MachineLearningEnsemble: Create an ensemble or comparison table of existing...
In 19900321/flipMultivariates: Multivariate models

Create an ensemble or comparison table of existing MachineLearning and/or Regression models

1
2
3

MachineLearningEnsemble(models, compare.only = FALSE,
  evaluation.subset = NULL, evaluation.weights = NULL,
  output = "Comparison", optimal.ensemble = FALSE)

`models`	A `list` of models, all of which are of class `MachineLearning` or `Regression`.
`compare.only`	Logical; whether to just produce a table comparing the models or additionally combine them to make a new ensemble model.
`evaluation.subset`	An optional vector specifying a subset of observations to be used for evaluating the models. If not specified, models will only be compared on the training data. If models are not trained on the whole sample To evaluate on the whole sample, a subset must still be specified.
`evaluation.weights`	An optional vector of weights to be used for evaluating the models. Ignored if no evaluation.subset is supplied. A warning is given if these differ from the training weights.
`output`	If `compare.only` is `FALSE`, one of `"Comparison"` which produces a table comparing the models, or `"Ensemble"` which produces a `ConfusionMatrix`.
`optimal.ensemble`	Logical; whether to find the ensemble with the best accuracy or r-squared, calculated on the `evaluation.subset` if given, else on the training data. Ignored if `compare.only` is TRUE.