statisticalmodel
A representation of a gaussian process, which is only defined on a discrete domain.
Represents a low-rank gaussian process, that is only defined at a finite, discrete set of points.
Represents a low-rank gaussian process, that is only defined at a finite, discrete set of points. It supports the same operations as the LowRankGaussianProcess class, but always returns instead a discrete representation. Furthermore, most operations are much more efficient, as they are implemented using fast matrix/vector operations.
Where the modeled functions in a LowRankGaussianProcess are of type Point[D]=>Vector[D], this discretized version is of type VectorPointData.
It is possible to convert a DiscreteLowRankGaussianProcess to a LowRankGaussianProcess by calling the interpolation method.
A gaussian process from a D dimensional input space, whose input values are points, to a DO dimensional output space.
A gaussian process from a D dimensional input space, whose input values are points, to a DO dimensional output space. The output space is a Euclidean vector space of dimensionality DO.
The dimensionality of the input space
The dimensionality of the output space
A gaussian process which is represented in terms of a (small) finite set of basis functions.
A gaussian process which is represented in terms of a (small) finite set of basis functions. The basis functions are the orthonormal basis functions given by a mercers' decomposition.
The dimensionality of the input space
The dimensionality of the output space
A StatisticalMeshModel is isomorphic to a DiscreteLowRankGaussianProcess.
A StatisticalMeshModel is isomorphic to a DiscreteLowRankGaussianProcess. The difference is that while the DiscreteLowRankGaussianProcess models defomation fields, the StatisticalMeshModel applies the deformation fields to a mesh, and warps the mesh with the deformation fields to produce a new mesh.
Factory methods for creating Gaussian processes
Factory methods for creating Low-rank gaussian processes, as well as generic algorithms to manipulate Gaussian processes.
Contains classes for handling datasets of registered meshes and building PCA model out of them.
Contains classes for handling datasets of registered meshes and building PCA model out of them.
The most important class of this package is the DataCollection.
A representation of a gaussian process, which is only defined on a discrete domain. While this is technically similar to a MultivariateNormalDistribution, we highlight with this class that we represent (discrete) functions, defined on the given domain.