1 #-*-coding:iso-8859-1-*-
3 # Copyright (C) 2008-2013 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, PlatformInfo
25 m = PlatformInfo.SystemUsage()
29 # ==============================================================================
30 class ElementaryAlgorithm(BasicObjects.Algorithm):
32 BasicObjects.Algorithm.__init__(self, "BLUE")
33 self.defineRequiredParameter(
34 name = "StoreInternalVariables",
37 message = "Stockage des variables internes ou intermédiaires du calcul",
39 self.defineRequiredParameter(
40 name = "StoreSupplementaryCalculations",
43 message = "Liste de calculs supplémentaires à stocker et/ou effectuer",
44 listval = ["APosterioriCovariance", "BMA", "OMA", "OMB", "Innovation", "SigmaBck2", "SigmaObs2", "MahalanobisConsistency", "SimulationQuantiles"]
46 self.defineRequiredParameter(
50 message = "Liste des valeurs de quantiles",
52 self.defineRequiredParameter(
54 typecast = numpy.random.seed,
55 message = "Graine fixée pour le générateur aléatoire",
57 self.defineRequiredParameter(
58 name = "NumberOfSamplesForQuantiles",
61 message = "Nombre d'échantillons simulés pour le calcul des quantiles",
64 self.defineRequiredParameter(
65 name = "SimulationForQuantiles",
68 message = "Type de simulation pour l'estimation des quantiles",
69 listval = ["Linear", "NonLinear"]
72 def run(self, Xb=None, Y=None, U=None, HO=None, EM=None, CM=None, R=None, B=None, Q=None, Parameters=None):
73 logging.debug("%s Lancement"%self._name)
74 logging.debug("%s Taille mémoire utilisée de %.1f Mo"%(self._name, m.getUsedMemory("M")))
76 # Paramètres de pilotage
77 # ----------------------
78 self.setParameters(Parameters)
80 # Opérateur d'observation
81 # -----------------------
82 Hm = HO["Tangent"].asMatrix(Xb)
83 Hm = Hm.reshape(Y.size,Xb.size) # ADAO & check shape
84 Ha = HO["Adjoint"].asMatrix(Xb)
85 Ha = Ha.reshape(Xb.size,Y.size) # ADAO & check shape
87 # Utilisation éventuelle d'un vecteur H(Xb) précalculé
88 # ----------------------------------------------------
89 if HO["AppliedToX"] is not None and HO["AppliedToX"].has_key("HXb"):
90 HXb = HO["AppliedToX"]["HXb"]
93 HXb = numpy.asmatrix(numpy.ravel( HXb )).T
95 # Précalcul des inversions de B et R
96 # ----------------------------------
100 # Calcul de l'innovation
101 # ----------------------
102 if Y.size != HXb.size:
103 raise ValueError("The size %i of observations Y and %i of observed calculation H(X) are different, they have to be identical."%(Y.size,HXb.size))
104 if max(Y.shape) != max(HXb.shape):
105 raise ValueError("The shapes %s of observations Y and %s of observed calculation H(X) are different, they have to be identical."%(Y.shape,HXb.shape))
108 # Calcul de la matrice de gain et de l'analyse
109 # --------------------------------------------
110 if Y.size <= Xb.size:
111 if Y.size > 100: # len(R)
113 _u = numpy.linalg.solve( _A , d )
114 Xa = Xb + B * Ha * _u
116 K = B * Ha * (R + Hm * B * Ha).I
119 if Y.size > 100: # len(R)
120 _A = BI + Ha * RI * Hm
121 _u = numpy.linalg.solve( _A , Ha * RI * d )
124 K = (BI + Ha * RI * Hm).I * Ha * RI
126 self.StoredVariables["Analysis"].store( Xa.A1 )
128 # Calcul de la fonction coût
129 # --------------------------
130 if self._parameters["StoreInternalVariables"] or \
131 "OMA" in self._parameters["StoreSupplementaryCalculations"] or \
132 "SigmaObs2" in self._parameters["StoreSupplementaryCalculations"] or \
133 "MahalanobisConsistency" in self._parameters["StoreSupplementaryCalculations"] or \
134 "SimulationQuantiles" in self._parameters["StoreSupplementaryCalculations"]:
137 if self._parameters["StoreInternalVariables"] or \
138 "MahalanobisConsistency" in self._parameters["StoreSupplementaryCalculations"]:
139 Jb = 0.5 * (Xa - Xb).T * BI * (Xa - Xb)
140 Jo = 0.5 * oma.T * RI * oma
141 J = float( Jb ) + float( Jo )
142 self.StoredVariables["CostFunctionJb"].store( Jb )
143 self.StoredVariables["CostFunctionJo"].store( Jo )
144 self.StoredVariables["CostFunctionJ" ].store( J )
146 # Calcul de la covariance d'analyse
147 # ---------------------------------
148 if "APosterioriCovariance" in self._parameters["StoreSupplementaryCalculations"] or \
149 "SimulationQuantiles" in self._parameters["StoreSupplementaryCalculations"]:
151 if min(A.shape) != max(A.shape):
152 raise ValueError("The %s a posteriori covariance matrix A is of shape %s, despites it has to be a squared matrix. There is an error in the observation operator, please check it."%(self._name,str(A.shape)))
153 if (numpy.diag(A) < 0).any():
154 raise ValueError("The %s a posteriori covariance matrix A has at least one negative value on its diagonal. There is an error in the observation operator, please check it."%(self._name,))
155 if logging.getLogger().level < logging.WARNING: # La verification n'a lieu qu'en debug
157 L = numpy.linalg.cholesky( A )
159 raise ValueError("The %s a posteriori covariance matrix A is not symmetric positive-definite. Please check your a priori covariances and your observation operator."%(self._name,))
160 self.StoredVariables["APosterioriCovariance"].store( A )
162 # Calculs et/ou stockages supplémentaires
163 # ---------------------------------------
164 if "Innovation" in self._parameters["StoreSupplementaryCalculations"]:
165 self.StoredVariables["Innovation"].store( numpy.ravel(d) )
166 if "BMA" in self._parameters["StoreSupplementaryCalculations"]:
167 self.StoredVariables["BMA"].store( numpy.ravel(Xb) - numpy.ravel(Xa) )
168 if "OMA" in self._parameters["StoreSupplementaryCalculations"]:
169 self.StoredVariables["OMA"].store( numpy.ravel(oma) )
170 if "OMB" in self._parameters["StoreSupplementaryCalculations"]:
171 self.StoredVariables["OMB"].store( numpy.ravel(d) )
172 if "SigmaObs2" in self._parameters["StoreSupplementaryCalculations"]:
173 TraceR = R.trace(Y.size)
174 self.StoredVariables["SigmaObs2"].store( float( (d.T * (numpy.asmatrix(numpy.ravel(oma)).T)) ) / TraceR )
175 if "SigmaBck2" in self._parameters["StoreSupplementaryCalculations"]:
176 self.StoredVariables["SigmaBck2"].store( float( (d.T * Hm * (Xa - Xb))/(Hm * B * Hm.T).trace() ) )
177 if "MahalanobisConsistency" in self._parameters["StoreSupplementaryCalculations"]:
178 self.StoredVariables["MahalanobisConsistency"].store( float( 2.*J/d.size ) )
179 if "SimulationQuantiles" in self._parameters["StoreSupplementaryCalculations"]:
180 Qtls = self._parameters["Quantiles"]
181 nech = self._parameters["NumberOfSamplesForQuantiles"]
183 for i in range(nech):
184 if self._parameters["SimulationForQuantiles"] == "Linear":
185 dXr = numpy.matrix(numpy.random.multivariate_normal(Xa.A1,A) - Xa.A1).T
186 dYr = numpy.matrix(numpy.ravel( Hm * dXr )).T
188 elif self._parameters["SimulationForQuantiles"] == "NonLinear":
189 Xr = numpy.matrix(numpy.random.multivariate_normal(Xa.A1,A)).T
190 Yr = numpy.matrix(numpy.ravel( Hm * Xr )).T
194 YfQ = numpy.hstack((YfQ,Yr))
197 for quantile in Qtls:
198 if not (0. <= quantile <= 1.): continue
199 indice = int(nech * quantile - 1./nech)
200 if YQ is None: YQ = YfQ[:,indice]
201 else: YQ = numpy.hstack((YQ,YfQ[:,indice]))
202 self.StoredVariables["SimulationQuantiles"].store( YQ )
204 logging.debug("%s Nombre d'évaluation(s) de l'opérateur d'observation direct/tangent/adjoint : %i/%i/%i"%(self._name, HO["Direct"].nbcalls()[0],HO["Tangent"].nbcalls()[0],HO["Adjoint"].nbcalls()[0]))
205 logging.debug("%s Taille mémoire utilisée de %.1f Mo"%(self._name, m.getUsedMemory("M")))
206 logging.debug("%s Terminé"%self._name)
210 # ==============================================================================
211 if __name__ == "__main__":
212 print '\n AUTODIAGNOSTIC \n'