Implements hyperband method for hyperparameter tuning. Various termination criteria can be set and combined. The class 'AutoTuner' provides a convenient way to perform nested resampling in combination with 'mlr3'. The hyperband algorithm was proposed by Lisha Li, Kevin Jamieson, Giulia DeSalvo, Afshin Rostamizadeh and Ameet Talwalkar (2018) <arXiv:1603.06560>.
Version: | 0.1.2 |
Depends: | mlr3tuning (≥ 0.6.0), R (≥ 3.1.0) |
Imports: | bbotk (≥ 0.3.0), checkmate (≥ 1.9.4), data.table, lgr, mlr3 (≥ 0.7.0), mlr3misc (≥ 0.7.0), paradox (≥ 0.7.0), R6 |
Suggests: | emoa, mlr3learners, mlr3pipelines, rpart, testthat (≥ 3.0.0), xgboost |
Published: | 2021-01-29 |
Author: | Marc Becker |
Maintainer: | Marc Becker <marcbecker at posteo.de> |
BugReports: | https://github.com/mlr-org/mlr3hyperband/issues |
License: | LGPL-3 |
URL: | https://mlr3hyperband.mlr-org.com, https://github.com/mlr-org/mlr3hyperband |
NeedsCompilation: | no |
Materials: | NEWS |
CRAN checks: | mlr3hyperband results |
Reference manual: | mlr3hyperband.pdf |
Package source: | mlr3hyperband_0.1.2.tar.gz |
Windows binaries: | r-devel: mlr3hyperband_0.1.2.zip, r-release: mlr3hyperband_0.1.2.zip, r-oldrel: mlr3hyperband_0.1.2.zip |
macOS binaries: | r-release: mlr3hyperband_0.1.2.tgz, r-oldrel: mlr3hyperband_0.1.2.tgz |
Old sources: | mlr3hyperband archive |
Please use the canonical form https://CRAN.R-project.org/package=mlr3hyperband to link to this page.