17 | The UQ Toolkit (UQTk) is a collection of libraries and tools for the quantification of uncertainty in numerical model predictions. Version 3.1.x offers intrusive and non-intrusive methods for propagating input uncertainties through computational models, tools for sensitivity analysis, methods for sparse surrogate construction, and Bayesian inference tools for inferring parameters from experimental data.
18 |
19 | For detailed information, please refer to the UQTk Manual.
20 |
21 | Explore the C++ source code documentation at UQTk C++ Doxygen Documentation.
22 |
23 | Visit the official UQTK website: https://www.sandia.gov/uqtoolkit/
24 |
25 |
26 |
27 |
28 |
--------------------------------------------------------------------------------
/examples/CMakeLists.txt:
--------------------------------------------------------------------------------
1 | project (UQTk)
2 |
3 | link_directories(${CMAKE_SUNDIALS_DIR}/lib)
4 | if( BUILD_SUNDIALS)
5 |
6 | link_directories("${PROJECT_BINARY_DIR}/../dep/sundials/src/cvode")
7 | link_directories("${PROJECT_BINARY_DIR}/../dep/sundials/src/nvector/serial")
8 | link_directories("${PROJECT_BINARY_DIR}/../dep/sundials/src/sunlinsol/dense")
9 | link_directories("${PROJECT_BINARY_DIR}/../dep/sundials/src/sunmatrix/dense")
10 | endif()
11 |
12 | add_subdirectory (ops)
13 | add_subdirectory (kle_ex1)
14 | add_subdirectory (line_infer)
15 | add_subdirectory (tmcmc_bimodal)
16 | add_subdirectory (tmcmc_umbridge)
17 | add_subdirectory (pce_bcs)
18 | add_subdirectory (surf_rxn)
19 | add_subdirectory (uqpc)
20 | add_subdirectory (sensMC)
21 | add_subdirectory (num_integ)
22 | add_subdirectory (heat_transfer_window)
23 | add_subdirectory (polynomial)
24 | add_subdirectory (d_spring_series)
25 | add_subdirectory (iuq)
26 | add_subdirectory (dfi)
27 | add_subdirectory (surrogate_genz)
28 | add_subdirectory (dfi_app)
29 | add_subdirectory (bare_bcs)
30 |
31 | # Add muq example only with the gnu compilers
32 | if ("${CMAKE_CXX_COMPILER_ID}" STREQUAL "GNU" AND EXISTS ${PATH2MUQ})
33 | # message(STATUS "Adding MUQ<->UQTk example")
34 | #add_subdirectory (muq)
35 | endif()
36 |
37 | # Copy over README file too
38 | INSTALL(FILES README
39 | PERMISSIONS OWNER_EXECUTE OWNER_WRITE OWNER_READ
40 | DESTINATION examples/)
41 |
--------------------------------------------------------------------------------
/examples/bare_bcs/CMakeLists.txt:
--------------------------------------------------------------------------------
1 | project (UQTk)
2 |
3 | SET(copy_EX_FILES
4 | bare_bcs.ipynb
5 | readme.md)
6 |
7 | INSTALL(FILES ${copy_EX_FILES}
8 | PERMISSIONS OWNER_WRITE OWNER_READ
9 | DESTINATION examples/bare_bcs)
10 |
--------------------------------------------------------------------------------
/examples/bare_bcs/readme.md:
--------------------------------------------------------------------------------
1 | # Example that calls BCS directly on an arbitrary set of basis functions
2 |
3 | The basis functions do not need to be part of a PCE Set
4 | So this is more general than the wiBCS interface that is set up specifically for PCEs
5 |
--------------------------------------------------------------------------------
/examples/d_spring_series/CMakeLists.txt:
--------------------------------------------------------------------------------
1 | project (UQTk)
2 |
3 | SET(copy_PY_FILES
4 | d_springs_tools.py
5 | run_d_springs.py
6 | README)
7 |
8 | INSTALL(FILES ${copy_PY_FILES}
9 | PERMISSIONS OWNER_EXECUTE OWNER_WRITE OWNER_READ
10 | DESTINATION examples/d_spring_series)
11 |
--------------------------------------------------------------------------------
/examples/d_spring_series/README:
--------------------------------------------------------------------------------
1 | d_spring_series example
2 |
3 |
4 | This examples contains several Python scripts that that propagate uncertainty
5 | in input parameters through a series springs model using basis adaptation
6 | approach, and is compared with Monte Carlo sampling method and non- intrusive
7 | spectral projection (NISP) via sparse quadrature method.
8 |
9 | For more information, look in UQTk manual under
10 | Forward Propagation of Uncertainty Using Basis Adaptation
11 |
12 | ===============================================================================
13 | FILES:
14 |
15 | run_d_springs.py: main script
16 |
17 | d_springs_tools.py: functions called by run_d_springs.py
18 |
--------------------------------------------------------------------------------
/examples/dfi/CMakeLists.txt:
--------------------------------------------------------------------------------
1 | project (UQTk)
2 |
3 | add_subdirectory (line)
4 |
5 | SET(copy_DOC_FILES
6 | README)
7 |
8 | INSTALL(FILES ${copy_DOC_FILES}
9 | PERMISSIONS OWNER_WRITE OWNER_READ
10 | DESTINATION examples/dfi)
11 |
--------------------------------------------------------------------------------
/examples/dfi/README:
--------------------------------------------------------------------------------
1 | ==============================
2 | dfi
3 | ==============================
4 | use data-free inference to recover an approximate correlated posterior
5 | distribution for parameters of a line model
6 | given summary statistics on marginals
7 |
8 | To run this example, compile it via "make" in the line directory, then
9 | run it using: "./dfi_line.exe datainference"
10 |
11 | Notes:
12 | 1) this example takes about 1 hour to run, and sends a lot of data to the screen.
13 | 2) if your UQTk build links against external libraries (e.g. SUNDIALS), you will
14 | have to make sure that your Makefile
15 | can find both those libraries and their include files. If the make fails,
16 | you may need to add the path to the cvode.h include
17 | files using a -I path_to_include_folder and the path to the external library
18 | through a -L path_to_external_library statement.
19 |
--------------------------------------------------------------------------------
/examples/dfi/line/CMakeLists.txt:
--------------------------------------------------------------------------------
1 | project (UQTk)
2 |
3 | SET(copy_FILES
4 | main.cpp
5 | Makefile
6 | userFunctions_line.cpp
7 | dfi.input
8 | )
9 |
10 | INSTALL(
11 | FILES ${copy_FILES}
12 | PERMISSIONS OWNER_EXECUTE OWNER_WRITE OWNER_READ
13 | DESTINATION examples/dfi/line
14 | )
15 |
16 |
--------------------------------------------------------------------------------
/examples/dfi/line/Makefile:
--------------------------------------------------------------------------------
1 | SHELL = /bin/bash
2 | TARGET = dfi_line.exe
3 |
4 | UQTK_LIBS= -L$(UQTK_INS)/lib \
5 | -luqtk -luqtkdfi -ldepuqtk -l sundials_cvode -lblas -llapack -lgfortran
6 | UQTK_INCL= -I$(UQTK_INS)/include/uqtk -I$(UQTK_INS)/include/dep -I$(UQTK_INS)/include
7 |
8 | CXX = g++
9 | CFLAGS= -std=c++14
10 | all: $(TARGET)
11 |
12 | userFunctions_line.o: userFunctions_line.cpp
13 | $(info compile the user functions...)
14 | $(CXX) $(CFLAGS) $(UQTK_INCL) -c userFunctions_line.cpp $(UQTK_LIBS)
15 | ar -crs libUserFunc_line.a userFunctions_line.o
16 |
17 | main.o: main.cpp
18 | $(info make the driver...)
19 | $(CXX) $(CFLAGS) $(UQTK_INCL) -c main.cpp $(UQTK_LIBS)
20 |
21 | $(TARGET): userFunctions_line.o main.o
22 | $(info build the dfi executeable...)
23 | $(CXX) $(UQTK_INCL) $(CFLAGS) -o $(TARGET) main.o $(UQTK_LIBS) libUserFunc_line.a
24 | clean:
25 | rm -f *.a *.o $(TARGET)
26 | rm -rf html latex
27 |
--------------------------------------------------------------------------------
/examples/dfi/line/dfi.input:
--------------------------------------------------------------------------------
1 | seed 13
2 | data_space_dimenion 10
3 | data_chain_burnin_samples 10
4 | data_chain_samples 10
5 | target_chain_accept 0.2
6 | parameter_chain_burnin_samples 10
7 | parameter_chain_samples 10
8 | error_optimization_chain_samples 1000
9 | data_chain_propcov_fac 0.1
10 | TEST
11 |
--------------------------------------------------------------------------------
/examples/dfi_app/CMakeLists.txt:
--------------------------------------------------------------------------------
1 | project (UQTk)
2 |
3 | SET(copy_FILES
4 | 1_quadratic_one_data_set.ipynb
5 | 2_quadratic_two_data_sets.ipynb
6 | 3_quadratic_two_data_sets_disagree.ipynb
7 | 4_quadratic_two_data_sets_different_nb_of_points.ipynb
8 | 5_quadratic_with_optimal_beta.ipynb
9 | utils.py
10 | README
11 | )
12 |
13 | INSTALL(FILES ${copy_FILES}
14 | PERMISSIONS OWNER_EXECUTE OWNER_WRITE OWNER_READ
15 | DESTINATION examples/dfi_app)
--------------------------------------------------------------------------------
/examples/dfi_app/utils.py:
--------------------------------------------------------------------------------
1 | import os
2 |
3 | # path to python
4 | path_to_python = "python"
5 |
6 | # path to uq_pc.py
7 | path_to_uq_pc = os.path.join(os.environ.get("UQTK_INS"), "examples", "uqpc", "uq_pc.py")
8 |
9 | # path to the dfi app
10 | path_to_dfi = os.path.join(os.environ.get("UQTK_INS"), "bin", "dfi")
11 |
12 | # settings to save figures if desired
13 | save = False
14 | save_location = "."
--------------------------------------------------------------------------------
/examples/heat_transfer_window/CMakeLists.txt:
--------------------------------------------------------------------------------
1 | project (UQTk)
2 |
3 | SET(copy_PY_FILES
4 | heat_transfer_window.ipynb
5 | window_tools.py
6 | img_window_1.png
7 | img_window_2.png
8 | README)
9 |
10 | INSTALL(FILES ${copy_PY_FILES}
11 | PERMISSIONS OWNER_EXECUTE OWNER_WRITE OWNER_READ
12 | DESTINATION examples/heat_transfer_window)
13 |
--------------------------------------------------------------------------------
/examples/heat_transfer_window/README:
--------------------------------------------------------------------------------
1 | window example
2 |
3 | contains a pair of Python scripts that propagate uncertainty in input parameters
4 | through a heat transfer model using both a Monte Carlo sampling approach
5 | and a non-intrusive spectral projection (NISP) via quadrature methods.
6 |
7 | For more information, look in UQTk manual under
8 | Forward Propagation of Uncertainty with PyUQTk
9 | (at time of writing this page. 72, but might change)
10 |
11 | ===============================================================================
12 | Files
13 |
14 | heat_transfer_window.ipynb: Jupyter notebook that produces a graph comparing PDF's
15 | of heat flux generated using NISP full and sparse quadrature methods
16 | and Monte Carlo sampling methods for three window models.
17 |
18 | window_tools.py: functions called by heat_transfer_window.ipynb
19 |
20 | img_window_1.png: supporting image
21 |
22 | img_window_2.png: supporting image
23 |
24 |
--------------------------------------------------------------------------------
/examples/heat_transfer_window/img_window_1.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/sandialabs/UQTk/00410672f7807dadc5c8bb2c34bfa64bb7762ad1/examples/heat_transfer_window/img_window_1.png
--------------------------------------------------------------------------------
/examples/heat_transfer_window/img_window_2.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/sandialabs/UQTk/00410672f7807dadc5c8bb2c34bfa64bb7762ad1/examples/heat_transfer_window/img_window_2.png
--------------------------------------------------------------------------------
/examples/iuq/CMakeLists.txt:
--------------------------------------------------------------------------------
1 | project (UQTk)
2 |
3 | SET(copy_FILES
4 | workflow.x
5 | prep_model1.x
6 | prep_data1.x
7 | prep_model2.x
8 | prep_data2.x
9 | prep_calib.x
10 | run_infer.x
11 | postp_infer.x
12 | get_postpred.py
13 | get_postsam.py
14 | plot_sens.py
15 | plot_pdfs.py
16 | plot_prpost.py
17 | plot_fit1d.py
18 | plot_shade.py
19 | README
20 | )
21 |
22 |
23 | INSTALL(FILES ${copy_FILES}
24 | PERMISSIONS OWNER_EXECUTE OWNER_WRITE OWNER_READ
25 | DESTINATION examples/iuq)
26 |
--------------------------------------------------------------------------------
/examples/iuq/README:
--------------------------------------------------------------------------------
1 | Surrogate-enabled inverse UQ workflow
2 |
3 | Requirements and Notes:
4 | =======================
5 | Please make sure the following environment variables are set, e.g.
6 | export UQTK_INS=$HOME/research/UQTk-install
7 | export PYTHONPATH="$PYTHONPATH:${UQTK_INS}"
8 |
9 | Create an empty directory, cd there and run workflow.x as an example
10 | Go through the steps of workflow.x
11 | Create your own workflow for problem-specific model ensemble, data generation and problem-specific postprocessing
12 |
13 |
14 |
--------------------------------------------------------------------------------
/examples/kle_ex1/README:
--------------------------------------------------------------------------------
1 | kle_ex1 example
2 |
3 | contains some examples of the construction of 1D and 2D Karhunen-Lo`eve (KL)
4 | expansions of a Gaussian stochastic process, based on sample realizations
5 | of this stochastic process.
6 |
7 | To run the examples, run the commands ./run1D.sh or ./run2D.sh A help menu will display
8 | the command line options. After you specify the proper options from the help menu, the
9 | scripts will run the 1D or 2D examples and created subdirectories with results and plots
10 |
11 | For more information, look in UQTk manual under
12 | Karhunen-Loéve Expansion of a Stochastic Process
13 |
--------------------------------------------------------------------------------
/examples/kle_ex1/data/scus.m:
--------------------------------------------------------------------------------
1 | global tx
2 | global ok
3 | global ks
4 |
5 | tx=load('tx.dat');
6 | ok=load('ok.dat');
7 | ks=load('ks.dat');
8 |
9 | splini=cvt_sample(2,100,100,3,true,1011);
10 | [ r1, seed, it_num, it_diff, energy ] = cvt ( 2, 2048, 100, 3, 3, 1000, 100, 10, 1011, splini);
11 |
--------------------------------------------------------------------------------
/examples/line_infer/README:
--------------------------------------------------------------------------------
1 | line_infer example
2 |
3 |
4 | contains example that the slope and intercept of a line from noisy data using
5 | Bayes’ rule. The C++ libraries are called directly from the driver program.
6 | By changing the likelihood function and the input data, this program can be
7 | tailored to other inference problems.
8 |
9 | For more information, look in UQTk manual under
10 | Bayesian Inference of a Line
11 |
12 | ================================================================================
13 | Files
14 |
15 | line_infer.py: main scripts, run this program to run the example
16 |
17 | line_infer.cpp: C++ code that produces some of the needed functions
18 |
19 | line_infer.xml.templ: template for the xml file needed to run example
20 |
21 | model.h
22 | posterior.h
23 | XMLreader.h
24 |
--------------------------------------------------------------------------------
/examples/muq/README:
--------------------------------------------------------------------------------
1 | muq
2 |
3 | interface between MUQ and UQTk
4 |
5 | This example does not yet have something written in the manual.
6 |
--------------------------------------------------------------------------------
/examples/num_integ/CMakeLists.txt:
--------------------------------------------------------------------------------
1 | project (UQTk)
2 |
3 | SET(copy_PY_FILES
4 | full_quad.py
5 | sparse_quad.py
6 | quad_tools.py
7 | README
8 | )
9 |
10 | INSTALL(
11 | FILES ${copy_PY_FILES}
12 | PERMISSIONS OWNER_EXECUTE OWNER_WRITE OWNER_READ
13 | DESTINATION examples/num_integ
14 | )
15 |
--------------------------------------------------------------------------------
/examples/num_integ/README:
--------------------------------------------------------------------------------
1 | num_integ
2 |
3 | contains a collection of Python scripts that can be used to perform numerical
4 | integration on six Genz functions: oscillatory, exponential, continuous,
5 | Gaussian, corner-peak, and product-peak. Quadrature and Monte Carlo integration
6 | methods are both employed in this example.
7 |
8 | For more information, look in UQTk manual under
9 | Numerical integration
10 |
11 | ================================================================================
12 | Files
13 |
14 | full_quad.py: produces a graph comparing full quadrature and Monte Carlo integration methods
15 |
16 | quad_tools.py: contains all functions needed for quad integration
17 |
18 | sparse_quad.py: produces a graph comparing sparse quadrature and Monte Carlo integration methods.
19 |
--------------------------------------------------------------------------------
/examples/ops/README:
--------------------------------------------------------------------------------
1 | ops example
2 |
3 | C++ code that illustrates various operations on PC variables.
4 | The code is self-documented.
5 |
6 | type "make examples" to compile the code and run some examples. The python
7 | script plots PDFs from sampled PC expansions.
8 |
9 | For more information, look in UQTk manual under
10 | Elementary Operations
11 |
--------------------------------------------------------------------------------
/examples/pce_bcs/README:
--------------------------------------------------------------------------------
1 | pce_bcs example
2 |
3 | construct sparse Polynomial Chaos expansions
4 |
--------------------------------------------------------------------------------
/examples/pce_bcs/pce_bcs.h:
--------------------------------------------------------------------------------
1 | /* =====================================================================================
2 |
3 | The UQ Toolkit (UQTk) version 3.1.5
4 | Copyright (2024) NTESS
5 | https://www.sandia.gov/UQToolkit/
6 | https://github.com/sandialabs/UQTk
7 |
8 | Copyright 2024 National Technology & Engineering Solutions of Sandia, LLC (NTESS).
9 | Under the terms of Contract DE-NA0003525 with NTESS, the U.S. Government
10 | retains certain rights in this software.
11 |
12 | This file is part of The UQ Toolkit (UQTk)
13 |
14 | UQTk is open source software: you can redistribute it and/or modify
15 | it under the terms of BSD 3-Clause License
16 |
17 | UQTk is distributed in the hope that it will be useful,
18 | but WITHOUT ANY WARRANTY; without even the implied warranty of
19 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
20 | BSD 3 Clause License for more details.
21 |
22 | You should have received a copy of the BSD 3 Clause License
23 | along with UQTk. If not, see https://choosealicense.com/licenses/bsd-3-clause/.
24 |
25 | Questions? Contact the UQTk Developers at https://github.com/sandialabs/UQTk/discussions
26 | Sandia National Laboratories, Livermore, CA, USA
27 | ===================================================================================== */
28 | #ifndef PCEBCS_H
29 | #define PCEBCS_H
30 |
31 | void func(Array2D