search

cog-imperial / OMLT

Represent trained machine learning models as Pyomo optimization formulations
Other
262 stars 57 forks source link

Add direct interfaces to LGBM, CatBoost, etc. to allow for categorical vars #37

Closed ThebTron closed 2 years ago

ThebTron commented 2 years ago

Currently ONNX doesn't support categorical vars which limits the use of tree ensembles in OMLT. By directly interfacing with these libraries, we can add support for categorical vars.

ThebTron commented 2 years ago

Some of the functions needed to create the optimization formulation of the tree ensemble with categorical vars are given in the following public branch: https://github.com/cog-imperial/OMLT/tree/cat_var_support