Question

How to best include hyperparameter optimization in MLInterfaces?

0

Entering edit mode

Guest User ★ 13k

@guest-user-4897

Last seen 9.6 years ago

Hi, I am currently working with MLInterfaces and have a question regarding the optimization of hyperparameters of the learning functions (that are given to MLearn by the .methods parameter). Is it somehow possible to do optimization of hyperparameters, e.g the k in knn? Or in other words is it intended to have hyperparameter optimization included in MLInterfaces? Thank you very much! -- output of sessionInfo(): R version 2.15.1 (2012-06-22) Platform: x86_64-unknown-linux-gnu (64-bit) locale: [1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C LC_TIME=en_US.UTF-8 LC_COLLATE=en_US.UTF-8 LC_MONETARY=en_US.UTF-8 [6] LC_MESSAGES=en_US.UTF-8 LC_PAPER=C LC_NAME=C LC_ADDRESS=C LC_TELEPHONE=C [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C attached base packages: [1] stats4 grid stats graphics grDevices utils datasets methods base other attached packages: [1] multicore_0.1-7 BIAutils_0.2.00 RPostgreSQL_0.3-3 DBI_0.2-5 TangoR_1.0.19 R.oo_1.10.1 [7] R.methodsS3_1.4.2 BIProlif_0.1.69 xtable_1.7-0 XML_3.95-0.1 drc_2.3-0 plotrix_3.4-5 [13] nlme_3.1-105 magic_1.5-2 abind_1.4-0 alr3_2.0.5 car_2.0-15 nnet_7.3-5 [19] RODBC_1.3-6 ROCR_1.0-4 gplots_2.11.0 KernSmooth_2.23-8 caTools_1.13 bitops_1.0-5 [25] gdata_2.12.0 gtools_2.7.0 lattice_0.20-10 logging_0.6-92 snow_0.3-10 Cairo_1.5-2 [31] MLInterfaces_1.38.0 sfsmisc_1.0-23 annotate_1.36.0 AnnotationDbi_1.20.3 rda_1.0.2-2 rpart_4.0-3 [37] genefilter_1.40.0 MASS_7.3-22 Biobase_2.18.0 BiocGenerics_0.4.0 ClassDiscovery_2.13.4 PreProcess_2.12.2 [43] oompaBase_2.15.0 mclust_4.0 cluster_1.14.3 loaded via a namespace (and not attached): [1] IRanges_1.16.4 Matrix_1.0-10 mboost_2.1-3 parallel_2.15.1 RSQLite_0.11.2 splines_2.15.1 survival_2.36-14 tools_2.15.1 -- Sent via the guest posting facility at bioconductor.org.

MLInterfaces MLInterfaces • 1.0k views

ADD COMMENT • link updated 11.0 years ago by Vincent J. Carey, Jr. 6.7k • written 11.0 years ago by Guest User ★ 13k

score 0 · Answer 1 · 2013-04-15

On Mon, Apr 15, 2013 at 4:43 AM, Andreas Bernthaler [guest] <guest at="" bioconductor.org=""> wrote: > > Hi, > > I am currently working with MLInterfaces and have a question regarding the optimization of hyperparameters of the learning functions (that are given to MLearn by the .methods parameter). Is it somehow possible to do optimization of hyperparameters, e.g the k in knn? Or in other words is it intended to have hyperparameter optimization included in MLInterfaces? > I have no immediate plans to introduce this in MLInterfaces. The CMA package does address hyperparameter tuning. > Thank you very much! > > -- output of sessionInfo(): > > R version 2.15.1 (2012-06-22) > Platform: x86_64-unknown-linux-gnu (64-bit) > > locale: > [1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C LC_TIME=en_US.UTF-8 LC_COLLATE=en_US.UTF-8 LC_MONETARY=en_US.UTF-8 > [6] LC_MESSAGES=en_US.UTF-8 LC_PAPER=C LC_NAME=C LC_ADDRESS=C LC_TELEPHONE=C > [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C > > attached base packages: > [1] stats4 grid stats graphics grDevices utils datasets methods base > > other attached packages: > [1] multicore_0.1-7 BIAutils_0.2.00 RPostgreSQL_0.3-3 DBI_0.2-5 TangoR_1.0.19 R.oo_1.10.1 > [7] R.methodsS3_1.4.2 BIProlif_0.1.69 xtable_1.7-0 XML_3.95-0.1 drc_2.3-0 plotrix_3.4-5 > [13] nlme_3.1-105 magic_1.5-2 abind_1.4-0 alr3_2.0.5 car_2.0-15 nnet_7.3-5 > [19] RODBC_1.3-6 ROCR_1.0-4 gplots_2.11.0 KernSmooth_2.23-8 caTools_1.13 bitops_1.0-5 > [25] gdata_2.12.0 gtools_2.7.0 lattice_0.20-10 logging_0.6-92 snow_0.3-10 Cairo_1.5-2 > [31] MLInterfaces_1.38.0 sfsmisc_1.0-23 annotate_1.36.0 AnnotationDbi_1.20.3 rda_1.0.2-2 rpart_4.0-3 > [37] genefilter_1.40.0 MASS_7.3-22 Biobase_2.18.0 BiocGenerics_0.4.0 ClassDiscovery_2.13.4 PreProcess_2.12.2 > [43] oompaBase_2.15.0 mclust_4.0 cluster_1.14.3 > > loaded via a namespace (and not attached): > [1] IRanges_1.16.4 Matrix_1.0-10 mboost_2.1-3 parallel_2.15.1 RSQLite_0.11.2 splines_2.15.1 survival_2.36-14 tools_2.15.1 > > -- > Sent via the guest posting facility at bioconductor.org.