< Master index Index for polypedal/gpml >

Index for polypedal/gpml

Matlab files in this directory:

 Contentsgpml: code from Rasmussen & Williams: Gaussian Processes for Machine Learning
 approxEPExpectation Propagation approximation to the posterior Gaussian Process.
 approxLALaplace approximation to the posterior Gaussian Process.
 approximationsapproximations: Exact inference for Gaussian process classification is
 binaryEPGPbinaryEPGP - The Expectation Propagation approximation for binary Gaussian
 binaryGPApproximate binary Gaussian Process classification. Two modes are possible:
 binaryLaplaceGPbinaryLaplaceGP - Laplace's approximation for binary Gaussian process
 covConstcovariance function for a constant function. The covariance function is
 covFunctionscovariance functions to be use by Gaussian process functions. There are two
 covLINardLinear covariance function with Automatic Relevance Determination (ARD). The
 covLINoneLinear covariance function with a single hyperparameter. The covariance
 covMatern3isoMatern covariance function with nu = 3/2 and isotropic distance measure. The
 covMatern5isoMatern covariance function with nu = 5/2 and isotropic distance measure. The
 covNNoneNeural network covariance function with a single parameter for the distance
 covNoiseIndependent covariance function, ie "white noise", with specified variance.
 covPeriodiccovariance function for a smooth periodic function, with unit period. The
 covProdcovProd - compose a covariance function as the product of other covariance
 covRQardRational Quadratic covariance function with Automatic Relevance Determination
 covRQisoRational Quadratic covariance function with isotropic distance measure. The
 covSEardSquared Exponential covariance function with Automatic Relevance Detemination
 covSEisoSquared Exponential covariance function with isotropic distance measure. The
 covSumcovSum - compose a covariance function as the sum of other covariance
 covTPeriodiccovariance function for a smooth periodic function with specified period.
 cumGausscumGauss - Cumulative Gaussian likelihood function. The expression for the
 gauhercompute abscissas and weight factors for Gaussian-Hermite quadrature
 gprgpr - Gaussian process regression, with a named covariance function. Two
 gprSRPPgprSRPP - Carries out approximate Gaussian process regression prediction
 likelihoodslikelihood: likelihood functions are provided to be used by the binaryGP
 logisticlogistic - logistic likelihood function. The expression for the likelihood is
 minimizeMinimize a differentiable multivariate function.
 negLogMLCompute the negative log marginal likelihood that the hyperparameters
 solve_cholsolve_chol - solve linear equations from the Cholesky factorization.
 sq_distsq_dist - a function to compute a matrix of all pairwise squared distances

Generated on Mon 02-Aug-2010 16:44:30 by m2html © 2003