Transform a parameter value

Apply a transformation function to a parameter value. This is used to map calibrated values back to their natural scale, e.g. when a parameter is calibrated on a log scale.

Usage

transform_parameter_value(x, fun)

Arguments

x

A numeric vector of parameter values, or NULL.

fun

A string naming the transformation to apply. One of "identity", "pow10" (10^x), "pow2" (2^x) or "exp" (exp(x)).

Value

The transformed values, or NULL if x is NULL.

Examples

transform_parameter_value(2, "pow10")
#> [1] 100
transform_parameter_value(c(1, 2, 3), "pow2")
#> [1] 2 4 8
transform_parameter_value(0.5, "identity")
#> [1] 0.5