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: 0.4.0
Imports: torch (≥ 0.9.0), magrittr, zeallot, rlang (≥ 1.0.0), coro, glue, progress, R6, generics, purrr, ellipsis, 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: 2023-04-17
DOI: 10.32614/CRAN.package.luz
Author: Daniel Falbel [aut, cre, cph], RStudio [cph]
Maintainer: Daniel Falbel <daniel at>
License: MIT + file LICENSE
NeedsCompilation: no
Materials: README NEWS
CRAN checks: luz results


Reference manual: luz.pdf
Vignettes: Accelerator API
Custom loops with luz
Get started with luz


Package source: luz_0.4.0.tar.gz
Windows binaries: r-devel:, r-release:, r-oldrel:
macOS binaries: r-release (arm64): luz_0.4.0.tgz, r-oldrel (arm64): luz_0.4.0.tgz, r-release (x86_64): luz_0.4.0.tgz, r-oldrel (x86_64): luz_0.4.0.tgz
Old sources: luz archive

Reverse dependencies:

Reverse suggests: bundle, innsight, nn2poly, sits, vetiver


Please use the canonical form to link to this page.