1 #-*-coding:iso-8859-1-*-
3 # Copyright (C) 2008-2011 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
23 from daCore import BasicObjects, PlatformInfo
24 m = PlatformInfo.SystemUsage()
29 if logging.getLogger().level < 30:
31 message = scipy.optimize.tnc.MSG_ALL
35 message = scipy.optimize.tnc.MSG_NONE
38 # ==============================================================================
39 class ElementaryAlgorithm(BasicObjects.Algorithm):
41 BasicObjects.Algorithm.__init__(self)
43 logging.debug("%s Initialisation"%self._name)
45 def run(self, Xb=None, Y=None, H=None, M=None, R=None, B=None, Q=None, Parameters=None):
47 Calcul de l'estimateur 3D-VAR
49 logging.debug("%s Lancement"%self._name)
50 logging.debug("%s Taille mémoire utilisée de %.1f Mo"%(self._name, m.getUsedMemory("Mo")))
52 # Opérateur d'observation
53 # -----------------------
54 Hm = H["Direct"].appliedTo
55 Ht = H["Adjoint"].appliedInXTo
57 # Utilisation éventuelle d'un vecteur H(Xb) précalculé
58 # ----------------------------------------------------
59 if H["AppliedToX"] is not None and H["AppliedToX"].has_key("HXb"):
60 logging.debug("%s Utilisation de HXb"%self._name)
61 HXb = H["AppliedToX"]["HXb"]
63 logging.debug("%s Calcul de Hm(Xb)"%self._name)
65 HXb = numpy.asmatrix(HXb).flatten().T
67 # Calcul du préconditionnement
68 # ----------------------------
69 # Bdemi = numpy.linalg.cholesky(B)
71 # Calcul de l'innovation
72 # ----------------------
74 logging.debug("%s Innovation d = %s"%(self._name, d))
76 # Précalcul des inversion appellée dans les fonction-coût et gradient
77 # -------------------------------------------------------------------
81 # Définition de la fonction-coût
82 # ------------------------------
84 _X = numpy.asmatrix(x).flatten().T
85 logging.info("%s CostFunction X = %s"%(self._name, numpy.asmatrix( _X ).flatten()))
87 _HX = numpy.asmatrix(_HX).flatten().T
88 Jb = 0.5 * (_X - Xb).T * BI * (_X - Xb)
89 Jo = 0.5 * (Y - _HX).T * RI * (Y - _HX)
90 J = float( Jb ) + float( Jo )
91 logging.info("%s CostFunction Jb = %s"%(self._name, Jb))
92 logging.info("%s CostFunction Jo = %s"%(self._name, Jo))
93 logging.info("%s CostFunction J = %s"%(self._name, J))
94 self.StoredVariables["CurrentState"].store( _X.A1 )
95 self.StoredVariables["CostFunctionJb"].store( Jb )
96 self.StoredVariables["CostFunctionJo"].store( Jo )
97 self.StoredVariables["CostFunctionJ" ].store( J )
100 def GradientOfCostFunction(x):
101 _X = numpy.asmatrix(x).flatten().T
102 logging.info("%s GradientOfCostFunction X = %s"%(self._name, numpy.asmatrix( _X ).flatten()))
104 _HX = numpy.asmatrix(_HX).flatten().T
105 GradJb = BI * (_X - Xb)
106 GradJo = - Ht( (_X, RI * (Y - _HX)) )
107 GradJ = numpy.asmatrix( GradJb ).flatten().T + numpy.asmatrix( GradJo ).flatten().T
108 logging.debug("%s GradientOfCostFunction GradJb = %s"%(self._name, numpy.asmatrix( GradJb ).flatten()))
109 logging.debug("%s GradientOfCostFunction GradJo = %s"%(self._name, numpy.asmatrix( GradJo ).flatten()))
110 logging.debug("%s GradientOfCostFunction GradJ = %s"%(self._name, numpy.asmatrix( GradJ ).flatten()))
113 # Point de démarrage de l'optimisation : Xini = Xb
114 # ------------------------------------
115 if type(Xb) is type(numpy.matrix([])):
116 Xini = Xb.A1.tolist()
119 logging.debug("%s Point de démarrage Xini = %s"%(self._name, Xini))
121 # Paramètres de pilotage
122 # ----------------------
123 # Potentiels : "Bounds", "Minimizer", "MaximumNumberOfSteps", "ProjectedGradientTolerance", "GradientNormTolerance", "InnerMinimizer"
124 if Parameters.has_key("Bounds") and (type(Parameters["Bounds"]) is type([]) or type(Parameters["Bounds"]) is type(())) and (len(Parameters["Bounds"]) > 0):
125 Bounds = Parameters["Bounds"]
128 MinimizerList = ["LBFGSB","TNC", "CG", "NCG", "BFGS"]
129 if Parameters.has_key("Minimizer") and (Parameters["Minimizer"] in MinimizerList):
130 Minimizer = str( Parameters["Minimizer"] )
132 logging.warning("%s Minimiseur inconnu ou non fourni, remplacé par la valeur par défaut"%self._name)
134 logging.debug("%s Minimiseur utilisé = %s"%(self._name, Minimizer))
135 if Parameters.has_key("MaximumNumberOfSteps") and (Parameters["MaximumNumberOfSteps"] > -1):
136 maxiter = int( Parameters["MaximumNumberOfSteps"] )
139 logging.debug("%s Nombre maximal de pas d'optimisation = %s"%(self._name, str(maxiter)))
140 if Parameters.has_key("CostDecrementTolerance") and (Parameters["CostDecrementTolerance"] > 0):
141 ftol = float(Parameters["CostDecrementTolerance"])
146 logging.debug("%s Diminution relative minimale du cout lors de l'arret = %s"%(self._name, str(1./factr)))
147 if Parameters.has_key("ProjectedGradientTolerance") and (Parameters["ProjectedGradientTolerance"] > -1):
148 pgtol = float(Parameters["ProjectedGradientTolerance"])
151 logging.debug("%s Maximum des composantes du gradient projete lors de l'arret = %s"%(self._name, str(pgtol)))
152 if Parameters.has_key("GradientNormTolerance") and (Parameters["GradientNormTolerance"] > -1):
153 gtol = float(Parameters["GradientNormTolerance"])
156 logging.debug("%s Maximum des composantes du gradient lors de l'arret = %s"%(self._name, str(gtol)))
157 InnerMinimizerList = ["CG", "NCG", "BFGS"]
158 if Parameters.has_key("InnerMinimizer") and (Parameters["InnerMinimizer"] in InnerMinimizerList):
159 InnerMinimizer = str( Parameters["Minimizer"] )
161 InnerMinimizer = "BFGS"
162 logging.debug("%s Minimiseur interne utilisé = %s"%(self._name, InnerMinimizer))
163 logging.debug("%s Norme du gradient lors de l'arret = %s"%(self._name, str(gtol)))
165 # Minimisation de la fonctionnelle
166 # --------------------------------
167 if Minimizer == "LBFGSB":
168 Minimum, J_optimal, Informations = scipy.optimize.fmin_l_bfgs_b(
171 fprime = GradientOfCostFunction,
179 nfeval = Informations['funcalls']
180 rc = Informations['warnflag']
181 elif Minimizer == "TNC":
182 Minimum, nfeval, rc = scipy.optimize.fmin_tnc(
185 fprime = GradientOfCostFunction,
193 elif Minimizer == "CG":
194 Minimum, fopt, nfeval, grad_calls, rc = scipy.optimize.fmin_cg(
197 fprime = GradientOfCostFunction,
204 elif Minimizer == "NCG":
205 Minimum, fopt, nfeval, grad_calls, hcalls, rc = scipy.optimize.fmin_ncg(
208 fprime = GradientOfCostFunction,
215 elif Minimizer == "BFGS":
216 Minimum, fopt, gopt, Hopt, nfeval, grad_calls, rc = scipy.optimize.fmin_bfgs(
219 fprime = GradientOfCostFunction,
227 raise ValueError("Error in Minimizer name: %s"%Minimizer)
229 # Correction pour pallier a un bug de TNC sur le retour du Minimum
230 # ----------------------------------------------------------------
231 StepMin = numpy.argmin( self.StoredVariables["CostFunctionJ"].valueserie() )
232 MinJ = self.StoredVariables["CostFunctionJ"].valueserie(step = StepMin)
233 Minimum = self.StoredVariables["CurrentState"].valueserie(step = StepMin)
235 logging.debug("%s %s Step of min cost = %s"%(self._name, Minimizer, StepMin))
236 logging.debug("%s %s Minimum cost = %s"%(self._name, Minimizer, MinJ))
237 logging.debug("%s %s Minimum state = %s"%(self._name, Minimizer, Minimum))
238 logging.debug("%s %s Nb of F = %s"%(self._name, Minimizer, nfeval))
239 logging.debug("%s %s RetCode = %s"%(self._name, Minimizer, rc))
241 # Calcul de l'analyse
242 # --------------------
243 Xa = numpy.asmatrix(Minimum).T
244 logging.debug("%s Analyse Xa = %s"%(self._name, Xa))
246 self.StoredVariables["Analysis"].store( Xa.A1 )
247 self.StoredVariables["Innovation"].store( d.A1 )
249 logging.debug("%s Taille mémoire utilisée de %.1f Mo"%(self._name, m.getUsedMemory("MB")))
250 logging.debug("%s Terminé"%self._name)
254 # ==============================================================================
255 if __name__ == "__main__":
256 print '\n AUTODIAGNOSTIC \n'