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: |
0.4.1 |
| Depends: |
mlr3tuning (≥ 0.13.0), R (≥ 3.1.0) |
| Imports: |
bbotk (≥ 0.5.2), checkmate (≥ 1.9.4), data.table, lgr, mlr3 (≥ 0.13.1), mlr3misc (≥ 0.10.0), paradox (≥ 0.9.0), R6 |
| Suggests: |
emoa, mlr3learners (≥ 0.5.2), mlr3pipelines, rpart, testthat (≥ 3.0.0), xgboost |
| Published: |
2022-05-04 |
| Author: |
Marc Becker  [aut,
cre],
Sebastian Gruber
[aut],
Jakob Richter
[aut],
Julia Moosbauer
[aut],
Bernd Bischl
[aut] |
| 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 |