Public Member Functions |
Static Public Member Functions |
Protected Member Functions |
List of all members
npstat::GridInterpolatedDistribution Class Reference
Inheritance diagram for npstat::GridInterpolatedDistribution:
Detailed DescriptionThis class interpolates between probability distributions placed at the points of a rectangular (not necessarily equidistant) grid. The weights in the interpolations are determined in the multilinear fashion. Multivariate quantile or copula interpolation method can be used, depending on the types of interpolated densities. Constructor & Destructor Documentation◆ GridInterpolatedDistribution()
Constructor arguments "axes" and "distributionDimension" are, respectively, the collection of parameter grid axes and the dimensionality of the associated distributions. If "interpolateCopulas" is true, copula interpolation method will be used (as implemented by class CopulaInterpolationND), otherwise unit map interpolation will be used. If "useNearestGridCellOnly" is true, interpolation will be disabled and only the distribution corresponding to the nearest grid cell will be used. Member Function Documentation◆ classId()
Prototype needed for I/O Implements npstat::AbsDistributionND. ◆ clone()
"Virtual copy constructor" Implements npstat::AbsDistributionND. ◆ density()
Method inherited from AbsDistributionND Implements npstat::AbsDistributionND. ◆ getGridDistro()
Get the distribution associated with the given parameter grid cell. Note that null pointer will be returned in case the cell has not been set yet. ◆ getLinearDistro()
Get the distribution associated with the given parameter grid cell using a linear index for the cell. Note that null pointer will be returned in case the cell has not been set yet. ◆ interpolateCopulas()
Switch between unit map interpolation and copula interpolation ◆ mappedByQuantiles()
The following method should return "true" in case the "unitMap" method is implemented by a sequence of conditional quantile functions. Distributions with such maps permit quantile-based interpolation procedures. Implements npstat::AbsDistributionND. ◆ marginalDensity()
This method works for copula interpolation only. Naturally, all interpolated distributions in this case must be of type (or must inherit from) CompositeDistributionND. ◆ nAxes()
Inspect object properties ◆ setGridCoords()
The following function must be called before calculating the density. The length of the coordinates array must be equal to the number of grid axes. ◆ setGridDistro()
Associate a distribution with the given parameter grid cell. This object will assume the ownership of the "distro" pointer if "assumeOwnership" is true and will make a copy of it if "assumeOwnership" is false. All distributions must be set before density can be calculated on all points inside and outside the grid. ◆ setLinearDistro()
Associate a distribution with the given parameter grid cell using a linear index for the cell ◆ unitMap()
Mapping from the unit hypercube into the density support region. Note that "bufLen" does not have to be equal to the dimensionality of the function. There may be an efficient way to generate just the leading dimensions in case "bufLen" is smaller than the dimensionality. Implements npstat::AbsDistributionND. ◆ useSingleCell()
Switch between normal interpolation and using just one cell The documentation for this class was generated from the following file: Generated by 1.9.1 |