Speaker
Lucio Anderlini
(Universita e INFN, Firenze (IT))
Description
Machine Learning plays a major role in computational phyics providing a mechanism to approximate arbitrarily complex functions with, for example, Artificial Neural Networks and Boosted Decision Trees.
Unfortunately, the integration of machine learning models trained with python frameworks in production code, often developed in C, C++ or FORTRAN, is notoriously a complicated task.
In this contribution we will present scikinC, a transpiler for scikit-learn and keras models into plain C, intended to be compiled into shared objects and dynamically linked to other applications.
Applications to the parametrization of the LHCb detector will be also presented.
Authors
Lucio Anderlini
(Universita e INFN, Firenze (IT))
Matteo Barbetti
(Universita e INFN, Firenze (IT))
