paramix (version 0.0.1)

interpolate_opts: Interpolation Options

Description

Creates and interpolation options object for use with alembic().

Usage

interpolate_opts(fun, kind = c("point", "integral"), ...)

Value

a list, with fun and kind keys, as well as whatever other valid keys appear in ....

Arguments

fun

a function

kind

a string; either "point" or "integral". How to interpret the x, y values being interpolated. Either as point observations of a function OR as the integral of the function over the interval.

...

arbitrary other arguments, but checked against signature of fun

Details

This method creates the interpolation object for use with alembic(); this is a convenience method, which does basic validation on arguments and ensures the information used in alembic() to do interpolation is available.

The ... arguments will be provided to fun when it is invoked to interpolate the tabular "functional" form of arguments to alembic(). If fun has an argument kind, that parameter will also be passed when invoking the function; if not, then the input data will be transformed to \(\{x, z\}\) pairs, such that \(x_{i+1}-x_{i} * z_i = y_i\) - i.e., transforming to a point value and a functional form which is assumed constant until the next partition.

Examples

interpolate_opts(
  fun = stats::splinefun, method = "natural", kind = "point"
)
interpolate_opts(
  fun = stats::approxfun, method = "constant", yleft = 0, yright = 0,
  kind = "integral"
)