)
>>> sas_code = pyml2ds(pkl)

Warning

This method is experimental and may be modified or removed without warning.