From: Eric Fayolle <eric.fayolle@edf.fr>
Date: Mon, 10 Oct 2022 14:40:02 +0000 (+0200)
Subject: Fusion des fonctionnalités Sortie Panda + Lancement fenêtré
X-Git-Tag: merge_uncertainty_odysee_1210~2
X-Git-Url: http://git.salome-platform.org/gitweb/?a=commitdiff_plain;h=61a900ee85d66be5483eafcde12135ca63fa67c7;p=tools%2Feficas.git

Fusion des fonctionnalités Sortie Panda + Lancement fenêtré
---

61a900ee85d66be5483eafcde12135ca63fa67c7
diff --cc generator/generator_UQ.py
index 69200c5f,ec984e76..506bae17
--- a/generator/generator_UQ.py
+++ b/generator/generator_UQ.py
@@@ -469,11 -411,16 +469,17 @@@ class UQGenerator(PythonGenerator)
          # print('txtResultCT:',txtResultCT)
          optionalResultList=getStrVarList(optionalResultNames)
          optionalResultStrList=getStrVarStrList(optionalResultNames)
+         post_csv_rnScript=os.path.basename(self.ScriptPosttraitement).split('.')[0]
+         post_csv_rnPath=os.path.dirname(self.ScriptPosttraitement)
+         print (post_csv_rnScript)
+         print (post_csv_rnPath)
          txtPrintResult=printResult.format(
+             post_csv_rnScript = post_csv_rnScript,
+             post_csv_rnPath = post_csv_rnPath,
              optionalResultList = optionalResultList,
              optionalResultStrList = optionalResultStrList,
 -            resultSkList = getStrVarList(self.resultSkList))
 +            resultSkList = getStrVarList(self.resultSkList),
 +            Uncertain_inputs = self.creeTexteInputVariablesSummary(self.indent1) )
          txtResult= txtResultCT.format( optionalResult=result )
          txtResult+=txtPrintResult
  
diff --cc generator/textePersalys.py
index e1e698b8,44317a2e..928d53ca
--- a/generator/textePersalys.py
+++ b/generator/textePersalys.py
@@@ -223,53 -217,13 +223,55 @@@ if __name__ == '__main__'
  #    inputs = [a, b]
  #    outputs = [d]
  
 -printResult="""    import numpy as np
 +# empiricalMean = ot.Sample(result.getMean())[nInputs:]
 +# empiricalStandardDeviation = ot.Sample(result.getStandardDeviation())[nInputs:]
 +# resultNp.reshape(nbindex,nbsorties)
 +
 +printResultMC="""    import numpy as np
-     from post_csv_rn import write_result_from_persalys
+     sys.path.append(os.path.abspath('{post_csv_rnPath}'))
+     from {post_csv_rnScript} import write_result_from_persalys
  
 -    #from post_csv_rn import write_result_from_persalys
 -    # import pandas as pnd
 -    # from functools import reduce
 +    #Noms de lignes
 +    #index = ['meanFirstOrder', 'meanSecondOrderOrder', 'standardDeviationFirstOrder' ]
 +    indexNp = [ {optionalResultStrList} ]
 +    nResult = len(indexNp)
 +
 +    #resultNp = np.array([meanFirstOrder, meanSecondOrderOrder, standardDeviationFirstOrder ])
 +    #Lignes de résultats demandés pour toutes les fcts d'aggrégation en colonne
 +    resultNp = np.array([ {optionalResultList} ])
 +    resultNp = resultNp.reshape(nResult, nOutputs)  #En MC à cause du ot.Sample( PointsCollection) il y a un niveau de liste en trop
 +
 +    #Tableau skyline localisant les variables (hors fctAgg) dans le tableau de résultat global  
 +    # ex pour deux variables et respectivement 7 et 3 fcts aggrégation :  resultSk = [0,7,10]
 +    #nOutVar = len(resSk)-1
 +    resSk = [ {resultSkList} ]
 +
 +    print('\\n\\n')
 +    print('*********************************************************************************\\n')
 +    print('                   UNCERTAINTY QUANTIFICATION RESULTS\\n')
 +   
 +    print(head_study,'\\n')
 +
 +    print('Uncertain inputs list :','\\n')
 +    print(
 +{Uncertain_inputs}'\\n'
 +    )
 +    print('\\nElapsed Time : ', centralTendency.getElapsedTime(),'\\n') #Monte Carlo ou Taylor
 +
 +    print('\\nDesign of Experiment :')
 +    print(result.getDesignOfExperiment().getInputSample(),'\\n') #TODO: générer un fichier csv
 +
 +    print('\\nCoefficient of Variation :')
 +    print(ot.Sample(result.getCoefficientOfVariation())[nInputs:],'\\n') #TODO: générer un fichier csv
 +    write_result_from_persalys(resultNp, indexNp, resSk, outputs)
 +
 +    print('\\n*********************************************************************************\\n')
 +"""
 +
 +
 +printResultTaylor="""    import numpy as np
-     from post_csv_rn import write_result_from_persalys
++    sys.path.append(os.path.abspath('{post_csv_rnPath}'))
++    from {post_csv_rnScript} import write_result_from_persalys
  
      #resultNp = np.array([meanFirstOrder, meanSecondOrderOrder, standardDeviationFirstOrder ])
      #Lignes de résultats demandés pour toutes les fcts d'aggrégation en colonne
@@@ -277,35 -231,41 +279,36 @@@
  
      #Noms de lignes
      #index = ['meanFirstOrder', 'meanSecondOrderOrder', 'standardDeviationFirstOrder' ]
 +    #nResult = len(indexNp)
      indexNp = [ {optionalResultStrList} ]
  
+     #nResult = len(indexNp)
  
      #Tableau skyline localisant les variables (hors fctAgg) dans le tableau de résultat global  
 -    #resultSk = [0,7,10]
 +    # ex pour deux variables et respectivement 7 et 3 fcts aggrégation :  resultSk = [0,7,10]
 +    #nOutVar = len(resSk)-1
      resSk = [ {resultSkList} ]
  
 -    #nOutVar = len(resSk)-1
 +    print('\\n\\n')
 +    print('*********************************************************************************\\n')
 +    print('                   UNCERTAINTY QUANTIFICATION RESULTS\\n')
 +   
 +    print(head_study,'\\n')
  
 -    write_result_from_persalys(resultNp, indexNp, resSk, outputs)
 +    print('Uncertain inputs list :','\\n')
 +    print(
 +{Uncertain_inputs}'\\n'
 +    )
 +    print('\\nElapsed Time : ', centralTendency.getElapsedTime(),'\\n') #Monte Carlo ou Taylor
  
 -    # print('\\n')
 -    # for i in range(nOutVar):
 -    #     # Récupère les couples (fctAggrégation,Nom de variable ss fct Agg)
 -    #     #  des champs de description des sorties.
 -    #     #  Les sorties concernées proviennent de la section de résultatNp
 -    #     #  qui doit correspondre à une même variable hors fct Agg
 -    #     fctAgg_outVarName = list( map(lambda o: eval(o.getDescription()),
 -    #                                   outputs[resSk[i]:resSk[i+1]]) )
 -    #     outVarName        = fctAgg_outVarName[0][1]
 -    #     checkOutVarName   = reduce( lambda n1,n2: n1 == n2,
 -    #                                 [True]+list( map(lambda o: o[1] == outVarName,
 -    #                                           fctAgg_outVarName) ) )
 -    #     assert(checkOutVarName == True)
 -    #     print(outVarName)
 -    #     columns  = list(map(lambda o1: o1[0], fctAgg_outVarName) )
 -    #     resultDf = pnd.DataFrame(resultNp[:,resSk[i]:(resSk[i+1])],
 -    #                             index = indexNp,
 -    #                             columns = columns )
 -    #     print(resultDf,'\\n')
 -    #     #name_csv = str.replace(str.replace(outVarName,"@","_"),' ','-')
 -    #     name_csv = outVarName
 -    #     resultDf.to_csv(name_csv+'-uncertainty.csv')
 +    # print('\\nDesign of Experiment :')
 +    # print(result.getDesignOfExperiment().getInputSample(),'\\n') #TODO: Activer uniquement en MC + fichier csv
  
 +    # print('\\nCoefficient of Variation :')
 +    # print(result.getCoefficientOfVariation(),'\\n') #TODO: Activer uniquement en MC + fichier csv
 +    write_result_from_persalys(resultNp, indexNp, resSk, outputs)
 +
 +    print('\\n*********************************************************************************\\n')
  """
  
  ## Tendance Centrale Taylor