Inference Code Generation for Recurrent Neural Networks

Toolkit for Multivariate Analysis (TMVA) is a multi-purpose machine learning toolkit integrated into the ROOT scientific software framework, used in many particle physics data analysis and applications. A fast inference system that takes takes ONNX model as input and produces compilation-ready standalone C++ scripts as output is under development in TMVA. This project will focus on the development of ONNX RNN, LSTM and GRU operators in the code generation format.