1 # -*- coding: utf-8 -*-
3 # Copyright (C) 2008-2023 EDF R&D
5 # This library is free software; you can redistribute it and/or
6 # modify it under the terms of the GNU Lesser General Public
7 # License as published by the Free Software Foundation; either
8 # version 2.1 of the License.
10 # This library is distributed in the hope that it will be useful,
11 # but WITHOUT ANY WARRANTY; without even the implied warranty of
12 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
13 # Lesser General Public License for more details.
15 # You should have received a copy of the GNU Lesser General Public
16 # License along with this library; if not, write to the Free Software
17 # Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
19 # See http://www.salome-platform.org/ or email : webmaster.salome@opencascade.com
21 # Author: Jean-Philippe Argaud, jean-philippe.argaud@edf.fr, EDF R&D
24 from daCore import BasicObjects, NumericObjects, PlatformInfo
25 mpr = PlatformInfo.PlatformInfo().MachinePrecision()
26 mfp = PlatformInfo.PlatformInfo().MaximumPrecision()
28 # ==============================================================================
29 class ElementaryAlgorithm(BasicObjects.Algorithm):
31 BasicObjects.Algorithm.__init__(self, "LINEARITYTEST")
32 self.defineRequiredParameter(
33 name = "ResiduFormula",
34 default = "CenteredDL",
36 message = "Formule de résidu utilisée",
37 listval = ["CenteredDL", "Taylor", "NominalTaylor", "NominalTaylorRMS"],
39 self.defineRequiredParameter(
40 name = "EpsilonMinimumExponent",
43 message = "Exposant minimal en puissance de 10 pour le multiplicateur d'incrément",
47 self.defineRequiredParameter(
48 name = "InitialDirection",
51 message = "Direction initiale de la dérivée directionnelle autour du point nominal",
53 self.defineRequiredParameter(
54 name = "AmplitudeOfInitialDirection",
57 message = "Amplitude de la direction initiale de la dérivée directionnelle autour du point nominal",
59 self.defineRequiredParameter(
60 name = "AmplitudeOfTangentPerturbation",
63 message = "Amplitude de la perturbation pour le calcul de la forme tangente",
67 self.defineRequiredParameter(
69 typecast = numpy.random.seed,
70 message = "Graine fixée pour le générateur aléatoire",
72 self.defineRequiredParameter(
73 name = "NumberOfPrintedDigits",
76 message = "Nombre de chiffres affichés pour les impressions de réels",
79 self.defineRequiredParameter(
83 message = "Titre du tableau et de la figure",
85 self.defineRequiredParameter(
86 name = "StoreSupplementaryCalculations",
89 message = "Liste de calculs supplémentaires à stocker et/ou effectuer",
93 "SimulatedObservationAtCurrentState",
96 self.requireInputArguments(
97 mandatory= ("Xb", "HO"),
99 self.setAttributes(tags=(
103 def run(self, Xb=None, Y=None, U=None, HO=None, EM=None, CM=None, R=None, B=None, Q=None, Parameters=None):
104 self._pre_run(Parameters, Xb, Y, U, HO, EM, CM, R, B, Q)
108 return math.sqrt( ((numpy.ravel(V2) - numpy.ravel(V1))**2).sum() / float(numpy.ravel(V1).size) )
110 Hm = HO["Direct"].appliedTo
111 if self._parameters["ResiduFormula"] in ["Taylor", "NominalTaylor", "NominalTaylorRMS"]:
112 Ht = HO["Tangent"].appliedInXTo
114 X0 = numpy.ravel( Xb ).reshape((-1,1))
117 __p = self._parameters["NumberOfPrintedDigits"]
122 if len(self._parameters["ResultTitle"]) > 0:
123 __rt = str(self._parameters["ResultTitle"])
124 msgs += (__marge + "====" + "="*len(__rt) + "====\n")
125 msgs += (__marge + " " + __rt + "\n")
126 msgs += (__marge + "====" + "="*len(__rt) + "====\n")
128 msgs += (__marge + "%s\n"%self._name)
129 msgs += (__marge + "%s\n"%("="*len(self._name),))
132 msgs += (__marge + "This test allows to analyze the linearity property of some given\n")
133 msgs += (__marge + "simulation operator F, applied to one single vector argument x.\n")
134 msgs += (__marge + "The output shows simple statistics related to its stability for various\n")
135 msgs += (__marge + "increments, around an input checking point X.\n")
137 msgs += (__flech + "Information before launching:\n")
138 msgs += (__marge + "-----------------------------\n")
140 msgs += (__marge + "Characteristics of input vector X, internally converted:\n")
141 msgs += (__marge + " Type...............: %s\n")%type( X0 )
142 msgs += (__marge + " Length of vector...: %i\n")%max(numpy.ravel( X0 ).shape)
143 msgs += (__marge + " Minimum value......: %."+str(__p)+"e\n")%numpy.min( X0 )
144 msgs += (__marge + " Maximum value......: %."+str(__p)+"e\n")%numpy.max( X0 )
145 msgs += (__marge + " Mean of vector.....: %."+str(__p)+"e\n")%numpy.mean( X0, dtype=mfp )
146 msgs += (__marge + " Standard error.....: %."+str(__p)+"e\n")%numpy.std( X0, dtype=mfp )
147 msgs += (__marge + " L2 norm of vector..: %."+str(__p)+"e\n")%numpy.linalg.norm( X0 )
149 msgs += (__marge + "%s\n\n"%("-"*75,))
150 msgs += (__flech + "Numerical quality indicators:\n")
151 msgs += (__marge + "-----------------------------\n")
153 msgs += (__marge + "Using the \"%s\" formula, one observes the residue R which is the\n"%self._parameters["ResiduFormula"])
154 msgs += (__marge + "following ratio or comparison:\n")
157 if self._parameters["ResiduFormula"] == "CenteredDL":
158 msgs += (__marge + " || F(X+Alpha*dX) + F(X-Alpha*dX) - 2*F(X) ||\n")
159 msgs += (__marge + " R(Alpha) = --------------------------------------------\n")
160 msgs += (__marge + " || F(X) ||\n")
162 msgs += (__marge + "If the residue remains always very small compared to 1, the linearity\n")
163 msgs += (__marge + "assumption of F is verified.\n")
165 msgs += (__marge + "If the residue varies a lot, or is of the order of 1 or more, and is\n")
166 msgs += (__marge + "small only under a certain order of Alpha increment, the linearity\n")
167 msgs += (__marge + "assumption of F is not verified.\n")
169 msgs += (__marge + "If the residue decreases and if the decay is in Alpha**2 according to\n")
170 msgs += (__marge + "Alpha, it means that the gradient is well calculated up to the stopping\n")
171 msgs += (__marge + "precision of the quadratic decay.\n")
173 __entete = u" i Alpha ||X|| ||F(X)|| | R(Alpha) log10( R )"
175 if self._parameters["ResiduFormula"] == "Taylor":
176 msgs += (__marge + " || F(X+Alpha*dX) - F(X) - Alpha * GradientF_X(dX) ||\n")
177 msgs += (__marge + " R(Alpha) = ----------------------------------------------------\n")
178 msgs += (__marge + " || F(X) ||\n")
180 msgs += (__marge + "If the residue remains always very small compared to 1, the linearity\n")
181 msgs += (__marge + "assumption of F is verified.\n")
183 msgs += (__marge + "If the residue varies a lot, or is of the order of 1 or more, and is\n")
184 msgs += (__marge + "small only under a certain order of Alpha increment, the linearity\n")
185 msgs += (__marge + "assumption of F is not verified.\n")
187 msgs += (__marge + "If the residue decreases and if the decay is in Alpha**2 according to\n")
188 msgs += (__marge + "Alpha, it means that the gradient is well calculated up to the stopping\n")
189 msgs += (__marge + "precision of the quadratic decay.\n")
191 __entete = u" i Alpha ||X|| ||F(X)|| | R(Alpha) log10( R )"
193 if self._parameters["ResiduFormula"] == "NominalTaylor":
194 msgs += (__marge + " R(Alpha) = max(\n")
195 msgs += (__marge + " || F(X+Alpha*dX) - Alpha * F(dX) || / || F(X) ||,\n")
196 msgs += (__marge + " || F(X-Alpha*dX) + Alpha * F(dX) || / || F(X) ||,\n")
197 msgs += (__marge + " )\n")
199 msgs += (__marge + "If the residue remains always equal to 1 within 2 or 3 percent (i.e.\n")
200 msgs += (__marge + "|R-1| remains equal to 2 or 3 percent), then the linearity assumption of\n")
201 msgs += (__marge + "F is verified.\n")
203 msgs += (__marge + "If the residue is equal to 1 over only a part of the range of variation\n")
204 msgs += (__marge + "of the Alpha increment, it is over this part that the linearity assumption\n")
205 msgs += (__marge + "of F is verified.\n")
207 __entete = u" i Alpha ||X|| ||F(X)|| | R(Alpha) |R-1| in %"
209 if self._parameters["ResiduFormula"] == "NominalTaylorRMS":
210 msgs += (__marge + " R(Alpha) = max(\n")
211 msgs += (__marge + " RMS( F(X), F(X+Alpha*dX) - Alpha * F(dX) ) / || F(X) ||,\n")
212 msgs += (__marge + " RMS( F(X), F(X-Alpha*dX) + Alpha * F(dX) ) / || F(X) ||,\n")
213 msgs += (__marge + " )\n")
215 msgs += (__marge + "If the residue remains always equal to 0 within 1 or 2 percent then the\n")
216 msgs += (__marge + "linearity assumption of F is verified.\n")
218 msgs += (__marge + "If the residue is equal to 0 over only a part of the range of variation\n")
219 msgs += (__marge + "of the Alpha increment, it is over this part that the linearity assumption\n")
220 msgs += (__marge + "of F is verified.\n")
222 __entete = u" i Alpha ||X|| ||F(X)|| | R(Alpha) |R| in %"
225 msgs += (__marge + "We take dX0 = Normal(0,X) and dX = Alpha*dX0. F is the calculation code.\n")
227 msgs += (__marge + "(Remark: numbers that are (about) under %.0e represent 0 to machine precision)\n"%mpr)
230 Perturbations = [ 10**i for i in range(self._parameters["EpsilonMinimumExponent"],1) ]
231 Perturbations.reverse()
233 FX = numpy.ravel( Hm( X0 ) ).reshape((-1,1))
234 NormeX = numpy.linalg.norm( X0 )
235 NormeFX = numpy.linalg.norm( FX )
236 if NormeFX < mpr: NormeFX = mpr
237 if self._toStore("CurrentState"):
238 self.StoredVariables["CurrentState"].store( X0 )
239 if self._toStore("SimulatedObservationAtCurrentState"):
240 self.StoredVariables["SimulatedObservationAtCurrentState"].store( FX )
242 dX0 = NumericObjects.SetInitialDirection(
243 self._parameters["InitialDirection"],
244 self._parameters["AmplitudeOfInitialDirection"],
248 if self._parameters["ResiduFormula"] in ["Taylor", "NominalTaylor", "NominalTaylorRMS"]:
249 dX1 = float(self._parameters["AmplitudeOfTangentPerturbation"]) * dX0
250 GradFxdX = Ht( (X0, dX1) )
251 GradFxdX = numpy.ravel( GradFxdX ).reshape((-1,1))
252 GradFxdX = float(1./self._parameters["AmplitudeOfTangentPerturbation"]) * GradFxdX
254 # Boucle sur les perturbations
255 # ----------------------------
256 __nbtirets = len(__entete) + 2
258 msgs += "\n" + __marge + "-"*__nbtirets
259 msgs += "\n" + __marge + __entete
260 msgs += "\n" + __marge + "-"*__nbtirets
263 for i,amplitude in enumerate(Perturbations):
264 dX = amplitude * dX0.reshape((-1,1))
266 if self._parameters["ResiduFormula"] == "CenteredDL":
267 if self._toStore("CurrentState"):
268 self.StoredVariables["CurrentState"].store( X0 + dX )
269 self.StoredVariables["CurrentState"].store( X0 - dX )
271 FX_plus_dX = numpy.ravel( Hm( X0 + dX ) ).reshape((-1,1))
272 FX_moins_dX = numpy.ravel( Hm( X0 - dX ) ).reshape((-1,1))
274 if self._toStore("SimulatedObservationAtCurrentState"):
275 self.StoredVariables["SimulatedObservationAtCurrentState"].store( FX_plus_dX )
276 self.StoredVariables["SimulatedObservationAtCurrentState"].store( FX_moins_dX )
278 Residu = numpy.linalg.norm( FX_plus_dX + FX_moins_dX - 2 * FX ) / NormeFX
280 self.StoredVariables["Residu"].store( Residu )
281 ttsep = " %2i %5.0e %9.3e %9.3e | %9.3e %4.0f\n"%(i,amplitude,NormeX,NormeFX,Residu,math.log10(max(1.e-99,Residu)))
282 msgs += __marge + ttsep
284 if self._parameters["ResiduFormula"] == "Taylor":
285 if self._toStore("CurrentState"):
286 self.StoredVariables["CurrentState"].store( X0 + dX )
288 FX_plus_dX = numpy.ravel( Hm( X0 + dX ) ).reshape((-1,1))
290 if self._toStore("SimulatedObservationAtCurrentState"):
291 self.StoredVariables["SimulatedObservationAtCurrentState"].store( FX_plus_dX )
293 Residu = numpy.linalg.norm( FX_plus_dX - FX - amplitude * GradFxdX ) / NormeFX
295 self.StoredVariables["Residu"].store( Residu )
296 ttsep = " %2i %5.0e %9.3e %9.3e | %9.3e %4.0f\n"%(i,amplitude,NormeX,NormeFX,Residu,math.log10(max(1.e-99,Residu)))
297 msgs += __marge + ttsep
299 if self._parameters["ResiduFormula"] == "NominalTaylor":
300 if self._toStore("CurrentState"):
301 self.StoredVariables["CurrentState"].store( X0 + dX )
302 self.StoredVariables["CurrentState"].store( X0 - dX )
303 self.StoredVariables["CurrentState"].store( dX )
305 FX_plus_dX = numpy.ravel( Hm( X0 + dX ) ).reshape((-1,1))
306 FX_moins_dX = numpy.ravel( Hm( X0 - dX ) ).reshape((-1,1))
307 FdX = numpy.ravel( Hm( dX ) ).reshape((-1,1))
309 if self._toStore("SimulatedObservationAtCurrentState"):
310 self.StoredVariables["SimulatedObservationAtCurrentState"].store( FX_plus_dX )
311 self.StoredVariables["SimulatedObservationAtCurrentState"].store( FX_moins_dX )
312 self.StoredVariables["SimulatedObservationAtCurrentState"].store( FdX )
315 numpy.linalg.norm( FX_plus_dX - amplitude * FdX ) / NormeFX,
316 numpy.linalg.norm( FX_moins_dX + amplitude * FdX ) / NormeFX,
319 self.StoredVariables["Residu"].store( Residu )
320 ttsep = " %2i %5.0e %9.3e %9.3e | %9.3e %5i %s\n"%(i,amplitude,NormeX,NormeFX,Residu,100.*abs(Residu-1.),"%")
321 msgs += __marge + ttsep
323 if self._parameters["ResiduFormula"] == "NominalTaylorRMS":
324 if self._toStore("CurrentState"):
325 self.StoredVariables["CurrentState"].store( X0 + dX )
326 self.StoredVariables["CurrentState"].store( X0 - dX )
327 self.StoredVariables["CurrentState"].store( dX )
329 FX_plus_dX = numpy.ravel( Hm( X0 + dX ) ).reshape((-1,1))
330 FX_moins_dX = numpy.ravel( Hm( X0 - dX ) ).reshape((-1,1))
331 FdX = numpy.ravel( Hm( dX ) ).reshape((-1,1))
333 if self._toStore("SimulatedObservationAtCurrentState"):
334 self.StoredVariables["SimulatedObservationAtCurrentState"].store( FX_plus_dX )
335 self.StoredVariables["SimulatedObservationAtCurrentState"].store( FX_moins_dX )
336 self.StoredVariables["SimulatedObservationAtCurrentState"].store( FdX )
339 RMS( FX, FX_plus_dX - amplitude * FdX ) / NormeFX,
340 RMS( FX, FX_moins_dX + amplitude * FdX ) / NormeFX,
343 self.StoredVariables["Residu"].store( Residu )
344 ttsep = " %2i %5.0e %9.3e %9.3e | %9.3e %5i %s\n"%(i,amplitude,NormeX,NormeFX,Residu,100.*Residu,"%")
345 msgs += __marge + ttsep
347 msgs += (__marge + "-"*__nbtirets + "\n\n")
348 msgs += (__marge + "End of the \"%s\" verification by the \"%s\" formula.\n\n"%(self._name,self._parameters["ResiduFormula"]))
349 msgs += (__marge + "%s\n"%("-"*75,))
355 # ==============================================================================
356 if __name__ == "__main__":
357 print('\n AUTODIAGNOSTIC\n')