1 | from __future__ import division, print_function |
---|
2 | # Make sasmodels available on the path |
---|
3 | import sys, os |
---|
4 | BETA_DIR = os.path.dirname(os.path.realpath(__file__)) |
---|
5 | SASMODELS_DIR = os.path.dirname(os.path.dirname(BETA_DIR)) |
---|
6 | sys.path.insert(0, SASMODELS_DIR) |
---|
7 | |
---|
8 | from collections import namedtuple |
---|
9 | |
---|
10 | from matplotlib import pyplot as plt |
---|
11 | import numpy as np |
---|
12 | from numpy import pi, sin, cos, sqrt, fabs |
---|
13 | from numpy.polynomial.legendre import leggauss |
---|
14 | from scipy.special import j1 as J1 |
---|
15 | from numpy import inf |
---|
16 | from scipy.special import gammaln # type: ignore |
---|
17 | |
---|
18 | Theory = namedtuple('Theory', 'Q F1 F2 P S I Seff Ibeta') |
---|
19 | Theory.__new__.__defaults__ = (None,) * len(Theory._fields) |
---|
20 | |
---|
21 | #Used to calculate F(q) for the cylinder, sphere, ellipsoid models |
---|
22 | def sas_sinx_x(x): |
---|
23 | with np.errstate(all='ignore'): |
---|
24 | retvalue = sin(x)/x |
---|
25 | retvalue[x == 0.] = 1. |
---|
26 | return retvalue |
---|
27 | |
---|
28 | def sas_2J1x_x(x): |
---|
29 | with np.errstate(all='ignore'): |
---|
30 | retvalue = 2*J1(x)/x |
---|
31 | retvalue[x == 0] = 1. |
---|
32 | return retvalue |
---|
33 | |
---|
34 | def sas_3j1x_x(x): |
---|
35 | """return 3*j1(x)/x""" |
---|
36 | retvalue = np.empty_like(x) |
---|
37 | with np.errstate(all='ignore'): |
---|
38 | # GSL bessel_j1 taylor expansion |
---|
39 | index = (x < 0.25) |
---|
40 | y = x[index]**2 |
---|
41 | c1 = -1.0/10.0 |
---|
42 | c2 = 1.0/280.0 |
---|
43 | c3 = -1.0/15120.0 |
---|
44 | c4 = 1.0/1330560.0 |
---|
45 | c5 = -1.0/172972800.0 |
---|
46 | retvalue[index] = 1.0 + y*(c1 + y*(c2 + y*(c3 + y*(c4 + y*c5)))) |
---|
47 | index = ~index |
---|
48 | y = x[index] |
---|
49 | retvalue[index] = 3*(sin(y) - y*cos(y))/y**3 |
---|
50 | retvalue[x == 0.] = 1. |
---|
51 | return retvalue |
---|
52 | |
---|
53 | #Used to cross check my models with sasview models |
---|
54 | def build_model(model_name, q, **pars): |
---|
55 | from sasmodels.core import load_model_info, build_model as build_sasmodel |
---|
56 | from sasmodels.data import empty_data1D |
---|
57 | from sasmodels.direct_model import DirectModel |
---|
58 | model_info = load_model_info(model_name) |
---|
59 | model = build_sasmodel(model_info, dtype='double!') |
---|
60 | data = empty_data1D(q) |
---|
61 | calculator = DirectModel(data, model,cutoff=0) |
---|
62 | calculator.pars = pars.copy() |
---|
63 | calculator.pars.setdefault('background', 1e-3) |
---|
64 | return calculator |
---|
65 | |
---|
66 | #gives the hardsphere structure factor that sasview uses |
---|
67 | def _hardsphere_simple(q, radius_effective, volfraction): |
---|
68 | CUTOFFHS=0.05 |
---|
69 | if fabs(radius_effective) < 1.E-12: |
---|
70 | HARDSPH=1.0 |
---|
71 | return HARDSPH |
---|
72 | X = 1.0/( 1.0 -volfraction) |
---|
73 | D= X*X |
---|
74 | A= (1.+2.*volfraction)*D |
---|
75 | A *=A |
---|
76 | X=fabs(q*radius_effective*2.0) |
---|
77 | if X < 5.E-06: |
---|
78 | HARDSPH=1./A |
---|
79 | return HARDSPH |
---|
80 | X2 =X*X |
---|
81 | B = (1.0 +0.5*volfraction)*D |
---|
82 | B *= B |
---|
83 | B *= -6.*volfraction |
---|
84 | G=0.5*volfraction*A |
---|
85 | if X < CUTOFFHS: |
---|
86 | FF = 8.0*A +6.0*B + 4.0*G + ( -0.8*A -B/1.5 -0.5*G +(A/35. +0.0125*B +0.02*G)*X2)*X2 |
---|
87 | HARDSPH= 1./(1. + volfraction*FF ) |
---|
88 | return HARDSPH |
---|
89 | X4=X2*X2 |
---|
90 | S, C = sin(X), cos(X) |
---|
91 | FF= (( G*( (4.*X2 -24.)*X*S -(X4 -12.*X2 +24.)*C +24. )/X2 + B*(2.*X*S -(X2-2.)*C -2.) )/X + A*(S-X*C))/X |
---|
92 | HARDSPH= 1./(1. + 24.*volfraction*FF/X2 ) |
---|
93 | return HARDSPH |
---|
94 | |
---|
95 | def hardsphere_simple(q, radius_effective, volfraction): |
---|
96 | SQ = [_hardsphere_simple(qk, radius_effective, volfraction) for qk in q] |
---|
97 | return np.array(SQ) |
---|
98 | |
---|
99 | #Used in gaussian quadrature for polydispersity |
---|
100 | #returns values and the probability of those values based on gaussian distribution |
---|
101 | N_GAUSS = 35 |
---|
102 | NSIGMA_GAUSS = 3 |
---|
103 | def gaussian_distribution(center, sigma, lb, ub): |
---|
104 | #3 standard deviations covers approx. 99.7% |
---|
105 | if sigma != 0: |
---|
106 | nsigmas = NSIGMA_GAUSS |
---|
107 | x = np.linspace(center-sigma*nsigmas, center+sigma*nsigmas, num=N_GAUSS) |
---|
108 | x= x[(x >= lb) & (x <= ub)] |
---|
109 | px = np.exp((x-center)**2 / (-2.0 * sigma * sigma)) |
---|
110 | return x, px |
---|
111 | else: |
---|
112 | return np.array([center]), np.array([1]) |
---|
113 | |
---|
114 | N_SCHULZ = 80 |
---|
115 | NSIGMA_SCHULZ = 8 |
---|
116 | def schulz_distribution(center, sigma, lb, ub): |
---|
117 | if sigma != 0: |
---|
118 | nsigmas = NSIGMA_SCHULZ |
---|
119 | x = np.linspace(center-sigma*nsigmas, center+sigma*nsigmas, num=N_SCHULZ) |
---|
120 | x= x[(x >= lb) & (x <= ub)] |
---|
121 | R = x/center |
---|
122 | z = (center/sigma)**2 |
---|
123 | arg = z*np.log(z) + (z-1)*np.log(R) - R*z - np.log(center) - gammaln(z) |
---|
124 | px = np.exp(arg) |
---|
125 | return x, px |
---|
126 | else: |
---|
127 | return np.array([center]), np.array([1]) |
---|
128 | |
---|
129 | #returns the effective radius used in sasview |
---|
130 | def ER_ellipsoid(radius_polar, radius_equatorial): |
---|
131 | ee = np.empty_like(radius_polar) |
---|
132 | if radius_polar > radius_equatorial: |
---|
133 | ee = (radius_polar**2 - radius_equatorial**2)/radius_polar**2 |
---|
134 | elif radius_polar < radius_equatorial: |
---|
135 | ee = (radius_equatorial**2 - radius_polar**2) / radius_equatorial**2 |
---|
136 | else: |
---|
137 | ee = 2*radius_polar |
---|
138 | if (radius_polar * radius_equatorial != 0): |
---|
139 | bd = 1.0 - ee |
---|
140 | e1 = np.sqrt(ee) |
---|
141 | b1 = 1.0 + np.arcsin(e1) / (e1*np.sqrt(bd)) |
---|
142 | bL = (1.0 + e1) / (1.0 - e1) |
---|
143 | b2 = 1.0 + bd / 2 / e1 * np.log(bL) |
---|
144 | delta = 0.75 * b1 * b2 |
---|
145 | ddd = np.zeros_like(radius_polar) |
---|
146 | ddd = 2.0*(delta + 1.0)*radius_polar*radius_equatorial**2 |
---|
147 | return 0.5*ddd**(1.0 / 3.0) |
---|
148 | |
---|
149 | def ellipsoid_volume(radius_polar,radius_equatorial): |
---|
150 | volume = (4./3.)*pi*radius_polar*radius_equatorial**2 |
---|
151 | return volume |
---|
152 | |
---|
153 | # F1 is F(q) |
---|
154 | # F2 is F(g)^2 |
---|
155 | #IQM is I(q) with monodispersity |
---|
156 | #IQSM is I(q) with structure factor S(q) and monodispersity |
---|
157 | #IQBM is I(q) with Beta Approximation and monodispersity |
---|
158 | #SQ is monodisperse approach for structure factor |
---|
159 | #SQ_EFF is the effective structure factor from beta approx |
---|
160 | def ellipsoid_theta(q, radius_polar, radius_equatorial, sld, sld_solvent, |
---|
161 | volfraction=0, radius_effective=None): |
---|
162 | #creates values z and corresponding probabilities w from legendre-gauss quadrature |
---|
163 | volume = ellipsoid_volume(radius_polar, radius_equatorial) |
---|
164 | z, w = leggauss(76) |
---|
165 | F1 = np.zeros_like(q) |
---|
166 | F2 = np.zeros_like(q) |
---|
167 | #use a u subsition(u=cos) and then u=(z+1)/2 to change integration from |
---|
168 | #0->2pi with respect to alpha to -1->1 with respect to z, allowing us to use |
---|
169 | #legendre-gauss quadrature |
---|
170 | for k, qk in enumerate(q): |
---|
171 | r = sqrt(radius_equatorial**2*(1-((z+1)/2)**2)+radius_polar**2*((z+1)/2)**2) |
---|
172 | form = (sld-sld_solvent)*volume*sas_3j1x_x(qk*r) |
---|
173 | F2[k] = np.sum(w*form**2) |
---|
174 | F1[k] = np.sum(w*form) |
---|
175 | #the 1/2 comes from the change of variables mentioned above |
---|
176 | F2 = F2/2.0 |
---|
177 | F1 = F1/2.0 |
---|
178 | if radius_effective is None: |
---|
179 | radius_effective = ER_ellipsoid(radius_polar,radius_equatorial) |
---|
180 | SQ = hardsphere_simple(q, radius_effective, volfraction) |
---|
181 | SQ_EFF = 1 + F1**2/F2*(SQ - 1) |
---|
182 | IQM = 1e-4*F2/volume |
---|
183 | IQSM = volfraction*IQM*SQ |
---|
184 | IQBM = volfraction*IQM*SQ_EFF |
---|
185 | return Theory(Q=q, F1=F1, F2=F2, P=IQM, S=SQ, I=IQSM, Seff=SQ_EFF, Ibeta=IQBM) |
---|
186 | |
---|
187 | #IQD is I(q) polydispursed, IQSD is I(q)S(q) polydispursed, etc. |
---|
188 | #IQBD HAS NOT BEEN CROSS CHECKED AT ALL |
---|
189 | def ellipsoid_pe(q, radius_polar, radius_equatorial, sld, sld_solvent, |
---|
190 | radius_polar_pd=0.1, radius_equatorial_pd=0.1, |
---|
191 | radius_polar_pd_type='gaussian', |
---|
192 | radius_equatorial_pd_type='gaussian', |
---|
193 | volfraction=0, radius_effective=None, |
---|
194 | background=0, scale=1, |
---|
195 | norm='sasview'): |
---|
196 | if norm not in ['sasview', 'sasfit', 'yun']: |
---|
197 | raise TypeError("unknown norm "+norm) |
---|
198 | if radius_polar_pd_type == 'gaussian': |
---|
199 | Rp_val, Rp_prob = gaussian_distribution(radius_polar, radius_polar_pd*radius_polar, 0, inf) |
---|
200 | elif radius_polar_pd_type == 'schulz': |
---|
201 | Rp_val, Rp_prob = schulz_distribution(radius_polar, radius_polar_pd*radius_polar, 0, inf) |
---|
202 | if radius_equatorial_pd_type == 'gaussian': |
---|
203 | Re_val, Re_prob = gaussian_distribution(radius_equatorial, radius_equatorial_pd*radius_equatorial, 0, inf) |
---|
204 | elif radius_equatorial_pd_type == 'schulz': |
---|
205 | Re_val, Re_prob = schulz_distribution(radius_equatorial, radius_equatorial_pd*radius_equatorial, 0, inf) |
---|
206 | total_weight = total_volume = 0 |
---|
207 | radius_eff = 0 |
---|
208 | F1, F2 = np.zeros_like(q), np.zeros_like(q) |
---|
209 | for k, Rpk in enumerate(Rp_val): |
---|
210 | for i, Rei in enumerate(Re_val): |
---|
211 | theory = ellipsoid_theta(q,Rpk,Rei,sld,sld_solvent) |
---|
212 | volume = ellipsoid_volume(Rpk, Rei) |
---|
213 | weight = Rp_prob[k]*Re_prob[i] |
---|
214 | total_weight += weight |
---|
215 | total_volume += weight*volume |
---|
216 | F1 += theory.F1*weight |
---|
217 | F2 += theory.F2*weight |
---|
218 | radius_eff += weight*ER_ellipsoid(Rpk,Rei) |
---|
219 | F1 /= total_weight |
---|
220 | F2 /= total_weight |
---|
221 | average_volume = total_volume/total_weight |
---|
222 | if radius_effective is None: |
---|
223 | radius_effective = radius_eff/total_weight |
---|
224 | if norm == 'sasfit': |
---|
225 | IQD = F2 |
---|
226 | elif norm == 'sasview': |
---|
227 | # Note: internally, sasview uses F2/total_volume because: |
---|
228 | # average_volume = total_volume/total_weight |
---|
229 | # F2/total_weight / average_volume |
---|
230 | # = F2/total_weight / total_volume/total_weight |
---|
231 | # = F2/total_volume |
---|
232 | IQD = F2/average_volume*1e-4*volfraction |
---|
233 | elif norm == 'yun': |
---|
234 | F1 *= 1e-6 # Yun is using sld in 1/A^2, not 1e-6/A^2 |
---|
235 | F2 *= 1e-12 |
---|
236 | IQD = F2/average_volume*1e8*volfraction |
---|
237 | SQ = hardsphere_simple(q, radius_effective, volfraction) |
---|
238 | beta = F1**2/F2 |
---|
239 | SQ_EFF = 1 + beta*(SQ - 1) |
---|
240 | IQSD = IQD*SQ |
---|
241 | IQBD = IQD*SQ_EFF |
---|
242 | return Theory(Q=q, F1=F1, F2=F2, P=IQD, S=SQ, I=IQSD, Seff=SQ_EFF, Ibeta=IQBD) |
---|
243 | |
---|
244 | #polydispersity for sphere |
---|
245 | def sphere_r(q,radius,sld,sld_solvent, |
---|
246 | radius_pd=0.1, radius_pd_type='gaussian', |
---|
247 | volfraction=0, radius_effective=None, |
---|
248 | background=0, scale=1, |
---|
249 | norm='sasview'): |
---|
250 | if norm not in ['sasview', 'sasfit', 'yun']: |
---|
251 | raise TypeError("unknown norm "+norm) |
---|
252 | if radius_pd_type == 'gaussian': |
---|
253 | radius_val, radius_prob = gaussian_distribution(radius, radius_pd*radius, 0, inf) |
---|
254 | elif radius_pd_type == 'schulz': |
---|
255 | radius_val, radius_prob = schulz_distribution(radius, radius_pd*radius, 0, inf) |
---|
256 | total_weight = total_volume = 0 |
---|
257 | F1 = np.zeros_like(q) |
---|
258 | F2 = np.zeros_like(q) |
---|
259 | for k, rk in enumerate(radius_val): |
---|
260 | volume = 4./3.*pi*rk**3 |
---|
261 | total_weight += radius_prob[k] |
---|
262 | total_volume += radius_prob[k]*volume |
---|
263 | form = (sld-sld_solvent)*volume*sas_3j1x_x(q*rk) |
---|
264 | F2 += radius_prob[k]*form**2 |
---|
265 | F1 += radius_prob[k]*form |
---|
266 | F1 /= total_weight |
---|
267 | F2 /= total_weight |
---|
268 | average_volume = total_volume/total_weight |
---|
269 | |
---|
270 | if radius_effective is None: |
---|
271 | radius_effective = radius |
---|
272 | average_volume = total_volume/total_weight |
---|
273 | if norm == 'sasfit': |
---|
274 | IQD = F2 |
---|
275 | elif norm == 'sasview': |
---|
276 | IQD = F2/average_volume*1e-4*volfraction |
---|
277 | elif norm == 'yun': |
---|
278 | F1 *= 1e-6 # Yun is using sld in 1/A^2, not 1e-6/A^2 |
---|
279 | F2 *= 1e-12 |
---|
280 | IQD = F2/average_volume*1e8*volfraction |
---|
281 | SQ = hardsphere_simple(q, radius_effective, volfraction) |
---|
282 | beta = F1**2/F2 |
---|
283 | SQ_EFF = 1 + beta*(SQ - 1) |
---|
284 | IQSD = IQD*SQ |
---|
285 | IQBD = IQD*SQ_EFF |
---|
286 | return Theory(Q=q, F1=F1, F2=F2, P=IQD, S=SQ, I=IQSD, Seff=SQ_EFF, Ibeta=IQBD) |
---|
287 | |
---|
288 | ############################################################################### |
---|
289 | ############################################################################### |
---|
290 | ############################################################################### |
---|
291 | ################## ################## |
---|
292 | ################## TESTS ################## |
---|
293 | ################## ################## |
---|
294 | ############################################################################### |
---|
295 | ############################################################################### |
---|
296 | ############################################################################### |
---|
297 | |
---|
298 | def popn(d, keys): |
---|
299 | """ |
---|
300 | Splits a dict into two, with any key of *d* which is in *keys* removed |
---|
301 | from *d* and added to *b*. Returns *b*. |
---|
302 | """ |
---|
303 | b = {} |
---|
304 | for k in keys: |
---|
305 | try: |
---|
306 | b[k] = d.pop(k) |
---|
307 | except KeyError: |
---|
308 | pass |
---|
309 | return b |
---|
310 | |
---|
311 | def sasmodels_theory(q, Pname, **pars): |
---|
312 | """ |
---|
313 | Call sasmodels to compute the model with and without a hard sphere |
---|
314 | structure factor. |
---|
315 | """ |
---|
316 | #mono_pars = {k: (0 if k.endswith('_pd') else v) for k, v in pars.items()} |
---|
317 | Ppars = pars.copy() |
---|
318 | Spars = popn(Ppars, ['radius_effective', 'volfraction']) |
---|
319 | Ipars = pars.copy() |
---|
320 | |
---|
321 | # Autofill npts and nsigmas for the given pd type |
---|
322 | for k, v in pars.items(): |
---|
323 | if k.endswith("_pd_type"): |
---|
324 | if v == "gaussian": |
---|
325 | n, nsigmas = N_GAUSS, NSIGMA_GAUSS |
---|
326 | elif v == "schulz": |
---|
327 | n, nsigmas = N_SCHULZ, NSIGMA_SCHULZ |
---|
328 | Ppars.setdefault(k.replace("_pd_type", "_pd_n"), n) |
---|
329 | Ppars.setdefault(k.replace("_pd_type", "_pd_nsigma"), nsigmas) |
---|
330 | Ipars.setdefault(k.replace("_pd_type", "_pd_n"), n) |
---|
331 | Ipars.setdefault(k.replace("_pd_type", "_pd_nsigma"), nsigmas) |
---|
332 | |
---|
333 | #Ppars['scale'] = Spars.get('volfraction', 1) |
---|
334 | P = build_model(Pname, q) |
---|
335 | S = build_model("hardsphere", q) |
---|
336 | I = build_model(Pname+"@hardsphere", q) |
---|
337 | Pq = P(**Ppars)*pars.get('volfraction', 1) |
---|
338 | #Sq = S(**Spars) |
---|
339 | Iq = I(**Ipars) |
---|
340 | #Iq = Pq*Sq*pars.get('volfraction', 1) |
---|
341 | Sq = Iq/Pq |
---|
342 | return Theory(Q=q, F1=None, F2=None, P=Pq, S=Sq, I=Iq, Seff=None, Ibeta=None) |
---|
343 | |
---|
344 | def compare(title, target, actual, fields='F1 F2 P S I Seff Ibeta'): |
---|
345 | """ |
---|
346 | Plot fields in common between target and actual, along with relative error. |
---|
347 | """ |
---|
348 | available = [s for s in fields.split() |
---|
349 | if getattr(target, s) is not None and getattr(actual, s) is not None] |
---|
350 | rows = len(available) |
---|
351 | for row, field in enumerate(available): |
---|
352 | Q = target.Q |
---|
353 | I1, I2 = getattr(target, field), getattr(actual, field) |
---|
354 | plt.subplot(rows, 2, 2*row+1) |
---|
355 | plt.loglog(Q, abs(I1), label="target "+field) |
---|
356 | plt.loglog(Q, abs(I2), label="value "+field) |
---|
357 | #plt.legend(loc="upper left", bbox_to_anchor=(1,1)) |
---|
358 | plt.legend(loc='lower left') |
---|
359 | plt.subplot(rows, 2, 2*row+2) |
---|
360 | plt.semilogx(Q, I2/I1 - 1, label="relative error") |
---|
361 | #plt.semilogx(Q, I1/I2 - 1, label="relative error") |
---|
362 | plt.tight_layout() |
---|
363 | plt.suptitle(title) |
---|
364 | plt.show() |
---|
365 | |
---|
366 | def data_file(name): |
---|
367 | return os.path.join(BETA_DIR, 'data_files', name) |
---|
368 | |
---|
369 | def load_sasfit(path): |
---|
370 | data = np.loadtxt(path, dtype=str, delimiter=';').T |
---|
371 | data = np.vstack((map(float, v) for v in data[0:2])) |
---|
372 | return data |
---|
373 | |
---|
374 | COMPARISON = {} # Type: Dict[(str,str,str)] -> Callable[(), None] |
---|
375 | |
---|
376 | def compare_sasview_sphere(pd_type='schulz'): |
---|
377 | q = np.logspace(-5, 0, 250) |
---|
378 | model = 'sphere' |
---|
379 | pars = dict( |
---|
380 | radius=20,sld=4,sld_solvent=1, |
---|
381 | background=0, |
---|
382 | radius_pd=.1, radius_pd_type=pd_type, |
---|
383 | volfraction=0.15, |
---|
384 | #radius_effective=12.59921049894873, # equivalent average sphere radius |
---|
385 | ) |
---|
386 | target = sasmodels_theory(q, model, **pars) |
---|
387 | actual = sphere_r(q, norm='sasview', **pars) |
---|
388 | title = " ".join(("sasmodels", model, pd_type)) |
---|
389 | compare(title, target, actual) |
---|
390 | COMPARISON[('sasview','sphere','gaussian')] = lambda: compare_sasview_sphere(pd_type='gaussian') |
---|
391 | COMPARISON[('sasview','sphere','schulz')] = lambda: compare_sasview_sphere(pd_type='schulz') |
---|
392 | |
---|
393 | def compare_sasview_ellipsoid(pd_type='gaussian'): |
---|
394 | q = np.logspace(-5, 0, 50) |
---|
395 | model = 'ellipsoid' |
---|
396 | pars = dict( |
---|
397 | radius_polar=20,radius_equatorial=400,sld=4,sld_solvent=1, |
---|
398 | background=0, |
---|
399 | radius_polar_pd=.1, radius_polar_pd_type=pd_type, |
---|
400 | radius_equatorial_pd=.1, radius_equatorial_pd_type=pd_type, |
---|
401 | volfraction=0.15, |
---|
402 | #radius_effective=12.59921049894873, |
---|
403 | ) |
---|
404 | target = sasmodels_theory(q, model, **pars) |
---|
405 | actual = ellipsoid_pe(q, norm='sasview', **pars) |
---|
406 | title = " ".join(("sasmodels", model, pd_type)) |
---|
407 | compare(title, target, actual) |
---|
408 | COMPARISON[('sasview','ellipsoid','gaussian')] = lambda: compare_sasview_ellipsoid(pd_type='gaussian') |
---|
409 | COMPARISON[('sasview','ellipsoid','schulz')] = lambda: compare_sasview_ellipsoid(pd_type='schulz') |
---|
410 | |
---|
411 | def compare_yun_ellipsoid_mono(): |
---|
412 | pars = { |
---|
413 | 'radius_polar': 20, 'radius_polar_pd': 0, 'radius_polar_pd_type': 'gaussian', |
---|
414 | 'radius_equatorial': 10, 'radius_equatorial_pd': 0, 'radius_equatorial_pd_type': 'gaussian', |
---|
415 | 'sld': 2, 'sld_solvent': 1, |
---|
416 | 'volfraction': 0.15, |
---|
417 | # Yun uses radius for same volume sphere for effective radius |
---|
418 | # whereas sasview uses the average curvature. |
---|
419 | 'radius_effective': 12.59921049894873, |
---|
420 | } |
---|
421 | |
---|
422 | data = np.loadtxt(data_file('yun_ellipsoid.dat'),skiprows=2).T |
---|
423 | Q = data[0] |
---|
424 | F1 = data[1] |
---|
425 | P = data[3]*pars['volfraction'] |
---|
426 | S = data[5] |
---|
427 | Seff = data[6] |
---|
428 | target = Theory(Q=Q, F1=F1, P=P, S=S, Seff=Seff) |
---|
429 | actual = ellipsoid_pe(Q, norm='yun', **pars) |
---|
430 | title = " ".join(("yun", "ellipsoid", "no pd")) |
---|
431 | #compare(title, target, actual, fields="P S I Seff Ibeta") |
---|
432 | compare(title, target, actual) |
---|
433 | COMPARISON[('yun','ellipsoid','gaussian')] = compare_yun_ellipsoid_mono |
---|
434 | COMPARISON[('yun','ellipsoid','schulz')] = compare_yun_ellipsoid_mono |
---|
435 | |
---|
436 | def compare_yun_sphere_gauss(): |
---|
437 | # Note: yun uses gauss limits from R0/10 to R0 + 5 sigma steps sigma/100 |
---|
438 | # With pd = 0.1, that's 14 sigma and 1400 points. |
---|
439 | pars = { |
---|
440 | 'radius': 20, 'radius_pd': 0.1, 'radius_pd_type': 'gaussian', |
---|
441 | 'sld': 6, 'sld_solvent': 0, |
---|
442 | 'volfraction': 0.1, |
---|
443 | } |
---|
444 | |
---|
445 | data = np.loadtxt(data_file('testPolydisperseGaussianSphere.dat'),skiprows=2).T |
---|
446 | Q = data[0] |
---|
447 | F1 = data[1] |
---|
448 | P = data[3] |
---|
449 | S = data[5] |
---|
450 | Seff = data[6] |
---|
451 | target = Theory(Q=Q, F1=F1, P=P, S=S, Seff=Seff) |
---|
452 | actual = sphere_r(Q, norm='yun', **pars) |
---|
453 | title = " ".join(("yun", "sphere", "10% dispersion 10% Vf")) |
---|
454 | compare(title, target, actual) |
---|
455 | data = np.loadtxt(data_file('testPolydisperseGaussianSphere2.dat'),skiprows=2).T |
---|
456 | pars.update(radius_pd=0.15) |
---|
457 | Q = data[0] |
---|
458 | F1 = data[1] |
---|
459 | P = data[3] |
---|
460 | S = data[5] |
---|
461 | Seff = data[6] |
---|
462 | target = Theory(Q=Q, F1=F1, P=P, S=S, Seff=Seff) |
---|
463 | actual = sphere_r(Q, norm='yun', **pars) |
---|
464 | title = " ".join(("yun", "sphere", "15% dispersion 10% Vf")) |
---|
465 | compare(title, target, actual) |
---|
466 | COMPARISON[('yun','sphere','gaussian')] = compare_yun_sphere_gauss |
---|
467 | |
---|
468 | |
---|
469 | def compare_sasfit_sphere_gauss(): |
---|
470 | #N=1,s=2,X0=20,distr radius R=20,eta_core=4,eta_solv=1,.3 |
---|
471 | pars = { |
---|
472 | 'radius': 20, 'radius_pd': 0.1, 'radius_pd_type': 'gaussian', |
---|
473 | 'sld': 4, 'sld_solvent': 1, |
---|
474 | 'volfraction': 0.3, |
---|
475 | } |
---|
476 | |
---|
477 | Q, IQ = load_sasfit(data_file('sasfit_sphere_IQD.txt')) |
---|
478 | Q, IQSD = load_sasfit(data_file('sasfit_sphere_IQSD.txt')) |
---|
479 | Q, IQBD = load_sasfit(data_file('sasfit_sphere_IQBD.txt')) |
---|
480 | Q, SQ = load_sasfit(data_file('sasfit_polydisperse_sphere_sq.txt')) |
---|
481 | Q, SQ_EFF = load_sasfit(data_file('sasfit_polydisperse_sphere_sqeff.txt')) |
---|
482 | target = Theory(Q=Q, F1=None, F2=None, P=IQ, S=SQ, I=IQSD, Seff=SQ_EFF, Ibeta=IQBD) |
---|
483 | actual = sphere_r(Q, norm="sasfit", **pars) |
---|
484 | title = " ".join(("sasfit", "sphere", "pd=10% gaussian")) |
---|
485 | compare(title, target, actual) |
---|
486 | #compare(title, target, actual, fields="P") |
---|
487 | COMPARISON[('sasfit','sphere','gaussian')] = compare_sasfit_sphere_gauss |
---|
488 | |
---|
489 | def compare_sasfit_sphere_schulz(): |
---|
490 | #radius=20,sld=4,sld_solvent=1,volfraction=0.3,radius_pd=0.1 |
---|
491 | #We have scaled the output from sasfit by 1e-4*volume*volfraction |
---|
492 | #0.10050378152592121 |
---|
493 | pars = { |
---|
494 | 'radius': 20, 'radius_pd': 0.1, 'radius_pd_type': 'schulz', |
---|
495 | 'sld': 4, 'sld_solvent': 1, |
---|
496 | 'volfraction': 0.3, |
---|
497 | } |
---|
498 | |
---|
499 | Q, IQ = load_sasfit(data_file('richard_test.txt')) |
---|
500 | Q, IQSD = load_sasfit(data_file('richard_test2.txt')) |
---|
501 | Q, IQBD = load_sasfit(data_file('richard_test3.txt')) |
---|
502 | target = Theory(Q=Q, F1=None, F2=None, P=IQ, S=None, I=IQSD, Seff=None, Ibeta=IQBD) |
---|
503 | actual = sphere_r(Q, norm="sasfit", **pars) |
---|
504 | title = " ".join(("sasfit", "sphere", "pd=10% schulz")) |
---|
505 | compare(title, target, actual) |
---|
506 | COMPARISON[('sasfit','sphere','schulz')] = compare_sasfit_sphere_schulz |
---|
507 | |
---|
508 | def compare_sasfit_ellipsoid_schulz(): |
---|
509 | #polarradius=20, equatorialradius=10, sld=4,sld_solvent=1,volfraction=0.3,radius_polar_pd=0.1 |
---|
510 | #Effective radius =13.1353356684 |
---|
511 | #We have scaled the output from sasfit by 1e-4*volume*volfraction |
---|
512 | #0.10050378152592121 |
---|
513 | pars = { |
---|
514 | 'radius_polar': 20, 'radius_polar_pd': 0.1, 'radius_polar_pd_type': 'schulz', |
---|
515 | 'radius_equatorial': 10, 'radius_equatorial_pd': 0., 'radius_equatorial_pd_type': 'schulz', |
---|
516 | 'sld': 4, 'sld_solvent': 1, |
---|
517 | 'volfraction': 0.3, 'radius_effective': 13.1353356684, |
---|
518 | } |
---|
519 | |
---|
520 | Q, IQ = load_sasfit(data_file('richard_test4.txt')) |
---|
521 | Q, IQSD = load_sasfit(data_file('richard_test5.txt')) |
---|
522 | Q, IQBD = load_sasfit(data_file('richard_test6.txt')) |
---|
523 | target = Theory(Q=Q, F1=None, F2=None, P=IQ, S=None, I=IQSD, Seff=None, Ibeta=IQBD) |
---|
524 | actual = ellipsoid_pe(Q, norm="sasfit", **pars) |
---|
525 | title = " ".join(("sasfit", "ellipsoid", "pd=10% schulz")) |
---|
526 | compare(title, target, actual) |
---|
527 | COMPARISON[('sasfit','ellipsoid','schulz')] = compare_sasfit_ellipsoid_schulz |
---|
528 | |
---|
529 | |
---|
530 | def compare_sasfit_ellipsoid_gaussian(): |
---|
531 | pars = { |
---|
532 | 'radius_polar': 20, 'radius_polar_pd': 0, 'radius_polar_pd_type': 'gaussian', |
---|
533 | 'radius_equatorial': 10, 'radius_equatorial_pd': 0, 'radius_equatorial_pd_type': 'gaussian', |
---|
534 | 'sld': 4, 'sld_solvent': 1, |
---|
535 | 'volfraction': 0, 'radius_effective': None, |
---|
536 | } |
---|
537 | |
---|
538 | #Rp=20,Re=10,eta_core=4,eta_solv=1 |
---|
539 | Q, PQ0 = load_sasfit(data_file('sasfit_ellipsoid_IQM.txt')) |
---|
540 | pars.update(volfraction=0, radius_polar_pd=0.0, radius_equatorial_pd=0, radius_effective=None) |
---|
541 | actual = ellipsoid_pe(Q, norm='sasfit', **pars) |
---|
542 | target = Theory(Q=Q, P=PQ0) |
---|
543 | compare("sasfit ellipsoid no poly", target, actual); plt.show() |
---|
544 | |
---|
545 | #N=1,s=2,X0=20,distr 10% polar Rp=20,Re=10,eta_core=4,eta_solv=1, no structure poly |
---|
546 | Q, PQ_Rp10 = load_sasfit(data_file('sasfit_ellipsoid_IQD.txt')) |
---|
547 | pars.update(volfraction=0, radius_polar_pd=0.1, radius_equatorial_pd=0.0, radius_effective=None) |
---|
548 | actual = ellipsoid_pe(Q, norm='sasfit', **pars) |
---|
549 | target = Theory(Q=Q, P=PQ_Rp10) |
---|
550 | compare("sasfit ellipsoid P(Q) 10% Rp", target, actual); plt.show() |
---|
551 | #N=1,s=1,X0=10,distr 10% equatorial Rp=20,Re=10,eta_core=4,eta_solv=1, no structure poly |
---|
552 | Q, PQ_Re10 = load_sasfit(data_file('sasfit_ellipsoid_IQD2.txt')) |
---|
553 | pars.update(volfraction=0, radius_polar_pd=0.0, radius_equatorial_pd=0.1, radius_effective=None) |
---|
554 | actual = ellipsoid_pe(Q, norm='sasfit', **pars) |
---|
555 | target = Theory(Q=Q, P=PQ_Re10) |
---|
556 | compare("sasfit ellipsoid P(Q) 10% Re", target, actual); plt.show() |
---|
557 | #N=1,s=6,X0=20,distr 30% polar Rp=20,Re=10,eta_core=4,eta_solv=1, no structure poly |
---|
558 | Q, PQ_Rp30 = load_sasfit(data_file('sasfit_ellipsoid_IQD3.txt')) |
---|
559 | pars.update(volfraction=0, radius_polar_pd=0.3, radius_equatorial_pd=0.0, radius_effective=None) |
---|
560 | actual = ellipsoid_pe(Q, norm='sasfit', **pars) |
---|
561 | target = Theory(Q=Q, P=PQ_Rp30) |
---|
562 | compare("sasfit ellipsoid P(Q) 30% Rp", target, actual); plt.show() |
---|
563 | #N=1,s=3,X0=10,distr 30% equatorial Rp=20,Re=10,eta_core=4,eta_solv=1, no structure poly |
---|
564 | Q, PQ_Re30 = load_sasfit(data_file('sasfit_ellipsoid_IQD4.txt')) |
---|
565 | pars.update(volfraction=0, radius_polar_pd=0.0, radius_equatorial_pd=0.3, radius_effective=None) |
---|
566 | actual = ellipsoid_pe(Q, norm='sasfit', **pars) |
---|
567 | target = Theory(Q=Q, P=PQ_Re30) |
---|
568 | compare("sasfit ellipsoid P(Q) 30% Re", target, actual); plt.show() |
---|
569 | #N=1,s=12,X0=20,distr 60% polar Rp=20,Re=10,eta_core=4,eta_solv=1, no structure poly |
---|
570 | Q, PQ_Rp60 = load_sasfit(data_file('sasfit_ellipsoid_IQD5.txt')) |
---|
571 | pars.update(volfraction=0, radius_polar_pd=0.6, radius_equatorial_pd=0.0, radius_effective=None) |
---|
572 | actual = ellipsoid_pe(Q, norm='sasfit', **pars) |
---|
573 | target = Theory(Q=Q, P=PQ_Rp60) |
---|
574 | compare("sasfit ellipsoid P(Q) 60% Rp", target, actual); plt.show() |
---|
575 | #N=1,s=6,X0=10,distr 60% equatorial Rp=20,Re=10,eta_core=4,eta_solv=1, no structure poly |
---|
576 | Q, PQ_Re60 = load_sasfit(data_file('sasfit_ellipsoid_IQD6.txt')) |
---|
577 | pars.update(volfraction=0, radius_polar_pd=0.0, radius_equatorial_pd=0.6, radius_effective=None) |
---|
578 | actual = ellipsoid_pe(Q, norm='sasfit', **pars) |
---|
579 | target = Theory(Q=Q, P=PQ_Re60) |
---|
580 | compare("sasfit ellipsoid P(Q) 60% Re", target, actual); plt.show() |
---|
581 | |
---|
582 | #N=1,s=2,X0=20,distr polar Rp=20,Re=10,eta_core=4,eta_solv=1, hardsphere ,13.1354236254,.15 |
---|
583 | Q, SQ = load_sasfit(data_file('sasfit_polydisperse_ellipsoid_sq.txt')) |
---|
584 | Q, SQ_EFF = load_sasfit(data_file('sasfit_polydisperse_ellipsoid_sqeff.txt')) |
---|
585 | pars.update(volfraction=0.15, radius_polar_pd=0.1, radius_equatorial_pd=0, radius_effective=13.1354236254) |
---|
586 | actual = ellipsoid_pe(Q, norm='sasfit', **pars) |
---|
587 | target = Theory(Q=Q, S=SQ, Seff=SQ_EFF) |
---|
588 | compare("sasfit ellipsoid P(Q) 10% Rp 15% Vf", target, actual); plt.show() |
---|
589 | #N=1,s=6,X0=20,distr polar Rp=20,Re=10,eta_core=4,eta_solv=1, hardsphere ,13.0901197149,.15 |
---|
590 | Q, SQ = load_sasfit(data_file('sasfit_polydisperse_ellipsoid_sq2.txt')) |
---|
591 | Q, SQ_EFF = load_sasfit(data_file('sasfit_polydisperse_ellipsoid_sqeff2.txt')) |
---|
592 | pars.update(volfraction=0.15, radius_polar_pd=0.3, radius_equatorial_pd=0, radius_effective=13.0901197149) |
---|
593 | actual = ellipsoid_pe(Q, norm='sasfit', **pars) |
---|
594 | target = Theory(Q=Q, S=SQ, Seff=SQ_EFF) |
---|
595 | compare("sasfit ellipsoid P(Q) 30% Rp 15% Vf", target, actual); plt.show() |
---|
596 | #N=1,s=12,X0=20,distr polar Rp=20,Re=10,eta_core=4,eta_solv=1, hardsphere ,13.336060917,.15 |
---|
597 | Q, SQ = load_sasfit(data_file('sasfit_polydisperse_ellipsoid_sq3.txt')) |
---|
598 | Q, SQ_EFF = load_sasfit(data_file('sasfit_polydisperse_ellipsoid_sqeff3.txt')) |
---|
599 | pars.update(volfraction=0.15, radius_polar_pd=0.6, radius_equatorial_pd=0, radius_effective=13.336060917) |
---|
600 | actual = ellipsoid_pe(Q, norm='sasfit', **pars) |
---|
601 | target = Theory(Q=Q, S=SQ, Seff=SQ_EFF) |
---|
602 | compare("sasfit ellipsoid P(Q) 60% Rp 15% Vf", target, actual); plt.show() |
---|
603 | |
---|
604 | #N=1,s=2,X0=20,distr polar Rp=20,Re=10,eta_core=4,eta_solv=1, hardsphere ,13.1354236254,.3 |
---|
605 | Q, SQ = load_sasfit(data_file('sasfit_polydisperse_ellipsoid_sq4.txt')) |
---|
606 | Q, SQ_EFF = load_sasfit(data_file('sasfit_polydisperse_ellipsoid_sqeff4.txt')) |
---|
607 | pars.update(volfraction=0.3, radius_polar_pd=0.1, radius_equatorial_pd=0, radius_effective=13.1354236254) |
---|
608 | actual = ellipsoid_pe(Q, norm='sasfit', **pars) |
---|
609 | target = Theory(Q=Q, S=SQ, Seff=SQ_EFF) |
---|
610 | compare("sasfit ellipsoid P(Q) 10% Rp 30% Vf", target, actual); plt.show() |
---|
611 | #N=1,s=6,X0=20,distr polar Rp=20,Re=10,eta_core=4,eta_solv=1, hardsphere ,13.0901197149,.3 |
---|
612 | Q, SQ = load_sasfit(data_file('sasfit_polydisperse_ellipsoid_sq5.txt')) |
---|
613 | Q, SQ_EFF = load_sasfit(data_file('sasfit_polydisperse_ellipsoid_sqeff5.txt')) |
---|
614 | pars.update(volfraction=0.3, radius_polar_pd=0.3, radius_equatorial_pd=0, radius_effective=13.0901197149) |
---|
615 | actual = ellipsoid_pe(Q, norm='sasfit', **pars) |
---|
616 | target = Theory(Q=Q, S=SQ, Seff=SQ_EFF) |
---|
617 | compare("sasfit ellipsoid P(Q) 30% Rp 30% Vf", target, actual); plt.show() |
---|
618 | #N=1,s=12,X0=20,distr polar Rp=20,Re=10,eta_core=4,eta_solv=1, hardsphere ,13.336060917,.3 |
---|
619 | Q, SQ = load_sasfit(data_file('sasfit_polydisperse_ellipsoid_sq6.txt')) |
---|
620 | Q, SQ_EFF = load_sasfit(data_file('sasfit_polydisperse_ellipsoid_sqeff6.txt')) |
---|
621 | pars.update(volfraction=0.3, radius_polar_pd=0.6, radius_equatorial_pd=0, radius_effective=13.336060917) |
---|
622 | actual = ellipsoid_pe(Q, norm='sasfit', **pars) |
---|
623 | target = Theory(Q=Q, S=SQ, Seff=SQ_EFF) |
---|
624 | compare("sasfit ellipsoid P(Q) 60% Rp 30% Vf", target, actual); plt.show() |
---|
625 | |
---|
626 | #N=1,s=2,X0=20,distr polar Rp=20,Re=10,eta_core=4,eta_solv=1, hardsphere ,13.1354236254,.6 |
---|
627 | Q, SQ = load_sasfit(data_file('sasfit_polydisperse_ellipsoid_sq7.txt')) |
---|
628 | Q, SQ_EFF = load_sasfit(data_file('sasfit_polydisperse_ellipsoid_sqeff7.txt')) |
---|
629 | pars.update(volfraction=0.6, radius_polar_pd=0.1, radius_equatorial_pd=0, radius_effective=13.1354236254) |
---|
630 | actual = ellipsoid_pe(Q, norm='sasfit', **pars) |
---|
631 | target = Theory(Q=Q, S=SQ, Seff=SQ_EFF) |
---|
632 | compare("sasfit ellipsoid P(Q) 10% Rp 60% Vf", target, actual); plt.show() |
---|
633 | #N=1,s=6,X0=20,distr polar Rp=20,Re=10,eta_core=4,eta_solv=1, hardsphere ,13.0901197149,.6 |
---|
634 | Q, SQ = load_sasfit(data_file('sasfit_polydisperse_ellipsoid_sq8.txt')) |
---|
635 | Q, SQ_EFF = load_sasfit(data_file('sasfit_polydisperse_ellipsoid_sqeff8.txt')) |
---|
636 | pars.update(volfraction=0.6, radius_polar_pd=0.3, radius_equatorial_pd=0, radius_effective=13.0901197149) |
---|
637 | actual = ellipsoid_pe(Q, norm='sasfit', **pars) |
---|
638 | target = Theory(Q=Q, S=SQ, Seff=SQ_EFF) |
---|
639 | compare("sasfit ellipsoid P(Q) 30% Rp 60% Vf", target, actual); plt.show() |
---|
640 | #N=1,s=12,X0=20,distr polar Rp=20,Re=10,eta_core=4,eta_solv=1, hardsphere ,13.336060917,.6 |
---|
641 | Q, SQ = load_sasfit(data_file('sasfit_polydisperse_ellipsoid_sq9.txt')) |
---|
642 | Q, SQ_EFF = load_sasfit(data_file('sasfit_polydisperse_ellipsoid_sqeff9.txt')) |
---|
643 | pars.update(volfraction=0.6, radius_polar_pd=0.6, radius_equatorial_pd=0, radius_effective=13.336060917) |
---|
644 | actual = ellipsoid_pe(Q, norm='sasfit', **pars) |
---|
645 | target = Theory(Q=Q, S=SQ, Seff=SQ_EFF) |
---|
646 | compare("sasfit ellipsoid P(Q) 60% Rp 60% Vf", target, actual); plt.show() |
---|
647 | COMPARISON[('sasfit','ellipsoid','gaussian')] = compare_sasfit_ellipsoid_gaussian |
---|
648 | |
---|
649 | def main(): |
---|
650 | key = tuple(sys.argv[1:]) |
---|
651 | if key not in COMPARISON: |
---|
652 | print("usage: sasfit_compare.py [sasview|sasfit|yun] [sphere|ellipsoid] [gaussian|schulz]") |
---|
653 | return |
---|
654 | comparison = COMPARISON[key] |
---|
655 | comparison() |
---|
656 | |
---|
657 | if __name__ == "__main__": |
---|
658 | main() |
---|