Merge pull request #300 from AdityaSavara/develop

Develop

Examples/Example00/runfile_Example_00c10_BPE_uninformed_MultiStart_ESS.py

@@ -0,0 +1,47 @@
+import sys; sys.path.insert(0, '../../');  import PEUQSE as PEUQSE
+import PEUQSE.UserInput as UserInput
+
+if __name__ == "__main__":
+    import observed_values_00  #Just a simple example. The user can also put the values in directly into the runfile or extract from a csv, for example.
+    import simulation_model_00 #Simple example.
+
+    UserInput.responses['responses_abscissa'] = observed_values_00.observed_data_x_values
+    UserInput.responses['responses_observed'] = observed_values_00.observed_data_y_values
+    UserInput.responses['responses_observed_uncertainties'] = observed_values_00.observed_data_y_values_uncertainties
+
+
+    UserInput.simulated_response_plot_settings['x_label'] = 'distance (m)'
+    UserInput.simulated_response_plot_settings['y_label'] = r'$time (s)$'
+
+
+    UserInput.model['parameterNamesAndMathTypeExpressionsDict'] = {'a':'a','b':'b'}
+    UserInput.model['InputParameterPriorValues'] = [200, 500] #prior expected values for a and b
+    UserInput.model['InputParametersPriorValuesUncertainties'] = [-1, -1] #required. #If user wants to use a prior with covariance, then this must be a 2D array/ list. To assume no covariance, a 1D
+    UserInput.model['InputParameterPriorValues_upperBounds'] = [1E6, 1E6] 
+    UserInput.model['InputParameterPriorValues_lowerBounds'] = [-1E6, -1E6]
+    UserInput.parameter_estimation_settings['scaling_uncertainties_type'] = "off"
+    UserInput.parameter_estimation_settings['mcmc_random_seed'] = 0
+        #UserInput.model['InputParameterInitialGuess'] = [150,400] #Can optionally change the initial guess to be different from prior means.
+
+    UserInput.model['simulateByInputParametersOnlyFunction'] = simulation_model_00.simulation_function_wrapper #This must simulate with *only* the parameters listed above, and no other arguments.
+    #UserInput.parameter_estimation_settings['mcmc_burn_in'] = 10000
+    #UserInput.parameter_estimation_settings['mcmc_length'] = 100000 #The uninformed prior int his example has a "bad" MCMC walker so requires lots of sampling to converge.
+    UserInput.parameter_estimation_settings['multistart_checkPointFrequency'] = 10000 #This example is long enough that it's good to get updates.
+
+    UserInput.parameter_estimation_settings['multistart_searchType'] = 'doEnsembleSliceSampling'
+    UserInput.parameter_estimation_settings['multistart_initialPointsDistributionType'] = 'uniform'
+    UserInput.parameter_estimation_settings['multistart_exportLog'] = True
+    UserInput.parameter_estimation_settings['multistart_gridsearch_threshold_filter_coefficient'] = 2.0 #The lower this is, the more the points become filtered. It is not recommended to go below 2.0.
+    UserInput.parameter_estimation_settings['multistart_numStartPoints'] = 30
+    UserInput.parameter_estimation_settings['mcmc_length'] = 1000
+    UserInput.parameter_estimation_settings['multistart_relativeInitialDistributionSpread'] = 0.50
+
+
+    #After making the UserInput, now we make a 'parameter_estimation' object from it.
+    PE_object = PEUQSE.parameter_estimation(UserInput)
+
+    PE_object.doMultiStart()
+    PE_object.createAllPlots() #This function calls each of the below functions so that the user does not have to.
+#    PE_object.makeHistogramsForEachParameter()    
+#    PE_object.makeSamplingScatterMatrixPlot()
+#    PE_object.createSimulatedResponsesPlots()

Examples/Example00/runfile_Example_00c11_BPE_uninformed_MultiStart_EJS.py

@@ -0,0 +1,47 @@
+import sys; sys.path.insert(0, '../../');  import PEUQSE as PEUQSE
+import PEUQSE.UserInput as UserInput
+
+if __name__ == "__main__":
+    import observed_values_00  #Just a simple example. The user can also put the values in directly into the runfile or extract from a csv, for example.
+    import simulation_model_00 #Simple example.
+
+    UserInput.responses['responses_abscissa'] = observed_values_00.observed_data_x_values
+    UserInput.responses['responses_observed'] = observed_values_00.observed_data_y_values
+    UserInput.responses['responses_observed_uncertainties'] = observed_values_00.observed_data_y_values_uncertainties
+
+
+    UserInput.simulated_response_plot_settings['x_label'] = 'distance (m)'
+    UserInput.simulated_response_plot_settings['y_label'] = r'$time (s)$'
+
+
+    UserInput.model['parameterNamesAndMathTypeExpressionsDict'] = {'a':'a','b':'b'}
+    UserInput.model['InputParameterPriorValues'] = [200, 500] #prior expected values for a and b
+    UserInput.model['InputParametersPriorValuesUncertainties'] = [-1, -1] #required. #If user wants to use a prior with covariance, then this must be a 2D array/ list. To assume no covariance, a 1D
+    UserInput.model['InputParameterPriorValues_upperBounds'] = [1E6, 1E6] 
+    UserInput.model['InputParameterPriorValues_lowerBounds'] = [-1E6, -1E6]
+    UserInput.parameter_estimation_settings['scaling_uncertainties_type'] = "off"
+    UserInput.parameter_estimation_settings['mcmc_random_seed'] = 0
+        #UserInput.model['InputParameterInitialGuess'] = [150,400] #Can optionally change the initial guess to be different from prior means.
+
+    UserInput.model['simulateByInputParametersOnlyFunction'] = simulation_model_00.simulation_function_wrapper #This must simulate with *only* the parameters listed above, and no other arguments.
+    #UserInput.parameter_estimation_settings['mcmc_burn_in'] = 10000
+    #UserInput.parameter_estimation_settings['mcmc_length'] = 100000 #The uninformed prior int his example has a "bad" MCMC walker so requires lots of sampling to converge.
+    UserInput.parameter_estimation_settings['multistart_checkPointFrequency'] = 10000 #This example is long enough that it's good to get updates.
+
+    UserInput.parameter_estimation_settings['multistart_searchType'] = 'doEnsembleJumpSampling'
+    UserInput.parameter_estimation_settings['multistart_initialPointsDistributionType'] = 'uniform'
+    UserInput.parameter_estimation_settings['multistart_exportLog'] = True
+    UserInput.parameter_estimation_settings['multistart_gridsearch_threshold_filter_coefficient'] = 2.0 #The lower this is, the more the points become filtered. It is not recommended to go below 2.0.
+    UserInput.parameter_estimation_settings['multistart_numStartPoints'] = 30
+    UserInput.parameter_estimation_settings['mcmc_length'] = 1000
+    UserInput.parameter_estimation_settings['multistart_relativeInitialDistributionSpread'] = 0.50
+
+
+    #After making the UserInput, now we make a 'parameter_estimation' object from it.
+    PE_object = PEUQSE.parameter_estimation(UserInput)
+
+    PE_object.doMultiStart()
+    PE_object.createAllPlots() #This function calls each of the below functions so that the user does not have to.
+#    PE_object.makeHistogramsForEachParameter()    
+#    PE_object.makeSamplingScatterMatrixPlot()
+#    PE_object.createSimulatedResponsesPlots()

Examples/Example00/runfile_Example_00c9_BPE_uninformed_MultiStart_MH.py

@@ -0,0 +1,47 @@
+import sys; sys.path.insert(0, '../../');  import PEUQSE as PEUQSE
+import PEUQSE.UserInput as UserInput
+
+if __name__ == "__main__":
+    import observed_values_00  #Just a simple example. The user can also put the values in directly into the runfile or extract from a csv, for example.
+    import simulation_model_00 #Simple example.
+
+    UserInput.responses['responses_abscissa'] = observed_values_00.observed_data_x_values
+    UserInput.responses['responses_observed'] = observed_values_00.observed_data_y_values
+    UserInput.responses['responses_observed_uncertainties'] = observed_values_00.observed_data_y_values_uncertainties
+
+
+    UserInput.simulated_response_plot_settings['x_label'] = 'distance (m)'
+    UserInput.simulated_response_plot_settings['y_label'] = r'$time (s)$'
+
+
+    UserInput.model['parameterNamesAndMathTypeExpressionsDict'] = {'a':'a','b':'b'}
+    UserInput.model['InputParameterPriorValues'] = [200, 500] #prior expected values for a and b
+    UserInput.model['InputParametersPriorValuesUncertainties'] = [-1, -1] #required. #If user wants to use a prior with covariance, then this must be a 2D array/ list. To assume no covariance, a 1D
+    UserInput.model['InputParameterPriorValues_upperBounds'] = [1E6, 1E6] 
+    UserInput.model['InputParameterPriorValues_lowerBounds'] = [-1E6, -1E6]
+    UserInput.parameter_estimation_settings['scaling_uncertainties_type'] = "off"
+    UserInput.parameter_estimation_settings['mcmc_random_seed'] = 0
+        #UserInput.model['InputParameterInitialGuess'] = [150,400] #Can optionally change the initial guess to be different from prior means.
+
+    UserInput.model['simulateByInputParametersOnlyFunction'] = simulation_model_00.simulation_function_wrapper #This must simulate with *only* the parameters listed above, and no other arguments.
+    #UserInput.parameter_estimation_settings['mcmc_burn_in'] = 10000
+    #UserInput.parameter_estimation_settings['mcmc_length'] = 100000 #The uninformed prior int his example has a "bad" MCMC walker so requires lots of sampling to converge.
+    UserInput.parameter_estimation_settings['multistart_checkPointFrequency'] = 10000 #This example is long enough that it's good to get updates.
+
+    UserInput.parameter_estimation_settings['multistart_searchType'] = 'doMetropolisHastings'
+    UserInput.parameter_estimation_settings['multistart_initialPointsDistributionType'] = 'uniform'
+    UserInput.parameter_estimation_settings['multistart_exportLog'] = True
+    UserInput.parameter_estimation_settings['multistart_gridsearch_threshold_filter_coefficient'] = 2.0 #The lower this is, the more the points become filtered. It is not recommended to go below 2.0.
+    UserInput.parameter_estimation_settings['multistart_numStartPoints'] = 30
+    UserInput.parameter_estimation_settings['mcmc_length'] = 1000
+    UserInput.parameter_estimation_settings['multistart_relativeInitialDistributionSpread'] = 0.50
+
+
+    #After making the UserInput, now we make a 'parameter_estimation' object from it.
+    PE_object = PEUQSE.parameter_estimation(UserInput)
+
+    PE_object.doMultiStart()
+    PE_object.createAllPlots() #This function calls each of the below functions so that the user does not have to.
+#    PE_object.makeHistogramsForEachParameter()    
+#    PE_object.makeSamplingScatterMatrixPlot()
+#    PE_object.createSimulatedResponsesPlots()

Examples/Example00/runfile_Example_00d3_BPE_EJS_continuedSamplingWithExternalFile.py

@@ -0,0 +1,45 @@
+import sys; sys.path.insert(0, '../../');  import PEUQSE as PEUQSE
+import PEUQSE.UserInput as UserInput
+
+if __name__ == "__main__":
+    """
+    This file is used to show how to continue sampling without keeping the previous data. 
+    This file works in conjunction with Example_00d4.
+    The example will start with a EJS mcmc run then use ESS as the second sampler in the d4 file. 
+    The number of walkers should be the same between runs for the last points to continue as the next sampling starting points. 
+    """
+    import observed_values_00  #Just a simple example. The user can also put the values in directly into the runfile or extract from a csv, for example.
+    import simulation_model_00 #Simple example.
+
+    UserInput.responses['responses_abscissa'] = observed_values_00.observed_data_x_values
+    UserInput.responses['responses_observed'] = observed_values_00.observed_data_y_values
+    UserInput.responses['responses_observed_uncertainties'] = observed_values_00.observed_data_y_values_uncertainties
+
+
+    UserInput.simulated_response_plot_settings['x_label'] = 'distance (m)'
+    UserInput.simulated_response_plot_settings['y_label'] = r'$time (s)$'
+    UserInput.simulated_response_plot_settings['fontdict'] = {'size':16}
+
+
+    UserInput.model['parameterNamesAndMathTypeExpressionsDict'] = {'a':'a','b':'b'}
+    UserInput.model['InputParameterPriorValues'] = [200, 500] #prior expected values for a and b
+    UserInput.model['InputParametersPriorValuesUncertainties'] = [100, 200] #required. #If user wants to use a prior with covariance, then this must be a 2D array/ list. To assume no covariance, a 1D
+    #UserInput.model['InputParameterInitialGuess'] = [150,400] #Can optionally change the initial guess to be different from prior means.
+    UserInput.parameter_estimation_settings['mcmc_nwalkers'] = 10
+
+    UserInput.model['simulateByInputParametersOnlyFunction'] = simulation_model_00.simulation_function_wrapper #This must simulate with *only* the parameters listed above, and no other arguments.
+
+
+    UserInput.parameter_estimation_settings['mcmc_threshold_filter_samples'] = True
+
+    UserInput.parameter_estimation_settings['mcmc_random_seed'] = None
+
+    UserInput.parameter_estimation_settings['mcmc_continueSampling'] = False # this example does not truly continue sampling since the old points are not saved.
+    #After making the UserInput, now we make a 'parameter_estimation' object from it.
+    PE_object = PEUQSE.parameter_estimation(UserInput)
+
+    PE_object.doEnsembleJumpSampling() #one can also use PE_object.doMetropolisHastings(continueSampling = True)
+    PE_object.createAllPlots() #This function calls each of the below functions so that the user does not have to.
+#    PE_object.makeHistogramsForEachParameter()    
+#    PE_object.makeSamplingScatterMatrixPlot()
+#    PE_object.createSimulatedResponsesPlots()

Examples/Example00/runfile_Example_00d4_BPE_ESS_continuedSamplingWithExternalFile.py

@@ -0,0 +1,47 @@
+import sys; sys.path.insert(0, '../../');  import PEUQSE as PEUQSE
+import PEUQSE.UserInput as UserInput
+
+if __name__ == "__main__":
+    """
+    This file is used to show how to continue sampling without keeping the previous data. 
+    This file works in conjunction with Example_00d4.
+    The example will start with a EJS mcmc run then use ESS as the second sampler in the d4 file. 
+    The number of walkers should be the same between runs for the last points to continue as the next sampling starting points. 
+    """
+    import observed_values_00  #Just a simple example. The user can also put the values in directly into the runfile or extract from a csv, for example.
+    import simulation_model_00 #Simple example.
+
+    UserInput.responses['responses_abscissa'] = observed_values_00.observed_data_x_values
+    UserInput.responses['responses_observed'] = observed_values_00.observed_data_y_values
+    UserInput.responses['responses_observed_uncertainties'] = observed_values_00.observed_data_y_values_uncertainties
+
+
+    UserInput.simulated_response_plot_settings['x_label'] = 'distance (m)'
+    UserInput.simulated_response_plot_settings['y_label'] = r'$time (s)$'
+    UserInput.simulated_response_plot_settings['fontdict'] = {'size':16}
+
+
+    UserInput.model['parameterNamesAndMathTypeExpressionsDict'] = {'a':'a','b':'b'}
+    UserInput.model['InputParameterPriorValues'] = [200, 500] #prior expected values for a and b
+    UserInput.model['InputParametersPriorValuesUncertainties'] = [100, 200] #required. #If user wants to use a prior with covariance, then this must be a 2D array/ list. To assume no covariance, a 1D
+    #UserInput.model['InputParameterInitialGuess'] = [150,400] #Can optionally change the initial guess to be different from prior means.
+    UserInput.parameter_estimation_settings['mcmc_nwalkers'] = 10
+
+    UserInput.model['simulateByInputParametersOnlyFunction'] = simulation_model_00.simulation_function_wrapper #This must simulate with *only* the parameters listed above, and no other arguments.
+
+
+    UserInput.parameter_estimation_settings['mcmc_threshold_filter_samples'] = True
+
+    UserInput.parameter_estimation_settings['mcmc_random_seed'] = None
+
+    UserInput.parameter_estimation_settings['mcmc_continueSampling'] = False # this example does not truly continue sampling since the old points are not saved.
+    # inputting a string into Initial guess, a pickle file is assumed. It will check if the file is in the cwd or in the pickles folder.
+    UserInput.model['InputParameterInitialGuess'] = 'mcmc_last_point_sampled.pkl' 
+    #After making the UserInput, now we make a 'parameter_estimation' object from it.
+    PE_object = PEUQSE.parameter_estimation(UserInput)
+
+    PE_object.doEnsembleSliceSampling() 
+    PE_object.createAllPlots() #This function calls each of the below functions so that the user does not have to.
+#    PE_object.makeHistogramsForEachParameter()    
+#    PE_object.makeSamplingScatterMatrixPlot()
+#    PE_object.createSimulatedResponsesPlots()