gptk: Gaussian Processes Tool-Kit

The gptk package implements a general-purpose toolkit for Gaussian process regression with a variety of covariance functions (e.g. RBF, Mattern, polynomial, etc). Based on a MATLAB implementation by Neil D. Lawrence. See inst/doc/index.html for more details.

AuthorAlfredo Kalaitzis <alkalait@gmail.com>, Antti Honkela <antti.honkela@tkk.fi>, Pei Gao <pg349@medschl.cam.ac.uk>, Neil D. Lawrence <N.Lawrence@dcs.shef.ac.uk>
Date of publication2014-03-07 17:44:33
MaintainerAlfredo Kalaitzis <alkalait@gmail.com>
LicenseBSD_2_clause + file LICENSE
Version1.08

View on CRAN

Man pages

basePlot: Plot a contour of the 2D Gaussian distribution with...

cmpndKernParamInit: CMPND kernel parameter initialisation.

cmpndNoiseParamInit: CMPND noise parameter initialisation.

demAutoOptimiseGp: Gaussian Process Optimisation Demo

demGpCov2D: Gaussian Process 2D Covariance Demo

demGpSample: Gaussian Process Sampling Demo

demInterpolation: Gaussian Process Interpolation Demo

demOptimiseGp: Gaussian Process Optimisation Demo

demRegression: Gaussian Process Regression Demo

expTransform: Constrains a parameter.

gaussianNoiseOut: Compute the output of the GAUSSIAN noise given the input mean...

gaussianNoiseParamInit: GAUSSIAN noise parameter initialisation.

gaussSamp: Sample from a Gaussian with a given covariance.

gpBlockIndices: Return indices of given block.

gpComputeAlpha: Update the vector 'alpha' for computing posterior mean...

gpComputeM: Compute the matrix m given the model.

gpCovGrads: Sparse objective function gradients wrt Covariance functions...

gpCovGradsTest: Test the gradients of the likelihood wrt the covariance.

gpCreate: Create a GP model with inducing variables/pseudo-inputs.

gpDataIndices: Return indices of present data.

gpExpandParam: Expand a parameter vector into a GP model.

gpExtractParam: Extract a parameter vector from a GP model.

gpGradient: Gradient wrapper for a GP model.

gpLogLikeGradients: Compute the gradients for the parameters and X.

gpLogLikelihood: Compute the log likelihood of a GP.

gpMeanFunctionGradient: Compute the log likelihood gradient wrt the scales.

gpObjective: Wrapper function for GP objective.

gpOptimise: Optimise the inducing variable based kernel.

gpOptions: Return default options for GP model.

gpOut: Evaluate the output of an Gaussian process model.

gpPlot: Gaussian Process Plotter

gpPosteriorMeanVar: Mean and variances of the posterior at points given by X.

gpPosteriorSample: Plot Samples from a GP Posterior.

gpSample: Plot Samples from a GP.

gpScaleBiasGradient: Compute the log likelihood gradient wrt the scales.

gpTest: Test the gradients of the gpCovGrads function and the gp...

gpUpdateAD: Update the representations of A and D associated with the...

gpUpdateKernels: Update the kernels that are needed.

kernCompute: Compute the kernel given the parameters and X.

kernCreate: Initialise a kernel structure.

kernDiagGradient: Compute the gradient of the kernel's parameters for the...

kernDiagGradX: Compute the gradient of the kernel wrt X.

kernGradient: Compute the gradient wrt the kernel parameters.

kernParamInit: Kernel parameter initialisation.

kernTest: Run some tests on the specified kernel.

modelDisplay: Display a model.

modelExpandParam: Update a model structure with new parameters or update the...

modelExtractParam: Extract the parameters of a model.

modelGradient: Model log-likelihood/objective error function and its...

modelGradientCheck: Check gradients of given model.

modelOut: Give the output of a model for given X.

modelOutputGrad: Compute derivatives with respect to params of model outputs.

multiKernParamInit: MULTI kernel parameter initialisation.

noiseCreate: Initialise a noise structure.

noiseOut: Give the output of the noise model given the mean and...

noiseParamInit: Noise model's parameter initialisation.

optimiDefaultConstraint: Returns function for parameter constraint.

rbfKernDiagGradX: Gradient of RBF kernel's diagonal with respect to X.

rbfKernGradX: Gradient of RBF kernel with respect to input locations.

rbfKernParamInit: RBF kernel parameter initialisation.

SCGoptim: Optimise the given function using (scaled) conjugate...

whiteKernDiagGradX: Gradient of WHITE kernel's diagonal with respect to X.

whiteKernGradX: Gradient of WHITE kernel with respect to input locations.

whiteKernParamInit: WHITE kernel parameter initialisation.

zeroAxes: A function to move the axes crossing point to the origin.

Files in this package

gptk
gptk/inst
gptk/inst/doc
gptk/inst/doc/demGpCov2D1_5.gif
gptk/inst/doc/index.html
gptk/inst/doc/gpSampleRbfSamples10InverseWidth16Variance1.png
gptk/inst/doc/demOptimiseGp2.gif
gptk/inst/doc/demInterpolation.gif
gptk/inst/doc/demRegression.gif
gptk/inst/doc/demGpCov2D1_2.gif
gptk/inst/doc/gpPosteriorSampleRbfSamples5InverseWidth16Variance1.png
gptk/inst/doc/gpSample.png
gptk/inst/doc/gpCovariance.png
gptk/inst/doc/gpSampleRbfSamples10InverseWidth1Variance1.png
gptk/inst/doc/gpPosteriorSampleRbfSamples5InverseWidth1Variance1.png
gptk/inst/doc/demOptimiseGp1.gif
gptk/NAMESPACE
gptk/R
gptk/R/kernDiagCompute.R gptk/R/expTransform.R gptk/R/rbfKernExtractParam.R gptk/R/multiKernParamInit.R gptk/R/noiseParamInit.R gptk/R/whiteKernParamInit.R gptk/R/gpLogLikeGradients.R gptk/R/rbfKernDisplay.R gptk/R/sigmoidTransform.R gptk/R/gpUpdateKernels.R gptk/R/rbfKernDiagGradX.R gptk/R/cmpndKernExtractParam.R gptk/R/kernExpandParam.R gptk/R/modelDisplay.R gptk/R/gpComputeM.R gptk/R/localCovarianceGradients.R gptk/R/gpTest.R gptk/R/whiteKernGradX.R gptk/R/whiteKernDisplay.R gptk/R/whiteKernDiagCompute.R gptk/R/whiteKernGradient.R gptk/R/modelGradientCheck.R gptk/R/gpBlockIndices.R gptk/R/gpScaleBiasGradient.R gptk/R/gpOut.R gptk/R/noiseOut.R gptk/R/multiKernGradient.R gptk/R/rbfKernParamInit.R gptk/R/gpCovGradsTest.R gptk/R/boundedTransform.R gptk/R/rbfKernCompute.R gptk/R/kernCreate.R gptk/R/CGoptim.R gptk/R/gpCreate.R gptk/R/multiKernDiagCompute.R gptk/R/modelOptimise.R gptk/R/gaussSamp.R gptk/R/modelExpandParam.R gptk/R/modelUpdateProcesses.R gptk/R/rbfKernExpandParam.R gptk/R/rbfKernGradX.R gptk/R/cmpndKernDiagGradX.R gptk/R/basePlot.R gptk/R/cmpndKernDisplay.R gptk/R/modelExtractParam.R gptk/R/gpMeanFunctionGradient.R gptk/R/whiteKernExpandParam.R gptk/R/localSCovarianceGradients.R gptk/R/cmpndKernGradX.R gptk/R/cmpndKernDiagCompute.R gptk/R/kernDiagGradient.R gptk/R/modelLogLikelihood.R gptk/R/gpLogLikelihood.R gptk/R/modelOut.R gptk/R/gptk-internal.R gptk/R/gpPosteriorSample.R gptk/R/rbfKernGradXpoint.R gptk/R/optimiDefaultOptions.R gptk/R/SCGoptim.R gptk/R/demRegression.R gptk/R/gpPosteriorMeanVar.R gptk/R/cmpndNoiseParamInit.R gptk/R/gpOptimise.R gptk/R/gaussianNoiseParamInit.R gptk/R/demGpSample.R gptk/R/gpDataIndices.R gptk/R/gpExpandParam.R gptk/R/multiKernDisplay.R gptk/R/optimiDefaultConstraint.R gptk/R/gpPlot.R gptk/R/multiKernCompute.R gptk/R/kernDiagGradX.R gptk/R/zeroAxes.R gptk/R/whiteKernExtractParam.R gptk/R/gpComputeAlpha.R gptk/R/cmpndKernCompute.R gptk/R/gaussianNoiseOut.R gptk/R/kernParamInit.R gptk/R/kernDisplay.R gptk/R/gpUpdateAD.R gptk/R/cmpndKernParamInit.R gptk/R/kernGradX.R gptk/R/gpExtractParam.R gptk/R/rbfKernGradient.R gptk/R/demGpCov2D.R gptk/R/modelObjective.R gptk/R/gpGradient.R gptk/R/demInterpolation.R gptk/R/gpSample.R gptk/R/kernGradient.R gptk/R/noiseCreate.R gptk/R/gpOptions.R gptk/R/demOptimiseGp.R gptk/R/whiteKernCompute.R gptk/R/kernTest.R gptk/R/whiteKernDiagGradX.R gptk/R/mlpOptions.R gptk/R/gpObjective.R gptk/R/multiKernExtractParam.R gptk/R/whiteXwhiteKernGradient.R gptk/R/modelOutputGrad.R gptk/R/whiteXwhiteKernCompute.R gptk/R/kernCompute.R gptk/R/kernExtractParam.R gptk/R/multiKernExpandParam.R gptk/R/gpCovGrads.R gptk/R/rbfKernDiagCompute.R gptk/R/cmpndKernGradient.R gptk/R/cmpndKernExpandParam.R gptk/R/demAutoOptimiseGp.R gptk/R/modelGradient.R
gptk/MD5
gptk/DESCRIPTION
gptk/man
gptk/man/gpTest.Rd gptk/man/gpExtractParam.Rd gptk/man/kernCreate.Rd gptk/man/gpOut.Rd gptk/man/modelExpandParam.Rd gptk/man/gpOptimise.Rd gptk/man/demRegression.Rd gptk/man/gpLogLikelihood.Rd gptk/man/whiteKernGradX.Rd gptk/man/noiseParamInit.Rd gptk/man/kernDiagGradX.Rd gptk/man/gpExpandParam.Rd gptk/man/gpObjective.Rd gptk/man/whiteKernDiagGradX.Rd gptk/man/demOptimiseGp.Rd gptk/man/noiseCreate.Rd gptk/man/kernTest.Rd gptk/man/gpScaleBiasGradient.Rd gptk/man/kernDiagGradient.Rd gptk/man/gpUpdateKernels.Rd gptk/man/kernCompute.Rd gptk/man/whiteKernParamInit.Rd gptk/man/gpCovGradsTest.Rd gptk/man/gpCreate.Rd gptk/man/modelGradientCheck.Rd gptk/man/gpUpdateAD.Rd gptk/man/gaussianNoiseParamInit.Rd gptk/man/noiseOut.Rd gptk/man/gpPlot.Rd gptk/man/gpGradient.Rd gptk/man/modelDisplay.Rd gptk/man/kernGradient.Rd gptk/man/zeroAxes.Rd gptk/man/multiKernParamInit.Rd gptk/man/gpOptions.Rd gptk/man/kernParamInit.Rd gptk/man/gpComputeAlpha.Rd gptk/man/demInterpolation.Rd gptk/man/gpPosteriorMeanVar.Rd gptk/man/modelExtractParam.Rd gptk/man/demGpCov2D.Rd gptk/man/rbfKernGradX.Rd gptk/man/expTransform.Rd gptk/man/modelOutputGrad.Rd gptk/man/gaussianNoiseOut.Rd gptk/man/rbfKernDiagGradX.Rd gptk/man/gpSample.Rd gptk/man/gpLogLikeGradients.Rd gptk/man/modelGradient.Rd gptk/man/gpBlockIndices.Rd gptk/man/gaussSamp.Rd gptk/man/demAutoOptimiseGp.Rd gptk/man/gpComputeM.Rd gptk/man/SCGoptim.Rd gptk/man/gpPosteriorSample.Rd gptk/man/rbfKernParamInit.Rd gptk/man/modelOut.Rd gptk/man/gpMeanFunctionGradient.Rd gptk/man/optimiDefaultConstraint.Rd gptk/man/gpDataIndices.Rd gptk/man/gpCovGrads.Rd gptk/man/demGpSample.Rd gptk/man/cmpndNoiseParamInit.Rd gptk/man/basePlot.Rd gptk/man/cmpndKernParamInit.Rd
gptk/LICENSE

Questions? Problems? Suggestions? or email at ian@mutexlabs.com.

All documentation is copyright its authors; we didn't write any of that.