File indexing completed on 2024-12-22 04:18:11

 /***************************************************************************
  *                                                                         *
  *   copyright : (C) 2007 The University of Toronto                        *
  *                   netterfield@astro.utoronto.ca                         *
  *   copyright : (C) 2005  University of British Columbia                  *
  *                   dscott@phas.ubc.ca                                    *
  *                                                                         *
  *   This program is free software; you can redistribute it and/or modify  *
  *   it under the terms of the GNU General Public License as published by  *
  *   the Free Software Foundation; either version 2 of the License, or     *
  *   (at your option) any later version.                                   *
  *                                                                         *
  ***************************************************************************/
 
 
 #include "fitgaussian_unweighted.h"
 #include "objectstore.h"
 #include "ui_fitgaussian_unweightedconfig.h"
 
 #define NUM_PARAMS 4
 #define MAX_NUM_ITERATIONS 500
 
 #include <gsl/gsl_fit.h>
 #include "../non_linear.h"
 
 static const QString& VECTOR_IN_X = "X Vector";
 static const QString& VECTOR_IN_Y = "Y Vector";
 static const QString& SCALAR_IN_OFFSET = "Offset";
 
 static const QString& VECTOR_OUT_Y_FITTED = "Fit";
 static const QString& VECTOR_OUT_Y_RESIDUALS = "Residuals";
 static const QString& VECTOR_OUT_Y_PARAMETERS = "Parameters Vector";
 static const QString& VECTOR_OUT_Y_COVARIANCE = "Covariance";
 static const QString& SCALAR_OUT = "chi^2/nu";
 
 class ConfigWidgetFitGaussianUnweightedPlugin : public Kst::DataObjectConfigWidget, public Ui_FitGaussian_UnweightedConfig {
   public:
     ConfigWidgetFitGaussianUnweightedPlugin(QSettings* cfg) : DataObjectConfigWidget(cfg), Ui_FitGaussian_UnweightedConfig() {
       _store = 0;
       setupUi(this);
     }
 
     ~ConfigWidgetFitGaussianUnweightedPlugin() {}
 
     void setObjectStore(Kst::ObjectStore* store) { 
       _store = store; 
       _vectorX->setObjectStore(store);
       _vectorY->setObjectStore(store);
       _scalarOffset->setObjectStore(store);
       _forceOffset->setChecked(false);
       _scalarOffset->setEnabled(false);
     }
 
     void setupSlots(QWidget* dialog) {
       if (dialog) {
         connect(_vectorX, SIGNAL(selectionChanged(QString)), dialog, SIGNAL(modified()));
         connect(_vectorY, SIGNAL(selectionChanged(QString)), dialog, SIGNAL(modified()));
         connect(_scalarOffset, SIGNAL(selectionChanged(QString)), dialog, SIGNAL(modified()));
         _scalarOffset->setDefaultValue(0.0);
         connect(_forceOffset, SIGNAL(toggled(bool)), dialog, SIGNAL(modified()));
         connect(_forceOffset, SIGNAL(toggled(bool)), _scalarOffset, SLOT(setEnabled(bool)));
       }
     }
 
 
     void setVectorX(Kst::VectorPtr vector) {
       setSelectedVectorX(vector);
     }
 
     void setVectorY(Kst::VectorPtr vector) {
       setSelectedVectorY(vector);
     }
 
     void setVectorsLocked(bool locked = true) {
       _vectorX->setEnabled(!locked);
       _vectorY->setEnabled(!locked);
     }
 
     Kst::VectorPtr selectedVectorX() { return _vectorX->selectedVector(); };
     void setSelectedVectorX(Kst::VectorPtr vector) { return _vectorX->setSelectedVector(vector); };
 
     Kst::ScalarPtr scalarOffset() {return _scalarOffset->selectedScalar(); };
     void setScalarOffset(Kst::ScalarPtr scalar) {_scalarOffset->setSelectedScalar(scalar);};
 
     Kst::VectorPtr selectedVectorY() { return _vectorY->selectedVector(); };
     void setSelectedVectorY(Kst::VectorPtr vector) { return _vectorY->setSelectedVector(vector); };
 
     virtual void setupFromObject(Kst::Object* dataObject) {
       if (FitGaussianUnweightedSource* source = static_cast<FitGaussianUnweightedSource*>(dataObject)) {
         setSelectedVectorX(source->vectorX());
         setSelectedVectorY(source->vectorY());
         _forceOffset->setChecked(source->_forceOffset);
         setScalarOffset(source->scalarOffset());
       }
     }
 
     virtual bool configurePropertiesFromXml(Kst::ObjectStore *store, QXmlStreamAttributes& attrs) {
       bool validTag = true;
 
       setObjectStore(store);
 
       QStringRef av;
       bool force_offset = false;
       av = attrs.value("ForceOffset");
       if (!av.isNull()) {
         force_offset = QVariant(av.toString()).toBool();
       }
       _forceOffset->setChecked(force_offset);
 //      if (force_offset) {
 //   The tickets will also be taken care of.     av = attrs.value("Offset");
 //        if (!av.isNull()) {
 //          QString name = av.toString();
 //          Kst::ObjectPtr object = store->retrieveObject(name);
 //          Kst::ScalarPtr scalar = Kst::kst_cast<Kst::Scalar>(object);
 //          setScalarOffset(scalar);
 //        }
 
 //      }
 
       return validTag;
     }
 
   public slots:
     virtual void save() {
       if (_cfg) {
         _cfg->beginGroup("Fit Gaussian Plugin");
         _cfg->setValue("Input Vector X", _vectorX->selectedVector()->Name());
         _cfg->setValue("Input Vector Y", _vectorY->selectedVector()->Name());
         _cfg->setValue("Force Offset", _forceOffset->isChecked());
         if (_forceOffset->isChecked()) {
           _cfg->setValue("Offset", _scalarOffset->selectedScalar()->Name());
         }
         _cfg->endGroup();
       }
     }
 
     virtual void load() {
       if (_cfg && _store) {
         _cfg->beginGroup("Fit Gaussian Plugin");
         QString vectorName = _cfg->value("Input Vector X").toString();
         Kst::Object* object = _store->retrieveObject(vectorName);
         Kst::Vector* vectorx = static_cast<Kst::Vector*>(object);
         if (vectorx) {
           setSelectedVectorX(vectorx);
         }
         vectorName = _cfg->value("Input Vector Y").toString();
         object = _store->retrieveObject(vectorName);
         Kst::Vector* vectory = static_cast<Kst::Vector*>(object);
         if (vectory) {
           setSelectedVectorX(vectory);
         }
         bool force_offset = _cfg->value("Force Offset").toBool();
         _forceOffset->setChecked(force_offset);
         if (force_offset) {
           QString scalarName = _cfg->value("Offset").toString();
           object = _store->retrieveObject(scalarName);
           Kst::Scalar* scalar = static_cast<Kst::Scalar*>(object);
           if (scalar) {
             setScalarOffset(scalar);
           }
         }
         _cfg->endGroup();
       }
     }
 
   private:
     Kst::ObjectStore *_store;
 
 };
 
 
 FitGaussianUnweightedSource::FitGaussianUnweightedSource(Kst::ObjectStore *store)
 : Kst::BasicPlugin(store) {
   _forceOffset = false;
 }
 
 
 FitGaussianUnweightedSource::~FitGaussianUnweightedSource() {
 }
 
 
 QString FitGaussianUnweightedSource::_automaticDescriptiveName() const {
     return tr("%1 Unweighted Gaussian").arg(vectorY()->descriptiveName());
 }
 
 
 void FitGaussianUnweightedSource::change(Kst::DataObjectConfigWidget *configWidget) {
   if (ConfigWidgetFitGaussianUnweightedPlugin* config = static_cast<ConfigWidgetFitGaussianUnweightedPlugin*>(configWidget)) {
     setInputVector(VECTOR_IN_X, config->selectedVectorX());
     setInputVector(VECTOR_IN_Y, config->selectedVectorY());
     setInputScalar(SCALAR_IN_OFFSET, config->scalarOffset());
     _forceOffset = config->_forceOffset->isChecked();
 //    _offset = config->_scalarOffset->selectedScalar();
   }
 }
 
 
 void FitGaussianUnweightedSource::setupOutputs() {
   setOutputVector(VECTOR_OUT_Y_FITTED, "");
   setOutputVector(VECTOR_OUT_Y_RESIDUALS, "");
   setOutputVector(VECTOR_OUT_Y_PARAMETERS, "");
   setOutputVector(VECTOR_OUT_Y_COVARIANCE, "");
   setOutputScalar(SCALAR_OUT, "");
 
   int i=0;
   for (QString paramName = parameterName(i); !paramName.isEmpty(); paramName = parameterName(++i)) {
     setOutputScalar(paramName, "");
   }
 }
 
 
 void function_initial_estimate( const double X[], const double Y[], int npts, double P[] ) {
   double min_y = 1E300;
   double max_y = -1E300;
   double min_x = 1E300;
   double max_x = -1E300;
   double mean_y = 0.0;
   double x_at_min_y=0;
   double x_at_max_y=0;
 
   double A, C, D;
 
   // find peak, valley, and mean
   for (int i = 0; i<npts; i++) {
     if (min_y > Y[i]) {
       min_y = Y[i];
       x_at_min_y = X[i];
     }
     if (max_y < Y[i]) {
       max_y = Y[i];
       x_at_max_y = X[i];
     }
     mean_y += Y[i];
 
     if (min_x > X[i]) {
       min_x = X[i];
     }
     if (max_x < X[i]) {
       max_x = X[i];
     }
   }
   if (npts>0) {
     mean_y /= double(npts);
   }
 
   // Heuristic for finding the sign of the : less time is spent in the peak than
   // in background if the range covers more than ~+- 2 sigma.
   // It will fail if you are
   // really zoomed into the gaussian.  Not sure what happens then :-(
   if (max_y - mean_y > mean_y - min_y) { // positive going gaussian
     A = max_y - min_y;
     D = min_y;
     C = x_at_max_y;
   } else { // negative going gaussian
     A = min_y - mean_y;
     D = max_y;
     C = x_at_min_y;
   }
   // guess that the width of the gaussian is around 1/10 of the x range (?)
 
   P[0] = A;
   P[1] = (max_x - min_x)*0.1;
   P[2] = C;
   P[3] = D;
 
 }
 
 
 double function_calculate(double x, double P[]) {
   double A = P[0];
   double B = 0.5/(P[1]*P[1]);
   double C = P[2];
   double D = offset_;
 
   if (n_params == 4) {
     D = P[3];
   }
   x -= C;
 
   return A*exp(-B*x*x) + D;
 }
 
 void function_derivative( double x, double P[], double dFdP[] ) {
   double A = P[0];
   double s = P[1];
   double B = 0.5/(s*s);
   double C = P[2];
   //double D = P[3];
 
   x -= C;
 
   double E = exp(-B*x*x);
 
   dFdP[0] = E;
   dFdP[1] = A*x*x*E/(s*s*s);
   dFdP[2] = 2*A*B*x*E;
   dFdP[3] = 1.0;
 
 }
 
 bool FitGaussianUnweightedSource::algorithm() {
   Kst::VectorPtr inputVectorX = _inputVectors[VECTOR_IN_X];
   Kst::VectorPtr inputVectorY = _inputVectors[VECTOR_IN_Y];
   Kst::ScalarPtr inputScalarOffset = _inputScalars[SCALAR_IN_OFFSET];
 
   Kst::VectorPtr outputVectorYFitted = _outputVectors[VECTOR_OUT_Y_FITTED];
   Kst::VectorPtr outputVectorYResiduals = _outputVectors[VECTOR_OUT_Y_RESIDUALS];
   Kst::VectorPtr outputVectorYParameters = _outputVectors[VECTOR_OUT_Y_PARAMETERS];
   Kst::VectorPtr outputVectorYCovariance = _outputVectors[VECTOR_OUT_Y_COVARIANCE];
   Kst::ScalarPtr outputScalar = _outputScalars[SCALAR_OUT];
 
 
   if (_forceOffset) {
     if (inputScalarOffset) {
       offset_ = inputScalarOffset->value();
     } else {
       offset_ = 0;
     }
     n_params = 3;
   } else {
     n_params = 4;
   }
 
   Kst::LabelInfo label_info = inputVectorY->labelInfo();
   label_info.name = tr("A\\cdotexp((x-x_o)^2/2\\sigma^2) + C fit to %1").arg(label_info.name);
   outputVectorYFitted->setLabelInfo(label_info);
 
   label_info.name = tr("Gaussian Fit Residuals");
   outputVectorYResiduals->setLabelInfo(label_info);
 
   bool bReturn = false;
 
   bReturn = kstfit_nonlinear( inputVectorX, inputVectorY,
                               outputVectorYFitted, outputVectorYResiduals, outputVectorYParameters,
                               outputVectorYCovariance, outputScalar );
   return bReturn;
 }
 
 
 Kst::VectorPtr FitGaussianUnweightedSource::vectorX() const {
   return _inputVectors[VECTOR_IN_X];
 }
 
 
 Kst::VectorPtr FitGaussianUnweightedSource::vectorY() const {
   return _inputVectors[VECTOR_IN_Y];
 }
 
 
 Kst::ScalarPtr FitGaussianUnweightedSource::scalarOffset() const {
   return _inputScalars[SCALAR_IN_OFFSET];
 }
 
 
 QStringList FitGaussianUnweightedSource::inputVectorList() const {
   QStringList vectors(VECTOR_IN_X);
   vectors += VECTOR_IN_Y;
   return vectors;
 }
 
 
 QStringList FitGaussianUnweightedSource::inputScalarList() const {
   QStringList scalars(SCALAR_IN_OFFSET);
   return scalars;
 }
 
 
 QStringList FitGaussianUnweightedSource::inputStringList() const {
   return QStringList( /*STRING_IN*/ );
 }
 
 
 QStringList FitGaussianUnweightedSource::outputVectorList() const {
   QStringList vectors(VECTOR_OUT_Y_FITTED);
   vectors += VECTOR_OUT_Y_RESIDUALS;
   vectors += VECTOR_OUT_Y_PARAMETERS;
   vectors += VECTOR_OUT_Y_COVARIANCE;
   vectors += VECTOR_OUT_Y_PARAMETERS;
   return vectors;
 }
 
 
 QStringList FitGaussianUnweightedSource::outputScalarList() const {
   return QStringList( SCALAR_OUT );
 }
 
 
 QStringList FitGaussianUnweightedSource::outputStringList() const {
   return QStringList( /*STRING_OUT*/ );
 }
 
 
 void FitGaussianUnweightedSource::saveProperties(QXmlStreamWriter &s) {
   QString force_offset;
   force_offset.setNum(_forceOffset);
   s.writeAttribute("ForceOffset", force_offset);
 }
 
 
 QString FitGaussianUnweightedSource::parameterName(int index) const {
   QString parameter;
   switch (index) {
   case 0:
     parameter = "A";
     break;
   case 1:
     parameter = "\\sigma";
     break;
   case 2:
     parameter = "x_o";
     break;
   case 3:
     parameter = "C";
     break;
   default:
     parameter = "";
   }
 
   return parameter;
 }
 
 
 // Name used to identify the plugin.  Used when loading the plugin.
 QString FitGaussianUnweightedPlugin::pluginName() const { return tr("Gaussian Fit"); }
 QString FitGaussianUnweightedPlugin::pluginDescription() const { return tr("Generates a gaussian fit for a set of data."); }
 
 
 Kst::DataObject *FitGaussianUnweightedPlugin::create(Kst::ObjectStore *store, Kst::DataObjectConfigWidget *configWidget, bool setupInputsOutputs) const {
 
   if (ConfigWidgetFitGaussianUnweightedPlugin* config = static_cast<ConfigWidgetFitGaussianUnweightedPlugin*>(configWidget)) {
 
     FitGaussianUnweightedSource* object = store->createObject<FitGaussianUnweightedSource>();
 
     object->_forceOffset = config->_forceOffset->isChecked();
 
     if (setupInputsOutputs) {
       object->setupOutputs();
       object->setInputVector(VECTOR_IN_X, config->selectedVectorX());
       object->setInputVector(VECTOR_IN_Y, config->selectedVectorY());
       object->setInputScalar(SCALAR_IN_OFFSET, config->scalarOffset());
     }
 
     object->setPluginName(pluginName());
 
     object->writeLock();
     object->registerChange();
     object->unlock();
 
     return object;
   }
   return 0;
 }
 
 
 Kst::DataObjectConfigWidget *FitGaussianUnweightedPlugin::configWidget(QSettings *settingsObject) const {
   ConfigWidgetFitGaussianUnweightedPlugin *widget = new ConfigWidgetFitGaussianUnweightedPlugin(settingsObject);
   return widget;
 }
 
 #ifndef QT5
 Q_EXPORT_PLUGIN2(kstplugin_FitGaussianUnweightedPlugin, FitGaussianUnweightedPlugin)
 #endif
 
 // vim: ts=2 sw=2 et