# Model predictions across many sub-models

Source:`R/aaa_multi_predict.R`

, `R/boost_tree.R`

, `R/glmnet-engines.R`

, and 3 more
`multi_predict.Rd`

For some models, predictions can be made on sub-models in the model object.

## Usage

```
multi_predict(object, ...)
# S3 method for default
multi_predict(object, ...)
# S3 method for `_xgb.Booster`
multi_predict(object, new_data, type = NULL, trees = NULL, ...)
# S3 method for `_C5.0`
multi_predict(object, new_data, type = NULL, trees = NULL, ...)
# S3 method for `_elnet`
multi_predict(object, new_data, type = NULL, penalty = NULL, ...)
# S3 method for `_lognet`
multi_predict(object, new_data, type = NULL, penalty = NULL, ...)
# S3 method for `_multnet`
multi_predict(object, new_data, type = NULL, penalty = NULL, ...)
# S3 method for `_glmnetfit`
multi_predict(object, new_data, type = NULL, penalty = NULL, ...)
# S3 method for `_earth`
multi_predict(object, new_data, type = NULL, num_terms = NULL, ...)
# S3 method for `_torch_mlp`
multi_predict(object, new_data, type = NULL, epochs = NULL, ...)
# S3 method for `_train.kknn`
multi_predict(object, new_data, type = NULL, neighbors = NULL, ...)
```

## Arguments

- object
A

`model_fit`

object.- ...
Optional arguments to pass to

`predict.model_fit(type = "raw")`

such as`type`

.- new_data
A rectangular data object, such as a data frame.

- type
A single character value or

`NULL`

. Possible values are`"numeric"`

,`"class"`

,`"prob"`

,`"conf_int"`

,`"pred_int"`

,`"quantile"`

, or`"raw"`

. When`NULL`

,`predict()`

will choose an appropriate value based on the model's mode.- trees
An integer vector for the number of trees in the ensemble.

- penalty
A numeric vector of penalty values.

- num_terms
An integer vector for the number of MARS terms to retain.

- epochs
An integer vector for the number of training epochs.

- neighbors
An integer vector for the number of nearest neighbors.

## Value

A tibble with the same number of rows as the data being predicted.
There is a list-column named `.pred`

that contains tibbles with
multiple rows per sub-model. Note that, within the tibbles, the column names
follow the usual standard based on prediction `type`

(i.e. `.pred_class`

for
`type = "class"`

and so on).