bespoke()
defines a model that makes predictions using a set function.
"Training" such a model only consists of verifying that the output is
consistent with the types of outputs seen in the training data; the function
is not tailored at all to better predict the training data. Such models can
be useful for establishing a baseline which any true models should
outperform.
Arguments
- mode
A single character string for the prediction outcome mode. Only "classification" is valid.
- engine
A single character string specifying what computational engine to use for "fitting." Only "bespoke" is valid.
- fn
A function that takes a data.frame as input and returns a vector (integer, character, or factor) indicating the outcomes as output (with one value per input row). We may someday extend this for probabilities.