| 1 | #/usr/bin/env python |
|---|
| 2 | # -*- coding: utf-8 -*- |
|---|
| 3 | ''' |
|---|
| 4 | *GSASII powder calculation module* |
|---|
| 5 | ================================== |
|---|
| 6 | |
|---|
| 7 | ''' |
|---|
| 8 | ########### SVN repository information ################### |
|---|
| 9 | # $Date: 2016-06-30 13:36:29 +0000 (Thu, 30 Jun 2016) $ |
|---|
| 10 | # $Author: vondreele $ |
|---|
| 11 | # $Revision: 2355 $ |
|---|
| 12 | # $URL: trunk/GSASIIpwd.py $ |
|---|
| 13 | # $Id: GSASIIpwd.py 2355 2016-06-30 13:36:29Z vondreele $ |
|---|
| 14 | ########### SVN repository information ################### |
|---|
| 15 | import sys |
|---|
| 16 | import math |
|---|
| 17 | import time |
|---|
| 18 | import os |
|---|
| 19 | import subprocess as subp |
|---|
| 20 | |
|---|
| 21 | import numpy as np |
|---|
| 22 | import scipy as sp |
|---|
| 23 | import numpy.linalg as nl |
|---|
| 24 | import numpy.ma as ma |
|---|
| 25 | import random as rand |
|---|
| 26 | from numpy.fft import ifft, fft, fftshift |
|---|
| 27 | import scipy.interpolate as si |
|---|
| 28 | import scipy.stats as st |
|---|
| 29 | import scipy.optimize as so |
|---|
| 30 | |
|---|
| 31 | import GSASIIpath |
|---|
| 32 | GSASIIpath.SetVersionNumber("$Revision: 2355 $") |
|---|
| 33 | import GSASIIlattice as G2lat |
|---|
| 34 | import GSASIIspc as G2spc |
|---|
| 35 | import GSASIIElem as G2elem |
|---|
| 36 | import GSASIIgrid as G2gd |
|---|
| 37 | import GSASIIIO as G2IO |
|---|
| 38 | import GSASIImath as G2mth |
|---|
| 39 | import pypowder as pyd |
|---|
| 40 | try: |
|---|
| 41 | import pydiffax as pyx |
|---|
| 42 | except ImportError: |
|---|
| 43 | print 'pydiffax is not available for this platform - under develpment' |
|---|
| 44 | |
|---|
| 45 | |
|---|
| 46 | # trig functions in degrees |
|---|
| 47 | sind = lambda x: math.sin(x*math.pi/180.) |
|---|
| 48 | asind = lambda x: 180.*math.asin(x)/math.pi |
|---|
| 49 | tand = lambda x: math.tan(x*math.pi/180.) |
|---|
| 50 | atand = lambda x: 180.*math.atan(x)/math.pi |
|---|
| 51 | atan2d = lambda y,x: 180.*math.atan2(y,x)/math.pi |
|---|
| 52 | cosd = lambda x: math.cos(x*math.pi/180.) |
|---|
| 53 | acosd = lambda x: 180.*math.acos(x)/math.pi |
|---|
| 54 | rdsq2d = lambda x,p: round(1.0/math.sqrt(x),p) |
|---|
| 55 | #numpy versions |
|---|
| 56 | npsind = lambda x: np.sin(x*np.pi/180.) |
|---|
| 57 | npasind = lambda x: 180.*np.arcsin(x)/math.pi |
|---|
| 58 | npcosd = lambda x: np.cos(x*math.pi/180.) |
|---|
| 59 | npacosd = lambda x: 180.*np.arccos(x)/math.pi |
|---|
| 60 | nptand = lambda x: np.tan(x*math.pi/180.) |
|---|
| 61 | npatand = lambda x: 180.*np.arctan(x)/np.pi |
|---|
| 62 | npatan2d = lambda y,x: 180.*np.arctan2(y,x)/np.pi |
|---|
| 63 | npT2stl = lambda tth, wave: 2.0*npsind(tth/2.0)/wave |
|---|
| 64 | npT2q = lambda tth,wave: 2.0*np.pi*npT2stl(tth,wave) |
|---|
| 65 | ateln2 = 8.0*math.log(2.0) |
|---|
| 66 | |
|---|
| 67 | #GSASII pdf calculation routines |
|---|
| 68 | |
|---|
| 69 | def Transmission(Geometry,Abs,Diam): |
|---|
| 70 | ''' |
|---|
| 71 | Calculate sample transmission |
|---|
| 72 | |
|---|
| 73 | :param str Geometry: one of 'Cylinder','Bragg-Brentano','Tilting flat plate in transmission','Fixed flat plate' |
|---|
| 74 | :param float Abs: absorption coeff in cm-1 |
|---|
| 75 | :param float Diam: sample thickness/diameter in mm |
|---|
| 76 | ''' |
|---|
| 77 | if 'Cylinder' in Geometry: #Lobanov & Alte da Veiga for 2-theta = 0; beam fully illuminates sample |
|---|
| 78 | MuR = Abs*Diam/20.0 |
|---|
| 79 | if MuR <= 3.0: |
|---|
| 80 | T0 = 16/(3.*math.pi) |
|---|
| 81 | T1 = -0.045780 |
|---|
| 82 | T2 = -0.02489 |
|---|
| 83 | T3 = 0.003045 |
|---|
| 84 | T = -T0*MuR-T1*MuR**2-T2*MuR**3-T3*MuR**4 |
|---|
| 85 | if T < -20.: |
|---|
| 86 | return 2.06e-9 |
|---|
| 87 | else: |
|---|
| 88 | return math.exp(T) |
|---|
| 89 | else: |
|---|
| 90 | T1 = 1.433902 |
|---|
| 91 | T2 = 0.013869+0.337894 |
|---|
| 92 | T3 = 1.933433+1.163198 |
|---|
| 93 | T4 = 0.044365-0.04259 |
|---|
| 94 | T = (T1-T4)/(1.0+T2*(MuR-3.0))**T3+T4 |
|---|
| 95 | return T/100. |
|---|
| 96 | elif 'plate' in Geometry: |
|---|
| 97 | MuR = Abs*Diam/10. |
|---|
| 98 | return math.exp(-MuR) |
|---|
| 99 | elif 'Bragg' in Geometry: |
|---|
| 100 | return 0.0 |
|---|
| 101 | |
|---|
| 102 | def SurfaceRough(SRA,SRB,Tth): |
|---|
| 103 | ''' Suortti (J. Appl. Cryst, 5,325-331, 1972) surface roughness correction |
|---|
| 104 | :param float SRA: Suortti surface roughness parameter |
|---|
| 105 | :param float SRB: Suortti surface roughness parameter |
|---|
| 106 | :param float Tth: 2-theta(deg) - can be numpy array |
|---|
| 107 | |
|---|
| 108 | ''' |
|---|
| 109 | sth = npsind(Tth/2.) |
|---|
| 110 | T1 = np.exp(-SRB/sth) |
|---|
| 111 | T2 = SRA+(1.-SRA)*np.exp(-SRB) |
|---|
| 112 | return (SRA+(1.-SRA)*T1)/T2 |
|---|
| 113 | |
|---|
| 114 | def SurfaceRoughDerv(SRA,SRB,Tth): |
|---|
| 115 | ''' Suortti surface roughness correction derivatives |
|---|
| 116 | :param float SRA: Suortti surface roughness parameter (dimensionless) |
|---|
| 117 | :param float SRB: Suortti surface roughness parameter (dimensionless) |
|---|
| 118 | :param float Tth: 2-theta(deg) - can be numpy array |
|---|
| 119 | :return list: [dydSRA,dydSRB] derivatives to be used for intensity derivative |
|---|
| 120 | ''' |
|---|
| 121 | sth = npsind(Tth/2.) |
|---|
| 122 | T1 = np.exp(-SRB/sth) |
|---|
| 123 | T2 = SRA+(1.-SRA)*np.exp(-SRB) |
|---|
| 124 | Trans = (SRA+(1.-SRA)*T1)/T2 |
|---|
| 125 | dydSRA = ((1.-T1)*T2-(1.-np.exp(-SRB))*Trans)/T2**2 |
|---|
| 126 | dydSRB = ((SRA-1.)*T1*T2/sth-Trans*(SRA-T2))/T2**2 |
|---|
| 127 | return [dydSRA,dydSRB] |
|---|
| 128 | |
|---|
| 129 | def Absorb(Geometry,MuR,Tth,Phi=0,Psi=0): |
|---|
| 130 | '''Calculate sample absorption |
|---|
| 131 | :param str Geometry: one of 'Cylinder','Bragg-Brentano','Tilting Flat Plate in transmission','Fixed flat plate' |
|---|
| 132 | :param float MuR: absorption coeff * sample thickness/2 or radius |
|---|
| 133 | :param Tth: 2-theta scattering angle - can be numpy array |
|---|
| 134 | :param float Phi: flat plate tilt angle - future |
|---|
| 135 | :param float Psi: flat plate tilt axis - future |
|---|
| 136 | ''' |
|---|
| 137 | |
|---|
| 138 | def muRunder3(MuR,Sth2): |
|---|
| 139 | T0 = 16.0/(3.*np.pi) |
|---|
| 140 | T1 = (25.99978-0.01911*Sth2**0.25)*np.exp(-0.024551*Sth2)+ \ |
|---|
| 141 | 0.109561*np.sqrt(Sth2)-26.04556 |
|---|
| 142 | T2 = -0.02489-0.39499*Sth2+1.219077*Sth2**1.5- \ |
|---|
| 143 | 1.31268*Sth2**2+0.871081*Sth2**2.5-0.2327*Sth2**3 |
|---|
| 144 | T3 = 0.003045+0.018167*Sth2-0.03305*Sth2**2 |
|---|
| 145 | Trns = -T0*MuR-T1*MuR**2-T2*MuR**3-T3*MuR**4 |
|---|
| 146 | return np.exp(Trns) |
|---|
| 147 | |
|---|
| 148 | def muRover3(MuR,Sth2): |
|---|
| 149 | T1 = 1.433902+11.07504*Sth2-8.77629*Sth2*Sth2+ \ |
|---|
| 150 | 10.02088*Sth2**3-3.36778*Sth2**4 |
|---|
| 151 | T2 = (0.013869-0.01249*Sth2)*np.exp(3.27094*Sth2)+ \ |
|---|
| 152 | (0.337894+13.77317*Sth2)/(1.0+11.53544*Sth2)**1.555039 |
|---|
| 153 | T3 = 1.933433/(1.0+23.12967*Sth2)**1.686715- \ |
|---|
| 154 | 0.13576*np.sqrt(Sth2)+1.163198 |
|---|
| 155 | T4 = 0.044365-0.04259/(1.0+0.41051*Sth2)**148.4202 |
|---|
| 156 | Trns = (T1-T4)/(1.0+T2*(MuR-3.0))**T3+T4 |
|---|
| 157 | return Trns/100. |
|---|
| 158 | |
|---|
| 159 | Sth2 = npsind(Tth/2.0)**2 |
|---|
| 160 | Cth2 = 1.-Sth2 |
|---|
| 161 | if 'Cylinder' in Geometry: #Lobanov & Alte da Veiga for 2-theta = 0; beam fully illuminates sample |
|---|
| 162 | if 'array' in str(type(MuR)): |
|---|
| 163 | MuRSTh2 = np.concatenate((MuR,Sth2)) |
|---|
| 164 | AbsCr = np.where(MuRSTh2[0]<=3.0,muRunder3(MuRSTh2[0],MuRSTh2[1]),muRover3(MuRSTh2[0],MuRSTh2[1])) |
|---|
| 165 | return AbsCr |
|---|
| 166 | else: |
|---|
| 167 | if MuR <= 3.0: |
|---|
| 168 | return muRunder3(MuR,Sth2) |
|---|
| 169 | else: |
|---|
| 170 | return muRover3(MuR,Sth2) |
|---|
| 171 | elif 'Bragg' in Geometry: |
|---|
| 172 | return 1.0 |
|---|
| 173 | elif 'Fixed' in Geometry: #assumes sample plane is perpendicular to incident beam |
|---|
| 174 | # and only defined for 2theta < 90 |
|---|
| 175 | MuT = 2.*MuR |
|---|
| 176 | T1 = np.exp(-MuT) |
|---|
| 177 | T2 = np.exp(-MuT/npcosd(Tth)) |
|---|
| 178 | Tb = MuT-MuT/npcosd(Tth) |
|---|
| 179 | return (T2-T1)/Tb |
|---|
| 180 | elif 'Tilting' in Geometry: #assumes symmetric tilt so sample plane is parallel to diffraction vector |
|---|
| 181 | MuT = 2.*MuR |
|---|
| 182 | cth = npcosd(Tth/2.0) |
|---|
| 183 | return np.exp(-MuT/cth)/cth |
|---|
| 184 | |
|---|
| 185 | def AbsorbDerv(Geometry,MuR,Tth,Phi=0,Psi=0): |
|---|
| 186 | 'needs a doc string' |
|---|
| 187 | dA = 0.001 |
|---|
| 188 | AbsP = Absorb(Geometry,MuR+dA,Tth,Phi,Psi) |
|---|
| 189 | if MuR: |
|---|
| 190 | AbsM = Absorb(Geometry,MuR-dA,Tth,Phi,Psi) |
|---|
| 191 | return (AbsP-AbsM)/(2.0*dA) |
|---|
| 192 | else: |
|---|
| 193 | return (AbsP-1.)/dA |
|---|
| 194 | |
|---|
| 195 | def Polarization(Pola,Tth,Azm=0.0): |
|---|
| 196 | """ Calculate angle dependent x-ray polarization correction (not scaled correctly!) |
|---|
| 197 | |
|---|
| 198 | :param Pola: polarization coefficient e.g 1.0 fully polarized, 0.5 unpolarized |
|---|
| 199 | :param Azm: azimuthal angle e.g. 0.0 in plane of polarization |
|---|
| 200 | :param Tth: 2-theta scattering angle - can be numpy array |
|---|
| 201 | which (if either) of these is "right"? |
|---|
| 202 | :return: (pola, dpdPola) |
|---|
| 203 | * pola = ((1-Pola)*npcosd(Azm)**2+Pola*npsind(Azm)**2)*npcosd(Tth)**2+ \ |
|---|
| 204 | (1-Pola)*npsind(Azm)**2+Pola*npcosd(Azm)**2 |
|---|
| 205 | * dpdPola: derivative needed for least squares |
|---|
| 206 | |
|---|
| 207 | """ |
|---|
| 208 | pola = ((1.0-Pola)*npcosd(Azm)**2+Pola*npsind(Azm)**2)*npcosd(Tth)**2+ \ |
|---|
| 209 | (1.0-Pola)*npsind(Azm)**2+Pola*npcosd(Azm)**2 |
|---|
| 210 | dpdPola = -npsind(Tth)**2*(npsind(Azm)**2-npcosd(Azm)**2) |
|---|
| 211 | return pola,dpdPola |
|---|
| 212 | |
|---|
| 213 | def Oblique(ObCoeff,Tth): |
|---|
| 214 | 'currently assumes detector is normal to beam' |
|---|
| 215 | if ObCoeff: |
|---|
| 216 | return (1.-ObCoeff)/(1.0-np.exp(np.log(ObCoeff)/npcosd(Tth))) |
|---|
| 217 | else: |
|---|
| 218 | return 1.0 |
|---|
| 219 | |
|---|
| 220 | def Ruland(RulCoff,wave,Q,Compton): |
|---|
| 221 | 'needs a doc string' |
|---|
| 222 | C = 2.9978e8 |
|---|
| 223 | D = 1.5e-3 |
|---|
| 224 | hmc = 0.024262734687 |
|---|
| 225 | sinth2 = (Q*wave/(4.0*np.pi))**2 |
|---|
| 226 | dlam = (wave**2)*Compton*Q/C |
|---|
| 227 | dlam_c = 2.0*hmc*sinth2-D*wave**2 |
|---|
| 228 | return 1.0/((1.0+dlam/RulCoff)*(1.0+(np.pi*dlam_c/(dlam+RulCoff))**2)) |
|---|
| 229 | |
|---|
| 230 | def LorchWeight(Q): |
|---|
| 231 | 'needs a doc string' |
|---|
| 232 | return np.sin(np.pi*(Q[-1]-Q)/(2.0*Q[-1])) |
|---|
| 233 | |
|---|
| 234 | def GetAsfMean(ElList,Sthl2): |
|---|
| 235 | '''Calculate various scattering factor terms for PDF calcs |
|---|
| 236 | |
|---|
| 237 | :param dict ElList: element dictionary contains scattering factor coefficients, etc. |
|---|
| 238 | :param np.array Sthl2: numpy array of sin theta/lambda squared values |
|---|
| 239 | :returns: mean(f^2), mean(f)^2, mean(compton) |
|---|
| 240 | ''' |
|---|
| 241 | sumNoAtoms = 0.0 |
|---|
| 242 | FF = np.zeros_like(Sthl2) |
|---|
| 243 | FF2 = np.zeros_like(Sthl2) |
|---|
| 244 | CF = np.zeros_like(Sthl2) |
|---|
| 245 | for El in ElList: |
|---|
| 246 | sumNoAtoms += ElList[El]['FormulaNo'] |
|---|
| 247 | for El in ElList: |
|---|
| 248 | el = ElList[El] |
|---|
| 249 | ff2 = (G2elem.ScatFac(el,Sthl2)+el['fp'])**2+el['fpp']**2 |
|---|
| 250 | cf = G2elem.ComptonFac(el,Sthl2) |
|---|
| 251 | FF += np.sqrt(ff2)*el['FormulaNo']/sumNoAtoms |
|---|
| 252 | FF2 += ff2*el['FormulaNo']/sumNoAtoms |
|---|
| 253 | CF += cf*el['FormulaNo']/sumNoAtoms |
|---|
| 254 | return FF2,FF**2,CF |
|---|
| 255 | |
|---|
| 256 | def GetNumDensity(ElList,Vol): |
|---|
| 257 | 'needs a doc string' |
|---|
| 258 | sumNoAtoms = 0.0 |
|---|
| 259 | for El in ElList: |
|---|
| 260 | sumNoAtoms += ElList[El]['FormulaNo'] |
|---|
| 261 | return sumNoAtoms/Vol |
|---|
| 262 | |
|---|
| 263 | def CalcPDF(data,inst,xydata): |
|---|
| 264 | 'needs a doc string' |
|---|
| 265 | auxPlot = [] |
|---|
| 266 | import copy |
|---|
| 267 | import scipy.fftpack as ft |
|---|
| 268 | #subtract backgrounds - if any |
|---|
| 269 | xydata['IofQ'] = copy.deepcopy(xydata['Sample']) |
|---|
| 270 | if data['Sample Bkg.']['Name']: |
|---|
| 271 | xydata['IofQ'][1][1] += (xydata['Sample Bkg.'][1][1]+ |
|---|
| 272 | data['Sample Bkg.']['Add'])*data['Sample Bkg.']['Mult'] |
|---|
| 273 | if data['Container']['Name']: |
|---|
| 274 | xycontainer = (xydata['Container'][1][1]+data['Container']['Add'])*data['Container']['Mult'] |
|---|
| 275 | if data['Container Bkg.']['Name']: |
|---|
| 276 | xycontainer += (xydata['Container Bkg.'][1][1]+ |
|---|
| 277 | data['Container Bkg.']['Add'])*data['Container Bkg.']['Mult'] |
|---|
| 278 | xydata['IofQ'][1][1] += xycontainer |
|---|
| 279 | #get element data & absorption coeff. |
|---|
| 280 | ElList = data['ElList'] |
|---|
| 281 | Abs = G2lat.CellAbsorption(ElList,data['Form Vol']) |
|---|
| 282 | #Apply angle dependent corrections |
|---|
| 283 | Tth = xydata['Sample'][1][0] |
|---|
| 284 | dt = (Tth[1]-Tth[0]) |
|---|
| 285 | MuR = Abs*data['Diam']/20.0 |
|---|
| 286 | xydata['IofQ'][1][1] /= Absorb(data['Geometry'],MuR,Tth) |
|---|
| 287 | if 'X' in inst['Type'][0]: |
|---|
| 288 | xydata['IofQ'][1][1] /= Polarization(inst['Polariz.'][1],Tth,Azm=inst['Azimuth'][1])[0] |
|---|
| 289 | if data['DetType'] == 'Image plate': |
|---|
| 290 | xydata['IofQ'][1][1] *= Oblique(data['ObliqCoeff'],Tth) |
|---|
| 291 | XY = xydata['IofQ'][1] |
|---|
| 292 | #convert to Q |
|---|
| 293 | hc = 12.397639 |
|---|
| 294 | wave = G2mth.getWave(inst) |
|---|
| 295 | keV = hc/wave |
|---|
| 296 | minQ = npT2q(Tth[0],wave) |
|---|
| 297 | maxQ = npT2q(Tth[-1],wave) |
|---|
| 298 | Qpoints = np.linspace(0.,maxQ,len(XY[0]),endpoint=True) |
|---|
| 299 | dq = Qpoints[1]-Qpoints[0] |
|---|
| 300 | XY[0] = npT2q(XY[0],wave) |
|---|
| 301 | # Qdata = np.nan_to_num(si.griddata(XY[0],XY[1],Qpoints,method='linear')) #only OK for scipy 0.9! |
|---|
| 302 | T = si.interp1d(XY[0],XY[1],bounds_error=False,fill_value=XY[1][0]) #OK for scipy 0.8 |
|---|
| 303 | Qdata = T(Qpoints) |
|---|
| 304 | Qdata -= np.min(Qdata)*data['BackRatio'] |
|---|
| 305 | |
|---|
| 306 | qLimits = data['QScaleLim'] |
|---|
| 307 | minQ = np.searchsorted(Qpoints,qLimits[0]) |
|---|
| 308 | maxQ = np.searchsorted(Qpoints,qLimits[1]) |
|---|
| 309 | newdata = [] |
|---|
| 310 | xydata['IofQ'][1][0] = Qpoints |
|---|
| 311 | xydata['IofQ'][1][1] = Qdata |
|---|
| 312 | for item in xydata['IofQ'][1]: |
|---|
| 313 | newdata.append(item[:maxQ]) |
|---|
| 314 | xydata['IofQ'][1] = newdata |
|---|
| 315 | |
|---|
| 316 | xydata['SofQ'] = copy.deepcopy(xydata['IofQ']) |
|---|
| 317 | FFSq,SqFF,CF = GetAsfMean(ElList,(xydata['SofQ'][1][0]/(4.0*np.pi))**2) #these are <f^2>,<f>^2,Cf |
|---|
| 318 | Q = xydata['SofQ'][1][0] |
|---|
| 319 | ruland = Ruland(data['Ruland'],wave,Q,CF) |
|---|
| 320 | # auxPlot.append([Q,ruland,'Ruland']) |
|---|
| 321 | CF *= ruland |
|---|
| 322 | # auxPlot.append([Q,CF,'CF-Corr']) |
|---|
| 323 | scale = np.sum((FFSq+CF)[minQ:maxQ])/np.sum(xydata['SofQ'][1][1][minQ:maxQ]) |
|---|
| 324 | xydata['SofQ'][1][1] *= scale |
|---|
| 325 | xydata['SofQ'][1][1] -= CF |
|---|
| 326 | xydata['SofQ'][1][1] = xydata['SofQ'][1][1]/SqFF |
|---|
| 327 | scale = len(xydata['SofQ'][1][1][minQ:maxQ])/np.sum(xydata['SofQ'][1][1][minQ:maxQ]) |
|---|
| 328 | xydata['SofQ'][1][1] *= scale |
|---|
| 329 | # if data.get('sinDamp',False): #not the right thing but leave as place holder |
|---|
| 330 | # sinDamp = np.sin(np.pi*xydata['SofQ'][1][0]/qLimits[1]) |
|---|
| 331 | # xydata['SofQ'][1][1] *= sinDamp |
|---|
| 332 | xydata['FofQ'] = copy.deepcopy(xydata['SofQ']) |
|---|
| 333 | xydata['FofQ'][1][1] = xydata['FofQ'][1][0]*(xydata['SofQ'][1][1]-1.0) |
|---|
| 334 | if data['Lorch']: |
|---|
| 335 | xydata['FofQ'][1][1] *= LorchWeight(Q) |
|---|
| 336 | xydata['GofR'] = copy.deepcopy(xydata['FofQ']) |
|---|
| 337 | nR = len(xydata['GofR'][1][1]) |
|---|
| 338 | xydata['GofR'][1][1] = -dq*np.imag(ft.fft(xydata['FofQ'][1][1],4*nR)[:nR]) |
|---|
| 339 | xydata['GofR'][1][0] = 0.5*np.pi*np.linspace(0,nR,nR)/qLimits[1] |
|---|
| 340 | return auxPlot |
|---|
| 341 | |
|---|
| 342 | def MakeRDF(RDFcontrols,background,inst,pwddata,xydata): |
|---|
| 343 | # GSASIIpath.IPyBreak() |
|---|
| 344 | auxPlot = [] |
|---|
| 345 | if 'C' in inst['Type'][0]: |
|---|
| 346 | Tth = pwddata[0] |
|---|
| 347 | hc = 12.397639 |
|---|
| 348 | wave = G2mth.getWave(inst) |
|---|
| 349 | keV = hc/wave |
|---|
| 350 | minQ = npT2q(Tth[0],wave) |
|---|
| 351 | maxQ = npT2q(Tth[-1],wave) |
|---|
| 352 | |
|---|
| 353 | |
|---|
| 354 | elif 'T' in inst['Type'][0]: |
|---|
| 355 | TOF = pwddata[0] |
|---|
| 356 | difC = inst['difC'][1] |
|---|
| 357 | minQ = 2.*np.pi*difC/TOF[-1] |
|---|
| 358 | maxQ = 2.*np.pi*difC/TOF[0] |
|---|
| 359 | |
|---|
| 360 | Qpoints = np.linspace(0.,maxQ,len(pwddata[0]),endpoint=True) |
|---|
| 361 | dq = Qpoints[1]-Qpoints[0] |
|---|
| 362 | T = si.interp1d(pwddata[0],pwddata[1],bounds_error=False,fill_value=0.0,kind=RDFcontrols['Smooth']) #OK for scipy 0.8 |
|---|
| 363 | Qdata = T(Qpoints) |
|---|
| 364 | auxPlot.append([Qpoints,Qdata,'interp1d:'+RDFcontrols['Smooth']]) |
|---|
| 365 | return auxPlot |
|---|
| 366 | print 'make RDF' |
|---|
| 367 | |
|---|
| 368 | ################################################################################ |
|---|
| 369 | #GSASII peak fitting routines: Finger, Cox & Jephcoat model |
|---|
| 370 | ################################################################################ |
|---|
| 371 | |
|---|
| 372 | def factorize(num): |
|---|
| 373 | ''' Provide prime number factors for integer num |
|---|
| 374 | :returns: dictionary of prime factors (keys) & power for each (data) |
|---|
| 375 | ''' |
|---|
| 376 | factors = {} |
|---|
| 377 | orig = num |
|---|
| 378 | |
|---|
| 379 | # we take advantage of the fact that (i +1)**2 = i**2 + 2*i +1 |
|---|
| 380 | i, sqi = 2, 4 |
|---|
| 381 | while sqi <= num: |
|---|
| 382 | while not num%i: |
|---|
| 383 | num /= i |
|---|
| 384 | factors[i] = factors.get(i, 0) + 1 |
|---|
| 385 | |
|---|
| 386 | sqi += 2*i + 1 |
|---|
| 387 | i += 1 |
|---|
| 388 | |
|---|
| 389 | if num != 1 and num != orig: |
|---|
| 390 | factors[num] = factors.get(num, 0) + 1 |
|---|
| 391 | |
|---|
| 392 | if factors: |
|---|
| 393 | return factors |
|---|
| 394 | else: |
|---|
| 395 | return {num:1} #a prime number! |
|---|
| 396 | |
|---|
| 397 | def makeFFTsizeList(nmin=1,nmax=1023,thresh=15): |
|---|
| 398 | ''' Provide list of optimal data sizes for FFT calculations |
|---|
| 399 | |
|---|
| 400 | :param int nmin: minimum data size >= 1 |
|---|
| 401 | :param int nmax: maximum data size > nmin |
|---|
| 402 | :param int thresh: maximum prime factor allowed |
|---|
| 403 | :Returns: list of data sizes where the maximum prime factor is < thresh |
|---|
| 404 | ''' |
|---|
| 405 | plist = [] |
|---|
| 406 | nmin = max(1,nmin) |
|---|
| 407 | nmax = max(nmin+1,nmax) |
|---|
| 408 | for p in range(nmin,nmax): |
|---|
| 409 | if max(factorize(p).keys()) < thresh: |
|---|
| 410 | plist.append(p) |
|---|
| 411 | return plist |
|---|
| 412 | |
|---|
| 413 | np.seterr(divide='ignore') |
|---|
| 414 | |
|---|
| 415 | # Normal distribution |
|---|
| 416 | |
|---|
| 417 | # loc = mu, scale = std |
|---|
| 418 | _norm_pdf_C = 1./math.sqrt(2*math.pi) |
|---|
| 419 | class norm_gen(st.rv_continuous): |
|---|
| 420 | 'needs a doc string' |
|---|
| 421 | |
|---|
| 422 | def pdf(self,x,*args,**kwds): |
|---|
| 423 | loc,scale=kwds['loc'],kwds['scale'] |
|---|
| 424 | x = (x-loc)/scale |
|---|
| 425 | return np.exp(-x**2/2.0) * _norm_pdf_C / scale |
|---|
| 426 | |
|---|
| 427 | norm = norm_gen(name='norm',longname='A normal',extradoc=""" |
|---|
| 428 | |
|---|
| 429 | Normal distribution |
|---|
| 430 | |
|---|
| 431 | The location (loc) keyword specifies the mean. |
|---|
| 432 | The scale (scale) keyword specifies the standard deviation. |
|---|
| 433 | |
|---|
| 434 | normal.pdf(x) = exp(-x**2/2)/sqrt(2*pi) |
|---|
| 435 | """) |
|---|
| 436 | |
|---|
| 437 | ## Cauchy |
|---|
| 438 | |
|---|
| 439 | # median = loc |
|---|
| 440 | |
|---|
| 441 | class cauchy_gen(st.rv_continuous): |
|---|
| 442 | 'needs a doc string' |
|---|
| 443 | |
|---|
| 444 | def pdf(self,x,*args,**kwds): |
|---|
| 445 | loc,scale=kwds['loc'],kwds['scale'] |
|---|
| 446 | x = (x-loc)/scale |
|---|
| 447 | return 1.0/np.pi/(1.0+x*x) / scale |
|---|
| 448 | |
|---|
| 449 | cauchy = cauchy_gen(name='cauchy',longname='Cauchy',extradoc=""" |
|---|
| 450 | |
|---|
| 451 | Cauchy distribution |
|---|
| 452 | |
|---|
| 453 | cauchy.pdf(x) = 1/(pi*(1+x**2)) |
|---|
| 454 | |
|---|
| 455 | This is the t distribution with one degree of freedom. |
|---|
| 456 | """) |
|---|
| 457 | |
|---|
| 458 | |
|---|
| 459 | #GSASII peak fitting routine: Finger, Cox & Jephcoat model |
|---|
| 460 | |
|---|
| 461 | |
|---|
| 462 | class fcjde_gen(st.rv_continuous): |
|---|
| 463 | """ |
|---|
| 464 | Finger-Cox-Jephcoat D(2phi,2th) function for S/L = H/L |
|---|
| 465 | Ref: J. Appl. Cryst. (1994) 27, 892-900. |
|---|
| 466 | |
|---|
| 467 | :param x: array -1 to 1 |
|---|
| 468 | :param t: 2-theta position of peak |
|---|
| 469 | :param s: sum(S/L,H/L); S: sample height, H: detector opening, |
|---|
| 470 | L: sample to detector opening distance |
|---|
| 471 | :param dx: 2-theta step size in deg |
|---|
| 472 | |
|---|
| 473 | :returns: for fcj.pdf |
|---|
| 474 | |
|---|
| 475 | * T = x*dx+t |
|---|
| 476 | * s = S/L+H/L |
|---|
| 477 | * if x < 0:: |
|---|
| 478 | |
|---|
| 479 | fcj.pdf = [1/sqrt({cos(T)**2/cos(t)**2}-1) - 1/s]/|cos(T)| |
|---|
| 480 | |
|---|
| 481 | * if x >= 0: fcj.pdf = 0 |
|---|
| 482 | """ |
|---|
| 483 | def _pdf(self,x,t,s,dx): |
|---|
| 484 | T = dx*x+t |
|---|
| 485 | ax2 = abs(npcosd(T)) |
|---|
| 486 | ax = ax2**2 |
|---|
| 487 | bx = npcosd(t)**2 |
|---|
| 488 | bx = np.where(ax>bx,bx,ax) |
|---|
| 489 | fx = np.where(ax>bx,(np.sqrt(bx/(ax-bx))-1./s)/ax2,0.0) |
|---|
| 490 | fx = np.where(fx > 0.,fx,0.0) |
|---|
| 491 | return fx |
|---|
| 492 | |
|---|
| 493 | def pdf(self,x,*args,**kwds): |
|---|
| 494 | loc=kwds['loc'] |
|---|
| 495 | return self._pdf(x-loc,*args) |
|---|
| 496 | |
|---|
| 497 | fcjde = fcjde_gen(name='fcjde',shapes='t,s,dx') |
|---|
| 498 | |
|---|
| 499 | def getWidthsCW(pos,sig,gam,shl): |
|---|
| 500 | '''Compute the peak widths used for computing the range of a peak |
|---|
| 501 | for constant wavelength data. On low-angle side, 10 FWHM are used, |
|---|
| 502 | on high-angle side 15 are used, low angle side extended by axial divergence |
|---|
| 503 | (for peaks above 90 deg, these are reversed.) |
|---|
| 504 | ''' |
|---|
| 505 | widths = [np.sqrt(sig)/100.,gam/100.] |
|---|
| 506 | fwhm = 2.355*widths[0]+widths[1] |
|---|
| 507 | fmin = 10.*(fwhm+shl*abs(npcosd(pos))) |
|---|
| 508 | fmax = 15.0*fwhm |
|---|
| 509 | if pos > 90: |
|---|
| 510 | fmin,fmax = [fmax,fmin] |
|---|
| 511 | return widths,fmin,fmax |
|---|
| 512 | |
|---|
| 513 | def getWidthsTOF(pos,alp,bet,sig,gam): |
|---|
| 514 | '''Compute the peak widths used for computing the range of a peak |
|---|
| 515 | for constant wavelength data. 10 FWHM are used on both sides each |
|---|
| 516 | extended by exponential coeff. |
|---|
| 517 | ''' |
|---|
| 518 | lnf = 3.3 # =log(0.001/2) |
|---|
| 519 | widths = [np.sqrt(sig),gam] |
|---|
| 520 | fwhm = 2.355*widths[0]+2.*widths[1] |
|---|
| 521 | fmin = 10.*fwhm*(1.+1./alp) |
|---|
| 522 | fmax = 10.*fwhm*(1.+1./bet) |
|---|
| 523 | return widths,fmin,fmax |
|---|
| 524 | |
|---|
| 525 | def getFWHM(pos,Inst): |
|---|
| 526 | '''Compute total FWHM from Thompson, Cox & Hastings (1987), J. Appl. Cryst. 20, 79-83 |
|---|
| 527 | via getgamFW(g,s). |
|---|
| 528 | |
|---|
| 529 | :param pos: float peak position in deg 2-theta or tof in musec |
|---|
| 530 | :param Inst: dict instrument parameters |
|---|
| 531 | |
|---|
| 532 | :returns float: total FWHM of pseudoVoigt in deg or musec |
|---|
| 533 | ''' |
|---|
| 534 | |
|---|
| 535 | sig = lambda Th,U,V,W: np.sqrt(max(0.001,U*tand(Th)**2+V*tand(Th)+W)) |
|---|
| 536 | sigTOF = lambda dsp,S0,S1,S2,Sq: S0+S1*dsp**2+S2*dsp**4+Sq/dsp**2 |
|---|
| 537 | gam = lambda Th,X,Y: (X/cosd(Th)+Y*tand(Th)) |
|---|
| 538 | gamTOF = lambda dsp,X,Y: X*dsp+Y*dsp**2 |
|---|
| 539 | if 'C' in Inst['Type'][0]: |
|---|
| 540 | s = sig(pos/2.,Inst['U'][1],Inst['V'][1],Inst['W'][1]) |
|---|
| 541 | g = gam(pos/2.,Inst['X'][1],Inst['Y'][1]) |
|---|
| 542 | return getgamFW(g,s)/100. #returns FWHM in deg |
|---|
| 543 | else: |
|---|
| 544 | dsp = pos/Inst['difC'][0] |
|---|
| 545 | s = sigTOF(dsp,Inst['sig-0'][1],Inst['sig-1'][1],Inst['sig-2'][1],Inst['sig-q'][1]) |
|---|
| 546 | g = gamTOF(dsp,Inst['X'][1],Inst['Y'][1]) |
|---|
| 547 | return getgamFW(g,s) |
|---|
| 548 | |
|---|
| 549 | def getgamFW(g,s): |
|---|
| 550 | '''Compute total FWHM from Thompson, Cox & Hastings (1987), J. Appl. Cryst. 20, 79-83 |
|---|
| 551 | lambda fxn needs FWHM for both Gaussian & Lorentzian components |
|---|
| 552 | |
|---|
| 553 | :param g: float Lorentzian gamma = FWHM(L) |
|---|
| 554 | :param s: float Gaussian sig |
|---|
| 555 | |
|---|
| 556 | :returns float: total FWHM of pseudoVoigt |
|---|
| 557 | ''' |
|---|
| 558 | gamFW = lambda s,g: np.exp(np.log(s**5+2.69269*s**4*g+2.42843*s**3*g**2+4.47163*s**2*g**3+0.07842*s*g**4+g**5)/5.) |
|---|
| 559 | return gamFW(2.35482*s,g) #sqrt(8ln2)*sig = FWHM(G) |
|---|
| 560 | |
|---|
| 561 | def getFCJVoigt(pos,intens,sig,gam,shl,xdata): |
|---|
| 562 | 'needs a doc string' |
|---|
| 563 | DX = xdata[1]-xdata[0] |
|---|
| 564 | widths,fmin,fmax = getWidthsCW(pos,sig,gam,shl) |
|---|
| 565 | x = np.linspace(pos-fmin,pos+fmin,256) |
|---|
| 566 | dx = x[1]-x[0] |
|---|
| 567 | Norm = norm.pdf(x,loc=pos,scale=widths[0]) |
|---|
| 568 | Cauchy = cauchy.pdf(x,loc=pos,scale=widths[1]) |
|---|
| 569 | arg = [pos,shl/57.2958,dx,] |
|---|
| 570 | FCJ = fcjde.pdf(x,*arg,loc=pos) |
|---|
| 571 | if len(np.nonzero(FCJ)[0])>5: |
|---|
| 572 | z = np.column_stack([Norm,Cauchy,FCJ]).T |
|---|
| 573 | Z = fft(z) |
|---|
| 574 | Df = ifft(Z.prod(axis=0)).real |
|---|
| 575 | else: |
|---|
| 576 | z = np.column_stack([Norm,Cauchy]).T |
|---|
| 577 | Z = fft(z) |
|---|
| 578 | Df = fftshift(ifft(Z.prod(axis=0))).real |
|---|
| 579 | Df /= np.sum(Df) |
|---|
| 580 | Df = si.interp1d(x,Df,bounds_error=False,fill_value=0.0) |
|---|
| 581 | return intens*Df(xdata)*DX/dx |
|---|
| 582 | |
|---|
| 583 | def getBackground(pfx,parmDict,bakType,dataType,xdata): |
|---|
| 584 | 'needs a doc string' |
|---|
| 585 | if 'T' in dataType: |
|---|
| 586 | q = 2.*np.pi*parmDict[pfx+'difC']/xdata |
|---|
| 587 | elif 'C' in dataType: |
|---|
| 588 | wave = parmDict.get(pfx+'Lam',parmDict.get(pfx+'Lam1',1.0)) |
|---|
| 589 | q = 2.*np.pi*npsind(xdata/2.)/wave |
|---|
| 590 | yb = np.zeros_like(xdata) |
|---|
| 591 | nBak = 0 |
|---|
| 592 | cw = np.diff(xdata) |
|---|
| 593 | cw = np.append(cw,cw[-1]) |
|---|
| 594 | sumBk = [0.,0.,0] |
|---|
| 595 | while True: |
|---|
| 596 | key = pfx+'Back;'+str(nBak) |
|---|
| 597 | if key in parmDict: |
|---|
| 598 | nBak += 1 |
|---|
| 599 | else: |
|---|
| 600 | break |
|---|
| 601 | #empirical functions |
|---|
| 602 | if bakType in ['chebyschev','cosine']: |
|---|
| 603 | dt = xdata[-1]-xdata[0] |
|---|
| 604 | for iBak in range(nBak): |
|---|
| 605 | key = pfx+'Back;'+str(iBak) |
|---|
| 606 | if bakType == 'chebyschev': |
|---|
| 607 | ybi = parmDict[key]*(2.*(xdata-xdata[0])/dt-1.)**iBak |
|---|
| 608 | elif bakType == 'cosine': |
|---|
| 609 | ybi = parmDict[key]*npcosd(xdata*iBak) |
|---|
| 610 | yb += ybi |
|---|
| 611 | sumBk[0] = np.sum(yb) |
|---|
| 612 | elif bakType in ['Q^2 power series','Q^-2 power series']: |
|---|
| 613 | QT = 1. |
|---|
| 614 | yb += np.ones_like(yb)*parmDict[pfx+'Back;0'] |
|---|
| 615 | for iBak in range(nBak-1): |
|---|
| 616 | key = pfx+'Back;'+str(iBak+1) |
|---|
| 617 | if '-2' in bakType: |
|---|
| 618 | QT *= (iBak+1)*q**-2 |
|---|
| 619 | else: |
|---|
| 620 | QT *= q**2/(iBak+1) |
|---|
| 621 | yb += QT*parmDict[key] |
|---|
| 622 | sumBk[0] = np.sum(yb) |
|---|
| 623 | elif bakType in ['lin interpolate','inv interpolate','log interpolate',]: |
|---|
| 624 | if nBak == 1: |
|---|
| 625 | yb = np.ones_like(xdata)*parmDict[pfx+'Back;0'] |
|---|
| 626 | elif nBak == 2: |
|---|
| 627 | dX = xdata[-1]-xdata[0] |
|---|
| 628 | T2 = (xdata-xdata[0])/dX |
|---|
| 629 | T1 = 1.0-T2 |
|---|
| 630 | yb = parmDict[pfx+'Back;0']*T1+parmDict[pfx+'Back;1']*T2 |
|---|
| 631 | else: |
|---|
| 632 | if bakType == 'lin interpolate': |
|---|
| 633 | bakPos = np.linspace(xdata[0],xdata[-1],nBak,True) |
|---|
| 634 | elif bakType == 'inv interpolate': |
|---|
| 635 | bakPos = 1./np.linspace(1./xdata[-1],1./xdata[0],nBak,True) |
|---|
| 636 | elif bakType == 'log interpolate': |
|---|
| 637 | bakPos = np.exp(np.linspace(np.log(xdata[0]),np.log(xdata[-1]),nBak,True)) |
|---|
| 638 | bakPos[0] = xdata[0] |
|---|
| 639 | bakPos[-1] = xdata[-1] |
|---|
| 640 | bakVals = np.zeros(nBak) |
|---|
| 641 | for i in range(nBak): |
|---|
| 642 | bakVals[i] = parmDict[pfx+'Back;'+str(i)] |
|---|
| 643 | bakInt = si.interp1d(bakPos,bakVals,'linear') |
|---|
| 644 | yb = bakInt(ma.getdata(xdata)) |
|---|
| 645 | sumBk[0] = np.sum(yb) |
|---|
| 646 | #Debye function |
|---|
| 647 | if pfx+'difC' in parmDict: |
|---|
| 648 | ff = 1. |
|---|
| 649 | else: |
|---|
| 650 | try: |
|---|
| 651 | wave = parmDict[pfx+'Lam'] |
|---|
| 652 | except KeyError: |
|---|
| 653 | wave = parmDict[pfx+'Lam1'] |
|---|
| 654 | SQ = (q/(4.*np.pi))**2 |
|---|
| 655 | FF = G2elem.GetFormFactorCoeff('Si')[0] |
|---|
| 656 | ff = np.array(G2elem.ScatFac(FF,SQ)[0])**2 |
|---|
| 657 | iD = 0 |
|---|
| 658 | while True: |
|---|
| 659 | try: |
|---|
| 660 | dbA = parmDict[pfx+'DebyeA;'+str(iD)] |
|---|
| 661 | dbR = parmDict[pfx+'DebyeR;'+str(iD)] |
|---|
| 662 | dbU = parmDict[pfx+'DebyeU;'+str(iD)] |
|---|
| 663 | ybi = ff*dbA*np.sin(q*dbR)*np.exp(-dbU*q**2)/(q*dbR) |
|---|
| 664 | yb += ybi |
|---|
| 665 | sumBk[1] += np.sum(ybi) |
|---|
| 666 | iD += 1 |
|---|
| 667 | except KeyError: |
|---|
| 668 | break |
|---|
| 669 | #peaks |
|---|
| 670 | iD = 0 |
|---|
| 671 | while True: |
|---|
| 672 | try: |
|---|
| 673 | pkP = parmDict[pfx+'BkPkpos;'+str(iD)] |
|---|
| 674 | pkI = parmDict[pfx+'BkPkint;'+str(iD)] |
|---|
| 675 | pkS = parmDict[pfx+'BkPksig;'+str(iD)] |
|---|
| 676 | pkG = parmDict[pfx+'BkPkgam;'+str(iD)] |
|---|
| 677 | if 'C' in dataType: |
|---|
| 678 | Wd,fmin,fmax = getWidthsCW(pkP,pkS,pkG,.002) |
|---|
| 679 | else: #'T'OF |
|---|
| 680 | Wd,fmin,fmax = getWidthsTOF(pkP,1.,1.,pkS,pkG) |
|---|
| 681 | iBeg = np.searchsorted(xdata,pkP-fmin) |
|---|
| 682 | iFin = np.searchsorted(xdata,pkP+fmax) |
|---|
| 683 | lenX = len(xdata) |
|---|
| 684 | if not iBeg: |
|---|
| 685 | iFin = np.searchsorted(xdata,pkP+fmax) |
|---|
| 686 | elif iBeg == lenX: |
|---|
| 687 | iFin = iBeg |
|---|
| 688 | else: |
|---|
| 689 | iFin = np.searchsorted(xdata,pkP+fmax) |
|---|
| 690 | if 'C' in dataType: |
|---|
| 691 | ybi = pkI*getFCJVoigt3(pkP,pkS,pkG,0.002,xdata[iBeg:iFin]) |
|---|
| 692 | yb[iBeg:iFin] += ybi |
|---|
| 693 | else: #'T'OF |
|---|
| 694 | ybi = pkI*getEpsVoigt(pkP,1.,1.,pkS,pkG,xdata[iBeg:iFin]) |
|---|
| 695 | yb[iBeg:iFin] += ybi |
|---|
| 696 | sumBk[2] += np.sum(ybi) |
|---|
| 697 | iD += 1 |
|---|
| 698 | except KeyError: |
|---|
| 699 | break |
|---|
| 700 | except ValueError: |
|---|
| 701 | print '**** WARNING - backround peak '+str(iD)+' sigma is negative; fix & try again ****' |
|---|
| 702 | break |
|---|
| 703 | return yb,sumBk |
|---|
| 704 | |
|---|
| 705 | def getBackgroundDerv(hfx,parmDict,bakType,dataType,xdata): |
|---|
| 706 | 'needs a doc string' |
|---|
| 707 | if 'T' in dataType: |
|---|
| 708 | q = 2.*np.pi*parmDict[hfx+'difC']/xdata |
|---|
| 709 | elif 'C' in dataType: |
|---|
| 710 | wave = parmDict.get(hfx+'Lam',parmDict.get(hfx+'Lam1',1.0)) |
|---|
| 711 | q = 2.*np.pi*npsind(xdata/2.)/wave |
|---|
| 712 | nBak = 0 |
|---|
| 713 | while True: |
|---|
| 714 | key = hfx+'Back;'+str(nBak) |
|---|
| 715 | if key in parmDict: |
|---|
| 716 | nBak += 1 |
|---|
| 717 | else: |
|---|
| 718 | break |
|---|
| 719 | dydb = np.zeros(shape=(nBak,len(xdata))) |
|---|
| 720 | dyddb = np.zeros(shape=(3*parmDict[hfx+'nDebye'],len(xdata))) |
|---|
| 721 | dydpk = np.zeros(shape=(4*parmDict[hfx+'nPeaks'],len(xdata))) |
|---|
| 722 | cw = np.diff(xdata) |
|---|
| 723 | cw = np.append(cw,cw[-1]) |
|---|
| 724 | |
|---|
| 725 | if bakType in ['chebyschev','cosine']: |
|---|
| 726 | dt = xdata[-1]-xdata[0] |
|---|
| 727 | for iBak in range(nBak): |
|---|
| 728 | if bakType == 'chebyschev': |
|---|
| 729 | dydb[iBak] = (2.*(xdata-xdata[0])/dt-1.)**iBak |
|---|
| 730 | elif bakType == 'cosine': |
|---|
| 731 | dydb[iBak] = npcosd(xdata*iBak) |
|---|
| 732 | elif bakType in ['Q^2 power series','Q^-2 power series']: |
|---|
| 733 | QT = 1. |
|---|
| 734 | dydb[0] = np.ones_like(xdata) |
|---|
| 735 | for iBak in range(nBak-1): |
|---|
| 736 | if '-2' in bakType: |
|---|
| 737 | QT *= (iBak+1)*q**-2 |
|---|
| 738 | else: |
|---|
| 739 | QT *= q**2/(iBak+1) |
|---|
| 740 | dydb[iBak+1] = QT |
|---|
| 741 | elif bakType in ['lin interpolate','inv interpolate','log interpolate',]: |
|---|
| 742 | if nBak == 1: |
|---|
| 743 | dydb[0] = np.ones_like(xdata) |
|---|
| 744 | elif nBak == 2: |
|---|
| 745 | dX = xdata[-1]-xdata[0] |
|---|
| 746 | T2 = (xdata-xdata[0])/dX |
|---|
| 747 | T1 = 1.0-T2 |
|---|
| 748 | dydb = [T1,T2] |
|---|
| 749 | else: |
|---|
| 750 | if bakType == 'lin interpolate': |
|---|
| 751 | bakPos = np.linspace(xdata[0],xdata[-1],nBak,True) |
|---|
| 752 | elif bakType == 'inv interpolate': |
|---|
| 753 | bakPos = 1./np.linspace(1./xdata[-1],1./xdata[0],nBak,True) |
|---|
| 754 | elif bakType == 'log interpolate': |
|---|
| 755 | bakPos = np.exp(np.linspace(np.log(xdata[0]),np.log(xdata[-1]),nBak,True)) |
|---|
| 756 | bakPos[0] = xdata[0] |
|---|
| 757 | bakPos[-1] = xdata[-1] |
|---|
| 758 | for i,pos in enumerate(bakPos): |
|---|
| 759 | if i == 0: |
|---|
| 760 | dydb[0] = np.where(xdata<bakPos[1],(bakPos[1]-xdata)/(bakPos[1]-bakPos[0]),0.) |
|---|
| 761 | elif i == len(bakPos)-1: |
|---|
| 762 | dydb[i] = np.where(xdata>bakPos[-2],(bakPos[-1]-xdata)/(bakPos[-1]-bakPos[-2]),0.) |
|---|
| 763 | else: |
|---|
| 764 | dydb[i] = np.where(xdata>bakPos[i], |
|---|
| 765 | np.where(xdata<bakPos[i+1],(bakPos[i+1]-xdata)/(bakPos[i+1]-bakPos[i]),0.), |
|---|
| 766 | np.where(xdata>bakPos[i-1],(xdata-bakPos[i-1])/(bakPos[i]-bakPos[i-1]),0.)) |
|---|
| 767 | if hfx+'difC' in parmDict: |
|---|
| 768 | ff = 1. |
|---|
| 769 | else: |
|---|
| 770 | try: |
|---|
| 771 | wave = parmDict[hfx+'Lam'] |
|---|
| 772 | except KeyError: |
|---|
| 773 | wave = parmDict[hfx+'Lam1'] |
|---|
| 774 | q = 4.0*np.pi*npsind(xdata/2.0)/wave |
|---|
| 775 | SQ = (q/(4*np.pi))**2 |
|---|
| 776 | FF = G2elem.GetFormFactorCoeff('Si')[0] |
|---|
| 777 | ff = np.array(G2elem.ScatFac(FF,SQ)[0]) |
|---|
| 778 | iD = 0 |
|---|
| 779 | while True: |
|---|
| 780 | try: |
|---|
| 781 | if hfx+'difC' in parmDict: |
|---|
| 782 | q = 2*np.pi*parmDict[hfx+'difC']/xdata |
|---|
| 783 | dbA = parmDict[hfx+'DebyeA;'+str(iD)] |
|---|
| 784 | dbR = parmDict[hfx+'DebyeR;'+str(iD)] |
|---|
| 785 | dbU = parmDict[hfx+'DebyeU;'+str(iD)] |
|---|
| 786 | sqr = np.sin(q*dbR)/(q*dbR) |
|---|
| 787 | cqr = np.cos(q*dbR) |
|---|
| 788 | temp = np.exp(-dbU*q**2) |
|---|
| 789 | dyddb[3*iD] = ff*sqr*temp/(np.pi*cw) |
|---|
| 790 | dyddb[3*iD+1] = ff*dbA*temp*(cqr-sqr)/(np.pi*dbR*cw) |
|---|
| 791 | dyddb[3*iD+2] = -ff*dbA*sqr*temp*q**2/(np.pi*cw) |
|---|
| 792 | iD += 1 #ff*dbA*np.sin(q*dbR)*np.exp(-dbU*q**2)/(q*dbR) |
|---|
| 793 | except KeyError: |
|---|
| 794 | break |
|---|
| 795 | iD = 0 |
|---|
| 796 | while True: |
|---|
| 797 | try: |
|---|
| 798 | pkP = parmDict[hfx+'BkPkpos;'+str(iD)] |
|---|
| 799 | pkI = parmDict[hfx+'BkPkint;'+str(iD)] |
|---|
| 800 | pkS = parmDict[hfx+'BkPksig;'+str(iD)] |
|---|
| 801 | pkG = parmDict[hfx+'BkPkgam;'+str(iD)] |
|---|
| 802 | if 'C' in dataType: |
|---|
| 803 | Wd,fmin,fmax = getWidthsCW(pkP,pkS,pkG,.002) |
|---|
| 804 | else: #'T'OF |
|---|
| 805 | Wd,fmin,fmax = getWidthsTOF(pkP,1.,1.,pkS,pkG) |
|---|
| 806 | iBeg = np.searchsorted(xdata,pkP-fmin) |
|---|
| 807 | iFin = np.searchsorted(xdata,pkP+fmax) |
|---|
| 808 | lenX = len(xdata) |
|---|
| 809 | if not iBeg: |
|---|
| 810 | iFin = np.searchsorted(xdata,pkP+fmax) |
|---|
| 811 | elif iBeg == lenX: |
|---|
| 812 | iFin = iBeg |
|---|
| 813 | else: |
|---|
| 814 | iFin = np.searchsorted(xdata,pkP+fmax) |
|---|
| 815 | if 'C' in dataType: |
|---|
| 816 | Df,dFdp,dFds,dFdg,x = getdFCJVoigt3(pkP,pkS,pkG,.002,xdata[iBeg:iFin]) |
|---|
| 817 | dydpk[4*iD][iBeg:iFin] += 100.*cw[iBeg:iFin]*pkI*dFdp |
|---|
| 818 | dydpk[4*iD+1][iBeg:iFin] += 100.*cw[iBeg:iFin]*Df |
|---|
| 819 | dydpk[4*iD+2][iBeg:iFin] += 100.*cw[iBeg:iFin]*pkI*dFds |
|---|
| 820 | dydpk[4*iD+3][iBeg:iFin] += 100.*cw[iBeg:iFin]*pkI*dFdg |
|---|
| 821 | else: #'T'OF |
|---|
| 822 | Df,dFdp,x,x,dFds,dFdg = getdEpsVoigt(pkP,1.,1.,pkS,pkG,xdata[iBeg:iFin]) |
|---|
| 823 | dydpk[4*iD][iBeg:iFin] += pkI*dFdp |
|---|
| 824 | dydpk[4*iD+1][iBeg:iFin] += Df |
|---|
| 825 | dydpk[4*iD+2][iBeg:iFin] += pkI*dFds |
|---|
| 826 | dydpk[4*iD+3][iBeg:iFin] += pkI*dFdg |
|---|
| 827 | iD += 1 |
|---|
| 828 | except KeyError: |
|---|
| 829 | break |
|---|
| 830 | except ValueError: |
|---|
| 831 | print '**** WARNING - backround peak '+str(iD)+' sigma is negative; fix & try again ****' |
|---|
| 832 | break |
|---|
| 833 | return dydb,dyddb,dydpk |
|---|
| 834 | |
|---|
| 835 | #use old fortran routine |
|---|
| 836 | def getFCJVoigt3(pos,sig,gam,shl,xdata): |
|---|
| 837 | 'needs a doc string' |
|---|
| 838 | |
|---|
| 839 | Df = pyd.pypsvfcj(len(xdata),xdata-pos,pos,sig,gam,shl) |
|---|
| 840 | # Df = pyd.pypsvfcjo(len(xdata),xdata-pos,pos,sig,gam,shl) |
|---|
| 841 | Df /= np.sum(Df) |
|---|
| 842 | return Df |
|---|
| 843 | |
|---|
| 844 | def getdFCJVoigt3(pos,sig,gam,shl,xdata): |
|---|
| 845 | 'needs a doc string' |
|---|
| 846 | |
|---|
| 847 | Df,dFdp,dFds,dFdg,dFdsh = pyd.pydpsvfcj(len(xdata),xdata-pos,pos,sig,gam,shl) |
|---|
| 848 | # Df,dFdp,dFds,dFdg,dFdsh = pyd.pydpsvfcjo(len(xdata),xdata-pos,pos,sig,gam,shl) |
|---|
| 849 | sumDf = np.sum(Df) |
|---|
| 850 | return Df,dFdp,dFds,dFdg,dFdsh |
|---|
| 851 | |
|---|
| 852 | def getPsVoigt(pos,sig,gam,xdata): |
|---|
| 853 | 'needs a doc string' |
|---|
| 854 | |
|---|
| 855 | Df = pyd.pypsvoigt(len(xdata),xdata-pos,sig,gam) |
|---|
| 856 | Df /= np.sum(Df) |
|---|
| 857 | return Df |
|---|
| 858 | |
|---|
| 859 | def getdPsVoigt(pos,sig,gam,xdata): |
|---|
| 860 | 'needs a doc string' |
|---|
| 861 | |
|---|
| 862 | Df,dFdp,dFds,dFdg = pyd.pydpsvoigt(len(xdata),xdata-pos,sig,gam) |
|---|
| 863 | sumDf = np.sum(Df) |
|---|
| 864 | return Df,dFdp,dFds,dFdg |
|---|
| 865 | |
|---|
| 866 | def getEpsVoigt(pos,alp,bet,sig,gam,xdata): |
|---|
| 867 | 'needs a doc string' |
|---|
| 868 | Df = pyd.pyepsvoigt(len(xdata),xdata-pos,alp,bet,sig,gam) |
|---|
| 869 | Df /= np.sum(Df) |
|---|
| 870 | return Df |
|---|
| 871 | |
|---|
| 872 | def getdEpsVoigt(pos,alp,bet,sig,gam,xdata): |
|---|
| 873 | 'needs a doc string' |
|---|
| 874 | Df,dFdp,dFda,dFdb,dFds,dFdg = pyd.pydepsvoigt(len(xdata),xdata-pos,alp,bet,sig,gam) |
|---|
| 875 | sumDf = np.sum(Df) |
|---|
| 876 | return Df,dFdp,dFda,dFdb,dFds,dFdg |
|---|
| 877 | |
|---|
| 878 | def ellipseSize(H,Sij,GB): |
|---|
| 879 | 'needs a doc string' |
|---|
| 880 | HX = np.inner(H.T,GB) |
|---|
| 881 | lenHX = np.sqrt(np.sum(HX**2)) |
|---|
| 882 | Esize,Rsize = nl.eigh(G2lat.U6toUij(Sij)) |
|---|
| 883 | R = np.inner(HX/lenHX,Rsize)*Esize #want column length for hkl in crystal |
|---|
| 884 | lenR = np.sqrt(np.sum(R**2)) |
|---|
| 885 | return lenR |
|---|
| 886 | |
|---|
| 887 | def ellipseSizeDerv(H,Sij,GB): |
|---|
| 888 | 'needs a doc string' |
|---|
| 889 | lenR = ellipseSize(H,Sij,GB) |
|---|
| 890 | delt = 0.001 |
|---|
| 891 | dRdS = np.zeros(6) |
|---|
| 892 | for i in range(6): |
|---|
| 893 | Sij[i] -= delt |
|---|
| 894 | lenM = ellipseSize(H,Sij,GB) |
|---|
| 895 | Sij[i] += 2.*delt |
|---|
| 896 | lenP = ellipseSize(H,Sij,GB) |
|---|
| 897 | Sij[i] -= delt |
|---|
| 898 | dRdS[i] = (lenP-lenM)/(2.*delt) |
|---|
| 899 | return lenR,dRdS |
|---|
| 900 | |
|---|
| 901 | def getHKLpeak(dmin,SGData,A,Inst=None): |
|---|
| 902 | 'needs a doc string' |
|---|
| 903 | HKL = G2lat.GenHLaue(dmin,SGData,A) |
|---|
| 904 | HKLs = [] |
|---|
| 905 | for h,k,l,d in HKL: |
|---|
| 906 | ext = G2spc.GenHKLf([h,k,l],SGData)[0] |
|---|
| 907 | if not ext: |
|---|
| 908 | if Inst == None: |
|---|
| 909 | HKLs.append([h,k,l,d,0,-1]) |
|---|
| 910 | else: |
|---|
| 911 | HKLs.append([h,k,l,d,G2lat.Dsp2pos(Inst,d),-1]) |
|---|
| 912 | return HKLs |
|---|
| 913 | |
|---|
| 914 | def getHKLMpeak(dmin,Inst,SGData,SSGData,Vec,maxH,A): |
|---|
| 915 | 'needs a doc string' |
|---|
| 916 | HKLs = [] |
|---|
| 917 | vec = np.array(Vec) |
|---|
| 918 | vstar = np.sqrt(G2lat.calc_rDsq(vec,A)) #find extra needed for -n SS reflections |
|---|
| 919 | dvec = 1./(maxH*vstar+1./dmin) |
|---|
| 920 | HKL = G2lat.GenHLaue(dvec,SGData,A) |
|---|
| 921 | SSdH = [vec*h for h in range(-maxH,maxH+1)] |
|---|
| 922 | SSdH = dict(zip(range(-maxH,maxH+1),SSdH)) |
|---|
| 923 | for h,k,l,d in HKL: |
|---|
| 924 | ext = G2spc.GenHKLf([h,k,l],SGData)[0] |
|---|
| 925 | if not ext and d >= dmin: |
|---|
| 926 | HKLs.append([h,k,l,0,d,G2lat.Dsp2pos(Inst,d),-1]) |
|---|
| 927 | for dH in SSdH: |
|---|
| 928 | if dH: |
|---|
| 929 | DH = SSdH[dH] |
|---|
| 930 | H = [h+DH[0],k+DH[1],l+DH[2]] |
|---|
| 931 | d = float(1/np.sqrt(G2lat.calc_rDsq(H,A))) |
|---|
| 932 | if d >= dmin: |
|---|
| 933 | HKLM = np.array([h,k,l,dH]) |
|---|
| 934 | if G2spc.checkSSextc(HKLM,SSGData): |
|---|
| 935 | HKLs.append([h,k,l,dH,d,G2lat.Dsp2pos(Inst,d),-1]) |
|---|
| 936 | # GSASIIpath.IPyBreak() |
|---|
| 937 | return G2lat.sortHKLd(HKLs,True,True,True) |
|---|
| 938 | |
|---|
| 939 | def getPeakProfile(dataType,parmDict,xdata,varyList,bakType): |
|---|
| 940 | 'needs a doc string' |
|---|
| 941 | |
|---|
| 942 | yb = getBackground('',parmDict,bakType,dataType,xdata)[0] |
|---|
| 943 | yc = np.zeros_like(yb) |
|---|
| 944 | cw = np.diff(xdata) |
|---|
| 945 | cw = np.append(cw,cw[-1]) |
|---|
| 946 | if 'C' in dataType: |
|---|
| 947 | shl = max(parmDict['SH/L'],0.002) |
|---|
| 948 | Ka2 = False |
|---|
| 949 | if 'Lam1' in parmDict.keys(): |
|---|
| 950 | Ka2 = True |
|---|
| 951 | lamRatio = 360*(parmDict['Lam2']-parmDict['Lam1'])/(np.pi*parmDict['Lam1']) |
|---|
| 952 | kRatio = parmDict['I(L2)/I(L1)'] |
|---|
| 953 | iPeak = 0 |
|---|
| 954 | while True: |
|---|
| 955 | try: |
|---|
| 956 | pos = parmDict['pos'+str(iPeak)] |
|---|
| 957 | tth = (pos-parmDict['Zero']) |
|---|
| 958 | intens = parmDict['int'+str(iPeak)] |
|---|
| 959 | sigName = 'sig'+str(iPeak) |
|---|
| 960 | if sigName in varyList: |
|---|
| 961 | sig = parmDict[sigName] |
|---|
| 962 | else: |
|---|
| 963 | sig = G2mth.getCWsig(parmDict,tth) |
|---|
| 964 | sig = max(sig,0.001) #avoid neg sigma^2 |
|---|
| 965 | gamName = 'gam'+str(iPeak) |
|---|
| 966 | if gamName in varyList: |
|---|
| 967 | gam = parmDict[gamName] |
|---|
| 968 | else: |
|---|
| 969 | gam = G2mth.getCWgam(parmDict,tth) |
|---|
| 970 | gam = max(gam,0.001) #avoid neg gamma |
|---|
| 971 | Wd,fmin,fmax = getWidthsCW(pos,sig,gam,shl) |
|---|
| 972 | iBeg = np.searchsorted(xdata,pos-fmin) |
|---|
| 973 | iFin = np.searchsorted(xdata,pos+fmin) |
|---|
| 974 | if not iBeg+iFin: #peak below low limit |
|---|
| 975 | iPeak += 1 |
|---|
| 976 | continue |
|---|
| 977 | elif not iBeg-iFin: #peak above high limit |
|---|
| 978 | return yb+yc |
|---|
| 979 | yc[iBeg:iFin] += intens*getFCJVoigt3(pos,sig,gam,shl,xdata[iBeg:iFin]) |
|---|
| 980 | if Ka2: |
|---|
| 981 | pos2 = pos+lamRatio*tand(pos/2.0) # + 360/pi * Dlam/lam * tan(th) |
|---|
| 982 | iBeg = np.searchsorted(xdata,pos2-fmin) |
|---|
| 983 | iFin = np.searchsorted(xdata,pos2+fmin) |
|---|
| 984 | if iBeg-iFin: |
|---|
| 985 | yc[iBeg:iFin] += intens*kRatio*getFCJVoigt3(pos2,sig,gam,shl,xdata[iBeg:iFin]) |
|---|
| 986 | iPeak += 1 |
|---|
| 987 | except KeyError: #no more peaks to process |
|---|
| 988 | return yb+yc |
|---|
| 989 | else: |
|---|
| 990 | Pdabc = parmDict['Pdabc'] |
|---|
| 991 | difC = parmDict['difC'] |
|---|
| 992 | iPeak = 0 |
|---|
| 993 | while True: |
|---|
| 994 | try: |
|---|
| 995 | pos = parmDict['pos'+str(iPeak)] |
|---|
| 996 | tof = pos-parmDict['Zero'] |
|---|
| 997 | dsp = tof/difC |
|---|
| 998 | intens = parmDict['int'+str(iPeak)] |
|---|
| 999 | alpName = 'alp'+str(iPeak) |
|---|
| 1000 | if alpName in varyList: |
|---|
| 1001 | alp = parmDict[alpName] |
|---|
| 1002 | else: |
|---|
| 1003 | if len(Pdabc): |
|---|
| 1004 | alp = np.interp(dsp,Pdabc[0],Pdabc[1]) |
|---|
| 1005 | else: |
|---|
| 1006 | alp = G2mth.getTOFalpha(parmDict,dsp) |
|---|
| 1007 | alp = max(0.0001,alp) |
|---|
| 1008 | betName = 'bet'+str(iPeak) |
|---|
| 1009 | if betName in varyList: |
|---|
| 1010 | bet = parmDict[betName] |
|---|
| 1011 | else: |
|---|
| 1012 | if len(Pdabc): |
|---|
| 1013 | bet = np.interp(dsp,Pdabc[0],Pdabc[2]) |
|---|
| 1014 | else: |
|---|
| 1015 | bet = G2mth.getTOFbeta(parmDict,dsp) |
|---|
| 1016 | bet = max(0.0001,bet) |
|---|
| 1017 | sigName = 'sig'+str(iPeak) |
|---|
| 1018 | if sigName in varyList: |
|---|
| 1019 | sig = parmDict[sigName] |
|---|
| 1020 | else: |
|---|
| 1021 | sig = G2mth.getTOFsig(parmDict,dsp) |
|---|
| 1022 | gamName = 'gam'+str(iPeak) |
|---|
| 1023 | if gamName in varyList: |
|---|
| 1024 | gam = parmDict[gamName] |
|---|
| 1025 | else: |
|---|
| 1026 | gam = G2mth.getTOFgamma(parmDict,dsp) |
|---|
| 1027 | gam = max(gam,0.001) #avoid neg gamma |
|---|
| 1028 | Wd,fmin,fmax = getWidthsTOF(pos,alp,bet,sig,gam) |
|---|
| 1029 | iBeg = np.searchsorted(xdata,pos-fmin) |
|---|
| 1030 | iFin = np.searchsorted(xdata,pos+fmax) |
|---|
| 1031 | lenX = len(xdata) |
|---|
| 1032 | if not iBeg: |
|---|
| 1033 | iFin = np.searchsorted(xdata,pos+fmax) |
|---|
| 1034 | elif iBeg == lenX: |
|---|
| 1035 | iFin = iBeg |
|---|
| 1036 | else: |
|---|
| 1037 | iFin = np.searchsorted(xdata,pos+fmax) |
|---|
| 1038 | if not iBeg+iFin: #peak below low limit |
|---|
| 1039 | iPeak += 1 |
|---|
| 1040 | continue |
|---|
| 1041 | elif not iBeg-iFin: #peak above high limit |
|---|
| 1042 | return yb+yc |
|---|
| 1043 | yc[iBeg:iFin] += intens*getEpsVoigt(pos,alp,bet,sig,gam,xdata[iBeg:iFin]) |
|---|
| 1044 | iPeak += 1 |
|---|
| 1045 | except KeyError: #no more peaks to process |
|---|
| 1046 | return yb+yc |
|---|
| 1047 | |
|---|
| 1048 | def getPeakProfileDerv(dataType,parmDict,xdata,varyList,bakType): |
|---|
| 1049 | 'needs a doc string' |
|---|
| 1050 | # needs to return np.array([dMdx1,dMdx2,...]) in same order as varylist = backVary,insVary,peakVary order |
|---|
| 1051 | dMdv = np.zeros(shape=(len(varyList),len(xdata))) |
|---|
| 1052 | dMdb,dMddb,dMdpk = getBackgroundDerv('',parmDict,bakType,dataType,xdata) |
|---|
| 1053 | if 'Back;0' in varyList: #background derivs are in front if present |
|---|
| 1054 | dMdv[0:len(dMdb)] = dMdb |
|---|
| 1055 | names = ['DebyeA','DebyeR','DebyeU'] |
|---|
| 1056 | for name in varyList: |
|---|
| 1057 | if 'Debye' in name: |
|---|
| 1058 | parm,id = name.split(';') |
|---|
| 1059 | ip = names.index(parm) |
|---|
| 1060 | dMdv[varyList.index(name)] = dMddb[3*int(id)+ip] |
|---|
| 1061 | names = ['BkPkpos','BkPkint','BkPksig','BkPkgam'] |
|---|
| 1062 | for name in varyList: |
|---|
| 1063 | if 'BkPk' in name: |
|---|
| 1064 | parm,id = name.split(';') |
|---|
| 1065 | ip = names.index(parm) |
|---|
| 1066 | dMdv[varyList.index(name)] = dMdpk[4*int(id)+ip] |
|---|
| 1067 | cw = np.diff(xdata) |
|---|
| 1068 | cw = np.append(cw,cw[-1]) |
|---|
| 1069 | if 'C' in dataType: |
|---|
| 1070 | shl = max(parmDict['SH/L'],0.002) |
|---|
| 1071 | Ka2 = False |
|---|
| 1072 | if 'Lam1' in parmDict.keys(): |
|---|
| 1073 | Ka2 = True |
|---|
| 1074 | lamRatio = 360*(parmDict['Lam2']-parmDict['Lam1'])/(np.pi*parmDict['Lam1']) |
|---|
| 1075 | kRatio = parmDict['I(L2)/I(L1)'] |
|---|
| 1076 | iPeak = 0 |
|---|
| 1077 | while True: |
|---|
| 1078 | try: |
|---|
| 1079 | pos = parmDict['pos'+str(iPeak)] |
|---|
| 1080 | tth = (pos-parmDict['Zero']) |
|---|
| 1081 | intens = parmDict['int'+str(iPeak)] |
|---|
| 1082 | sigName = 'sig'+str(iPeak) |
|---|
| 1083 | if sigName in varyList: |
|---|
| 1084 | sig = parmDict[sigName] |
|---|
| 1085 | dsdU = dsdV = dsdW = 0 |
|---|
| 1086 | else: |
|---|
| 1087 | sig = G2mth.getCWsig(parmDict,tth) |
|---|
| 1088 | dsdU,dsdV,dsdW = G2mth.getCWsigDeriv(tth) |
|---|
| 1089 | sig = max(sig,0.001) #avoid neg sigma |
|---|
| 1090 | gamName = 'gam'+str(iPeak) |
|---|
| 1091 | if gamName in varyList: |
|---|
| 1092 | gam = parmDict[gamName] |
|---|
| 1093 | dgdX = dgdY = 0 |
|---|
| 1094 | else: |
|---|
| 1095 | gam = G2mth.getCWgam(parmDict,tth) |
|---|
| 1096 | dgdX,dgdY = G2mth.getCWgamDeriv(tth) |
|---|
| 1097 | gam = max(gam,0.001) #avoid neg gamma |
|---|
| 1098 | Wd,fmin,fmax = getWidthsCW(pos,sig,gam,shl) |
|---|
| 1099 | iBeg = np.searchsorted(xdata,pos-fmin) |
|---|
| 1100 | iFin = np.searchsorted(xdata,pos+fmin) |
|---|
| 1101 | if not iBeg+iFin: #peak below low limit |
|---|
| 1102 | iPeak += 1 |
|---|
| 1103 | continue |
|---|
| 1104 | elif not iBeg-iFin: #peak above high limit |
|---|
| 1105 | break |
|---|
| 1106 | dMdpk = np.zeros(shape=(6,len(xdata))) |
|---|
| 1107 | dMdipk = getdFCJVoigt3(pos,sig,gam,shl,xdata[iBeg:iFin]) |
|---|
| 1108 | for i in range(1,5): |
|---|
| 1109 | dMdpk[i][iBeg:iFin] += 100.*cw[iBeg:iFin]*intens*dMdipk[i] |
|---|
| 1110 | dMdpk[0][iBeg:iFin] += 100.*cw[iBeg:iFin]*dMdipk[0] |
|---|
| 1111 | dervDict = {'int':dMdpk[0],'pos':dMdpk[1],'sig':dMdpk[2],'gam':dMdpk[3],'shl':dMdpk[4]} |
|---|
| 1112 | if Ka2: |
|---|
| 1113 | pos2 = pos+lamRatio*tand(pos/2.0) # + 360/pi * Dlam/lam * tan(th) |
|---|
| 1114 | iBeg = np.searchsorted(xdata,pos2-fmin) |
|---|
| 1115 | iFin = np.searchsorted(xdata,pos2+fmin) |
|---|
| 1116 | if iBeg-iFin: |
|---|
| 1117 | dMdipk2 = getdFCJVoigt3(pos2,sig,gam,shl,xdata[iBeg:iFin]) |
|---|
| 1118 | for i in range(1,5): |
|---|
| 1119 | dMdpk[i][iBeg:iFin] += 100.*cw[iBeg:iFin]*intens*kRatio*dMdipk2[i] |
|---|
| 1120 | dMdpk[0][iBeg:iFin] += 100.*cw[iBeg:iFin]*kRatio*dMdipk2[0] |
|---|
| 1121 | dMdpk[5][iBeg:iFin] += 100.*cw[iBeg:iFin]*dMdipk2[0] |
|---|
| 1122 | dervDict = {'int':dMdpk[0],'pos':dMdpk[1],'sig':dMdpk[2],'gam':dMdpk[3],'shl':dMdpk[4],'L1/L2':dMdpk[5]*intens} |
|---|
| 1123 | for parmName in ['pos','int','sig','gam']: |
|---|
| 1124 | try: |
|---|
| 1125 | idx = varyList.index(parmName+str(iPeak)) |
|---|
| 1126 | dMdv[idx] = dervDict[parmName] |
|---|
| 1127 | except ValueError: |
|---|
| 1128 | pass |
|---|
| 1129 | if 'U' in varyList: |
|---|
| 1130 | dMdv[varyList.index('U')] += dsdU*dervDict['sig'] |
|---|
| 1131 | if 'V' in varyList: |
|---|
| 1132 | dMdv[varyList.index('V')] += dsdV*dervDict['sig'] |
|---|
| 1133 | if 'W' in varyList: |
|---|
| 1134 | dMdv[varyList.index('W')] += dsdW*dervDict['sig'] |
|---|
| 1135 | if 'X' in varyList: |
|---|
| 1136 | dMdv[varyList.index('X')] += dgdX*dervDict['gam'] |
|---|
| 1137 | if 'Y' in varyList: |
|---|
| 1138 | dMdv[varyList.index('Y')] += dgdY*dervDict['gam'] |
|---|
| 1139 | if 'SH/L' in varyList: |
|---|
| 1140 | dMdv[varyList.index('SH/L')] += dervDict['shl'] #problem here |
|---|
| 1141 | if 'I(L2)/I(L1)' in varyList: |
|---|
| 1142 | dMdv[varyList.index('I(L2)/I(L1)')] += dervDict['L1/L2'] |
|---|
| 1143 | iPeak += 1 |
|---|
| 1144 | except KeyError: #no more peaks to process |
|---|
| 1145 | break |
|---|
| 1146 | else: |
|---|
| 1147 | Pdabc = parmDict['Pdabc'] |
|---|
| 1148 | difC = parmDict['difC'] |
|---|
| 1149 | iPeak = 0 |
|---|
| 1150 | while True: |
|---|
| 1151 | try: |
|---|
| 1152 | pos = parmDict['pos'+str(iPeak)] |
|---|
| 1153 | tof = pos-parmDict['Zero'] |
|---|
| 1154 | dsp = tof/difC |
|---|
| 1155 | intens = parmDict['int'+str(iPeak)] |
|---|
| 1156 | alpName = 'alp'+str(iPeak) |
|---|
| 1157 | if alpName in varyList: |
|---|
| 1158 | alp = parmDict[alpName] |
|---|
| 1159 | else: |
|---|
| 1160 | if len(Pdabc): |
|---|
| 1161 | alp = np.interp(dsp,Pdabc[0],Pdabc[1]) |
|---|
| 1162 | dad0 = 0 |
|---|
| 1163 | else: |
|---|
| 1164 | alp = G2mth.getTOFalpha(parmDict,dsp) |
|---|
| 1165 | dada0 = G2mth.getTOFalphaDeriv(dsp) |
|---|
| 1166 | betName = 'bet'+str(iPeak) |
|---|
| 1167 | if betName in varyList: |
|---|
| 1168 | bet = parmDict[betName] |
|---|
| 1169 | else: |
|---|
| 1170 | if len(Pdabc): |
|---|
| 1171 | bet = np.interp(dsp,Pdabc[0],Pdabc[2]) |
|---|
| 1172 | dbdb0 = dbdb1 = dbdb2 = 0 |
|---|
| 1173 | else: |
|---|
| 1174 | bet = G2mth.getTOFbeta(parmDict,dsp) |
|---|
| 1175 | dbdb0,dbdb1,dbdb2 = G2mth.getTOFbetaDeriv(dsp) |
|---|
| 1176 | sigName = 'sig'+str(iPeak) |
|---|
| 1177 | if sigName in varyList: |
|---|
| 1178 | sig = parmDict[sigName] |
|---|
| 1179 | dsds0 = dsds1 = dsds2 = dsds3 = 0 |
|---|
| 1180 | else: |
|---|
| 1181 | sig = G2mth.getTOFsig(parmDict,dsp) |
|---|
| 1182 | dsds0,dsds1,dsds2,dsds3 = G2mth.getTOFsigDeriv(dsp) |
|---|
| 1183 | gamName = 'gam'+str(iPeak) |
|---|
| 1184 | if gamName in varyList: |
|---|
| 1185 | gam = parmDict[gamName] |
|---|
| 1186 | dsdX = dsdY = 0 |
|---|
| 1187 | else: |
|---|
| 1188 | gam = G2mth.getTOFgamma(parmDict,dsp) |
|---|
| 1189 | dsdX,dsdY = G2mth.getTOFgammaDeriv(dsp) |
|---|
| 1190 | gam = max(gam,0.001) #avoid neg gamma |
|---|
| 1191 | Wd,fmin,fmax = getWidthsTOF(pos,alp,bet,sig,gam) |
|---|
| 1192 | iBeg = np.searchsorted(xdata,pos-fmin) |
|---|
| 1193 | lenX = len(xdata) |
|---|
| 1194 | if not iBeg: |
|---|
| 1195 | iFin = np.searchsorted(xdata,pos+fmax) |
|---|
| 1196 | elif iBeg == lenX: |
|---|
| 1197 | iFin = iBeg |
|---|
| 1198 | else: |
|---|
| 1199 | iFin = np.searchsorted(xdata,pos+fmax) |
|---|
| 1200 | if not iBeg+iFin: #peak below low limit |
|---|
| 1201 | iPeak += 1 |
|---|
| 1202 | continue |
|---|
| 1203 | elif not iBeg-iFin: #peak above high limit |
|---|
| 1204 | break |
|---|
| 1205 | dMdpk = np.zeros(shape=(7,len(xdata))) |
|---|
| 1206 | dMdipk = getdEpsVoigt(pos,alp,bet,sig,gam,xdata[iBeg:iFin]) |
|---|
| 1207 | for i in range(1,6): |
|---|
| 1208 | dMdpk[i][iBeg:iFin] += intens*cw[iBeg:iFin]*dMdipk[i] |
|---|
| 1209 | dMdpk[0][iBeg:iFin] += cw[iBeg:iFin]*dMdipk[0] |
|---|
| 1210 | dervDict = {'int':dMdpk[0],'pos':dMdpk[1],'alp':dMdpk[2],'bet':dMdpk[3],'sig':dMdpk[4],'gam':dMdpk[5]} |
|---|
| 1211 | for parmName in ['pos','int','alp','bet','sig','gam']: |
|---|
| 1212 | try: |
|---|
| 1213 | idx = varyList.index(parmName+str(iPeak)) |
|---|
| 1214 | dMdv[idx] = dervDict[parmName] |
|---|
| 1215 | except ValueError: |
|---|
| 1216 | pass |
|---|
| 1217 | if 'alpha' in varyList: |
|---|
| 1218 | dMdv[varyList.index('alpha')] += dada0*dervDict['alp'] |
|---|
| 1219 | if 'beta-0' in varyList: |
|---|
| 1220 | dMdv[varyList.index('beta-0')] += dbdb0*dervDict['bet'] |
|---|
| 1221 | if 'beta-1' in varyList: |
|---|
| 1222 | dMdv[varyList.index('beta-1')] += dbdb1*dervDict['bet'] |
|---|
| 1223 | if 'beta-q' in varyList: |
|---|
| 1224 | dMdv[varyList.index('beta-q')] += dbdb2*dervDict['bet'] |
|---|
| 1225 | if 'sig-0' in varyList: |
|---|
| 1226 | dMdv[varyList.index('sig-0')] += dsds0*dervDict['sig'] |
|---|
| 1227 | if 'sig-1' in varyList: |
|---|
| 1228 | dMdv[varyList.index('sig-1')] += dsds1*dervDict['sig'] |
|---|
| 1229 | if 'sig-2' in varyList: |
|---|
| 1230 | dMdv[varyList.index('sig-2')] += dsds2*dervDict['sig'] |
|---|
| 1231 | if 'sig-q' in varyList: |
|---|
| 1232 | dMdv[varyList.index('sig-q')] += dsds3*dervDict['sig'] |
|---|
| 1233 | if 'X' in varyList: |
|---|
| 1234 | dMdv[varyList.index('X')] += dsdX*dervDict['gam'] |
|---|
| 1235 | if 'Y' in varyList: |
|---|
| 1236 | dMdv[varyList.index('Y')] += dsdY*dervDict['gam'] #problem here |
|---|
| 1237 | iPeak += 1 |
|---|
| 1238 | except KeyError: #no more peaks to process |
|---|
| 1239 | break |
|---|
| 1240 | return dMdv |
|---|
| 1241 | |
|---|
| 1242 | def Dict2Values(parmdict, varylist): |
|---|
| 1243 | '''Use before call to leastsq to setup list of values for the parameters |
|---|
| 1244 | in parmdict, as selected by key in varylist''' |
|---|
| 1245 | return [parmdict[key] for key in varylist] |
|---|
| 1246 | |
|---|
| 1247 | def Values2Dict(parmdict, varylist, values): |
|---|
| 1248 | ''' Use after call to leastsq to update the parameter dictionary with |
|---|
| 1249 | values corresponding to keys in varylist''' |
|---|
| 1250 | parmdict.update(zip(varylist,values)) |
|---|
| 1251 | |
|---|
| 1252 | def SetBackgroundParms(Background): |
|---|
| 1253 | 'needs a doc string' |
|---|
| 1254 | if len(Background) == 1: # fix up old backgrounds |
|---|
| 1255 | BackGround.append({'nDebye':0,'debyeTerms':[]}) |
|---|
| 1256 | bakType,bakFlag = Background[0][:2] |
|---|
| 1257 | backVals = Background[0][3:] |
|---|
| 1258 | backNames = ['Back;'+str(i) for i in range(len(backVals))] |
|---|
| 1259 | Debye = Background[1] #also has background peaks stuff |
|---|
| 1260 | backDict = dict(zip(backNames,backVals)) |
|---|
| 1261 | backVary = [] |
|---|
| 1262 | if bakFlag: |
|---|
| 1263 | backVary = backNames |
|---|
| 1264 | |
|---|
| 1265 | backDict['nDebye'] = Debye['nDebye'] |
|---|
| 1266 | debyeDict = {} |
|---|
| 1267 | debyeList = [] |
|---|
| 1268 | for i in range(Debye['nDebye']): |
|---|
| 1269 | debyeNames = ['DebyeA;'+str(i),'DebyeR;'+str(i),'DebyeU;'+str(i)] |
|---|
| 1270 | debyeDict.update(dict(zip(debyeNames,Debye['debyeTerms'][i][::2]))) |
|---|
| 1271 | debyeList += zip(debyeNames,Debye['debyeTerms'][i][1::2]) |
|---|
| 1272 | debyeVary = [] |
|---|
| 1273 | for item in debyeList: |
|---|
| 1274 | if item[1]: |
|---|
| 1275 | debyeVary.append(item[0]) |
|---|
| 1276 | backDict.update(debyeDict) |
|---|
| 1277 | backVary += debyeVary |
|---|
| 1278 | |
|---|
| 1279 | backDict['nPeaks'] = Debye['nPeaks'] |
|---|
| 1280 | peaksDict = {} |
|---|
| 1281 | peaksList = [] |
|---|
| 1282 | for i in range(Debye['nPeaks']): |
|---|
| 1283 | peaksNames = ['BkPkpos;'+str(i),'BkPkint;'+str(i),'BkPksig;'+str(i),'BkPkgam;'+str(i)] |
|---|
| 1284 | peaksDict.update(dict(zip(peaksNames,Debye['peaksList'][i][::2]))) |
|---|
| 1285 | peaksList += zip(peaksNames,Debye['peaksList'][i][1::2]) |
|---|
| 1286 | peaksVary = [] |
|---|
| 1287 | for item in peaksList: |
|---|
| 1288 | if item[1]: |
|---|
| 1289 | peaksVary.append(item[0]) |
|---|
| 1290 | backDict.update(peaksDict) |
|---|
| 1291 | backVary += peaksVary |
|---|
| 1292 | return bakType,backDict,backVary |
|---|
| 1293 | |
|---|
| 1294 | def DoCalibInst(IndexPeaks,Inst): |
|---|
| 1295 | |
|---|
| 1296 | def SetInstParms(): |
|---|
| 1297 | dataType = Inst['Type'][0] |
|---|
| 1298 | insVary = [] |
|---|
| 1299 | insNames = [] |
|---|
| 1300 | insVals = [] |
|---|
| 1301 | for parm in Inst: |
|---|
| 1302 | insNames.append(parm) |
|---|
| 1303 | insVals.append(Inst[parm][1]) |
|---|
| 1304 | if parm in ['Lam','difC','difA','difB','Zero',]: |
|---|
| 1305 | if Inst[parm][2]: |
|---|
| 1306 | insVary.append(parm) |
|---|
| 1307 | instDict = dict(zip(insNames,insVals)) |
|---|
| 1308 | return dataType,instDict,insVary |
|---|
| 1309 | |
|---|
| 1310 | def GetInstParms(parmDict,Inst,varyList): |
|---|
| 1311 | for name in Inst: |
|---|
| 1312 | Inst[name][1] = parmDict[name] |
|---|
| 1313 | |
|---|
| 1314 | def InstPrint(Inst,sigDict): |
|---|
| 1315 | print 'Instrument Parameters:' |
|---|
| 1316 | if 'C' in Inst['Type'][0]: |
|---|
| 1317 | ptfmt = "%12.6f" |
|---|
| 1318 | else: |
|---|
| 1319 | ptfmt = "%12.3f" |
|---|
| 1320 | ptlbls = 'names :' |
|---|
| 1321 | ptstr = 'values:' |
|---|
| 1322 | sigstr = 'esds :' |
|---|
| 1323 | for parm in Inst: |
|---|
| 1324 | if parm in ['Lam','difC','difA','difB','Zero',]: |
|---|
| 1325 | ptlbls += "%s" % (parm.center(12)) |
|---|
| 1326 | ptstr += ptfmt % (Inst[parm][1]) |
|---|
| 1327 | if parm in sigDict: |
|---|
| 1328 | sigstr += ptfmt % (sigDict[parm]) |
|---|
| 1329 | else: |
|---|
| 1330 | sigstr += 12*' ' |
|---|
| 1331 | print ptlbls |
|---|
| 1332 | print ptstr |
|---|
| 1333 | print sigstr |
|---|
| 1334 | |
|---|
| 1335 | def errPeakPos(values,peakDsp,peakPos,peakWt,dataType,parmDict,varyList): |
|---|
| 1336 | parmDict.update(zip(varyList,values)) |
|---|
| 1337 | return np.sqrt(peakWt)*(G2lat.getPeakPos(dataType,parmDict,peakDsp)-peakPos) |
|---|
| 1338 | |
|---|
| 1339 | peakPos = [] |
|---|
| 1340 | peakDsp = [] |
|---|
| 1341 | peakWt = [] |
|---|
| 1342 | for peak,sig in zip(IndexPeaks[0],IndexPeaks[1]): |
|---|
| 1343 | if peak[2] and peak[3] and sig > 0.: |
|---|
| 1344 | peakPos.append(peak[0]) |
|---|
| 1345 | peakDsp.append(peak[-1]) #d-calc |
|---|
| 1346 | peakWt.append(peak[-1]**2/sig**2) #weight by d**2 |
|---|
| 1347 | peakPos = np.array(peakPos) |
|---|
| 1348 | peakDsp = np.array(peakDsp) |
|---|
| 1349 | peakWt = np.array(peakWt) |
|---|
| 1350 | dataType,insDict,insVary = SetInstParms() |
|---|
| 1351 | parmDict = {} |
|---|
| 1352 | parmDict.update(insDict) |
|---|
| 1353 | varyList = insVary |
|---|
| 1354 | if not len(varyList): |
|---|
| 1355 | print '**** ERROR - nothing to refine! ****' |
|---|
| 1356 | return False |
|---|
| 1357 | while True: |
|---|
| 1358 | begin = time.time() |
|---|
| 1359 | values = np.array(Dict2Values(parmDict, varyList)) |
|---|
| 1360 | result = so.leastsq(errPeakPos,values,full_output=True,ftol=0.000001, |
|---|
| 1361 | args=(peakDsp,peakPos,peakWt,dataType,parmDict,varyList)) |
|---|
| 1362 | ncyc = int(result[2]['nfev']/2) |
|---|
| 1363 | runtime = time.time()-begin |
|---|
| 1364 | chisq = np.sum(result[2]['fvec']**2) |
|---|
| 1365 | Values2Dict(parmDict, varyList, result[0]) |
|---|
| 1366 | GOF = chisq/(len(peakPos)-len(varyList)) #reduced chi^2 |
|---|
| 1367 | print 'Number of function calls:',result[2]['nfev'],' Number of observations: ',len(peakPos),' Number of parameters: ',len(varyList) |
|---|
| 1368 | print 'calib time = %8.3fs, %8.3fs/cycle'%(runtime,runtime/ncyc) |
|---|
| 1369 | print 'chi**2 = %12.6g, reduced chi**2 = %6.2f'%(chisq,GOF) |
|---|
| 1370 | try: |
|---|
| 1371 | sig = np.sqrt(np.diag(result[1])*GOF) |
|---|
| 1372 | if np.any(np.isnan(sig)): |
|---|
| 1373 | print '*** Least squares aborted - some invalid esds possible ***' |
|---|
| 1374 | break #refinement succeeded - finish up! |
|---|
| 1375 | except ValueError: #result[1] is None on singular matrix |
|---|
| 1376 | print '**** Refinement failed - singular matrix ****' |
|---|
| 1377 | |
|---|
| 1378 | sigDict = dict(zip(varyList,sig)) |
|---|
| 1379 | GetInstParms(parmDict,Inst,varyList) |
|---|
| 1380 | InstPrint(Inst,sigDict) |
|---|
| 1381 | return True |
|---|
| 1382 | |
|---|
| 1383 | def DoPeakFit(FitPgm,Peaks,Background,Limits,Inst,Inst2,data,prevVaryList=[],oneCycle=False,controls=None,dlg=None): |
|---|
| 1384 | '''Called to perform a peak fit, refining the selected items in the peak |
|---|
| 1385 | table as well as selected items in the background. |
|---|
| 1386 | |
|---|
| 1387 | :param str FitPgm: type of fit to perform. At present this is ignored. |
|---|
| 1388 | :param list Peaks: a list of peaks. Each peak entry is a list with 8 values: |
|---|
| 1389 | four values followed by a refine flag where the values are: position, intensity, |
|---|
| 1390 | sigma (Gaussian width) and gamma (Lorentzian width). From the Histogram/"Peak List" |
|---|
| 1391 | tree entry, dict item "peaks" |
|---|
| 1392 | :param list Background: describes the background. List with two items. |
|---|
| 1393 | Item 0 specifies a background model and coefficients. Item 1 is a dict. |
|---|
| 1394 | From the Histogram/Background tree entry. |
|---|
| 1395 | :param list Limits: min and max x-value to use |
|---|
| 1396 | :param dict Inst: Instrument parameters |
|---|
| 1397 | :param dict Inst2: more Instrument parameters |
|---|
| 1398 | :param numpy.array data: a 5xn array. data[0] is the x-values, |
|---|
| 1399 | data[1] is the y-values, data[2] are weight values, data[3], [4] and [5] are |
|---|
| 1400 | calc, background and difference intensities, respectively. |
|---|
| 1401 | :param list prevVaryList: Used in sequential refinements to override the |
|---|
| 1402 | variable list. Defaults as an empty list. |
|---|
| 1403 | :param bool oneCycle: True if only one cycle of fitting should be performed |
|---|
| 1404 | :param dict controls: a dict specifying two values, Ftol = controls['min dM/M'] |
|---|
| 1405 | and derivType = controls['deriv type']. If None default values are used. |
|---|
| 1406 | :param wx.Dialog dlg: A dialog box that is updated with progress from the fit. |
|---|
| 1407 | Defaults to None, which means no updates are done. |
|---|
| 1408 | ''' |
|---|
| 1409 | def GetBackgroundParms(parmList,Background): |
|---|
| 1410 | iBak = 0 |
|---|
| 1411 | while True: |
|---|
| 1412 | try: |
|---|
| 1413 | bakName = 'Back;'+str(iBak) |
|---|
| 1414 | Background[0][iBak+3] = parmList[bakName] |
|---|
| 1415 | iBak += 1 |
|---|
| 1416 | except KeyError: |
|---|
| 1417 | break |
|---|
| 1418 | iDb = 0 |
|---|
| 1419 | while True: |
|---|
| 1420 | names = ['DebyeA;','DebyeR;','DebyeU;'] |
|---|
| 1421 | try: |
|---|
| 1422 | for i,name in enumerate(names): |
|---|
| 1423 | val = parmList[name+str(iDb)] |
|---|
| 1424 | Background[1]['debyeTerms'][iDb][2*i] = val |
|---|
| 1425 | iDb += 1 |
|---|
| 1426 | except KeyError: |
|---|
| 1427 | break |
|---|
| 1428 | iDb = 0 |
|---|
| 1429 | while True: |
|---|
| 1430 | names = ['BkPkpos;','BkPkint;','BkPksig;','BkPkgam;'] |
|---|
| 1431 | try: |
|---|
| 1432 | for i,name in enumerate(names): |
|---|
| 1433 | val = parmList[name+str(iDb)] |
|---|
| 1434 | Background[1]['peaksList'][iDb][2*i] = val |
|---|
| 1435 | iDb += 1 |
|---|
| 1436 | except KeyError: |
|---|
| 1437 | break |
|---|
| 1438 | |
|---|
| 1439 | def BackgroundPrint(Background,sigDict): |
|---|
| 1440 | print 'Background coefficients for',Background[0][0],'function' |
|---|
| 1441 | ptfmt = "%12.5f" |
|---|
| 1442 | ptstr = 'value: ' |
|---|
| 1443 | sigstr = 'esd : ' |
|---|
| 1444 | for i,back in enumerate(Background[0][3:]): |
|---|
| 1445 | ptstr += ptfmt % (back) |
|---|
| 1446 | if Background[0][1]: |
|---|
| 1447 | prm = 'Back;'+str(i) |
|---|
| 1448 | if prm in sigDict: |
|---|
| 1449 | sigstr += ptfmt % (sigDict[prm]) |
|---|
| 1450 | else: |
|---|
| 1451 | sigstr += " "*12 |
|---|
| 1452 | if len(ptstr) > 75: |
|---|
| 1453 | print ptstr |
|---|
| 1454 | if Background[0][1]: print sigstr |
|---|
| 1455 | ptstr = 'value: ' |
|---|
| 1456 | sigstr = 'esd : ' |
|---|
| 1457 | if len(ptstr) > 8: |
|---|
| 1458 | print ptstr |
|---|
| 1459 | if Background[0][1]: print sigstr |
|---|
| 1460 | |
|---|
| 1461 | if Background[1]['nDebye']: |
|---|
| 1462 | parms = ['DebyeA;','DebyeR;','DebyeU;'] |
|---|
| 1463 | print 'Debye diffuse scattering coefficients' |
|---|
| 1464 | ptfmt = "%12.5f" |
|---|
| 1465 | print ' term DebyeA esd DebyeR esd DebyeU esd' |
|---|
| 1466 | for term in range(Background[1]['nDebye']): |
|---|
| 1467 | line = ' term %d'%(term) |
|---|
| 1468 | for ip,name in enumerate(parms): |
|---|
| 1469 | line += ptfmt%(Background[1]['debyeTerms'][term][2*ip]) |
|---|
| 1470 | if name+str(term) in sigDict: |
|---|
| 1471 | line += ptfmt%(sigDict[name+str(term)]) |
|---|
| 1472 | else: |
|---|
| 1473 | line += " "*12 |
|---|
| 1474 | print line |
|---|
| 1475 | if Background[1]['nPeaks']: |
|---|
| 1476 | print 'Coefficients for Background Peaks' |
|---|
| 1477 | ptfmt = "%15.3f" |
|---|
| 1478 | for j,pl in enumerate(Background[1]['peaksList']): |
|---|
| 1479 | names = 'peak %3d:'%(j+1) |
|---|
| 1480 | ptstr = 'values :' |
|---|
| 1481 | sigstr = 'esds :' |
|---|
| 1482 | for i,lbl in enumerate(['BkPkpos','BkPkint','BkPksig','BkPkgam']): |
|---|
| 1483 | val = pl[2*i] |
|---|
| 1484 | prm = lbl+";"+str(j) |
|---|
| 1485 | names += '%15s'%(prm) |
|---|
| 1486 | ptstr += ptfmt%(val) |
|---|
| 1487 | if prm in sigDict: |
|---|
| 1488 | sigstr += ptfmt%(sigDict[prm]) |
|---|
| 1489 | else: |
|---|
| 1490 | sigstr += " "*15 |
|---|
| 1491 | print names |
|---|
| 1492 | print ptstr |
|---|
| 1493 | print sigstr |
|---|
| 1494 | |
|---|
| 1495 | def SetInstParms(Inst): |
|---|
| 1496 | dataType = Inst['Type'][0] |
|---|
| 1497 | insVary = [] |
|---|
| 1498 | insNames = [] |
|---|
| 1499 | insVals = [] |
|---|
| 1500 | for parm in Inst: |
|---|
| 1501 | insNames.append(parm) |
|---|
| 1502 | insVals.append(Inst[parm][1]) |
|---|
| 1503 | if parm in ['U','V','W','X','Y','SH/L','I(L2)/I(L1)','alpha', |
|---|
| 1504 | 'beta-0','beta-1','beta-q','sig-0','sig-1','sig-2','sig-q',] and Inst[parm][2]: |
|---|
| 1505 | insVary.append(parm) |
|---|
| 1506 | instDict = dict(zip(insNames,insVals)) |
|---|
| 1507 | instDict['X'] = max(instDict['X'],0.01) |
|---|
| 1508 | instDict['Y'] = max(instDict['Y'],0.01) |
|---|
| 1509 | if 'SH/L' in instDict: |
|---|
| 1510 | instDict['SH/L'] = max(instDict['SH/L'],0.002) |
|---|
| 1511 | return dataType,instDict,insVary |
|---|
| 1512 | |
|---|
| 1513 | def GetInstParms(parmDict,Inst,varyList,Peaks): |
|---|
| 1514 | for name in Inst: |
|---|
| 1515 | Inst[name][1] = parmDict[name] |
|---|
| 1516 | iPeak = 0 |
|---|
| 1517 | while True: |
|---|
| 1518 | try: |
|---|
| 1519 | sigName = 'sig'+str(iPeak) |
|---|
| 1520 | pos = parmDict['pos'+str(iPeak)] |
|---|
| 1521 | if sigName not in varyList: |
|---|
| 1522 | if 'C' in Inst['Type'][0]: |
|---|
| 1523 | parmDict[sigName] = G2mth.getCWsig(parmDict,pos) |
|---|
| 1524 | else: |
|---|
| 1525 | dsp = G2lat.Pos2dsp(Inst,pos) |
|---|
| 1526 | parmDict[sigName] = G2mth.getTOFsig(parmDict,dsp) |
|---|
| 1527 | gamName = 'gam'+str(iPeak) |
|---|
| 1528 | if gamName not in varyList: |
|---|
| 1529 | if 'C' in Inst['Type'][0]: |
|---|
| 1530 | parmDict[gamName] = G2mth.getCWgam(parmDict,pos) |
|---|
| 1531 | else: |
|---|
| 1532 | dsp = G2lat.Pos2dsp(Inst,pos) |
|---|
| 1533 | parmDict[gamName] = G2mth.getTOFgamma(parmDict,dsp) |
|---|
| 1534 | iPeak += 1 |
|---|
| 1535 | except KeyError: |
|---|
| 1536 | break |
|---|
| 1537 | |
|---|
| 1538 | def InstPrint(Inst,sigDict): |
|---|
| 1539 | print 'Instrument Parameters:' |
|---|
| 1540 | ptfmt = "%12.6f" |
|---|
| 1541 | ptlbls = 'names :' |
|---|
| 1542 | ptstr = 'values:' |
|---|
| 1543 | sigstr = 'esds :' |
|---|
| 1544 | for parm in Inst: |
|---|
| 1545 | if parm in ['U','V','W','X','Y','SH/L','I(L2)/I(L1)','alpha', |
|---|
| 1546 | 'beta-0','beta-1','beta-q','sig-0','sig-1','sig-2','sig-q',]: |
|---|
| 1547 | ptlbls += "%s" % (parm.center(12)) |
|---|
| 1548 | ptstr += ptfmt % (Inst[parm][1]) |
|---|
| 1549 | if parm in sigDict: |
|---|
| 1550 | sigstr += ptfmt % (sigDict[parm]) |
|---|
| 1551 | else: |
|---|
| 1552 | sigstr += 12*' ' |
|---|
| 1553 | print ptlbls |
|---|
| 1554 | print ptstr |
|---|
| 1555 | print sigstr |
|---|
| 1556 | |
|---|
| 1557 | def SetPeaksParms(dataType,Peaks): |
|---|
| 1558 | peakNames = [] |
|---|
| 1559 | peakVary = [] |
|---|
| 1560 | peakVals = [] |
|---|
| 1561 | if 'C' in dataType: |
|---|
| 1562 | names = ['pos','int','sig','gam'] |
|---|
| 1563 | else: |
|---|
| 1564 | names = ['pos','int','alp','bet','sig','gam'] |
|---|
| 1565 | for i,peak in enumerate(Peaks): |
|---|
| 1566 | for j,name in enumerate(names): |
|---|
| 1567 | peakVals.append(peak[2*j]) |
|---|
| 1568 | parName = name+str(i) |
|---|
| 1569 | peakNames.append(parName) |
|---|
| 1570 | if peak[2*j+1]: |
|---|
| 1571 | peakVary.append(parName) |
|---|
| 1572 | return dict(zip(peakNames,peakVals)),peakVary |
|---|
| 1573 | |
|---|
| 1574 | def GetPeaksParms(Inst,parmDict,Peaks,varyList): |
|---|
| 1575 | if 'C' in Inst['Type'][0]: |
|---|
| 1576 | names = ['pos','int','sig','gam'] |
|---|
| 1577 | else: #'T' |
|---|
| 1578 | names = ['pos','int','alp','bet','sig','gam'] |
|---|
| 1579 | for i,peak in enumerate(Peaks): |
|---|
| 1580 | pos = parmDict['pos'+str(i)] |
|---|
| 1581 | if 'difC' in Inst: |
|---|
| 1582 | dsp = pos/Inst['difC'][1] |
|---|
| 1583 | for j in range(len(names)): |
|---|
| 1584 | parName = names[j]+str(i) |
|---|
| 1585 | if parName in varyList: |
|---|
| 1586 | peak[2*j] = parmDict[parName] |
|---|
| 1587 | elif 'alpha' in parName: |
|---|
| 1588 | peak[2*j] = parmDict['alpha']/dsp |
|---|
| 1589 | elif 'beta' in parName: |
|---|
| 1590 | peak[2*j] = G2mth.getTOFbeta(parmDict,dsp) |
|---|
| 1591 | elif 'sig' in parName: |
|---|
| 1592 | if 'C' in Inst['Type'][0]: |
|---|
| 1593 | peak[2*j] = G2mth.getCWsig(parmDict,pos) |
|---|
| 1594 | else: |
|---|
| 1595 | peak[2*j] = G2mth.getTOFsig(parmDict,dsp) |
|---|
| 1596 | elif 'gam' in parName: |
|---|
| 1597 | if 'C' in Inst['Type'][0]: |
|---|
| 1598 | peak[2*j] = G2mth.getCWgam(parmDict,pos) |
|---|
| 1599 | else: |
|---|
| 1600 | peak[2*j] = G2mth.getTOFgamma(parmDict,dsp) |
|---|
| 1601 | |
|---|
| 1602 | def PeaksPrint(dataType,parmDict,sigDict,varyList): |
|---|
| 1603 | print 'Peak coefficients:' |
|---|
| 1604 | if 'C' in dataType: |
|---|
| 1605 | names = ['pos','int','sig','gam'] |
|---|
| 1606 | else: #'T' |
|---|
| 1607 | names = ['pos','int','alp','bet','sig','gam'] |
|---|
| 1608 | head = 13*' ' |
|---|
| 1609 | for name in names: |
|---|
| 1610 | if name in ['alp','bet']: |
|---|
| 1611 | head += name.center(8)+'esd'.center(8) |
|---|
| 1612 | else: |
|---|
| 1613 | head += name.center(10)+'esd'.center(10) |
|---|
| 1614 | print head |
|---|
| 1615 | if 'C' in dataType: |
|---|
| 1616 | ptfmt = {'pos':"%10.5f",'int':"%10.1f",'sig':"%10.3f",'gam':"%10.3f"} |
|---|
| 1617 | else: |
|---|
| 1618 | ptfmt = {'pos':"%10.2f",'int':"%10.4f",'alp':"%8.3f",'bet':"%8.5f",'sig':"%10.3f",'gam':"%10.3f"} |
|---|
| 1619 | for i,peak in enumerate(Peaks): |
|---|
| 1620 | ptstr = ':' |
|---|
| 1621 | for j in range(len(names)): |
|---|
| 1622 | name = names[j] |
|---|
| 1623 | parName = name+str(i) |
|---|
| 1624 | ptstr += ptfmt[name] % (parmDict[parName]) |
|---|
| 1625 | if parName in varyList: |
|---|
| 1626 | ptstr += ptfmt[name] % (sigDict[parName]) |
|---|
| 1627 | else: |
|---|
| 1628 | if name in ['alp','bet']: |
|---|
| 1629 | ptstr += 8*' ' |
|---|
| 1630 | else: |
|---|
| 1631 | ptstr += 10*' ' |
|---|
| 1632 | print '%s'%(('Peak'+str(i+1)).center(8)),ptstr |
|---|
| 1633 | |
|---|
| 1634 | def devPeakProfile(values,xdata,ydata, weights,dataType,parmdict,varylist,bakType,dlg): |
|---|
| 1635 | parmdict.update(zip(varylist,values)) |
|---|
| 1636 | return np.sqrt(weights)*getPeakProfileDerv(dataType,parmdict,xdata,varylist,bakType) |
|---|
| 1637 | |
|---|
| 1638 | def errPeakProfile(values,xdata,ydata,weights,dataType,parmdict,varylist,bakType,dlg): |
|---|
| 1639 | parmdict.update(zip(varylist,values)) |
|---|
| 1640 | M = np.sqrt(weights)*(getPeakProfile(dataType,parmdict,xdata,varylist,bakType)-ydata) |
|---|
| 1641 | Rwp = min(100.,np.sqrt(np.sum(M**2)/np.sum(weights*ydata**2))*100.) |
|---|
| 1642 | if dlg: |
|---|
| 1643 | GoOn = dlg.Update(Rwp,newmsg='%s%8.3f%s'%('Peak fit Rwp =',Rwp,'%'))[0] |
|---|
| 1644 | if not GoOn: |
|---|
| 1645 | return -M #abort!! |
|---|
| 1646 | return M |
|---|
| 1647 | |
|---|
| 1648 | if controls: |
|---|
| 1649 | Ftol = controls['min dM/M'] |
|---|
| 1650 | derivType = controls['deriv type'] |
|---|
| 1651 | else: |
|---|
| 1652 | Ftol = 0.0001 |
|---|
| 1653 | derivType = 'analytic' |
|---|
| 1654 | if oneCycle: |
|---|
| 1655 | Ftol = 1.0 |
|---|
| 1656 | x,y,w,yc,yb,yd = data #these are numpy arrays! |
|---|
| 1657 | yc *= 0. #set calcd ones to zero |
|---|
| 1658 | yb *= 0. |
|---|
| 1659 | yd *= 0. |
|---|
| 1660 | xBeg = np.searchsorted(x,Limits[0]) |
|---|
| 1661 | xFin = np.searchsorted(x,Limits[1]) |
|---|
| 1662 | bakType,bakDict,bakVary = SetBackgroundParms(Background) |
|---|
| 1663 | dataType,insDict,insVary = SetInstParms(Inst) |
|---|
| 1664 | peakDict,peakVary = SetPeaksParms(Inst['Type'][0],Peaks) |
|---|
| 1665 | parmDict = {} |
|---|
| 1666 | parmDict.update(bakDict) |
|---|
| 1667 | parmDict.update(insDict) |
|---|
| 1668 | parmDict.update(peakDict) |
|---|
| 1669 | parmDict['Pdabc'] = [] #dummy Pdabc |
|---|
| 1670 | parmDict.update(Inst2) #put in real one if there |
|---|
| 1671 | if prevVaryList: |
|---|
| 1672 | varyList = prevVaryList[:] |
|---|
| 1673 | else: |
|---|
| 1674 | varyList = bakVary+insVary+peakVary |
|---|
| 1675 | fullvaryList = varyList[:] |
|---|
| 1676 | while True: |
|---|
| 1677 | begin = time.time() |
|---|
| 1678 | values = np.array(Dict2Values(parmDict, varyList)) |
|---|
| 1679 | Rvals = {} |
|---|
| 1680 | badVary = [] |
|---|
| 1681 | result = so.leastsq(errPeakProfile,values,Dfun=devPeakProfile,full_output=True,ftol=Ftol,col_deriv=True, |
|---|
| 1682 | args=(x[xBeg:xFin],y[xBeg:xFin],w[xBeg:xFin],dataType,parmDict,varyList,bakType,dlg)) |
|---|
| 1683 | ncyc = int(result[2]['nfev']/2) |
|---|
| 1684 | runtime = time.time()-begin |
|---|
| 1685 | chisq = np.sum(result[2]['fvec']**2) |
|---|
| 1686 | Values2Dict(parmDict, varyList, result[0]) |
|---|
| 1687 | Rvals['Rwp'] = np.sqrt(chisq/np.sum(w[xBeg:xFin]*y[xBeg:xFin]**2))*100. #to % |
|---|
| 1688 | Rvals['GOF'] = chisq/(xFin-xBeg-len(varyList)) #reduced chi^2 |
|---|
| 1689 | print 'Number of function calls:',result[2]['nfev'],' Number of observations: ',xFin-xBeg,' Number of parameters: ',len(varyList) |
|---|
| 1690 | if ncyc: |
|---|
| 1691 | print 'fitpeak time = %8.3fs, %8.3fs/cycle'%(runtime,runtime/ncyc) |
|---|
| 1692 | print 'Rwp = %7.2f%%, chi**2 = %12.6g, reduced chi**2 = %6.2f'%(Rvals['Rwp'],chisq,Rvals['GOF']) |
|---|
| 1693 | sig = [0]*len(varyList) |
|---|
| 1694 | if len(varyList) == 0: break # if nothing was refined |
|---|
| 1695 | try: |
|---|
| 1696 | sig = np.sqrt(np.diag(result[1])*Rvals['GOF']) |
|---|
| 1697 | if np.any(np.isnan(sig)): |
|---|
| 1698 | print '*** Least squares aborted - some invalid esds possible ***' |
|---|
| 1699 | break #refinement succeeded - finish up! |
|---|
| 1700 | except ValueError: #result[1] is None on singular matrix |
|---|
| 1701 | print '**** Refinement failed - singular matrix ****' |
|---|
| 1702 | Ipvt = result[2]['ipvt'] |
|---|
| 1703 | for i,ipvt in enumerate(Ipvt): |
|---|
| 1704 | if not np.sum(result[2]['fjac'],axis=1)[i]: |
|---|
| 1705 | print 'Removing parameter: ',varyList[ipvt-1] |
|---|
| 1706 | badVary.append(varyList[ipvt-1]) |
|---|
| 1707 | del(varyList[ipvt-1]) |
|---|
| 1708 | break |
|---|
| 1709 | else: # nothing removed |
|---|
| 1710 | break |
|---|
| 1711 | if dlg: dlg.Destroy() |
|---|
| 1712 | sigDict = dict(zip(varyList,sig)) |
|---|
| 1713 | yb[xBeg:xFin] = getBackground('',parmDict,bakType,dataType,x[xBeg:xFin])[0] |
|---|
| 1714 | yc[xBeg:xFin] = getPeakProfile(dataType,parmDict,x[xBeg:xFin],varyList,bakType) |
|---|
| 1715 | yd[xBeg:xFin] = y[xBeg:xFin]-yc[xBeg:xFin] |
|---|
| 1716 | GetBackgroundParms(parmDict,Background) |
|---|
| 1717 | if bakVary: BackgroundPrint(Background,sigDict) |
|---|
| 1718 | GetInstParms(parmDict,Inst,varyList,Peaks) |
|---|
| 1719 | if insVary: InstPrint(Inst,sigDict) |
|---|
| 1720 | GetPeaksParms(Inst,parmDict,Peaks,varyList) |
|---|
| 1721 | if peakVary: PeaksPrint(dataType,parmDict,sigDict,varyList) |
|---|
| 1722 | return sigDict,result,sig,Rvals,varyList,parmDict,fullvaryList,badVary |
|---|
| 1723 | |
|---|
| 1724 | def calcIncident(Iparm,xdata): |
|---|
| 1725 | 'needs a doc string' |
|---|
| 1726 | |
|---|
| 1727 | def IfunAdv(Iparm,xdata): |
|---|
| 1728 | Itype = Iparm['Itype'] |
|---|
| 1729 | Icoef = Iparm['Icoeff'] |
|---|
| 1730 | DYI = np.ones((12,xdata.shape[0])) |
|---|
| 1731 | YI = np.ones_like(xdata)*Icoef[0] |
|---|
| 1732 | |
|---|
| 1733 | x = xdata/1000. #expressions are in ms |
|---|
| 1734 | if Itype == 'Exponential': |
|---|
| 1735 | for i in [1,3,5,7,9]: |
|---|
| 1736 | Eterm = np.exp(-Icoef[i+1]*x**((i+1)/2)) |
|---|
| 1737 | YI += Icoef[i]*Eterm |
|---|
| 1738 | DYI[i] *= Eterm |
|---|
| 1739 | DYI[i+1] *= -Icoef[i]*Eterm*x**((i+1)/2) |
|---|
| 1740 | elif 'Maxwell'in Itype: |
|---|
| 1741 | Eterm = np.exp(-Icoef[2]/x**2) |
|---|
| 1742 | DYI[1] = Eterm/x**5 |
|---|
| 1743 | DYI[2] = -Icoef[1]*DYI[1]/x**2 |
|---|
| 1744 | YI += (Icoef[1]*Eterm/x**5) |
|---|
| 1745 | if 'Exponential' in Itype: |
|---|
| 1746 | for i in range(3,12,2): |
|---|
| 1747 | Eterm = np.exp(-Icoef[i+1]*x**((i+1)/2)) |
|---|
| 1748 | YI += Icoef[i]*Eterm |
|---|
| 1749 | DYI[i] *= Eterm |
|---|
| 1750 | DYI[i+1] *= -Icoef[i]*Eterm*x**((i+1)/2) |
|---|
| 1751 | else: #Chebyschev |
|---|
| 1752 | T = (2./x)-1. |
|---|
| 1753 | Ccof = np.ones((12,xdata.shape[0])) |
|---|
| 1754 | Ccof[1] = T |
|---|
| 1755 | for i in range(2,12): |
|---|
| 1756 | Ccof[i] = 2*T*Ccof[i-1]-Ccof[i-2] |
|---|
| 1757 | for i in range(1,10): |
|---|
| 1758 | YI += Ccof[i]*Icoef[i+2] |
|---|
| 1759 | DYI[i+2] =Ccof[i] |
|---|
| 1760 | return YI,DYI |
|---|
| 1761 | |
|---|
| 1762 | Iesd = np.array(Iparm['Iesd']) |
|---|
| 1763 | Icovar = Iparm['Icovar'] |
|---|
| 1764 | YI,DYI = IfunAdv(Iparm,xdata) |
|---|
| 1765 | YI = np.where(YI>0,YI,1.) |
|---|
| 1766 | WYI = np.zeros_like(xdata) |
|---|
| 1767 | vcov = np.zeros((12,12)) |
|---|
| 1768 | k = 0 |
|---|
| 1769 | for i in range(12): |
|---|
| 1770 | for j in range(i,12): |
|---|
| 1771 | vcov[i][j] = Icovar[k]*Iesd[i]*Iesd[j] |
|---|
| 1772 | vcov[j][i] = Icovar[k]*Iesd[i]*Iesd[j] |
|---|
| 1773 | k += 1 |
|---|
| 1774 | M = np.inner(vcov,DYI.T) |
|---|
| 1775 | WYI = np.sum(M*DYI,axis=0) |
|---|
| 1776 | WYI = np.where(WYI>0.,WYI,0.) |
|---|
| 1777 | return YI,WYI |
|---|
| 1778 | |
|---|
| 1779 | ################################################################################ |
|---|
| 1780 | # Stacking fault simulation codes |
|---|
| 1781 | ################################################################################ |
|---|
| 1782 | |
|---|
| 1783 | def GetStackParms(Layers): |
|---|
| 1784 | |
|---|
| 1785 | Parms = [] |
|---|
| 1786 | #cell parms |
|---|
| 1787 | cell = Layers['Cell'] |
|---|
| 1788 | if Layers['Laue'] in ['-3','-3m','4/m','4/mmm','6/m','6/mmm']: |
|---|
| 1789 | Parms.append('cellA') |
|---|
| 1790 | Parms.append('cellC') |
|---|
| 1791 | else: |
|---|
| 1792 | Parms.append('cellA') |
|---|
| 1793 | Parms.append('cellB') |
|---|
| 1794 | Parms.append('cellC') |
|---|
| 1795 | if Layers['Laue'] != 'mmm': |
|---|
| 1796 | Parms.append('cellG') |
|---|
| 1797 | #Transition parms |
|---|
| 1798 | Trans = Layers['Transitions'] |
|---|
| 1799 | for iY in range(len(Layers['Layers'])): |
|---|
| 1800 | for iX in range(len(Layers['Layers'])): |
|---|
| 1801 | Parms.append('TransP;%d;%d'%(iY,iX)) |
|---|
| 1802 | Parms.append('TransX;%d;%d'%(iY,iX)) |
|---|
| 1803 | Parms.append('TransY;%d;%d'%(iY,iX)) |
|---|
| 1804 | Parms.append('TransZ;%d;%d'%(iY,iX)) |
|---|
| 1805 | return Parms |
|---|
| 1806 | |
|---|
| 1807 | def StackSim(Layers,ctrls,scale=0.,background={},limits=[],inst={},profile=[]): |
|---|
| 1808 | '''Simulate powder or selected area diffraction pattern from stacking faults using DIFFaX |
|---|
| 1809 | |
|---|
| 1810 | param: Layers dict: 'Laue':'-1','Cell':[False,1.,1.,1.,90.,90.,90,1.], |
|---|
| 1811 | 'Width':[[10.,10.],[False,False]],'Toler':0.01,'AtInfo':{}, |
|---|
| 1812 | 'Layers':[],'Stacking':[],'Transitions':[]} |
|---|
| 1813 | param: ctrls string: controls string to be written on DIFFaX controls.dif file |
|---|
| 1814 | param: scale float: scale factor |
|---|
| 1815 | param: background dict: background parameters |
|---|
| 1816 | param: limits list: min/max 2-theta to be calculated |
|---|
| 1817 | param: inst dict: instrument parameters dictionary |
|---|
| 1818 | param: profile list: powder pattern data |
|---|
| 1819 | |
|---|
| 1820 | all updated in place |
|---|
| 1821 | ''' |
|---|
| 1822 | import atmdata |
|---|
| 1823 | path = sys.path |
|---|
| 1824 | for name in path: |
|---|
| 1825 | if 'bin' in name: |
|---|
| 1826 | DIFFaX = name+'/DIFFaX.exe' |
|---|
| 1827 | print ' Execute ',DIFFaX |
|---|
| 1828 | break |
|---|
| 1829 | # make form factor file that DIFFaX wants - atom types are GSASII style |
|---|
| 1830 | sf = open('data.sfc','w') |
|---|
| 1831 | sf.write('GSASII special form factor file for DIFFaX\n\n') |
|---|
| 1832 | atTypes = Layers['AtInfo'].keys() |
|---|
| 1833 | if 'H' not in atTypes: |
|---|
| 1834 | atTypes.insert(0,'H') |
|---|
| 1835 | for atType in atTypes: |
|---|
| 1836 | if atType == 'H': |
|---|
| 1837 | blen = -.3741 |
|---|
| 1838 | else: |
|---|
| 1839 | blen = Layers['AtInfo'][atType]['Isotopes']['Nat. Abund.']['SL'][0] |
|---|
| 1840 | Adat = atmdata.XrayFF[atType] |
|---|
| 1841 | text = '%4s'%(atType.ljust(4)) |
|---|
| 1842 | for i in range(4): |
|---|
| 1843 | text += '%11.6f%11.6f'%(Adat['fa'][i],Adat['fb'][i]) |
|---|
| 1844 | text += '%11.6f%11.6f'%(Adat['fc'],blen) |
|---|
| 1845 | text += '%3d\n'%(Adat['Z']) |
|---|
| 1846 | sf.write(text) |
|---|
| 1847 | sf.close() |
|---|
| 1848 | #make DIFFaX control.dif file - future use GUI to set some of these flags |
|---|
| 1849 | cf = open('control.dif','w') |
|---|
| 1850 | if ctrls == '0\n0\n3\n' or ctrls == '0\n1\n3\n': |
|---|
| 1851 | x0 = profile[0] |
|---|
| 1852 | iBeg = np.searchsorted(x0,limits[0]) |
|---|
| 1853 | iFin = np.searchsorted(x0,limits[1]) |
|---|
| 1854 | if iFin-iBeg > 20000: |
|---|
| 1855 | iFin = iBeg+20000 |
|---|
| 1856 | Dx = (x0[iFin]-x0[iBeg])/(iFin-iBeg) |
|---|
| 1857 | cf.write('GSASII-DIFFaX.dat\n'+ctrls) |
|---|
| 1858 | cf.write('%.6f %.6f %.6f\n1\n1\nend\n'%(x0[iBeg],x0[iFin],Dx)) |
|---|
| 1859 | else: |
|---|
| 1860 | cf.write('GSASII-DIFFaX.dat\n'+ctrls) |
|---|
| 1861 | inst = {'Type':['XSC','XSC',]} |
|---|
| 1862 | cf.close() |
|---|
| 1863 | #make DIFFaX data file |
|---|
| 1864 | df = open('GSASII-DIFFaX.dat','w') |
|---|
| 1865 | df.write('INSTRUMENTAL\n') |
|---|
| 1866 | if 'X' in inst['Type'][0]: |
|---|
| 1867 | df.write('X-RAY\n') |
|---|
| 1868 | elif 'N' in inst['Type'][0]: |
|---|
| 1869 | df.write('NEUTRON\n') |
|---|
| 1870 | if ctrls == '0\n0\n3\n' or ctrls == '0\n1\n3\n': |
|---|
| 1871 | df.write('%.4f\n'%(G2mth.getMeanWave(inst))) |
|---|
| 1872 | U = ateln2*inst['U'][1]/10000. |
|---|
| 1873 | V = ateln2*inst['V'][1]/10000. |
|---|
| 1874 | W = ateln2*inst['W'][1]/10000. |
|---|
| 1875 | HWHM = U*nptand(x0[iBeg:iFin]/2.)**2+V*nptand(x0[iBeg:iFin]/2.)+W |
|---|
| 1876 | HW = np.sqrt(np.mean(HWHM)) |
|---|
| 1877 | # df.write('PSEUDO-VOIGT 0.015 -0.0036 0.009 0.605 TRIM\n') |
|---|
| 1878 | if 'Mean' in Layers['selInst']: |
|---|
| 1879 | df.write('GAUSSIAN %.6f TRIM\n'%(HW)) #fast option - might not really matter |
|---|
| 1880 | elif 'Gaussian' in Layers['selInst']: |
|---|
| 1881 | df.write('GAUSSIAN %.6f %.6f %.6f TRIM\n'%(U,V,W)) #slow - make a GUI option? |
|---|
| 1882 | else: |
|---|
| 1883 | df.write('None\n') |
|---|
| 1884 | else: |
|---|
| 1885 | df.write('0.10\nNone\n') |
|---|
| 1886 | df.write('STRUCTURAL\n') |
|---|
| 1887 | a,b,c = Layers['Cell'][1:4] |
|---|
| 1888 | gam = Layers['Cell'][6] |
|---|
| 1889 | df.write('%.4f %.4f %.4f %.3f\n'%(a,b,c,gam)) |
|---|
| 1890 | laue = Layers['Laue'] |
|---|
| 1891 | if laue == '2/m(ab)': |
|---|
| 1892 | laue = '2/m(1)' |
|---|
| 1893 | elif laue == '2/m(c)': |
|---|
| 1894 | laue = '2/m(2)' |
|---|
| 1895 | if 'unknown' in Layers['Laue']: |
|---|
| 1896 | df.write('%s %.3f\n'%(laue,Layers['Toler'])) |
|---|
| 1897 | else: |
|---|
| 1898 | df.write('%s\n'%(laue)) |
|---|
| 1899 | df.write('%d\n'%(len(Layers['Layers']))) |
|---|
| 1900 | if Layers['Width'][0][0] < 1. or Layers['Width'][0][1] < 1.: |
|---|
| 1901 | df.write('%.1f %.1f\n'%(Layers['Width'][0][0]*10000.,Layers['Width'][0][0]*10000.)) #mum to A |
|---|
| 1902 | layerNames = [] |
|---|
| 1903 | for layer in Layers['Layers']: |
|---|
| 1904 | layerNames.append(layer['Name']) |
|---|
| 1905 | for il,layer in enumerate(Layers['Layers']): |
|---|
| 1906 | if layer['SameAs']: |
|---|
| 1907 | df.write('LAYER %d = %d\n'%(il+1,layerNames.index(layer['SameAs'])+1)) |
|---|
| 1908 | continue |
|---|
| 1909 | df.write('LAYER %d\n'%(il+1)) |
|---|
| 1910 | if '-1' in layer['Symm']: |
|---|
| 1911 | df.write('CENTROSYMMETRIC\n') |
|---|
| 1912 | else: |
|---|
| 1913 | df.write('NONE\n') |
|---|
| 1914 | for ia,atom in enumerate(layer['Atoms']): |
|---|
| 1915 | [name,atype,x,y,z,frac,Uiso] = atom |
|---|
| 1916 | if '-1' in layer['Symm'] and [x,y,z] == [0.,0.,0.]: |
|---|
| 1917 | frac /= 2. |
|---|
| 1918 | df.write('%4s %3d %.5f %.5f %.5f %.4f %.2f\n'%(atype.ljust(6),ia,x,y,z,78.9568*Uiso,frac)) |
|---|
| 1919 | df.write('STACKING\n') |
|---|
| 1920 | df.write('%s\n'%(Layers['Stacking'][0])) |
|---|
| 1921 | if 'recursive' in Layers['Stacking'][0]: |
|---|
| 1922 | df.write('%s\n'%Layers['Stacking'][1]) |
|---|
| 1923 | else: |
|---|
| 1924 | if 'list' in Layers['Stacking'][1]: |
|---|
| 1925 | Slen = len(Layers['Stacking'][2]) |
|---|
| 1926 | iB = 0 |
|---|
| 1927 | iF = 0 |
|---|
| 1928 | while True: |
|---|
| 1929 | iF += 68 |
|---|
| 1930 | if iF >= Slen: |
|---|
| 1931 | break |
|---|
| 1932 | iF = min(iF,Slen) |
|---|
| 1933 | df.write('%s\n'%(Layers['Stacking'][2][iB:iF])) |
|---|
| 1934 | iB = iF |
|---|
| 1935 | else: |
|---|
| 1936 | df.write('%s\n'%Layers['Stacking'][1]) |
|---|
| 1937 | df.write('TRANSITIONS\n') |
|---|
| 1938 | for iY in range(len(Layers['Layers'])): |
|---|
| 1939 | sumPx = 0. |
|---|
| 1940 | for iX in range(len(Layers['Layers'])): |
|---|
| 1941 | p,dx,dy,dz = Layers['Transitions'][iY][iX][:4] |
|---|
| 1942 | p = round(p,3) |
|---|
| 1943 | df.write('%.3f %.5f %.5f %.5f\n'%(p,dx,dy,dz)) |
|---|
| 1944 | sumPx += p |
|---|
| 1945 | if sumPx != 1.0: #this has to be picky since DIFFaX is. |
|---|
| 1946 | print 'ERROR - Layer probabilities sum to ',sumPx,' DIFFaX will insist it = 1.0' |
|---|
| 1947 | df.close() |
|---|
| 1948 | os.remove('data.sfc') |
|---|
| 1949 | os.remove('control.dif') |
|---|
| 1950 | os.remove('GSASII-DIFFaX.dat') |
|---|
| 1951 | return |
|---|
| 1952 | df.close() |
|---|
| 1953 | time0 = time.time() |
|---|
| 1954 | try: |
|---|
| 1955 | subp.call(DIFFaX) |
|---|
| 1956 | except OSError: |
|---|
| 1957 | print ' DIFFax.exe is not available for this platform - under development' |
|---|
| 1958 | print ' DIFFaX time = %.2fs'%(time.time()-time0) |
|---|
| 1959 | if os.path.exists('GSASII-DIFFaX.spc'): |
|---|
| 1960 | Xpat = np.loadtxt('GSASII-DIFFaX.spc').T |
|---|
| 1961 | iFin = iBeg+Xpat.shape[1] |
|---|
| 1962 | bakType,backDict,backVary = SetBackgroundParms(background) |
|---|
| 1963 | backDict['Lam1'] = G2mth.getWave(inst) |
|---|
| 1964 | profile[4][iBeg:iFin] = getBackground('',backDict,bakType,inst['Type'][0],profile[0][iBeg:iFin])[0] |
|---|
| 1965 | profile[3][iBeg:iFin] = Xpat[-1]*scale+profile[4][iBeg:iFin] |
|---|
| 1966 | if not np.any(profile[1]): #fill dummy data x,y,w,yc,yb,yd |
|---|
| 1967 | rv = st.poisson(profile[3][iBeg:iFin]) |
|---|
| 1968 | profile[1][iBeg:iFin] = rv.rvs() |
|---|
| 1969 | Z = np.ones_like(profile[3][iBeg:iFin]) |
|---|
| 1970 | Z[1::2] *= -1 |
|---|
| 1971 | profile[1][iBeg:iFin] = profile[3][iBeg:iFin]+np.abs(profile[1][iBeg:iFin]-profile[3][iBeg:iFin])*Z |
|---|
| 1972 | profile[2][iBeg:iFin] = np.where(profile[1][iBeg:iFin]>0.,1./profile[1][iBeg:iFin],1.0) |
|---|
| 1973 | profile[5][iBeg:iFin] = profile[1][iBeg:iFin]-profile[3][iBeg:iFin] |
|---|
| 1974 | #cleanup files.. |
|---|
| 1975 | os.remove('GSASII-DIFFaX.spc') |
|---|
| 1976 | elif os.path.exists('GSASII-DIFFaX.sadp'): |
|---|
| 1977 | Sadp = np.fromfile('GSASII-DIFFaX.sadp','>u2') |
|---|
| 1978 | Sadp = np.reshape(Sadp,(256,-1)) |
|---|
| 1979 | Layers['Sadp']['Img'] = Sadp |
|---|
| 1980 | os.remove('GSASII-DIFFaX.sadp') |
|---|
| 1981 | os.remove('data.sfc') |
|---|
| 1982 | os.remove('control.dif') |
|---|
| 1983 | os.remove('GSASII-DIFFaX.dat') |
|---|
| 1984 | |
|---|
| 1985 | def SetPWDRscan(inst,limits,profile): |
|---|
| 1986 | |
|---|
| 1987 | wave = G2mth.getMeanWave(inst) |
|---|
| 1988 | x0 = profile[0] |
|---|
| 1989 | iBeg = np.searchsorted(x0,limits[0]) |
|---|
| 1990 | iFin = np.searchsorted(x0,limits[1]) |
|---|
| 1991 | if iFin-iBeg > 20000: |
|---|
| 1992 | iFin = iBeg+20000 |
|---|
| 1993 | Dx = (x0[iFin]-x0[iBeg])/(iFin-iBeg) |
|---|
| 1994 | pyx.pygetinst(wave,x0[iBeg],x0[iFin],Dx) |
|---|
| 1995 | return iFin-iBeg |
|---|
| 1996 | |
|---|
| 1997 | def SetStackingSF(Layers,debug): |
|---|
| 1998 | # Load scattering factors into DIFFaX arrays |
|---|
| 1999 | import atmdata |
|---|
| 2000 | atTypes = Layers['AtInfo'].keys() |
|---|
| 2001 | aTypes = [] |
|---|
| 2002 | for atype in atTypes: |
|---|
| 2003 | aTypes.append('%4s'%(atype.ljust(4))) |
|---|
| 2004 | SFdat = [] |
|---|
| 2005 | for atType in atTypes: |
|---|
| 2006 | if atType == 'H': |
|---|
| 2007 | blen = -.3741 |
|---|
| 2008 | else: |
|---|
| 2009 | blen = Layers['AtInfo'][atType]['Isotopes']['Nat. Abund.']['SL'][0] |
|---|
| 2010 | Adat = atmdata.XrayFF[atType] |
|---|
| 2011 | SF = np.zeros(9) |
|---|
| 2012 | SF[:8:2] = Adat['fa'] |
|---|
| 2013 | SF[1:8:2] = Adat['fb'] |
|---|
| 2014 | SF[8] = Adat['fc'] |
|---|
| 2015 | SFdat.append(SF) |
|---|
| 2016 | SFdat = np.array(SFdat) |
|---|
| 2017 | pyx.pyloadscf(len(atTypes),aTypes,SFdat.T,debug) |
|---|
| 2018 | |
|---|
| 2019 | def SetStackingClay(Layers,Type): |
|---|
| 2020 | # Controls |
|---|
| 2021 | rand.seed() |
|---|
| 2022 | ranSeed = rand.randint(1,2**16-1) |
|---|
| 2023 | try: |
|---|
| 2024 | laueId = ['-1','2/m(ab)','2/m(c)','mmm','-3','-3m','4/m','4/mmm', |
|---|
| 2025 | '6/m','6/mmm'].index(Layers['Laue'])+1 |
|---|
| 2026 | except ValueError: #for 'unknown' |
|---|
| 2027 | laueId = -1 |
|---|
| 2028 | if 'SADP' in Type: |
|---|
| 2029 | planeId = ['h0l','0kl','hhl','h-hl'].index(Layers['Sadp']['Plane'])+1 |
|---|
| 2030 | lmax = int(Layers['Sadp']['Lmax']) |
|---|
| 2031 | else: |
|---|
| 2032 | planeId = 0 |
|---|
| 2033 | lmax = 0 |
|---|
| 2034 | # Sequences |
|---|
| 2035 | StkType = ['recursive','explicit'].index(Layers['Stacking'][0]) |
|---|
| 2036 | try: |
|---|
| 2037 | StkParm = ['infinite','random','list'].index(Layers['Stacking'][1]) |
|---|
| 2038 | except ValueError: |
|---|
| 2039 | StkParm = -1 |
|---|
| 2040 | if StkParm == 2: #list |
|---|
| 2041 | StkSeq = [int(val) for val in Layers['Stacking'][2].split()] |
|---|
| 2042 | Nstk = len(StkSeq) |
|---|
| 2043 | else: |
|---|
| 2044 | Nstk = 1 |
|---|
| 2045 | StkSeq = [0,] |
|---|
| 2046 | if StkParm == -1: |
|---|
| 2047 | StkParm = int(Layers['Stacking'][1]) |
|---|
| 2048 | Wdth = Layers['Width'][0] |
|---|
| 2049 | mult = 1 |
|---|
| 2050 | controls = [laueId,planeId,lmax,mult,StkType,StkParm,ranSeed] |
|---|
| 2051 | LaueSym = Layers['Laue'].ljust(12) |
|---|
| 2052 | pyx.pygetclay(controls,LaueSym,Wdth,Nstk,StkSeq) |
|---|
| 2053 | return laueId,controls |
|---|
| 2054 | |
|---|
| 2055 | def SetCellAtoms(Layers): |
|---|
| 2056 | Cell = Layers['Cell'][1:4]+Layers['Cell'][6:7] |
|---|
| 2057 | # atoms in layers |
|---|
| 2058 | atTypes = Layers['AtInfo'].keys() |
|---|
| 2059 | AtomXOU = [] |
|---|
| 2060 | AtomTp = [] |
|---|
| 2061 | LayerSymm = [] |
|---|
| 2062 | LayerNum = [] |
|---|
| 2063 | layerNames = [] |
|---|
| 2064 | Natm = 0 |
|---|
| 2065 | Nuniq = 0 |
|---|
| 2066 | for layer in Layers['Layers']: |
|---|
| 2067 | layerNames.append(layer['Name']) |
|---|
| 2068 | for il,layer in enumerate(Layers['Layers']): |
|---|
| 2069 | if layer['SameAs']: |
|---|
| 2070 | LayerNum.append(layerNames.index(layer['SameAs'])+1) |
|---|
| 2071 | continue |
|---|
| 2072 | else: |
|---|
| 2073 | LayerNum.append(il+1) |
|---|
| 2074 | Nuniq += 1 |
|---|
| 2075 | if '-1' in layer['Symm']: |
|---|
| 2076 | LayerSymm.append(1) |
|---|
| 2077 | else: |
|---|
| 2078 | LayerSymm.append(0) |
|---|
| 2079 | for ia,atom in enumerate(layer['Atoms']): |
|---|
| 2080 | [name,atype,x,y,z,frac,Uiso] = atom |
|---|
| 2081 | Natm += 1 |
|---|
| 2082 | AtomTp.append('%4s'%(atype.ljust(4))) |
|---|
| 2083 | Ta = atTypes.index(atype)+1 |
|---|
| 2084 | AtomXOU.append([float(Nuniq),float(ia+1),float(Ta),x,y,z,frac,Uiso*78.9568]) |
|---|
| 2085 | AtomXOU = np.array(AtomXOU) |
|---|
| 2086 | Nlayers = len(layerNames) |
|---|
| 2087 | pyx.pycellayer(Cell,Natm,AtomTp,AtomXOU.T,Nuniq,LayerSymm,Nlayers,LayerNum) |
|---|
| 2088 | return Nlayers |
|---|
| 2089 | |
|---|
| 2090 | def SetStackingTrans(Layers,Nlayers): |
|---|
| 2091 | # Transitions |
|---|
| 2092 | TransX = [] |
|---|
| 2093 | TransP = [] |
|---|
| 2094 | for Ytrans in Layers['Transitions']: |
|---|
| 2095 | TransP.append([trans[0] for trans in Ytrans]) #get just the numbers |
|---|
| 2096 | TransX.append([trans[1:4] for trans in Ytrans]) #get just the numbers |
|---|
| 2097 | TransP = np.array(TransP,dtype='float').T |
|---|
| 2098 | TransX = np.array(TransX,dtype='float') |
|---|
| 2099 | # GSASIIpath.IPyBreak() |
|---|
| 2100 | pyx.pygettrans(Nlayers,TransP,TransX) |
|---|
| 2101 | |
|---|
| 2102 | def CalcStackingPWDR(Layers,scale,background,limits,inst,profile,debug): |
|---|
| 2103 | # Scattering factors |
|---|
| 2104 | SetStackingSF(Layers,debug) |
|---|
| 2105 | # Controls & sequences |
|---|
| 2106 | laueId,controls = SetStackingClay(Layers,'PWDR') |
|---|
| 2107 | # cell & atoms |
|---|
| 2108 | Nlayers = SetCellAtoms(Layers) |
|---|
| 2109 | Volume = Layers['Cell'][7] |
|---|
| 2110 | # Transitions |
|---|
| 2111 | SetStackingTrans(Layers,Nlayers) |
|---|
| 2112 | # PWDR scan |
|---|
| 2113 | Nsteps = SetPWDRscan(inst,limits,profile) |
|---|
| 2114 | # result as Spec |
|---|
| 2115 | x0 = profile[0] |
|---|
| 2116 | profile[3] = np.zeros(len(profile[0])) |
|---|
| 2117 | profile[4] = np.zeros(len(profile[0])) |
|---|
| 2118 | profile[5] = np.zeros(len(profile[0])) |
|---|
| 2119 | iBeg = np.searchsorted(x0,limits[0]) |
|---|
| 2120 | iFin = np.searchsorted(x0,limits[1]) |
|---|
| 2121 | if iFin-iBeg > 20000: |
|---|
| 2122 | iFin = iBeg+20000 |
|---|
| 2123 | Nspec = 20001 |
|---|
| 2124 | spec = np.zeros(Nspec,dtype='double') |
|---|
| 2125 | time0 = time.time() |
|---|
| 2126 | pyx.pygetspc(controls,Nspec,spec) |
|---|
| 2127 | print ' GETSPC time = %.2fs'%(time.time()-time0) |
|---|
| 2128 | time0 = time.time() |
|---|
| 2129 | U = ateln2*inst['U'][1]/10000. |
|---|
| 2130 | V = ateln2*inst['V'][1]/10000. |
|---|
| 2131 | W = ateln2*inst['W'][1]/10000. |
|---|
| 2132 | HWHM = U*nptand(x0[iBeg:iFin]/2.)**2+V*nptand(x0[iBeg:iFin]/2.)+W |
|---|
| 2133 | HW = np.sqrt(np.mean(HWHM)) |
|---|
| 2134 | BrdSpec = np.zeros(Nsteps) |
|---|
| 2135 | if 'Mean' in Layers['selInst']: |
|---|
| 2136 | pyx.pyprofile(U,V,W,HW,1,Nsteps,BrdSpec) |
|---|
| 2137 | elif 'Gaussian' in Layers['selInst']: |
|---|
| 2138 | pyx.pyprofile(U,V,W,HW,4,Nsteps,BrdSpec) |
|---|
| 2139 | else: |
|---|
| 2140 | BrdSpec = spec[:Nsteps] |
|---|
| 2141 | BrdSpec /= Volume |
|---|
| 2142 | iFin = iBeg+Nsteps |
|---|
| 2143 | bakType,backDict,backVary = SetBackgroundParms(background) |
|---|
| 2144 | backDict['Lam1'] = G2mth.getWave(inst) |
|---|
| 2145 | profile[4][iBeg:iFin] = getBackground('',backDict,bakType,inst['Type'][0],profile[0][iBeg:iFin])[0] |
|---|
| 2146 | profile[3][iBeg:iFin] = BrdSpec*scale+profile[4][iBeg:iFin] |
|---|
| 2147 | if not np.any(profile[1]): #fill dummy data x,y,w,yc,yb,yd |
|---|
| 2148 | rv = st.poisson(profile[3][iBeg:iFin]) |
|---|
| 2149 | profile[1][iBeg:iFin] = rv.rvs() |
|---|
| 2150 | Z = np.ones_like(profile[3][iBeg:iFin]) |
|---|
| 2151 | Z[1::2] *= -1 |
|---|
| 2152 | profile[1][iBeg:iFin] = profile[3][iBeg:iFin]+np.abs(profile[1][iBeg:iFin]-profile[3][iBeg:iFin])*Z |
|---|
| 2153 | profile[2][iBeg:iFin] = np.where(profile[1][iBeg:iFin]>0.,1./profile[1][iBeg:iFin],1.0) |
|---|
| 2154 | profile[5][iBeg:iFin] = profile[1][iBeg:iFin]-profile[3][iBeg:iFin] |
|---|
| 2155 | print ' Broadening time = %.2fs'%(time.time()-time0) |
|---|
| 2156 | |
|---|
| 2157 | def CalcStackingSADP(Layers,debug): |
|---|
| 2158 | |
|---|
| 2159 | # Scattering factors |
|---|
| 2160 | SetStackingSF(Layers,debug) |
|---|
| 2161 | # Controls & sequences |
|---|
| 2162 | laueId,controls = SetStackingClay(Layers,'SADP') |
|---|
| 2163 | # cell & atoms |
|---|
| 2164 | Nlayers = SetCellAtoms(Layers) |
|---|
| 2165 | # Transitions |
|---|
| 2166 | SetStackingTrans(Layers,Nlayers) |
|---|
| 2167 | # result as Sadp |
|---|
| 2168 | mirror = laueId in [-1,2,3,7,8,9,10] |
|---|
| 2169 | Nspec = 20001 |
|---|
| 2170 | spec = np.zeros(Nspec,dtype='double') |
|---|
| 2171 | time0 = time.time() |
|---|
| 2172 | hkLim,Incr,Nblk = pyx.pygetsadp(controls,Nspec,spec) |
|---|
| 2173 | Sapd = np.zeros((256,256)) |
|---|
| 2174 | iB = 0 |
|---|
| 2175 | for i in range(hkLim): |
|---|
| 2176 | iF = iB+Nblk |
|---|
| 2177 | p1 = 127+int(i*Incr) |
|---|
| 2178 | p2 = 128-int(i*Incr) |
|---|
| 2179 | if Nblk == 128: |
|---|
| 2180 | if i: |
|---|
| 2181 | Sapd[128:,p1] = spec[iB:iF] |
|---|
| 2182 | Sapd[:128,p1] = spec[iF:iB:-1] |
|---|
| 2183 | Sapd[128:,p2] = spec[iB:iF] |
|---|
| 2184 | Sapd[:128,p2] = spec[iF:iB:-1] |
|---|
| 2185 | else: |
|---|
| 2186 | if i: |
|---|
| 2187 | Sapd[:,p1] = spec[iB:iF] |
|---|
| 2188 | Sapd[:,p2] = spec[iB:iF] |
|---|
| 2189 | iB += Nblk |
|---|
| 2190 | Layers['Sadp']['Img'] = Sapd |
|---|
| 2191 | print ' GETSAD time = %.2fs'%(time.time()-time0) |
|---|
| 2192 | # GSASIIpath.IPyBreak() |
|---|
| 2193 | |
|---|
| 2194 | #testing data |
|---|
| 2195 | NeedTestData = True |
|---|
| 2196 | def TestData(): |
|---|
| 2197 | 'needs a doc string' |
|---|
| 2198 | # global NeedTestData |
|---|
| 2199 | NeedTestData = False |
|---|
| 2200 | global bakType |
|---|
| 2201 | bakType = 'chebyschev' |
|---|
| 2202 | global xdata |
|---|
| 2203 | xdata = np.linspace(4.0,40.0,36000) |
|---|
| 2204 | global parmDict0 |
|---|
| 2205 | parmDict0 = { |
|---|
| 2206 | 'pos0':5.6964,'int0':8835.8,'sig0':1.0,'gam0':1.0, |
|---|
| 2207 | 'pos1':11.4074,'int1':3922.3,'sig1':1.0,'gam1':1.0, |
|---|
| 2208 | 'pos2':20.6426,'int2':1573.7,'sig2':1.0,'gam2':1.0, |
|---|
| 2209 | 'pos3':26.9568,'int3':925.1,'sig3':1.0,'gam3':1.0, |
|---|
| 2210 | 'U':1.163,'V':-0.605,'W':0.093,'X':0.0,'Y':2.183,'SH/L':0.002, |
|---|
| 2211 | 'Back0':5.384,'Back1':-0.015,'Back2':.004, |
|---|
| 2212 | } |
|---|
| 2213 | global parmDict1 |
|---|
| 2214 | parmDict1 = { |
|---|
| 2215 | 'pos0':13.4924,'int0':48697.6,'sig0':1.0,'gam0':1.0, |
|---|
| 2216 | 'pos1':23.4360,'int1':43685.5,'sig1':1.0,'gam1':1.0, |
|---|
| 2217 | 'pos2':27.1152,'int2':123712.6,'sig2':1.0,'gam2':1.0, |
|---|
| 2218 | 'pos3':33.7196,'int3':65349.4,'sig3':1.0,'gam3':1.0, |
|---|
| 2219 | 'pos4':36.1119,'int4':115829.8,'sig4':1.0,'gam4':1.0, |
|---|
| 2220 | 'pos5':39.0122,'int5':6916.9,'sig5':1.0,'gam5':1.0, |
|---|
| 2221 | 'U':22.75,'V':-17.596,'W':10.594,'X':1.577,'Y':5.778,'SH/L':0.002, |
|---|
| 2222 | 'Back0':36.897,'Back1':-0.508,'Back2':.006, |
|---|
| 2223 | 'Lam1':1.540500,'Lam2':1.544300,'I(L2)/I(L1)':0.5, |
|---|
| 2224 | } |
|---|
| 2225 | global parmDict2 |
|---|
| 2226 | parmDict2 = { |
|---|
| 2227 | 'pos0':5.7,'int0':1000.0,'sig0':0.5,'gam0':0.5, |
|---|
| 2228 | 'U':2.,'V':-2.,'W':5.,'X':0.5,'Y':0.5,'SH/L':0.02, |
|---|
| 2229 | 'Back0':5.,'Back1':-0.02,'Back2':.004, |
|---|
| 2230 | # 'Lam1':1.540500,'Lam2':1.544300,'I(L2)/I(L1)':0.5, |
|---|
| 2231 | } |
|---|
| 2232 | global varyList |
|---|
| 2233 | varyList = [] |
|---|
| 2234 | |
|---|
| 2235 | def test0(): |
|---|
| 2236 | if NeedTestData: TestData() |
|---|
| 2237 | msg = 'test ' |
|---|
| 2238 | gplot = plotter.add('FCJ-Voigt, 11BM').gca() |
|---|
| 2239 | gplot.plot(xdata,getBackground('',parmDict0,bakType,'PXC',xdata)[0]) |
|---|
| 2240 | gplot.plot(xdata,getPeakProfile(parmDict0,xdata,varyList,bakType)) |
|---|
| 2241 | fplot = plotter.add('FCJ-Voigt, Ka1+2').gca() |
|---|
| 2242 | fplot.plot(xdata,getBackground('',parmDict1,bakType,'PXC',xdata)[0]) |
|---|
| 2243 | fplot.plot(xdata,getPeakProfile(parmDict1,xdata,varyList,bakType)) |
|---|
| 2244 | |
|---|
| 2245 | def test1(): |
|---|
| 2246 | if NeedTestData: TestData() |
|---|
| 2247 | time0 = time.time() |
|---|
| 2248 | for i in range(100): |
|---|
| 2249 | y = getPeakProfile(parmDict1,xdata,varyList,bakType) |
|---|
| 2250 | print '100+6*Ka1-2 peaks=1200 peaks',time.time()-time0 |
|---|
| 2251 | |
|---|
| 2252 | def test2(name,delt): |
|---|
| 2253 | if NeedTestData: TestData() |
|---|
| 2254 | varyList = [name,] |
|---|
| 2255 | xdata = np.linspace(5.6,5.8,400) |
|---|
| 2256 | hplot = plotter.add('derivatives test for '+name).gca() |
|---|
| 2257 | hplot.plot(xdata,getPeakProfileDerv(parmDict2,xdata,varyList,bakType)[0]) |
|---|
| 2258 | y0 = getPeakProfile(parmDict2,xdata,varyList,bakType) |
|---|
| 2259 | parmDict2[name] += delt |
|---|
| 2260 | y1 = getPeakProfile(parmDict2,xdata,varyList,bakType) |
|---|
| 2261 | hplot.plot(xdata,(y1-y0)/delt,'r+') |
|---|
| 2262 | |
|---|
| 2263 | def test3(name,delt): |
|---|
| 2264 | if NeedTestData: TestData() |
|---|
| 2265 | names = ['pos','sig','gam','shl'] |
|---|
| 2266 | idx = names.index(name) |
|---|
| 2267 | myDict = {'pos':parmDict2['pos0'],'sig':parmDict2['sig0'],'gam':parmDict2['gam0'],'shl':parmDict2['SH/L']} |
|---|
| 2268 | xdata = np.linspace(5.6,5.8,800) |
|---|
| 2269 | dx = xdata[1]-xdata[0] |
|---|
| 2270 | hplot = plotter.add('derivatives test for '+name).gca() |
|---|
| 2271 | hplot.plot(xdata,100.*dx*getdFCJVoigt3(myDict['pos'],myDict['sig'],myDict['gam'],myDict['shl'],xdata)[idx+1]) |
|---|
| 2272 | y0 = getFCJVoigt3(myDict['pos'],myDict['sig'],myDict['gam'],myDict['shl'],xdata) |
|---|
| 2273 | myDict[name] += delt |
|---|
| 2274 | y1 = getFCJVoigt3(myDict['pos'],myDict['sig'],myDict['gam'],myDict['shl'],xdata) |
|---|
| 2275 | hplot.plot(xdata,(y1-y0)/delt,'r+') |
|---|
| 2276 | |
|---|
| 2277 | if __name__ == '__main__': |
|---|
| 2278 | import GSASIItestplot as plot |
|---|
| 2279 | global plotter |
|---|
| 2280 | plotter = plot.PlotNotebook() |
|---|
| 2281 | # test0() |
|---|
| 2282 | # for name in ['int0','pos0','sig0','gam0','U','V','W','X','Y','SH/L','I(L2)/I(L1)']: |
|---|
| 2283 | for name,shft in [['int0',0.1],['pos0',0.0001],['sig0',0.01],['gam0',0.00001], |
|---|
| 2284 | ['U',0.1],['V',0.01],['W',0.01],['X',0.0001],['Y',0.0001],['SH/L',0.00005]]: |
|---|
| 2285 | test2(name,shft) |
|---|
| 2286 | for name,shft in [['pos',0.0001],['sig',0.01],['gam',0.0001],['shl',0.00005]]: |
|---|
| 2287 | test3(name,shft) |
|---|
| 2288 | print "OK" |
|---|
| 2289 | plotter.StartEventLoop() |
|---|
