11#include <framework/core/Module.h>
12#include <analysis/dbobjects/PIDNeuralNetworkParameters.h>
Module that creates PID neural network parameters and uploads them to the DB.
PIDNNInputsToCut m_inputsToCut
overwrite certain input variables
void initialize() override
Initialize the module.
std::vector< float > m_standardDeviations
standard deviations of inputs
PIDNeuralNetworkParametersCreatorModule()
Constructor: Sets the description, the properties and the parameters of the module.
PIDNNMissingInputs m_handleMissingInputs
handling of NaN inputs
int m_runLow
Lowest valid run #.
std::vector< std::string > m_inputNames
list of input names
std::string m_neuralNetworkParametersName
Name of the set of parameters.
std::string m_modelDefinition
neural network string for frugally-deep
std::vector< float > m_meanValues
mean values of inputs
std::vector< int > m_outputSpeciesPdg
PDG codes of hypotheses of neural-network output.
int m_experimentLow
Lowest valid experiment #.
std::string m_description
description of neural network parameters
int m_experimentHigh
Highest valid experiment #.
int m_runHigh
Highest valid run #.
std::vector< std::tuple< size_t, size_t, double, double, double > > PIDNNInputsToCut
Stores information on whether and how to overwrite certain inputs.
std::vector< std::tuple< size_t, float > > PIDNNMissingInputs
Stores information on how to handle missing inputs, i.e.
Abstract base class for different kinds of events.