io.github.mandar2812.dynaml.models.gp
Create an instance of GPBasisFuncRegressionModel for a particular data type T
Create an instance of GPBasisFuncRegressionModel for a particular data type T
The type of the training data
The type of the input patterns in the data set of type T
The covariance function
The noise covariance function
A DataPipe transforming the input features to basis function components.
A MultGaussianRV which is the prior distribution on basis function coefficients
The actual data set of type T
An implicit conversion from T to Seq represented as a DataPipe
Create an instance of AbstractGPRegressionModel for a particular data type T
Create an instance of AbstractGPRegressionModel for a particular data type T
The type of the training data
The type of the input patterns in the data set of type T
The covariance function
The noise covariance function
The trend or mean function
The actual data set of type T
An implicit conversion from T to Seq represented as a DataPipe
Calculate the marginal log likelihood of the training data for a pre-initialized kernel and noise matrices.
Calculate the marginal log likelihood of the training data for a pre-initialized kernel and noise matrices.
The function values assimilated as a DenseVector
The kernel matrix of the training features
Calculate the marginal log likelihood of the training data for a pre-initialized kernel and noise matrices.
Calculate the marginal log likelihood of the training data for a pre-initialized kernel and noise matrices.
The function values assimilated as a DenseVector
The kernel matrix of the training features
Calculate the parameters of the posterior predictive distribution for a multivariate gaussian model.