1 | |
---|
2 | ##################################################################### |
---|
3 | #This software was developed by the University of Tennessee as part of the |
---|
4 | #Distributed Data Analysis of Neutron Scattering Experiments (DANSE) |
---|
5 | #project funded by the US National Science Foundation. |
---|
6 | #See the license text in license.txt |
---|
7 | #copyright 2008, University of Tennessee |
---|
8 | ###################################################################### |
---|
9 | import numpy |
---|
10 | #import math |
---|
11 | import logging |
---|
12 | import sys |
---|
13 | import DataLoader.extensions.smearer as smearer |
---|
14 | from DataLoader.smearing_2d import Smearer2D |
---|
15 | |
---|
16 | def smear_selection(data1D): |
---|
17 | """ |
---|
18 | Creates the right type of smearer according |
---|
19 | to the data. |
---|
20 | |
---|
21 | The canSAS format has a rule that either |
---|
22 | slit smearing data OR resolution smearing data |
---|
23 | is available. |
---|
24 | |
---|
25 | For the present purpose, we choose the one that |
---|
26 | has none-zero data. If both slit and resolution |
---|
27 | smearing arrays are filled with good data |
---|
28 | (which should not happen), then we choose the |
---|
29 | resolution smearing data. |
---|
30 | |
---|
31 | :param data1D: Data1D object |
---|
32 | """ |
---|
33 | # Sanity check. If we are not dealing with a SANS Data1D |
---|
34 | # object, just return None |
---|
35 | if data1D.__class__.__name__ not in ['Data1D', 'Theory1D']: |
---|
36 | if data1D == None: |
---|
37 | return None |
---|
38 | elif data1D.dqx_data == None or data1D.dqy_data == None: |
---|
39 | return None |
---|
40 | return Smearer2D(data1D) |
---|
41 | |
---|
42 | if not hasattr(data1D, "dx") and not hasattr(data1D, "dxl")\ |
---|
43 | and not hasattr(data1D, "dxw"): |
---|
44 | return None |
---|
45 | |
---|
46 | # Look for resolution smearing data |
---|
47 | _found_resolution = False |
---|
48 | if data1D.dx is not None and len(data1D.dx) == len(data1D.x): |
---|
49 | |
---|
50 | # Check that we have non-zero data |
---|
51 | if data1D.dx[0] > 0.0: |
---|
52 | _found_resolution = True |
---|
53 | #print "_found_resolution",_found_resolution |
---|
54 | #print "data1D.dx[0]",data1D.dx[0],data1D.dxl[0] |
---|
55 | # If we found resolution smearing data, return a QSmearer |
---|
56 | if _found_resolution == True: |
---|
57 | return QSmearer(data1D) |
---|
58 | |
---|
59 | # Look for slit smearing data |
---|
60 | _found_slit = False |
---|
61 | if data1D.dxl is not None and len(data1D.dxl) == len(data1D.x) \ |
---|
62 | and data1D.dxw is not None and len(data1D.dxw) == len(data1D.x): |
---|
63 | |
---|
64 | # Check that we have non-zero data |
---|
65 | if data1D.dxl[0] > 0.0 or data1D.dxw[0] > 0.0: |
---|
66 | _found_slit = True |
---|
67 | |
---|
68 | # Sanity check: all data should be the same as a function of Q |
---|
69 | for item in data1D.dxl: |
---|
70 | if data1D.dxl[0] != item: |
---|
71 | _found_resolution = False |
---|
72 | break |
---|
73 | |
---|
74 | for item in data1D.dxw: |
---|
75 | if data1D.dxw[0] != item: |
---|
76 | _found_resolution = False |
---|
77 | break |
---|
78 | # If we found slit smearing data, return a slit smearer |
---|
79 | if _found_slit == True: |
---|
80 | return SlitSmearer(data1D) |
---|
81 | return None |
---|
82 | |
---|
83 | |
---|
84 | class _BaseSmearer(object): |
---|
85 | |
---|
86 | def __init__(self): |
---|
87 | self.nbins = 0 |
---|
88 | self._weights = None |
---|
89 | ## Internal flag to keep track of C++ smearer initialization |
---|
90 | self._init_complete = False |
---|
91 | self._smearer = None |
---|
92 | |
---|
93 | def __deepcopy__(self, memo={}): |
---|
94 | """ |
---|
95 | Return a valid copy of self. |
---|
96 | Avoid copying the _smearer C object and force a matrix recompute |
---|
97 | when the copy is used. |
---|
98 | """ |
---|
99 | result = _BaseSmearer() |
---|
100 | result.nbins = self.nbins |
---|
101 | return result |
---|
102 | |
---|
103 | def _compute_matrix(self): |
---|
104 | """ |
---|
105 | """ |
---|
106 | return NotImplemented |
---|
107 | |
---|
108 | def get_bin_range(self, q_min=None, q_max=None): |
---|
109 | """ |
---|
110 | |
---|
111 | :param q_min: minimum q-value to smear |
---|
112 | :param q_max: maximum q-value to smear |
---|
113 | |
---|
114 | """ |
---|
115 | # If this is the first time we call for smearing, |
---|
116 | # initialize the C++ smearer object first |
---|
117 | if not self._init_complete: |
---|
118 | self._initialize_smearer() |
---|
119 | if q_min == None: |
---|
120 | q_min = self.min |
---|
121 | if q_max == None: |
---|
122 | q_max = self.max |
---|
123 | _qmin_unsmeared, _qmax_unsmeared = self.get_unsmeared_range(q_min, |
---|
124 | q_max) |
---|
125 | _first_bin = None |
---|
126 | _last_bin = None |
---|
127 | |
---|
128 | step = (self.max - self.min) / (self.nbins - 1.0) |
---|
129 | try: |
---|
130 | for i in range(self.nbins): |
---|
131 | q_i = smearer.get_q(self._smearer, i) |
---|
132 | if (q_i >= _qmin_unsmeared) and (q_i <= _qmax_unsmeared): |
---|
133 | # Identify first and last bin |
---|
134 | if _first_bin is None: |
---|
135 | _first_bin = i |
---|
136 | else: |
---|
137 | _last_bin = i |
---|
138 | except: |
---|
139 | msg = "_BaseSmearer.get_bin_range: " |
---|
140 | msg += " error getting range\n %s" % sys.exc_value |
---|
141 | raise RuntimeError, msg |
---|
142 | |
---|
143 | return _first_bin, _last_bin |
---|
144 | |
---|
145 | def __call__(self, iq_in, first_bin=0, last_bin=None): |
---|
146 | """ |
---|
147 | Perform smearing |
---|
148 | """ |
---|
149 | # If this is the first time we call for smearing, |
---|
150 | # initialize the C++ smearer object first |
---|
151 | if not self._init_complete: |
---|
152 | self._initialize_smearer() |
---|
153 | |
---|
154 | # Get the max value for the last bin |
---|
155 | if last_bin is None or last_bin >= len(iq_in): |
---|
156 | last_bin = len(iq_in) - 1 |
---|
157 | # Check that the first bin is positive |
---|
158 | if first_bin < 0: |
---|
159 | first_bin = 0 |
---|
160 | |
---|
161 | # Sanity check |
---|
162 | if len(iq_in) != self.nbins: |
---|
163 | msg = "Invalid I(q) vector: inconsistent array " |
---|
164 | msg += " length %d != %s" % (len(iq_in), str(self.nbins)) |
---|
165 | raise RuntimeError, msg |
---|
166 | |
---|
167 | # Storage for smeared I(q) |
---|
168 | iq_out = numpy.zeros(self.nbins) |
---|
169 | smear_output = smearer.smear(self._smearer, iq_in, iq_out, |
---|
170 | first_bin, last_bin) |
---|
171 | if smear_output < 0: |
---|
172 | msg = "_BaseSmearer: could not smear, code = %g" % smear_output |
---|
173 | raise RuntimeError, msg |
---|
174 | return iq_out |
---|
175 | |
---|
176 | def _initialize_smearer(self): |
---|
177 | """ |
---|
178 | """ |
---|
179 | return NotImplemented |
---|
180 | |
---|
181 | class _SlitSmearer(_BaseSmearer): |
---|
182 | """ |
---|
183 | Slit smearing for I(q) array |
---|
184 | """ |
---|
185 | |
---|
186 | def __init__(self, nbins=None, width=None, height=None, min=None, max=None): |
---|
187 | """ |
---|
188 | Initialization |
---|
189 | |
---|
190 | :param iq: I(q) array [cm-1] |
---|
191 | :param width: slit width [A-1] |
---|
192 | :param height: slit height [A-1] |
---|
193 | :param min: Q_min [A-1] |
---|
194 | :param max: Q_max [A-1] |
---|
195 | |
---|
196 | """ |
---|
197 | _BaseSmearer.__init__(self) |
---|
198 | ## Slit width in Q units |
---|
199 | self.width = width |
---|
200 | ## Slit height in Q units |
---|
201 | self.height = height |
---|
202 | ## Q_min (Min Q-value for I(q)) |
---|
203 | self.min = min |
---|
204 | ## Q_max (Max Q_value for I(q)) |
---|
205 | self.max = max |
---|
206 | ## Number of Q bins |
---|
207 | self.nbins = nbins |
---|
208 | ## Number of points used in the smearing computation |
---|
209 | self.npts = 1000 |
---|
210 | ## Smearing matrix |
---|
211 | self._weights = None |
---|
212 | self.qvalues = None |
---|
213 | |
---|
214 | def _initialize_smearer(self): |
---|
215 | """ |
---|
216 | Initialize the C++ smearer object. |
---|
217 | This method HAS to be called before smearing |
---|
218 | """ |
---|
219 | #self._smearer = smearer.new_slit_smearer(self.width, |
---|
220 | # self.height, self.min, self.max, self.nbins) |
---|
221 | self._smearer = smearer.new_slit_smearer_with_q(self.width, |
---|
222 | self.height, self.qvalues) |
---|
223 | self._init_complete = True |
---|
224 | |
---|
225 | def get_unsmeared_range(self, q_min, q_max): |
---|
226 | """ |
---|
227 | Determine the range needed in unsmeared-Q to cover |
---|
228 | the smeared Q range |
---|
229 | """ |
---|
230 | # Range used for input to smearing |
---|
231 | _qmin_unsmeared = q_min |
---|
232 | _qmax_unsmeared = q_max |
---|
233 | try: |
---|
234 | _qmin_unsmeared = self.min |
---|
235 | _qmax_unsmeared = self.max |
---|
236 | except: |
---|
237 | logging.error("_SlitSmearer.get_bin_range: %s" % sys.exc_value) |
---|
238 | return _qmin_unsmeared, _qmax_unsmeared |
---|
239 | |
---|
240 | class SlitSmearer(_SlitSmearer): |
---|
241 | """ |
---|
242 | Adaptor for slit smearing class and SANS data |
---|
243 | """ |
---|
244 | def __init__(self, data1D): |
---|
245 | """ |
---|
246 | Assumption: equally spaced bins of increasing q-values. |
---|
247 | |
---|
248 | :param data1D: data used to set the smearing parameters |
---|
249 | """ |
---|
250 | # Initialization from parent class |
---|
251 | super(SlitSmearer, self).__init__() |
---|
252 | |
---|
253 | ## Slit width |
---|
254 | self.width = 0 |
---|
255 | if data1D.dxw is not None and len(data1D.dxw) == len(data1D.x): |
---|
256 | self.width = data1D.dxw[0] |
---|
257 | # Sanity check |
---|
258 | for value in data1D.dxw: |
---|
259 | if value != self.width: |
---|
260 | msg = "Slit smearing parameters must " |
---|
261 | msg += " be the same for all data" |
---|
262 | raise RuntimeError, msg |
---|
263 | ## Slit height |
---|
264 | self.height = 0 |
---|
265 | if data1D.dxl is not None and len(data1D.dxl) == len(data1D.x): |
---|
266 | self.height = data1D.dxl[0] |
---|
267 | # Sanity check |
---|
268 | for value in data1D.dxl: |
---|
269 | if value != self.height: |
---|
270 | msg = "Slit smearing parameters must be" |
---|
271 | msg += " the same for all data" |
---|
272 | raise RuntimeError, msg |
---|
273 | |
---|
274 | ## Number of Q bins |
---|
275 | self.nbins = len(data1D.x) |
---|
276 | ## Minimum Q |
---|
277 | self.min = min(data1D.x) |
---|
278 | ## Maximum |
---|
279 | self.max = max(data1D.x) |
---|
280 | ## Q-values |
---|
281 | self.qvalues = data1D.x |
---|
282 | |
---|
283 | |
---|
284 | class _QSmearer(_BaseSmearer): |
---|
285 | """ |
---|
286 | Perform Gaussian Q smearing |
---|
287 | """ |
---|
288 | |
---|
289 | def __init__(self, nbins=None, width=None, min=None, max=None): |
---|
290 | """ |
---|
291 | Initialization |
---|
292 | |
---|
293 | :param nbins: number of Q bins |
---|
294 | :param width: array standard deviation in Q [A-1] |
---|
295 | :param min: Q_min [A-1] |
---|
296 | :param max: Q_max [A-1] |
---|
297 | """ |
---|
298 | _BaseSmearer.__init__(self) |
---|
299 | ## Standard deviation in Q [A-1] |
---|
300 | self.width = width |
---|
301 | ## Q_min (Min Q-value for I(q)) |
---|
302 | self.min = min |
---|
303 | ## Q_max (Max Q_value for I(q)) |
---|
304 | self.max = max |
---|
305 | ## Number of Q bins |
---|
306 | self.nbins = nbins |
---|
307 | ## Smearing matrix |
---|
308 | self._weights = None |
---|
309 | self.qvalues = None |
---|
310 | |
---|
311 | def _initialize_smearer(self): |
---|
312 | """ |
---|
313 | Initialize the C++ smearer object. |
---|
314 | This method HAS to be called before smearing |
---|
315 | """ |
---|
316 | #self._smearer = smearer.new_q_smearer(numpy.asarray(self.width), |
---|
317 | # self.min, self.max, self.nbins) |
---|
318 | self._smearer = smearer.new_q_smearer_with_q(numpy.asarray(self.width), |
---|
319 | self.qvalues) |
---|
320 | self._init_complete = True |
---|
321 | |
---|
322 | def get_unsmeared_range(self, q_min, q_max): |
---|
323 | """ |
---|
324 | Determine the range needed in unsmeared-Q to cover |
---|
325 | the smeared Q range |
---|
326 | Take 3 sigmas as the offset between smeared and unsmeared space |
---|
327 | """ |
---|
328 | # Range used for input to smearing |
---|
329 | _qmin_unsmeared = q_min |
---|
330 | _qmax_unsmeared = q_max |
---|
331 | try: |
---|
332 | offset = 3.0 * max(self.width) |
---|
333 | _qmin_unsmeared = max([self.min, q_min - offset]) |
---|
334 | _qmax_unsmeared = min([self.max, q_max + offset]) |
---|
335 | except: |
---|
336 | logging.error("_QSmearer.get_bin_range: %s" % sys.exc_value) |
---|
337 | return _qmin_unsmeared, _qmax_unsmeared |
---|
338 | |
---|
339 | |
---|
340 | class QSmearer(_QSmearer): |
---|
341 | """ |
---|
342 | Adaptor for Gaussian Q smearing class and SANS data |
---|
343 | """ |
---|
344 | def __init__(self, data1D): |
---|
345 | """ |
---|
346 | Assumption: equally spaced bins of increasing q-values. |
---|
347 | |
---|
348 | :param data1D: data used to set the smearing parameters |
---|
349 | """ |
---|
350 | # Initialization from parent class |
---|
351 | super(QSmearer, self).__init__() |
---|
352 | |
---|
353 | ## Resolution |
---|
354 | self.width = numpy.zeros(len(data1D.x)) |
---|
355 | if data1D.dx is not None and len(data1D.dx) == len(data1D.x): |
---|
356 | self.width = data1D.dx |
---|
357 | |
---|
358 | ## Number of Q bins |
---|
359 | self.nbins = len(data1D.x) |
---|
360 | ## Minimum Q |
---|
361 | self.min = min(data1D.x) |
---|
362 | ## Maximum |
---|
363 | self.max = max(data1D.x) |
---|
364 | ## Q-values |
---|
365 | self.qvalues = data1D.x |
---|
366 | |
---|
367 | |
---|
368 | if __name__ == '__main__': |
---|
369 | x = 0.001 * numpy.arange(1, 11) |
---|
370 | y = 12.0 - numpy.arange(1, 11) |
---|
371 | print x |
---|
372 | #for i in range(10): print i, 0.001 + i*0.008/9.0 |
---|
373 | #for i in range(100): print i, int(math.floor( (i/ (100/9.0)) )) |
---|
374 | s = _SlitSmearer(nbins=10, width=0.0, height=0.005, min=0.001, max=0.010) |
---|
375 | #s = _QSmearer(nbins=10, width=0.001, min=0.001, max=0.010) |
---|
376 | s._compute_matrix() |
---|
377 | |
---|
378 | sy = s(y) |
---|
379 | print sy |
---|
380 | |
---|
381 | if True: |
---|
382 | for i in range(10): |
---|
383 | print x[i], y[i], sy[i] |
---|
384 | #print q, ' : ', s.weight(q), s._compute_iq(q) |
---|
385 | #print q, ' : ', s(q), s._compute_iq(q) |
---|
386 | #s._compute_iq(q) |
---|
387 | |
---|
388 | |
---|
389 | |
---|
390 | |
---|