iamDecode / sklearn-pmml-model

A library to parse and convert PMML models into Scikit-learn estimators.
BSD 2-Clause "Simplified" License
76 stars 15 forks source link

Import PMML directly from compressed format #27

Open iamDecode opened 3 years ago

iamDecode commented 3 years ago

Random forest and gradient boosting PMML files can easily exceed 100s of mbs in size, which takes a while to load. As the format is XML-based, its well suited for compression, and can reduce the file size many-fold. It is worth exploring whether loading the file directly from gzip or zip format improves loading speeds. Fewer disk operations are required, however the decompression will take some more cpu cycles.