luz: Higher Level 'API' for 'torch'

A high level interface for 'torch' providing utilities to reduce the the amount of code needed for common tasks, abstract away torch details and make the same code work on both the 'CPU' and 'GPU'. It's flexible enough to support expressing a large range of models. It's heavily inspired by 'fastai' by Howard et al. (2020) <doi:10.48550/arXiv.2002.04688>, 'Keras' by Chollet et al. (2015) and 'PyTorch Lightning' by Falcon et al. (2019) <doi:10.5281/zenodo.3828935>.

Version:	0.5.1
Imports:	torch (≥ 0.11.9000), magrittr, zeallot, rlang (≥ 1.0.0), coro, glue, progress, R6, generics, purrr, fs, prettyunits, cli
Suggests:	knitr, rmarkdown, testthat (≥ 3.0.0), covr, Metrics, withr, vdiffr, ggplot2 (≥ 3.0.0), dplyr, torchvision, tfevents (≥ 0.0.2), tidyr
Published:	2025-10-30
DOI:	10.32614/CRAN.package.luz
Author:	Daniel Falbel [aut, cre, cph], Christophe Regouby [ctb], RStudio [cph]
Maintainer:	Daniel Falbel <daniel at rstudio.com>
BugReports:	https://github.com/mlverse/luz/issues
License:	MIT + file LICENSE
URL:	https://mlverse.github.io/luz/, https://github.com/mlverse/luz
NeedsCompilation:	no
Materials:	README, NEWS
CRAN checks:	luz results

Documentation:

Reference manual:	luz.html , luz.pdf
Vignettes:	Accelerator API (source, R code) Custom loops with luz (source, R code) Get started with luz (source, R code)

Downloads:

Package source:	luz_0.5.1.tar.gz
Windows binaries:	r-devel: luz_0.5.1.zip, r-release: luz_0.5.1.zip, r-oldrel: luz_0.5.1.zip
macOS binaries:	r-release (arm64): luz_0.5.1.tgz, r-oldrel (arm64): luz_0.5.1.tgz, r-release (x86_64): luz_0.5.1.tgz, r-oldrel (x86_64): luz_0.5.1.tgz
Old sources:	luz archive

Reverse dependencies:

Reverse imports:	geodl, NNMoMo, sits
Reverse suggests:	bundle, innsight, nn2poly, vetiver

Linking:

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