PunchThroughG4Classifier.cxx

Go to the documentation of this file.

/*
  Copyright (C) 2002-2026 CERN for the benefit of the ATLAS collaboration
*/

// PunchThroughG4Classifier.cxx, (c) ATLAS Detector software
#include "PunchThroughG4Classifier.h"
 
#include <fstream>
 
// PathResolver
#include "PathResolver/PathResolver.h"
 
// Geant4
#include "G4FastTrack.hh"
#include "G4FastStep.hh"
 
//LWTNN
#include "lwtnn/parse_json.hh"
 
#include "XMLCoreParser/XMLCoreParser.h"
#include "XMLCoreParser/XMLCoreNode.h"
 
 
PunchThroughG4Classifier::PunchThroughG4Classifier(const std::string& type, const std::string& name, const IInterface*  parent)
    : base_class(type, name, parent) {
}
 
StatusCode PunchThroughG4Classifier::initialize(){
 
    ATH_MSG_DEBUG( "[ punchthroughclassifier ] Initializing PunchThroughG4Classifier" );
 
    std::string resolvedScalerFileName = PathResolverFindCalibFile (m_scalerConfigFileName);
    ATH_CHECK ( initializeScaler(resolvedScalerFileName) );
 
    std::string resolvedNetworkFileName = PathResolverFindCalibFile (m_networkConfigFileName);
    ATH_CHECK ( initializeNetwork(resolvedNetworkFileName) );
 
    std::string resolvedCalibratorFileName = PathResolverFindCalibFile (m_calibratorConfigFileName);
    ATH_CHECK ( initializeCalibrator(resolvedCalibratorFileName) );
 
    return StatusCode::SUCCESS;
}
 
StatusCode PunchThroughG4Classifier::finalize(){
 
    ATH_MSG_DEBUG( "[punchthroughclassifier] finalize() successful" );
 
    return StatusCode::SUCCESS;
}
 
StatusCode PunchThroughG4Classifier::initializeScaler(const std::string & scalerConfigFile){
    XMLCoreParser p;
    std::unique_ptr<XMLCoreNode> doc = p.parse (scalerConfigFile);
 
    ATH_MSG_DEBUG( "[ punchthroughclassifier ] Loading scaler: " << scalerConfigFile);
 
    for (const XMLCoreNode* node : doc->get_children ("Transformations/*"))
    {
      if (node->get_name() == "ScalerValues") {
        m_scalerMin = node->get_double_attrib ("min");
        m_scalerMax = node->get_double_attrib ("max");
      }
      else if (node->get_name() == "VarScales") {
        std::string name = node->get_attrib ("name");
        m_scalerMinMap[name] = node->get_double_attrib ("min");
        m_scalerMaxMap[name] = node->get_double_attrib ("max");
      }
    }
 
    return StatusCode::SUCCESS;
}
 
StatusCode PunchThroughG4Classifier::initializeNetwork(const std::string & networkConfigFile){
 
    ATH_MSG_DEBUG( "[ punchthroughclassifier ] Loading classifier: " << networkConfigFile);
 
    std::ifstream input(networkConfigFile);
    if(!input){
        ATH_MSG_ERROR("Could not find json file " << networkConfigFile );
        return StatusCode::FAILURE;
    }
 
    m_graph  = std::make_unique<lwt::LightweightGraph>(lwt::parse_json_graph(input));
    if(!m_graph){
        ATH_MSG_ERROR("Could not parse graph json file " << networkConfigFile );
        return StatusCode::FAILURE;
    }
 
 
    return StatusCode::SUCCESS;
}
 
StatusCode PunchThroughG4Classifier::initializeCalibrator(const std::string & calibratorConfigFile){
    XMLCoreParser p;
    std::unique_ptr<XMLCoreNode> doc = p.parse (calibratorConfigFile);
 
    //parse xml that contains config for isotonic regressor used to calibrate the network output
    ATH_MSG_DEBUG( "[ punchthroughclassifier ] Loading calibrator: " << calibratorConfigFile);
 
    for (const XMLCoreNode* node : doc->get_children ("Transformations/*"))
    {
      if (node->get_name() == "LimitValues") {
        m_calibrationMin = node->get_double_attrib ("min");
        m_calibrationMax = node->get_double_attrib ("max");
      }
      else if (node->get_name() == "LinearNorm") {
        double orig = node->get_double_attrib ("orig");
        double norm = node->get_double_attrib ("norm");
        m_calibrationMap[orig] = norm;
      }
    }
 
    return StatusCode::SUCCESS;
}
 
double PunchThroughG4Classifier::computePunchThroughProbability(const G4FastTrack& fastTrack, const double simE, const std::vector<double> & simEfrac) const {
 
    std::map<std::string, std::map<std::string, double> > networkInputs = computeInputs(fastTrack, simE, simEfrac); //compute inputs
 
    networkInputs = scaleInputs(networkInputs); //scale inputs
 
    std::map<std::string, double> networkOutputs = m_graph->compute(networkInputs); //call neural network on inputs
 
    double calibratedOutput = calibrateOutput(networkOutputs["out_0"]); //calibrate neural network output
 
    return calibratedOutput;
}
 
std::map<std::string, std::map<std::string, double> > PunchThroughG4Classifier::computeInputs(const G4FastTrack& fastTrack, const double simE, const std::vector<double> & simEfrac) {
 
    //calculate inputs for NN
 
    std::map<std::string, std::map<std::string, double> > networkInputs;
 
    //add initial particle and total energy variables
    networkInputs["node_0"] = {
        {"variable_0", fastTrack.GetPrimaryTrack()->GetMomentum().mag() },
        {"variable_1", std::abs(fastTrack.GetPrimaryTrack()->GetPosition().eta()) },
        {"variable_2", fastTrack.GetPrimaryTrack()->GetPosition().phi() },
        {"variable_3", simE},
    };
 
    //add energy fraction variables
    for (unsigned int i = 0; i < simEfrac.size(); i++) { //from 0 to 23, 24 layers
        networkInputs["node_0"].insert({"variable_" + std::to_string(i + 4), simEfrac[i]});
    }
 
    return networkInputs;
}
 
std::map<std::string, std::map<std::string, double> > PunchThroughG4Classifier::scaleInputs(std::map<std::string, std::map<std::string, double> >& inputs) const{
 
    //apply MinMaxScaler to network inputs
 
    for (auto& var : inputs["node_0"]) {
 
        double x_std;
        if(m_scalerMaxMap.at(var.first) != m_scalerMinMap.at(var.first)){
            x_std = (var.second - m_scalerMinMap.at(var.first)) / (m_scalerMaxMap.at(var.first) - m_scalerMinMap.at(var.first));
        }
        else{
            x_std = (var.second - m_scalerMinMap.at(var.first));
        }
        var.second = x_std * (m_scalerMax - m_scalerMin) + m_scalerMin;
    }
 
    return inputs;
}
 
double PunchThroughG4Classifier::calibrateOutput(double& networkOutput) const {
 
    //calibrate output of network using isotonic regressor model
 
    //if network output is outside of the range of isotonic regressor then return min and max values
    if (networkOutput < m_calibrationMin){
        return m_calibrationMin;
    }
    else if (networkOutput > m_calibrationMax){
        return m_calibrationMax;
    }
 
    //otherwise find neighbouring points in isotonic regressor
    auto upper = m_calibrationMap.upper_bound(networkOutput);
    auto lower = upper--;
 
    //Perform linear interpolation between points
    double m = (upper->second - lower->second)/(upper->first - lower->first);
    double c = lower->second - m * lower->first;
    double calibrated = m * networkOutput + c;
 
    return calibrated;
}