Mads.jl

Matrix Factorization using Levenberg Marquardt

Methods:

• Mads.MFlm(X::AbstractMatrix{T}, nk::Integer; method, log_W, log_H, retries, initW, initH, tolX, tolG, tolOF, tolOFcount, minOF, maxEval, maxIter, maxJacobians, lambda, lambda_mu, np_lambda, show_trace, quiet) where T<:Number : /home/travis/build/madsjulia/Mads.jl/src/MadsBlindSourceSeparation.jl:136
• Mads.MFlm(X::AbstractMatrix{T}, range::AbstractRange{Int64}; kw...) where T<:Number : /home/travis/build/madsjulia/Mads.jl/src/MadsBlindSourceSeparation.jl:103

Arguments:

• X::AbstractMatrix{T} : matrix to factorize
• nk::Integer : number of features to extract
• range::AbstractRange{Int64}

Keywords:

• initH : initial H (feature) matrix
• initW : initial W (weight) matrix
• lambda
• lambda_mu
• log_H : log-transform H (feature) matrix[default=false]
• log_W : log-transform W (weight) matrix [default=false]
• maxEval
• maxIter
• maxJacobians
• method
• minOF
• np_lambda
• quiet
• retries : number of solution retries [default=1]
• show_trace
• tolG : parameter space update tolerance [default=1e-6]
• tolOF : objective function update tolerance [default=1e-3]
• tolOFcount : number of Jacobian runs with small objective function change [default=5]
• tolX

Returns:

• NMF results

Non-negative Matrix Factorization using JuMP/Ipopt

Methods:

• Mads.NMFipopt(X::AbstractMatrix, nk::Integer) : /home/travis/build/madsjulia/Mads.jl/src/MadsBlindSourceSeparation.jl:60
• Mads.NMFipopt(X::AbstractMatrix, nk::Integer, retries::Integer; random, maxiter, maxguess, initW, initH, verbosity, quiet) : /home/travis/build/madsjulia/Mads.jl/src/MadsBlindSourceSeparation.jl:60

Arguments:

• X::AbstractMatrix : matrix to factorize
• nk::Integer : number of features to extract
• retries::Integer : number of solution retries [default=1]

Keywords:

• initH : initial H (feature) matrix
• initW : initial W (weight) matrix
• maxguess : guess about the maximum for the H (feature) matrix [default=1]
• maxiter : maximum number of iterations [default=100000]
• quiet : quiet [default=false]
• random : random initial guesses [default=false]
• verbosity : verbosity output level [default=0]

Returns:

• NMF results

Non-negative Matrix Factorization using NMF

Methods:

• Mads.NMFm(X::Array, nk::Integer) : /home/travis/build/madsjulia/Mads.jl/src/MadsBlindSourceSeparation.jl:21
• Mads.NMFm(X::Array, nk::Integer, retries::Integer; tol, maxiter) : /home/travis/build/madsjulia/Mads.jl/src/MadsBlindSourceSeparation.jl:21

Arguments:

• X::Array : matrix to factorize
• nk::Integer : number of features to extract
• retries::Integer : number of solution retries [default=1]

Keywords:

• maxiter : maximum number of iterations [default=10000]
• tol : solution tolerance [default=1.0e-9]

Returns:

• NMF results

Add a keyword in a class within the Mads dictionary madsdata

Methods:

• Mads.addkeyword!(madsdata::AbstractDict, class::AbstractString, keyword::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsHelpers.jl:314
• Mads.addkeyword!(madsdata::AbstractDict, keyword::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsHelpers.jl:310

Arguments:

• class::AbstractString : dictionary class; if not provided searches for keyword in Problem class
• keyword::AbstractString : dictionary key
• madsdata::AbstractDict : MADS problem dictionary

Add an additional contamination source

Methods:

• Mads.addsource!(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsAnasol.jl:17
• Mads.addsource!(madsdata::AbstractDict, sourceid::Int64; dict) : /home/travis/build/madsjulia/Mads.jl/src/MadsAnasol.jl:17

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• sourceid::Int64 : source id [default=0]

Keywords:

• dict

Add contaminant source parameters

Methods:

• Mads.addsourceparameters!(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsAnasol.jl:74

Arguments:

• madsdata::AbstractDict : MADS problem dictionary

Turn off all the wells in the MADS problem dictionary

Methods:

• Mads.allwellsoff!(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:618

Arguments:

• madsdata::AbstractDict : MADS problem dictionary

Turn on all the wells in the MADS problem dictionary

Methods:

• Mads.allwellson!(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:560

Arguments:

• madsdata::AbstractDict : MADS problem dictionary

Execute Amanzi external groundwater flow and transport simulator

Methods:

• Mads.amanzi(filename::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsSimulators.jl:12
• Mads.amanzi(filename::AbstractString, nproc::Int64) : /home/travis/build/madsjulia/Mads.jl/src/MadsSimulators.jl:12
• Mads.amanzi(filename::AbstractString, nproc::Int64, quiet::Bool) : /home/travis/build/madsjulia/Mads.jl/src/MadsSimulators.jl:12
• Mads.amanzi(filename::AbstractString, nproc::Int64, quiet::Bool, observation_filename::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsSimulators.jl:12
• Mads.amanzi(filename::AbstractString, nproc::Int64, quiet::Bool, observation_filename::AbstractString, setup::AbstractString; amanzi_exe) : /home/travis/build/madsjulia/Mads.jl/src/MadsSimulators.jl:12

Arguments:

• filename::AbstractString : amanzi input file name
• nproc::Int64 : number of processor to be used by Amanzi [default=Mads.nprocs_per_task_default]
• observation_filename::AbstractString : Amanzi observation file name [default="observations.out"]
• quiet::Bool : suppress output [default=Mads.quiet]
• setup::AbstractString : bash script to setup Amanzi environmental variables [default="source-amanzi-setup"]

Keywords:

• amanzi_exe : full path to the Amanzi executable

Parse Amanzi output provided in an external file (filename)

Methods:

• Mads.amanzi_output_parser() : /home/travis/build/madsjulia/Mads.jl/src/MadsParsers.jl:20
• Mads.amanzi_output_parser(filename::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsParsers.jl:20

Arguments:

• filename::AbstractString : external file name [default="observations.out"]

Returns:

• dictionary with model observations following MADS requirements

Example:

Mads.amanzi_output_parser()
Mads.amanzi_output_parser("observations.out")

Arcsine transformation of model parameters

Methods:

• Mads.asinetransform(madsdata::AbstractDict, params::AbstractVector) : /home/travis/build/madsjulia/Mads.jl/src/MadsSineTransformations.jl:1
• Mads.asinetransform(params::AbstractVector, lowerbounds::AbstractVector, upperbounds::AbstractVector, indexlogtransformed::AbstractVector) : /home/travis/build/madsjulia/Mads.jl/src/MadsSineTransformations.jl:11

Arguments:

• indexlogtransformed::AbstractVector : index vector of log-transformed parameters
• lowerbounds::AbstractVector : lower bounds
• madsdata::AbstractDict : MADS problem dictionary
• params::AbstractVector : model parameters
• upperbounds::AbstractVector : upper bounds

Returns:

• Arcsine transformation of model parameters

Perform Bayesian Information Gap Decision Theory (BIG-DT) analysis

Methods:

• Mads.bigdt(madsdata::AbstractDict, nummodelruns::Int64; numhorizons, maxHorizon, numlikelihoods) : /home/travis/build/madsjulia/Mads.jl/src/MadsBayesInfoGap.jl:121

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• nummodelruns::Int64 : number of model runs

Keywords:

• maxHorizon : maximum info-gap horizons of uncertainty [default=3]
• numhorizons : number of info-gap horizons of uncertainty [default=100]
• numlikelihoods : number of Bayesian likelihoods [default=25]

Returns:

• dictionary with BIG-DT results

Bound model parameters based on their ranges

Methods:

• Mads.boundparameters!(madsdata::AbstractDict, pardict::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsParameters.jl:817
• Mads.boundparameters!(madsdata::AbstractDict, parvec::AbstractVector) : /home/travis/build/madsjulia/Mads.jl/src/MadsParameters.jl:805

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• pardict::AbstractDict : Parameter dictionary
• parvec::AbstractVector : Parameter vector

Calibrate Mads model using a constrained Levenberg-Marquardt technique

Mads.calibrate(madsdata; tolX=1e-3, tolG=1e-6, maxEval=1000, maxIter=100, maxJacobians=100, lambda=100.0, lambda_mu=10.0, np_lambda=10, show_trace=false, usenaive=false)

Methods:

• Mads.calibrate(madsdata::AbstractDict; tolX, tolG, tolOF, tolOFcount, minOF, maxEval, maxIter, maxJacobians, lambda, lambda_mu, np_lambda, show_trace, quiet, usenaive, save_results, localsa, parallel_optimization) : /home/travis/build/madsjulia/Mads.jl/src/MadsCalibrate.jl:195

Arguments:

• madsdata::AbstractDict : MADS problem dictionary

Keywords:

• lambda : initial Levenberg-Marquardt lambda [default=100.0]
• lambda_mu : lambda multiplication factor [default=10.0]
• localsa : perform local sensitivity analysis [default=false]
• maxEval : maximum number of model evaluations [default=1000]
• maxIter : maximum number of optimization iterations [default=100]
• maxJacobians : maximum number of Jacobian solves [default=100]
• minOF : objective function update tolerance [default=1e-3]
• np_lambda : number of parallel lambda solves [default=10]
• parallel_optimization
• quiet
• save_results : save intermediate results [default=true]
• show_trace : shows solution trace [default=false]
• tolG : parameter space update tolerance [default=1e-6]
• tolOF : objective function update tolerance [default=1e-3]
• tolOFcount : number of Jacobian runs with small objective function change [default=5]
• tolX : parameter space tolerance [default=1e-4]
• usenaive : use naive Levenberg-Marquardt solver [default=false]

Returns:

• model parameter dictionary with the optimal values at the minimum
• optimization algorithm results (e.g. results.minimizer)

Calibrate with random initial guesses

Methods:

• Mads.calibraterandom(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsCalibrate.jl:43
• Mads.calibraterandom(madsdata::AbstractDict, numberofsamples::Integer; tolX, tolG, tolOF, tolOFcount, minOF, maxEval, maxIter, maxJacobians, lambda, lambda_mu, np_lambda, show_trace, usenaive, seed, rng, quiet, all, save_results, first_init) : /home/travis/build/madsjulia/Mads.jl/src/MadsCalibrate.jl:43

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• numberofsamples::Integer : number of random initial samples [default=1]

Keywords:

• all : all model results are returned [default=false]
• first_init
• lambda : initial Levenberg-Marquardt lambda [default=100.0]
• lambda_mu : lambda multiplication factor [default=10.0]
• maxEval : maximum number of model evaluations [default=1000]
• maxIter : maximum number of optimization iterations [default=100]
• maxJacobians : maximum number of Jacobian solves [default=100]
• minOF
• np_lambda : number of parallel lambda solves [default=10]
• quiet : [default=true]
• rng
• save_results : save intermediate results [default=true]
• seed : random seed [default=0]
• show_trace : shows solution trace [default=false]
• tolG : parameter space update tolerance [default=1e-6]
• tolOF : objective function update tolerance [default=1e-3]
• tolOFcount : number of Jacobian runs with small objective function change [default=5]
• tolX : parameter space tolerance [default=1e-4]
• usenaive : use naive Levenberg-Marquardt solver [default=false]

Returns:

• model parameter dictionary with the optimal values at the minimum
• optimization algorithm results (e.g. bestresult[2].minimizer)

Example:

Mads.calibraterandom(madsdata; tolX=1e-3, tolG=1e-6, maxEval=1000, maxIter=100, maxJacobians=100, lambda=100.0, lambda_mu=10.0, np_lambda=10, show_trace=false, usenaive=false)
Mads.calibraterandom(madsdata, numberofsamples; tolX=1e-3, tolG=1e-6, maxEval=1000, maxIter=100, maxJacobians=100, lambda=100.0, lambda_mu=10.0, np_lambda=10, show_trace=false, usenaive=false)

Calibrate with random initial guesses in parallel

Methods:

• Mads.calibraterandom_parallel(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsCalibrate.jl:119
• Mads.calibraterandom_parallel(madsdata::AbstractDict, numberofsamples::Integer; tolX, tolG, tolOF, tolOFcount, minOF, maxEval, maxIter, maxJacobians, lambda, lambda_mu, np_lambda, show_trace, usenaive, seed, rng, quiet, save_results, first_init, localsa, all_results) : /home/travis/build/madsjulia/Mads.jl/src/MadsCalibrate.jl:119

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• numberofsamples::Integer : number of random initial samples [default=1]

Keywords:

• all_results
• first_init
• lambda : initial Levenberg-Marquardt lambda [default=100.0]
• lambda_mu : lambda multiplication factor [default=10.0]
• localsa : perform local sensitivity analysis [default=false]
• maxEval : maximum number of model evaluations [default=1000]
• maxIter : maximum number of optimization iterations [default=100]
• maxJacobians : maximum number of Jacobian solves [default=100]
• minOF
• np_lambda : number of parallel lambda solves [default=10]
• quiet : suppress output [default=true]
• rng
• save_results : save intermediate results [default=true]
• seed : random seed [default=0]
• show_trace : shows solution trace [default=false]
• tolG : parameter space update tolerance [default=1e-6]
• tolOF : objective function update tolerance [default=1e-3]
• tolOFcount : number of Jacobian runs with small objective function change [default=5]
• tolX : parameter space tolerance [default=1e-4]
• usenaive : use naive Levenberg-Marquardt solver [default=false]

Returns:

• vector with all objective function values
• boolean vector (converged/not converged)
• array with estimate model parameters

Make MADS not capture

Methods:

• Mads.captureoff() : /home/travis/build/madsjulia/Mads.jl/src/MadsHelpers.jl:148

Make MADS capture

Methods:

• Mads.captureon() : /home/travis/build/madsjulia/Mads.jl/src/MadsHelpers.jl:139

Check the directories where model outputs should be saved for MADS

Methods:

• Mads.checkmodeloutputdirs(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:766

Arguments:

• madsdata::AbstractDict : MADS problem dictionary

Returns:

• true or false

Check if a directory is readable

Methods:

• Mads.checknodedir(dir::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsExecute.jl:10
• Mads.checknodedir(dir::AbstractString, waittime::Float64) : /home/travis/build/madsjulia/Mads.jl/src/MadsExecute.jl:10
• Mads.checknodedir(node::AbstractString, dir::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsExecute.jl:1
• Mads.checknodedir(node::AbstractString, dir::AbstractString, waittime::Float64) : /home/travis/build/madsjulia/Mads.jl/src/MadsExecute.jl:1

Arguments:

• dir::AbstractString : directory
• node::AbstractString : computational node name (e.g. madsmax, wf03, or 127.0.0.1)
• waittime::Float64 : wait time in seconds [default=10]

Returns:

• true if the directory is readable, false otherwise

Check parameter ranges for model parameters

Methods:

• Mads.checkobservationranges(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:811

Arguments:

• madsdata::AbstractDict : MADS problem dictionary

Checkout (pull) the latest version of Mads modules

Methods:

• Mads.checkout() : /home/travis/build/madsjulia/Mads.jl/src/MadsModules.jl:107
• Mads.checkout(modulename::AbstractString; git, master, force, pull, required, all) : /home/travis/build/madsjulia/Mads.jl/src/MadsModules.jl:107

Arguments:

• modulename::AbstractString : module name

Keywords:

• all : whether to checkout all the modules [default=false]
• force : whether to overwrite local changes when checkout [default=false]
• git : whether to use "git checkout" [default=true]
• master : whether on master branch [default=false]
• pull : whether to run "git pull" [default=true]
• required : whether only checkout Mads.required modules [default=false]

Check parameter ranges for model parameters

Methods:

• Mads.checkparameterranges(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsParameters.jl:741

Arguments:

• madsdata::AbstractDict : MADS problem dictionary

Remove Mads coverage files

Methods:

• Mads.cleancoverage() : /home/travis/build/madsjulia/Mads.jl/src/MadsTest.jl:22

Call C MADS ins_obs() function from MADS dynamic library

Methods:

• Mads.cmadsins_obs(obsid::AbstractVector, instructionfilename::AbstractString, inputfilename::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsCMads.jl:38

Arguments:

• inputfilename::AbstractString : input file name
• instructionfilename::AbstractString : instruction file name
• obsid::AbstractVector : observation id

Return:

• observations

Commit the latest version of Mads modules in the repository

Methods:

• Mads.commit(commitmsg::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsModules.jl:255
• Mads.commit(commitmsg::AbstractString, modulename::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsModules.jl:255

Arguments:

• commitmsg::AbstractString : commit message
• modulename::AbstractString : module name

Compute injected/reduced contaminant mass (for a given set of mads input files when "path" is provided)

Methods:

• Mads.computemass(madsdata::AbstractDict; time) : /home/travis/build/madsjulia/Mads.jl/src/MadsAnasol.jl:463
• Mads.computemass(madsfiles::Union{Regex, String}; time, path) : /home/travis/build/madsjulia/Mads.jl/src/MadsAnasol.jl:490

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• madsfiles::Union{Regex, String} : matching pattern for Mads input files (string or regular expression accepted)

Keywords:

• path : search directory for the mads input files [default="."]
• time : computational time [default=0]

Returns:

• array with all the lambda values
• array with associated total injected mass
• array with associated total reduced mass

Example:

Mads.computemass(madsfiles; time=0, path=".")

Compute sensitivities for each model parameter; averaging the sensitivity indices over the entire observation range

Methods:

• Mads.computeparametersensitities(madsdata::AbstractDict, saresults::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsSensitivityAnalysis.jl:867

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• saresults::AbstractDict : dictionary with sensitivity analysis results

Compute concentration for a point in space and time (x,y,z,t)

Methods:

• Mads.contamination(wellx::Number, welly::Number, wellz::Number, n::Number, lambda::Number, theta::Number, vx::Number, vy::Number, vz::Number, ax::Number, ay::Number, az::Number, H::Number, x::Number, y::Number, z::Number, dx::Number, dy::Number, dz::Number, f::Number, t0::Number, t1::Number, t::AbstractVector, anasolfunction::Function) : /home/travis/build/madsjulia/Mads.jl/src/MadsAnasol.jl:433

Arguments:

• H::Number : Hurst coefficient for Fractional Brownian dispersion
• anasolfunction::Function
• ax::Number : dispersivity in X direction (longitudinal)
• ay::Number : dispersivity in Y direction (transverse horizontal)
• az::Number : dispersivity in Y direction (transverse vertical)
• dx::Number : source size (extent) in X direction
• dy::Number : source size (extent) in Y direction
• dz::Number : source size (extent) in Z direction
• f::Number : source mass flux
• lambda::Number : first-order reaction rate
• n::Number : porosity
• t0::Number : source starting time
• t1::Number : source termination time
• t::AbstractVector : vector of times to compute concentration at the observation point
• theta::Number : groundwater flow direction
• vx::Number : advective transport velocity in X direction
• vy::Number : advective transport velocity in Y direction
• vz::Number : advective transport velocity in Z direction
• wellx::Number : observation point (well) X coordinate
• welly::Number : observation point (well) Y coordinate
• wellz::Number : observation point (well) Z coordinate
• x::Number : X coordinate of contaminant source location
• y::Number : Y coordinate of contaminant source location
• z::Number : Z coordinate of contaminant source location

Returns:

• a vector of predicted concentration at (wellx, welly, wellz, t)

Copy aquifer parameters to become contaminant source parameters

Methods:

• Mads.copyaquifer2sourceparameters!(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsAnasol.jl:113

Arguments:

• madsdata::AbstractDict : MADS problem dictionary

Produce MADS copyright information

Methods:

• Mads.copyright() : /home/travis/build/madsjulia/Mads.jl/src/MadsHelp.jl:44

Turn off the generation of MADS tests (default)

Methods:

• Mads.create_tests_off() : /home/travis/build/madsjulia/Mads.jl/src/MadsHelpers.jl:193

Turn on the generation of MADS tests (dangerous)

Methods:

• Mads.create_tests_on() : /home/travis/build/madsjulia/Mads.jl/src/MadsHelpers.jl:184

Create Mads dictionary of observations and instruction file

Methods:

• Mads.createobservations(nrow::Int64) : /home/travis/build/madsjulia/Mads.jl/src/MadsCreate.jl:25
• Mads.createobservations(nrow::Int64, ncol::Int64; obstring, pretext, prestring, poststring, filename) : /home/travis/build/madsjulia/Mads.jl/src/MadsCreate.jl:25
• Mads.createobservations(obs::AbstractMatrix; key, weight, time) : /home/travis/build/madsjulia/Mads.jl/src/MadsCreate.jl:80
• Mads.createobservations(obs::AbstractVector; key, weight, time, min, max, minorig, maxorig, dist, distribution) : /home/travis/build/madsjulia/Mads.jl/src/MadsCreate.jl:43

Arguments:

• ncol::Int64 : number of columns [default 1]
• nrow::Int64 : number of rows
• obs::AbstractMatrix
• obs::AbstractVector

Keywords:

• dist
• distribution
• filename : file name
• key
• max
• maxorig
• min
• minorig
• obstring : observation string
• poststring : post instruction file string
• prestring : pre instruction file string
• pretext : preamble instructions
• time
• weight

)

Returns:

• observation dictionary

Create observations in the MADS problem dictionary based on time and observation vectors

Methods:

• Mads.createobservations!(madsdata::AbstractDict, observation::AbstractDict; logtransform, weight_type, weight) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:499
• Mads.createobservations!(madsdata::AbstractDict, time::AbstractVector, observation::AbstractVector; logtransform, weight_type, weight) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:455
• Mads.createobservations!(md::AbstractDict, obs::AbstractVecOrMat; kw...) : /home/travis/build/madsjulia/Mads.jl/src/MadsCreate.jl:115

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• md::AbstractDict
• obs::AbstractVecOrMat
• observation::AbstractDict : dictionary of observations
• observation::AbstractVector : dictionary of observations
• time::AbstractVector : vector of observation times

Keywords:

• logtransform : log transform observations [default=false]
• weight : weight value [default=1]
• weight_type : weight type [default=constant]

Create a new Mads problem where the observation targets are computed based on the model predictions

Methods:

• Mads.createproblem(in::Integer, out::Integer, f::Union{AbstractString, Function}; kw...) : /home/travis/build/madsjulia/Mads.jl/src/MadsCreate.jl:223
• Mads.createproblem(infilename::AbstractString, outfilename::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsCreate.jl:237
• Mads.createproblem(madsdata::AbstractDict, outfilename::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsCreate.jl:262
• Mads.createproblem(madsdata::AbstractDict, predictions::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsCreate.jl:273
• Mads.createproblem(madsdata::AbstractDict, predictions::AbstractDict, outfilename::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsCreate.jl:268
• Mads.createproblem(param::AbstractVector, obs::AbstractVecOrMat, f::Union{AbstractString, Function, Symbol}; problemname, paramkey, paramname, paramplotname, paramtype, parammin, parammax, paramminorig, parammaxorig, paramdist, distribution, expressions, paramlog, obskey, obsweight, obstime, obsmin, obsmax, obsminorig, obsmaxorig, obsdist) : /home/travis/build/madsjulia/Mads.jl/src/MadsCreate.jl:227
• Mads.createproblem(paramfile::AbstractString, obsfile::AbstractString, f::Union{AbstractString, Function}; kw...) : /home/travis/build/madsjulia/Mads.jl/src/MadsCreate.jl:211

Arguments:

• f::Union{AbstractString, Function, Symbol}
• f::Union{AbstractString, Function}
• in::Integer
• infilename::AbstractString : input Mads file
• madsdata::AbstractDict : MADS problem dictionary
• obs::AbstractVecOrMat
• obsfile::AbstractString
• out::Integer
• outfilename::AbstractString : output Mads file
• param::AbstractVector
• paramfile::AbstractString
• predictions::AbstractDict : dictionary of model predictions

Keywords:

• distribution
• expressions
• obsdist
• obskey
• obsmax
• obsmaxorig
• obsmin
• obsminorig
• obstime
• obsweight
• paramdist
• paramkey
• paramlog
• parammax
• parammaxorig
• parammin
• paramminorig
• paramname
• paramplotname
• paramtype
• problemname

Returns:

• new MADS problem dictionary

Create temporary directory

Methods:

• Mads.createtempdir(tempdirname::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:1449

Arguments:

• tempdirname::AbstractString : temporary directory name

Delete rows with NaN in a dataframe df

Methods:

• Mads.deleteNaN!(df::DataFrames.DataFrame) : /home/travis/build/madsjulia/Mads.jl/src/MadsSensitivityAnalysis.jl:1093

Arguments:

• df::DataFrames.DataFrame : dataframe

Delete a keyword in a class within the Mads dictionary madsdata

Methods:

• Mads.deletekeyword!(madsdata::AbstractDict, class::AbstractString, keyword::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsHelpers.jl:343
• Mads.deletekeyword!(madsdata::AbstractDict, keyword::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsHelpers.jl:337

Arguments:

• class::AbstractString : dictionary class; if not provided searches for keyword in Problem class
• keyword::AbstractString : dictionary key
• madsdata::AbstractDict : MADS problem dictionary

Delete all wells marked as being off in the MADS problem dictionary

Methods:

• Mads.welloff!(madsdata::AbstractDict, wellname::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:632

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• wellname::AbstractString : name of the well to be turned off

Delete all times in the MADS problem dictionary in a given list.

Methods:

• Mads.welloff!(madsdata::AbstractDict, wellname::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:632

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• wellname::AbstractString : name of the well to be turned off

Lists module dependents on a module (Mads by default)

Methods:

• Mads.dependents() : /home/travis/build/madsjulia/Mads.jl/src/MadsModules.jl:71
• Mads.dependents(modulename::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsModules.jl:71
• Mads.dependents(modulename::AbstractString, filter::Bool) : /home/travis/build/madsjulia/Mads.jl/src/MadsModules.jl:71

Arguments:

• filter::Bool : whether to filter modules [default=false]
• modulename::AbstractString : module name [default="Mads"]

Returns:

• modules that are dependents of the input module

Diff the latest version of Mads modules in the repository

Methods:

• Mads.diff() : /home/travis/build/madsjulia/Mads.jl/src/MadsModules.jl:198
• Mads.diff(modulename::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsModules.jl:198

Arguments:

• modulename::AbstractString : module name

Create web documentation

Methods:

• Mads.documentation_create() : /home/travis/build/madsjulia/Mads.jl/src/MadsPublish.jl:9
• Mads.documentation_create(modules_doc) : /home/travis/build/madsjulia/Mads.jl/src/MadsPublish.jl:9
• Mads.documentation_create(modules_doc, modules_load) : /home/travis/build/madsjulia/Mads.jl/src/MadsPublish.jl:9

Arguments:

• modules_doc
• modules_load

Create web documentation

Methods:

• Mads.documentation_deploy(; deploy_config) : /home/travis/build/madsjulia/Mads.jl/src/MadsPublish.jl:40

Keywords:

• deploy_config

Dump ASCII file

Methods:

• Mads.dumpasciifile(filename::AbstractString, data) : /home/travis/build/madsjulia/Mads.jl/src/MadsASCII.jl:29

Arguments:

• data : data to dump
• filename::AbstractString : ASCII file name

Dumps:

• ASCII file with the name in "filename"

Dump a JSON file

Methods:

• Mads.dumpjsonfile(filename::AbstractString, data) : /home/travis/build/madsjulia/Mads.jl/src/MadsJSON.jl:38

Arguments:

• data : data to dump
• filename::AbstractString : JSON file name

Dumps:

• JSON file with the name in "filename"

Dump well data from MADS problem dictionary into a ASCII file

Methods:

• Mads.dumpwelldata(madsdata::AbstractDict, filename::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:1315

Arguments:

• filename::AbstractString : output file name
• madsdata::AbstractDict : MADS problem dictionary

Dumps:

• filename : a ASCII file

Dump YAML file

Methods:

• Mads.dumpyamlfile(filename::AbstractString, data) : /home/travis/build/madsjulia/Mads.jl/src/MadsYAML.jl:29

Arguments:

• data : YAML data
• filename::AbstractString : output file name

Dump YAML Mads file

Methods:

• Mads.dumpyamlmadsfile(madsdata::AbstractDict, filename::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsYAML.jl:41

Arguments:

• filename::AbstractString : output file name
• madsdata::AbstractDict : MADS problem dictionary

Sensitivity analysis using Saltelli's extended Fourier Amplitude Sensitivity Testing (eFAST) method

Methods:

• Mads.efast(md::AbstractDict; N, M, gamma, seed, checkpointfrequency, save, load, restartdir, restart, rng) : /home/travis/build/madsjulia/Mads.jl/src/MadsSensitivityAnalysis.jl:1136

Arguments:

• md::AbstractDict : MADS problem dictionary

Keywords:

• M : maximum number of harmonics [default=6]
• N : number of samples [default=100]
• checkpointfrequency : check point frequency [default=N]
• gamma : multiplication factor (Saltelli 1999 recommends gamma = 2 or 4) [default=4]
• load
• restart : save restart information [default=false]
• restartdir : directory where files will be stored containing model results for the efast simulation restarts [default="efastcheckpoints"]
• rng
• save
• seed : random seed [default=0]

Bayesian sampling with Goodman & Weare's Affine Invariant Markov chain Monte Carlo (MCMC) Ensemble sampler (aka Emcee)

Methods:

• Mads.emceesampling(madsdata::AbstractDict, p0::Array; numwalkers, nsteps, burnin, thinning, seed, weightfactor, rng, distributed_function) : /home/travis/build/madsjulia/Mads.jl/src/MadsMonteCarlo.jl:47
• Mads.emceesampling(madsdata::AbstractDict; numwalkers, sigma, seed, rng, kw...) : /home/travis/build/madsjulia/Mads.jl/src/MadsMonteCarlo.jl:24

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• p0::Array : initial parameters (matrix of size (number of parameters, number of walkers) or (length(Mads.getoptparamkeys(madsdata)), numwalkers))

Keywords:

• burnin : number of initial realizations before the MCMC are recorded [default=10]
• distributed_function
• nsteps : number of final realizations in the chain [default=100]
• numwalkers : number of walkers (if in parallel this can be the number of available processors; in general, the higher the number of walkers, the better the results and computational time [default=10]
• rng
• seed : random seed [default=0]
• sigma : a standard deviation parameter used to initialize the walkers [default=0.01]
• thinning : removal of any thinning realization [default=1]
• weightfactor : weight factor [default=1.0]

Returns:

• MCMC chain
• log likelihoods of the final samples in the chain

Examples:

Mads.emceesampling(madsdata; numwalkers=10, nsteps=100, burnin=100, thinning=1, seed=2016, sigma=0.01)
Mads.emceesampling(madsdata, p0; numwalkers=10, nsteps=100, burnin=10, thinning=1, seed=2016)

Estimate kriging error

Methods:

• Mads.estimationerror(w::AbstractVector, covmat::AbstractMatrix, covvec::AbstractVector, cov0::Number) : /home/travis/build/madsjulia/Mads.jl/src/MadsKriging.jl:205
• Mads.estimationerror(w::AbstractVector, x0::AbstractVector, X::AbstractMatrix, covfn::Function) : /home/travis/build/madsjulia/Mads.jl/src/MadsKriging.jl:198

Arguments:

• X::AbstractMatrix : observation matrix
• cov0::Number : zero-separation covariance
• covfn::Function
• covmat::AbstractMatrix : covariance matrix
• covvec::AbstractVector : covariance vector
• w::AbstractVector : kriging weights
• x0::AbstractVector : estimated locations

Returns:

• estimation kriging error

Evaluate an expression string based on a parameter dictionary

Methods:

• Mads.evaluatemadsexpression(expressionstring::AbstractString, parameters::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsMisc.jl:154

Arguments:

• expressionstring::AbstractString : expression string
• parameters::AbstractDict : parameter dictionary applied to evaluate the expression string

Returns:

• dictionary containing the expression names as keys, and the values of the expression as values

Evaluate all the expressions in the Mads problem dictiorany based on a parameter dictionary

Methods:

• Mads.evaluatemadsexpressions(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsMisc.jl:173
• Mads.evaluatemadsexpressions(madsdata::AbstractDict, parameters::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsMisc.jl:173

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• parameters::AbstractDict : parameter dictionary applied to evaluate the expression strings

Returns:

• dictionary containing the parameter and expression names as keys, and the values of the expression as values

List available examples

Methods:

• Mads.examples() : /home/travis/build/madsjulia/Mads.jl/src/MadsExamples.jl:6

List available examples

Methods:

• Mads.examples() : /home/travis/build/madsjulia/Mads.jl/src/MadsExamples.jl:6

Exponential spatial covariance function

Methods:

• Mads.expcov(h::Number, maxcov::Number, scale::Number) : /home/travis/build/madsjulia/Mads.jl/src/MadsKriging.jl:31

Arguments:

• h::Number : separation distance
• maxcov::Number : maximum covariance
• scale::Number : scale

Returns:

• covariance

Exponential variogram

Methods:

• Mads.exponentialvariogram(h::Number, sill::Number, range::Number, nugget::Number) : /home/travis/build/madsjulia/Mads.jl/src/MadsKriging.jl:83

Arguments:

• h::Number : separation distance
• nugget::Number : nugget
• range::Number : range
• sill::Number : sill

Returns:

• Exponential variogram

Filter dictionary keys based on a string or regular expression

Methods:

• Mads.filterkeys(dict::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:979
• Mads.filterkeys(dict::AbstractDict, key::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:979
• Mads.filterkeys(dict::AbstractDict, key::Regex) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:978

Arguments:

• dict::AbstractDict : dictionary
• key::AbstractString : the regular expression or string used to filter dictionary keys
• key::Regex : the regular expression or string used to filter dictionary keys

Perform a forward run using the initial or provided values for the model parameters

Methods:

• Mads.forward(madsdata::AbstractDict, paramarray::AbstractArray; all, checkpointfrequency, checkpointfilename) : /home/travis/build/madsjulia/Mads.jl/src/MadsForward.jl:46
• Mads.forward(madsdata::AbstractDict, paramdict::AbstractDict; all, checkpointfrequency, checkpointfilename) : /home/travis/build/madsjulia/Mads.jl/src/MadsForward.jl:11
• Mads.forward(madsdata::AbstractDict; all) : /home/travis/build/madsjulia/Mads.jl/src/MadsForward.jl:7

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• paramarray::AbstractArray : array of model parameter values
• paramdict::AbstractDict : dictionary of model parameter values

Keywords:

• all : all model results are returned [default=false]
• checkpointfilename : check point file name [default="checkpoint_forward"]
• checkpointfrequency : check point frequency for storing restart information [default=0]

Returns:

• dictionary of model predictions

Perform a forward run over a 3D grid defined in madsdata using the initial or provided values for the model parameters

Methods:

• Mads.forwardgrid(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsForward.jl:134
• Mads.forwardgrid(madsdatain::AbstractDict, paramvalues::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsForward.jl:139

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• madsdatain::AbstractDict : MADS problem dictionary
• paramvalues::AbstractDict : dictionary of model parameter values

Returns:

• 3D array with model predictions along a 3D grid

Free Mads modules

Methods:

• Mads.free() : /home/travis/build/madsjulia/Mads.jl/src/MadsModules.jl:231
• Mads.free(modulename::AbstractString; required, all) : /home/travis/build/madsjulia/Mads.jl/src/MadsModules.jl:231

Arguments:

• modulename::AbstractString : module name

Keywords:

• all : free all the modules [default=false]
• required : only free Mads.required modules [default=false]

List available functions in the MADS modules:

Methods:

• Mads.functions() : /home/travis/build/madsjulia/Mads.jl/src/MadsHelp.jl:57
• Mads.functions(m::Union{Module, Symbol}) : /home/travis/build/madsjulia/Mads.jl/src/MadsHelp.jl:96
• Mads.functions(m::Union{Module, Symbol}, re::Regex; shortoutput, quiet) : /home/travis/build/madsjulia/Mads.jl/src/MadsHelp.jl:66
• Mads.functions(m::Union{Module, Symbol}, string::AbstractString; shortoutput, quiet) : /home/travis/build/madsjulia/Mads.jl/src/MadsHelp.jl:96
• Mads.functions(re::Regex; shortoutput, quiet) : /home/travis/build/madsjulia/Mads.jl/src/MadsHelp.jl:48
• Mads.functions(string::AbstractString; shortoutput, quiet) : /home/travis/build/madsjulia/Mads.jl/src/MadsHelp.jl:57

Arguments:

• m::Union{Module, Symbol} : MADS module
• re::Regex
• string::AbstractString : string to display functions with matching names

Keywords:

• quiet
• shortoutput

Examples:

Mads.functions()
Mads.functions(BIGUQ)
Mads.functions("get")
Mads.functions(Mads, "get")

Gaussian spatial covariance function

Methods:

• Mads.gaussiancov(h::Number, maxcov::Number, scale::Number) : /home/travis/build/madsjulia/Mads.jl/src/MadsKriging.jl:17

Arguments:

• h::Number : separation distance
• maxcov::Number : maximum covariance
• scale::Number : scale

Returns:

• covariance

Gaussian variogram

Methods:

• Mads.gaussianvariogram(h::Number, sill::Number, range::Number, nugget::Number) : /home/travis/build/madsjulia/Mads.jl/src/MadsKriging.jl:104

Arguments:

• h::Number : separation distance
• nugget::Number : nugget
• range::Number : range
• sill::Number : sill

Returns:

• Gaussian variogram

Get spatial covariance matrix

Methods:

• Mads.getcovmat(X::AbstractMatrix, covfunction::Function) : /home/travis/build/madsjulia/Mads.jl/src/MadsKriging.jl:160

Arguments:

• X::AbstractMatrix : matrix with coordinates of the data points (x or y)
• covfunction::Function

Returns:

• spatial covariance matrix

Get spatial covariance vector

Methods:

• Mads.getcovvec!(covvec::AbstractVector, x0::AbstractVector, X::AbstractMatrix, covfn::Function) : /home/travis/build/madsjulia/Mads.jl/src/MadsKriging.jl:186

Arguments:

• X::AbstractMatrix : matrix with coordinates of the data points
• covfn::Function : spatial covariance function
• covvec::AbstractVector : spatial covariance vector
• x0::AbstractVector : vector with coordinates of the estimation points (x or y)

Returns:

• spatial covariance vector

Get dictionary values for keys based on a string or regular expression

Methods:

• Mads.getdictvalues(dict::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:1001
• Mads.getdictvalues(dict::AbstractDict, key::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:1001
• Mads.getdictvalues(dict::AbstractDict, key::Regex) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:1000

Arguments:

• dict::AbstractDict : dictionary
• key::AbstractString : the key to find value for
• key::Regex : the key to find value for

Get directory

Methods:

• Mads.getdir(filename::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:570

Arguments:

• filename::AbstractString : file name

Returns:

• directory in file name

Example:

d = Mads.getdir("a.mads") # d = "."
d = Mads.getdir("test/a.mads") # d = "test"

Parse parameter distribution from a string

Methods:

• Mads.getdistribution(dist::AbstractString, i::AbstractString, inputtype::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsMisc.jl:207

Arguments:

• dist::AbstractString : parameter distribution
• i::AbstractString
• inputtype::AbstractString : input type (parameter or observation)

Returns:

• distribution

Get file name extension

Methods:

• Mads.getextension(filename::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:746

Arguments:

• filename::AbstractString : file name

Returns:

• file name extension

Example:

ext = Mads.getextension("a.mads") # ext = "mads"

Get file names by expanding wildcards

Methods:

• Mads.getfilenames(cmdstring::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:103

Arguments:

• cmdstring::AbstractString

Get important samples

Methods:

• Mads.getimportantsamples(samples::Array, llhoods::AbstractVector) : /home/travis/build/madsjulia/Mads.jl/src/MadsSensitivityAnalysis.jl:357

Arguments:

• llhoods::AbstractVector : vector of log-likelihoods
• samples::Array : array of samples

Returns:

• array of important samples

Get the keys in the MADS problem dictionary for parameters that are log-transformed (log)

Get the default MADS input file set as a MADS global variable using setmadsinputfile(filename)

Methods:

• Mads.getmadsinputfile() : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:522

Returns:

• input file name (e.g. input_file_name.mads)

Get the directory where the Mads data file is located

Methods:

• Mads.getmadsproblemdir(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:593

Arguments:

• madsdata::AbstractDict : MADS problem dictionary

Example:

madsdata = Mads.loadmadsfile("../../a.mads")
madsproblemdir = Mads.getmadsproblemdir(madsdata)

where madsproblemdir = "../../"

Get the MADS problem root name

Methods:

• Mads.getmadsrootname(madsdata::AbstractDict; first, version) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:544

Arguments:

• madsdata::AbstractDict : MADS problem dictionary

Keywords:

• first : use the first . in filename as the seperator between root name and extention [default=true]
• version : delete version information from filename for the returned rootname [default=false]

Example:

madsrootname = Mads.getmadsrootname(madsdata)

Returns:

• root of file name

Get next mads file name

Methods:

• Mads.getnextmadsfilename(filename::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:709

Arguments:

• filename::AbstractString : file name

Returns:

• next mads file name

Get the keys in the MADS problem dictionary for parameters that are NOT log-transformed (log)

Get the keys in the MADS problem dictionary for parameters that are NOT optimized (opt)

Get an array with dist values for observations in the MADS problem dictionary defined by obskeys

Get an array with dist values for all observations in the MADS problem dictionary

Get keys for all observations in the MADS problem dictionary

Methods:

• Mads.getobskeys(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:43

Arguments:

• madsdata::AbstractDict : MADS problem dictionary

Returns:

• keys for all observations in the MADS problem dictionary

Get an array with log values for observations in the MADS problem dictionary defined by obskeys

Get an array with log values for all observations in the MADS problem dictionary

Get an array with max values for observations in the MADS problem dictionary defined by obskeys

Get an array with max values for all observations in the MADS problem dictionary

Get an array with min values for observations in the MADS problem dictionary defined by obskeys

Get an array with min values for all observations in the MADS problem dictionary

Get an array with target values for observations in the MADS problem dictionary defined by obskeys

Get an array with target values for all observations in the MADS problem dictionary

Get an array with time values for observations in the MADS problem dictionary defined by obskeys

Get an array with time values for all observations in the MADS problem dictionary

Get an array with weight values for observations in the MADS problem dictionary defined by obskeys

Get an array with weight values for all observations in the MADS problem dictionary

Get the keys in the MADS problem dictionary for parameters that are optimized (opt)

Get optimizable parameters

Methods:

• Mads.getoptparams(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsParameters.jl:364
• Mads.getoptparams(madsdata::AbstractDict, parameterarray::Array) : /home/travis/build/madsjulia/Mads.jl/src/MadsParameters.jl:364
• Mads.getoptparams(madsdata::AbstractDict, parameterarray::Array, optparameterkey::Array) : /home/travis/build/madsjulia/Mads.jl/src/MadsParameters.jl:364

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• optparameterkey::Array : optimizable parameter keys
• parameterarray::Array : parameter array

Returns:

• parameter array

Get dictionary with all parameters and their respective initial values

Methods:

• Mads.getparamdict(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsParameters.jl:61

Arguments:

• madsdata::AbstractDict : MADS problem dictionary

Returns:

• dictionary with all parameters and their respective initial values

Get probabilistic distributions of all parameters in the MADS problem dictionary

Note:

Probabilistic distribution of parameters can be defined only if dist or min/max model parameter fields are specified in the MADS problem dictionary madsdata.

Methods:

• Mads.getparamdistributions(madsdata::AbstractDict; init_dist) : /home/travis/build/madsjulia/Mads.jl/src/MadsParameters.jl:696

Arguments:

• madsdata::AbstractDict : MADS problem dictionary

Keywords:

• init_dist : if true use the distribution defined for initialization in the MADS problem dictionary (defined using init_dist parameter field); else use the regular distribution defined in the MADS problem dictionary (defined using dist parameter field [default=false]

Returns:

• probabilistic distributions

Get keys of all parameters in the MADS problem dictionary

Methods:

• Mads.getparamkeys(madsdata::AbstractDict; filter) : /home/travis/build/madsjulia/Mads.jl/src/MadsParameters.jl:42

Arguments:

• madsdata::AbstractDict : MADS problem dictionary

Keywords:

• filter : parameter filter

Returns:

• array with the keys of all parameters in the MADS problem dictionary

Get independent sampling of model parameters defined in the MADS problem dictionary

Methods:

• Mads.getparamrandom(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsSensitivityAnalysis.jl:393
• Mads.getparamrandom(madsdata::AbstractDict, numsamples::Integer) : /home/travis/build/madsjulia/Mads.jl/src/MadsSensitivityAnalysis.jl:393
• Mads.getparamrandom(madsdata::AbstractDict, numsamples::Integer, parameterkey::AbstractString; init_dist) : /home/travis/build/madsjulia/Mads.jl/src/MadsSensitivityAnalysis.jl:393
• Mads.getparamrandom(madsdata::AbstractDict, parameterkey::AbstractString; numsamples, paramdist, init_dist) : /home/travis/build/madsjulia/Mads.jl/src/MadsSensitivityAnalysis.jl:410

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• numsamples::Integer : number of samples, [default=1]
• parameterkey::AbstractString : model parameter key

Keywords:

• init_dist : if true use the distribution set for initialization in the MADS problem dictionary (defined using init_dist parameter field); if false (default) use the regular distribution set in the MADS problem dictionary (defined using dist parameter field)
• numsamples : number of samples
• paramdist : dictionary of parameter distributions

Returns:

• generated sample

Get an array with init values for parameters defined by paramkeys

Get an array with init values for all the MADS model parameters

Get an array with init_max values for parameters defined by paramkeys

Methods:

• Mads.getparamsinit_max(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsParameters.jl:275
• Mads.getparamsinit_max(madsdata::AbstractDict, paramkeys::AbstractVector) : /home/travis/build/madsjulia/Mads.jl/src/MadsParameters.jl:275

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• paramkeys::AbstractVector : parameter keys

Returns:

• the parameter values

Get an array with init_min values for parameters

Methods:

• Mads.getparamsinit_min(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsParameters.jl:229
• Mads.getparamsinit_min(madsdata::AbstractDict, paramkeys::AbstractVector) : /home/travis/build/madsjulia/Mads.jl/src/MadsParameters.jl:229

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• paramkeys::AbstractVector : parameter keys

Returns:

• the parameter values

Get an array with log values for parameters defined by paramkeys

Get an array with log values for all the MADS model parameters

Get an array with longname values for parameters defined by paramkeys

Get an array with longname values for all the MADS model parameters

Get an array with max values for parameters defined by paramkeys

Methods:

• Mads.getparamsmax(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsParameters.jl:195
• Mads.getparamsmax(madsdata::AbstractDict, paramkeys::AbstractVector) : /home/travis/build/madsjulia/Mads.jl/src/MadsParameters.jl:195

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• paramkeys::AbstractVector : parameter keys

Returns:

• returns the parameter values

Get an array with min values for parameters defined by paramkeys

Methods:

• Mads.getparamsmin(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsParameters.jl:161
• Mads.getparamsmin(madsdata::AbstractDict, paramkeys::AbstractVector) : /home/travis/build/madsjulia/Mads.jl/src/MadsParameters.jl:161

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• paramkeys::AbstractVector : parameter keys

Returns:

• the parameter values

Get an array with plotname values for parameters defined by paramkeys

Get an array with plotname values for all the MADS model parameters

Get an array with step values for parameters defined by paramkeys

Get an array with step values for all the MADS model parameters

Get an array with type values for parameters defined by paramkeys

Get an array with type values for all the MADS model parameters

Get the directory where currently Mads is running

Methods:

• Mads.getproblemdir() : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:616

Example:

problemdir = Mads.getproblemdir()

Returns:

• Mads problem directory

Get the number of processors

Methods:

• Mads.getprocs() : /home/travis/build/madsjulia/Mads.jl/src/MadsParallel.jl:27

Get MADS restart status

Methods:

• Mads.getrestart(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsHelpers.jl:94

Arguments:

• madsdata::AbstractDict : MADS problem dictionary

Get the directory where Mads restarts will be stored

Methods:

• Mads.getrestartdir(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsFunc.jl:387
• Mads.getrestartdir(madsdata::AbstractDict, suffix::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsFunc.jl:387

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• suffix::AbstractString : Suffix to be added to the name of restart directory

Returns:

• restart directory where reusable model results will be stored

Get file name root

Methods:

• Mads.getrootname(filename::AbstractString; first, version) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:646

Arguments:

• filename::AbstractString : file name

Keywords:

• first : use the first . in filename as the seperator between root name and extention [default=true]
• version : delete version information from filename for the returned rootname [default=false]

Returns:

• root of file name

Example:

r = Mads.getrootname("a.rnd.dat") # r = "a"
r = Mads.getrootname("a.rnd.dat", first=false) # r = "a.rnd"

Get and return current random seed.

Methods:

• Mads.getseed() : /home/travis/build/madsjulia/Mads.jl/src/MadsHelpers.jl:496

Get sin-space dx

Methods:

• Mads.getsindx(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsHelpers.jl:375

Arguments:

• madsdata::AbstractDict : MADS problem dictionary

Returns:

• sin-space dx value

Get keys of all source parameters in the MADS problem dictionary

Methods:

• Mads.getsourcekeys(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsParameters.jl:79

Arguments:

• madsdata::AbstractDict : MADS problem dictionary

Returns:

• array with keys of all source parameters in the MADS problem dictionary

Get observation target

Methods:

• Mads.gettarget(o::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:221

Arguments:

• o::AbstractDict : observation data

Returns:

• observation target

Get keys for all targets (observations with weights greater than zero) in the MADS problem dictionary

Methods:

• Mads.gettargetkeys(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:57

Arguments:

• madsdata::AbstractDict : MADS problem dictionary

Returns:

• keys for all targets in the MADS problem dictionary

Get observation time

Methods:

• Mads.gettime(o::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:144

Arguments:

• o::AbstractDict : observation data

Returns:

• observation time ("NaN" it time is missing)

Get observation weight

Methods:

• Mads.getweight(o::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:182

Arguments:

• o::AbstractDict : observation data

Returns:

• observation weight ("NaN" when weight is missing)

Get spatial and temporal data in the Wells class

Methods:

• Mads.getwelldata(madsdata::AbstractDict; time) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:727

Arguments:

• madsdata::AbstractDict : Mads problem dictionary

Keywords:

• time : get observation times [default=false]

Returns:

• array with spatial and temporal data in the Wells class

Get keys for all wells in the MADS problem dictionary

Methods:

• Mads.getwellkeys(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:74

Arguments:

• madsdata::AbstractDict : MADS problem dictionary

Returns:

• keys for all wells in the MADS problem dictionary

Methods:

• Mads.getwelltargets(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:761

Arguments:

• madsdata::AbstractDict : Mads problem dictionary

Returns:

• array with targets in the Wells class

MADS graph output off

Methods:

• Mads.graphoff() : /home/travis/build/madsjulia/Mads.jl/src/MadsHelpers.jl:166

MADS graph output on

Methods:

• Mads.graphon() : /home/travis/build/madsjulia/Mads.jl/src/MadsHelpers.jl:157

Check for a keyword in a class within the Mads dictionary madsdata

Methods:

• Mads.haskeyword(madsdata::AbstractDict, class::AbstractString, keyword::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsHelpers.jl:275
• Mads.haskeyword(madsdata::AbstractDict, keyword::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsHelpers.jl:272

Arguments:

• class::AbstractString : dictionary class; if not provided searches for keyword in Problem class
• keyword::AbstractString : dictionary key
• madsdata::AbstractDict : MADS problem dictionary

Returns: true or false

Examples:

- `Mads.haskeyword(madsdata, "disp")` ... searches in `Problem` class by default
- `Mads.haskeyword(madsdata, "Wells", "R-28")` ... searches in `Wells` class for a keyword "R-28"

Produce MADS help information

Methods:

• Mads.help() : /home/travis/build/madsjulia/Mads.jl/src/MadsHelp.jl:35

Import Julia function everywhere from a file. The first function in the Julia input file is the one that will be targeted by Mads for execution.

Methods:

• Mads.importeverywhere(filename::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsFunc.jl:439

Arguments:

• filename::AbstractString : file name

Returns:

• Julia function to execute the model

Find indexes for dictionary keys based on a string or regular expression

Methods:

• Mads.indexkeys(dict::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:990
• Mads.indexkeys(dict::AbstractDict, key::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:990
• Mads.indexkeys(dict::AbstractDict, key::Regex) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:989

Arguments:

• dict::AbstractDict : dictionary
• key::AbstractString : the key to find index for
• key::Regex : the key to find index for

Information Gap Decision Analysis using JuMP

Methods:

• Mads.infogap_jump() : /home/travis/build/madsjulia/Mads.jl/src/MadsInfoGap.jl:23
• Mads.infogap_jump(madsdata::AbstractDict; horizons, retries, random, maxiter, verbosity, seed) : /home/travis/build/madsjulia/Mads.jl/src/MadsInfoGap.jl:23

Arguments:

• madsdata::AbstractDict : Mads problem dictionary

Keywords:

• horizons : info-gap horizons of uncertainty [default=[0.05, 0.1, 0.2, 0.5]]
• maxiter : maximum number of iterations [default=3000]
• random : random initial guesses [default=false]
• retries : number of solution retries [default=1]
• seed : random seed [default=0]
• verbosity : verbosity output level [default=0]

Information Gap Decision Analysis using JuMP

Methods:

• Mads.infogap_jump_polynomial() : /home/travis/build/madsjulia/Mads.jl/src/MadsInfoGap.jl:128
• Mads.infogap_jump_polynomial(madsdata::AbstractDict; horizons, retries, random, maxiter, verbosity, quiet, plot, model, seed) : /home/travis/build/madsjulia/Mads.jl/src/MadsInfoGap.jl:128

Arguments:

• madsdata::AbstractDict : Mads problem dictionary

Keywords:

• horizons : info-gap horizons of uncertainty [default=[0.05, 0.1, 0.2, 0.5]]
• maxiter : maximum number of iterations [default=3000]
• model : model id [default=1]
• plot : activate plotting [default=false]
• quiet : quiet [default=false]
• random : random initial guesses [default=false]
• retries : number of solution retries [default=1]
• seed : random seed [default=0]
• verbosity : verbosity output level [default=0]

Returns:

• hmin, hmax

Information Gap Decision Analysis using MathOptInterface

Methods:

• Mads.infogap_moi_lin() : /home/travis/build/madsjulia/Mads.jl/src/MadsInfoGap.jl:442
• Mads.infogap_moi_lin(madsdata::AbstractDict; horizons, retries, random, maxiter, verbosity, seed, pinit) : /home/travis/build/madsjulia/Mads.jl/src/MadsInfoGap.jl:442

Arguments:

• madsdata::AbstractDict : Mads problem dictionary

Keywords:

• horizons : info-gap horizons of uncertainty [default=[0.05, 0.1, 0.2, 0.5]]
• maxiter : maximum number of iterations [default=3000]
• pinit : vector with initial parameters
• random : random initial guesses [default=false]
• retries : number of solution retries [default=1]
• seed : random seed [default=0]
• verbosity : verbosity output level [default=0]

Information Gap Decision Analysis using MathOptInterface

Methods:

• Mads.infogap_moi_polynomial() : /home/travis/build/madsjulia/Mads.jl/src/MadsInfoGap.jl:301
• Mads.infogap_moi_polynomial(madsdata::AbstractDict; horizons, retries, random, maxiter, verbosity, seed, rng, pinit) : /home/travis/build/madsjulia/Mads.jl/src/MadsInfoGap.jl:301

Arguments:

• madsdata::AbstractDict : Mads problem dictionary

Keywords:

• horizons : info-gap horizons of uncertainty [default=[0.05, 0.1, 0.2, 0.5]]
• maxiter : maximum number of iterations [default=3000]
• pinit : vector with initial parameters
• random : random initial guesses [default=false]
• retries : number of solution retries [default=1]
• rng
• seed : random seed [default=0]
• verbosity : verbosity output level [default=0]

Apply Mads instruction file instructionfilename to read model output file modeloutputfilename

Methods:

• Mads.ins_obs(instructionfilename::AbstractString, modeloutputfilename::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:1208

Arguments:

• instructionfilename::AbstractString : instruction file name
• modeloutputfilename::AbstractString : model output file name

Returns:

• obsdict : observation dictionary with the model outputs

Convert an instruction line in the Mads instruction file into regular expressions

Methods:

• Mads.instline2regexs(instline::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:1108

Arguments:

• instline::AbstractString : instruction line

Returns:

• regexs : regular expressions
• obsnames : observation names
• getparamhere : parameters

Set inversely proportional observation weights in the MADS problem dictionary

Methods:

• Mads.invobsweights!(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:327
• Mads.invobsweights!(madsdata::AbstractDict, multiplier::Number) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:327
• Mads.invobsweights!(madsdata::AbstractDict, multiplier::Number, obskeys::AbstractVector) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:327

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• multiplier::Number : weight multiplier
• obskeys::AbstractVector

Set inversely proportional well weights in the MADS problem dictionary

Methods:

• Mads.invwellweights!(madsdata::AbstractDict, multiplier::Number) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:379
• Mads.invwellweights!(madsdata::AbstractDict, multiplier::Number, wellkeys::AbstractVector) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:379

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• multiplier::Number : weight multiplier
• wellkeys::AbstractVector

Is parameter with key parameterkey log-transformed?

Methods:

• Mads.islog(madsdata::AbstractDict, parameterkey::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsParameters.jl:437

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• parameterkey::AbstractString : parameter key

Returns:

• true if log-transformed, false otherwise

Is a dictionary containing all the observations

Methods:

• Mads.isobs(madsdata::AbstractDict, dict::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:17

Arguments:

• dict::AbstractDict : dictionary
• madsdata::AbstractDict : MADS problem dictionary

Returns:

• true if the dictionary contain all the observations, false otherwise

Is parameter with key parameterkey optimizable?

Methods:

• Mads.isopt(madsdata::AbstractDict, parameterkey::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsParameters.jl:417

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• parameterkey::AbstractString : parameter key

Returns:

• true if optimizable, false if not

Check if a dictionary containing all the Mads model parameters

Methods:

• Mads.isparam(madsdata::AbstractDict, dict::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsParameters.jl:15

Arguments:

• dict::AbstractDict : dictionary
• madsdata::AbstractDict : MADS problem dictionary

Returns:

• true if the dictionary containing all the parameters, false otherwise

Checks if package is available

Methods:

• Mads.ispkgavailable(modulename::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsHelpers.jl:591

Arguments:

• modulename::AbstractString : module name

Returns:

• true or false

Checks if package is available

Methods:

• Mads.ispkgavailable_old(modulename::AbstractString; quiet) : /home/travis/build/madsjulia/Mads.jl/src/MadsHelpers.jl:569

Arguments:

• modulename::AbstractString : module name

Keywords:

• quiet

Returns:

• true or false

Kriging

Methods:

• Mads.krige(x0mat::AbstractMatrix, X::AbstractMatrix, Z::AbstractVector, covfn::Function) : /home/travis/build/madsjulia/Mads.jl/src/MadsKriging.jl:125

Arguments:

• X::AbstractMatrix : coordinates of the observation (conditioning) data
• Z::AbstractVector : values for the observation (conditioning) data
• covfn::Function : spatial covariance function
• x0mat::AbstractMatrix : point coordinates at which to obtain kriging estimates

Returns:

• kriging estimates at x0mat

Levenberg-Marquardt optimization

Methods:

• Mads.levenberg_marquardt(f::Function, g::Function, x0) : /home/travis/build/madsjulia/Mads.jl/src/MadsLevenbergMarquardt.jl:357
• Mads.levenberg_marquardt(f::Function, g::Function, x0, o::Function; root, tolX, tolG, tolOF, tolOFcount, minOF, maxEval, maxIter, maxJacobians, lambda, lambda_scale, lambda_mu, lambda_nu, np_lambda, show_trace, quiet, callbackinitial, callbackiteration, callbackjacobian, callbackfinal, parallel_execution, center_provided) : /home/travis/build/madsjulia/Mads.jl/src/MadsLevenbergMarquardt.jl:357

Arguments:

• f::Function : forward model function
• g::Function : gradient function for the forward model
• o::Function : objective function [default=x->(x'*x)[1]]
• x0 : initial parameter guess

Keywords:

• callbackfinal : final call back function [default=(best_x::AbstractVector, of::Number, lambda::Number)->nothing]
• callbackinitial
• callbackiteration : call back function for each iteration [default=(best_x::AbstractVector, of::Number, lambda::Number)->nothing]
• callbackjacobian : call back function for each Jacobian [default=(x::AbstractVector, J::AbstractMatrix)->nothing]
• center_provided
• lambda : initial Levenberg-Marquardt lambda [default=eps(Float32)]
• lambda_mu : lambda multiplication factor μ [default=10]
• lambda_nu : lambda multiplication factor ν [default=2]
• lambda_scale : lambda scaling factor [default=1e-3,]
• maxEval : maximum number of model evaluations [default=1001]
• maxIter : maximum number of optimization iterations [default=100]
• maxJacobians : maximum number of Jacobian solves [default=100]
• minOF : objective function update tolerance [default=1e-3]
• np_lambda : number of parallel lambda solves [default=10]
• parallel_execution
• quiet
• root : Mads problem root name
• show_trace : shows solution trace [default=false]
• tolG : parameter space update tolerance [default=1e-6]
• tolOF : objective function update tolerance [default=1e-3]
• tolOFcount : number of Jacobian runs with small objective function change [default=5]
• tolX : parameter space tolerance [default=1e-4]

Link files in a temporary directory

Methods:

• Mads.linktempdir(madsproblemdir::AbstractString, tempdirname::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:1475

Arguments:

• madsproblemdir::AbstractString : Mads problem directory
• tempdirname::AbstractString : temporary directory name

Load ASCII file

Methods:

• Mads.loadasciifile(filename::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsASCII.jl:14

Arguments:

• filename::AbstractString : ASCII file name

Returns:

• data from the file

Load BIG YAML input file

Methods:

• Mads.loadmadsfile(filename::AbstractString; bigfile, format, quiet) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:141

Arguments:

• filename::AbstractString : input file name (e.g. input_file_name.mads)

Keywords:

• bigfile
• format
• quiet

Returns:

• MADS problem dictionary

Load a JSON file

Methods:

• Mads.loadjsonfile(filename::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsJSON.jl:16

Arguments:

• filename::AbstractString : JSON file name

Returns:

• data from the JSON file

Load MADS input file defining a MADS problem dictionary

Methods:

• Mads.loadmadsfile(filename::AbstractString; bigfile, format, quiet) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:141

Arguments:

• filename::AbstractString : input file name (e.g. input_file_name.mads)

Keywords:

• bigfile
• format : acceptable formats are yaml and json [default=yaml]
• quiet

Returns:

• MADS problem dictionary

Example:

md = Mads.loadmadsfile("input_file_name.mads")

Load a predefined Mads problem

Methods:

• Mads.loadmadsproblem(name::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsCreate.jl:14

Arguments:

• name::AbstractString : predefined MADS problem name

Returns:

• MADS problem dictionary

Load Saltelli sensitivity analysis results for fast simulation restarts

Methods:

• Mads.loadsaltellirestart!(evalmat::Array, matname::AbstractString, restartdir::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsSensitivityAnalysis.jl:604

Arguments:

• evalmat::Array : loaded array
• matname::AbstractString : matrix (array) name (defines the name of the loaded file)
• restartdir::AbstractString : directory where files will be stored containing model results for fast simulation restarts

Returns:

• true when successfully loaded, false when it is not

Load YAML file

Methods:

• Mads.loadyamlfile(filename::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsYAML.jl:16

Arguments:

• filename::AbstractString : file name

Returns:

• data in the yaml input file

Local sensitivity analysis based on eigen analysis of the parameter covariance matrix

Methods:

• Mads.localsa(madsdata::AbstractDict; sinspace, keyword, filename, format, datafiles, imagefiles, par, obs, J) : /home/travis/build/madsjulia/Mads.jl/src/MadsSensitivityAnalysis.jl:124

Arguments:

• madsdata::AbstractDict : MADS problem dictionary

Keywords:

• J : Jacobian matrix
• datafiles : flag to write data files [default=true]
• filename : output file name
• format : output plot format (png, pdf, etc.)
• imagefiles : flag to create image files [default=Mads.graphoutput]
• keyword : keyword to be added in the filename root
• obs : observations for the parameter set
• par : parameter set
• sinspace : apply sin transformation [default=true]

Dumps:

• filename : output plot file

Turn off execution of long MADS tests (default)

Methods:

• Mads.long_tests_off() : /home/travis/build/madsjulia/Mads.jl/src/MadsHelpers.jl:211

Turn on execution of long MADS tests

Methods:

• Mads.long_tests_on() : /home/travis/build/madsjulia/Mads.jl/src/MadsHelpers.jl:202

Define MadsModel type applied for Mads execution using MathOptInterface

Methods:

• Mads.madsMathOptInterface() : /home/travis/build/madsjulia/Mads.jl/src/MadsMathOptInterface.jl:16
• Mads.madsMathOptInterface(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsMathOptInterface.jl:16

Arguments:

• madsdata::AbstractDict : MADS problem dictionary [default=Dict()]

Check the number of processors on a series of servers

Methods:

• Mads.madscores() : /home/travis/build/madsjulia/Mads.jl/src/MadsParallel.jl:306
• Mads.madscores(nodenames::Vector{String}) : /home/travis/build/madsjulia/Mads.jl/src/MadsParallel.jl:306

Arguments:

• nodenames::Vector{String} : array with names of machines/nodes [default=madsservers]

MADS critical error messages

Methods:

• Mads.madscritical(message::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsLog.jl:70

Arguments:

• message::AbstractString : critical error message

MADS debug messages (controlled by quiet and debuglevel)

Methods:

• Mads.madsdebug(message::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsLog.jl:23
• Mads.madsdebug(message::AbstractString, level::Int64) : /home/travis/build/madsjulia/Mads.jl/src/MadsLog.jl:23

Arguments:

• level::Int64 : output verbosity level [default=0]
• message::AbstractString : debug message

Change the current directory to the Mads source dictionary

Methods:

• Mads.madsdir() : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:112

MADS error messages

Methods:

• Mads.madserror(message::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsLog.jl:60

Arguments:

• message::AbstractString : error message

MADS information/status messages (controlled by quietandverbositylevel`)

Methods:

• Mads.madsinfo(message::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsLog.jl:38
• Mads.madsinfo(message::AbstractString, level::Int64) : /home/travis/build/madsjulia/Mads.jl/src/MadsLog.jl:38

Arguments:

• level::Int64 : output verbosity level [default=0]
• message::AbstractString : information/status message

Check the load of a series of servers

Methods:

• Mads.madsload() : /home/travis/build/madsjulia/Mads.jl/src/MadsParallel.jl:326
• Mads.madsload(nodenames::Vector{String}) : /home/travis/build/madsjulia/Mads.jl/src/MadsParallel.jl:326

Arguments:

• nodenames::Vector{String} : array with names of machines/nodes [default=madsservers]

MADS output (controlled by quiet and verbositylevel)

Methods:

• Mads.madsoutput(message::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsLog.jl:8
• Mads.madsoutput(message::AbstractString, level::Int64) : /home/travis/build/madsjulia/Mads.jl/src/MadsLog.jl:8

Arguments:

• level::Int64 : output verbosity level [default=0]
• message::AbstractString : output message

Check the uptime of a series of servers

Methods:

• Mads.madsup() : /home/travis/build/madsjulia/Mads.jl/src/MadsParallel.jl:316
• Mads.madsup(nodenames::Vector{String}) : /home/travis/build/madsjulia/Mads.jl/src/MadsParallel.jl:316

Arguments:

• nodenames::Vector{String} : array with names of machines/nodes [default=madsservers]

MADS warning messages

Methods:

• Mads.madswarn(message::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsLog.jl:50

Arguments:

• message::AbstractString : warning message

Make a conditional log likelihood function that accepts an array containing the optimal parameter values

Methods:

• Mads.makearrayconditionalloglikelihood(madsdata::AbstractDict, conditionalloglikelihood) : /home/travis/build/madsjulia/Mads.jl/src/MadsMisc.jl:104

Arguments:

• conditionalloglikelihood : conditional log likelihood
• madsdata::AbstractDict : MADS problem dictionary

Returns:

• a conditional log likelihood function that accepts an array

Make array of conditional log-likelihoods

Methods:

• Mads.makearrayconditionalloglikelihood(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsBayesInfoGap.jl:158
• Mads.makearrayconditionalloglikelihood(madsdata::AbstractDict, conditionalloglikelihood) : /home/travis/build/madsjulia/Mads.jl/src/MadsMisc.jl:104

Arguments:

• conditionalloglikelihood
• madsdata::AbstractDict : MADS problem dictionary

Returns:

• array of conditional log-likelihoods

Make a version of the function f that accepts an array containing the optimal parameter values

Methods:

• Mads.makearrayfunction(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsMisc.jl:31
• Mads.makearrayfunction(madsdata::AbstractDict, f::Function) : /home/travis/build/madsjulia/Mads.jl/src/MadsMisc.jl:31

Arguments:

• f::Function : function [default=makemadscommandfunction(madsdata)]
• madsdata::AbstractDict : MADS problem dictionary

Returns:

• function accepting an array containing the optimal parameter values

Make a log likelihood function that accepts an array containing the optimal parameter values

Methods:

• Mads.makearrayloglikelihood(madsdata::AbstractDict, loglikelihood) : /home/travis/build/madsjulia/Mads.jl/src/MadsMisc.jl:127

Arguments:

• loglikelihood : log likelihood
• madsdata::AbstractDict : MADS problem dictionary

Returns:

• a log likelihood function that accepts an array

Setup Bayesian Information Gap Decision Theory (BIG-DT) problem

Methods:

• Mads.makebigdt!(madsdata::AbstractDict, choice::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsBayesInfoGap.jl:33

Arguments:

• choice::AbstractDict : dictionary of BIG-DT choices (scenarios)
• madsdata::AbstractDict : MADS problem dictionary

Returns:

• BIG-DT problem type

Setup Bayesian Information Gap Decision Theory (BIG-DT) problem

Methods:

• Mads.makebigdt(madsdata::AbstractDict, choice::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsBayesInfoGap.jl:18

Arguments:

• choice::AbstractDict : dictionary of BIG-DT choices (scenarios)
• madsdata::AbstractDict : MADS problem dictionary

Returns:

• BIG-DT problem type

Create a function to compute concentrations for all the observation points using Anasol

Methods:

• Mads.makecomputeconcentrations(madsdata::AbstractDict; calczeroweightobs, calcpredictions) : /home/travis/build/madsjulia/Mads.jl/src/MadsAnasol.jl:177

Arguments:

• madsdata::AbstractDict : MADS problem dictionary

Keywords:

• calcpredictions : calculate zero weight predictions [default=true]
• calczeroweightobs : calculate zero weight observations[default=false]

Returns:

• function to compute concentrations; the new function returns a dictionary of observations and model predicted concentrations

Examples:

computeconcentrations = Mads.makecomputeconcentrations(madsdata)
paramkeys = Mads.getparamkeys(madsdata)
paramdict = OrderedDict(zip(paramkeys, map(key->madsdata["Parameters"][key]["init"], paramkeys)))
forward_preds = computeconcentrations(paramdict)

Make dixon price

Methods:

• Mads.makedixonprice(n::Integer) : /home/travis/build/madsjulia/Mads.jl/src/MadsTestFunctions.jl:257

Arguments:

• n::Integer : number of observations

Returns:

• dixon price

Methods:

• Mads.makedixonprice(n::Integer) : /home/travis/build/madsjulia/Mads.jl/src/MadsTestFunctions.jl:257

Arguments:

• n::Integer : number of observations

Returns:

• dixon price gradient

Make a version of the function f that accepts an array containing the optimal parameter values, and returns an array of observations

Methods:

• Mads.makedoublearrayfunction(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsMisc.jl:77
• Mads.makedoublearrayfunction(madsdata::AbstractDict, f::Function) : /home/travis/build/madsjulia/Mads.jl/src/MadsMisc.jl:77

Arguments:

• f::Function : function [default=makemadscommandfunction(madsdata)]
• madsdata::AbstractDict : MADS problem dictionary

Returns:

• function accepting an array containing the optimal parameter values, and returning an array of observations

Make forward model, gradient, objective functions needed for Levenberg-Marquardt optimization

Methods:

• Mads.makelmfunctions(f::Function) : /home/travis/build/madsjulia/Mads.jl/src/MadsLevenbergMarquardt.jl:107
• Mads.makelmfunctions(madsdata::AbstractDict; parallel_gradients) : /home/travis/build/madsjulia/Mads.jl/src/MadsLevenbergMarquardt.jl:128

Arguments:

• f::Function : Function
• madsdata::AbstractDict : MADS problem dictionary

Keywords:

• parallel_gradients

Returns:

• forward model, gradient, objective functions

Make gradient function needed for local sensitivity analysis

Methods:

• Mads.makelocalsafunction(madsdata::AbstractDict; multiplycenterbyweights) : /home/travis/build/madsjulia/Mads.jl/src/MadsSensitivityAnalysis.jl:25

Arguments:

• madsdata::AbstractDict : MADS problem dictionary

Keywords:

• multiplycenterbyweights : multiply center by observation weights [default=true]

Returns:

• gradient function

Make a function to compute the prior log-likelihood of the model parameters listed in the MADS problem dictionary madsdata

Methods:

• Mads.makelogprior(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsFunc.jl:467

Arguments:

• madsdata::AbstractDict : MADS problem dictionary

Return:

• the prior log-likelihood of the model parameters listed in the MADS problem dictionary madsdata

Make MADS function to execute the model defined in the input MADS problem dictionary

Methods:

• Mads.makemadscommandfunction(madsdata_in::AbstractDict; obskeys, calczeroweightobs, calcpredictions, quiet) : /home/travis/build/madsjulia/Mads.jl/src/MadsFunc.jl:68

Arguments:

• madsdata_in::AbstractDict

Keywords:

• calcpredictions : Calculate predictions [default=true]
• calczeroweightobs : Calculate zero weight observations [default=false]
• obskeys
• quiet

Example:

Mads.makemadscommandfunction(madsdata)

MADS can be coupled with any internal or external model. The model coupling is defined in the MADS problem dictionary. The expectations is that for a given set of model inputs, the model will produce a model output that will be provided to MADS. The fields in the MADS problem dictionary that can be used to define the model coupling are:

• Model : execute a Julia function defined in an external input Julia file. The function that should accept a parameter dictionary with all the model parameters as an input argument and should return an observation dictionary with all the model predicted observations. MADS will execute the first function defined in the file.

• MADS model : create a Julia function based on an external input Julia file. The input file should contain a function that accepts as an argument the MADS problem dictionary. MADS will execute the first function defined in the file. This function should create a Julia function that will accept a parameter dictionary with all the model parameters as an input argument and will return an observation dictionary with all the model predicted observations.

• Julia model : execute an internal Julia function that accepts a parameter dictionary with all the model parameters as an input argument and will return an observation dictionary with all the model predicted observations.

• Julia function : execute an internal Julia function that accepts a parameter vector with all the model parameters as an input argument and will return an observation vector with all the model predicted observations.

• Command : execute an external UNIX command or script that will execute an external model.

• Julia command : execute a Julia script that will execute an external model. The Julia script is defined in an external Julia file. The input file should contain a function that accepts a parameter dictionary with all the model parameters as an input argument; MADS will execute the first function defined in the file. The Julia script should be capable to (1) execute the model (making a system call of an external model), (2) parse the model outputs, (3) return an observation dictionary with model predictions.

Both Command and Julia command can use different approaches to pass model parameters to the external model.

Only Command uses different approaches to get back the model outputs.

The script defined under Julia command parses the model outputs using Julia.

The available options for writing model inputs and reading model outputs are as follows.

Options for writing model inputs:

• Templates : template files for writing model input files as defined at http://madsjulia.github.io
• ASCIIParameters : model parameters written in a ASCII file
• JLDParameters : model parameters written in a JLD file
• YAMLParameters : model parameters written in a YAML file
• JSONParameters : model parameters written in a JSON file

Options for reading model outputs:

• Instructions : instruction files for reading model output files as defined at http://madsjulia.github.io
• ASCIIPredictions : model predictions read from a ASCII file
• JLDPredictions : model predictions read from a JLD file
• YAMLPredictions : model predictions read from a YAML file
• JSONPredictions : model predictions read from a JSON file

Returns:

• Mads function to execute a forward model simulation

Make a function to compute the conditional log-likelihood of the model parameters conditioned on the model predictions/observations. Model parameters and observations are defined in the MADS problem dictionary madsdata.

Methods:

• Mads.makemadsconditionalloglikelihood(madsdata::AbstractDict; weightfactor) : /home/travis/build/madsjulia/Mads.jl/src/MadsFunc.jl:490

Arguments:

• madsdata::AbstractDict : MADS problem dictionary

Keywords:

• weightfactor : Weight factor [default=1]

Return:

• the conditional log-likelihood

Make a function to compute the log-likelihood for a given set of model parameters, associated model predictions and existing observations. By default, the Log-likelihood function computed internally. The Log-likelihood can be constructed from an external Julia function defined the MADS problem dictionary under LogLikelihood or ConditionalLogLikelihood.

In the case of a LogLikelihood external Julia function, the first function in the file provided should be a function that takes as arguments:

• dictionary of model parameters
• dictionary of model predictions
• dictionary of respective observations

In the case of a ConditionalLogLikelihood external Julia function, the first function in the file provided should be a function that takes as arguments:

• dictionary of model predictions
• dictionary of respective observations

Methods:

• Mads.makemadsloglikelihood(madsdata::AbstractDict; weightfactor) : /home/travis/build/madsjulia/Mads.jl/src/MadsFunc.jl:535

Arguments:

• madsdata::AbstractDict : MADS problem dictionary

Keywords:

• weightfactor : Weight factor [default=1]

Returns:

• the log-likelihood for a given set of model parameters

Make Reusable Mads function to execute a forward model simulation (automatically restarts if restart data exists)

Methods:

• Mads.makemadsreusablefunction(madsdata::AbstractDict, madscommandfunction::Function) : /home/travis/build/madsjulia/Mads.jl/src/MadsFunc.jl:339
• Mads.makemadsreusablefunction(madsdata::AbstractDict, madscommandfunction::Function, suffix::AbstractString; usedict) : /home/travis/build/madsjulia/Mads.jl/src/MadsFunc.jl:339
• Mads.makemadsreusablefunction(paramkeys::AbstractVector, obskeys::AbstractVector, madsdatarestart::Union{Bool, String}, madscommandfunction::Function, restartdir::AbstractString; usedict) : /home/travis/build/madsjulia/Mads.jl/src/MadsFunc.jl:342

Arguments:

• madscommandfunction::Function : Mads function to execute a forward model simulation
• madsdata::AbstractDict : MADS problem dictionary
• madsdatarestart::Union{Bool, String} : Restart type (memory/disk) or on/off status
• obskeys::AbstractVector : Dictionary of observation keys
• paramkeys::AbstractVector : Dictionary of parameter keys
• restartdir::AbstractString : Restart directory where the reusable model results are stored
• suffix::AbstractString : Suffix to be added to the name of restart directory

Keywords:

• usedict : Use dictionary [default=true]

Returns:

• Reusable Mads function to execute a forward model simulation (automatically restarts if restart data exists)

Make forward model, gradient, objective functions needed for MathOptInterface optimization

Methods:

• Mads.makemoifunctions(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsMathOptInterface.jl:90

Arguments:

• madsdata::AbstractDict : MADS problem dictionary

Returns:

• forward model, gradient, objective functions

Make Powell test function for LM optimization

Methods:

• Mads.makepowell(n::Integer) : /home/travis/build/madsjulia/Mads.jl/src/MadsTestFunctions.jl:160

Arguments:

• n::Integer : number of observations

Returns:

• Powell test function for LM optimization

ake parameter gradients of the Powell test function for LM optimization

Methods:

• Mads.makepowell_gradient(n::Integer) : /home/travis/build/madsjulia/Mads.jl/src/MadsTestFunctions.jl:184

Arguments:

• n::Integer : number of observations

Returns:

• arameter gradients of the Powell test function for LM optimization

Make Rosenbrock test function for LM optimization

Methods:

• Mads.makerosenbrock(n::Integer) : /home/travis/build/madsjulia/Mads.jl/src/MadsTestFunctions.jl:115

Arguments:

• n::Integer : number of observations

Returns:

• Rosenbrock test function for LM optimization

Make parameter gradients of the Rosenbrock test function for LM optimization

Methods:

• Mads.makerosenbrock_gradient(n::Integer) : /home/travis/build/madsjulia/Mads.jl/src/MadsTestFunctions.jl:137

Arguments:

• n::Integer : number of observations

Returns:

• parameter gradients of the Rosenbrock test function for LM optimization

Methods:

• Mads.makerotatedhyperellipsoid(n::Integer) : /home/travis/build/madsjulia/Mads.jl/src/MadsTestFunctions.jl:336

Arguments:

• n::Integer : number of observations

Returns:

• rotated hyperellipsoid

Methods:

• Mads.makerotatedhyperellipsoid_gradient(n::Integer) : /home/travis/build/madsjulia/Mads.jl/src/MadsTestFunctions.jl:360

Arguments:

• n::Integer : number of observations

Returns:

• rotated hyperellipsoid gradient

Make sphere

Methods:

• Mads.makesphere(n::Integer) : /home/travis/build/madsjulia/Mads.jl/src/MadsTestFunctions.jl:215

Arguments:

• n::Integer : number of observations

Returns:

• sphere

Make sphere gradient

Methods:

• Mads.makesphere_gradient(n::Integer) : /home/travis/build/madsjulia/Mads.jl/src/MadsTestFunctions.jl:236

Arguments:

• n::Integer : number of observations

Returns:

• sphere gradient

Methods:

• Mads.makesumsquares(n::Integer) : /home/travis/build/madsjulia/Mads.jl/src/MadsTestFunctions.jl:298

Arguments:

• n::Integer : number of observations

Returns:

• sumsquares

Methods:

• Mads.makesumsquares_gradient(n::Integer) : /home/travis/build/madsjulia/Mads.jl/src/MadsTestFunctions.jl:317

Arguments:

• n::Integer : number of observations

Returns:

• sumsquares gradient

Make SVR model functions (executor and cleaner)

Methods:

• Mads.makesvrmodel(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsSVR.jl:204
• Mads.makesvrmodel(madsdata::AbstractDict, numberofsamples::Integer; check, addminmax, loadsvr, savesvr, svm_type, kernel_type, degree, gamma, coef0, C, nu, epsilon, cache_size, tol, shrinking, probability, verbose, seed) : /home/travis/build/madsjulia/Mads.jl/src/MadsSVR.jl:204

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• numberofsamples::Integer : number of samples [default=100]

Keywords:

• C : cost; penalty parameter of the error term [default=1000.0]
• addminmax : add parameter minimum / maximum range values in the training set [default=true]
• cache_size : size of the kernel cache [default=100.0]
• check : check SVR performance [default=false]
• coef0 : independent term in kernel function; important only in POLY and SIGMOND kernel types [default=0]
• degree : degree of the polynomial kernel [default=3]
• epsilon : epsilon in the EPSILON_SVR model; defines an epsilon-tube within which no penalty is associated in the training loss function with points predicted within a distance epsilon from the actual value [default=0.001]
• gamma : coefficient for RBF, POLY and SIGMOND kernel types [default=1/numberofsamples]
• kernel_type : kernel type[default=SVR.RBF]
• loadsvr : load SVR models [default=false]
• nu : upper bound on the fraction of training errors / lower bound of the fraction of support vectors; acceptable range (0, 1]; applied if NU_SVR model [default=0.5]
• probability : train to estimate probabilities [default=false]
• savesvr : save SVR models [default=false]
• seed : random seed [default=0]
• shrinking : apply shrinking heuristic [default=true]
• svm_type : SVM type [default=SVR.EPSILON_SVR]
• tol : tolerance of termination criterion [default=0.001]
• verbose : verbose output [default=false]

Returns:

• function performing SVR predictions
• function loading existing SVR models
• function saving SVR models
• function removing SVR models from the memory

Scale down values larger than max(Float32) in a dataframe df so that Gadfly can plot the data

Methods:

• Mads.maxtofloatmax!(df::DataFrames.DataFrame) : /home/travis/build/madsjulia/Mads.jl/src/MadsSensitivityAnalysis.jl:1110

Arguments:

• df::DataFrames.DataFrame : dataframe

Create mesh grid

Methods:

• Mads.meshgrid(nx::Number, ny::Number) : /home/travis/build/madsjulia/Mads.jl/src/MadsHelpers.jl:454
• Mads.meshgrid(x::AbstractVector, y::AbstractVector) : /home/travis/build/madsjulia/Mads.jl/src/MadsHelpers.jl:447

Arguments:

• nx::Number
• ny::Number
• x::AbstractVector : vector of grid x coordinates
• y::AbstractVector : vector of grid y coordinates

Returns:

• 2D grid coordinates based on the coordinates contained in vectors x and y

Minimize Julia function using a constrained Levenberg-Marquardt technique

Mads.calibrate(madsdata; tolX=1e-3, tolG=1e-6, maxEval=1000, maxIter=100, maxJacobians=100, lambda=100.0, lambda_mu=10.0, np_lambda=10, show_trace=false, usenaive=false)

Methods:

• Mads.calibrate(madsdata::AbstractDict; tolX, tolG, tolOF, tolOFcount, minOF, maxEval, maxIter, maxJacobians, lambda, lambda_mu, np_lambda, show_trace, quiet, usenaive, save_results, localsa, parallel_optimization) : /home/travis/build/madsjulia/Mads.jl/src/MadsCalibrate.jl:195

Arguments:

• madsdata::AbstractDict

Keywords:

• lambda : initial Levenberg-Marquardt lambda [default=100.0]
• lambda_mu : lambda multiplication factor [default=10.0]
• localsa
• maxEval : maximum number of model evaluations [default=1000]
• maxIter : maximum number of optimization iterations [default=100]
• maxJacobians : maximum number of Jacobian solves [default=100]
• minOF : objective function update tolerance [default=1e-3]
• np_lambda : number of parallel lambda solves [default=10]
• parallel_optimization
• quiet
• save_results
• show_trace : shows solution trace [default=false]
• tolG : parameter space update tolerance [default=1e-6]
• tolOF : objective function update tolerance [default=1e-3]
• tolOFcount : number of Jacobian runs with small objective function change [default=5]
• tolX : parameter space tolerance [default=1e-4]
• usenaive

Returns:

• vector with the optimal parameter values at the minimum
• optimization algorithm results (e.g. results.minimizer)

Create a directory (if does not already exist)

Methods:

• Mads.mkdir(dirname::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:1502

Arguments:

• dirname::AbstractString : directory

Model section information criteria

Methods:

• Mads.modelinformationcriteria(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsModelSelection.jl:11
• Mads.modelinformationcriteria(madsdata::AbstractDict, par::Array{Float64}) : /home/travis/build/madsjulia/Mads.jl/src/MadsModelSelection.jl:11

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• par::Array{Float64} : parameter array

Modify (multiply) observation weights in the MADS problem dictionary

Methods:

• Mads.modobsweights!(madsdata::AbstractDict, value::Number) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:314
• Mads.modobsweights!(madsdata::AbstractDict, value::Number, obskeys::AbstractVector) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:314

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• obskeys::AbstractVector
• value::Number : value for modifing observation weights

Modify (multiply) well weights in the MADS problem dictionary

Methods:

• Mads.modwellweights!(madsdata::AbstractDict, value::Number) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:361
• Mads.modwellweights!(madsdata::AbstractDict, value::Number, wellkeys::AbstractVector) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:361

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• value::Number : value for well weights
• wellkeys::AbstractVector

Monte Carlo analysis

Methods:

• Mads.montecarlo(madsdata::AbstractDict; compute, N, filename) : /home/travis/build/madsjulia/Mads.jl/src/MadsMonteCarlo.jl:208

Arguments:

• madsdata::AbstractDict : MADS problem dictionary

Keywords:

• N : number of samples [default=100]
• compute
• filename : file name to save Monte-Carlo results

Returns:

• parameter dictionary containing the data arrays

Dumps:

• YAML output file with the parameter dictionary containing the data arrays

Example:

Mads.montecarlo(madsdata; N=100)

Naive Levenberg-Marquardt optimization: get the LM parameter space step

Methods:

• Mads.naive_get_deltax(JpJ::AbstractMatrix{Float64}, Jp::AbstractMatrix{Float64}, f0::AbstractVector{Float64}, lambda::Number) : /home/travis/build/madsjulia/Mads.jl/src/MadsLevenbergMarquardt.jl:264

Arguments:

• Jp::AbstractMatrix{Float64} : Jacobian matrix times model parameters
• JpJ::AbstractMatrix{Float64} : Jacobian matrix times model parameters times transposed Jacobian matrix
• f0::AbstractVector{Float64} : initial model observations
• lambda::Number : Levenberg-Marquardt lambda

Returns:

• the LM parameter space step

Naive Levenberg-Marquardt optimization

Methods:

• Mads.naive_levenberg_marquardt(f::Function, g::Function, x0::AbstractVector{Float64}) : /home/travis/build/madsjulia/Mads.jl/src/MadsLevenbergMarquardt.jl:314
• Mads.naive_levenberg_marquardt(f::Function, g::Function, x0::AbstractVector{Float64}, o::Function; maxIter, maxEval, lambda, lambda_mu, np_lambda) : /home/travis/build/madsjulia/Mads.jl/src/MadsLevenbergMarquardt.jl:314

Arguments:

• f::Function : forward model function
• g::Function : gradient function for the forward model
• o::Function : objective function [default=x->(x'*x)[1]]
• x0::AbstractVector{Float64} : initial parameter guess

Keywords:

• lambda : initial Levenberg-Marquardt lambda [default=100]
• lambda_mu : lambda multiplication factor μ [default=10]
• maxEval : maximum number of model evaluations [default=101]
• maxIter : maximum number of optimization iterations [default=10]
• np_lambda : number of parallel lambda solves [default=10]

Returns:

Naive Levenberg-Marquardt optimization: perform LM iteration

Methods:

• Mads.naive_lm_iteration(f::Function, g::Function, o::Function, x0::AbstractVector{Float64}, f0::AbstractVector{Float64}, lambdas::AbstractVector{Float64}) : /home/travis/build/madsjulia/Mads.jl/src/MadsLevenbergMarquardt.jl:285

Arguments:

• f0::AbstractVector{Float64} : initial model observations
• f::Function : forward model function
• g::Function : gradient function for the forward model
• lambdas::AbstractVector{Float64} : Levenberg-Marquardt lambdas
• o::Function : objective function
• x0::AbstractVector{Float64} : initial parameter guess

Returns:

Disable MADS plotting

Methods:

• Mads.noplot() : /home/travis/build/madsjulia/Mads.jl/src/MadsParallel.jl:239

Execute Jupyter notebook in IJulia or as a script

Methods:

• Mads.notebook(rootname::AbstractString; script, notebook_directory, check) : /home/travis/build/madsjulia/Mads.jl/src/MadsNotebooks.jl:20

Arguments:

• rootname::AbstractString : notebook root name

Keywords:

• check : check of notebook exists
• notebook_directory : notebook directory
• script : execute as a script

Check is Jupyter notebook exists

Methods:

• Mads.notebook_check(rootname::AbstractString; notebook_directory) : /home/travis/build/madsjulia/Mads.jl/src/MadsNotebooks.jl:101

Arguments:

• rootname::AbstractString : notebook root name

Keywords:

• notebook_directory : notebook directory

Export Jupyter notebook in html, markdown, latex, and script versions

Methods:

• Mads.notebook_export(rootname::AbstractString; notebook_directory) : /home/travis/build/madsjulia/Mads.jl/src/MadsNotebooks.jl:67

Arguments:

• rootname::AbstractString : notebook root name

Keywords:

• notebook_directory : notebook directory

Open Jupyter in the Mads notebook directory

Methods:

• Mads.notebooks(; notebook_directory) : /home/travis/build/madsjulia/Mads.jl/src/MadsNotebooks.jl:52

Keywords:

• notebook_directory : notebook directory

Execute Jupyter notebook as a script

Methods:

• Mads.notebookscript(a...; script, notebook_directory, k...) : /home/travis/build/madsjulia/Mads.jl/src/MadsNotebooks.jl:9

Keywords:

• notebook_directory : notebook directory
• script : execute as a script

Match an instruction line in the Mads instruction file with model input file

Methods:

• Mads.obslineoccursin(obsline::AbstractString, regexs::Vector{Regex}) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:1157

Arguments:

• obsline::AbstractString : instruction line
• regexs::Vector{Regex} : regular expressions

Returns:

• true or false

Compute objective function

Methods:

• Mads.of(madsdata::AbstractDict, d::AbstractDict; filter) : /home/travis/build/madsjulia/Mads.jl/src/MadsLevenbergMarquardt.jl:54
• Mads.of(madsdata::AbstractDict, resultvec::AbstractVector; filter) : /home/travis/build/madsjulia/Mads.jl/src/MadsLevenbergMarquardt.jl:50
• Mads.of(madsdata::AbstractDict; filter) : /home/travis/build/madsjulia/Mads.jl/src/MadsLevenbergMarquardt.jl:65

Arguments:

• d::AbstractDict
• madsdata::AbstractDict : MADS problem dictionary
• resultvec::AbstractVector : result vector

Keywords:

• filter

Turn off parallel optimization of jacobians and lambdas

Methods:

• Mads.parallel_optimization_off() : /home/travis/build/madsjulia/Mads.jl/src/MadsHelpers.jl:229

Turn on parallel optimization of jacobians and lambdas

Methods:

• Mads.parallel_optimization_on() : /home/travis/build/madsjulia/Mads.jl/src/MadsHelpers.jl:220

Convert a parameter array to a parameter dictionary of arrays

Methods:

• Mads.paramarray2dict(madsdata::AbstractDict, array::Array) : /home/travis/build/madsjulia/Mads.jl/src/MadsMonteCarlo.jl:278

Arguments:

• array::Array : parameter array
• madsdata::AbstractDict : MADS problem dictionary

Returns:

• a parameter dictionary of arrays

Convert a parameter dictionary of arrays to a parameter array

Methods:

• Mads.paramdict2array(dict::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsMonteCarlo.jl:297

Arguments:

• dict::AbstractDict : parameter dictionary of arrays

Returns:

• a parameter array

Parse loaded MADS problem dictionary

Methods:

• Mads.parsemadsdata!(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:288

Arguments:

• madsdata::AbstractDict : MADS problem dictionary

Parse string with node names defined in SLURM

Methods:

• Mads.parsenodenames(nodenames::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsParallel.jl:208
• Mads.parsenodenames(nodenames::AbstractString, ntasks_per_node::Integer) : /home/travis/build/madsjulia/Mads.jl/src/MadsParallel.jl:208

Arguments:

• nodenames::AbstractString : string with node names defined in SLURM
• ntasks_per_node::Integer : number of parallel tasks per node [default=1]

Returns:

• vector with names of compute nodes (hosts)

Compute the sum of squared residuals for observations that match a regular expression

Methods:

• Mads.partialof(madsdata::AbstractDict, resultdict::AbstractDict, regex::Regex) : /home/travis/build/madsjulia/Mads.jl/src/MadsLevenbergMarquardt.jl:91

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• regex::Regex : regular expression
• resultdict::AbstractDict : result dictionary

Returns:

• the sum of squared residuals for observations that match the regular expression

Get package version

Methods:

• Mads.pkgversion_old(modulestr::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsHelpers.jl:545

Arguments:

• modulestr::AbstractString

Returns:

• package version

Print sensitivity analysis results

Methods:

• Mads.printSAresults(madsdata::AbstractDict, results::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsSensitivityAnalysis.jl:946

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• results::AbstractDict : dictionary with sensitivity analysis results

Print sensitivity analysis results (method 2)

Methods:

• Mads.printSAresults2(madsdata::AbstractDict, results::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsSensitivityAnalysis.jl:1028

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• results::AbstractDict : dictionary with sensitivity analysis results

Print error message

Methods:

• Mads.printerrormsg(errmsg) : /home/travis/build/madsjulia/Mads.jl/src/MadsHelpers.jl:438

Arguments:

• errmsg : error message

Print (emit) observations in the MADS problem dictionary

Methods:

• Mads.printobservations(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:435
• Mads.printobservations(madsdata::AbstractDict, filename::AbstractString; json) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:443
• Mads.printobservations(madsdata::AbstractDict, io::IO) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:435
• Mads.printobservations(madsdata::AbstractDict, io::IO, obskeys::AbstractVector) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:435

Arguments:

• filename::AbstractString : output file name
• io::IO : output stream
• madsdata::AbstractDict : MADS problem dictionary
• obskeys::AbstractVector

Keywords:

• json

Pull (checkout) the latest version of Mads modules

Methods:

• Mads.pull() : /home/travis/build/madsjulia/Mads.jl/src/MadsModules.jl:91
• Mads.pull(modulename::AbstractString; kw...) : /home/travis/build/madsjulia/Mads.jl/src/MadsModules.jl:91

Arguments:

• modulename::AbstractString : module name

Push the latest version of Mads modules in the default remote repository

Methods:

• Mads.push() : /home/travis/build/madsjulia/Mads.jl/src/MadsModules.jl:166
• Mads.push(modulename::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsModules.jl:166

Arguments:

• modulename::AbstractString : module name

Make MADS not quiet

Methods:

• Mads.quietoff() : /home/travis/build/madsjulia/Mads.jl/src/MadsHelpers.jl:112

Make MADS quiet

Methods:

• Mads.quieton() : /home/travis/build/madsjulia/Mads.jl/src/MadsHelpers.jl:103

Read MADS predictions from an ASCII file

Methods:

• Mads.readasciipredictions(filename::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsASCII.jl:43

Arguments:

• filename::AbstractString : ASCII file name

Returns:

• MADS predictions

Read model outputs saved for MADS

Methods:

• Mads.readmodeloutput(madsdata::AbstractDict; obskeys) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:910

Arguments:

• madsdata::AbstractDict : MADS problem dictionary

Keywords:

• obskeys : observation keys [default=getobskeys(madsdata)]

Read observations

Methods:

• Mads.readobservations(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:1277
• Mads.readobservations(madsdata::AbstractDict, obskeys::AbstractVector) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:1277

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• obskeys::AbstractVector : observation keys [default=getobskeys(madsdata)]

Returns:

• dictionary with Mads observations

Read observations using C MADS dynamic library

Methods:

• Mads.readobservations_cmads(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsCMads.jl:13

Arguments:

• madsdata::AbstractDict : Mads problem dictionary

Returns:

• observations

Read MADS model predictions from a YAML file filename

Methods:

• Mads.readyamlpredictions(filename::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsYAML.jl:104

Arguments:

• filename::AbstractString : file name

Returns:

• data in yaml input file

Create directories recursively (if does not already exist)

Methods:

• Mads.recursivemkdir(s::AbstractString; filename) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:1514

Arguments:

• s::AbstractString

Keywords:

• filename

Remove directories recursively

Methods:

• Mads.recursivermdir(s::AbstractString; filename) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:1559

Arguments:

• s::AbstractString

Keywords:

• filename

Get observations for a set of regular expressions

Methods:

• Mads.regexs2obs(obsline::AbstractString, regexs::Vector{Regex}, obsnames::Vector{String}, getparamhere::Vector{Bool}) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:1178

Arguments:

• getparamhere::Vector{Bool} : parameters
• obsline::AbstractString : observation line
• obsnames::Vector{String} : observation names
• regexs::Vector{Regex} : regular expressions

Returns:

• obsdict : observations

Remove a contamination source

Methods:

• Mads.removesource!(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsAnasol.jl:49
• Mads.removesource!(madsdata::AbstractDict, sourceid::Int64) : /home/travis/build/madsjulia/Mads.jl/src/MadsAnasol.jl:49

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• sourceid::Int64 : source id [default=0]

Remove contaminant source parameters

Methods:

• Mads.removesourceparameters!(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsAnasol.jl:134

Arguments:

• madsdata::AbstractDict : MADS problem dictionary

Lists modules required by a module (Mads by default)

Methods:

• Mads.required() : /home/travis/build/madsjulia/Mads.jl/src/MadsModules.jl:45
• Mads.required(modulename::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsModules.jl:45
• Mads.required(modulename::AbstractString, filtermodule::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsModules.jl:45

Arguments:

• filtermodule::AbstractString : filter module name
• modulename::AbstractString : module name [default="Mads"]

Returns:

• filtered modules

Reset the model runs count to be equal to zero

Methods:

• Mads.resetmodelruns() : /home/travis/build/madsjulia/Mads.jl/src/MadsHelpers.jl:268

Compute residuals

Methods:

• Mads.residuals(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsLevenbergMarquardt.jl:32
• Mads.residuals(madsdata::AbstractDict, resultdict::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsLevenbergMarquardt.jl:29
• Mads.residuals(madsdata::AbstractDict, resultvec::AbstractVector) : /home/travis/build/madsjulia/Mads.jl/src/MadsLevenbergMarquardt.jl:6

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• resultdict::AbstractDict : result dictionary
• resultvec::AbstractVector : result vector

Returns:

MADS restart off

Methods:

• Mads.restartoff() : /home/travis/build/madsjulia/Mads.jl/src/MadsHelpers.jl:84

MADS restart on

Methods:

• Mads.restarton() : /home/travis/build/madsjulia/Mads.jl/src/MadsHelpers.jl:75

Reweigh samples using importance sampling – returns a vector of log-likelihoods after reweighing

Methods:

• Mads.reweighsamples(madsdata::AbstractDict, predictions::Array, oldllhoods::AbstractVector) : /home/travis/build/madsjulia/Mads.jl/src/MadsSensitivityAnalysis.jl:331

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• oldllhoods::AbstractVector : the log likelihoods of the parameters in the old distribution
• predictions::Array : the model predictions for each of the samples

Returns:

• vector of log-likelihoods after reweighing

Remove directory

Methods:

• Mads.rmdir(dir::AbstractString; path) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:1374

Arguments:

• dir::AbstractString : directory to be removed

Keywords:

• path : path of the directory [default=current path]

Remove file

Methods:

• Mads.rmfile(filename::AbstractString; path) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:1390

Arguments:

• filename::AbstractString : file to be removed

Keywords:

• path : path of the file [default=current path]

Remove files

Methods:

• Mads.rmfile(filename::AbstractString; path) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:1390

Arguments:

• filename::AbstractString

Keywords:

• path : path of the file [default=current path]

Remove files with extension ext

Methods:

• Mads.rmfiles_ext(ext::AbstractString; path) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:1419

Arguments:

• ext::AbstractString : extension

Keywords:

• path : path of the files to be removed [default=.]

Remove files with root root

Methods:

• Mads.rmfiles_root(root::AbstractString; path) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:1432

Arguments:

• root::AbstractString : root

Keywords:

• path : path of the files to be removed [default=.]

Rosenbrock test function

Methods:

• Mads.rosenbrock(x::AbstractVector) : /home/travis/build/madsjulia/Mads.jl/src/MadsTestFunctions.jl:40

Arguments:

• x::AbstractVector : parameter vector

Returns:

• test result

Parameter gradients of the Rosenbrock test function

Methods:

• Mads.rosenbrock2_gradient_lm(x::AbstractVector) : /home/travis/build/madsjulia/Mads.jl/src/MadsTestFunctions.jl:21

Arguments:

• x::AbstractVector : parameter vector

Returns:

• parameter gradients

Rosenbrock test function (more difficult to solve)

Methods:

• Mads.rosenbrock2_lm(x::AbstractVector) : /home/travis/build/madsjulia/Mads.jl/src/MadsTestFunctions.jl:7

Arguments:

• x::AbstractVector : parameter vector

Parameter gradients of the Rosenbrock test function

Methods:

• Mads.rosenbrock_gradient!(x::AbstractVector, grad::AbstractVector) : /home/travis/build/madsjulia/Mads.jl/src/MadsTestFunctions.jl:65

Arguments:

• grad::AbstractVector : gradient vector
• x::AbstractVector : parameter vector

Parameter gradients of the Rosenbrock test function for LM optimization (returns the gradients for the 2 components separately)

Methods:

• Mads.rosenbrock_gradient_lm(x::AbstractVector; dx, center) : /home/travis/build/madsjulia/Mads.jl/src/MadsTestFunctions.jl:82

Arguments:

• x::AbstractVector : parameter vector

Keywords:

• center : array with parameter observations at the center applied to compute numerical derivatives [default=Vector{Float64}(undef, 0)]
• dx : apply parameter step to compute numerical derivatives [default=false]

Returns:

• parameter gradients

Parameter Hessian of the Rosenbrock test function

Methods:

• Mads.rosenbrock_hessian!(x::AbstractVector, hess::AbstractMatrix) : /home/travis/build/madsjulia/Mads.jl/src/MadsTestFunctions.jl:98

Arguments:

• hess::AbstractMatrix : Hessian matrix
• x::AbstractVector : parameter vector

Rosenbrock test function for LM optimization (returns the 2 components separately)

Methods:

• Mads.rosenbrock_lm(x::AbstractVector) : /home/travis/build/madsjulia/Mads.jl/src/MadsTestFunctions.jl:54

Arguments:

• x::AbstractVector : parameter vector

Returns:

• test result

Run external command and pipe stdout and stderr

Methods:

• Mads.runcmd(cmd::Cmd; quiet, pipe, waittime) : /home/travis/build/madsjulia/Mads.jl/src/MadsExecute.jl:39
• Mads.runcmd(cmdstring::AbstractString; quiet, pipe, waittime) : /home/travis/build/madsjulia/Mads.jl/src/MadsExecute.jl:98

Arguments:

• cmd::Cmd : command (as a julia command; e.g. `ls`)
• cmdstring::AbstractString : command (as a string; e.g. "ls")

Keywords:

• pipe : [default=false]
• quiet : [default=Mads.quiet]
• waittime : wait time is second [default=Mads.executionwaittime]

Returns:

• command output
• command error message

Run remote command on a series of servers

Methods:

• Mads.runremote(cmd::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsParallel.jl:284
• Mads.runremote(cmd::AbstractString, nodenames::Vector{String}) : /home/travis/build/madsjulia/Mads.jl/src/MadsParallel.jl:284

Arguments:

• cmd::AbstractString : remote command
• nodenames::Vector{String} : names of machines/nodes [default=madsservers]

Returns:

• output of running remote command

Saltelli sensitivity analysis

Methods:

• Mads.saltelli(madsdata::AbstractDict; N, seed, rng, restartdir, parallel, checkpointfrequency, save, load) : /home/travis/build/madsjulia/Mads.jl/src/MadsSensitivityAnalysis.jl:644

Arguments:

• madsdata::AbstractDict : MADS problem dictionary

Keywords:

• N : number of samples [default=100]
• checkpointfrequency : check point frequency [default=N]
• load
• parallel : set to true if the model runs should be performed in parallel [default=false]
• restartdir : directory where files will be stored containing model results for fast simulation restarts
• rng
• save
• seed : random seed [default=0]

Saltelli sensitivity analysis (brute force)

Methods:

• Mads.saltellibrute(madsdata::AbstractDict; N, seed, rng, restartdir) : /home/travis/build/madsjulia/Mads.jl/src/MadsSensitivityAnalysis.jl:456

Arguments:

• madsdata::AbstractDict : MADS problem dictionary

Keywords:

• N : number of samples [default=1000]
• restartdir : directory where files will be stored containing model results for fast simulation restarts
• rng
• seed : random seed [default=0]

Parallel version of saltellibrute

Parallel version of saltelli

Methods:

• Mads.sampling(param::AbstractVector, J::Array, numsamples::Number; seed, rng, scale) : /home/travis/build/madsjulia/Mads.jl/src/MadsSensitivityAnalysis.jl:280

Arguments:

• J::Array : Jacobian matrix
• numsamples::Number : Number of samples
• param::AbstractVector : Parameter vector

Keywords:

• rng
• scale : data scaling [default=1]
• seed : random esee [default=0]

Returns:

• generated samples (vector or array)
• vector of log-likelihoods

Save MADS problem dictionary madsdata in MADS input file filename

Methods:

• Mads.savemadsfile(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:432
• Mads.savemadsfile(madsdata::AbstractDict, filename::AbstractString; observations_separate, filenameobs) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:432
• Mads.savemadsfile(madsdata::AbstractDict, parameters::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:449
• Mads.savemadsfile(madsdata::AbstractDict, parameters::AbstractDict, filename::AbstractString; explicit, observations_separate) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:449

Arguments:

• filename::AbstractString : input file name (e.g. input_file_name.mads)
• madsdata::AbstractDict : MADS problem dictionary
• parameters::AbstractDict : Dictionary with parameters (optional)

Keywords:

• explicit : if true ignores MADS YAML file modifications and rereads the original input file [default=false]
• filenameobs
• observations_separate

Example:

Mads.savemadsfile(madsdata)
Mads.savemadsfile(madsdata, "test.mads")
Mads.savemadsfile(madsdata, parameters, "test.mads")
Mads.savemadsfile(madsdata, parameters, "test.mads", explicit=true)

Save MCMC chain in a file

Methods:

• Mads.savemcmcresults(chain::Array, filename::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsMonteCarlo.jl:163

Arguments:

• chain::Array : MCMC chain
• filename::AbstractString : file name

Dumps:

• the file containing MCMC chain

Save Saltelli sensitivity analysis results for fast simulation restarts

Methods:

• Mads.savesaltellirestart(evalmat::Array, matname::AbstractString, restartdir::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsSensitivityAnalysis.jl:625

Arguments:

• evalmat::Array : saved array
• matname::AbstractString : matrix (array) name (defines the name of the loaded file)
• restartdir::AbstractString : directory where files will be stored containing model results for fast simulation restarts

Get files in the current directory or in a directory defined by path matching pattern key which can be a string or regular expression

Methods:

• Mads.searchdir(key::AbstractString; path) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:956
• Mads.searchdir(key::Regex; path) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:955

Arguments:

• key::AbstractString : matching pattern for Mads input files (string or regular expression accepted)
• key::Regex : matching pattern for Mads input files (string or regular expression accepted)

Keywords:

• path : search directory for the mads input files [default=.]

Returns:

• filename : an array with file names matching the pattern in the specified directory

Examples:

- `Mads.searchdir("a")`
- `Mads.searchdir(r"[A-B]"; path = ".")`
- `Mads.searchdir(r".*.cov"; path = ".")`

Set number of processors needed for each parallel task at each node

Methods:

• Mads.set_nprocs_per_task() : /home/travis/build/madsjulia/Mads.jl/src/MadsHelpers.jl:66
• Mads.set_nprocs_per_task(local_nprocs_per_task::Integer) : /home/travis/build/madsjulia/Mads.jl/src/MadsHelpers.jl:66

Arguments:

• local_nprocs_per_task::Integer

Set all parameters OFF

Methods:

• Mads.setallparamsoff!(madsdata::AbstractDict; filter) : /home/travis/build/madsjulia/Mads.jl/src/MadsParameters.jl:466

Arguments:

• madsdata::AbstractDict : MADS problem dictionary

Keywords:

• filter : parameter filter

Set all parameters ON

Methods:

• Mads.setallparamson!(madsdata::AbstractDict; filter) : /home/travis/build/madsjulia/Mads.jl/src/MadsParameters.jl:452

Arguments:

• madsdata::AbstractDict : MADS problem dictionary

Keywords:

• filter : parameter filter

Set MADS debug level

Methods:

• Mads.setdebuglevel(level::Int64) : /home/travis/build/madsjulia/Mads.jl/src/MadsHelpers.jl:239

Arguments:

• level::Int64 : debug level

Set the working directory (for parallel environments)

Methods:

• Mads.setdir() : /home/travis/build/madsjulia/Mads.jl/src/MadsParallel.jl:254
• Mads.setdir(dir) : /home/travis/build/madsjulia/Mads.jl/src/MadsParallel.jl:249

Arguments:

• dir : directory

Example:

@Distributed.everywhere Mads.setdir()
@Distributed.everywhere Mads.setdir("/home/monty")

Set image dpi

Methods:

• Mads.setdpi(dpi::Integer) : /home/travis/build/madsjulia/Mads.jl/src/MadsHelpers.jl:175

Arguments:

• dpi::Integer

Set maximum execution wait time for forward model runs in seconds

Methods:

• Mads.setexecutionwaittime(waitime::Float64) : /home/travis/build/madsjulia/Mads.jl/src/MadsHelpers.jl:259

Arguments:

• waitime::Float64 : maximum execution wait time for forward model runs in seconds

Set a default MADS input file

Methods:

• Mads.setmadsinputfile(filename::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:509

Arguments:

• filename::AbstractString : input file name (e.g. input_file_name.mads)

Generate a list of Mads servers

Methods:

• Mads.setmadsservers() : /home/travis/build/madsjulia/Mads.jl/src/MadsParallel.jl:339
• Mads.setmadsservers(first::Int64) : /home/travis/build/madsjulia/Mads.jl/src/MadsParallel.jl:339
• Mads.setmadsservers(first::Int64, last::Int64) : /home/travis/build/madsjulia/Mads.jl/src/MadsParallel.jl:339

Arguments:

• first::Int64 : first [default=0]
• last::Int64 : last [default=18]

Returns

• array string of mads servers

Set model input files; delete files where model output should be saved for MADS

Methods:

• Mads.setmodelinputs(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:822
• Mads.setmodelinputs(madsdata::AbstractDict, parameters::AbstractDict; path) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:822

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• parameters::AbstractDict : parameters

Keywords:

• path : path for the files [default=.]

Set new mads file name

Methods:

• Mads.setnewmadsfilename(filename::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:672
• Mads.setnewmadsfilename(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsIO.jl:669

Arguments:

• filename::AbstractString : file name
• madsdata::AbstractDict : MADS problem dictionary

Returns:

• new file name

Set observations (calibration targets) in the MADS problem dictionary based on a predictions dictionary

Methods:

• Mads.setobservationtargets!(madsdata::AbstractDict, predictions::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:539

Arguments:

• madsdata::AbstractDict : Mads problem dictionary
• predictions::AbstractDict : dictionary with model predictions

Set observation time based on the observation name in the MADS problem dictionary

Methods:

• Mads.setobstime!(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:250
• Mads.setobstime!(madsdata::AbstractDict, rx::Regex) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:260
• Mads.setobstime!(madsdata::AbstractDict, rx::Regex, obskeys::AbstractVector) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:260
• Mads.setobstime!(madsdata::AbstractDict, separator::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:250
• Mads.setobstime!(madsdata::AbstractDict, separator::AbstractString, obskeys::AbstractVector) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:250

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• obskeys::AbstractVector
• rx::Regex : regular expression to match
• separator::AbstractString : separator [default=_]

Examples:

Mads.setobstime!(madsdata, "_t")
Mads.setobstime!(madsdata, r"[A-x]*_t([0-9,.]+)")

Set observation weights in the MADS problem dictionary

Methods:

• Mads.setobsweights!(madsdata::AbstractDict, v::AbstractVector) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:292
• Mads.setobsweights!(madsdata::AbstractDict, v::AbstractVector, obskeys::AbstractVector) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:292
• Mads.setobsweights!(madsdata::AbstractDict, value::Number) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:287
• Mads.setobsweights!(madsdata::AbstractDict, value::Number, obskeys::AbstractVector) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:287

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• obskeys::AbstractVector
• v::AbstractVector : vector of observation weights
• value::Number : value for observation weights

Set a specific parameter with a key parameterkey OFF

Methods:

• Mads.setparamoff!(madsdata::AbstractDict, parameterkey::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsParameters.jl:491

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• parameterkey::AbstractString : parameter key

Set a specific parameter with a key parameterkey ON

Methods:

• Mads.setparamon!(madsdata::AbstractDict, parameterkey::AbstractString) : /home/travis/build/madsjulia/Mads.jl/src/MadsParameters.jl:480

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• parameterkey::AbstractString : parameter key

Set normal parameter distributions for all the model parameters in the MADS problem dictionary

Methods:

• Mads.setparamsdistnormal!(madsdata::AbstractDict, mean::AbstractVector, stddev::AbstractVector) : /home/travis/build/madsjulia/Mads.jl/src/MadsParameters.jl:503

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• mean::AbstractVector : array with the mean values
• stddev::AbstractVector : array with the standard deviation values

Set uniform parameter distributions for all the model parameters in the MADS problem dictionary

Methods:

• Mads.setparamsdistuniform!(madsdata::AbstractDict, min::AbstractVector, max::AbstractVector) : /home/travis/build/madsjulia/Mads.jl/src/MadsParameters.jl:518

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• max::AbstractVector : array with the maximum values
• min::AbstractVector : array with the minimum values

Set initial optimized parameter guesses in the MADS problem dictionary

Methods:

• Mads.setparamsinit!(madsdata::AbstractDict, paramdict::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsParameters.jl:319
• Mads.setparamsinit!(madsdata::AbstractDict, paramdict::AbstractDict, idx::Int64) : /home/travis/build/madsjulia/Mads.jl/src/MadsParameters.jl:319

Arguments:

• idx::Int64 : index of the dictionary of arrays with initial model parameter values
• madsdata::AbstractDict : MADS problem dictionary
• paramdict::AbstractDict : dictionary with initial model parameter values

Set the available processors based on environmental variables (supports SLURM only at the moment)

Methods:

• Mads.setprocs(; ntasks_per_node, nprocs_per_task, nodenames, mads_servers, test, quiet, veryquiet, dir, exename) : /home/travis/build/madsjulia/Mads.jl/src/MadsParallel.jl:47
• Mads.setprocs(np::Integer) : /home/travis/build/madsjulia/Mads.jl/src/MadsParallel.jl:44
• Mads.setprocs(np::Integer, nt::Integer) : /home/travis/build/madsjulia/Mads.jl/src/MadsParallel.jl:31

Arguments:

• np::Integer : number of processors [default=1]
• nt::Integer : number of threads[default=1]

Keywords:

• dir : common directory shared by all the jobs
• exename : location of the julia executable (the same version of julia is needed on all the workers)
• mads_servers : if true use MADS servers [default=false]
• nodenames : array with names of machines/nodes to be invoked
• nprocs_per_task : number of processors needed for each parallel task at each node [default=Mads.nprocs_per_task]
• ntasks_per_node : number of parallel tasks per node [default=0]
• quiet : suppress output [default=Mads.quiet]
• test : test the servers and connect to each one ones at a time [default=false]
• veryquiet

Returns:

• vector with names of compute nodes (hosts)

Example:

Mads.setprocs()
Mads.setprocs(4)
Mads.setprocs(4, 8)
Mads.setprocs(ntasks_per_node=4)
Mads.setprocs(ntasks_per_node=32, mads_servers=true)
Mads.setprocs(ntasks_per_node=64, nodenames=madsservers)
Mads.setprocs(ntasks_per_node=64, nodenames=["madsmax", "madszem"])
Mads.setprocs(ntasks_per_node=64, nodenames="wc[096-157,160,175]")
Mads.setprocs(ntasks_per_node=64, mads_servers=true, exename="/home/monty/bin/julia", dir="/home/monty")

Set / get current random seed. seed < 0 gets seed, anything else sets it.

Methods:

• Mads.setseed() : /home/travis/build/madsjulia/Mads.jl/src/MadsHelpers.jl:480
• Mads.setseed(seed::Integer) : /home/travis/build/madsjulia/Mads.jl/src/MadsHelpers.jl:480
• Mads.setseed(seed::Integer, quiet::Bool; rng) : /home/travis/build/madsjulia/Mads.jl/src/MadsHelpers.jl:480

Arguments:

• quiet::Bool : [default=true]
• seed::Integer : random seed

Keywords:

• rng

Set initial optimized parameter guesses in the MADS problem dictionary for the Source class

Methods:

• Mads.setparamsinit!(madsdata::AbstractDict, paramdict::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsParameters.jl:319
• Mads.setparamsinit!(madsdata::AbstractDict, paramdict::AbstractDict, idx::Int64) : /home/travis/build/madsjulia/Mads.jl/src/MadsParameters.jl:319

Arguments:

• idx::Int64 : index of the dictionary of arrays with initial model parameter values
• madsdata::AbstractDict : MADS problem dictionary
• paramdict::AbstractDict : dictionary with initial model parameter values

Set observation target

Methods:

• Mads.settarget!(o::AbstractDict, target::Number) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:240

Arguments:

• o::AbstractDict : observation data
• target::Number : observation target

Set observation time

Methods:

• Mads.settime!(o::AbstractDict, time::Number) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:162

Arguments:

• o::AbstractDict : observation data
• time::Number : observation time

Set MADS verbosity level

Methods:

• Mads.setverbositylevel(level::Int64) : /home/travis/build/madsjulia/Mads.jl/src/MadsHelpers.jl:249

Arguments:

• level::Int64 : debug level

Set observation weight

Methods:

• Mads.setweight!(o::AbstractDict, weight::Number) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:201

Arguments:

• o::AbstractDict : observation data
• weight::Number : observation weight

Set well weights in the MADS problem dictionary

Methods:

• Mads.setwellweights!(madsdata::AbstractDict, value::Number) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:343
• Mads.setwellweights!(madsdata::AbstractDict, value::Number, wellkeys::AbstractVector) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:343

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• value::Number : value for well weights
• wellkeys::AbstractVector

Show all parameters in the MADS problem dictionary

Methods:

• Mads.showallparameters(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsParameters.jl:579
• Mads.showallparameters(madsdata::AbstractDict, parkeys::AbstractVector) : /home/travis/build/madsjulia/Mads.jl/src/MadsParameters.jl:579
• Mads.showallparameters(madsdata::AbstractDict, result::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsParameters.jl:583

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• parkeys::AbstractVector
• result::AbstractDict

Show observations in the MADS problem dictionary

Methods:

• Mads.showobservations(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:399
• Mads.showobservations(madsdata::AbstractDict, obskeys::AbstractVector) : /home/travis/build/madsjulia/Mads.jl/src/MadsObservations.jl:399

Arguments:

• madsdata::AbstractDict : MADS problem dictionary
• obskeys::AbstractVector

Show parameters in the MADS problem dictionary

Methods:

• Mads.showparameters(madsdata::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsParameters.jl:565
• Mads.showparameters(madsdata::AbstractDict, parkeys::AbstractVector) : /home/travis/build/madsjulia/Mads.jl/src/MadsParameters.jl:565
• Mads.showparameters(madsdata::AbstractDict, parkeys::AbstractVector, all::Bool) : /home/travis/build/madsjulia/Mads.jl/src/MadsParameters.jl:565
• Mads.showparameters(madsdata::AbstractDict, result::AbstractDict) : /home/travis/build/madsjulia/Mads.jl/src/MadsParameters.jl:560

Arguments:

• all::Bool
• madsdata::AbstractDict : MADS problem dictionary
• parkeys::AbstractVector
• result::AbstractDict

Sine transformation of model parameters

Methods:

• Mads.sinetransform(madsdata::AbstractDict, params::AbstractVector) : /home/travis/build/madsjulia/Mads.jl/src/MadsSineTransformations.jl:33
• Mads.sinetransform(sineparams::AbstractVector, lowerbounds::AbstractVector, upperbounds::AbstractVector, indexlogtransformed::AbstractVector) : /home/travis/build/madsjulia/Mads.jl/src/MadsSineTransformations.jl:43

Arguments:

• indexlogtransformed::AbstractVector : index vector of log-transformed parameters
• lowerbounds::AbstractVector : lower bounds
• madsdata::AbstractDict : MADS problem dictionary
• params::AbstractVector
• sineparams::AbstractVector : model parameters
• upperbounds::AbstractVector : upper bounds

Returns:

• Sine transformation of model parameters

Sine transformation of a function

Methods:

• Mads.sinetransformfunction(f::Function, lowerbounds::AbstractVector, upperbounds::AbstractVector, indexlogtransformed::AbstractVector) : /home/travis/build/madsjulia/Mads.jl/src/MadsSineTransformations.jl:77

Arguments:

• f::Function : function
• indexlogtransformed::AbstractVector : index vector of log-transformed parameters
• lowerbounds::AbstractVector : lower bounds
• upperbounds::AbstractVector : upper bounds

Returns:

• Sine transformation