140abb35be9f969d7a73481eeed2fca59451fb4b
[agghoo.git] / DESCRIPTION
1 Package: agghoo
2 Title: Aggregated Hold-Out Cross Validation
3 Date: 2022-08-30
4 Version: 0.1-0
5 Description: The 'agghoo' procedure is an alternative to usual cross-validation.
6 Instead of choosing the best model trained on V subsamples, it determines
7 a winner model for each subsample, and then aggregate the V outputs.
8 For the details, see "Aggregated hold-out" by Guillaume Maillard,
9 Sylvain Arlot, Matthieu Lerasle (2021) <arXiv:1909.04890>
10 published in Journal of Machine Learning Research 22(20):1--55.
11 Author: Sylvain Arlot <sylvain.arlot@universite-paris-saclay.fr> [cph,ctb],
12 Benjamin Auder <benjamin.auder@universite-paris-saclay.fr> [aut,cre,cph],
13 Melina Gallopin <melina.gallopin@universite-paris-saclay.fr> [cph,ctb],
14 Matthieu Lerasle <matthieu.lerasle@universite-paris-saclay.fr> [cph,ctb],
15 Guillaume Maillard <guillaume.maillard@uni.lu> [cph,ctb]
16 Maintainer: Benjamin Auder <benjamin.auder@universite-paris-saclay.fr>
17 Depends:
18 R (>= 3.5.0)
19 Imports:
20 class,
21 parallel,
22 R6,
23 rpart,
24 FNN
25 Suggests:
26 roxygen2
27 URL: https://git.auder.net/?p=agghoo.git
28 License: MIT + file LICENSE
29 RoxygenNote: 7.2.1
30 Collate:
31 'compareTo.R'
32 'agghoo.R'
33 'R6_AgghooCV.R'
34 'R6_Model.R'
35 'checks.R'
36 'utils.R'
37 'A_NAMESPACE.R'