comma fixes

examples/anasol/runtests.jl

@@ -162,7 +162,7 @@ good_params_init_max = JLD2.load(joinpath(d, "params_init_max.jld2"), "m4")
 good_targetkeys = JLD2.load(joinpath(d, "targetkeys.jld2"), "tk")
 
 Test.@testset "Anasol" begin
-	# Test Mads.forward(md, all=true)
+	# Test Mads.forward(md; all=true)
 	ssr = 0.
 	for obskey in union(Set(keys(forward_results)), Set(keys(good_forward_results)))
 		ssr += (forward_results[obskey] - good_forward_results[obskey])^2
@@ -214,7 +214,7 @@ end
 
 md = Mads.loadmadsfile(joinpath(workdir, "w01shortexp.mads"))
 md["Restart"] = true
-Mads.localsa(md, filename="w01shortexp.mads", par=Mads.getparamsinit(md, Mads.getoptparamkeys(md)))
+Mads.localsa(md; filename="w01shortexp.png", par=Mads.getparamsinit(md, Mads.getoptparamkeys(md)))
 if !haskey(ENV, "MADS_NO_GADFLY")
 	Mads.graphoff()
 	Mads.plotlocalsa("w01shortexp")

examples/arctic/test.jl

@@ -92,6 +92,6 @@ Mads.showobservations(md)
 
 forward_model = Mads.forward(md)
 
-calib_param, calib_information = Mads.calibrate(md, tolOF=0.01, tolOFcount=4)
+calib_param, calib_information = Mads.calibrate(md; tolOF=0.01, tolOFcount=4)
 
-Mads.plotmatches(md, calib_param, xtitle="# of observations", ytitle = "GPP",filename=mads_config["mads_problemname"]*".png")
+Mads.plotmatches(md, calib_param; xtitle="# of observations", ytitle = "GPP",filename=mads_config["mads_problemname"]*".png")

examples/bayesian_sampling/bayesian_sampling_anasol.jl

@@ -21,7 +21,7 @@ Mads.setparamsinit!(md, opt_param)
 Mads.plotmatches(md)
 
 Mads.madsinfo("Bayesian sampling of contaminant transport problem ...")
-mcmcchains = Mads.bayessampling(md, seed=2016)
+mcmcchains = Mads.bayessampling(md; seed=2016)
 
 Mads.madsinfo("Bayesian scatter plots ...")
 Mads.scatterplotsamples(md, permutedims(mcmcchains.value), rootname * "-bayes-results.svg")

examples/contamination/contamination-efast.jl

@@ -11,5 +11,5 @@ Mads.wellon!(md, "w20a") # use well w20a
 if !isdefined(Mads, :efastresult)
 	efast_results = Mads.efast(md; N=1000, seed=2016)
 end
-Mads.plotobsSAresults(md, efastresult, filter=r"w13a", filename="w13a-efast.svg", xtitle = "Time [a]", ytitle = "Concentration [ppb]", separate_files=true)
-Mads.plotobsSAresults(md, efastresult, filter=r"w20a", filename="w20a-efast.svg", xtitle = "Time [a]", ytitle = "Concentration [ppb]", separate_files=true)
+Mads.plotobsSAresults(md, efastresult; filter=r"w13a", filename="w13a-efast.svg", xtitle = "Time [a]", ytitle = "Concentration [ppb]", separate_files=true)
+Mads.plotobsSAresults(md, efastresult; filter=r"w20a", filename="w20a-efast.svg", xtitle = "Time [a]", ytitle = "Concentration [ppb]", separate_files=true)

examples/contamination/contamination-saltelli.jl

@@ -11,5 +11,5 @@ Mads.wellon!(md, "w20a") # use well w20a
 if !isdefined(Mads, :saltelliresult)
 	saltelli_results = Mads.saltelli(md; N=10000, seed=2016)
 end
-Mads.plotobsSAresults(md, saltelliresult, filter=r"w13a", filename="w13a-saltelli.svg", xtitle = "Time [a]", ytitle = "Concentration [ppb]", separate_files=true)
-Mads.plotobsSAresults(md, saltelliresult, filter=r"w20a", filename="w20a-saltelli.svg", xtitle = "Time [a]", ytitle = "Concentration [ppb]", separate_files=true)
+Mads.plotobsSAresults(md, saltelliresult; filter=r"w13a", filename="w13a-saltelli.svg", xtitle = "Time [a]", ytitle = "Concentration [ppb]", separate_files=true)
+Mads.plotobsSAresults(md, saltelliresult; filter=r"w20a", filename="w20a-saltelli.svg", xtitle = "Time [a]", ytitle = "Concentration [ppb]", separate_files=true)

examples/contamination/contamination-w03-w13-svr-efast-mads.jl

@@ -15,18 +15,18 @@ paramarray = hcat(map(i->collect(paramdict[i]), keys(paramdict))...)
 @time predictions = Mads.forward(md, paramdict)'
 
 Mads.madsinfo("Model predictions ...")
-Mads.spaghettiplot(md, predictions, keyword="w13a-model", format="PNG")
+Mads.spaghettiplot(md, predictions; keyword="w13a-model", format="PNG")
 Mads.display("$rootname-w13a-model-$numberofsamples-spaghetti.png")
 
 @time svrpredictions = svrexec(paramarray)
 @info("SVR discrepancy $(maximum(abs.(svrpredictions .- predictions)))")
 
 Mads.madsinfo("SVR predictions ...")
-Mads.spaghettiplot(md, svrpredictions, keyword="w13a-svr", format="PNG")
+Mads.spaghettiplot(md, svrpredictions; keyword="w13a-svr", format="PNG")
 Mads.display("$rootname-w13a-svr-$numberofsamples-spaghetti.png")
 
 sa = Mads.efast(md)
-Mads.plotobsSAresults(md, sa, format="PNG")
+Mads.plotobsSAresults(md, sa; format="PNG")
 Mads.display("$rootname-efast-385.png")
 
 mdsvr = deepcopy(md)

examples/contamination/contamination-w03-w13-svr-efast.jl

@@ -15,7 +15,7 @@ if !isdefined(Mads, :predictions) || size(predictions) != numberofsamples
 end
 
 Mads.madsinfo("Training set spaghetti plot ...")
-Mads.spaghettiplot(md, predictions, keyword="w13a-training-set", format="PNG")
+Mads.spaghettiplot(md, predictions; keyword="w13a-training-set", format="PNG")
 Mads.display("$rootname-w13a-training-set-$numberofsamples-spaghetti.png")
 
 svrpredictions = Array{Float64}(undef, 0, numberofsamples)
@@ -28,5 +28,5 @@ end
 @show maximum(abs.(svrpredictions .- predictions))
 
 Mads.madsinfo("SVR predictions ...")
-Mads.spaghettiplot(md, svrpredictions, keyword="w13a-svr", format="PNG")
+Mads.spaghettiplot(md, svrpredictions; keyword="w13a-svr", format="PNG")
 Mads.display("$rootname-w13a-svr-$numberofsamples-spaghetti.png")

examples/contamination/contamination-w03.jl

@@ -3,7 +3,7 @@ import Mads
 md = Mads.loadmadsfile("w03.mads") # load Mads input file into Julia Dictionary
 rootname = Mads.getmadsrootname(md) # get problem rootname
 
-Mads.plotmadsproblem(md, keyword="all_wells") # display the well locations and the initial source location
+Mads.plotmadsproblem(md; keyword="all_wells") # display the well locations and the initial source location
 Mads.plotmatches(md) # plot initial matches
 
 Mads.allwellsoff!(md) # turn off all wells
@@ -24,14 +24,14 @@ wd = Mads.getwelldata(md, time=true)'
 
 # Sensitivity analysis: spaghetti plots based on prior parameter uncertainty ranges
 Mads.madsinfo("Prior spaghetti plot ...")
-Mads.spaghettiplot(md, predictions, keyword="w13a_w20a-prior")
+Mads.spaghettiplot(md, predictions; keyword="w13a_w20a-prior")
 
 Mads.madsinfo("Bayesian sampling ...")
-mcmcchain = Mads.bayessampling(md, seed=20151001)
+mcmcchain = Mads.bayessampling(md; seed=20151001)
 
 Mads.madsinfo("Bayesian scatter plots ...")
 Mads.scatterplotsamples(md, permutedims(mcmcchain.value), rootname * "-bayes.png")
 
 posterior_predictions = Mads.forward(md, permutedims(mcmcchain.value))
 Mads.madsinfo("Posterior (Bayesian) spaghetti plot ...")
-Mads.spaghettiplot(md, posterior_predictions, keyword="w13a_w20a-posterior", format="PNG")
+Mads.spaghettiplot(md, posterior_predictions; keyword="w13a_w20a-posterior", format="PNG")

examples/contamination/contamination-w04-w13-svr-efast-mads.jl

@@ -26,7 +26,7 @@ Mads.spaghettiplot(md, svrpredictions, keyword="w13a-svr", format="PNG")
 Mads.display("$rootname-w13a-svr-$numberofsamples-spaghetti.png")
 
 sa = Mads.efast(md)
-Mads.plotobsSAresults(md, sa, format="PNG")
+Mads.plotobsSAresults(md, sa; format="PNG")
 Mads.display("$rootname-efast-385.png")
 
 mdsvr = deepcopy(md)

examples/contamination/contamination-w04-w13-svr-efast.jl

@@ -15,7 +15,7 @@ if !isdefined(Mads, :predictions) || size(predictions) != numberofsamples
 end
 
 Mads.madsinfo("Training set spaghetti plot ...")
-Mads.spaghettiplot(md, predictions, keyword="w13a-training-set", format="PNG")
+Mads.spaghettiplot(md, predictions; keyword="w13a-training-set", format="PNG")
 Mads.display("$rootname-w13a-training-set-$numberofsamples-spaghetti.png")
 
 svrpredictions = Array{Float64}(undef, 0, numberofsamples)
@@ -28,5 +28,5 @@ end
 @show maximum(abs.(svrpredictions .- predictions))
 
 Mads.madsinfo("SVR predictions ...")
-Mads.spaghettiplot(md, svrpredictions, keyword="w13a-svr", format="PNG")
+Mads.spaghettiplot(md, svrpredictions; keyword="w13a-svr", format="PNG")
 Mads.display("$rootname-w13a-svr-$numberofsamples-spaghetti.png")

examples/contamination/contamination.jl

@@ -1,4 +1,4 @@
-import Mads
+using Mads: Mads
 
 currentdir = pwd()
 
@@ -17,18 +17,18 @@ Mads.showallparameters(md) # show all the model parameters
 Mads.showparameters(md) # show all the adjustable model parameters
 
 # use all wells
-Mads.plotmadsproblem(md, keyword="all_wells") # display the well locations and the initial source location
+Mads.plotmadsproblem(md; keyword = "all_wells") # display the well locations and the initial source location
 
 forward_predictions = Mads.forward(md) # execute forward model simulation based on initial parameter guesses
 
-Mads.plotmatches(md, forward_predictions, xtitle = "Time [a]", ytitle = "Concentration [ppb]") # plot initial matches
+Mads.plotmatches(md, forward_predictions; xtitle = "Time [a]", ytitle = "Concentration [ppb]") # plot initial matches
 Mads.display("w01-match.svg")
 
 inverse_parameters, inverse_results = Mads.calibrate(md) # perform model calibration
 
 inverse_predictions = Mads.forward(md, inverse_parameters) # execute forward model simulation based on calibrated values
 
-Mads.plotmatches(md, inverse_predictions, xtitle = "Time [a]", ytitle = "Concentration [ppb]") # plot calibrated matches
+Mads.plotmatches(md, inverse_predictions; xtitle = "Time [a]", ytitle = "Concentration [ppb]") # plot calibrated matches
 Mads.display("w01-match.svg")
 
 # use only two wells
@@ -37,28 +37,28 @@ Mads.wellon!(md, "w13a") # use well w13a
 Mads.wellon!(md, "w20a") # use well w20a
 Mads.showobservations(md) # show all the observations
 
-Mads.plotmadsproblem(md, keyword="w13a_w20a") # display the well locations and the initial source location
+Mads.plotmadsproblem(md; keyword = "w13a_w20a") # display the well locations and the initial source location
 
 forward_predictions = Mads.forward(md) # execute a forward model simulation based on the initial parameter guesses
 
-Mads.plotmatches(md, forward_predictions, filename=rootname * "-w13a_w20a-init-match.svg", xtitle = "Time [a]", ytitle = "Concentration [ppb]") # plot initial matches
+Mads.plotmatches(md, forward_predictions, filename = rootname * "-w13a_w20a-init-match.svg", xtitle = "Time [a]", ytitle = "Concentration [ppb]") # plot initial matches
 Mads.display("w01-w13a_w20a-init-match.svg")
 
 inverse_parameters, inverse_results = Mads.calibrate(md) # perform model calibration
 
 inverse_predictions = Mads.forward(md, inverse_parameters) # execute a forward model simulation based on the calibrated values
 
-Mads.plotmatches(md, inverse_predictions, filename=rootname * "-w13a_w20a-calib-match.svg", xtitle = "Time [a]", ytitle = "Concentration [ppb]") # plot calibrated matches
+Mads.plotmatches(md, inverse_predictions; filename = rootname * "-w13a_w20a-calib-match.svg", xtitle = "Time [a]", ytitle = "Concentration [ppb]") # plot calibrated matches
 Mads.display("w01-w13a_w20a-calib-match.svg")
 
 # Sensitivity analysis: spaghetti plots based on prior parameter uncertainty ranges
 Mads.madsinfo("Prior spaghetti plot ...")
 paramvalues = Mads.getparamrandom(md, 100)
-Mads.spaghettiplot(md, paramvalues, keyword="w13a_w20a-prior", xtitle = "Time [a]", ytitle = "Concentration [ppb]")
+Mads.spaghettiplot(md, paramvalues; keyword = "w13a_w20a-prior", xtitle = "Time [a]", ytitle = "Concentration [ppb]")
 Mads.display("w01-w13a_w20a-prior-100-spaghetti.svg")
 
 Mads.madsinfo("Bayesian sampling ...")
-mcmcchain = Mads.bayessampling(md, seed=20151001)
+mcmcchain = Mads.bayessampling(md, seed = 20151001)
 
 Mads.madsinfo("Bayesian scatter plots ...")
 Mads.scatterplotsamples(md, permutedims(mcmcchain.value), rootname * "-bayes.png")
@@ -68,7 +68,7 @@ Mads.display(rootname * "-bayes.png")
 mcmcvalues = Mads.paramarray2dict(md, permutedims(mcmcchain.value))
 
 Mads.madsinfo("Posterior (Bayesian) spaghetti plot ...")
-Mads.spaghettiplot(md, mcmcvalues, keyword="w13a_w20a-posterior", format="PNG", xtitle = "Time [a]", ytitle = "Concentration [ppb]")
+Mads.spaghettiplot(md, mcmcvalues, keyword = "w13a_w20a-posterior", format = "PNG", xtitle = "Time [a]", ytitle = "Concentration [ppb]")
 Mads.display("w01-w13a_w20a-posterior-1000-spaghetti.png")
 
 # Create a new problem (example)
@@ -90,27 +90,27 @@ md_new["Sources"][1]["gauss"]["dy"]["init"] = 0 # set a point source
 md_new["Sources"][1]["gauss"]["dz"]["init"] = 0 # set a point source
 new_forward_predictions = Mads.forward(md_new) # execute a forward model simulation
 Mads.setobservationtargets!(md_new, new_forward_predictions) # set calibration targets to match the forward model predictions
-Mads.plotmatches(md_new, new_forward_predictions, filename=rootname * "-new-problem.svg", xtitle = "Time [a]", ytitle = "Concentration [ppb]")
+Mads.plotmatches(md_new, new_forward_predictions; filename = rootname * "-new-problem.svg", xtitle = "Time [a]", ytitle = "Concentration [ppb]")
 Mads.display("w01-new-problem.svg")
 Mads.dumpyamlmadsfile(md_new, "w01-new-problem.mads") # write out a new mads input file
 
 # Calibrate with random initial guesses
 Mads.allwellsoff!(md_new) # turn off all wells
 Mads.wellon!(md_new, "w13a") # use well w13a
-Mads.calibraterandom(md_new, 10, seed=20151001) # calibrate 10 times with random initial guesses
+Mads.calibraterandom(md_new, 10; seed = 20151001) # calibrate 10 times with random initial guesses
 
 # Global sensitivity analysis using Saltelli's method
-saltelli_results = Mads.saltelli(md; N=1000, seed=2016)
-Mads.plotobsSAresults(md, saltelliresult, filter=r"w13a", filename="w13a-saltelli.svg", xtitle = "Time [a]", ytitle = "Concentration [ppb]")
+saltelli_results = Mads.saltelli(md; N = 1000, seed = 2016)
+Mads.plotobsSAresults(md, saltelliresult; filter = r"w13a", filename = "w13a-saltelli.svg", xtitle = "Time [a]", ytitle = "Concentration [ppb]")
 Mads.display("w13a-saltelli.svg")
-Mads.plotobsSAresults(md, saltelliresult, filter=r"w20a", filename="w20a-saltelli.svg", xtitle = "Time [a]", ytitle = "Concentration [ppb]")
+Mads.plotobsSAresults(md, saltelliresult; filter = r"w20a", filename = "w20a-saltelli.svg", xtitle = "Time [a]", ytitle = "Concentration [ppb]")
 Mads.display("w20a-saltelli.svg")
 
 # Global sensitivity analysis using Saltelli's eFAST method
-efast_results = Mads.efast(md; N=1000, seed=2016)
-Mads.plotobsSAresults(md, efastresult, filter=r"w13a", filename="w13a-efast.svg", xtitle = "Time [a]", ytitle = "Concentration [ppb]")
+efast_results = Mads.efast(md; N = 1000, seed = 2016)
+Mads.plotobsSAresults(md, efastresult, filter = r"w13a", filename = "w13a-efast.svg", xtitle = "Time [a]", ytitle = "Concentration [ppb]")
 Mads.display("w13a-efast.svg")
-Mads.plotobsSAresults(md, efastresult, filter=r"w20a", filename="w20a-efast.svg", xtitle = "Time [a]", ytitle = "Concentration [ppb]")
+Mads.plotobsSAresults(md, efastresult; filter = r"w20a", filename = "w20a-efast.svg", xtitle = "Time [a]", ytitle = "Concentration [ppb]")
 Mads.display("w20a-efast.svg")
 
-cd(currentdir)
+cd(currentdir)

examples/contamination/runtests.jl

@@ -79,8 +79,8 @@ if !haskey(ENV, "MADS_NO_GADFLY")
 	Mads.plotmatches(md, inverse_predictions; display=false) # plot calibrated matches
 	Mads.rmfile(joinpath(workdir, "w01-w13a_w20a-match.svg"))
 
-	Mads.spaghettiplots(md, paramvalues, keyword="w13a_w20a")
-	Mads.spaghettiplot(md, paramvalues, keyword="w13a_w20a")
+	Mads.spaghettiplots(md, paramvalues; keyword="w13a_w20a")
+	Mads.spaghettiplot(md, paramvalues; keyword="w13a_w20a")
 	s = splitdir(rootname)
 	for filesinadir in Mads.searchdir(Regex(string(s[2], "-w13a_w20a", "[.]*", "spaghetti.svg")), path=s[1])
 		Mads.rmfile(filesinadir, path=s[1])
@@ -89,7 +89,7 @@ if !haskey(ENV, "MADS_NO_GADFLY")
 	sa_results = Mads.saltelli(md; N=5, seed=2015)
 	Mads.plotwellSAresults(md, sa_results)
 	Mads.plotobsSAresults(md, sa_results)
-	Mads.plotobsSAresults(md, sa_results, separate_files=true)
+	Mads.plotobsSAresults(md, sa_results; separate_files=true)
 	Mads.rmfile(joinpath(workdir, "w01-w13a_w20a-saltelli-5-main_effect.svg"))
 	Mads.rmfile(joinpath(workdir, "w01-w13a_w20a-saltelli-5-total_effect.svg"))
 	Mads.rmfile(joinpath(workdir, "w01-w13a_w20a-saltelli-5-variance.svg"))

examples/contamination/w01short.jl

@@ -60,7 +60,7 @@ numberofsamples = 10
 paramvalues=Mads.getparamrandom(md, numberofsamples)
 Mads.allwellsoff!(md)
 Mads.wellon!(md, "w1a")
-Mads.spaghettiplots(md, paramvalues, keyword="w1a")
+Mads.spaghettiplots(md, paramvalues; keyword="w1a")
 
 # solve the inverse problem
 result = Mads.calibrate(md; show_trace=true)

examples/external_execution/exec.jl

@@ -3,5 +3,5 @@ import Mads
 Mads.madsinfo("TEST Saltelli sensitivity analysis:")
 md = Mads.loadmadsfile("exec.mads")
 results = Mads.saltelli(md; N=100)
-results = Mads.saltelliparallel(md, N=50, 2)
+results = Mads.saltelliparallel(md, 2; N=50)
 Mads.printSAresults(md, results)

examples/model_analysis/emcee.jl

@@ -6,6 +6,6 @@ Mads.mkdir("emcee_results")
 
 Mads.setobsweights!(md, 10)
 chain, llhoods = Mads.emceesampling(md; numwalkers=100, nsteps=1000000, burnin=100000, thinning=100, seed=2016, sigma=0.01)
-Mads.scatterplotsamples(md, chain', joinpath("emcee_results", "emcee_scatter.png"))
+Mads.scatterplotsamples(md, chain'; joinpath("emcee_results", "emcee_scatter.png"))
 o = Mads.forward(md, chain')
-Mads.spaghettiplot(md, o, filename=joinpath("emcee_results", "emcee_spaghetti.png"))
+Mads.spaghettiplot(md, o; filename=joinpath("emcee_results", "emcee_spaghetti.png"))

examples/model_analysis/sensitivity.jl

@@ -5,7 +5,7 @@ Mads.mkdir("sensitivity_results")
 
 n = 100
 @info("Calibration using $n random initial guesses for model parameters")
-calib_random_results = Mads.calibraterandom(md, n, all=true, seed=2016, save_results=false)
+calib_random_results = Mads.calibraterandom(md, n; all=true, seed=2016, save_results=false)
 pnames = collect(keys(r[1,3]))
 calib_random_estimates = hcat(map(i->collect(values(calib_random_results[i,3])), 1:n)...)
 np = length(pnames)
@@ -30,10 +30,10 @@ end
 @info("Local sensitivity analysis for the 3 different global optima")
 for i = 1:3
 	Mads.setparamsinit!(md, r[v[i],3])
-	Mads.localsa(md, filename="sensitivity_results/sensitivity_local_$(optnames[i]).png", datafiles=false)
+	Mads.localsa(md; filename="sensitivity_results/sensitivity_local_$(optnames[i]).png", datafiles=false)
 end
 Mads.setparamsinit!(md, pinit)
 
 @info("Global sensitivity analysis")
 efast_results = Mads.efast(md; N=1000, seed=2016)
-Mads.plotobsSAresults(md, efastresult, filename="sensitivity_results/sensitivity_global.png", xtitle = "x", ytitle = "y")
+Mads.plotobsSAresults(md, efastresult; filename="sensitivity_results/sensitivity_global.png", xtitle = "x", ytitle = "y")

examples/model_analysis/svr.jl

@@ -14,14 +14,14 @@ paramarray = hcat(map(i->collect(paramdict[i]), keys(paramdict))...)
 predictions = Mads.forward(md, paramdict)'
 
 Mads.madsinfo("Model predictions ...")
-Mads.spaghettiplot(md, predictions, keyword="w13a-model", format="PNG")
+Mads.spaghettiplot(md, predictions; keyword="w13a-model", format="PNG")
 Mads.display("$rootname-w13a-model-$numberofsamples-spaghetti.png")
 
 @time svrpredictions = svrexec(paramarray)
 @info("SVR discrepancy $(maximum(abs.(svrpredictions .- predictions)))")
 
 Mads.madsinfo("SVR predictions ...")
-Mads.spaghettiplot(md, svrpredictions, keyword="w13a-svr", format="PNG")
+Mads.spaghettiplot(md, svrpredictions; keyword="w13a-svr", format="PNG")
 Mads.display("$rootname-w13a-svr-$numberofsamples-spaghetti.png")
 
 sa = Mads.efast(md)

examples/model_analysis/uncertainty.jl

@@ -50,10 +50,10 @@ display(s_var)
 # JLD2.save("uncertainty_results/variance-important-sampling.jld2", "goodoprime", goodoprime)
 
 @info("Spaghetti plot of posterior predictions")
-Mads.spaghettiplot(md, o, filename="uncertainty_results/spaghetti-$(problem).png")
+Mads.spaghettiplot(md, o; filename="uncertainty_results/spaghetti-$(problem).png")
 
 @info("Spaghetti plot of posterior predictions using importance sampling")
-Mads.spaghettiplot(md, goodoprime', filename="uncertainty_results/spaghetti-$(problem)-importance-sampling.png")
+Mads.spaghettiplot(md, goodoprime'; filename="uncertainty_results/spaghetti-$(problem)-importance-sampling.png")
 Mads.display("uncertainty_results/spaghetti-$(problem)-importance-sampling.png")
 
 @info("Histogram of `o5` predictions")
@@ -68,7 +68,7 @@ Gadfly.draw(Gadfly.PNG("uncertainty_results/histogram-$(problem)-importance-samp
 Mads.setparamsinit!(md, p)
 mcmcchain = Mads.bayessampling(md; nsteps=10000, burnin=1000, thinning=1, seed=2016)
 ob = Mads.forward(md, mcmcchain.value)
-Mads.spaghettiplot(md, ob, filename="uncertainty_results/spaghetti-$(problem)-bayes.png")
+Mads.spaghettiplot(md, ob; filename="uncertainty_results/spaghetti-$(problem)-bayes.png")
 Mads.display("uncertainty_results/spaghetti-$(problem)-bayes.png")
 
 @info("Histogram of `o5` predictions using Bayesian analysis")