mod_eval_fun: Internal functions for evaluating models

In mosaicModel: An Interface to Statistical Modeling Independent of Model Architecture

Description

These functions are the interface to the various model types for mod_eval(), and through that to all the other mod_ functions that need to evaluate models, e.g. mod_effect(), mod_cv(), and so on.

Usage

mod_eval_fun(model, data = NULL, interval = "none", ...)

Arguments

model	A model object of the classes permitted
data	Usually, a data table specifying the inputs to the model. But if not specified, the training data will be used.
interval	One of "none", "confidence", or "prediction". Not all model types support "prediction" or even "confidence".
...	additional arguments

Details

All of the eval_ functions are ex These functions return a numerical vector (for regression types) or a matrix of probabilities (for classifiers)

mosaicModel documentation built on May 2, 2019, 7:59 a.m.