Returns the criteria that optTrain can optimize, with the name to give to its
method argument, the function that scores a given training set, the direction of
optimization, and whether the criterion uses a test set.
Details
Criteria marked uses_test = FALSE depend on the training set only. When a
test set is given to optTrain, such a criterion ignores it (with a warning).
One page per criterion on the package website gives the definition, the original
reference, and the license review.
Examples
criteria()
#> method name type
#> 1 rScore r-score model-based
#> 2 PEV PEVmean (ridge, PCs) model-based
#> 3 CD CD (ridge, PCs) model-based
#> 4 AOpt A-optimality optimal design
#> 5 DOpt D-optimality optimal design
#> 6 EOpt E-optimality optimal design
#> 7 PEVmax PEVmax model-based
#> 8 GOpt G-optimality (GOPTPEV) optimal design
#> 9 MSPE MSPE of ridge regression model-based
#> 10 GVoverall GVoverall genomic variation
#> 11 GVaverage GVaverage genomic variation
#> 12 AvgGRM Avg_GRM relationship
#> 13 AvgGRMself Avg_GRM_self relationship
#> 14 AvgGRMminmax Avg_GRM_MinMax relationship
#> 15 CDmean CDmean model-based
#> 16 CDmeanV2 CDmean(v2) model-based
#> 17 PAM Partitioning around medoids (PAM) clustering
#> 18 CDmax CDmax (ridge, PCs) model-based
#> score_fun maximize uses_test
#> 1 r_score TRUE TRUE
#> 2 pev_score FALSE TRUE
#> 3 cd_score TRUE TRUE
#> 4 aopt_score FALSE FALSE
#> 5 dopt_score TRUE FALSE
#> 6 eopt_score FALSE FALSE
#> 7 pevmax_score FALSE TRUE
#> 8 gopt_score FALSE TRUE
#> 9 mspe_score FALSE TRUE
#> 10 gv_overall_score TRUE FALSE
#> 11 gv_average_score TRUE FALSE
#> 12 avg_grm_score TRUE TRUE
#> 13 avg_grm_self_score FALSE FALSE
#> 14 avg_grm_minmax_score TRUE TRUE
#> 15 cdmean_score TRUE TRUE
#> 16 cdmean2_score TRUE TRUE
#> 17 pam_score FALSE TRUE
#> 18 cdmax_score TRUE TRUE