1 | # This program is public domain |
---|
2 | """ |
---|
3 | An assembly is a collection of fitting functions. This provides |
---|
4 | the model representation that is the basis of the park fitting engine. |
---|
5 | |
---|
6 | Models can range from very simple one dimensional theory functions |
---|
7 | to complex assemblies of multidimensional datasets from different |
---|
8 | experimental techniques, each with their own theory function and |
---|
9 | a common underlying physical model. |
---|
10 | |
---|
11 | Usage |
---|
12 | ===== |
---|
13 | |
---|
14 | First define the models you want to work with. In the example |
---|
15 | below we will use an example of a simple multilayer system measured by |
---|
16 | specular reflection of xrays and neutrons. The gold depth is the only |
---|
17 | fitting parameter, ranging from 10-30 A. The interface depths are |
---|
18 | tied together using expressions. In this case the expression is |
---|
19 | a simple copy, but any standard math functions can be used. Some |
---|
20 | model developers may provide additional functions for use with the |
---|
21 | expression. |
---|
22 | |
---|
23 | Example models:: |
---|
24 | |
---|
25 | import reflectometry.model1d as refl |
---|
26 | xray = refl.model('xray') |
---|
27 | xray.incident('Air',rho=0) |
---|
28 | xray.interface('iAu',sigma=5) |
---|
29 | xray.layer('Au',rho=124.68,depth=[10,30]) |
---|
30 | xray.interface('iSi',sigma=5) |
---|
31 | xray.substrate('Si',rho=20.07) |
---|
32 | datax = refl.data('xray.dat') |
---|
33 | |
---|
34 | neutron = refl.model('neutron') |
---|
35 | neutron.incident('Air',rho=0) |
---|
36 | neutron.interface('iAu',sigma='xray.iAu') |
---|
37 | neutron.layer('Au',rho=4.66,depth='xray.Au.depth') |
---|
38 | neutron.interface('iSi',sigma='xray.iSi') |
---|
39 | neutron.substrate('Si',rho=2.07) |
---|
40 | datan = refl.data('neutron.dat') |
---|
41 | |
---|
42 | As you can see from the above, parameters can be set to a value if |
---|
43 | the parameter is fixed, to a range if the parametemr is fitted, or |
---|
44 | to a string expression if the parameter is calculated from other |
---|
45 | parameters. See park.Parameter.set for further details. |
---|
46 | |
---|
47 | Having constructed the models, we can now create an assembly:: |
---|
48 | |
---|
49 | import park |
---|
50 | assembly = park.Assembly([(xray,datax), (neutron,datan)]) |
---|
51 | |
---|
52 | Note: this would normally be done in the context of a fit |
---|
53 | using fit = park.Fit([(xray,datax), (neutron,datan)]), and later referenced |
---|
54 | using fit.assembly. |
---|
55 | |
---|
56 | Individual parts of the assembly are accessable using the |
---|
57 | model number 0, 1, 2... or by the model name. In the above, |
---|
58 | assembly[0] and assembly['xray'] refer to the same model. |
---|
59 | Assemblies have insert and append functions for adding new |
---|
60 | models, and "del model[idx]" for removing them. |
---|
61 | |
---|
62 | Once the assembly is created computing the values for the system |
---|
63 | is a matter of calling:: |
---|
64 | |
---|
65 | assembly.eval() |
---|
66 | print "Chi**2",assembly.chisq |
---|
67 | print "Reduced chi**2",assembly.chisq/assembly.degrees_of_freedom |
---|
68 | plot(arange(len(assembly.residuals)), assembly.residuals) |
---|
69 | |
---|
70 | This defines the attributes residuals, degrees_of_freedom and chisq, |
---|
71 | which is what the optimizer uses as the cost function to minimize. |
---|
72 | |
---|
73 | assembly.eval uses the current values for the parameters in the |
---|
74 | individual models. These parameters can be changed directly |
---|
75 | in the model. In the reflectometry example above, you could |
---|
76 | set the gold thickness using xray.layer.Au.depth=156, or |
---|
77 | something similar (the details are model specific). Parameters |
---|
78 | can also be changed through the assembly parameter set. In the same |
---|
79 | example, this would be assembly.parameterset['xray']['Au']['depth']. |
---|
80 | See parameter set for details. |
---|
81 | |
---|
82 | In the process of modeling data, particularly with multiple |
---|
83 | datasets, you will sometimes want to temporarily ignore |
---|
84 | how well one of the datasets matches so that you |
---|
85 | can more quickly refine the model for the other datasets, |
---|
86 | or see how particular models are influencing the fit. To |
---|
87 | temporarily ignore the xray data in the example above use:: |
---|
88 | |
---|
89 | assembly.parts[0].isfitted = False |
---|
90 | |
---|
91 | The model itself isn't ignored since its parameters may be |
---|
92 | needed to compute the parameters for other models. To |
---|
93 | reenable checking against the xray data, you would assign |
---|
94 | a True value instead. More subtle weighting of the models |
---|
95 | can be controlled using assembly.parts[idx].weight, but |
---|
96 | see below for a note on model weighting. |
---|
97 | |
---|
98 | A note on model weighting |
---|
99 | ------------------------- |
---|
100 | |
---|
101 | Changing the weight is equivalent to scaling the error bars |
---|
102 | on the given model by a factor of weight/n where n is the |
---|
103 | number of data points. It is better to set the correct error |
---|
104 | bars on your data in the first place than to adjust the weights. |
---|
105 | If you have the correct error bars, then you should expect |
---|
106 | roughly 2/3 of the data points to lie within one error bar of |
---|
107 | the theory curve. If consecutive data points have largely |
---|
108 | overlapping errorbars, then your uncertainty is overestimated. |
---|
109 | |
---|
110 | Another case where weights are adjusted (abused?) is to |
---|
111 | compensate for systematic errors in the data by forcing the |
---|
112 | errorbars to be large enough to cover the systematic bias. |
---|
113 | This is a poor approach to the problem. A better strategy |
---|
114 | is to capture the systematic effects in the model, and treat |
---|
115 | the measurement of the independent variable as an additional |
---|
116 | data point in the fit. This is still not statistically sound |
---|
117 | as there is likely to be a large correlation between the |
---|
118 | uncertainty of the measurement and the values of all the |
---|
119 | other variables. |
---|
120 | |
---|
121 | That said, adjusting the weight on a dataset is a quick way |
---|
122 | of reducing its influence on the entire fit. Please use it |
---|
123 | with care. |
---|
124 | """ |
---|
125 | |
---|
126 | __all__ = ['Assembly', 'Fitness'] |
---|
127 | import numpy |
---|
128 | |
---|
129 | import park |
---|
130 | from park.parameter import Parameter,ParameterSet |
---|
131 | from park.fitresult import FitParameter |
---|
132 | import park.expression |
---|
133 | |
---|
134 | |
---|
135 | |
---|
136 | class Fitness(object): |
---|
137 | """ |
---|
138 | Container for theory and data. |
---|
139 | |
---|
140 | The fit object compares theory with data. |
---|
141 | |
---|
142 | TODO: what to do with fittable metadata (e.g., footprint correction)? |
---|
143 | """ |
---|
144 | data = None |
---|
145 | model = None |
---|
146 | def __init__(self, model=None,data=None): |
---|
147 | self.data,self.model = data,model |
---|
148 | def _parameterset(self): |
---|
149 | return self.model.parameterset |
---|
150 | parameterset = property(_parameterset) |
---|
151 | def residuals(self): |
---|
152 | return self.data.residuals(self.model.eval) |
---|
153 | def residuals_deriv(self, pars=[]): |
---|
154 | return self.data.residuals_deriv(self.model.eval_derivs,pars=pars) |
---|
155 | def set(self, **kw): |
---|
156 | """ |
---|
157 | Set parameters in the model. |
---|
158 | |
---|
159 | User convenience function. This allows a user with an assembly |
---|
160 | of models in a script to for example set the fit range for |
---|
161 | parameter 'a' of the model:: |
---|
162 | assembly[0].set(a=[5,6]) |
---|
163 | |
---|
164 | Raises KeyError if the parameter is not in parameterset. |
---|
165 | """ |
---|
166 | self.model.set(**kw) |
---|
167 | def abort(self): |
---|
168 | if hasattr(self.model,'abort'): self.model.abort() |
---|
169 | |
---|
170 | class Part(object): |
---|
171 | """ |
---|
172 | Part of a fitting assembly. Part holds the model itself and |
---|
173 | associated data. The part can be initialized with a fitness |
---|
174 | object or with a pair (model,data) for the default fitness function. |
---|
175 | |
---|
176 | fitness (Fitness) |
---|
177 | object implementing the `park.assembly.Fitness` interface. In |
---|
178 | particular, fitness should provide a parameterset attribute |
---|
179 | containing a ParameterSet and a residuals method returning a vector |
---|
180 | of residuals. |
---|
181 | weight (dimensionless) |
---|
182 | weight for the model. See comments in assembly.py for details. |
---|
183 | isfitted (boolean) |
---|
184 | True if the model residuals should be included in the fit. |
---|
185 | The model parameters may still be used in parameter |
---|
186 | expressions, but there will be no comparison to the data. |
---|
187 | residuals (vector) |
---|
188 | Residuals for the model if they have been calculated, or None |
---|
189 | degrees_of_freedom |
---|
190 | Number of residuals minus number of fitted parameters. |
---|
191 | Degrees of freedom for individual models does not make |
---|
192 | sense in the presence of expressions combining models, |
---|
193 | particularly in the case where a model has many parameters |
---|
194 | but no data or many computed parameters. The degrees of |
---|
195 | freedom for the model is set to be at least one. |
---|
196 | chisq |
---|
197 | sum(residuals**2); use chisq/degrees_of_freedom to |
---|
198 | get the reduced chisq value. |
---|
199 | |
---|
200 | Get/set the weight on the given model. |
---|
201 | |
---|
202 | assembly.weight(3) returns the weight on model 3 (0-origin) |
---|
203 | assembly.weight(3,0.5) sets the weight on model 3 (0-origin) |
---|
204 | """ |
---|
205 | |
---|
206 | def __init__(self, fitness, weight=1., isfitted=True): |
---|
207 | if isinstance(fitness, tuple): |
---|
208 | fitness = park.Fitness(*fitness) |
---|
209 | self.fitness = fitness |
---|
210 | self.weight = weight |
---|
211 | self.isfitted = isfitted |
---|
212 | self.residuals = None |
---|
213 | self.chisq = numpy.Inf |
---|
214 | self.degrees_of_freedom = 1 |
---|
215 | |
---|
216 | |
---|
217 | class Assembly(object): |
---|
218 | """ |
---|
219 | Collection of fit models. |
---|
220 | |
---|
221 | Assembly implements the `park.fit.Objective` interface. |
---|
222 | |
---|
223 | See `park.assembly` for usage. |
---|
224 | |
---|
225 | Instance variables: |
---|
226 | |
---|
227 | residuals : array |
---|
228 | a vector of residuals spanning all models, with model |
---|
229 | weights applied as appropriate. |
---|
230 | degrees_of_freedom : integer |
---|
231 | length of the residuals - number of fitted parameters |
---|
232 | chisq : float |
---|
233 | sum squared residuals; this is not the reduced chisq, which |
---|
234 | you can get using chisq/degrees_of_freedom |
---|
235 | |
---|
236 | These fields are defined for the individual models as well, with |
---|
237 | degrees of freedom adjusted to the length of the individual data |
---|
238 | set. If the model is not fitted or the weight is zero, the residual |
---|
239 | will not be calculated. |
---|
240 | |
---|
241 | The residuals fields are available only after the model has been |
---|
242 | evaluated. |
---|
243 | """ |
---|
244 | |
---|
245 | def __init__(self, models=[]): |
---|
246 | """Build an assembly from a list of models.""" |
---|
247 | self.parts = [] |
---|
248 | for m in models: |
---|
249 | self.parts.append(Part(m)) |
---|
250 | self._reset() |
---|
251 | |
---|
252 | def __iter__(self): |
---|
253 | """Iterate through the models in order""" |
---|
254 | for m in self.parts: yield m |
---|
255 | |
---|
256 | def __getitem__(self, n): |
---|
257 | """Return the nth model""" |
---|
258 | return self.parts[n].fitness |
---|
259 | |
---|
260 | def __setitem__(self, n, fitness): |
---|
261 | """Replace the nth model""" |
---|
262 | self.parts[n].fitness = fitness |
---|
263 | self._reset() |
---|
264 | |
---|
265 | def __delitem__(self, n): |
---|
266 | """Delete the nth model""" |
---|
267 | del self.parts[n] |
---|
268 | self._reset() |
---|
269 | |
---|
270 | def weight(self, idx, value=None): |
---|
271 | """ |
---|
272 | Query the weight on a particular model. |
---|
273 | |
---|
274 | Set weight to value if value is supplied. |
---|
275 | |
---|
276 | :Parameters: |
---|
277 | idx : integer |
---|
278 | model number |
---|
279 | value : float |
---|
280 | model weight |
---|
281 | :return: model weight |
---|
282 | """ |
---|
283 | if value is not None: |
---|
284 | self.parts[idx].weight = value |
---|
285 | return self.parts[idx].weight |
---|
286 | |
---|
287 | def isfitted(self, idx, value=None): |
---|
288 | """ |
---|
289 | Query if a particular model is fitted. |
---|
290 | |
---|
291 | Set isfitted to value if value is supplied. |
---|
292 | |
---|
293 | :param idx: model number |
---|
294 | :type idx: integer |
---|
295 | :param value: |
---|
296 | """ |
---|
297 | if value is not None: |
---|
298 | self.parts[idx].isfitted = value |
---|
299 | return self.parts[idx].isfitted |
---|
300 | |
---|
301 | def append(self, fitness, weight=1.0, isfitted=True): |
---|
302 | """ |
---|
303 | Add a model to the end of set. |
---|
304 | |
---|
305 | :param fitness: the fitting model |
---|
306 | The fitting model can be an instance of `park.assembly.Fitness`, |
---|
307 | or a tuple of (`park.model.Model`,`park.data.Data1D`) |
---|
308 | :param weight: model weighting (usually 1.0) |
---|
309 | :param isfitted: whether model should be fit (equivalent to weight 0.) |
---|
310 | """ |
---|
311 | self.parts.append(Part(fitness,weight,isfitted)) |
---|
312 | self._reset() |
---|
313 | |
---|
314 | def insert(self, idx, fitness, weight=1.0, isfitted=True): |
---|
315 | """Add a model to a particular position in the set.""" |
---|
316 | self.parts.insert(idx,Part(fitness,weight,isfitted)) |
---|
317 | self._reset() |
---|
318 | |
---|
319 | def _reset(self): |
---|
320 | """Adjust the parameter set after the addition of a new model.""" |
---|
321 | subsets = [m.fitness.parameterset for m in self] |
---|
322 | self.parameterset = ParameterSet('root',subsets) |
---|
323 | self.parameterset.setprefix() |
---|
324 | #print [p.path for p in self.parameterset.flatten()] |
---|
325 | |
---|
326 | def eval(self): |
---|
327 | """ |
---|
328 | Recalculate the theory functions, and from them, the |
---|
329 | residuals and chisq. |
---|
330 | |
---|
331 | :note: Call this after the parameters have been updated. |
---|
332 | """ |
---|
333 | # Handle abort from a separate thread. |
---|
334 | self._cancel = False |
---|
335 | |
---|
336 | # Evaluate the computed parameters |
---|
337 | self._fitexpression() |
---|
338 | |
---|
339 | # Check that the resulting parameters are in a feasible region. |
---|
340 | if not self.isfeasible(): return numpy.inf |
---|
341 | |
---|
342 | resid = [] |
---|
343 | k = len(self._fitparameters) |
---|
344 | for m in self.parts: |
---|
345 | # In order to support abort, need to be able to propagate an |
---|
346 | # external abort signal from self.abort() into an abort signal |
---|
347 | # for the particular model. Can't see a way to do this which |
---|
348 | # doesn't involve setting a state variable. |
---|
349 | self._current_model = m |
---|
350 | if self._cancel: return numpy.inf |
---|
351 | if m.isfitted and m.weight != 0: |
---|
352 | m.residuals = m.fitness.residuals() |
---|
353 | N = len(m.residuals) |
---|
354 | m.degrees_of_freedom = N-k if N>k else 1 |
---|
355 | m.chisq = numpy.sum(m.residuals**2) |
---|
356 | resid.append(m.weight*m.residuals) |
---|
357 | self.residuals = numpy.hstack(resid) |
---|
358 | N = len(self.residuals) |
---|
359 | self.degrees_of_freedom = N-k if N>k else 1 |
---|
360 | self.chisq = numpy.sum(self.residuals**2) |
---|
361 | return self.chisq |
---|
362 | |
---|
363 | def jacobian(self, pvec, step=1e-8): |
---|
364 | """ |
---|
365 | Returns the derivative wrt the fit parameters at point p. |
---|
366 | |
---|
367 | Numeric derivatives are calculated based on step, where step is |
---|
368 | the portion of the total range for parameter j, or the portion of |
---|
369 | point value p_j if the range on parameter j is infinite. |
---|
370 | """ |
---|
371 | # Make sure the input vector is an array |
---|
372 | pvec = numpy.asarray(pvec) |
---|
373 | # We are being lazy here. We can precompute the bounds, we can |
---|
374 | # use the residuals_deriv from the sub-models which have analytic |
---|
375 | # derivatives and we need only recompute the models which depend |
---|
376 | # on the varying parameters. |
---|
377 | # Meanwhile, let's compute the numeric derivative using the |
---|
378 | # three point formula. |
---|
379 | # We are not checking that the varied parameter in numeric |
---|
380 | # differentiation is indeed feasible in the interval of interest. |
---|
381 | range = zip(*[p.range for p in self._fitparameters]) |
---|
382 | lo,hi = [numpy.asarray(v) for v in range] |
---|
383 | delta = (hi-lo)*step |
---|
384 | # For infinite ranges, use p*1e-8 for the step size |
---|
385 | idx = numpy.isinf(delta) |
---|
386 | #print "J",idx,delta,pvec,type(idx),type(delta),type(pvec) |
---|
387 | delta[idx] = pvec[idx]*step |
---|
388 | delta[delta==0] = step |
---|
389 | |
---|
390 | # Set the initial value |
---|
391 | for k,v in enumerate(pvec): |
---|
392 | self._fitparameters[k].value = v |
---|
393 | # Gather the residuals |
---|
394 | r = [] |
---|
395 | for k,v in enumerate(pvec): |
---|
396 | # Center point formula: |
---|
397 | # df/dv = lim_{h->0} ( f(v+h)-f(v-h) ) / ( 2h ) |
---|
398 | h = delta[k] |
---|
399 | self._fitparameters[k].value = v + h |
---|
400 | self.eval() |
---|
401 | rk = self.residuals |
---|
402 | self._fitparameters[k].value = v - h |
---|
403 | self.eval() |
---|
404 | rk -= self.residuals |
---|
405 | self._fitparameters[k].value = v |
---|
406 | r.append(rk/(2*h)) |
---|
407 | # return the jacobian |
---|
408 | return numpy.vstack(r).T |
---|
409 | |
---|
410 | |
---|
411 | def cov(self, pvec): |
---|
412 | """ |
---|
413 | Return the covariance matrix inv(J'J) at point p. |
---|
414 | """ |
---|
415 | |
---|
416 | # Find cov of f at p |
---|
417 | # cov(f,p) = inv(J'J) |
---|
418 | # Use SVD |
---|
419 | # J = U S V' |
---|
420 | # J'J = (U S V')' (U S V') |
---|
421 | # = V S' U' U S V' |
---|
422 | # = V S S V' |
---|
423 | # inv(J'J) = inv(V S S V') |
---|
424 | # = inv(V') inv(S S) inv(V) |
---|
425 | # = V inv (S S) V' |
---|
426 | J = self.jacobian(pvec) |
---|
427 | u,s,vh = numpy.linalg.svd(J,0) |
---|
428 | JTJinv = numpy.dot(vh.T.conj()/s**2,vh) |
---|
429 | return JTJinv |
---|
430 | |
---|
431 | def stderr(self, pvec): |
---|
432 | """ |
---|
433 | Return parameter uncertainty. |
---|
434 | |
---|
435 | This is just the sqrt diagonal of covariance matrix inv(J'J) at point p. |
---|
436 | """ |
---|
437 | return numpy.sqrt(numpy.diag(self.cov(pvec))) |
---|
438 | |
---|
439 | def isfeasible(self): |
---|
440 | """ |
---|
441 | Returns true if the parameter set is in a feasible region of the |
---|
442 | modeling space. |
---|
443 | """ |
---|
444 | return True |
---|
445 | |
---|
446 | # Fitting service interface |
---|
447 | def fit_parameters(self): |
---|
448 | """ |
---|
449 | Return an alphabetical list of the fitting parameters. |
---|
450 | |
---|
451 | This function is called once at the beginning of a fit, |
---|
452 | and serves as a convenient place to precalculate what |
---|
453 | can be precalculated such as the set of fitting parameters |
---|
454 | and the parameter expressions evaluator. |
---|
455 | """ |
---|
456 | self.parameterset.setprefix() |
---|
457 | self._fitparameters = self.parameterset.fitted |
---|
458 | self._restraints = self.parameterset.restrained |
---|
459 | pars = self.parameterset.flatten() |
---|
460 | context = self.parameterset.gather_context() |
---|
461 | self._fitexpression = park.expression.build_eval(pars,context) |
---|
462 | #print "constraints",self._fitexpression.__doc__ |
---|
463 | |
---|
464 | self._fitparameters.sort(lambda a,b: cmp(a.path,b.path)) |
---|
465 | # Convert to fitparameter a object |
---|
466 | fitpars = [FitParameter(p.path,p.range,p.value) |
---|
467 | for p in self._fitparameters] |
---|
468 | return fitpars |
---|
469 | |
---|
470 | def set_result(self, result): |
---|
471 | """ |
---|
472 | Set the parameters resulting from the fit into the parameter set, |
---|
473 | and update the calculated expression. |
---|
474 | |
---|
475 | The parameter values may be retrieved by walking the assembly.parameterset |
---|
476 | tree, checking each parameter for isfitted, iscomputed, or isfixed. |
---|
477 | For example:: |
---|
478 | |
---|
479 | assembly.set_result(result) |
---|
480 | for p in assembly.parameterset.flatten(): |
---|
481 | if p.isfitted(): |
---|
482 | print "%s %g in [%g,%g]"%(p.path,p.value,p.range[0],p.range[1]) |
---|
483 | elif p.iscomputed(): |
---|
484 | print "%s computed as %g"%(p.path.p.value) |
---|
485 | |
---|
486 | This does not calculate the function or the residuals for these parameters. |
---|
487 | You can call assembly.eval() to do this. The residuals will be set in |
---|
488 | assembly[i].residuals. The theory and data are model specific, and can |
---|
489 | be found in assembly[i].fitness.data. |
---|
490 | """ |
---|
491 | for n,p in enumerate(result.parameters): |
---|
492 | self._fitparameters[n] = p.value |
---|
493 | self._fitexpression() |
---|
494 | |
---|
495 | def all_results(self, result): |
---|
496 | """ |
---|
497 | Extend result from the fit with the calculated parameters. |
---|
498 | """ |
---|
499 | calcpars = [FitParameter(p.path,p.range,p.value) |
---|
500 | for p in self.parameterset.computed] |
---|
501 | result.parameters += calcpars |
---|
502 | |
---|
503 | def result(self, status='step'): |
---|
504 | """ |
---|
505 | Details to send back to the fitting client on an improved fit. |
---|
506 | |
---|
507 | status is 'start', 'step' or 'end' depending if this is the |
---|
508 | first result to return, an improved result, or the final result. |
---|
509 | |
---|
510 | [Not implemented] |
---|
511 | """ |
---|
512 | return None |
---|
513 | |
---|
514 | def fresiduals(self, pvec): |
---|
515 | chisq = self.__call__(pvec) |
---|
516 | return self.residuals |
---|
517 | |
---|
518 | def __call__(self, pvec): |
---|
519 | """ |
---|
520 | Cost function. |
---|
521 | |
---|
522 | Evaluate the system for the parameter vector pvec, returning chisq |
---|
523 | as the cost function to be minimized. |
---|
524 | |
---|
525 | Raises a runtime error if the number of fit parameters is |
---|
526 | different than the length of the vector. |
---|
527 | """ |
---|
528 | # Plug fit parameters into model |
---|
529 | #print "Trying",pvec |
---|
530 | pars = self._fitparameters |
---|
531 | if len(pvec) != len(pars): |
---|
532 | raise RuntimeError("Unexpected number of parameters") |
---|
533 | for n,value in enumerate(pvec): |
---|
534 | pars[n].value = value |
---|
535 | # Evaluate model |
---|
536 | chisq = self.eval() |
---|
537 | # Evaluate additional restraints based on parameter value |
---|
538 | # likelihood |
---|
539 | restraints_penalty = 0 |
---|
540 | for p in self._restraints: |
---|
541 | restraints_penalty += p.likelihood(p.value) |
---|
542 | # Return total cost function |
---|
543 | return self.chisq + restraints_penalty |
---|
544 | |
---|
545 | def abort(self): |
---|
546 | """ |
---|
547 | Interrupt the current function evaluation. |
---|
548 | |
---|
549 | Forward this to the currently executing model if possible. |
---|
550 | """ |
---|
551 | self._cancel = True |
---|
552 | if hasattr(self._current_model,'abort'): |
---|
553 | self._current_model.abort() |
---|
554 | |
---|
555 | class _Exp(park.Model): |
---|
556 | """ |
---|
557 | Sample model for testing assembly. |
---|
558 | """ |
---|
559 | parameters = ['a','c'] |
---|
560 | def eval(self,x): |
---|
561 | return self.a*numpy.exp(self.c*x) |
---|
562 | class _Linear(park.Model): |
---|
563 | parameters = ['a','c'] |
---|
564 | def eval(self,x): |
---|
565 | #print "eval",self.a,self.c,x,self.a*x+self.c |
---|
566 | return self.a*x+self.c |
---|
567 | def example(): |
---|
568 | """ |
---|
569 | Return an example assembly consisting of a pair of functions, |
---|
570 | M1.a*exp(M1.c*x), M2.a*exp(2*M1.c*x) |
---|
571 | and ideal data for |
---|
572 | M1.a=1, M1.c=1.5, M2.a=2.5 |
---|
573 | """ |
---|
574 | import numpy |
---|
575 | import park |
---|
576 | from numpy import inf |
---|
577 | # Make some fake data |
---|
578 | x1 = numpy.linspace(0,1,11) |
---|
579 | x2 = numpy.linspace(0,1,12) |
---|
580 | # Define a shared model |
---|
581 | if True: # Exp model |
---|
582 | y1,y2 = numpy.exp(1.5*x1),2.5*numpy.exp(3*x2) |
---|
583 | M1 = _Exp('M1',a=[1,3],c=[1,3]) |
---|
584 | M2 = _Exp('M2',a=[1,3],c='2*M1.c') |
---|
585 | #M2 = _Exp('M2',a=[1,3],c=3) |
---|
586 | else: # Linear model |
---|
587 | y1,y2 = x1+1.5, 2.5*x2+3 |
---|
588 | M1 = _Linear('M1',a=[1,3],c=[1,3]) |
---|
589 | M2 = _Linear('M2',a=[1,3],c='2*M1.c') |
---|
590 | if False: # Unbounded |
---|
591 | M1.a = [-inf,inf] |
---|
592 | M1.c = [-inf,inf] |
---|
593 | M2.a = [-inf,inf] |
---|
594 | D1 = park.Data1D(x=x1, y=y1) |
---|
595 | D2 = park.Data1D(x=x2, y=y2) |
---|
596 | # Construct the assembly |
---|
597 | assembly = park.Assembly([(M1,D1),(M2,D2)]) |
---|
598 | return assembly |
---|
599 | |
---|
600 | class _Sphere(park.Model): |
---|
601 | parameters = ['a','b','c','d','e'] |
---|
602 | def eval(self,x): |
---|
603 | return self.a*x**2+self.b*x+self.c + exp(self.d) - 3*sin(self.e) |
---|
604 | |
---|
605 | def example5(): |
---|
606 | import numpy |
---|
607 | import park |
---|
608 | from numpy import inf |
---|
609 | # Make some fake data |
---|
610 | x = numpy.linspace(0,1,11) |
---|
611 | # Define a shared model |
---|
612 | S = _Sphere(a=1,b=2,c=3,d=4,e=5) |
---|
613 | y = S.eval(x1) |
---|
614 | Sfit = _Sphere(a=[-inf,inf],b=[-inf,inf],c=[-inf,inf],d=[-inf,inf],e=[-inf,inf]) |
---|
615 | D = park.Data1D(x=x, y=y) |
---|
616 | # Construct the assembly |
---|
617 | assembly = park.Assembly([(Sfit,D)]) |
---|
618 | return assembly |
---|
619 | |
---|
620 | def test(): |
---|
621 | assembly = example() |
---|
622 | assert assembly[0].parameterset.name == 'M1' |
---|
623 | |
---|
624 | # extract the fitting parameters |
---|
625 | pars = [p.name for p in assembly.fit_parameters()] |
---|
626 | assert set(pars) == set(['M1.a','M1.c','M2.a']) |
---|
627 | # Compute chisq and verify constraints are updated properly |
---|
628 | assert assembly([1,1.5,2.5]) == 0 |
---|
629 | assert assembly[0].model.c == 1.5 and assembly[1].model.c == 3 |
---|
630 | |
---|
631 | # Try without constraints |
---|
632 | assembly[1].set(c=3) |
---|
633 | assembly.fit_parameters() # Fit parameters have changed |
---|
634 | assert assembly([1,1.5,2.5]) == 0 |
---|
635 | |
---|
636 | # Check that assembly.cov runs ... still need to check that it is correct! |
---|
637 | C = assembly.cov(numpy.array([1,1.5,2.5])) |
---|
638 | |
---|
639 | if __name__ == "__main__": test() |
---|