Muon catboost
A new approach to catboost model evaluation. It doesn't requires catboost and flatbuffers packages, but it utilizes library for reading model from json format: https://gitlab.cern.ch/lhcb-parallelization/cuda_hlt/blob/muon_features_gpu/main/include/json.hpp
The current pipeline is:
- reading muon common hits from binaries (master)
- reading catboost model from json
- creating mock track
- calculating catboost features on gpu
- evaluating catboost model on gpu
Closes #17 (closed)
Edited by Dorothea Vom Bruch