LabTools3 1.1.3.11__tar.gz → 1.1.3.14__tar.gz

{LabTools3-1.1.3.11 → labtools3-1.1.3.14}/LT/box.py

@@ -247,6 +247,8 @@ class histo:
     title          Set the title
     xlabel         Set the x-label
     ylabel         Set the y-label
+    calc_w2        If True calccultae the sum of the square of the weights for each bin to calculate the final bin error
+    weights        Array of weight to increment the histogram with
     ============   =====================================================
 
     Additional keyword arguments are passed to the :func:`numpy.histogram` function
@@ -265,8 +267,11 @@ class histo:
                  title = 'my histogram', \
                  xlabel = 'x-bin', \
                  ylabel = 'content', \
+                 calc_w2 = False, \
                  **kwargs):
         self.res = None
+        self.has_weights = 'weights' in kwargs
+        self.calc_w2 = calc_w2
         self.fit_dict = {}
         # initialize fitting
         self.b0 = Parameter(0., 'b0')
@@ -342,10 +347,13 @@ class histo:
         """
         if not add:
             # a new filling
-            try:
-                self.res = np.histogram(y, new = None, **kwargs)
-            except:
-                self.res = np.histogram(y, **kwargs)
+            self.res = np.histogram(y,  **kwargs)
+            if self.calc_w2 and self.has_weights:
+                w_loc = kwargs['weights']**2
+                # histogram with weights squared
+                kwargs1 = copy.copy(kwargs)
+                kwargs1['weights'] = w_loc
+                self.w2 = np.histogram(y, **kwargs1)[0] # save only the bin content
             self.__setup_bins(error = None)
         else:
             # the bins have already been defined continue
@@ -353,12 +361,18 @@ class histo:
             if self.res is None:
                 print("no binning information: try fill with add = False ")
                 return
-            try:
-                res = np.histogram(y, new = True, bins = self.res[1], **kwargs)
-            except:
-                res = np.histogram(y, bins = self.res[1], **kwargs)
+
+            res = np.histogram(y, bins = self.res[1], **kwargs)
+            if self.calc_w2 and self.has_weights:
+                w2 = kwargs['weights']**2
+                # histogram with weights squared
+                kwargs1 = copy.copy(kwargs)
+                kwargs1['weights'] = w2
+                w2 = np.histogram(y, **kwargs1)[0] # save only the bin content
             # add the new bin content to the old one
             self.res = (self.res[0] + res[0], self.res[1])
+            if self.calc_w2 and self.has_weights:
+                self.w2 += w2
             # update the histogram information
             self.__setup_bins(error = None)
         # end of fill
@@ -399,24 +413,24 @@ class histo:
         if (xmin is None) and (xmax is None):
             if self.window_set:
                 sel = self.window_sel
-                sum = self.bin_content[sel].sum()
+                sum_v = self.bin_content[sel].sum()
                 sum_err = np.sqrt( (self.bin_error[sel]**2).sum())
             else:                
-                sum = self.bin_content.sum()
+                sum_v = self.bin_content.sum()
                 sum_err = np.sqrt( (self.bin_error**2).sum())
         elif (xmin is None):
             sel = (self.bin_center <= xmax)
-            sum = self.bin_content[sel].sum()
+            sum_v = self.bin_content[sel].sum()
             sum_err = np.sqrt( (self.bin_error[sel]**2).sum())
         elif (xmax is None):
             sel = (xmin <= self.bin_center)
-            sum = (self.bin_content[sel]).sum()
+            sum_v = (self.bin_content[sel]).sum()
             sum_err = np.sqrt( (self.bin_error[sel]**2).sum())
         else:
             sel = (xmin <= self.bin_center) & (self.bin_center <= xmax)
-            sum = (self.bin_content[sel]).sum()
+            sum_v = (self.bin_content[sel]).sum()
             sum_err = np.sqrt( (self.bin_error[sel]**2).sum())
-        return (sum, sum_err)
+        return (sum_v, sum_err)
 
     def copy(self):
         """
@@ -800,7 +814,7 @@ class histo:
         # is there a range given, or is a window set
         sel_all = np.ones_like(self.bin_center, dtype = 'bool')
         if init:
-            self.init_gauss(xmin, xmax)
+            self.init_gauss(xmin, xmax, ignore_zeros)
         if (xmin is None) and (xmax is None):
             # check if a window is set
             if self.window_set:
@@ -835,9 +849,10 @@ class histo:
                 self.fit_indx, = np.where(sel_w)
             else:
                 self.fit_indx, = np.where(sel)
-        # set minimal error to 1
-        is_zero = np.where(self.bin_error == 0.)
-        self.bin_error[is_zero] = 1.
+        if not ignore_zeros:
+            # set minimal error to 1
+            is_zero = np.where(self.bin_error == 0.)
+            self.bin_error[is_zero] = 1.
         # store the limits of fit_indx
         self.fit_index_min = self.fit_indx.min()
         self.fit_index_max = self.fit_indx.max()
@@ -909,7 +924,7 @@ class histo:
         self.sigma.set(1.)
         self.A.set(1.)
 
-    def init_gauss(self, xmin = None, xmax = None):
+    def init_gauss(self, xmin = None, xmax = None, ignore_zeros = True):
         """
 
         Calculate the initial parameter guess for a gaussian. These parameters
@@ -952,9 +967,10 @@ class histo:
                 self.fit_indx, = np.where(sel_w)
             else:
                 self.fit_indx, = np.where(sel)
-        # set minimal error to 1
-        is_zero = np.where(self.bin_error == 0.)
-        self.bin_error[is_zero] = 1.
+        if not ignore_zeros:
+            # set minimal error to 1
+            s_zero = np.where(self.bin_error == 0.)
+            self.bin_error[is_zero] = 1.
         # do the fit
         bin_content = self.bin_content[self.fit_indx]
         bin_center = self.bin_center[self.fit_indx]
@@ -1008,6 +1024,44 @@ class histo:
         fit_val = (self.b2()*x + self.b1())*x + self.b0() + \
             self.A()*np.exp(-0.5*((x-self.mean())/self.sigma() )**2)
         return fit_val
+    
+    def gauss(self, x):
+        """
+        
+        Return the value of the gauss function using the fitted parameters at x
+        
+
+        Parameters
+        ----------
+        x : float
+
+        Returns
+        -------
+        gauss(x) : float
+            fitted gaussian at x.
+
+        """
+        g_val = self.A()*np.exp(-0.5*((x-self.mean())/self.sigma() )**2)
+        return g_val
+    
+    def bkg(self, x):
+        """
+        
+        Return the value of the fitted background at x
+
+        Parameters
+        ----------
+        x : float
+
+        Returns
+        -------
+        b_val(x) : float
+            fitted backgound at x.
+
+        """
+        b_val = (self.b2()*x + self.b1())*x + self.b0()
+        return b_val
+        
     def apply_calibration(self, cal):
         """
 
@@ -1028,7 +1082,10 @@ class histo:
         self.bin_content = self.res[0]
         self.bins = self.res[1]
         if error is None:
-            self.bin_error = np.sqrt(self.bin_content)
+            if not self.has_weights:  # no weights specified
+                self.bin_error = np.sqrt(self.bin_content)
+            elif self.calc_w2:   # weights given and sum(weights**2) calculated for each bin
+                self.bin_error = np.sqrt(self.w2)
         else:
             self.bin_error = error
         self.__prepare_histo_plot()
@@ -1226,6 +1283,8 @@ class histo2d:
     xlabel         Set the x-label
     ylabel         Set the y-label
     zlabel         Set the z-label
+    calc_w2        If True calccultae the sum of the square of the weights for each bin to calculate the final bin error
+    weights        Array of weight to increment the histogram with
     colorbar       if True, plot a colorbar
     bad_color      Set the color for plot for bins below zmin (default: w)
     logz           if True plot content on log scale
@@ -1252,10 +1311,13 @@ class histo2d:
                  bad_color = 'w',\
                  colorbar = True, \
                  logz = False,\
+                 calc_w2 = False,\
                  **kwargs):
+        self.has_weights = 'weights' in kwargs
         self.bad_color = bad_color # color for bad pixels
         self.colorbar = colorbar
         self.logz = logz
+        self.calc_w2 = calc_w2
         # initialize fitting
         if (x_values is not None) and (y_values is not None):
             # values have been given for filling
@@ -1329,10 +1391,13 @@ class histo2d:
         """
         if not add:
             # a new filling
-            try:
-                self.res = np.histogram2d(x, y, new = None, **kwargs)
-            except:
-                self.res = np.histogram2d(x, y, **kwargs)
+            self.res = np.histogram2d(x, y, **kwargs)
+            if self.calc_w2 and self.has_weights:
+                w_loc = kwargs['weights']**2
+                # histogram with weights squared
+                kwargs1 = copy.copy(kwargs)
+                kwargs1['weights'] = w_loc
+                self.w2 = np.histogram2d(x,y, **kwargs1)[0] # save only the bin content
             self.__setup_bins(error = None)
         else:
             # the bins have already been defined continue
@@ -1341,8 +1406,15 @@ class histo2d:
                 print("no binning information: try fill with add = False ")
                 return
             res = np.histogram2d(x, y, bins = [self.x_bins,self.y_bins], **kwargs)
+            if self.calc_w2 and self.has_weights:
+                w_loc = kwargs['weights']**2
+                # histogram with weights squared
+                kwargs1 = copy.copy(kwargs)
+                kwargs1['weights'] = w_loc
+                w2 = np.histogram2d(x,y, bins = [self.x_bins,self.y_bins], **kwargs1)[0] # save only the bin content            
             # add the new bin content to the old one
             self.res = (self.res[0] + res[0], self.res[1], self.res[2])
+            self.w2 += w2
             # update the histogram information
             self.__setup_bins(error = None)
         # end of fill
@@ -1774,7 +1846,12 @@ class histo2d:
         self.x_bins = self.res[1]
         self.y_bins = self.res[2]
         if error is None:
-            self.bin_error = np.sqrt(self.bin_content)
+            if not self.has_weights:  # no weights specified
+                self.bin_error = np.sqrt(self.bin_content)
+            elif self.calc_w2:   # weights given and sum(weights**2) calculated for each bin
+                self.bin_error = np.sqrt(self.w2)
+            else:
+                self.bin_error = np.sqrt(self.bin_content)
         else:
             self.bin_error = error
         self.__prepare_histo_plot()

{LabTools3-1.1.3.11 → labtools3-1.1.3.14}/LT/plotting.py

@@ -48,6 +48,7 @@ def plot_exp(x,\
                  # set general line width
                  linewidth=2,\
                  logy = False,\
+                 logx = False, \
                  min_val = None,\
                  scale = 1.,\
                  # labelling
@@ -82,6 +83,7 @@ def plot_exp(x,\
     marker         marker type (see :func:`~matplotlib.pyplot.plot`)
     linestyle      line style (see :func:`~matplotlib.pyplot.plot`)
     logy           use log y-scale (True/False)
+    logx           use log x-scale (True/False)
     label          label for data (used in :func:`~matplotlib.pyplot.legend` )  
     min_val        min. values to be plotted
     scale          scale ally-values (including errrors ) by this factor
@@ -105,7 +107,9 @@ def plot_exp(x,\
         axes = pl.gca()
     if logy:
         axes.set_yscale("log", nonpositive='clip')
-    if dy == []:
+    if logx:
+        axes.set_xscale("log", nonpositive='clip')
+    if len(dy) == 0:
         # no error bars
         e=axes.plot(xx, yy,\
                    linestyle=linestyle,\
@@ -121,7 +125,7 @@ def plot_exp(x,\
             axes.set_title(plot_title)
         return e
     # plot second (total) error bar behind first (statistical) one
-    if dyt != []:
+    if len(dyt) != 0:
         # in case ther are several different errors
         et=axes.errorbar(xx, \
                         yy, \
@@ -175,6 +179,7 @@ def plot_line(x,\
              y, \
              label='_nolegend_',\
              logy = False, \
+             logx = False, \
              convx = 1., \
              convy = 1., \
               axes = None, \
@@ -195,6 +200,7 @@ def plot_line(x,\
     ============   ===================================================== 
     label          label for curve (used in :func:`~matplotlib.pyplot.legend` ) 
     logy           use log y-scale (True/False) 
+    logx           use log x-scale (True/False) 
     convx          scale all x-values by this factor 
     convy          scale all y-values by this factor 
     ============   =====================================================
@@ -206,10 +212,11 @@ def plot_line(x,\
         axes = pl.gca()
     xx = np.array(x) * convx
     yy = np.array(y) * convy
+    s=axes.plot(xx, yy, marker='None',label=label,**kwargs)
     if logy:
-        s=axes.semilogy(xx, yy, marker='None',label=label,**kwargs)
-    else:
-        s=axes.plot(xx, yy, marker='None',label=label,**kwargs)
+        axes.set_yscale("log", nonpositive='clip')
+    if logx:
+        axes.set_xscale("log", nonpositive='clip')
     return s
 
 #---------------------------------------------------------------------- 
@@ -225,6 +232,7 @@ def plot_spline(x, \
                 convx = 1., \
                 convy = 1., \
                 logy = False,\
+                logx = False, \
                 axes = None, \
                 **kwargs):
     """
@@ -243,6 +251,7 @@ def plot_spline(x, \
     ============   ===================================================== 
     label          label for curve (used in :func:`~matplotlib.pyplot.legend` ) 
     logy           use log y-scale (True/False)
+    logx           use log x-scale (True/False)
     nstep          factor by which the number of interpolated data points is increased 
     convx          scale all x-values by this factor 
     convy          scale all y-values by this factor 
@@ -266,15 +275,15 @@ def plot_spline(x, \
     # now get interpolation coefficients
     yvar_cj = splrep(xvar,yvar)
     new_yvar = splev(new_xvar, yvar_cj)
+    s=axes.plot(new_xvar, new_yvar, marker=marker,\
+               label=label,\
+               **kwargs)
     if logy:
-        s=axes.semilogy(new_xvar, new_yvar, marker=marker,\
-                   label=label,\
-                   **kwargs)
-    else:
-        s=axes.plot(new_xvar, new_yvar, marker=marker,\
-                   label=label,\
-                   **kwargs)
+        axes.set_yscale("log", nonpositive='clip')
+    if logx:
+        axes.set_xscale("log", nonpositive='clip')
     return s
+
 #---------------------------------------------------------------------- 
 #---------------------------------------------------------------------- 
 # plot spline lines only, this has to be used carefully
@@ -287,6 +296,7 @@ def log_plot_spline(x, \
                     nstep = 5,\
                     conv = 1., \
                     axes = None, \
+                    logx = False, \
                     **kwargs):
     if (axes == None):
         axes = pl.gca()
@@ -306,6 +316,8 @@ def log_plot_spline(x, \
     s=axes.semilogy(new_xvar, new_yvar, marker=marker,\
                label=label,\
                **kwargs)
+    if logx:
+        axes.set_xscale("log", nonpositive='clip')
     return s
 #---------------------------------------------------------------------- 
 # plot data read using data file
@@ -332,6 +344,7 @@ def datafile_plot_exp(set,\
                  linewidth=2,\
                  # set for semilog plot
                  logy = False,\
+                 logx = False, \
                  # minimum value for semilog plot
                  min_val = None,\
                  scale = 1.,\
@@ -369,7 +382,8 @@ def datafile_plot_exp(set,\
     marker         marker type (see :func:`~matplotlib.pyplot.plot`)
     linestyle      line style (see :func:`~matplotlib.pyplot.plot`)
     label          label for data (used in :func:`~matplotlib.pyplot.legend` )  
-    logy           use log y-scale (True/False) 
+    logy           use log y-scale (True/False)
+    logx           use log x-scale (True/False)
     min_val        min. values to be plotted
     scale          scale all y-values (including errrors ) by this factor 
     x_label        label for x-axis
@@ -393,6 +407,8 @@ def datafile_plot_exp(set,\
     yy = np.array(set.get_data(y))*scale
     if logy:
         axes.set_yscale("log", nonpositive='clip')
+    if logx:
+        axes.set_xscale("log", nonpositive='clip')
     if dy != None:
         dyy = np.array(set.get_data(dy))*scale
     dyyt = []
@@ -407,7 +423,7 @@ def datafile_plot_exp(set,\
                         label=label, \
                         **kwargs)
     # plot second (total) error bar behind first (statistical) one
-    elif dyyt != []:
+    elif len(dyyt) != 0:
         # in case ther are several different errors
         et=axes.errorbar(xx, \
                         yy, \
@@ -474,6 +490,7 @@ def datafile_plot_theory(set,\
                          convx = 1.,\
                          convy = 1.,\
                          logy = False,\
+                         logx = False, \
                          axes = None, \
                          **kwargs):
     """
@@ -487,19 +504,17 @@ def datafile_plot_theory(set,\
     xvar = np.array(set.get_data(x) )*convx
     yvar = np.array(set.get_data(y))*convy
 
+    s=axes.plot(xvar, yvar, marker=marker,color=color, \
+               label=label,\
+               **kwargs)
+
     if logy:
-        for i in range(len(xvar)):
-            if yvar[i] <= 0. :
-                yvar[i] =  min_val
-        s=axes.semilogy(xvar, yvar, marker=marker,color=color, \
-                   label=label,\
-                   **kwargs)
-    else:
-        s=axes.plot(xvar, yvar, marker=marker,color=color, \
-                   label=label,\
-                   **kwargs)
+        axes.set_yscale("log", nonpositive='clip')
+    if logx:
+        axes.set_xscale("log", nonpositive='clip')
     return s
 
+
 #---------------------------------------------------------------------- 
 #plot spline lines only, this has to be used carefully
 def datafile_spline_plot_theory(set,\
@@ -513,6 +528,7 @@ def datafile_spline_plot_theory(set,\
                                 convx = 1.,\
                                 convy = 1.,\
                                 logy = False,\
+                                logx = False, \
                                 axes = None, \
                                 **kwargs):
     """
@@ -528,23 +544,18 @@ def datafile_spline_plot_theory(set,\
     # number of interpolation steps
     new_xvar = np.linspace(xvar[0], xvar[-1:], int(nstep*len(xvar)) ) # create a new range with more points
     # handle 0 and neg values for log scale
-    if logy:
-        for i in range(len(xvar)):
-            if yvar[i] <= 0. :
-                yvar[i] = min_val
     # now get interpolation coefficients
     yvar_cj = splrep(xvar,yvar)
     new_yvar = splev(new_xvar, yvar_cj)
+    s=axes.plot(new_xvar, new_yvar, marker=marker,\
+               color=color,\
+               label=label,\
+               **kwargs)
     if logy:
-        s=axes.semilogy(new_xvar, new_yvar, marker=marker,\
-                      color=color,\
-                      label=label,\
-                      **kwargs)
-    else:
-        s=axes.plot(new_xvar, new_yvar, marker=marker,\
-                   color=color,\
-                   label=label,\
-                   **kwargs)
+        axes.set_yscale("log", nonpositive='clip')
+    if logx:
+        axes.set_xscale("log", nonpositive='clip')
+
     return s
 
 #----------------------------------------------------------------------

{LabTools3-1.1.3.11 → labtools3-1.1.3.14}/LT_Fit/linear_fit.py

@@ -30,6 +30,7 @@ def linfit(function, y, x = None, y_err = None,  nplot = 100):
     if x is None: x = arange(y.shape[0])
     #
     # pdb.set_trace()
+
     A = function(x).transpose()
     if y_err is None:
         weight = ones_like(y)
@@ -39,14 +40,21 @@ def linfit(function, y, x = None, y_err = None,  nplot = 100):
     # store the parameter values in a list to be passed to the fitting function 
     M = dot(Aw.transpose(),A)
     # invert the matrix
-    Mi = linalg.inv(M)
+    if np.isscalar(M):
+        # check if only one parameter
+        Mi = np.expand_dims(np.array([1./M]), 0)
+    else:
+        Mi = linalg.inv(M)
     b=dot(Aw.transpose(),y)
     par = dot(Mi,b)
     # for the calculation of errors
     # sigma = sqrt( diag(Mi) )
     # final total chi square
     # number of degrees of freedom
-    yfit = dot(par,function(x))
+    fux = function(x)
+    if fux.ndim == 1:
+        fux = np.expand_dims(fux, 0)
+    yfit = dot(par,fux)
     n_dof = len(y) - len(par)
     diff = yfit - y
     chi2 = sum( power(diff, 2 )*weight )
@@ -56,7 +64,12 @@ def linfit(function, y, x = None, y_err = None,  nplot = 100):
     # create an array with calculated values for plotting
     if (nplot > 0):
         xpl = linspace(x.min(), x.max(), nplot+1)
-        ypl = dot(par,function(xpl))
+        fux = function(xpl)
+        if fux.ndim ==1 :
+            fux = np.expand_dims(fux, 0)
+            ypl = np.squeeze(dot(par,fux))
+        else:
+            ypl = dot(par,fux)
     stat = {'chisquare': chi2, \
                 'red. chisquare':chi2_red, \
                 'conf. level':CL, \

{LabTools3-1.1.3.11 → labtools3-1.1.3.14/LabTools3.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: LabTools3
-Version: 1.1.3.11
+Version: 1.1.3.14
 Summary: Python 3 Package of modules for typical analysis tasks analyzing physics data
 Home-page: http://wanda.fiu.edu/LabTools3
 Author: Werner Boeglin
@@ -44,3 +44,9 @@ Version 1.1.3.9: Fixed a bug in *histo* where histogram window variable was not
 Version 1.1.3.10: 2d histo now accepts axes as keywords. Make sure for 3d display the axes is a Axes3DSubplot.
 
 Version 1.1.3.11: new control on title and axes labels for plotting
+
+Version 1.1.3.12: bug fixes, peak and background functions are available for histogram fits.
+
+Version 1.1.3.13: fitting of only one parameters enabled for linear fits
+
+Version 1.1.3.14: error calcualtion for filling histograms with weights include sum of weight**2 

{LabTools3-1.1.3.11/LabTools3.egg-info → labtools3-1.1.3.14}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: LabTools3
-Version: 1.1.3.11
+Version: 1.1.3.14
 Summary: Python 3 Package of modules for typical analysis tasks analyzing physics data
 Home-page: http://wanda.fiu.edu/LabTools3
 Author: Werner Boeglin
@@ -44,3 +44,9 @@ Version 1.1.3.9: Fixed a bug in *histo* where histogram window variable was not
 Version 1.1.3.10: 2d histo now accepts axes as keywords. Make sure for 3d display the axes is a Axes3DSubplot.
 
 Version 1.1.3.11: new control on title and axes labels for plotting
+
+Version 1.1.3.12: bug fixes, peak and background functions are available for histogram fits.
+
+Version 1.1.3.13: fitting of only one parameters enabled for linear fits
+
+Version 1.1.3.14: error calcualtion for filling histograms with weights include sum of weight**2 

{LabTools3-1.1.3.11 → labtools3-1.1.3.14}/README.md

@@ -29,4 +29,10 @@ Version 1.1.3.9: Fixed a bug in *histo* where histogram window variable was not
 
 Version 1.1.3.10: 2d histo now accepts axes as keywords. Make sure for 3d display the axes is a Axes3DSubplot.
 
-Version 1.1.3.11: new control on title and axes labels for plotting
+Version 1.1.3.11: new control on title and axes labels for plotting
+
+Version 1.1.3.12: bug fixes, peak and background functions are available for histogram fits.
+
+Version 1.1.3.13: fitting of only one parameters enabled for linear fits
+
+Version 1.1.3.14: error calcualtion for filling histograms with weights include sum of weight**2 

{LabTools3-1.1.3.11 → labtools3-1.1.3.14}/setup.py

@@ -5,7 +5,7 @@ with open("README.md", "r") as fh:
 
 setup(
     name = "LabTools3",
-    version = "1.1.3.11",
+    version = "1.1.3.14",
     packages = find_packages(),
     # add additional files
     package_data = {'':['*.bat','*.command']},