GP-init: Gaussian Process
In italo-goncalves/geomod3D: Implicit 3D Geological Modeling and Geostatistics

Description Usage Arguments Details See Also

Implementation of the Gaussian Process model for 3D spatial interpolation.

1
2
3

GP(data, model, value, mean = NULL, trend = NULL,
  force.interp = numeric(), reg.v = 1e-09, tangents = NULL,
  reg.t = 1e-12, nugget.t = 0)

`data`	A `spatial3DDataFrame` object containing the data one wishes to model.
`model`	The covariance model. A `covarianceModel3D` object.
`value`	The column name of the variable to be modeled. It is assumed the column does not contain missing values.
`mean`	The global mean. Irrelevant if a trend is provided.
`trend`	The model's trend component. A formula in character format.
`force.interp`	Indices of points that must be interpolated exactly.
`reg.v`	Regularization to improve stability. A single value or a vector with length matching the number of data points.
`tangents`	A `directions3DDataFrame` object containing structural geology data. Most likely generated with the `GetLineDirections()` method.
`reg.t`	Regularization for structural data. A single value or a vector with length matching the number of structural data.
`weights`	The importance of each data point in the model (a vector with values between 0 and 1)

This method builds a GP object with all the information needed to make preditions at new data points.

trend must be a character string with a formula as a function of uppercase X, Y, and Z. The most common is the linear trend, "~ X + Y + Z". For ordinary kriging, use "~1". If neither trend nor mean are given, it is assumed that the global mean is the mean of the data values.

If any point index is given in force.interp, the predicted mean function will pass exactly through those points, but the predictive variance will still be computed as usual. This is in contrast to what is usually done by geostatistics softwares, which assign a variance of 0 to those points.

weights can be used to "tune down" the effect of some data points. The smaller the weight, the less effect a point will have on the predicted funtion, to the limit that a weight of 0 filters the point completely.

Note that this implementation uses all the data provided to make predictions, which may be too memory intensive for large datasets.

GP-class, SPGP-class

italo-goncalves/geomod3D documentation built on May 24, 2019, 2:49 p.m.