A C++ Implementation of DeltaBM

Features:

• Boost random forests in a parallel way using OpenMP.
• Support different base learners in an automated manner, including trees, linear regressors, DPC_stairs, Kmeans2d and Splines.
• Directly use deltas, instead of gradients, to do the boosting.

Build:

• cmake CMakeLists.txt
• make -j4

APIs:

• Python & R

Remark:

• Details refer to the document.
• Set MKLROOT to the root of Intel MKL in the environment manually before building.