Add a model for bagged neural networks

[topepo]

Engines in tidymodels/baguette#64

[simonpcouch]

The model spec print method will need a new row in this table to describe this model.🙂

parsnip/R/print.R

Lines 39 to 40 in 3c0a0e2

	model_descs <- tibble::tribble(
	~cls, ~desc,

[simonpcouch]

Looks good! A few small changes in the PR, and 2 questions:

1. Why is epochs a tunable parameter for mlp but not for bag_mlp?

2. If I mistakenly try to tune epochs, I get the parsnip error about guarded arguments:

# pak::pak("tidymodels/baguette#64")

library(baguette)
#> Loading required package: parsnip

bm <-
  bag_mlp(epochs = tune()) %>%
  set_engine("nnet") %>%
  set_mode("regression")

bm_tune <-
  tune::tune_grid(
    bm,
    mpg ~ .,
    rsample::bootstraps(mtcars, 2),
    grid = 2
  )
#> ! Bootstrap1: preprocessor 1/1, model 1/2: The following arguments cannot be manually modified and were removed: ma...
#> ! Bootstrap1: preprocessor 1/1, model 2/2: The following arguments cannot be manually modified and were removed: ma...
#> ! Bootstrap2: preprocessor 1/1, model 1/2: The following arguments cannot be manually modified and were removed: ma...
#> ! Bootstrap2: preprocessor 1/1, model 2/2: The following arguments cannot be manually modified and were removed: ma...

tune::show_notes(bm_tune)
#> unique notes:
#> ────────────────────────────────────────────────────────────────────────────
#> The following arguments cannot be manually modified and were removed: maxit.

^{Created on 2022-09-22 by the reprex package (v2.0.1)}

Is this the usual error when tune() is mistakenly supplied to a main model org that isn't tunable?

[topepo]

The epoch issue was hard to track down but fixed in baguette.

[github-actions]

This pull request has been automatically locked. If you believe you have found a related problem, please file a new issue (with a reprex: https://reprex.tidyverse.org) and link to this issue.