BOSS 7.1.2
BESIII Offline Software System
Loading...
Searching...
No Matches
Public Member Functions | List of all members
ElectronCorrection Class Reference

#include <ElectronCorrection.h>

Public Member Functions

 ElectronCorrection ()
 
 ElectronCorrection (TString constfilename, TString infilename, double removeLowest, double removeHighest)
 
virtual ~ElectronCorrection ()
 
void initializeParameters ()
 
void process (TFile *outfile)
 
double RunGainCorrection (double &dedx) const
 
double WireGainCorrection (int wireID, double &dedx) const
 
double StandardCorrection (int wireID, double costheta, double dedx) const
 
double HadronCorrection (double costheta, double dedx) const
 
void calculateMeans (double &mean, double &truncatedMean, double &truncatedMeanErr, double dedx[], int size) const
 
void calculateMeans (double &mean, double &truncatedMean, double &truncatedMeanErr, std::vector< double > dedx, int size) const
 

Detailed Description

Definition at line 19 of file ElectronCorrection.h.

Constructor & Destructor Documentation

◆ ElectronCorrection() [1/2]

ElectronCorrection::ElectronCorrection ( )

Definition at line 3 of file ElectronWidget/ElectronCorrection.cc.

3 :
4 m_constfilename("dEdxConstants.root"),
5 m_filename(""),
6 m_removeLowest(0.05),
7 m_removeHighest(0.25)
8{
9 initializeParameters();
10}

◆ ElectronCorrection() [2/2]

ElectronCorrection::ElectronCorrection ( TString constfilename,
TString infilename,
double removeLowest,
double removeHighest )

Definition at line 13 of file ElectronWidget/ElectronCorrection.cc.

13 :
14 m_constfilename(constfilename),
15 m_filename(infilename),
16 m_removeLowest(removeLowest),
17 m_removeHighest(removeHighest)
18{
19 initializeParameters();
20}

◆ ~ElectronCorrection()

virtual ElectronCorrection::~ElectronCorrection ( )
inlinevirtual

Definition at line 25 of file ElectronCorrection.h.

25{};

Member Function Documentation

◆ calculateMeans() [1/2]

void ElectronCorrection::calculateMeans ( double & mean,
double & truncatedMean,
double & truncatedMeanErr,
double dedx[],
int size ) const

Perform the truncation

Definition at line 168 of file ElectronWidget/ElectronCorrection.cc.

170{
171 // Calculate the truncated average by skipping the lowest & highest
172 // events in the array of dE/dx values
173 std::vector<double> sortedDedx;
174 for( int i = 0; i < size; ++i )
175 sortedDedx.push_back(dedx[i]);
176 std::sort(sortedDedx.begin(), sortedDedx.end());
177
178 double truncatedMeanTmp = 0.0;
179 double meanTmp = 0.0;
180 double sum_of_squares = 0.0;
181 int numValuesTrunc = 0;
182 const int numDedx = sortedDedx.size();
183 const int lowEdgeTrunc = int(numDedx * m_removeLowest);
184 const int highEdgeTrunc = int(numDedx * (1 - m_removeHighest));
185 for (int i = 0; i < numDedx; i++) {
186 std::cout << sortedDedx[i] << "+\t";
187 meanTmp += sortedDedx[i];
188 if (i >= lowEdgeTrunc and i < highEdgeTrunc) {
189 truncatedMeanTmp += sortedDedx[i];
190 sum_of_squares += sortedDedx[i] * sortedDedx[i];
191 numValuesTrunc++;
192 }
193 }
194 std::cout << std::endl;
195
196 if (numDedx != 0) {
197 meanTmp /= numDedx;
198 }
199 if (numValuesTrunc != 0) {
200 truncatedMeanTmp /= numValuesTrunc;
201 } else {
202 truncatedMeanTmp = meanTmp;
203 }
204
205 mean = meanTmp;
206 truncatedMean = truncatedMeanTmp;
207
208 if (numValuesTrunc > 1) {
209 truncatedMeanErr = sqrt(sum_of_squares / double(numValuesTrunc) - truncatedMeanTmp * truncatedMeanTmp) / double(numValuesTrunc - 1);
210 } else {
211 truncatedMeanErr = 0;
212 }
213}

Referenced by process(), and ElectronCalibration::TwoDCorrection().

◆ calculateMeans() [2/2]

void ElectronCorrection::calculateMeans ( double & mean,
double & truncatedMean,
double & truncatedMeanErr,
std::vector< double > dedx,
int size ) const

Definition at line 216 of file ElectronWidget/ElectronCorrection.cc.

218{
219 // Calculate the truncated average by skipping the lowest & highest
220 // events in the array of dE/dx values
221 std::vector<double> sortedDedx = dedx;
222 std::sort(sortedDedx.begin(), sortedDedx.end());
223
224 double truncatedMeanTmp = 0.0;
225 double meanTmp = 0.0;
226 double sum_of_squares = 0.0;
227 int numValuesTrunc = 0;
228 const int numDedx = sortedDedx.size();
229 const int lowEdgeTrunc = int(numDedx * m_removeLowest);
230 const int highEdgeTrunc = int(numDedx * (1 - m_removeHighest));
231 for (int i = 0; i < numDedx; i++) {
232 std::cout << sortedDedx[i] << "+\t";
233 meanTmp += sortedDedx[i];
234 if (i >= lowEdgeTrunc and i < highEdgeTrunc) {
235 truncatedMeanTmp += sortedDedx[i];
236 sum_of_squares += sortedDedx[i] * sortedDedx[i];
237 numValuesTrunc++;
238 }
239 }
240 std::cout << std::endl;
241
242 if (numDedx != 0) {
243 meanTmp /= numDedx;
244 }
245 if (numValuesTrunc != 0) {
246 truncatedMeanTmp /= numValuesTrunc;
247 } else {
248 truncatedMeanTmp = meanTmp;
249 }
250
251 mean = meanTmp;
252 truncatedMean = truncatedMeanTmp;
253
254 if (numValuesTrunc > 1) {
255 truncatedMeanErr = sqrt(sum_of_squares / double(numValuesTrunc) - truncatedMeanTmp * truncatedMeanTmp) / double(numValuesTrunc - 1);
256 } else {
257 truncatedMeanErr = 0;
258 }
259}

◆ HadronCorrection()

double ElectronCorrection::HadronCorrection ( double costheta,
double dedx ) const

Perform a hadron saturation correction. (Set the peak of the truncated mean for electrons to 1)

Definition at line 145 of file ElectronWidget/ElectronCorrection.cc.

146{
147
148 double newDedx = m_hadsat.D2I(costheta,m_hadsat.I2D(costheta,1.0)*dedx);
149 return newDedx;
150}
HadronSaturation::I2D
double I2D(double cosTheta, double I, double alpha, double gamma, double delta, double power, double ratio) const
Definition HadronSaturation.cc:133
HadronSaturation::D2I
double D2I(double cosTheta, double D, double alpha, double gamma, double delta, double power, double ratio) const
Definition HadronSaturation.cc:119
costheta
float costheta
Definition prepare_data.cxx:28

◆ initializeParameters()

void ElectronCorrection::initializeParameters ( )

Retrieve the calibration constants (placeholder for now)

Definition at line 96 of file ElectronWidget/ElectronCorrection.cc.

97{
98
99 std::cout << "ElectronCorrection: initializing calibration constants..." << std::endl;
100
101 // For now just initialize the parameters to an arbitrary values for
102 // debugging. Eventually, this should get the constants from the
103 // calibration database.
104 m_runGain = 1.0;
105 for (int i = 0; i < c_NCDCWires; ++i) {
106 m_wireGain[i] = 1.0;
107 m_valid[i] = 1.0;
108 }
109
110 // get the hadron saturation parameters
111 // if the parameters do not exist, use the values in the default constructor
112 double hadpar[5];
113 std::ifstream parfile("sat-pars.txt");
114 if( !parfile.fail() ){
115 for( int i = 0; i < 5; ++i ){
116 parfile >> hadpar[i];
117 }
118 parfile.close();
119 m_hadsat.setParameters(hadpar);
120 }
121}
c_NCDCWires
const int c_NCDCWires
Definition ElectronCorrection.h:17
HadronSaturation::setParameters
void setParameters(double par[])
Definition HadronSaturation.cc:37

Referenced by ElectronCorrection(), and ElectronCorrection().

◆ process()

void ElectronCorrection::process ( TFile * outfile)

Apply corrections to events in infile and write to outfile

Definition at line 23 of file ElectronWidget/ElectronCorrection.cc.

24{
25
26 TFile* dedxInput = new TFile(m_filename);
27 TTree* track = (TTree*)dedxInput->Get("track");
28
29 // --------------------------------------------------
30 // INITIALIZE CONTAINERS
31 // --------------------------------------------------
32
33 double dedxpub; // dE/dx without electron saturation correction
34 double dedx; // dE/dx truncated mean with electron saturation correction
35 double dedxerr; // dE/dx error with electron saturation correction
36 double mean; // dE/dx mean with electron saturation correction
37 double p; // track momentum
38 double bg; // track beta-gamma
39 double costh; // cosine of track polar angle
40 int nhits; // number of hits on this track
41
42 const int maxhits = 100;
43 int wire[maxhits]; // wire number of this hit
44 double dedxhit[maxhits]; // dE/dx for this hit
45
46 track->SetBranchAddress("mean", &mean);
47 track->SetBranchAddress("dedx", &dedxpub);
48 track->SetBranchAddress("dedxsat", &dedx);
49 track->SetBranchAddress("dedxerr", &dedxerr);
50 track->SetBranchAddress("pF", &p);
51 track->SetBranchAddress("costh", &costh);
52 track->SetBranchAddress("numGoodLayerHits", &nhits);
53 track->SetBranchAddress("wire", wire);
54 track->SetBranchAddress("dedxhit", dedxhit);
55
56 outFile->cd();
57 TTree* newtrack = track->CloneTree(0);
58
59 // --------------------------------------------------
60 // LOOP OVER EACH DEDX MEASUREMENT (TRACK LEVEL)
61 // --------------------------------------------------
62
63 for( unsigned int i = 0; i < track->GetEntries(); ++i ){
64 track->GetEvent(i);
65
66 if( nhits <= 0 ){
67 std::cout << "No good hits on this track...";
68 continue;
69 }
70
71
72 // --------------------------------------------------
73 // LOOP OVER EACH DEDX MEASUREMENT (HIT LEVEL)
74 // --------------------------------------------------
75
76 for (int j = 0; j < nhits; ++j) {
77 double newdedx = StandardCorrection(wire[j], costh, dedxhit[j]);
78 dedxhit[j] = newdedx;
79 } // end loop over hits
80
81 // recalculate means -> mean, dedx, and dedxerr will be replaced
82 std::cout << "MEANS = " << dedx << "\t";
83 calculateMeans(mean, dedx, dedxerr, dedxhit, nhits);
84 std::cout << dedx << std::endl << std::endl;
85
86 newtrack->Fill();
87 } // end loop over tracks
88
89 outFile->cd();
90 newtrack->AutoSave();
91 outFile->Close();
92 dedxInput->Close();
93}
ElectronCorrection::calculateMeans
void calculateMeans(double &mean, double &truncatedMean, double &truncatedMeanErr, double dedx[], int size) const
Definition ElectronWidget/ElectronCorrection.cc:168
ElectronCorrection::StandardCorrection
double StandardCorrection(int wireID, double costheta, double dedx) const
Definition ElectronWidget/ElectronCorrection.cc:153
nhits
int nhits
Definition prepare_data.cxx:29
bg
float bg
Definition prepare_data.cxx:28

Referenced by ElectronInterface::ApplyCorrections().

◆ RunGainCorrection()

double ElectronCorrection::RunGainCorrection ( double & dedx) const

Perform a run gain correction

Definition at line 124 of file ElectronWidget/ElectronCorrection.cc.

125{
126
127 if (m_runGain != 0) {
128 double newDedx = dedx / m_runGain;
129 return newDedx;
130 } else
131 return dedx;
132}

Referenced by StandardCorrection().

◆ StandardCorrection()

double ElectronCorrection::StandardCorrection ( int wireID,
double costheta,
double dedx ) const

Perform a standard set of corrections

Definition at line 153 of file ElectronWidget/ElectronCorrection.cc.

154{
155
156 double temp = dedx;
157
158 temp = RunGainCorrection(temp);
159
160 temp = WireGainCorrection(wireID, temp);
161
162 // temp = HadronCorrection(costheta, temp);
163
164 return temp;
165}
ElectronCorrection::RunGainCorrection
double RunGainCorrection(double &dedx) const
Definition ElectronWidget/ElectronCorrection.cc:124
ElectronCorrection::WireGainCorrection
double WireGainCorrection(int wireID, double &dedx) const
Definition ElectronWidget/ElectronCorrection.cc:134

Referenced by process().

◆ WireGainCorrection()

double ElectronCorrection::WireGainCorrection ( int wireID,
double & dedx ) const

Perform a wire gain correction

Definition at line 134 of file ElectronWidget/ElectronCorrection.cc.

135{
136
137 if (m_valid[wireID] && m_wireGain[wireID] != 0) {
138 double newDedx = dedx / m_wireGain[wireID];
139 return newDedx;
140 } else
141 return dedx;
142}

Referenced by StandardCorrection().

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