Additional Information: http://www.public.iastate.edu/~gdancik/mlegp/

Installation Instructions: http://www.public.iastate.edu/~gdancik/mlegp/mlegp_install

]]>Additional Information: http://www.public.iastate.edu/~gdancik/mlegp/

Installation Instructions: http://www.public.iastate.edu/~gdancik/mlegp/mlegp_install

]]>Gaussian processes (GPs) are commonly used as surrogate statistical models for predicting output of computer experiments (Santner et al., 2003). Even more generlly, GPs are both interpolators and smoothers of data and are effective when the response surface of interest is a smooth function of the parameter space. The package finds maximum likelihood estimates of Gaussian processes for univariate and multi-dimensional responses, for Gaussian processes with product exponential correlation structures; constant or linear regression mean functions; no nugget term, constant nugget terms, or a nugget matrix that can be specifed up to a multiplicative constant. The latter is an extension of previous Gaussian process models and provides some exibility for using GPs to model heteroscedastic responses. Diagnostic plotting functions, and the sensitivity analysis tools of Functional Analysis of Variance (FANOVA) decomposition, and plotting of main and two-way factor interaction effects are implemented. Multi-dimensional output can be modelled by fitting independent GPs to each dimension of output, or to the most principle component weights following singular value decomposition of the output. Plotting of main effects for functional output is also implemented.

Once the library is installed, the library is loaded into R by calling 'library(mlegp)' from within R. A complete list of functions and vignettes can be obtained by calling `library(help = "mlegp")'.

Once the library is installed, the library is loaded into R by calling 'library(mlegp)' from within R. A complete list of functions and vignettes can be obtained by calling `library(help = "mlegp")'.

Gaussian processes (GPs) are commonly used as surrogate statistical models for predicting output of computer experiments (Santner et al., 2003). Even more generlly, GPs are both interpolators and smoothers of data and are effective when the response surface of interest is a smooth function of the parameter space. The package finds maximum likelihood estimates of Gaussian processes for univariate and multi-dimensional responses, for Gaussian processes with product exponential correlation structures; constant or linear regression mean functions; no nugget term, constant nugget terms, or a nugget matrix that can be specifed up to a multiplicative constant. The latter is an extension of previous Gaussian process models and provides some exibility for using GPs to model heteroscedastic responses. Diagnostic plotting functions, and the sensitivity analysis tools of Functional Analysis of Variance (FANOVA) decomposition, and plotting of main and two-way factor interaction effects are implemented. Multi-dimensional output can be modelled by fitting independent GPs to each dimension of output, or to the most principle component weights following singular value decomposition of the output. Plotting of main effects for functional output is also implemented. From within R, a complete list of functions and vignettes can be obtained by calling `library(help = "mlegp")'.

]]>Gaussian processes (GPs) are commonly used as surrogate statistical models for predicting output of computer experiments (Santner et al., 2003). Even more generlly, GPs are both interpolators and smoothers of data and are effective when the response surface of interest is a smooth function of the parameter space. The package finds maximum likelihood estimates of Gaussian processes for univariate and multi-dimensional responses, for Gaussian processes with product exponential correlation structures; constant or linear regression mean functions; no nugget term, constant nugget terms, or a nugget matrix that can be specifed up to a multiplicative constant. The latter is an extension of previous Gaussian process models and provides some exibility for using GPs to model heteroscedastic responses. Diagnostic plotting functions, and the sensitivity analysis tools of Functional Analysis of Variance (FANOVA) decomposition, and plotting of main and two-way factor interaction effects are implemented. Multi-dimensional output can be modelled by fitting independent GPs to each dimension of output, or to the most principle component weights following singular value decomposition of the output. Plotting of main effects for functional output is also implemented.

Once the library is installed, the library is loaded into R by calling 'library(mlegp)' from within R. A complete list of functions and vignettes can be obtained by calling `library(help = "mlegp")'.

Once the library is installed, the library is loaded into R by calling 'library(mlegp)' from within R. A complete list of functions and vignettes can be obtained by calling `library(help = "mlegp")'.

Gaussian processes (GPs) are commonly used as surrogate statistical models for predicting output of computer experiments (Santner et al., 2003). Even more generlly, GPs are both interpolators and smoothers of data and are effective when the response surface of interest is a smooth function of the parameter space. The package finds maximum likelihood estimates of Gaussian processes for univariate and multi-dimensional responses, for Gaussian processes with product exponential correlation structures; constant or linear regression mean functions; no nugget term, constant nugget terms, or a nugget matrix that can be specifed up to a multiplicative constant. The latter is an extension of previous Gaussian process models and provides some exibility for using GPs to model heteroscedastic responses. Diagnostic plotting functions, and the sensitivity analysis tools of Functional Analysis of Variance (FANOVA) decomposition, and plotting of main and two-way factor interaction effects are implemented. Multi-dimensional output can be modelled by fitting independent GPs to each dimension of output, or to the most principle component weights following singular value decomposition of the output. Plotting of main effects for functional output is also implemented. From within R, a complete list of functions and vignettes can be obtained by calling `library(help = "mlegp")'.

]]>]]>

]]>

]]>

]]>

The package *mlegp* provides maximum likelihood Gaussian process modeling for univariate and multi-dimensional outputs with diagnostic plots and sensitivity analysis.

Gaussian processes (GPs) are commonly used as surrogate statistical models for predicting output of computer experiments (Santner et al., 2003). Even more generlly, GPs are both interpolators and smoothers of data and are effective when the response surface of interest is a smooth function of the parameter space. The package finds maximum likelihood estimates of Gaussian processes for univariate and multi-dimensional responses, for Gaussian processes with product exponential correlation structures; constant or linear regression mean functions; no nugget term, constant nugget terms, or a nugget matrix that can be specifed up to a multiplicative constant. The latter is an extension of previous Gaussian process models and provides some exibility for using GPs to model heteroscedastic responses. Diagnostic plotting functions, and the sensitivity analysis tools of Functional Analysis of Variance (FANOVA) decomposition, and plotting of main and two-way factor interaction effects are implemented. Multi-dimensional output can be modelled by fitting independent GPs to each dimension of output, or to the most principle component weights following singular value decomposition of the output. Plotting of main effects for functional output is also implemented. From within R, a complete list of functions and vignettes can be obtained by calling `library(help = "mlegp")'.

Download mlegp from CRAN: http://cran.r-project.org/src/contrib/Descriptions/mlegp.html

Installation Instructions: http://www.public.iastate.edu/~gdancik/mlegp/mlegp_install

Gaussian processes (GPs) are commonly used as surrogate statistical models for predicting output of computer experiments (Santner et al., 2003). Even more generlly, GPs are both interpolators and smoothers of data and are effective when the response surface of interest is a smooth function of the parameter space. The package finds maximum likelihood estimates of Gaussian processes for univariate and multi-dimensional responses, for Gaussian processes with product exponential correlation structures; constant or linear regression mean functions; no nugget term, constant nugget terms, or a nugget matrix that can be specifed up to a multiplicative constant. The latter is an extension of previous Gaussian process models and provides some exibility for using GPs to model heteroscedastic responses. Diagnostic plotting functions, and the sensitivity analysis tools of Functional Analysis of Variance (FANOVA) decomposition, and plotting of main and two-way factor interaction effects are implemented. Multi-dimensional output can be modelled by fitting independent GPs to each dimension of output, or to the most principle component weights following singular value decomposition of the output. Plotting of main effects for functional output is also implemented. From within R, a complete list of functions and vignettes can be obtained by calling `library(help = "mlegp")'.

Download mlegp from CRAN: http://cran.r-project.org/src/contrib/Descriptions/mlegp.html

Installation Instructions: http://www.public.iastate.edu/~gdancik/mlegp/mlegp_install

All necessary files can be found on rumi.gdcb.iastate.edu in the directory

Type R from the command line to open a R session

Load the necessary R code and C library:

> source("mlegp.R")

> source("PCWeightsGP.R")

> dyn.load("mlegp.so")

Read in the design matrix and output matrix:

> X = as.matrix(read.table("design2"))

> Z = as.matrix(read.table("output"))

Save the workspace:

> save.image("annie.RData")

Just in case something bad happens, we can recover the current workspace using load("annie.RData")

If we do this we must also reload the C library: > dyn.load("mlegp.so")

The first step of the sensitivity analysis is to fit gaussian processes to the computer model output:

> fit1 = mlegp(X,Z)

If this results in errors, it is probably best to see me. You might also try looking at a few parameters at a time, and a few outputs at a time. For example, to look at parameters 3 and 4 and the second output, use

> fit1 = mlegp(X[,3:4], Z[,2])

Do the sensitivity analysis:

> library('adapt')

> R2 = FANOVADecomposition(fit1, Interaction = F)

If R gets stuck for more than 3 minutes at any point, use Ctrl-C to cancel, and call the function again using

> R2 = FANOVADecomposition(fit1, Interaction = F, maxpts = 20000)

Each column of R2 corresponds to a computer model output; each element of this table corresponds to the percentage of total variance accounted for by each parameter; parameters are named p1, p2, ... by default.

Setting Interaction = T will calculate all 2-way interaction effects (for a total of 55 effects for 10 parameters)

If elements of R2 are in scientific notation, you can use the following to get a prettier table:

> nice(R2)

You may be interested in plotting main effects for parameters that are interesting (make sure you are using VNC to access rumi, otherwise these commands may cause the R session to crash).

To plot the main effects of parameter 8 for all outputs:

> plotMainEffects(fit1, 8)

To plot the main effects effects of parameter 6 for the first output:

> plotMainEffects(fit11, 6)

]]>/home/gdancik/annie

The commands that follow assume you are working from within the directory that contains the necessary files. Otherwise you will have to specify the path to the files.Type R from the command line to open a R session

Load the necessary R code and C library:

> source("mlegp.R")

> source("PCWeightsGP.R")

> dyn.load("mlegp.so")

Read in the design matrix and output matrix:

> X = as.matrix(read.table("design2"))

> Z = as.matrix(read.table("output"))

Save the workspace:

> save.image("annie.RData")

Just in case something bad happens, we can recover the current workspace using load("annie.RData")

If we do this we must also reload the C library: > dyn.load("mlegp.so")

The first step of the sensitivity analysis is to fit gaussian processes to the computer model output:

> fit1 = mlegp(X,Z)

If this results in errors, it is probably best to see me. You might also try looking at a few parameters at a time, and a few outputs at a time. For example, to look at parameters 3 and 4 and the second output, use

> fit1 = mlegp(X[,3:4], Z[,2])

Do the sensitivity analysis:

> library('adapt')

> R2 = FANOVADecomposition(fit1, Interaction = F)

If R gets stuck for more than 3 minutes at any point, use Ctrl-C to cancel, and call the function again using

> R2 = FANOVADecomposition(fit1, Interaction = F, maxpts = 20000)

Each column of R2 corresponds to a computer model output; each element of this table corresponds to the percentage of total variance accounted for by each parameter; parameters are named p1, p2, ... by default.

Setting Interaction = T will calculate all 2-way interaction effects (for a total of 55 effects for 10 parameters)

If elements of R2 are in scientific notation, you can use the following to get a prettier table:

> nice(R2)

You may be interested in plotting main effects for parameters that are interesting (make sure you are using VNC to access rumi, otherwise these commands may cause the R session to crash).

To plot the main effects of parameter 8 for all outputs:

> plotMainEffects(fit1, 8)

To plot the main effects effects of parameter 6 for the first output:

> plotMainEffects(fit11, 6)

The package *mlegp* provides maximum likelihood Gaussian process modeling for univariate and multi-dimensional outputs with diagnostic plots and sensitivity analysis.

Gaussian processes (GPs) are commonly used as surrogate statistical models for predicting output of computer experiments (Santner et al., 2003). Even more generlly, GPs are both interpolators and smoothers of data and are effective when the response surface of interest is a smooth function of the parameter space. The package finds maximum likelihood estimates of Gaussian processes for univariate and multi-dimensional responses, for Gaussian processes with product exponential correlation structures; constant or linear regression mean functions; no nugget term, constant nugget terms, or a nugget matrix that can be specifed up to a multiplicative constant. The latter is an extension of previous Gaussian process models and provides some exibility for using GPs to model heteroscedastic responses. Diagnostic plotting functions, and the sensitivity analysis tools of Functional Analysis of Variance (FANOVA) decomposition, and plotting of main and two-way factor interaction effects are implemented. Multi-dimensional output can be modelled by fitting independent GPs to each dimension of output, or to the most principle component weights following singular value decomposition of the output. Plotting of main effects for functional output is also implemented. From within R, a complete list of functions and vignettes can be obtained by calling `library(help = "mlegp")'.

Download mlegp from CRAN: http://cran.r-project.org/src/contrib/Descriptions/mlegp.html

Installation Instructions: http://www.public.iastate.edu/~gdancik/mlegp/mlegp_install

Gaussian processes (GPs) are commonly used as surrogate statistical models for predicting output of computer experiments (Santner et al., 2003). Even more generlly, GPs are both interpolators and smoothers of data and are effective when the response surface of interest is a smooth function of the parameter space. The package finds maximum likelihood estimates of Gaussian processes for univariate and multi-dimensional responses, for Gaussian processes with product exponential correlation structures; constant or linear regression mean functions; no nugget term, constant nugget terms, or a nugget matrix that can be specifed up to a multiplicative constant. The latter is an extension of previous Gaussian process models and provides some exibility for using GPs to model heteroscedastic responses. Diagnostic plotting functions, and the sensitivity analysis tools of Functional Analysis of Variance (FANOVA) decomposition, and plotting of main and two-way factor interaction effects are implemented. Multi-dimensional output can be modelled by fitting independent GPs to each dimension of output, or to the most principle component weights following singular value decomposition of the output. Plotting of main effects for functional output is also implemented. From within R, a complete list of functions and vignettes can be obtained by calling `library(help = "mlegp")'.

Download mlegp from CRAN: http://cran.r-project.org/src/contrib/Descriptions/mlegp.html

Installation Instructions: http://www.public.iastate.edu/~gdancik/mlegp/mlegp_install

All necessary files can be found on rumi.gdcb.iastate.edu in the directory

Type R from the command line to open a R session

Load the necessary R code and C library:

> source("mlegp.R")

> source("PCWeightsGP.R")

> dyn.load("mlegp.so")

Read in the design matrix and output matrix:

> X = as.matrix(read.table("design2"))

> Z = as.matrix(read.table("output"))

Save the workspace:

> save.image("annie.RData")

Just in case something bad happens, we can recover the current workspace using load("annie.RData")

If we do this we must also reload the C library: > dyn.load("mlegp.so")

The first step of the sensitivity analysis is to fit gaussian processes to the computer model output:

> fit1 = mlegp(X,Z)

If this results in errors, it is probably best to see me. You might also try looking at a few parameters at a time, and a few outputs at a time. For example, to look at parameters 3 and 4 and the second output, use

> fit1 = mlegp(X[,3:4], Z[,2])

Do the sensitivity analysis:

> library('adapt')

> R2 = FANOVADecomposition(fit1, Interaction = F)

If R gets stuck for more than 3 minutes at any point, use Ctrl-C to cancel, and call the function again using

> R2 = FANOVADecomposition(fit1, Interaction = F, maxpts = 20000)

Each column of R2 corresponds to a computer model output; each element of this table corresponds to the percentage of total variance accounted for by each parameter; parameters are named p1, p2, ... by default.

Setting Interaction = T will calculate all 2-way interaction effects (for a total of 55 effects for 10 parameters)

If elements of R2 are in scientific notation, you can use the following to get a prettier table:

> nice(R2)

You may be interested in plotting main effects for parameters that are interesting (make sure you are using VNC to access rumi, otherwise these commands may cause the R session to crash).

To plot the main effects of parameter 8 for all outputs:

> plotMainEffects(fit1, 8)

To plot the main effects effects of parameter 6 for the first output:

> plotMainEffects(fit11, 6)

]]>/home/gdancik/annie

The commands that follow assume you are working from within the directory that contains the necessary files. Otherwise you will have to specify the path to the files.Type R from the command line to open a R session

Load the necessary R code and C library:

> source("mlegp.R")

> source("PCWeightsGP.R")

> dyn.load("mlegp.so")

Read in the design matrix and output matrix:

> X = as.matrix(read.table("design2"))

> Z = as.matrix(read.table("output"))

Save the workspace:

> save.image("annie.RData")

Just in case something bad happens, we can recover the current workspace using load("annie.RData")

If we do this we must also reload the C library: > dyn.load("mlegp.so")

The first step of the sensitivity analysis is to fit gaussian processes to the computer model output:

> fit1 = mlegp(X,Z)

If this results in errors, it is probably best to see me. You might also try looking at a few parameters at a time, and a few outputs at a time. For example, to look at parameters 3 and 4 and the second output, use

> fit1 = mlegp(X[,3:4], Z[,2])

Do the sensitivity analysis:

> library('adapt')

> R2 = FANOVADecomposition(fit1, Interaction = F)

If R gets stuck for more than 3 minutes at any point, use Ctrl-C to cancel, and call the function again using

> R2 = FANOVADecomposition(fit1, Interaction = F, maxpts = 20000)

Each column of R2 corresponds to a computer model output; each element of this table corresponds to the percentage of total variance accounted for by each parameter; parameters are named p1, p2, ... by default.

Setting Interaction = T will calculate all 2-way interaction effects (for a total of 55 effects for 10 parameters)

If elements of R2 are in scientific notation, you can use the following to get a prettier table:

> nice(R2)

You may be interested in plotting main effects for parameters that are interesting (make sure you are using VNC to access rumi, otherwise these commands may cause the R session to crash).

To plot the main effects of parameter 8 for all outputs:

> plotMainEffects(fit1, 8)

To plot the main effects effects of parameter 6 for the first output:

> plotMainEffects(fit11, 6)

> plotMainEffects(fit11, 6)

> plotMainEffects(fit11?, 6)

]]>> plotMainEffects(fit11, 6)

> plotMainEffects(fit11?, 6)

]]>> source("PCWeightsGP.R")

> dyn.load("mlegp.so")

> X = as.matrix(read.table("design2"))

> Z = as.matrix(read.table("output"))

> save.image("annie.RData")

If we do this we must also reload the C library: > dyn.load("mlegp.so")

> fit1 = mlegp(X,Z)

If this results in errors, it is probably best to see me. You might also try looking at a few parameters at a time, and a few outputs at a time. For example, to look at parameters 3 and 4 and the second output, use

> fit1 = mlegp(X[,3:4], Z[,2])

> library('adapt')

> R2 = FANOVADecomposition(fit1, Interaction = F)

> R2 = FANOVADecomposition(fit1, Interaction = F, maxpts = 20000)

> nice(R2)

> plotMainEffects(fit1, 8)

> plotMainEffects(fit11?, 6)

> dyn.load("mlegp.so")

> X = as.matrix(read.table("design2"))

> Z = as.matrix(read.table("output"))

> save.image("annie.RData")

If we do this we must also reload the C library: > dyn.load("mlegp.so")

> fit1 = mlegp(X,Z)

If this results in errors, it is probably best to see me. You might also try looking at a few parameters at a time, and a few outputs at a time. For example, to look at parameters 3 and 4 and the second output, use

> fit1 = mlegp(X[,3:4], Z[,2])

> library('adapt')

> R2 = FANOVADecomposition(fit1, Interaction = F)

> R2 = FANOVADecomposition(fit1, Interaction = F, maxpts = 20000)

> nice(R2)

> plotMainEffects(fit1, 8)

> plotMainEffects(fit11?, 6)

source("PCWeightsGP.R")

dyn.load("mlegp.so")

X = as.matrix(read.table("design2"))

Z = as.matrix(read.table("output"))

save.image("annie.RData")

If we do this we must also reload the C library: > dyn.load("mlegp.so") dyn.load("mlegp.so")

X = as.matrix(read.table("design2"))

Z = as.matrix(read.table("output"))

save.image("annie.RData")

fit1 = mlegp(X,Z)

If this results in errors, try looking at a few parameters at a time, and a few outputs at a time. For example, to look at parameters 3 and 4 and the second output, use fit1 = mlegp(X[,3:4], Z[,2]

library('adapt')

R2 = FANOVADecomposition(fit1, Interaction = F)

R2 = FANOVADecomposition(fit1, Interaction = F, maxpts = 20000)

nice(R2)

plotMainEffects(fit1, 8)

plotMainEffects(fit11?, 6)

library('adapt')

R2 = FANOVADecomposition(fit1, Interaction = F)

R2 = FANOVADecomposition(fit1, Interaction = F, maxpts = 20000)

nice(R2)

plotMainEffects(fit1, 8)

plotMainEffects(fit11?, 6)

> source("PCWeightsGP.R")

> dyn.load("mlegp.so")

> X = as.matrix(read.table("design2"))

> Z = as.matrix(read.table("output"))

> save.image("annie.RData")

If we do this we must also reload the C library: > dyn.load("mlegp.so")

> fit1 = mlegp(X,Z)

If this results in errors, it is probably best to see me. You might also try looking at a few parameters at a time, and a few outputs at a time. For example, to look at parameters 3 and 4 and the second output, use

> fit1 = mlegp(X[,3:4], Z[,2])

> library('adapt')

> R2 = FANOVADecomposition(fit1, Interaction = F)

> R2 = FANOVADecomposition(fit1, Interaction = F, maxpts = 20000)

> nice(R2)

> plotMainEffects(fit1, 8)

> plotMainEffects(fit11?, 6)

> dyn.load("mlegp.so")

> X = as.matrix(read.table("design2"))

> Z = as.matrix(read.table("output"))

> save.image("annie.RData")

If we do this we must also reload the C library: > dyn.load("mlegp.so")

> fit1 = mlegp(X,Z)

If this results in errors, it is probably best to see me. You might also try looking at a few parameters at a time, and a few outputs at a time. For example, to look at parameters 3 and 4 and the second output, use

> fit1 = mlegp(X[,3:4], Z[,2])

> library('adapt')

> R2 = FANOVADecomposition(fit1, Interaction = F)

> R2 = FANOVADecomposition(fit1, Interaction = F, maxpts = 20000)

> nice(R2)

> plotMainEffects(fit1, 8)

> plotMainEffects(fit11?, 6)

source("PCWeightsGP.R")

dyn.load("mlegp.so")

X = as.matrix(read.table("design2"))

Z = as.matrix(read.table("output"))

save.image("annie.RData")

If we do this we must also reload the C library: > dyn.load("mlegp.so") dyn.load("mlegp.so")

X = as.matrix(read.table("design2"))

Z = as.matrix(read.table("output"))

save.image("annie.RData")

fit1 = mlegp(X,Z)

If this results in errors, try looking at a few parameters at a time, and a few outputs at a time. For example, to look at parameters 3 and 4 and the second output, use fit1 = mlegp(X[,3:4], Z[,2]

library('adapt')

R2 = FANOVADecomposition(fit1, Interaction = F)

R2 = FANOVADecomposition(fit1, Interaction = F, maxpts = 20000)

nice(R2)

plotMainEffects(fit1, 8)

plotMainEffects(fit11?, 6)

library('adapt')

R2 = FANOVADecomposition(fit1, Interaction = F)

R2 = FANOVADecomposition(fit1, Interaction = F, maxpts = 20000)

nice(R2)

plotMainEffects(fit1, 8)

plotMainEffects(fit11?, 6)

> source("mlegp.R")

">" source("mlegp.R")

]]>> source("mlegp.R")

">" source("mlegp.R")

]]>Setting Interaction = T will calculate all 2-way interaction effects (for a total of 55 effects for 10 parameters)

If elements of R2 are in scientific notation, you can use the following to get a prettier table:

To plot the main effects of parameter 8 for all outputs:

If elements of R2 are in scientific notation, you can use the following to get a prettier table:

nice(R2)

You may be interested in plotting main effects for parameters that are interesting (make sure you are using VNC to access rumi, otherwise these commands may cause the R session to crash).To plot the main effects of parameter 8 for all outputs:

plotMainEffects(fit1, 8)

To plot the main effects effects of parameter 6 for the first output: plotMainEffects(fit11?, 6)

Setting Interaction = T will calculate all 2-way interaction effects

]]>Setting Interaction = T will calculate all 2-way interaction effects (for a total of 55 effects for 10 parameters)

If elements of R2 are in scientific notation, you can use the following to get a prettier table:

To plot the main effects of parameter 8 for all outputs:

If elements of R2 are in scientific notation, you can use the following to get a prettier table:

nice(R2)

You may be interested in plotting main effects for parameters that are interesting (make sure you are using VNC to access rumi, otherwise these commands may cause the R session to crash).To plot the main effects of parameter 8 for all outputs:

plotMainEffects(fit1, 8)

To plot the main effects effects of parameter 6 for the first output: plotMainEffects(fit11?, 6)

Setting Interaction = T will calculate all 2-way interaction effects

]]>">" source("mlegp.R")

source("mlegp.R")

">" source("mlegp.R")

source("mlegp.R")

source("mlegp.R")

/> source("mlegp.R")

]]> source("mlegp.R")

/> source("mlegp.R")

]]>/> source("mlegp.R")

'>' source("mlegp.R")

]]>/> source("mlegp.R")

'>' source("mlegp.R")

]]>'>' source("mlegp.R")

source("mlegp.R")

'>' source("mlegp.R")

source("mlegp.R")

The commands that follow assume you are working from within the directory that contains the necessary files. Otherwise you will have to specify the path to the files.

The commands that follow assume you are working from within the directory that contains the necessary files.

]]>The commands that follow assume you are working from within the directory that contains the necessary files. Otherwise you will have to specify the path to the files.

The commands that follow assume you are working from within the directory that contains the necessary files.

]]>