Tmva 1 0 3_ Application

This tutorial illustrates how you can conveniently apply BDTs in C++ using the fast tree inference engine offered by TMVA. Supported workflows are event-by-event inference, batch inference and pipelines with RDataFrame.

Author: Stefan Wunsch
This notebook tutorial was automatically generated with ROOTBOOK-izer from the macro found in the ROOT repository on Thursday, June 17, 2021 at 05:51 PM.

In [1]:
using namespace TMVA::Experimental;

Load bdt model remotely from a webserver

In [2]:
RBDT<> bdt("myBDT", "http://root.cern/files/tmva101.root");

Apply model on a single input

In [3]:
auto y1 = bdt.Compute({1.0, 2.0, 3.0, 4.0});

std::cout << "Apply model on a single input vector: " << y1[0] << std::endl;
Apply model on a single input vector: 0.0302787

Apply model on a batch of inputs

In [4]:
float data[8] = {1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0};
RTensor<float> x(data, {2, 4});
auto y2 = bdt.Compute(x);

std::cout << "Apply model on an input tensor: " << y2 << std::endl;
Apply model on an input tensor: { { 0.0302787 } { 0.19114 } }

Apply model as part of an rdataframe workflow

In [5]:
ROOT::RDataFrame df("Events", "root://eospublic.cern.ch//eos/root-eos/cms_opendata_2012_nanoaod/SMHiggsToZZTo4L.root");
auto df2 = df.Filter("nMuon >= 2")
             .Filter("nElectron >= 2")
             .Define("Muon_pt_1", "Muon_pt[0]")
             .Define("Muon_pt_2", "Muon_pt[1]")
             .Define("Electron_pt_1", "Electron_pt[0]")
             .Define("Electron_pt_2", "Electron_pt[1]")
             .Define("y",
                     Compute<4, float>(bdt),
                     {"Muon_pt_1", "Muon_pt_2", "Electron_pt_1", "Electron_pt_2"});

std::cout << "Mean response on the signal sample: " << *df2.Mean("y") << std::endl;
Mean response on the signal sample: 0.625916
Plugin No such file or directory loading sec.protocol libXrdSeckrb5-4.so