kerasnip: A Bridge Between 'keras' and 'tidymodels'

Provides a seamless bridge between 'keras' and the 'tidymodels' frameworks. It allows for the dynamic creation of 'parsnip' model specifications for 'keras' models.

Version:	0.0.3
Depends:	R (≥ 4.1.0)
Imports:	abind, parsnip (≥ 1.0.0), rlang, keras3, tibble, purrr, dplyr, cli, recipes, reticulate
Suggests:	testthat (≥ 3.0.0), modeldata, tidymodels, finetune, tune, dials, workflows, rsample, knitr, lme4, rmarkdown, future, ggplot2
Published:	2025-09-18
DOI:	10.32614/CRAN.package.kerasnip
Author:	David Díaz [aut, cre]
Maintainer:	David Díaz <daviddrsch at gmail.com>
License:	MIT + file LICENSE
NeedsCompilation:	no
Materials:	README
CRAN checks:	kerasnip results

Documentation:

Reference manual:

kerasnip.html , kerasnip.pdf

Vignettes:

Transfer Learning with Keras Applications (source, R code) Tuning Multiple Similar Parameters: Ensuring 'autoplot' Uniqueness (source, R code) Building Functional Models with kerasnip (source, R code) Getting Started with kerasnip (source, R code) The Sequential Model with kerasnip (source, R code) Tuning Fit and Compile Arguments (source, R code) Tidymodels Workflow with Functional Keras Models (Multi-Input) (source, R code) Tidymodels Workflow with Sequential Keras Models (source, R code)

Downloads:

Package source:	kerasnip_0.0.3.tar.gz
Windows binaries:	r-devel: kerasnip_0.0.3.zip, r-release: kerasnip_0.0.3.zip, r-oldrel: not available
macOS binaries:	r-release (arm64): kerasnip_0.0.3.tgz, r-oldrel (arm64): kerasnip_0.0.3.tgz, r-release (x86_64): kerasnip_0.0.3.tgz, r-oldrel (x86_64): kerasnip_0.0.3.tgz
Old sources:	kerasnip archive

Linking:

Please use the canonical form https://CRAN.R-project.org/package=kerasnip to link to this page.