LauResultsExtractor Class Reference

Utility class to allow the extraction of the best fit from a series of fits to a given data sample. More...

#include <LauResultsExtractor.hh>

Public Member Functions
	LauResultsExtractor (const TString &inputFileName, const TString &outputFileName, const TString &treeName)
	Constructor. More...
virtual	~LauResultsExtractor ()
	Destructor.
void	process (const Int_t numExpts)
	Run the calculations. More...

Protected Member Functions
void	setupInputTree ()
	Create storage for leaves and call SetBranchAddress for each.
void	setupOutputTree (TTree *tree)
	Create branches in the output tree.
void	setInputTreeBranchStatus (const Bool_t status)
	Toggle branch status of input tree branches (except iExpt, fitStatus and NLL)
void	clearMaps ()
	Clear all information.
void	writeFile ()
	Write the output file.

Private Attributes
TString	inputFileName_
	Name of text file containing list of input files.
TString	outputFileName_
	Name of output ROOT file.
TString	treeName_
	Name of tree in input ROOT files.
TChain *	inputTree_
	Chain of inputs.
TFile *	outputFile_
	Output file.
TTree *	outputTree_
	Output tree.
Int_t	nEntries_
	Number of entries in the input chain.
Int_t	iExpt_
	Storage for experiment ID variable.
Int_t	fitStatus_
	Storage for fit status variable.
Double_t	NLL_
	Storage for NLL variable.
Double_t	EDM_
	Storage for EDM variable.
std::map< TString, Double_t >	otherVars_
	Storage for other input variables.
std::map< Int_t, std::pair< Double_t, Int_t > >	bestNLL_
	Best NLL and corresponding tree entries for each experiment.
std::map< Int_t, std::pair< Double_t, Int_t > >	worstNLL_
	Worst NLL and corresponding tree entries for each experiment.
std::map< Int_t, std::vector< Double_t > >	allNLLs_
	All NLL values for each experiment.
std::map< Int_t, TH1 * >	nllHistos_
	Histograms of the NLL values for each experiment.

Detailed Description

Utility class to allow the extraction of the best fit from a series of fits to a given data sample.

A utility class to allow the extraction of the best fit from a series of fits to a given data sample.

When fitting amplitude models, the likelihood parameter space is highly complex. Hence the fitter can often wander into local minima. To mitigate this effect a data sample can be fitted many times with randomised starting values of the isobar parameters. It is then necessary to determine which of these fits gives the best solution, i.e. the minimum NLL. This class performs this task, reading in a series of fits for each data sample and writing out a single file that contains the results of the best fit for each data sample.

Definition at line 51 of file LauResultsExtractor.hh.

Constructor & Destructor Documentation

LauResultsExtractor()

LauResultsExtractor::LauResultsExtractor	(	const TString &	inputFileName,
		const TString &	outputFileName,
		const TString &	treeName
	)

Constructor.

Parameters

[in]	inputFileName	name of text file containing the input ROOT files
[in]	outputFileName	name of the file to which the best fit info should be written
[in]	treeName	name of the tree to read from the input files

Definition at line 44 of file LauResultsExtractor.cc.

Member Function Documentation

process()

void LauResultsExtractor::process

(

const Int_t

numExpts

)

Run the calculations.

Parameters

[in]

numExpts

the number of experiments to process

Definition at line 153 of file LauResultsExtractor.cc.

---

The documentation for this class was generated from the following files:

• /home/epp/phsdba/Development/Laura++/master-dev/laura/inc/LauResultsExtractor.hh
• /home/epp/phsdba/Development/Laura++/master-dev/laura/src/LauResultsExtractor.cc