src/daComposant/daAlgorithms/DerivativeFreeOptimization.py

   1 #-*-coding:iso-8859-1-*-
   2 #
   3 #  Copyright (C) 2008-2015 EDF R&D
   4 #
   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.
   9 #
  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.
  14 #
  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
  18 #
  19 #  See http://www.salome-platform.org/ or email : webmaster.salome@opencascade.com
  20 #
  21 #  Author: Jean-Philippe Argaud, jean-philippe.argaud@edf.fr, EDF R&D
  22
  23 import logging
  24 from daCore import BasicObjects
  25 import numpy, scipy.optimize
  26
  27 # ==============================================================================
  28 class ElementaryAlgorithm(BasicObjects.Algorithm):
  29     def __init__(self):
  30         BasicObjects.Algorithm.__init__(self, "DERIVATIVEFREEOPTIMIZATION")
  31         self.defineRequiredParameter(
  32             name     = "Minimizer",
  33             default  = "POWELL",
  34             typecast = str,
  35             message  = "Minimiseur utilisé",
  36             listval  = ["POWELL", "SIMPLEX"],
  37             )
  38         self.defineRequiredParameter(
  39             name     = "MaximumNumberOfSteps",
  40             default  = 15000,
  41             typecast = int,
  42             message  = "Nombre maximal de pas d'optimisation",
  43             minval   = -1,
  44             )
  45         self.defineRequiredParameter(
  46             name     = "MaximumNumberOfFunctionEvaluations",
  47             default  = 15000,
  48             typecast = int,
  49             message  = "Nombre maximal de d'évaluations de la function",
  50             minval   = -1,
  51             )
  52         self.defineRequiredParameter(
  53             name     = "StateVariationTolerance",
  54             default  = 1.e-4,
  55             typecast = float,
  56             message  = "Variation relative maximale de l'état lors de l'arrêt",
  57             )
  58         self.defineRequiredParameter(
  59             name     = "CostDecrementTolerance",
  60             default  = 1.e-7,
  61             typecast = float,
  62             message  = "Diminution relative minimale du cout lors de l'arrêt",
  63             )
  64         self.defineRequiredParameter(
  65             name     = "QualityCriterion",
  66             default  = "AugmentedWeightedLeastSquares",
  67             typecast = str,
  68             message  = "Critère de qualité utilisé",
  69             listval  = ["AugmentedWeightedLeastSquares","AWLS","DA",
  70                         "WeightedLeastSquares","WLS",
  71                         "LeastSquares","LS","L2",
  72                         "AbsoluteValue","L1",
  73                         "MaximumError","ME"],
  74             )
  75         self.defineRequiredParameter(
  76             name     = "StoreInternalVariables",
  77             default  = False,
  78             typecast = bool,
  79             message  = "Stockage des variables internes ou intermédiaires du calcul",
  80             )
  81         self.defineRequiredParameter(
  82             name     = "StoreSupplementaryCalculations",
  83             default  = [],
  84             typecast = tuple,
  85             message  = "Liste de calculs supplémentaires à stocker et/ou effectuer",
  86             listval  = ["CurrentState", "CostFunctionJ", "CostFunctionJAtCurrentOptimum", "CurrentOptimum", "IndexOfOptimum", "InnovationAtCurrentState", "BMA", "OMA", "OMB", "SimulatedObservationAtBackground", "SimulatedObservationAtCurrentOptimum", "SimulatedObservationAtCurrentState", "SimulatedObservationAtOptimum"]
  87             )
  88
  89     def run(self, Xb=None, Y=None, U=None, HO=None, EM=None, CM=None, R=None, B=None, Q=None, Parameters=None):
  90         self._pre_run()
  91         if logging.getLogger().level < logging.WARNING:
  92             self.__disp = 1
  93         else:
  94             self.__disp = 0
  95         #
  96         # Paramètres de pilotage
  97         # ----------------------
  98         self.setParameters(Parameters)
  99         #
 100         # Opérateurs
 101         # ----------
 102         Hm = HO["Direct"].appliedTo
 103         #
 104         # Précalcul des inversions de B et R
 105         # ----------------------------------
 106         BI = B.getI()
 107         RI = R.getI()
 108         #
 109         # Définition de la fonction-coût
 110         # ------------------------------
 111         def CostFunction(x, QualityMeasure="AugmentedWeightedLeastSquares"):
 112             _X  = numpy.asmatrix(numpy.ravel( x )).T
 113             self.StoredVariables["CurrentState"].store( _X )
 114             _HX = Hm( _X )
 115             _HX = numpy.asmatrix(numpy.ravel( _HX )).T
 116             _Innovation = Y - _HX
 117             if "SimulatedObservationAtCurrentState" in self._parameters["StoreSupplementaryCalculations"] or \
 118                "SimulatedObservationAtCurrentOptimum" in self._parameters["StoreSupplementaryCalculations"]:
 119                 self.StoredVariables["SimulatedObservationAtCurrentState"].store( _HX )
 120             if "InnovationAtCurrentState" in self._parameters["StoreSupplementaryCalculations"]:
 121                 self.StoredVariables["InnovationAtCurrentState"].store( _Innovation )
 122             #
 123             if QualityMeasure in ["AugmentedWeightedLeastSquares","AWLS","DA"]:
 124                 if BI is None or RI is None:
 125                     raise ValueError("Background and Observation error covariance matrix has to be properly defined!")
 126                 Jb  = 0.5 * (_X - Xb).T * BI * (_X - Xb)
 127                 Jo  = 0.5 * (_Innovation).T * RI * (_Innovation)
 128             elif QualityMeasure in ["WeightedLeastSquares","WLS"]:
 129                 if RI is None:
 130                     raise ValueError("Observation error covariance matrix has to be properly defined!")
 131                 Jb  = 0.
 132                 Jo  = 0.5 * (_Innovation).T * RI * (_Innovation)
 133             elif QualityMeasure in ["LeastSquares","LS","L2"]:
 134                 Jb  = 0.
 135                 Jo  = 0.5 * (_Innovation).T * (_Innovation)
 136             elif QualityMeasure in ["AbsoluteValue","L1"]:
 137                 Jb  = 0.
 138                 Jo  = numpy.sum( numpy.abs(_Innovation) )
 139             elif QualityMeasure in ["MaximumError","ME"]:
 140                 Jb  = 0.
 141                 Jo  = numpy.max( numpy.abs(_Innovation) )
 142             #
 143             J   = float( Jb ) + float( Jo )
 144             #
 145             self.StoredVariables["CostFunctionJb"].store( Jb )
 146             self.StoredVariables["CostFunctionJo"].store( Jo )
 147             self.StoredVariables["CostFunctionJ" ].store( J )
 148             if "IndexOfOptimum" in self._parameters["StoreSupplementaryCalculations"] or \
 149                "CurrentOptimum" in self._parameters["StoreSupplementaryCalculations"] or \
 150                "CostFunctionJAtCurrentOptimum" in self._parameters["StoreSupplementaryCalculations"] or \
 151                "SimulatedObservationAtCurrentOptimum" in self._parameters["StoreSupplementaryCalculations"]:
 152                 IndexMin = numpy.argmin( self.StoredVariables["CostFunctionJ"][nbPreviousSteps:] ) + nbPreviousSteps
 153             if "IndexOfOptimum" in self._parameters["StoreSupplementaryCalculations"]:
 154                 self.StoredVariables["IndexOfOptimum"].store( IndexMin )
 155             if "CurrentOptimum" in self._parameters["StoreSupplementaryCalculations"]:
 156                 self.StoredVariables["CurrentOptimum"].store( self.StoredVariables["CurrentState"][IndexMin] )
 157             if "SimulatedObservationAtCurrentOptimum" in self._parameters["StoreSupplementaryCalculations"]:
 158                 self.StoredVariables["SimulatedObservationAtCurrentOptimum"].store( self.StoredVariables["SimulatedObservationAtCurrentState"][IndexMin] )
 159             if "CostFunctionJAtCurrentOptimum" in self._parameters["StoreSupplementaryCalculations"]:
 160                 self.StoredVariables["CostFunctionJbAtCurrentOptimum"].store( self.StoredVariables["CostFunctionJb"][IndexMin] )
 161                 self.StoredVariables["CostFunctionJoAtCurrentOptimum"].store( self.StoredVariables["CostFunctionJo"][IndexMin] )
 162                 self.StoredVariables["CostFunctionJAtCurrentOptimum" ].store( self.StoredVariables["CostFunctionJ" ][IndexMin] )
 163             return J
 164         #
 165         # Point de démarrage de l'optimisation : Xini = Xb
 166         # ------------------------------------
 167         Xini = numpy.ravel(Xb)
 168         #
 169         # Minimisation de la fonctionnelle
 170         # --------------------------------
 171         nbPreviousSteps = self.StoredVariables["CostFunctionJ"].stepnumber()
 172         #
 173         if self._parameters["Minimizer"] == "POWELL":
 174             Minimum, J_optimal, direc, niter, nfeval, rc = scipy.optimize.fmin_powell(
 175                 func        = CostFunction,
 176                 x0          = Xini,
 177                 args        = (self._parameters["QualityCriterion"],),
 178                 maxiter     = self._parameters["MaximumNumberOfSteps"]-1,
 179                 maxfun      = self._parameters["MaximumNumberOfFunctionEvaluations"]-1,
 180                 xtol        = self._parameters["StateVariationTolerance"],
 181                 ftol        = self._parameters["CostDecrementTolerance"],
 182                 full_output = True,
 183                 disp        = self.__disp,
 184                 )
 185         elif self._parameters["Minimizer"] == "SIMPLEX":
 186             Minimum, J_optimal, niter, nfeval, rc = scipy.optimize.fmin(
 187                 func        = CostFunction,
 188                 x0          = Xini,
 189                 args        = (self._parameters["QualityCriterion"],),
 190                 maxiter     = self._parameters["MaximumNumberOfSteps"]-1,
 191                 maxfun      = self._parameters["MaximumNumberOfFunctionEvaluations"]-1,
 192                 xtol        = self._parameters["StateVariationTolerance"],
 193                 ftol        = self._parameters["CostDecrementTolerance"],
 194                 full_output = True,
 195                 disp        = self.__disp,
 196                 )
 197         else:
 198             raise ValueError("Error in Minimizer name: %s"%self._parameters["Minimizer"])
 199         #
 200         IndexMin = numpy.argmin( self.StoredVariables["CostFunctionJ"][nbPreviousSteps:] ) + nbPreviousSteps
 201         MinJ     = self.StoredVariables["CostFunctionJ"][IndexMin]
 202         Minimum  = self.StoredVariables["CurrentState"][IndexMin]
 203         #
 204         # Obtention de l'analyse
 205         # ----------------------
 206         Xa = numpy.asmatrix(numpy.ravel( Minimum )).T
 207         #
 208         self.StoredVariables["Analysis"].store( Xa.A1 )
 209         #
 210         if "OMA"                           in self._parameters["StoreSupplementaryCalculations"] or \
 211            "SimulatedObservationAtOptimum" in self._parameters["StoreSupplementaryCalculations"]:
 212             if "SimulatedObservationAtCurrentState" in self._parameters["StoreSupplementaryCalculations"]:
 213                 HXa = self.StoredVariables["SimulatedObservationAtCurrentState"][IndexMin]
 214             elif "SimulatedObservationAtCurrentOptimum" in self._parameters["StoreSupplementaryCalculations"]:
 215                 HXa = self.StoredVariables["SimulatedObservationAtCurrentOptimum"][-1]
 216             else:
 217                 HXa = Hm(Xa)
 218         #
 219         if "Innovation" in self._parameters["StoreSupplementaryCalculations"]:
 220             self.StoredVariables["Innovation"].store( numpy.ravel(d) )
 221         if "OMB" in self._parameters["StoreSupplementaryCalculations"]:
 222             self.StoredVariables["OMB"].store( numpy.ravel(d) )
 223         if "BMA" in self._parameters["StoreSupplementaryCalculations"]:
 224             self.StoredVariables["BMA"].store( numpy.ravel(Xb) - numpy.ravel(Xa) )
 225         if "OMA" in self._parameters["StoreSupplementaryCalculations"]:
 226             self.StoredVariables["OMA"].store( numpy.ravel(Y) - numpy.ravel(HXa) )
 227         if "SimulatedObservationAtBackground" in self._parameters["StoreSupplementaryCalculations"]:
 228             self.StoredVariables["SimulatedObservationAtBackground"].store( numpy.ravel(Hm(Xb)) )
 229         if "SimulatedObservationAtOptimum" in self._parameters["StoreSupplementaryCalculations"]:
 230             self.StoredVariables["SimulatedObservationAtOptimum"].store( numpy.ravel(HXa) )
 231         #
 232         self._post_run()
 233         return 0
 234
 235 # ==============================================================================
 236 if __name__ == "__main__":
 237     print '\n AUTODIAGNOSTIC \n'