Skip to contents

Multiclass Learners in {nadir}

Source: R/multiclass_learners.R
multiclass_learners.Rd

  • lnr_multinomial_nnet

  • lnr_multinomial_vglm

Value

Every learner function shares the same structure: when called with (data, formula, ...) it fits the underlying model and returns a prediction closure which is a function of newdata returning a numeric vector of predictions (predicted probabilities of the second factor level for binary learners; predicted densities for density learners; and a vector of the predicted class probabilities for the observed classes for multiclass learners).

Details

Suppose one of these is trained on some data and the fit learner is stored. Suppose we are going to call it on newdata and newdata$class is the outcome variable being predicting.

The important thing to know about multiclass learners is that they produce predictions that the outcome class is equal to newdata$class given the covariates specified in newdata.

This means that newdata passed to the returned prediction closure must contain the outcome column, or else an error is produced.

Similar to density estimation, we want to use determine_weights_using_neg_log_loss in our calls to super_learner(). This can be done automatically by declaring outcome_type = 'multiclass' in calling super_learner()

See also

density_learners binary_learners learners

Examples


super_learner(
  data = iris,
  learners = list(lnr_multinomial_vglm, lnr_multinomial_vglm, lnr_multinomial_nnet),
  formulas = list(
    .default = Species ~ .,
    multinomial_vglm_2 = Species ~ Petal.Length * Petal.Width + .
  ),
  outcome_type = "multiclass"
)
#> 
#> Super Learner (nadir_sl_model)
#>   outcome:      Species (multiclass)
#>   observations: 150   CV folds: 5
#>   ensemble weights:
#>     multinomial_vglm_2  1.000
#>     multinomial_vglm_1  0.000
#>     multinomial_nnet    0.000
#>   captured conditions (see $errors_from_*, $warnings_from_*):
#>     [warning] multinomial_vglm_1 @ cv-training: 11 diagonal elements of the working weights variable 'wz'... (x2)
#>     [warning] multinomial_vglm_1 @ cv-training: 17 diagonal elements of the working weights variable 'wz'...
#>     [warning] multinomial_vglm_1 @ cv-training: 24 diagonal elements of the working weights variable 'wz'...
#>     [warning] multinomial_vglm_1 @ cv-training: 32 diagonal elements of the working weights variable 'wz'... (x3)
#>     ... and 100 more unique conditions not shown
#> 
#> Methods: predict(x, newdata), plot(x), summary(x), coef(x), fitted(x), ...