added bounds and initial guess to spec_plot ui, modified fit_funcs, implemented double gaussian

Author: LAKESIDE\LindaT18

PythonGUI_apps/Spectrum_analysis/Spectra_fit_funcs.py

@@ -56,13 +56,14 @@ def gaussian_model(self):
         result = gmodel.fit(y, pars, x=x, nan_policy='propagate')
         return result
 
-    def gaussian_model_w_lims(self, center_min=None, center_max=None):
+    def gaussian_model_w_lims(self, center_initial_guess=None, sigma_initial_guess=None, center_min=None, center_max=None):
         x,y = self.background_correction()
         gmodel = GaussianModel(prefix = 'g1_') # calling gaussian model
         pars = gmodel.guess(y, x=x) # parameters - center, width, height
-        pars['g1_center'].set(min=center_min, max=center_max)
+        pars['g1_center'].set(center_initial_guess, min=center_min, max=center_max)
+        pars['g1_sigma'].set(sigma_initial_guess)
         result = gmodel.fit(y, pars, x=x, nan_policy='propagate')
-        return result
+        return result #770 760 780   sigma 15 
 
 class Single_Lorentzian(Spectra_Fit):
     """Fit a single Lorentzian to the spectrum
@@ -79,11 +80,20 @@ def lorentzian_model(self):
         result = lmodel.fit(y, pars, x=x, nan_policy='propagate')
         return result
 
-    def lorentzian_model_w_lims(self, center_min = None, center_max = None):
+    # def lorentzian_model_w_lims(self, center_min = None, center_max = None):
+    #     x,y = self.background_correction()
+    #     lmodel = LorentzianModel(prefix = 'l1_') # calling lorentzian model
+    #     pars = lmodel.guess(y, x=x) # parameters - center, width, height
+    #     pars['l1_center'].set(min = center_min, max = center_max)
+    #     result = lmodel.fit(y, pars, x=x, nan_policy='propagate')
+    #     return result
+
+    def lorentzian_model_w_lims(self, center_initial_guess=None, sigma_initial_guess=None, center_min = None, center_max = None):
         x,y = self.background_correction()
         lmodel = LorentzianModel(prefix = 'l1_') # calling lorentzian model
         pars = lmodel.guess(y, x=x) # parameters - center, width, height
-        pars['l1_center'].set(min = center_min, max = center_max)
+        pars['l1_center'].set(center_initial_guess, min = center_min, max = center_max)
+        pars['l1_sigma'].set(sigma_initial_guess)
         result = lmodel.fit(y, pars, x=x, nan_policy='propagate')
         return result
 
@@ -96,18 +106,35 @@ class Double_Gaussian(Spectra_Fit):
     """
 
     def gaussian_model(self):
+
+        x,y = self.background_correction()
+        gmodel_1 = GaussianModel(prefix='g1_') # calling gaussian model
+        pars = gmodel_1.guess(y, x=x) # parameters - center, width, height
+
+        gmodel_2 = GaussianModel(prefix='g2_')
+        pars.update(gmodel_2.make_params()) # update parameters - center, width, height
+
+        gmodel = gmodel_1 + gmodel_2
+        result = gmodel.fit(y, pars, x=x, nan_policy='propagate')
+        return result
+
+    def gaussian_model_w_lims(self, center_initial_guesses=None, sigma_initial_guesses=None, min_max_range=None):
+        #center_initial_guesses - list containing initial guesses for peak centers. [center_guess1, center_guess2]
+        #sigma_initial_guesses - list containing initial guesses for sigma. [sigma1, sigma2]
+        #min_max_range - list containing lists of min and max for peak center. [ [min1, max1], [min2, max2] ] 
+        
         x,y = self.background_correction()
         gmodel_1 = GaussianModel(prefix='g1_') # calling gaussian model
         pars = gmodel_1.guess(y, x=x) # parameters - center, width, height
-        pars['g1_center'].set(800, min = 795, max = 820)
-        pars['g1_sigma'].set(15)
+        pars['g1_center'].set(center_initial_guesses[0], min = min_max_range[0][0], max = min_max_range[0][1])
+        pars['g1_sigma'].set(sigma_initial_guesses[0])
         pars['g1_amplitude'].set(min=0)
 
         gmodel_2 = GaussianModel(prefix='g2_')
         pars.update(gmodel_2.make_params()) # update parameters - center, width, height
-        pars['g2_center'].set(767, min = 760, max = 775)
+        pars['g2_center'].set(center_initial_guesses[1], min = min_max_range[1][0], max = min_max_range[1][1])
+        pars['g2_sigma'].set(sigma_initial_guesses[1])
         pars['g2_amplitude'].set(min = 0)
-        pars['g2_sigma'].set(min = pars['g1_sigma'].value)
 
         gmodel = gmodel_1 + gmodel_2
         result = gmodel.fit(y, pars, x=x, nan_policy='propagate')
@@ -146,7 +173,7 @@ def multi_gaussian(self):
             else:
                 composite_pars.update(model.make_params())
                 composite_pars['g'+str(i+1)+'_center'].set(self.peak_pos[i],
-                                                          min = self.min_max_range[1][0], max = self.min_max_range[1][1])
+                                                          min = self.min_max_range[i][0], max = self.min_max_range[i][1])
                 composite_pars['g'+str(i+1)+'_sigma'].set(min = composite_pars['g1_sigma'].value)
                 composite_pars['g'+str(i+1)+'_amplitude'].set(min = 0)