-#-*-coding:iso-8859-1-*-
+# -*- coding: utf-8 -*-
#
-# Copyright (C) 2008-2016 EDF R&D
+# Copyright (C) 2008-2023 EDF R&D
#
# This library is free software; you can redistribute it and/or
# modify it under the terms of the GNU Lesser General Public
# Author: Jean-Philippe Argaud, jean-philippe.argaud@edf.fr, EDF R&D
"Verification de la disponibilite de l'ensemble des algorithmes"
+import sys
+import unittest
+import numpy
+from adao import adaoBuilder
+
# ==============================================================================
-import adaoBuilder, numpy
-def test1():
- """Verification de la disponibilite de l'ensemble des algorithmes"""
- for algo in ("3DVAR", "Blue", "ExtendedBlue", "LinearLeastSquares", "NonLinearLeastSquares", ):
- print
- msg = "Algorithme en test : %s"%algo
- print msg+"\n"+"="*len(msg)
+class Test_Adao(unittest.TestCase):
+ def test1(self):
+ """Verification de la disponibilite de l'ensemble des algorithmes\n(Utilisation d'un operateur matriciel)"""
+ print(self.test1.__doc__.strip()+"\n")
+ Xa = {}
+ for algo in ("3DVAR", "Blue", "ExtendedBlue", "LinearLeastSquares", "NonLinearLeastSquares", "DerivativeFreeOptimization"):
+ print("")
+ msg = "Algorithme en test : %s"%algo
+ print(msg+"\n"+"-"*len(msg))
+ #
+ adaopy = adaoBuilder.New()
+ adaopy.setAlgorithmParameters(Algorithm=algo, Parameters={"EpsilonMinimumExponent":-10, "Bounds":[[-1,10.],[-1,10.],[-1,10.]]})
+ adaopy.setBackground (Vector = [0,1,2])
+ adaopy.setBackgroundError (ScalarSparseMatrix = 1.)
+ adaopy.setObservation (Vector = [0.5,1.5,2.5])
+ adaopy.setObservationError (DiagonalSparseMatrix = "1 1 1")
+ adaopy.setObservationOperator(Matrix = "1 0 0;0 2 0;0 0 3")
+ adaopy.setObserver("Analysis",Template="ValuePrinter")
+ adaopy.execute()
+ Xa[algo] = adaopy.get("Analysis")[-1]
+ del adaopy
+ #
+ for algo in ("ExtendedKalmanFilter", "KalmanFilter", "UnscentedKalmanFilter", "EnsembleKalmanFilter", "4DVAR"):
+ print("")
+ msg = "Algorithme en test : %s"%algo
+ print(msg+"\n"+"-"*len(msg))
+ #
+ adaopy = adaoBuilder.New()
+ adaopy.setAlgorithmParameters(Algorithm=algo, Parameters={"EpsilonMinimumExponent":-10, "SetSeed":1000})
+ adaopy.setBackground (Vector = [0,1,2])
+ adaopy.setBackgroundError (ScalarSparseMatrix = 1.)
+ adaopy.setObservation (Vector = [0.5,1.5,2.5])
+ adaopy.setObservationError (DiagonalSparseMatrix = "1 1 1")
+ adaopy.setObservationOperator(Matrix = "1 0 0;0 2 0;0 0 3")
+ adaopy.setEvolutionError (ScalarSparseMatrix = 1.)
+ adaopy.setEvolutionModel (Matrix = "1 0 0;0 1 0;0 0 1")
+ adaopy.setObserver("Analysis",Template="ValuePrinter")
+ adaopy.execute()
+ Xa[algo] = adaopy.get("Analysis")[-1]
+ del adaopy
#
- adaopy = adaoBuilder.New()
- adaopy.setAlgorithmParameters(Algorithm=algo, Parameters={"EpsilonMinimumExponent":-10, })
- adaopy.setBackground (Vector = [0,1,2])
- adaopy.setBackgroundError (ScalarSparseMatrix = 1.)
- adaopy.setObservation (Vector = [0.5,1.5,2.5])
- adaopy.setObservationError (DiagonalSparseMatrix = "1 1 1")
- adaopy.setObservationOperator(Matrix = "1 0 0;0 2 0;0 0 3")
- adaopy.setObserver("Analysis",Template="ValuePrinter")
- adaopy.execute()
- del adaopy
- #
- for algo in ("ExtendedKalmanFilter", "KalmanFilter", "UnscentedKalmanFilter", "4DVAR"):
- print
- msg = "Algorithme en test : %s"%algo
- print msg+"\n"+"="*len(msg)
+ for algo in ("ParticleSwarmOptimization", "QuantileRegression", ):
+ print("")
+ msg = "Algorithme en test : %s"%algo
+ print(msg+"\n"+"-"*len(msg))
+ #
+ adaopy = adaoBuilder.New()
+ adaopy.setAlgorithmParameters(Algorithm=algo, Parameters={"BoxBounds":3*[[-1,3]], "SetSeed":1000})
+ adaopy.setBackground (Vector = [0,1,2])
+ adaopy.setBackgroundError (ScalarSparseMatrix = 1.)
+ adaopy.setObservation (Vector = [0.5,1.5,2.5])
+ adaopy.setObservationError (DiagonalSparseMatrix = "1 2 3")
+ adaopy.setObservationOperator(Matrix = "1 0 0;0 1 0;0 0 1")
+ adaopy.setObserver("Analysis",Template="ValuePrinter")
+ adaopy.execute()
+ Xa[algo] = adaopy.get("Analysis")[-1]
+ del adaopy
#
- adaopy = adaoBuilder.New()
- adaopy.setAlgorithmParameters(Algorithm=algo, Parameters={"EpsilonMinimumExponent":-10, })
- adaopy.setBackground (Vector = [0,1,2])
- adaopy.setBackgroundError (ScalarSparseMatrix = 1.)
- adaopy.setObservation (Vector = [0.5,1.5,2.5])
- adaopy.setObservationError (DiagonalSparseMatrix = "1 1 1")
- adaopy.setObservationOperator(Matrix = "1 0 0;0 1 0;0 0 1")
- adaopy.setEvolutionModel (Matrix = "1 0 0;0 1 0;0 0 1")
- adaopy.setEvolutionError (ScalarSparseMatrix = 1.)
- adaopy.setObserver("Analysis",Template="ValuePrinter")
- adaopy.execute()
- del adaopy
- #
- for algo in ("ParticleSwarmOptimization", "QuantileRegression", ):
- print
- msg = "Algorithme en test : %s"%algo
- print msg+"\n"+"="*len(msg)
+ for algo in ("EnsembleBlue", ):
+ print("")
+ msg = "Algorithme en test : %s"%algo
+ print(msg+"\n"+"-"*len(msg))
+ #
+ adaopy = adaoBuilder.New()
+ adaopy.setAlgorithmParameters(Algorithm=algo, Parameters={"SetSeed":1000, })
+ adaopy.setBackground (VectorSerie = 100*[[0,1,2]])
+ adaopy.setBackgroundError (ScalarSparseMatrix = 1.)
+ adaopy.setObservation (Vector = [0.5,1.5,2.5])
+ adaopy.setObservationError (DiagonalSparseMatrix = "1 2 3")
+ adaopy.setObservationOperator(Matrix = "1 0 0;0 1 0;0 0 1")
+ adaopy.setObserver("Analysis",Template="ValuePrinter")
+ adaopy.execute()
+ Xa[algo] = adaopy.get("Analysis")[-1]
+ del adaopy
+ #
+ print("")
+ msg = "Tests des ecarts attendus :"
+ print(msg+"\n"+"="*len(msg))
+ verify_similarity_of_algo_results(("3DVAR", "Blue", "ExtendedBlue", "4DVAR", "DerivativeFreeOptimization"), Xa, 5.e-5)
+ verify_similarity_of_algo_results(("LinearLeastSquares", "NonLinearLeastSquares"), Xa, 5.e-7)
+ verify_similarity_of_algo_results(("KalmanFilter", "ExtendedKalmanFilter", "UnscentedKalmanFilter"), Xa, 1.e-14)
+ verify_similarity_of_algo_results(("KalmanFilter", "EnsembleKalmanFilter"), Xa, 2.e-1)
+ print(" Les resultats obtenus sont corrects.")
+ print("")
+
+ def test2(self):
+ """Verification de la disponibilite de l'ensemble des algorithmes\n(Utilisation d'un operateur fonctionnel)"""
+ print(self.test2.__doc__)
+ Xa = {}
+ M = numpy.diag([1.,2.,3.])
+ def H(x): return M @ numpy.ravel( x )
+ for algo in ("3DVAR", "Blue", "ExtendedBlue", "NonLinearLeastSquares", "DerivativeFreeOptimization"):
+ print("")
+ msg = "Algorithme en test : %s"%algo
+ print(msg+"\n"+"-"*len(msg))
+ #
+ adaopy = adaoBuilder.New()
+ adaopy.setAlgorithmParameters(Algorithm=algo, Parameters={"EpsilonMinimumExponent":-10, "Bounds":[[-1,10.],[-1,10.],[-1,10.]]})
+ adaopy.setBackground (Vector = [0,1,2])
+ adaopy.setBackgroundError (ScalarSparseMatrix = 1.)
+ adaopy.setObservation (Vector = [0.5,1.5,2.5])
+ adaopy.setObservationError (DiagonalSparseMatrix = "1 1 1")
+ adaopy.setObservationOperator(OneFunction = H)
+ adaopy.setObserver("Analysis",Template="ValuePrinter")
+ adaopy.execute()
+ Xa[algo] = adaopy.get("Analysis")[-1]
+ del adaopy
#
- adaopy = adaoBuilder.New()
- adaopy.setAlgorithmParameters(Algorithm=algo, Parameters={"BoxBounds":3*[[-1,3]], "SetSeed":1000, })
- adaopy.setBackground (Vector = [0,1,2])
- adaopy.setBackgroundError (ScalarSparseMatrix = 1.)
- adaopy.setObservation (Vector = [0.5,1.5,2.5])
- adaopy.setObservationError (DiagonalSparseMatrix = "1 1 1")
- adaopy.setObservationOperator(Matrix = "1 0 0;0 1 0;0 0 1")
- adaopy.setEvolutionModel (Matrix = "1 0 0;0 1 0;0 0 1")
- adaopy.setEvolutionError (ScalarSparseMatrix = 1.)
- adaopy.setObserver("Analysis",Template="ValuePrinter")
- adaopy.execute()
- del adaopy
- #
- for algo in ("EnsembleBlue", ):
- print
- msg = "Algorithme en test : %s"%algo
- print msg+"\n"+"="*len(msg)
+ M = numpy.diag([1.,2.,3.])
+ def H(x): return M @ numpy.ravel( x )
+ for algo in ("ExtendedKalmanFilter", "KalmanFilter", "EnsembleKalmanFilter", "UnscentedKalmanFilter", "4DVAR"):
+ print("")
+ msg = "Algorithme en test : %s"%algo
+ print(msg+"\n"+"-"*len(msg))
+ #
+ adaopy = adaoBuilder.New()
+ adaopy.setAlgorithmParameters(Algorithm=algo, Parameters={"EpsilonMinimumExponent":-10, "SetSeed":1000})
+ adaopy.setBackground (Vector = [0,1,2])
+ adaopy.setBackgroundError (ScalarSparseMatrix = 1.)
+ adaopy.setObservation (Vector = [0.5,1.5,2.5])
+ adaopy.setObservationError (DiagonalSparseMatrix = "1 1 1")
+ adaopy.setObservationOperator(OneFunction = H)
+ adaopy.setEvolutionError (ScalarSparseMatrix = 1.)
+ adaopy.setEvolutionModel (Matrix = "1 0 0;0 1 0;0 0 1")
+ adaopy.setObserver("Analysis",Template="ValuePrinter")
+ adaopy.execute()
+ Xa[algo] = adaopy.get("Analysis")[-1]
+ del adaopy
#
- adaopy = adaoBuilder.New()
- adaopy.setAlgorithmParameters(Algorithm=algo, Parameters={"SetSeed":1000, })
- adaopy.setBackground (VectorSerie = 100*[[0,1,2]])
- adaopy.setBackgroundError (ScalarSparseMatrix = 1.)
- adaopy.setObservation (Vector = [0.5,1.5,2.5])
- adaopy.setObservationError (DiagonalSparseMatrix = "1 1 1")
- adaopy.setObservationOperator(Matrix = "1 0 0;0 1 0;0 0 1")
- adaopy.setEvolutionModel (Matrix = "1 0 0;0 1 0;0 0 1")
- adaopy.setEvolutionError (ScalarSparseMatrix = 1.)
- adaopy.setObserver("Analysis",Template="ValuePrinter")
- adaopy.execute()
- del adaopy
- #
- return 0
+ M = numpy.identity(3)
+ def H(x): return M @ numpy.ravel( x )
+ for algo in ("ParticleSwarmOptimization", "QuantileRegression", ):
+ print("")
+ msg = "Algorithme en test : %s"%algo
+ print(msg+"\n"+"-"*len(msg))
+ #
+ adaopy = adaoBuilder.New()
+ adaopy.setAlgorithmParameters(Algorithm=algo, Parameters={"BoxBounds":3*[[-1,3]], "SetSeed":1000})
+ adaopy.setBackground (Vector = [0,1,2])
+ adaopy.setBackgroundError (ScalarSparseMatrix = 1.)
+ adaopy.setObservation (Vector = [0.5,1.5,2.5])
+ adaopy.setObservationError (DiagonalSparseMatrix = "1 2 3")
+ adaopy.setObservationOperator(OneFunction = H)
+ adaopy.setObserver("Analysis",Template="ValuePrinter")
+ adaopy.execute()
+ Xa[algo] = adaopy.get("Analysis")[-1]
+ del adaopy
+ #
+ print("")
+ msg = "Tests des ecarts attendus :"
+ print(msg+"\n"+"="*len(msg))
+ verify_similarity_of_algo_results(("3DVAR", "Blue", "ExtendedBlue", "4DVAR", "DerivativeFreeOptimization"), Xa, 5.e-5)
+ verify_similarity_of_algo_results(("KalmanFilter", "ExtendedKalmanFilter", "UnscentedKalmanFilter"), Xa, 2.e-14)
+ verify_similarity_of_algo_results(("KalmanFilter", "EnsembleKalmanFilter"), Xa, 2e-1)
+ print(" Les resultats obtenus sont corrects.")
+ print("")
+
+def almost_equal_vectors(v1, v2, precision = 1.e-15, msg = ""):
+ """Comparaison de deux vecteurs"""
+ print(" Difference maximale %s: %.2e"%(msg, max(abs(v2 - v1))))
+ return max(abs(v2 - v1)) < precision
+
+def verify_similarity_of_algo_results(serie = [], Xa = {}, precision = 1.e-15):
+ print(" Comparaisons :")
+ for algo1 in serie:
+ for algo2 in serie:
+ if algo1 is algo2: break
+ assert almost_equal_vectors( Xa[algo1], Xa[algo2], precision, "entre %s et %s "%(algo1, algo2) )
+ print(" Algorithmes dont les resultats sont similaires a %.0e : %s\n"%(precision, serie,))
+ sys.stdout.flush()
#===============================================================================
if __name__ == "__main__":
- print '\n AUTODIAGNOSTIC \n'
- test1()
+ print("\nAUTODIAGNOSTIC\n==============")
+ sys.stderr = sys.stdout
+ unittest.main(verbosity=2)
