1 | #/usr/bin/env python |
---|
2 | # -*- coding: utf-8 -*- |
---|
3 | #GSASII powder calculation module |
---|
4 | ########### SVN repository information ################### |
---|
5 | # $Date: 2011-04-20 13:09:53 -0500 (Wed, 20 Apr 2011) $ |
---|
6 | # $Author: vondreele $ |
---|
7 | # $Revision: 267 $ |
---|
8 | # $URL: https://subversion.xor.aps.anl.gov/pyGSAS/trunk/GSASIIpwd.py $ |
---|
9 | # $Id: GSASIIpwd.py 267 2011-04-20 18:09:53Z vondreele $ |
---|
10 | ########### SVN repository information ################### |
---|
11 | import sys |
---|
12 | import math |
---|
13 | import wx |
---|
14 | import time |
---|
15 | |
---|
16 | import numpy as np |
---|
17 | import scipy as sp |
---|
18 | import numpy.linalg as nl |
---|
19 | from numpy.fft import ifft, fft, fftshift |
---|
20 | import scipy.interpolate as si |
---|
21 | import scipy.stats as st |
---|
22 | import scipy.optimize as so |
---|
23 | |
---|
24 | import GSASIIpath |
---|
25 | import GSASIIplot as G2plt |
---|
26 | import GSASIIlattice as G2lat |
---|
27 | import GSASIIElem as G2elem |
---|
28 | import GSASIIgrid as G2gd |
---|
29 | import GSASIIIO as G2IO |
---|
30 | import pypowder as pyd |
---|
31 | |
---|
32 | # trig functions in degrees |
---|
33 | sind = lambda x: math.sin(x*math.pi/180.) |
---|
34 | asind = lambda x: 180.*math.asin(x)/math.pi |
---|
35 | tand = lambda x: math.tan(x*math.pi/180.) |
---|
36 | atand = lambda x: 180.*math.atan(x)/math.pi |
---|
37 | atan2d = lambda y,x: 180.*math.atan2(y,x)/math.pi |
---|
38 | cosd = lambda x: math.cos(x*math.pi/180.) |
---|
39 | acosd = lambda x: 180.*math.acos(x)/math.pi |
---|
40 | rdsq2d = lambda x,p: round(1.0/math.sqrt(x),p) |
---|
41 | #numpy versions |
---|
42 | npsind = lambda x: np.sin(x*np.pi/180.) |
---|
43 | npasind = lambda x: 180.*np.arcsin(x)/math.pi |
---|
44 | npcosd = lambda x: np.cos(x*math.pi/180.) |
---|
45 | npacosd = lambda x: 180.*np.arccos(x)/math.pi |
---|
46 | nptand = lambda x: np.tan(x*math.pi/180.) |
---|
47 | npatand = lambda x: 180.*np.arctan(x)/np.pi |
---|
48 | npatan2d = lambda y,x: 180.*np.arctan2(y,x)/np.pi |
---|
49 | npT2stl = lambda tth, wave: 2.0*npsind(tth/2.0)/wave |
---|
50 | npT2q = lambda tth,wave: 2.0*np.pi*npT2stl(tth,wave) |
---|
51 | |
---|
52 | #GSASII pdf calculation routines |
---|
53 | |
---|
54 | def Transmission(Geometry,Abs,Diam): |
---|
55 | #Calculate sample transmission |
---|
56 | # Geometry: one of 'Cylinder','Bragg-Brentano','Tilting flat plate in transmission','Fixed flat plate' |
---|
57 | # Abs: absorption coeff in cm-1 |
---|
58 | # Diam: sample thickness/diameter in mm |
---|
59 | if 'Cylinder' in Geometry: #Lobanov & Alte da Veiga for 2-theta = 0; beam fully illuminates sample |
---|
60 | MuR = Abs*Diam/20.0 |
---|
61 | if MuR <= 3.0: |
---|
62 | T0 = 16/(3.*math.pi) |
---|
63 | T1 = -0.045780 |
---|
64 | T2 = -0.02489 |
---|
65 | T3 = 0.003045 |
---|
66 | T = -T0*MuR-T1*MuR**2-T2*MuR**3-T3*MuR**4 |
---|
67 | if T < -20.: |
---|
68 | return 2.06e-9 |
---|
69 | else: |
---|
70 | return math.exp(T) |
---|
71 | else: |
---|
72 | T1 = 1.433902 |
---|
73 | T2 = 0.013869+0.337894 |
---|
74 | T3 = 1.933433+1.163198 |
---|
75 | T4 = 0.044365-0.04259 |
---|
76 | T = (T1-T4)/(1.0+T2*(MuR-3.0))**T3+T4 |
---|
77 | return T/100. |
---|
78 | elif 'plate' in Geometry: |
---|
79 | MuR = Abs*Diam/10. |
---|
80 | return math.exp(-MuR) |
---|
81 | elif 'Bragg' in Geometry: |
---|
82 | return 0.0 |
---|
83 | |
---|
84 | def Absorb(Geometry,Abs,Diam,Tth,Phi=0,Psi=0): |
---|
85 | #Calculate sample absorption |
---|
86 | # Geometry: one of 'Cylinder','Bragg-Brentano','Tilting Flat Plate in transmission','Fixed flat plate' |
---|
87 | # Abs: absorption coeff in cm-1 |
---|
88 | # Diam: sample thickness/diameter in mm |
---|
89 | # Tth: 2-theta scattering angle - can be numpy array |
---|
90 | # Phi: flat plate tilt angle - future |
---|
91 | # Psi: flat plate tilt axis - future |
---|
92 | Sth2 = npsind(Tth/2.0)**2 |
---|
93 | Cth2 = 1.-Sth2 |
---|
94 | if 'Cylinder' in Geometry: #Lobanov & Alte da Veiga for 2-theta = 0; beam fully illuminates sample |
---|
95 | MuR = Abs*Diam/20.0 |
---|
96 | if MuR < 3.0: |
---|
97 | T0 = 16.0/(3*np.pi) |
---|
98 | T1 = (25.99978-0.01911*Sth2**0.25)*np.exp(-0.024551*Sth2)+ \ |
---|
99 | 0.109561*np.sqrt(Sth2)-26.04556 |
---|
100 | T2 = -0.02489-0.39499*Sth2+1.219077*Sth2**1.5- \ |
---|
101 | 1.31268*Sth2**2+0.871081*Sth2**2.5-0.2327*Sth2**3 |
---|
102 | T3 = 0.003045+0.018167*Sth2-0.03305*Sth2**2 |
---|
103 | Trns = -T0*MuR-T1*MuR**2-T2*MuR**3-T3*MuR**4 |
---|
104 | return np.exp(Trns) |
---|
105 | else: |
---|
106 | T1 = 1.433902+11.07504*Sth2-8.77629*Sth2*Sth2+ \ |
---|
107 | 10.02088*Sth2**3-3.36778*Sth2**4 |
---|
108 | T2 = (0.013869-0.01249*Sth2)*np.exp(3.27094*Sth2)+ \ |
---|
109 | (0.337894+13.77317*Sth2)/(1.0+11.53544*Sth2)**1.555039 |
---|
110 | T3 = 1.933433/(1.0+23.12967*Sth2)**1.686715- \ |
---|
111 | 0.13576*np.sqrt(Sth2)+1.163198 |
---|
112 | T4 = 0.044365-0.04259/(1.0+0.41051*Sth2)**148.4202 |
---|
113 | Trns = (T1-T4)/(1.0+T2*(MuR-3.0))**T3+T4 |
---|
114 | return Trns/100. |
---|
115 | elif 'Bragg' in Geometry: |
---|
116 | return 1.0 |
---|
117 | elif 'Fixed' in Geometry: #assumes sample plane is perpendicular to incident beam |
---|
118 | # and only defined for 2theta < 90 |
---|
119 | MuR = Abs*Diam/10.0 |
---|
120 | T1 = np.exp(-MuR) |
---|
121 | T2 = np.exp(-MuR/npcosd(Tth)) |
---|
122 | Tb = MuR-MuR/npcosd(Tth) |
---|
123 | return (T2-T1)/Tb |
---|
124 | elif 'Tilting' in Geometry: #assumes symmetric tilt so sample plane is parallel to diffraction vector |
---|
125 | MuR = Abs*Diam/10.0 |
---|
126 | cth = npcosd(Tth/2.0) |
---|
127 | return np.exp(-MuR/cth)/cth |
---|
128 | |
---|
129 | def Polarization(Pola,Tth,Azm=0.0): |
---|
130 | # Calculate angle dependent x-ray polarization correction (not scaled correctly!) |
---|
131 | # Pola: polarization coefficient e.g 1.0 fully polarized, 0.5 unpolarized |
---|
132 | # Azm: azimuthal angle e.g. 0.0 in plane of polarization |
---|
133 | # Tth: 2-theta scattering angle - can be numpy array |
---|
134 | # which (if either) of these is "right"? |
---|
135 | # return (Pola*npcosd(Azm)**2+(1.-Pola)*npsind(Azm)**2)*npcosd(Tth)**2+ \ |
---|
136 | # Pola*npsind(Azm)**2+(1.-Pola)*npcosd(Azm)**2 |
---|
137 | pola = ((1.0-Pola)*npcosd(Azm)**2+Pola*npsind(Azm)**2)*npcosd(Tth)**2+ \ |
---|
138 | (1.0-Pola)*npsind(Azm)**2+Pola*npcosd(Azm)**2 |
---|
139 | dpdPola = npsind(Azm)**2*npsind(Tth)**2 |
---|
140 | return pola,dpdPola |
---|
141 | |
---|
142 | def Oblique(ObCoeff,Tth): |
---|
143 | if ObCoeff: |
---|
144 | return (1.-ObCoeff)/(1.0-np.exp(np.log(ObCoeff)/npcosd(Tth))) |
---|
145 | else: |
---|
146 | return 1.0 |
---|
147 | |
---|
148 | def Ruland(RulCoff,wave,Q,Compton): |
---|
149 | C = 2.9978e8 |
---|
150 | D = 1.5e-3 |
---|
151 | hmc = 0.024262734687 |
---|
152 | sinth2 = (Q*wave/(4.0*np.pi))**2 |
---|
153 | dlam = (wave**2)*Compton*Q/C |
---|
154 | dlam_c = 2.0*hmc*sinth2-D*wave**2 |
---|
155 | return 1.0/((1.0+dlam/RulCoff)*(1.0+(np.pi*dlam_c/(dlam+RulCoff))**2)) |
---|
156 | |
---|
157 | def LorchWeight(Q): |
---|
158 | return np.sin(np.pi*(Q[-1]-Q)/(2.0*Q[-1])) |
---|
159 | |
---|
160 | def GetAsfMean(ElList,Sthl2): |
---|
161 | # Calculate various scattering factor terms for PDF calcs |
---|
162 | # ElList: element dictionary contains scattering factor coefficients, etc. |
---|
163 | # Sthl2: numpy array of sin theta/lambda squared values |
---|
164 | # returns: mean(f^2), mean(f)^2, mean(compton) |
---|
165 | sumNoAtoms = 0.0 |
---|
166 | FF = np.zeros_like(Sthl2) |
---|
167 | FF2 = np.zeros_like(Sthl2) |
---|
168 | CF = np.zeros_like(Sthl2) |
---|
169 | for El in ElList: |
---|
170 | sumNoAtoms += ElList[El]['FormulaNo'] |
---|
171 | for El in ElList: |
---|
172 | el = ElList[El] |
---|
173 | ff2 = (G2elem.ScatFac(el,Sthl2)+el['fp'])**2+el['fpp']**2 |
---|
174 | cf = G2elem.ComptonFac(el,Sthl2) |
---|
175 | FF += np.sqrt(ff2)*el['FormulaNo']/sumNoAtoms |
---|
176 | FF2 += ff2*el['FormulaNo']/sumNoAtoms |
---|
177 | CF += cf*el['FormulaNo']/sumNoAtoms |
---|
178 | return FF2,FF**2,CF |
---|
179 | |
---|
180 | def GetNumDensity(ElList,Vol): |
---|
181 | sumNoAtoms = 0.0 |
---|
182 | for El in ElList: |
---|
183 | sumNoAtoms += ElList[El]['FormulaNo'] |
---|
184 | return sumNoAtoms/Vol |
---|
185 | |
---|
186 | def MultGetQ(Tth,MuT,Geometry,b=88.0,a=0.01): |
---|
187 | NS = 500 |
---|
188 | Gama = np.linspace(0.,np.pi/2.,NS,False)[1:] |
---|
189 | Cgama = np.cos(Gama)[:,np.newaxis] |
---|
190 | Sgama = np.sin(Gama)[:,np.newaxis] |
---|
191 | CSgama = 1.0/Sgama |
---|
192 | Delt = Gama[1]-Gama[0] |
---|
193 | emc = 7.94e-26 |
---|
194 | Navo = 6.023e23 |
---|
195 | Cth = npcosd(Tth/2.0) |
---|
196 | CTth = npcosd(Tth) |
---|
197 | Sth = npcosd(Tth/2.0) |
---|
198 | STth = npsind(Tth) |
---|
199 | CSth = 1.0/Sth |
---|
200 | CSTth = 1.0/STth |
---|
201 | SCth = 1.0/Cth |
---|
202 | SCTth = 1.0/CTth |
---|
203 | if 'Bragg' in Geometry: |
---|
204 | G = 8.0*Delt*Navo*emc*Sth/((1.0-CTth**2)*(1.0-np.exp(-2.0*MuT*CSth))) |
---|
205 | Y1 = np.pi |
---|
206 | Y2 = np.pi/2.0 |
---|
207 | Y3 = 3.*np.pi/8. #3pi/4? |
---|
208 | W = 1.0/(Sth+np.fabs(Sgama)) |
---|
209 | W += np.exp(-MuT*CSth)*(2.0*np.fabs(Sgama)*np.exp(-MuT*np.fabs(CSgama))- |
---|
210 | (Sth+np.fabs(Sgama))*np.exp(-MuT*CSth))/(Sth**2-Sgama**2) |
---|
211 | Fac0 = Sth**2*Sgama**2 |
---|
212 | X = Fac0+(Fac0+CTth)**2/2 |
---|
213 | Y = Cgama**2*Cth**2*(1.0-Fac0-CTth) |
---|
214 | Z = Cgama**4*Cth**4/2.0 |
---|
215 | E = 2.0*(1.0-a)/(b*Cgama/Cth) |
---|
216 | F1 = (2.0+b*(1.0+Sth*Sgama))/(b*Cth*Cgama) #trouble if < 1 |
---|
217 | F2 = (2.0+b*(1.0-Sth*Sgama))/(b*Cth*Cgama) |
---|
218 | T1 = np.pi/np.sqrt(F1**2-1.0) |
---|
219 | T2 = np.pi/np.sqrt(F2**2-1.0) |
---|
220 | Y4 = T1+T2 |
---|
221 | Y5 = F1**2*T1+F2**2*T2-np.pi*(F1+F2) |
---|
222 | Y6 = F1**4*T1+F2**4*T2-np.pi*(F1+F2)/2.0-np.pi*(F1**3+F2**3) |
---|
223 | Y7 = (T2-T1)/(F1-F2) |
---|
224 | YT = F2**2*T2-F1**2*T1 |
---|
225 | Y8 = Y1+YT/(F1-F2) |
---|
226 | Y9 = Y2+(F2**4*T2-F1**4*T1)/(F1-F2)+Y1*((F1+F2)**2-F1*F2) |
---|
227 | M = (a**2*(X*Y1+Y*Y2+Z*Y3)+a*E*(X*Y4+Y*Y5+Z*Y6)+E**2*(X*Y7+Y*Y8+Z*Y9))*Cgama |
---|
228 | |
---|
229 | Q = np.sum(W*M,axis=0) |
---|
230 | return Q*G*NS/(NS-1.) |
---|
231 | # |
---|
232 | # cos2th=2.0d*costh^2 - 1.0d |
---|
233 | # G= delta * 8.0d * Na * emc * sinth/(1.0d + cos2th^2)/(1.0d - exp(-2.0d*mut*cscth)) |
---|
234 | # Y1=3.1415926d |
---|
235 | # Y2=Y1*0.5d |
---|
236 | # Y3=Y2*0.75d |
---|
237 | # for i=1,num_steps-1 do begin |
---|
238 | # cosgama=double(cos(gama[i])) |
---|
239 | # singama=double(sin(gama[i])) |
---|
240 | # cscgama=1.0d / singama |
---|
241 | # |
---|
242 | # W=1.0d/(sinth+abs(singama)) |
---|
243 | # W=W+exp(-1.0*mut*cscth)*(2.0d*abs(singama)*exp(-1.0d*mut*abs(cscgama))- $ |
---|
244 | # (sinth+abs(singama))*exp(-1.0d*mut*cscth))/(sinth^2-singama^2) |
---|
245 | # |
---|
246 | # factor0=sinth^2*singama^2 |
---|
247 | # X=factor0+(factor0+cos2th)^2/2.0d |
---|
248 | # Y=cosgama^2*(1.0d - factor0-cos2th)*costh^2 |
---|
249 | # Z=cosgama^4/2.0d*costh^4 |
---|
250 | # E=2.0d*(1.0-a)/b/cosgama/costh |
---|
251 | # |
---|
252 | # F1=1.0d/b/cosgama*(2.0d + b*(1.0+sinth*singama))/costh |
---|
253 | # F2=1.0d/b/cosgama*(2.0d + b*(1.0-sinth*singama))/costh |
---|
254 | # |
---|
255 | # T1=3.14159/sqrt(F1^2-1.0d) |
---|
256 | # T2=3.14159/sqrt(F2^2-1.0d) |
---|
257 | # Y4=T1+T2 |
---|
258 | # Y5=F1^2*T1+F2^2*T2-3.14159*(F1+F2) |
---|
259 | # Y6=F1^4*T1+F2^4*T2-3.14159*(F1+F2)/2.0-3.14159*(F1^3+F2^3) |
---|
260 | # Y7=(T2-T1)/(F1-F2) |
---|
261 | # Y8=Y1+(F2^2*T2-F1^2*T1)/(F1-F2) |
---|
262 | # Y9=Y2+(F2^4*T2-F1^4*T1)/(F1-F2)+Y1*((F1+F2)^2-F1*F2) |
---|
263 | # M=(a^2*(X*Y1+Y*Y2+Z*Y3)+a*E*(X*Y4+Y*Y5+Z*Y6)+E^2* $ |
---|
264 | # (X*Y7+Y*Y8+Z*Y9))*cosgama |
---|
265 | # |
---|
266 | # Q=Q+W*M |
---|
267 | # |
---|
268 | # endfor |
---|
269 | # Q=double(num_steps)/Double(num_steps-1)*Q*G |
---|
270 | # end |
---|
271 | elif 'Cylinder' in Geometry: |
---|
272 | Q = 0. |
---|
273 | elif 'Fixed' in Geometry: #Dwiggens & Park, Acta Cryst. A27, 264 (1971) with corrections |
---|
274 | EMA = np.exp(-MuT*(1.0-SCTth)) |
---|
275 | Fac1 = (1.-EMA)/(1.0-SCTth) |
---|
276 | G = 2.0*Delt*Navo*emc/((1.0+CTth**2)*Fac1) |
---|
277 | Fac0 = Cgama/(1-Sgama*SCTth) |
---|
278 | Wp = Fac0*(Fac1-(EMA-np.exp(-MuT*(CSgama-SCTth)))/(CSgama-1.0)) |
---|
279 | Fac0 = Cgama/(1.0+Sgama*SCTth) |
---|
280 | Wm = Fac0*(Fac1+(np.exp(-MuT*(1.0+CSgama))-1.0)/(CSgama+1.0)) |
---|
281 | X = (Sgama**2+CTth**2*(1.0-Sgama**2+Sgama**4))/2.0 |
---|
282 | Y = Sgama**3*Cgama*STth*CTth |
---|
283 | Z = Cgama**2*(1.0+Sgama**2)*STth**2/2.0 |
---|
284 | Fac2 = 1.0/(b*Cgama*STth) |
---|
285 | U = 2.0*(1.0-a)*Fac2 |
---|
286 | V = (2.0+b*(1.0-CTth*Sgama))*Fac2 |
---|
287 | Mp = 2.0*np.pi*(a+2.0*(1.0-a)/(2.0+b*(1.0-Sgama)))*(a*X+a*Z/2.0-U*Y+U*(X+Y*V+Z*V**2)/np.sqrt(V**2-1.0)-U*Z*V) |
---|
288 | V = (2.0+b*(1.0+CTth*Sgama))*Fac2 |
---|
289 | Y = -Y |
---|
290 | Mm = 2.0*np.pi*(a+2.0*(1.0-a)/(2.0+b*(1.0+Sgama)))*(a*X+a*Z/2.0-U*Y+U*(X+Y*V+Z*V**2)/np.sqrt(V**2-1.0)-U*Z*V) |
---|
291 | Q = np.sum(Wp*Mp+Wm*Mm,axis=0) |
---|
292 | return Q*G*NS/(NS-1.) |
---|
293 | elif 'Tilting' in Geometry: |
---|
294 | EMA = np.exp(-MuT*SCth) |
---|
295 | G = 2.0*Delt*Navo*emc/((1.0+CTth**2)*EMA) |
---|
296 | # Wplus = expmutsecth/(1.0d - singama*secth) + singama/mut/(1.0 -singama*secth)/(1.0-singama*secth)* $ |
---|
297 | # (Exp(-1.0*mut*cscgama) - expmutsecth) |
---|
298 | # Wminus = expmutsecth/(1.0d + singama*secth) - singama/mut/(1.0 +singama*secth)/(1.0+singama*secth)* $ |
---|
299 | # expmutsecth*(1.0d - expmutsecth*Exp(-1.0*mut*cscgama)) |
---|
300 | Wp = EMA/(1.0-Sgama*SCth)+Sgama/MuT/(1.0-Sgama*SCth)/(1.0-Sgama*SCth)*(np.exp(-MuT*CSgama)-EMA) |
---|
301 | # Wp = EMA/(1.0-Sgama*SCth)+Sgama/MuT/(1.0-Sgama*SCth)**2*(np.exp(-MuT*CSgama)-EMA) |
---|
302 | Wm = EMA/(1.0+Sgama*SCth)-Sgama/MuT/(1.0+Sgama*SCth)/(1.0+Sgama*SCth)*EMA*(1.0-EMA*np.exp(-MuT*CSgama)) |
---|
303 | # Wm = EMA/(1.0+Sgama*SCth)-Sgama/MuT/(1.0+Sgama*SCth)**2*EMA*(1.0-EMA*np.exp(-MuT*CSgama)) |
---|
304 | X = 0.5*(Cth**2*(Cth**2*Sgama**4-4.0*Sth**2*Cgama**2)+1.0) |
---|
305 | Y = Cgama**2*(1.0+Cgama**2)*Cth**2*Sth**2 |
---|
306 | Z = 0.5*Cgama**4*Sth**4 |
---|
307 | # X = 0.5*(costh*costh*(costh*costh*singama*singama*singama*singama - $ |
---|
308 | # 4.0*sinth*sinth*cosgama*cosgama) +1.0d) |
---|
309 | # |
---|
310 | # Y = cosgama*cosgama*(1.0 + cosgama*cosgama)*costh*costh*sinth*sinth |
---|
311 | # |
---|
312 | # Z= 0.5 *cosgama*cosgama*cosgama*cosgama* (sinth^4) |
---|
313 | # |
---|
314 | AlP = 2.0+b*(1.0-Cth*Sgama) |
---|
315 | AlM = 2.0+b*(1.0+Cth*Sgama) |
---|
316 | # alphaplus = 2.0 + b*(1.0 - costh*singama) |
---|
317 | # alphaminus = 2.0 + b*(1.0 + costh*singama) |
---|
318 | BeP = np.sqrt(np.fabs(AlP**2-(b*Cgama*Sth)**2)) |
---|
319 | BeM = np.sqrt(np.fabs(AlM**2-(b*Cgama*Sth)**2)) |
---|
320 | # betaplus = Sqrt(Abs(alphaplus*alphaplus - b*b*cosgama*cosgama*sinth*sinth)) |
---|
321 | # betaminus = Sqrt(Abs(alphaminus*alphaminus - b*b*cosgama*cosgama*sinth*sinth)) |
---|
322 | Mp = Cgama*(np.pi*a**2*(2.0*X+Y+0.75*Z)+(2.0*np.pi*(1.0-a))*(1.0-a+a*AlP)* \ |
---|
323 | (4.0*X/AlP/BeP+(4.0*(1.0+Cgama**2)/b/b*Cth**2)*(AlP/BeP-1.0)+ |
---|
324 | 2.0/b**4*AlP/BeP*AlP**2-2.0/b**4*AlP**2-Cgama**2/b/b*Sth*2)) |
---|
325 | # Mplus = cosgama*(!DPI * a * a * (2.0*x + y + 0.75*z) + $ |
---|
326 | # (2.0*!DPI*(1.0 - a)) *(1.0 - a + a*alphaplus)* $ |
---|
327 | # (4.0*x/alphaplus/betaplus + (4.0*(1.0+cosgama*cosgama)/b/b*costh*costh)*(alphaplus/betaplus -1.0) + $ |
---|
328 | # 2.0/(b^4)*alphaplus/betaplus*alphaplus*alphaplus - 2.0/(b^4)*alphaplus*alphaplus - $ |
---|
329 | # cosgama*cosgama/b/b*sinth*sinth)) |
---|
330 | Mm =Cgama*(np.pi*a**2*(2.0*X+Y+0.75*Z)+(2.0*np.pi*(1.0-a))*(1.0-a+a*AlM)* \ |
---|
331 | (4.0*X/AlM/BeM+(4.0*(1.0+Cgama**2)/b/b*Cth**2)*(AlM/BeM-1.0)+ |
---|
332 | 2.0/b**4*AlM/BeM*AlM**2-2.0/b**4*AlM**2-Cgama**2/b/b*Sth*2)) |
---|
333 | # Mminus = cosgama*(!DPI * a * a * (2.0*x + y + 0.75*z) + $ |
---|
334 | # (2.0*!DPI*(1.0 - a)) *(1.0 - a + a*alphaminus)* $ |
---|
335 | # (4.0*x/alphaminus/betaminus + (4.0*(1.0+cosgama*cosgama)/b/b*costh*costh)*(alphaminus/betaminus -1.0) + $ |
---|
336 | # 2.0/(b^4)*alphaminus/betaminus*alphaminus*alphaminus - 2.0/(b^4)*alphaminus*alphaminus - $ |
---|
337 | # cosgama*cosgama/b/b*sinth*sinth)) |
---|
338 | Q = np.sum(Wp*Mp+Wm*Mm,axis=0) |
---|
339 | return Q*G*NS/(NS-1.) |
---|
340 | # expmutsecth = Exp(-1.0*mut*secth) |
---|
341 | # G= delta * 2.0 * Na * emc /(1.0+costth^2)/expmutsecth |
---|
342 | # for i=1, num_steps-1 do begin |
---|
343 | # cosgama=double(cos(gama[i])) |
---|
344 | # singama=double(sin(gama[i])) |
---|
345 | # cscgama=1.0d/singama |
---|
346 | # |
---|
347 | # |
---|
348 | # ; print, "W", min(wplus), max(wplus), min(wminus), max(wminus) |
---|
349 | # |
---|
350 | # |
---|
351 | # |
---|
352 | # |
---|
353 | # ; print, a,b |
---|
354 | # ; print, "M", min(mplus), max(mplus), min(mminus), max(mminus) |
---|
355 | # Q=Q+ Wplus*Mplus + Wminus*Mminus |
---|
356 | # endfor |
---|
357 | # Q=double(num_steps)/double(num_steps-1)*Q*G |
---|
358 | # ; print, min(q), max(q) |
---|
359 | # end |
---|
360 | |
---|
361 | def MultiScattering(Geometry,ElList,Tth): |
---|
362 | BN = BD = 0.0 |
---|
363 | Amu = 0.0 |
---|
364 | for El in ElList: |
---|
365 | el = ElList[El] |
---|
366 | BN += el['Z']*el['FormulaNo'] |
---|
367 | BD += el['FormulaNo'] |
---|
368 | Amu += el['FormulaNo']*el['mu'] |
---|
369 | |
---|
370 | def CalcPDF(data,inst,xydata): |
---|
371 | auxPlot = [] |
---|
372 | import copy |
---|
373 | import scipy.fftpack as ft |
---|
374 | #subtract backgrounds - if any |
---|
375 | xydata['IofQ'] = copy.deepcopy(xydata['Sample']) |
---|
376 | if data['Sample Bkg.']['Name']: |
---|
377 | xydata['IofQ'][1][1] += (xydata['Sample Bkg.'][1][1]+ |
---|
378 | data['Sample Bkg.']['Add'])*data['Sample Bkg.']['Mult'] |
---|
379 | if data['Container']['Name']: |
---|
380 | xycontainer = (xydata['Container'][1][1]+data['Container']['Add'])*data['Container']['Mult'] |
---|
381 | if data['Container Bkg.']['Name']: |
---|
382 | xycontainer += (xydata['Container Bkg.'][1][1]+ |
---|
383 | data['Container Bkg.']['Add'])*data['Container Bkg.']['Mult'] |
---|
384 | xydata['IofQ'][1][1] += xycontainer |
---|
385 | #get element data & absorption coeff. |
---|
386 | ElList = data['ElList'] |
---|
387 | Abs = G2lat.CellAbsorption(ElList,data['Form Vol']) |
---|
388 | #Apply angle dependent corrections |
---|
389 | Tth = xydata['Sample'][1][0] |
---|
390 | dt = (Tth[1]-Tth[0]) |
---|
391 | xydata['IofQ'][1][1] /= Absorb(data['Geometry'],Abs,data['Diam'],Tth) |
---|
392 | xydata['IofQ'][1][1] /= Polarization(inst['Polariz.'],Tth,Azm=inst['Azimuth'])[0] |
---|
393 | if data['DetType'] == 'Image plate': |
---|
394 | xydata['IofQ'][1][1] *= Oblique(data['ObliqCoeff'],Tth) |
---|
395 | XY = xydata['IofQ'][1] |
---|
396 | #convert to Q |
---|
397 | hc = 12.397639 |
---|
398 | if 'Lam' in inst: |
---|
399 | wave = inst['Lam'] |
---|
400 | else: |
---|
401 | wave = inst['Lam1'] |
---|
402 | keV = hc/wave |
---|
403 | minQ = npT2q(Tth[0],wave) |
---|
404 | maxQ = npT2q(Tth[-1],wave) |
---|
405 | Qpoints = np.linspace(0.,maxQ,len(XY[0]),endpoint=True) |
---|
406 | dq = Qpoints[1]-Qpoints[0] |
---|
407 | XY[0] = npT2q(XY[0],wave) |
---|
408 | # Qdata = np.nan_to_num(si.griddata(XY[0],XY[1],Qpoints,method='linear')) #only OK for scipy 0.9! |
---|
409 | T = si.interp1d(XY[0],XY[1],bounds_error=False,fill_value=0.0) #OK for scipy 0.8 |
---|
410 | Qdata = T(Qpoints) |
---|
411 | |
---|
412 | qLimits = data['QScaleLim'] |
---|
413 | minQ = np.searchsorted(Qpoints,qLimits[0]) |
---|
414 | maxQ = np.searchsorted(Qpoints,qLimits[1]) |
---|
415 | newdata = [] |
---|
416 | xydata['IofQ'][1][0] = Qpoints |
---|
417 | xydata['IofQ'][1][1] = Qdata |
---|
418 | for item in xydata['IofQ'][1]: |
---|
419 | newdata.append(item[:maxQ]) |
---|
420 | xydata['IofQ'][1] = newdata |
---|
421 | |
---|
422 | |
---|
423 | xydata['SofQ'] = copy.deepcopy(xydata['IofQ']) |
---|
424 | FFSq,SqFF,CF = GetAsfMean(ElList,(xydata['SofQ'][1][0]/(4.0*np.pi))**2) #these are <f^2>,<f>^2,Cf |
---|
425 | Q = xydata['SofQ'][1][0] |
---|
426 | ruland = Ruland(data['Ruland'],wave,Q,CF) |
---|
427 | # auxPlot.append([Q,ruland,'Ruland']) |
---|
428 | CF *= ruland |
---|
429 | # auxPlot.append([Q,CF,'CF-Corr']) |
---|
430 | scale = np.sum((FFSq+CF)[minQ:maxQ])/np.sum(xydata['SofQ'][1][1][minQ:maxQ]) |
---|
431 | xydata['SofQ'][1][1] *= scale |
---|
432 | xydata['SofQ'][1][1] -= CF |
---|
433 | xydata['SofQ'][1][1] = xydata['SofQ'][1][1]/SqFF |
---|
434 | scale = len(xydata['SofQ'][1][1][minQ:maxQ])/np.sum(xydata['SofQ'][1][1][minQ:maxQ]) |
---|
435 | xydata['SofQ'][1][1] *= scale |
---|
436 | |
---|
437 | xydata['FofQ'] = copy.deepcopy(xydata['SofQ']) |
---|
438 | xydata['FofQ'][1][1] = xydata['FofQ'][1][0]*(xydata['SofQ'][1][1]-1.0) |
---|
439 | if data['Lorch']: |
---|
440 | xydata['FofQ'][1][1] *= LorchWeight(Q) |
---|
441 | |
---|
442 | xydata['GofR'] = copy.deepcopy(xydata['FofQ']) |
---|
443 | nR = len(xydata['GofR'][1][1]) |
---|
444 | xydata['GofR'][1][1] = -dq*np.imag(ft.fft(xydata['FofQ'][1][1],4*nR)[:nR]) |
---|
445 | xydata['GofR'][1][0] = 0.5*np.pi*np.linspace(0,nR,nR)/qLimits[1] |
---|
446 | |
---|
447 | |
---|
448 | return auxPlot |
---|
449 | |
---|
450 | #GSASII peak fitting routines: Finger, Cox & Jephcoat model |
---|
451 | |
---|
452 | def factorize(num): |
---|
453 | ''' Provide prime number factors for integer num |
---|
454 | Returns dictionary of prime factors (keys) & power for each (data) |
---|
455 | ''' |
---|
456 | factors = {} |
---|
457 | orig = num |
---|
458 | |
---|
459 | # we take advantage of the fact that (i +1)**2 = i**2 + 2*i +1 |
---|
460 | i, sqi = 2, 4 |
---|
461 | while sqi <= num: |
---|
462 | while not num%i: |
---|
463 | num /= i |
---|
464 | factors[i] = factors.get(i, 0) + 1 |
---|
465 | |
---|
466 | sqi += 2*i + 1 |
---|
467 | i += 1 |
---|
468 | |
---|
469 | if num != 1 and num != orig: |
---|
470 | factors[num] = factors.get(num, 0) + 1 |
---|
471 | |
---|
472 | if factors: |
---|
473 | return factors |
---|
474 | else: |
---|
475 | return {num:1} #a prime number! |
---|
476 | |
---|
477 | def makeFFTsizeList(nmin=1,nmax=1023,thresh=15): |
---|
478 | ''' Provide list of optimal data sizes for FFT calculations |
---|
479 | Input: |
---|
480 | nmin: minimum data size >= 1 |
---|
481 | nmax: maximum data size > nmin |
---|
482 | thresh: maximum prime factor allowed |
---|
483 | Returns: |
---|
484 | list of data sizes where the maximum prime factor is < thresh |
---|
485 | ''' |
---|
486 | plist = [] |
---|
487 | nmin = max(1,nmin) |
---|
488 | nmax = max(nmin+1,nmax) |
---|
489 | for p in range(nmin,nmax): |
---|
490 | if max(factorize(p).keys()) < thresh: |
---|
491 | plist.append(p) |
---|
492 | return plist |
---|
493 | |
---|
494 | np.seterr(divide='ignore') |
---|
495 | |
---|
496 | # Normal distribution |
---|
497 | |
---|
498 | # loc = mu, scale = std |
---|
499 | _norm_pdf_C = 1./math.sqrt(2*math.pi) |
---|
500 | class norm_gen(st.rv_continuous): |
---|
501 | |
---|
502 | def pdf(self,x,*args,**kwds): |
---|
503 | loc,scale=kwds['loc'],kwds['scale'] |
---|
504 | x = (x-loc)/scale |
---|
505 | return np.exp(-x**2/2.0) * _norm_pdf_C / scale |
---|
506 | |
---|
507 | norm = norm_gen(name='norm',longname='A normal',extradoc=""" |
---|
508 | |
---|
509 | Normal distribution |
---|
510 | |
---|
511 | The location (loc) keyword specifies the mean. |
---|
512 | The scale (scale) keyword specifies the standard deviation. |
---|
513 | |
---|
514 | normal.pdf(x) = exp(-x**2/2)/sqrt(2*pi) |
---|
515 | """) |
---|
516 | |
---|
517 | ## Cauchy |
---|
518 | |
---|
519 | # median = loc |
---|
520 | |
---|
521 | class cauchy_gen(st.rv_continuous): |
---|
522 | |
---|
523 | def pdf(self,x,*args,**kwds): |
---|
524 | loc,scale=kwds['loc'],kwds['scale'] |
---|
525 | x = (x-loc)/scale |
---|
526 | return 1.0/np.pi/(1.0+x*x) / scale |
---|
527 | |
---|
528 | cauchy = cauchy_gen(name='cauchy',longname='Cauchy',extradoc=""" |
---|
529 | |
---|
530 | Cauchy distribution |
---|
531 | |
---|
532 | cauchy.pdf(x) = 1/(pi*(1+x**2)) |
---|
533 | |
---|
534 | This is the t distribution with one degree of freedom. |
---|
535 | """) |
---|
536 | |
---|
537 | |
---|
538 | #GSASII peak fitting routine: Finger, Cox & Jephcoat model |
---|
539 | |
---|
540 | |
---|
541 | class fcjde_gen(st.rv_continuous): |
---|
542 | """ |
---|
543 | Finger-Cox-Jephcoat D(2phi,2th) function for S/L = H/L |
---|
544 | Ref: J. Appl. Cryst. (1994) 27, 892-900. |
---|
545 | Parameters |
---|
546 | ----------------------------------------- |
---|
547 | x: array -1 to 1 |
---|
548 | t: 2-theta position of peak |
---|
549 | s: sum(S/L,H/L); S: sample height, H: detector opening, |
---|
550 | L: sample to detector opening distance |
---|
551 | dx: 2-theta step size in deg |
---|
552 | Result for fcj.pdf |
---|
553 | ----------------------------------------- |
---|
554 | T = x*dx+t |
---|
555 | s = S/L+H/L |
---|
556 | if x < 0: |
---|
557 | fcj.pdf = [1/sqrt({cos(T)**2/cos(t)**2}-1) - 1/s]/|cos(T)| |
---|
558 | if x >= 0: |
---|
559 | fcj.pdf = 0 |
---|
560 | """ |
---|
561 | def _pdf(self,x,t,s,dx): |
---|
562 | T = dx*x+t |
---|
563 | ax2 = abs(npcosd(T)) |
---|
564 | ax = ax2**2 |
---|
565 | bx = npcosd(t)**2 |
---|
566 | bx = np.where(ax>bx,bx,ax) |
---|
567 | fx = np.where(ax>bx,(np.sqrt(bx/(ax-bx))-1./s)/ax2,0.0) |
---|
568 | fx = np.where(fx > 0.,fx,0.0) |
---|
569 | return fx |
---|
570 | |
---|
571 | def pdf(self,x,*args,**kwds): |
---|
572 | loc=kwds['loc'] |
---|
573 | return self._pdf(x-loc,*args) |
---|
574 | |
---|
575 | fcjde = fcjde_gen(name='fcjde',shapes='t,s,dx') |
---|
576 | |
---|
577 | def getWidths(pos,sig,gam,shl): |
---|
578 | widths = [np.sqrt(sig)/100.,gam/200.] |
---|
579 | fwhm = 2.355*widths[0]+2.*widths[1] |
---|
580 | fmin = 10.*(fwhm+shl*abs(npcosd(pos))) |
---|
581 | fmax = 15.0*fwhm |
---|
582 | if pos > 90: |
---|
583 | fmin,fmax = [fmax,fmin] |
---|
584 | return widths,fmin,fmax |
---|
585 | |
---|
586 | def getFCJVoigt(pos,intens,sig,gam,shl,xdata): |
---|
587 | DX = xdata[1]-xdata[0] |
---|
588 | widths,fmin,fmax = getWidths(pos,sig,gam,shl) |
---|
589 | x = np.linspace(pos-fmin,pos+fmin,256) |
---|
590 | dx = x[1]-x[0] |
---|
591 | Norm = norm.pdf(x,loc=pos,scale=widths[0]) |
---|
592 | Cauchy = cauchy.pdf(x,loc=pos,scale=widths[1]) |
---|
593 | arg = [pos,shl/57.2958,dx,] |
---|
594 | FCJ = fcjde.pdf(x,*arg,loc=pos) |
---|
595 | if len(np.nonzero(FCJ)[0])>5: |
---|
596 | z = np.column_stack([Norm,Cauchy,FCJ]).T |
---|
597 | Z = fft(z) |
---|
598 | Df = ifft(Z.prod(axis=0)).real |
---|
599 | else: |
---|
600 | z = np.column_stack([Norm,Cauchy]).T |
---|
601 | Z = fft(z) |
---|
602 | Df = fftshift(ifft(Z.prod(axis=0))).real |
---|
603 | Df /= np.sum(Df) |
---|
604 | Df = si.interp1d(x,Df,bounds_error=False,fill_value=0.0) |
---|
605 | return intens*Df(xdata)*DX/dx |
---|
606 | |
---|
607 | def getBackground(pfx,parmDict,bakType,xdata): |
---|
608 | yb = np.zeros_like(xdata) |
---|
609 | if bakType == 'chebyschev': |
---|
610 | iBak = 0 |
---|
611 | while True: |
---|
612 | key = pfx+'Back:'+str(iBak) |
---|
613 | try: |
---|
614 | yb += parmDict[key]*(xdata-xdata[0])**iBak |
---|
615 | iBak += 1 |
---|
616 | except KeyError: |
---|
617 | break |
---|
618 | return yb |
---|
619 | |
---|
620 | def getBackgroundDerv(pfx,parmDict,bakType,xdata): |
---|
621 | dydb = [] |
---|
622 | if bakType == 'chebyschev': |
---|
623 | iBak = 0 |
---|
624 | while True: |
---|
625 | if pfx+'Back:'+str(iBak) in parmDict: |
---|
626 | dydb.append((xdata-xdata[0])**iBak) |
---|
627 | iBak += 1 |
---|
628 | else: |
---|
629 | break |
---|
630 | return dydb |
---|
631 | |
---|
632 | #use old fortran routine |
---|
633 | def getFCJVoigt3(pos,sig,gam,shl,xdata): |
---|
634 | |
---|
635 | Df = pyd.pypsvfcj(len(xdata),xdata-pos,pos,sig,gam,shl) |
---|
636 | Df /= np.sum(Df) |
---|
637 | return Df |
---|
638 | |
---|
639 | def getdFCJVoigt3(pos,sig,gam,shl,xdata): |
---|
640 | |
---|
641 | Df,dFdp,dFds,dFdg,dFdsh = pyd.pydpsvfcj(len(xdata),xdata-pos,pos,sig,gam,shl) |
---|
642 | sumDf = np.sum(Df) |
---|
643 | return Df,dFdp,dFds,dFdg,dFdsh |
---|
644 | |
---|
645 | |
---|
646 | def getPeakProfile(parmDict,xdata,varyList,bakType): |
---|
647 | |
---|
648 | yb = getBackground('',parmDict,bakType,xdata) |
---|
649 | yc = np.zeros_like(yb) |
---|
650 | dx = xdata[1]-xdata[0] |
---|
651 | U = parmDict['U'] |
---|
652 | V = parmDict['V'] |
---|
653 | W = parmDict['W'] |
---|
654 | X = parmDict['X'] |
---|
655 | Y = parmDict['Y'] |
---|
656 | shl = max(parmDict['SH/L'],0.002) |
---|
657 | Ka2 = False |
---|
658 | if 'Lam1' in parmDict.keys(): |
---|
659 | Ka2 = True |
---|
660 | lamRatio = 360*(parmDict['Lam2']-parmDict['Lam1'])/(np.pi*parmDict['Lam1']) |
---|
661 | kRatio = parmDict['I(L2)/I(L1)'] |
---|
662 | iPeak = 0 |
---|
663 | while True: |
---|
664 | try: |
---|
665 | pos = parmDict['pos'+str(iPeak)] |
---|
666 | intens = parmDict['int'+str(iPeak)] |
---|
667 | sigName = 'sig'+str(iPeak) |
---|
668 | if sigName in varyList: |
---|
669 | sig = parmDict[sigName] |
---|
670 | else: |
---|
671 | sig = U*tand(pos/2.0)**2+V*tand(pos/2.0)+W |
---|
672 | sig = max(sig,0.001) #avoid neg sigma |
---|
673 | gamName = 'gam'+str(iPeak) |
---|
674 | if gamName in varyList: |
---|
675 | gam = parmDict[gamName] |
---|
676 | else: |
---|
677 | gam = X/cosd(pos/2.0)+Y*tand(pos/2.0) |
---|
678 | gam = max(gam,0.001) #avoid neg gamma |
---|
679 | Wd,fmin,fmax = getWidths(pos,sig,gam,shl) |
---|
680 | iBeg = np.searchsorted(xdata,pos-fmin) |
---|
681 | lenX = len(xdata) |
---|
682 | if not iBeg: |
---|
683 | iFin = np.searchsorted(xdata,pos+fmin) |
---|
684 | elif iBeg == lenX: |
---|
685 | iFin = iBeg |
---|
686 | else: |
---|
687 | iFin = min(lenX,iBeg+int((fmin+fmax)/dx)) |
---|
688 | if not iBeg+iFin: #peak below low limit |
---|
689 | iPeak += 1 |
---|
690 | continue |
---|
691 | elif not iBeg-iFin: #peak above high limit |
---|
692 | return yb+yc |
---|
693 | yc[iBeg:iFin] += intens*getFCJVoigt3(pos,sig,gam,shl,xdata[iBeg:iFin]) |
---|
694 | if Ka2: |
---|
695 | pos2 = pos+lamRatio*tand(pos/2.0) # + 360/pi * Dlam/lam * tan(th) |
---|
696 | kdelt = int((pos2-pos)/dx) |
---|
697 | iBeg = min(lenX,iBeg+kdelt) |
---|
698 | iFin = min(lenX,iFin+kdelt) |
---|
699 | if iBeg-iFin: |
---|
700 | yc[iBeg:iFin] += intens*kRatio*getFCJVoigt3(pos2,sig,gam,shl,xdata[iBeg:iFin]) |
---|
701 | iPeak += 1 |
---|
702 | except KeyError: #no more peaks to process |
---|
703 | return yb+yc |
---|
704 | |
---|
705 | def getPeakProfileDerv(parmDict,xdata,varyList,bakType): |
---|
706 | # needs to return np.array([dMdx1,dMdx2,...]) in same order as varylist = backVary,insVary,peakVary order |
---|
707 | dMdv = np.zeros(shape=(len(varyList),len(xdata))) |
---|
708 | if 'Back:0' in varyList: #background derivs are in front if present |
---|
709 | dMdb = getBackgroundDerv('',parmDict,bakType,xdata) |
---|
710 | dMdv[0:len(dMdb)] = dMdb |
---|
711 | |
---|
712 | dx = xdata[1]-xdata[0] |
---|
713 | U = parmDict['U'] |
---|
714 | V = parmDict['V'] |
---|
715 | W = parmDict['W'] |
---|
716 | X = parmDict['X'] |
---|
717 | Y = parmDict['Y'] |
---|
718 | shl = max(parmDict['SH/L'],0.002) |
---|
719 | Ka2 = False |
---|
720 | if 'Lam1' in parmDict.keys(): |
---|
721 | Ka2 = True |
---|
722 | lamRatio = 360*(parmDict['Lam2']-parmDict['Lam1'])/(np.pi*parmDict['Lam1']) |
---|
723 | kRatio = parmDict['I(L2)/I(L1)'] |
---|
724 | iPeak = 0 |
---|
725 | while True: |
---|
726 | try: |
---|
727 | pos = parmDict['pos'+str(iPeak)] |
---|
728 | intens = parmDict['int'+str(iPeak)] |
---|
729 | sigName = 'sig'+str(iPeak) |
---|
730 | tanth = tand(pos/2.0) |
---|
731 | costh = cosd(pos/2.0) |
---|
732 | if sigName in varyList: |
---|
733 | sig = parmDict[sigName] |
---|
734 | else: |
---|
735 | sig = U*tanth**2+V*tanth+W |
---|
736 | dsdU = tanth**2 |
---|
737 | dsdV = tanth |
---|
738 | dsdW = 1.0 |
---|
739 | sig = max(sig,0.001) #avoid neg sigma |
---|
740 | gamName = 'gam'+str(iPeak) |
---|
741 | if gamName in varyList: |
---|
742 | gam = parmDict[gamName] |
---|
743 | else: |
---|
744 | gam = X/costh+Y*tanth |
---|
745 | dgdX = 1.0/costh |
---|
746 | dgdY = tanth |
---|
747 | gam = max(gam,0.001) #avoid neg gamma |
---|
748 | Wd,fmin,fmax = getWidths(pos,sig,gam,shl) |
---|
749 | iBeg = np.searchsorted(xdata,pos-fmin) |
---|
750 | lenX = len(xdata) |
---|
751 | if not iBeg: |
---|
752 | iFin = np.searchsorted(xdata,pos+fmin) |
---|
753 | elif iBeg == lenX: |
---|
754 | iFin = iBeg |
---|
755 | else: |
---|
756 | iFin = min(lenX,iBeg+int((fmin+fmax)/dx)) |
---|
757 | if not iBeg+iFin: #peak below low limit |
---|
758 | iPeak += 1 |
---|
759 | continue |
---|
760 | elif not iBeg-iFin: #peak above high limit |
---|
761 | break |
---|
762 | dMdpk = np.zeros(shape=(6,len(xdata))) |
---|
763 | dMdipk = getdFCJVoigt3(pos,sig,gam,shl,xdata[iBeg:iFin]) |
---|
764 | for i in range(1,5): |
---|
765 | dMdpk[i][iBeg:iFin] += 100.*dx*intens*dMdipk[i] |
---|
766 | dMdpk[0][iBeg:iFin] += 100.*dx*dMdipk[0] |
---|
767 | dervDict = {'int':dMdpk[0],'pos':dMdpk[1],'sig':dMdpk[2],'gam':dMdpk[3],'shl':dMdpk[4]} |
---|
768 | if Ka2: |
---|
769 | pos2 = pos+lamRatio*tand(pos/2.0) # + 360/pi * Dlam/lam * tan(th) |
---|
770 | kdelt = int((pos2-pos)/dx) |
---|
771 | iBeg = min(lenX,iBeg+kdelt) |
---|
772 | iFin = min(lenX,iFin+kdelt) |
---|
773 | if iBeg-iFin: |
---|
774 | dMdipk2 = getdFCJVoigt3(pos2,sig,gam,shl,xdata[iBeg:iFin]) |
---|
775 | for i in range(1,5): |
---|
776 | dMdpk[i][iBeg:iFin] += 100.*dx*intens*kRatio*dMdipk2[i] |
---|
777 | dMdpk[0][iBeg:iFin] += 100.*dx*kRatio*dMdipk2[0] |
---|
778 | dMdpk[5][iBeg:iFin] += 100.*dx*dMdipk2[0] |
---|
779 | dervDict = {'int':dMdpk[0],'pos':dMdpk[1],'sig':dMdpk[2],'gam':dMdpk[3],'shl':dMdpk[4],'L1/L2':dMdpk[5]*intens} |
---|
780 | for parmName in ['pos','int','sig','gam']: |
---|
781 | try: |
---|
782 | idx = varyList.index(parmName+str(iPeak)) |
---|
783 | dMdv[idx] = dervDict[parmName] |
---|
784 | except ValueError: |
---|
785 | pass |
---|
786 | if 'U' in varyList: |
---|
787 | dMdv[varyList.index('U')] += dsdU*dervDict['sig'] |
---|
788 | if 'V' in varyList: |
---|
789 | dMdv[varyList.index('V')] += dsdV*dervDict['sig'] |
---|
790 | if 'W' in varyList: |
---|
791 | dMdv[varyList.index('W')] += dsdW*dervDict['sig'] |
---|
792 | if 'X' in varyList: |
---|
793 | dMdv[varyList.index('X')] += dgdX*dervDict['gam'] |
---|
794 | if 'Y' in varyList: |
---|
795 | dMdv[varyList.index('Y')] += dgdY*dervDict['gam'] |
---|
796 | if 'SH/L' in varyList: |
---|
797 | dMdv[varyList.index('SH/L')] += dervDict['shl'] #problem here |
---|
798 | if 'I(L2)/I(L1)' in varyList: |
---|
799 | dMdv[varyList.index('I(L2)/I(L1)')] += dervDict['L1/L2'] |
---|
800 | iPeak += 1 |
---|
801 | except KeyError: #no more peaks to process |
---|
802 | break |
---|
803 | return dMdv |
---|
804 | |
---|
805 | def Dict2Values(parmdict, varylist): |
---|
806 | '''Use before call to leastsq to setup list of values for the parameters |
---|
807 | in parmdict, as selected by key in varylist''' |
---|
808 | return [parmdict[key] for key in varylist] |
---|
809 | |
---|
810 | def Values2Dict(parmdict, varylist, values): |
---|
811 | ''' Use after call to leastsq to update the parameter dictionary with |
---|
812 | values corresponding to keys in varylist''' |
---|
813 | parmdict.update(zip(varylist,values)) |
---|
814 | |
---|
815 | def DoPeakFit(FitPgm,Peaks,Background,Limits,Inst,data,oneCycle=False,controls=None): |
---|
816 | |
---|
817 | def SetBackgroundParms(Background): |
---|
818 | bakType,bakFlag = Background[:2] |
---|
819 | backVals = Background[3:] |
---|
820 | backNames = ['Back:'+str(i) for i in range(len(backVals))] |
---|
821 | if bakFlag: #returns backNames as varyList = backNames |
---|
822 | return bakType,dict(zip(backNames,backVals)),backNames |
---|
823 | else: #no background varied; varyList = [] |
---|
824 | return bakType,dict(zip(backNames,backVals)),[] |
---|
825 | |
---|
826 | def GetBackgroundParms(parmList,Background): |
---|
827 | iBak = 0 |
---|
828 | while True: |
---|
829 | try: |
---|
830 | bakName = 'Back:'+str(iBak) |
---|
831 | Background[iBak+3] = parmList[bakName] |
---|
832 | iBak += 1 |
---|
833 | except KeyError: |
---|
834 | break |
---|
835 | |
---|
836 | def BackgroundPrint(Background,sigDict): |
---|
837 | if Background[1]: |
---|
838 | print 'Background coefficients for',Background[0],'function' |
---|
839 | ptfmt = "%12.5f" |
---|
840 | ptstr = 'values:' |
---|
841 | sigstr = 'esds :' |
---|
842 | for i,back in enumerate(Background[3:]): |
---|
843 | ptstr += ptfmt % (back) |
---|
844 | sigstr += ptfmt % (sigDict['Back:'+str(i)]) |
---|
845 | print ptstr |
---|
846 | print sigstr |
---|
847 | else: |
---|
848 | print 'Background not refined' |
---|
849 | |
---|
850 | def SetInstParms(Inst): |
---|
851 | insVals,insFlags,insNames = Inst[1:4] |
---|
852 | dataType = insVals[0] |
---|
853 | insVary = [] |
---|
854 | for i,flag in enumerate(insFlags): |
---|
855 | if flag and insNames[i] in ['U','V','W','X','Y','SH/L','I(L2)/I(L1)']: |
---|
856 | insVary.append(insNames[i]) |
---|
857 | instDict = dict(zip(insNames,insVals)) |
---|
858 | instDict['X'] = max(instDict['X'],0.01) |
---|
859 | instDict['Y'] = max(instDict['Y'],0.01) |
---|
860 | instDict['SH/L'] = max(instDict['SH/L'],0.002) |
---|
861 | return dataType,instDict,insVary |
---|
862 | |
---|
863 | def GetInstParms(parmDict,Inst,varyList,Peaks): |
---|
864 | instNames = Inst[3] |
---|
865 | for i,name in enumerate(instNames): |
---|
866 | Inst[1][i] = parmDict[name] |
---|
867 | iPeak = 0 |
---|
868 | while True: |
---|
869 | try: |
---|
870 | sigName = 'sig'+str(iPeak) |
---|
871 | pos = parmDict['pos'+str(iPeak)] |
---|
872 | if sigName not in varyList: |
---|
873 | parmDict[sigName] = parmDict['U']*tand(pos/2.0)**2+parmDict['V']*tand(pos/2.0)+parmDict['W'] |
---|
874 | gamName = 'gam'+str(iPeak) |
---|
875 | if gamName not in varyList: |
---|
876 | parmDict[gamName] = parmDict['X']/cosd(pos/2.0)+parmDict['Y']*tand(pos/2.0) |
---|
877 | iPeak += 1 |
---|
878 | except KeyError: |
---|
879 | break |
---|
880 | |
---|
881 | def InstPrint(Inst,sigDict): |
---|
882 | print 'Instrument Parameters:' |
---|
883 | ptfmt = "%12.6f" |
---|
884 | ptlbls = 'names :' |
---|
885 | ptstr = 'values:' |
---|
886 | sigstr = 'esds :' |
---|
887 | instNames = Inst[3][1:] |
---|
888 | for i,name in enumerate(instNames): |
---|
889 | ptlbls += "%s" % (name.center(12)) |
---|
890 | ptstr += ptfmt % (Inst[1][i+1]) |
---|
891 | if name in sigDict: |
---|
892 | sigstr += ptfmt % (sigDict[name]) |
---|
893 | else: |
---|
894 | sigstr += 12*' ' |
---|
895 | print ptlbls |
---|
896 | print ptstr |
---|
897 | print sigstr |
---|
898 | |
---|
899 | def SetPeaksParms(Peaks): |
---|
900 | peakNames = [] |
---|
901 | peakVary = [] |
---|
902 | peakVals = [] |
---|
903 | names = ['pos','int','sig','gam'] |
---|
904 | for i,peak in enumerate(Peaks): |
---|
905 | for j in range(4): |
---|
906 | peakVals.append(peak[2*j]) |
---|
907 | parName = names[j]+str(i) |
---|
908 | peakNames.append(parName) |
---|
909 | if peak[2*j+1]: |
---|
910 | peakVary.append(parName) |
---|
911 | return dict(zip(peakNames,peakVals)),peakVary |
---|
912 | |
---|
913 | def GetPeaksParms(parmDict,Peaks,varyList): |
---|
914 | names = ['pos','int','sig','gam'] |
---|
915 | for i,peak in enumerate(Peaks): |
---|
916 | for j in range(4): |
---|
917 | pos = parmDict['pos'+str(i)] |
---|
918 | parName = names[j]+str(i) |
---|
919 | if parName in varyList: |
---|
920 | peak[2*j] = parmDict[parName] |
---|
921 | elif 'sig' in parName: |
---|
922 | peak[2*j] = parmDict['U']*tand(pos/2.0)**2+parmDict['V']*tand(pos/2.0)+parmDict['W'] |
---|
923 | elif 'gam' in parName: |
---|
924 | peak[2*j] = parmDict['X']/cosd(pos/2.0)+parmDict['Y']*tand(pos/2.0) |
---|
925 | |
---|
926 | def PeaksPrint(parmDict,sigDict,varyList): |
---|
927 | print 'Peak coefficients:' |
---|
928 | names = ['pos','int','sig','gam'] |
---|
929 | head = 15*' ' |
---|
930 | for name in names: |
---|
931 | head += name.center(12)+'esd'.center(12) |
---|
932 | print head |
---|
933 | ptfmt = {'pos':"%12.5f",'int':"%12.1f",'sig':"%12.3f",'gam':"%12.3f"} |
---|
934 | for i,peak in enumerate(Peaks): |
---|
935 | ptstr = ':' |
---|
936 | for j in range(4): |
---|
937 | name = names[j] |
---|
938 | parName = name+str(i) |
---|
939 | ptstr += ptfmt[name] % (parmDict[parName]) |
---|
940 | if parName in varyList: |
---|
941 | # ptstr += G2IO.ValEsd(parmDict[parName],sigDict[parName]) |
---|
942 | ptstr += ptfmt[name] % (sigDict[parName]) |
---|
943 | else: |
---|
944 | # ptstr += G2IO.ValEsd(parmDict[parName],0.0) |
---|
945 | ptstr += 12*' ' |
---|
946 | print '%s'%(('Peak'+str(i+1)).center(8)),ptstr |
---|
947 | |
---|
948 | def devPeakProfile(values, xdata, ydata, weights, parmdict, varylist,bakType,dlg): |
---|
949 | parmdict.update(zip(varylist,values)) |
---|
950 | return np.sqrt(weights)*getPeakProfileDerv(parmdict,xdata,varylist,bakType) |
---|
951 | |
---|
952 | def errPeakProfile(values, xdata, ydata, weights, parmdict, varylist,bakType,dlg): |
---|
953 | parmdict.update(zip(varylist,values)) |
---|
954 | M = np.sqrt(weights)*(getPeakProfile(parmdict,xdata,varylist,bakType)-ydata) |
---|
955 | Rwp = min(100.,np.sqrt(np.sum(M**2)/np.sum(weights*ydata**2))*100.) |
---|
956 | if dlg: |
---|
957 | GoOn = dlg.Update(Rwp,newmsg='%s%8.3f%s'%('Peak fit Rwp =',Rwp,'%'))[0] |
---|
958 | if not GoOn: |
---|
959 | return -M #abort!! |
---|
960 | return M |
---|
961 | |
---|
962 | if controls: |
---|
963 | Ftol = controls['min dM/M'] |
---|
964 | derivType = controls['deriv type'] |
---|
965 | else: |
---|
966 | Ftol = 0.0001 |
---|
967 | derivType = 'analytic' |
---|
968 | if oneCycle: |
---|
969 | Ftol = 1.0 |
---|
970 | x,y,w,yc,yb,yd = data #these are numpy arrays! |
---|
971 | xBeg = np.searchsorted(x,Limits[0]) |
---|
972 | xFin = np.searchsorted(x,Limits[1]) |
---|
973 | bakType,bakDict,bakVary = SetBackgroundParms(Background) |
---|
974 | dataType,insDict,insVary = SetInstParms(Inst) |
---|
975 | peakDict,peakVary = SetPeaksParms(Peaks) |
---|
976 | parmDict = {} |
---|
977 | parmDict.update(bakDict) |
---|
978 | parmDict.update(insDict) |
---|
979 | parmDict.update(peakDict) |
---|
980 | varyList = bakVary+insVary+peakVary |
---|
981 | while True: |
---|
982 | begin = time.time() |
---|
983 | values = np.array(Dict2Values(parmDict, varyList)) |
---|
984 | if FitPgm == 'LSQ': |
---|
985 | dlg = wx.ProgressDialog('Residual','Peak fit Rwp = ',101.0, |
---|
986 | style = wx.PD_ELAPSED_TIME|wx.PD_AUTO_HIDE|wx.PD_REMAINING_TIME|wx.PD_CAN_ABORT) |
---|
987 | screenSize = wx.ClientDisplayRect() |
---|
988 | Size = dlg.GetSize() |
---|
989 | dlg.SetPosition(wx.Point(screenSize[2]-Size[0]-305,screenSize[1]+5)) |
---|
990 | try: |
---|
991 | if derivType == 'analytic': |
---|
992 | result = so.leastsq(errPeakProfile,values,Dfun=devPeakProfile,full_output=True,ftol=Ftol,col_deriv=True, |
---|
993 | args=(x[xBeg:xFin],y[xBeg:xFin],w[xBeg:xFin],parmDict,varyList,bakType,dlg)) |
---|
994 | ncyc = int(result[2]['nfev']/2) |
---|
995 | else: |
---|
996 | result = so.leastsq(errPeakProfile,values,full_output=True,ftol=Ftol,epsfcn=1.e-8, |
---|
997 | args=(x[xBeg:xFin],y[xBeg:xFin],w[xBeg:xFin],parmDict,varyList,bakType,dlg)) |
---|
998 | ncyc = int(result[2]['nfev']/len(varyList)) |
---|
999 | finally: |
---|
1000 | dlg.Destroy() |
---|
1001 | runtime = time.time()-begin |
---|
1002 | chisq = np.sum(result[2]['fvec']**2) |
---|
1003 | Values2Dict(parmDict, varyList, result[0]) |
---|
1004 | Rwp = np.sqrt(chisq/np.sum(w[xBeg:xFin]*y[xBeg:xFin]**2))*100. #to % |
---|
1005 | GOF = chisq/(xFin-xBeg-len(varyList)) |
---|
1006 | print 'Number of function calls:',result[2]['nfev'],' Number of observations: ',xFin-xBeg,' Number of parameters: ',len(varyList) |
---|
1007 | print 'fitpeak time = %8.3fs, %8.3fs/cycle'%(runtime,runtime/ncyc) |
---|
1008 | print 'Rwp = %7.2f%%, chi**2 = %12.6g, reduced chi**2 = %6.2f'%(Rwp,chisq,GOF) |
---|
1009 | try: |
---|
1010 | sig = np.sqrt(np.diag(result[1])*GOF) |
---|
1011 | if np.any(np.isnan(sig)): |
---|
1012 | print '*** Least squares aborted - some invalid esds possible ***' |
---|
1013 | break #refinement succeeded - finish up! |
---|
1014 | except ValueError: #result[1] is None on singular matrix |
---|
1015 | print '**** Refinement failed - singular matrix ****' |
---|
1016 | Ipvt = result[2]['ipvt'] |
---|
1017 | for i,ipvt in enumerate(Ipvt): |
---|
1018 | if not np.sum(result[2]['fjac'],axis=1)[i]: |
---|
1019 | print 'Removing parameter: ',varyList[ipvt-1] |
---|
1020 | del(varyList[ipvt-1]) |
---|
1021 | break |
---|
1022 | elif FitPgm == 'BFGS': |
---|
1023 | print 'Other program here' |
---|
1024 | return |
---|
1025 | |
---|
1026 | sigDict = dict(zip(varyList,sig)) |
---|
1027 | yb[xBeg:xFin] = getBackground('',parmDict,bakType,x[xBeg:xFin]) |
---|
1028 | yc[xBeg:xFin] = getPeakProfile(parmDict,x[xBeg:xFin],varyList,bakType) |
---|
1029 | yd[xBeg:xFin] = y[xBeg:xFin]-yc[xBeg:xFin] |
---|
1030 | GetBackgroundParms(parmDict,Background) |
---|
1031 | BackgroundPrint(Background,sigDict) |
---|
1032 | GetInstParms(parmDict,Inst,varyList,Peaks) |
---|
1033 | InstPrint(Inst,sigDict) |
---|
1034 | GetPeaksParms(parmDict,Peaks,varyList) |
---|
1035 | PeaksPrint(parmDict,sigDict,varyList) |
---|
1036 | |
---|
1037 | #testing data |
---|
1038 | NeedTestData = True |
---|
1039 | def TestData(): |
---|
1040 | # global NeedTestData |
---|
1041 | NeedTestData = False |
---|
1042 | global bakType |
---|
1043 | bakType = 'chebyschev' |
---|
1044 | global xdata |
---|
1045 | xdata = np.linspace(4.0,40.0,36000) |
---|
1046 | global parmDict0 |
---|
1047 | parmDict0 = { |
---|
1048 | 'pos0':5.6964,'int0':8835.8,'sig0':1.0,'gam0':1.0, |
---|
1049 | 'pos1':11.4074,'int1':3922.3,'sig1':1.0,'gam1':1.0, |
---|
1050 | 'pos2':20.6426,'int2':1573.7,'sig2':1.0,'gam2':1.0, |
---|
1051 | 'pos3':26.9568,'int3':925.1,'sig3':1.0,'gam3':1.0, |
---|
1052 | 'U':1.163,'V':-0.605,'W':0.093,'X':0.0,'Y':2.183,'SH/L':0.002, |
---|
1053 | 'Back0':5.384,'Back1':-0.015,'Back2':.004, |
---|
1054 | } |
---|
1055 | global parmDict1 |
---|
1056 | parmDict1 = { |
---|
1057 | 'pos0':13.4924,'int0':48697.6,'sig0':1.0,'gam0':1.0, |
---|
1058 | 'pos1':23.4360,'int1':43685.5,'sig1':1.0,'gam1':1.0, |
---|
1059 | 'pos2':27.1152,'int2':123712.6,'sig2':1.0,'gam2':1.0, |
---|
1060 | 'pos3':33.7196,'int3':65349.4,'sig3':1.0,'gam3':1.0, |
---|
1061 | 'pos4':36.1119,'int4':115829.8,'sig4':1.0,'gam4':1.0, |
---|
1062 | 'pos5':39.0122,'int5':6916.9,'sig5':1.0,'gam5':1.0, |
---|
1063 | 'U':22.75,'V':-17.596,'W':10.594,'X':1.577,'Y':5.778,'SH/L':0.002, |
---|
1064 | 'Back0':36.897,'Back1':-0.508,'Back2':.006, |
---|
1065 | 'Lam1':1.540500,'Lam2':1.544300,'I(L2)/I(L1)':0.5, |
---|
1066 | } |
---|
1067 | global parmDict2 |
---|
1068 | parmDict2 = { |
---|
1069 | 'pos0':5.7,'int0':1000.0,'sig0':0.5,'gam0':0.5, |
---|
1070 | 'U':2.,'V':-2.,'W':5.,'X':0.5,'Y':0.5,'SH/L':0.02, |
---|
1071 | 'Back0':5.,'Back1':-0.02,'Back2':.004, |
---|
1072 | # 'Lam1':1.540500,'Lam2':1.544300,'I(L2)/I(L1)':0.5, |
---|
1073 | } |
---|
1074 | global varyList |
---|
1075 | varyList = [] |
---|
1076 | |
---|
1077 | def test0(): |
---|
1078 | if NeedTestData: TestData() |
---|
1079 | msg = 'test ' |
---|
1080 | gplot = plotter.add('FCJ-Voigt, 11BM').gca() |
---|
1081 | gplot.plot(xdata,getBackground('',parmDict0,bakType,xdata)) |
---|
1082 | gplot.plot(xdata,getPeakProfile(parmDict0,xdata,varyList,bakType)) |
---|
1083 | fplot = plotter.add('FCJ-Voigt, Ka1+2').gca() |
---|
1084 | fplot.plot(xdata,getBackground('',parmDict1,bakType,xdata)) |
---|
1085 | fplot.plot(xdata,getPeakProfile(parmDict1,xdata,varyList,bakType)) |
---|
1086 | |
---|
1087 | def test1(): |
---|
1088 | if NeedTestData: TestData() |
---|
1089 | time0 = time.time() |
---|
1090 | for i in range(100): |
---|
1091 | y = getPeakProfile(parmDict1,xdata,varyList,bakType) |
---|
1092 | print '100+6*Ka1-2 peaks=1200 peaks',time.time()-time0 |
---|
1093 | |
---|
1094 | def test2(name,delt): |
---|
1095 | if NeedTestData: TestData() |
---|
1096 | varyList = [name,] |
---|
1097 | xdata = np.linspace(5.6,5.8,400) |
---|
1098 | hplot = plotter.add('derivatives test for '+name).gca() |
---|
1099 | hplot.plot(xdata,getPeakProfileDerv(parmDict2,xdata,varyList,bakType)[0]) |
---|
1100 | y0 = getPeakProfile(parmDict2,xdata,varyList,bakType) |
---|
1101 | parmDict2[name] += delt |
---|
1102 | y1 = getPeakProfile(parmDict2,xdata,varyList,bakType) |
---|
1103 | hplot.plot(xdata,(y1-y0)/delt,'r+') |
---|
1104 | |
---|
1105 | def test3(name,delt): |
---|
1106 | if NeedTestData: TestData() |
---|
1107 | names = ['pos','sig','gam','shl'] |
---|
1108 | idx = names.index(name) |
---|
1109 | myDict = {'pos':parmDict2['pos0'],'sig':parmDict2['sig0'],'gam':parmDict2['gam0'],'shl':parmDict2['SH/L']} |
---|
1110 | xdata = np.linspace(5.6,5.8,800) |
---|
1111 | dx = xdata[1]-xdata[0] |
---|
1112 | hplot = plotter.add('derivatives test for '+name).gca() |
---|
1113 | hplot.plot(xdata,100.*dx*getdFCJVoigt3(myDict['pos'],myDict['sig'],myDict['gam'],myDict['shl'],xdata)[idx+1]) |
---|
1114 | y0 = getFCJVoigt3(myDict['pos'],myDict['sig'],myDict['gam'],myDict['shl'],xdata) |
---|
1115 | myDict[name] += delt |
---|
1116 | y1 = getFCJVoigt3(myDict['pos'],myDict['sig'],myDict['gam'],myDict['shl'],xdata) |
---|
1117 | hplot.plot(xdata,(y1-y0)/delt,'r+') |
---|
1118 | |
---|
1119 | if __name__ == '__main__': |
---|
1120 | import GSASIItestplot as plot |
---|
1121 | global plotter |
---|
1122 | plotter = plot.PlotNotebook() |
---|
1123 | # test0() |
---|
1124 | # for name in ['int0','pos0','sig0','gam0','U','V','W','X','Y','SH/L','I(L2)/I(L1)']: |
---|
1125 | for name,shft in [['int0',0.1],['pos0',0.0001],['sig0',0.01],['gam0',0.00001], |
---|
1126 | ['U',0.1],['V',0.01],['W',0.01],['X',0.0001],['Y',0.0001],['SH/L',0.00005]]: |
---|
1127 | test2(name,shft) |
---|
1128 | for name,shft in [['pos',0.0001],['sig',0.01],['gam',0.0001],['shl',0.00005]]: |
---|
1129 | test3(name,shft) |
---|
1130 | print "OK" |
---|
1131 | plotter.StartEventLoop() |
---|