diff --git a/dev/pkgdown.yml b/dev/pkgdown.yml index 5a44d4f4..c7eae0f3 100644 --- a/dev/pkgdown.yml +++ b/dev/pkgdown.yml @@ -2,7 +2,7 @@ pandoc: 3.1.11 pkgdown: 2.1.0 pkgdown_sha: ~ articles: {} -last_built: 2024-07-24T11:01Z +last_built: 2024-07-24T12:00Z urls: reference: https://mlr3fselect.mlr-org.com/reference article: https://mlr3fselect.mlr-org.com/articles diff --git a/dev/reference/FSelectInstanceBatchMultiCrit.html b/dev/reference/FSelectInstanceBatchMultiCrit.html index 57b4e950..a8c0406d 100644 --- a/dev/reference/FSelectInstanceBatchMultiCrit.html +++ b/dev/reference/FSelectInstanceBatchMultiCrit.html @@ -249,27 +249,26 @@

Examplesfselector$optimize(instance) #> bill_depth bill_length body_mass flipper_length island sex year #> <lgcl> <lgcl> <lgcl> <lgcl> <lgcl> <lgcl> <lgcl> -#> 1: TRUE FALSE TRUE TRUE FALSE TRUE TRUE -#> 2: TRUE TRUE TRUE TRUE TRUE TRUE TRUE +#> 1: TRUE TRUE TRUE TRUE TRUE TRUE TRUE +#> 2: FALSE TRUE TRUE TRUE FALSE FALSE FALSE #> features n_features #> <list> <int> -#> 1: bill_depth,body_mass,flipper_length,sex,year 5 -#> 2: bill_depth,bill_length,body_mass,flipper_length,island,sex,... 5 -#> classif.ce time_train -#> <num> <num> -#> 1: 0.2035850 0.002666667 -#> 2: 0.0698449 0.003000000 +#> 1: bill_depth,bill_length,body_mass,flipper_length,island,sex,... 7 +#> 2: bill_length,body_mass,flipper_length 7 +#> classif.ce time_train +#> <num> <num> +#> 1: 0.06984490 0.003 +#> 2: 0.08151538 0.002 # Optimal feature sets instance$result_feature_set #> [[1]] -#> [1] "bill_depth" "body_mass" "flipper_length" "sex" -#> [5] "year" -#> -#> [[2]] #> [1] "bill_depth" "bill_length" "body_mass" "flipper_length" #> [5] "island" "sex" "year" #> +#> [[2]] +#> [1] "bill_length" "body_mass" "flipper_length" +#> # Inspect all evaluated sets as.data.table(instance$archive) @@ -281,10 +280,10 @@

Examples#> 4: FALSE TRUE TRUE TRUE FALSE FALSE FALSE #> classif.ce time_train runtime_learners timestamp batch_nr #> <num> <num> <num> <POSc> <int> -#> 1: 0.08151538 0.003000000 0.016 2024-07-24 11:01:20 1 -#> 2: 0.20358505 0.002666667 0.014 2024-07-24 11:01:20 1 -#> 3: 0.06984490 0.003000000 0.017 2024-07-24 11:01:20 2 -#> 4: 0.08151538 0.007666667 0.029 2024-07-24 11:01:20 2 +#> 1: 0.08151538 0.003333333 0.016 2024-07-24 12:00:46 1 +#> 2: 0.20358505 0.002666667 0.014 2024-07-24 12:00:46 1 +#> 3: 0.06984490 0.003000000 0.016 2024-07-24 12:00:46 2 +#> 4: 0.08151538 0.002000000 0.013 2024-07-24 12:00:46 2 #> warnings errors #> <int> <int> #> 1: 0 0 diff --git a/dev/reference/FSelectInstanceBatchSingleCrit.html b/dev/reference/FSelectInstanceBatchSingleCrit.html index 2e802dcf..02ecaa1c 100644 --- a/dev/reference/FSelectInstanceBatchSingleCrit.html +++ b/dev/reference/FSelectInstanceBatchSingleCrit.html @@ -317,10 +317,10 @@

Examples#> 4: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> classif.ce runtime_learners timestamp batch_nr warnings errors #> <num> <num> <POSc> <int> <int> <int> -#> 1: 0.11929825 0.013 2024-07-24 11:01:21 1 0 0 -#> 2: 0.06117468 0.015 2024-07-24 11:01:21 1 0 0 -#> 3: 0.04652937 0.015 2024-07-24 11:01:21 2 0 0 -#> 4: 0.04652937 0.033 2024-07-24 11:01:21 2 0 0 +#> 1: 0.11929825 0.010 2024-07-24 12:00:47 1 0 0 +#> 2: 0.06117468 0.016 2024-07-24 12:00:47 1 0 0 +#> 3: 0.04652937 0.014 2024-07-24 12:00:47 2 0 0 +#> 4: 0.04652937 0.033 2024-07-24 12:00:47 2 0 0 #> features n_features #> <list> <list> #> 1: bill_length,body_mass 2 diff --git a/dev/reference/extract_inner_fselect_archives.html b/dev/reference/extract_inner_fselect_archives.html index 36bb0347..0512fede 100644 --- a/dev/reference/extract_inner_fselect_archives.html +++ b/dev/reference/extract_inner_fselect_archives.html @@ -145,26 +145,26 @@

Examples#> iteration bill_depth bill_length body_mass flipper_length island sex #> year classif.ce runtime_learners timestamp batch_nr warnings #> <lgcl> <num> <num> <POSc> <int> <int> -#> 1: TRUE 0.03508772 0.006 2024-07-24 11:01:37 1 0 -#> 2: TRUE 0.19298246 0.005 2024-07-24 11:01:37 1 0 -#> 3: FALSE 0.03508772 0.005 2024-07-24 11:01:37 1 0 -#> 4: FALSE 0.03508772 0.006 2024-07-24 11:01:37 1 0 -#> 5: TRUE 0.29824561 0.004 2024-07-24 11:01:37 1 0 -#> 6: TRUE 0.03508772 0.005 2024-07-24 11:01:37 1 0 -#> 7: FALSE 0.03508772 0.005 2024-07-24 11:01:37 1 0 -#> 8: FALSE 0.10526316 0.005 2024-07-24 11:01:37 1 0 -#> 9: FALSE 0.15789474 0.005 2024-07-24 11:01:37 1 0 -#> 10: FALSE 0.17543860 0.004 2024-07-24 11:01:37 1 0 -#> 11: TRUE 0.57894737 0.004 2024-07-24 11:01:37 1 0 -#> 12: FALSE 0.19298246 0.004 2024-07-24 11:01:37 1 0 -#> 13: FALSE 0.28070175 0.005 2024-07-24 11:01:37 1 0 -#> 14: TRUE 0.14035088 0.005 2024-07-24 11:01:37 1 0 -#> 15: FALSE 0.21052632 0.005 2024-07-24 11:01:37 1 0 -#> 16: TRUE 0.08771930 0.005 2024-07-24 11:01:37 1 0 -#> 17: FALSE 0.33333333 0.003 2024-07-24 11:01:37 1 0 -#> 18: FALSE 0.14035088 0.005 2024-07-24 11:01:37 1 0 -#> 19: TRUE 0.12280702 0.005 2024-07-24 11:01:37 1 0 -#> 20: FALSE 0.31578947 0.004 2024-07-24 11:01:37 1 0 +#> 1: TRUE 0.03508772 0.007 2024-07-24 12:01:03 1 0 +#> 2: TRUE 0.19298246 0.005 2024-07-24 12:01:03 1 0 +#> 3: FALSE 0.03508772 0.005 2024-07-24 12:01:03 1 0 +#> 4: FALSE 0.03508772 0.005 2024-07-24 12:01:03 1 0 +#> 5: TRUE 0.29824561 0.004 2024-07-24 12:01:03 1 0 +#> 6: TRUE 0.03508772 0.005 2024-07-24 12:01:03 1 0 +#> 7: FALSE 0.03508772 0.005 2024-07-24 12:01:03 1 0 +#> 8: FALSE 0.10526316 0.005 2024-07-24 12:01:03 1 0 +#> 9: FALSE 0.15789474 0.005 2024-07-24 12:01:03 1 0 +#> 10: FALSE 0.17543860 0.004 2024-07-24 12:01:03 1 0 +#> 11: TRUE 0.57894737 0.005 2024-07-24 12:01:03 1 0 +#> 12: FALSE 0.19298246 0.004 2024-07-24 12:01:03 1 0 +#> 13: FALSE 0.28070175 0.005 2024-07-24 12:01:03 1 0 +#> 14: TRUE 0.14035088 0.005 2024-07-24 12:01:03 1 0 +#> 15: FALSE 0.21052632 0.005 2024-07-24 12:01:03 1 0 +#> 16: TRUE 0.08771930 0.005 2024-07-24 12:01:03 1 0 +#> 17: FALSE 0.33333333 0.005 2024-07-24 12:01:03 1 0 +#> 18: FALSE 0.14035088 0.005 2024-07-24 12:01:03 1 0 +#> 19: TRUE 0.12280702 0.004 2024-07-24 12:01:03 1 0 +#> 20: FALSE 0.31578947 0.004 2024-07-24 12:01:03 1 0 #> year classif.ce runtime_learners timestamp batch_nr warnings #> errors features #> <int> <list> diff --git a/dev/reference/fselect.html b/dev/reference/fselect.html index 837fe223..07aaefb7 100644 --- a/dev/reference/fselect.html +++ b/dev/reference/fselect.html @@ -225,16 +225,16 @@

Examples#> 10: TRUE TRUE TRUE TRUE TRUE FALSE TRUE FALSE 0.2734375 #> runtime_learners timestamp batch_nr warnings errors #> <num> <POSc> <int> <int> <int> -#> 1: 0.009 2024-07-24 11:01:39 1 0 0 -#> 2: 0.007 2024-07-24 11:01:39 1 0 0 -#> 3: 0.007 2024-07-24 11:01:39 1 0 0 -#> 4: 0.007 2024-07-24 11:01:39 1 0 0 -#> 5: 0.008 2024-07-24 11:01:39 1 0 0 -#> 6: 0.008 2024-07-24 11:01:39 1 0 0 -#> 7: 0.009 2024-07-24 11:01:39 1 0 0 -#> 8: 0.009 2024-07-24 11:01:39 1 0 0 -#> 9: 0.008 2024-07-24 11:01:39 1 0 0 -#> 10: 0.007 2024-07-24 11:01:39 1 0 0 +#> 1: 0.009 2024-07-24 12:01:05 1 0 0 +#> 2: 0.009 2024-07-24 12:01:05 1 0 0 +#> 3: 0.006 2024-07-24 12:01:05 1 0 0 +#> 4: 0.006 2024-07-24 12:01:05 1 0 0 +#> 5: 0.008 2024-07-24 12:01:05 1 0 0 +#> 6: 0.007 2024-07-24 12:01:05 1 0 0 +#> 7: 0.009 2024-07-24 12:01:05 1 0 0 +#> 8: 0.008 2024-07-24 12:01:05 1 0 0 +#> 9: 0.007 2024-07-24 12:01:05 1 0 0 +#> 10: 0.008 2024-07-24 12:01:05 1 0 0 #> features n_features resample_result #> <list> <list> <list> #> 1: age,glucose,insulin,mass,pedigree,pregnant,... 8 <ResampleResult> diff --git a/dev/reference/fsi.html b/dev/reference/fsi.html index 17b5b235..c08a3c54 100644 --- a/dev/reference/fsi.html +++ b/dev/reference/fsi.html @@ -185,10 +185,10 @@

Examples#> 4: TRUE TRUE FALSE FALSE TRUE TRUE TRUE #> classif.ce runtime_learners timestamp batch_nr warnings errors #> <num> <num> <POSc> <int> <int> <int> -#> 1: 0.06114925 0.015 2024-07-24 11:01:41 1 0 0 -#> 2: 0.19471142 0.015 2024-07-24 11:01:41 1 0 0 -#> 3: 0.06687007 0.016 2024-07-24 11:01:41 2 0 0 -#> 4: 0.06114925 0.017 2024-07-24 11:01:41 2 0 0 +#> 1: 0.06114925 0.015 2024-07-24 12:01:07 1 0 0 +#> 2: 0.19471142 0.014 2024-07-24 12:01:07 1 0 0 +#> 3: 0.06687007 0.015 2024-07-24 12:01:07 2 0 0 +#> 4: 0.06114925 0.015 2024-07-24 12:01:07 2 0 0 #> features n_features resample_result #> <list> <list> <list> #> 1: bill_depth,bill_length,body_mass,island,sex 5 <ResampleResult> diff --git a/dev/reference/mlr_fselectors_design_points.html b/dev/reference/mlr_fselectors_design_points.html index aebe34ad..e5b264d2 100644 --- a/dev/reference/mlr_fselectors_design_points.html +++ b/dev/reference/mlr_fselectors_design_points.html @@ -176,10 +176,10 @@

Examples#> 4: TRUE FALSE TRUE TRUE FALSE TRUE TRUE TRUE 0.2617188 #> runtime_learners timestamp batch_nr warnings errors #> <num> <POSc> <int> <int> <int> -#> 1: 0.007 2024-07-24 11:01:46 1 0 0 -#> 2: 0.008 2024-07-24 11:01:46 2 0 0 -#> 3: 0.008 2024-07-24 11:01:46 3 0 0 -#> 4: 0.008 2024-07-24 11:01:46 4 0 0 +#> 1: 0.009 2024-07-24 12:01:12 1 0 0 +#> 2: 0.008 2024-07-24 12:01:12 2 0 0 +#> 3: 0.008 2024-07-24 12:01:12 3 0 0 +#> 4: 0.008 2024-07-24 12:01:12 4 0 0 #> features n_features resample_result #> <list> <list> <list> #> 1: age,insulin,mass,pregnant,triceps 5 <ResampleResult> diff --git a/dev/reference/mlr_fselectors_exhaustive_search.html b/dev/reference/mlr_fselectors_exhaustive_search.html index fcc9208c..11b99da2 100644 --- a/dev/reference/mlr_fselectors_exhaustive_search.html +++ b/dev/reference/mlr_fselectors_exhaustive_search.html @@ -172,16 +172,16 @@

Examples#> 10: TRUE FALSE FALSE TRUE FALSE FALSE FALSE #> classif.ce runtime_learners timestamp batch_nr warnings errors #> <num> <num> <POSc> <int> <int> <int> -#> 1: 0.26086957 0.004 2024-07-24 11:01:47 1 0 0 -#> 2: 0.21739130 0.005 2024-07-24 11:01:47 1 0 0 -#> 3: 0.29565217 0.004 2024-07-24 11:01:47 1 0 0 -#> 4: 0.15652174 0.005 2024-07-24 11:01:47 1 0 0 -#> 5: 0.27826087 0.004 2024-07-24 11:01:47 1 0 0 -#> 6: 0.57391304 0.005 2024-07-24 11:01:47 1 0 0 -#> 7: 0.57391304 0.003 2024-07-24 11:01:47 1 0 0 -#> 8: 0.09565217 0.004 2024-07-24 11:01:47 1 0 0 -#> 9: 0.25217391 0.004 2024-07-24 11:01:47 1 0 0 -#> 10: 0.15652174 0.004 2024-07-24 11:01:47 1 0 0 +#> 1: 0.26086957 0.005 2024-07-24 12:01:13 1 0 0 +#> 2: 0.21739130 0.003 2024-07-24 12:01:13 1 0 0 +#> 3: 0.29565217 0.005 2024-07-24 12:01:13 1 0 0 +#> 4: 0.15652174 0.005 2024-07-24 12:01:13 1 0 0 +#> 5: 0.27826087 0.004 2024-07-24 12:01:13 1 0 0 +#> 6: 0.57391304 0.004 2024-07-24 12:01:13 1 0 0 +#> 7: 0.57391304 0.004 2024-07-24 12:01:13 1 0 0 +#> 8: 0.09565217 0.005 2024-07-24 12:01:13 1 0 0 +#> 9: 0.25217391 0.004 2024-07-24 12:01:13 1 0 0 +#> 10: 0.15652174 0.003 2024-07-24 12:01:13 1 0 0 #> features n_features resample_result #> <list> <list> <list> #> 1: bill_depth 1 <ResampleResult> diff --git a/dev/reference/mlr_fselectors_genetic_search.html b/dev/reference/mlr_fselectors_genetic_search.html index 44cbdc3d..e573658b 100644 --- a/dev/reference/mlr_fselectors_genetic_search.html +++ b/dev/reference/mlr_fselectors_genetic_search.html @@ -162,16 +162,16 @@

Examples#> 10: FALSE TRUE FALSE TRUE TRUE FALSE FALSE #> classif.ce runtime_learners timestamp batch_nr warnings errors #> <num> <num> <POSc> <int> <int> <int> -#> 1: 0.30434783 0.004 2024-07-24 11:01:48 1 0 0 -#> 2: 0.31304348 0.005 2024-07-24 11:01:48 2 0 0 -#> 3: 0.24347826 0.003 2024-07-24 11:01:48 3 0 0 -#> 4: 0.24347826 0.003 2024-07-24 11:01:48 4 0 0 -#> 5: 0.22608696 0.004 2024-07-24 11:01:48 5 0 0 -#> 6: 0.30434783 0.004 2024-07-24 11:01:48 6 0 0 -#> 7: 0.03478261 0.005 2024-07-24 11:01:48 7 0 0 -#> 8: 0.24347826 0.008 2024-07-24 11:01:48 8 0 0 -#> 9: 0.37391304 0.004 2024-07-24 11:01:48 9 0 0 -#> 10: 0.06086957 0.006 2024-07-24 11:01:48 10 0 0 +#> 1: 0.30434783 0.005 2024-07-24 12:01:14 1 0 0 +#> 2: 0.31304348 0.004 2024-07-24 12:01:14 2 0 0 +#> 3: 0.24347826 0.005 2024-07-24 12:01:14 3 0 0 +#> 4: 0.24347826 0.004 2024-07-24 12:01:14 4 0 0 +#> 5: 0.22608696 0.004 2024-07-24 12:01:14 5 0 0 +#> 6: 0.30434783 0.005 2024-07-24 12:01:14 6 0 0 +#> 7: 0.03478261 0.005 2024-07-24 12:01:14 7 0 0 +#> 8: 0.24347826 0.005 2024-07-24 12:01:14 8 0 0 +#> 9: 0.37391304 0.005 2024-07-24 12:01:14 9 0 0 +#> 10: 0.06086957 0.005 2024-07-24 12:01:14 10 0 0 #> features n_features resample_result #> <list> <list> <list> #> 1: bill_depth 1 <ResampleResult> diff --git a/dev/reference/mlr_fselectors_random_search.html b/dev/reference/mlr_fselectors_random_search.html index 0051b1e0..6c628b8e 100644 --- a/dev/reference/mlr_fselectors_random_search.html +++ b/dev/reference/mlr_fselectors_random_search.html @@ -181,16 +181,16 @@

Examples#> 10: FALSE FALSE FALSE TRUE FALSE FALSE FALSE #> classif.ce runtime_learners timestamp batch_nr warnings errors #> <num> <num> <POSc> <int> <int> <int> -#> 1: 0.08695652 0.006 2024-07-24 11:01:49 1 0 0 -#> 2: 0.27826087 0.006 2024-07-24 11:01:49 1 0 0 -#> 3: 0.08695652 0.005 2024-07-24 11:01:49 1 0 0 -#> 4: 0.08695652 0.006 2024-07-24 11:01:49 1 0 0 -#> 5: 0.21739130 0.005 2024-07-24 11:01:49 1 0 0 -#> 6: 0.08695652 0.005 2024-07-24 11:01:49 1 0 0 -#> 7: 0.23478261 0.005 2024-07-24 11:01:49 1 0 0 -#> 8: 0.23478261 0.005 2024-07-24 11:01:49 1 0 0 -#> 9: 0.08695652 0.005 2024-07-24 11:01:49 1 0 0 -#> 10: 0.23478261 0.004 2024-07-24 11:01:49 1 0 0 +#> 1: 0.08695652 0.006 2024-07-24 12:01:15 1 0 0 +#> 2: 0.27826087 0.004 2024-07-24 12:01:15 1 0 0 +#> 3: 0.08695652 0.005 2024-07-24 12:01:15 1 0 0 +#> 4: 0.08695652 0.005 2024-07-24 12:01:15 1 0 0 +#> 5: 0.21739130 0.005 2024-07-24 12:01:15 1 0 0 +#> 6: 0.08695652 0.005 2024-07-24 12:01:15 1 0 0 +#> 7: 0.23478261 0.003 2024-07-24 12:01:15 1 0 0 +#> 8: 0.23478261 0.004 2024-07-24 12:01:15 1 0 0 +#> 9: 0.08695652 0.006 2024-07-24 12:01:15 1 0 0 +#> 10: 0.23478261 0.004 2024-07-24 12:01:15 1 0 0 #> features n_features #> <list> <list> #> 1: bill_depth,bill_length,body_mass,flipper_length,island,sex,... 7 diff --git a/dev/reference/mlr_fselectors_rfe.html b/dev/reference/mlr_fselectors_rfe.html index 2c222ce8..36e0e246 100644 --- a/dev/reference/mlr_fselectors_rfe.html +++ b/dev/reference/mlr_fselectors_rfe.html @@ -226,8 +226,8 @@

Examples#> 2: TRUE TRUE FALSE TRUE FALSE FALSE FALSE #> classif.ce runtime_learners timestamp batch_nr warnings errors #> <num> <num> <POSc> <int> <int> <int> -#> 1: 0.08695652 0.006 2024-07-24 11:01:50 1 0 0 -#> 2: 0.10434783 0.006 2024-07-24 11:01:50 2 0 0 +#> 1: 0.08695652 0.006 2024-07-24 12:01:16 1 0 0 +#> 2: 0.10434783 0.005 2024-07-24 12:01:16 2 0 0 #> importance #> <list> #> 1: 7,6,5,4,3,2,... diff --git a/dev/reference/mlr_fselectors_rfecv.html b/dev/reference/mlr_fselectors_rfecv.html index 3c03e22c..da05695b 100644 --- a/dev/reference/mlr_fselectors_rfecv.html +++ b/dev/reference/mlr_fselectors_rfecv.html @@ -221,14 +221,14 @@

Examples#> 8: TRUE TRUE FALSE TRUE FALSE FALSE FALSE #> classif.ce runtime_learners timestamp batch_nr warnings errors #> <num> <num> <POSc> <int> <int> <int> -#> 1: 0.08695652 0.006 2024-07-24 11:01:51 1 0 0 -#> 2: 0.03478261 0.005 2024-07-24 11:01:51 1 0 0 -#> 3: 0.03508772 0.005 2024-07-24 11:01:51 1 0 0 -#> 4: 0.11304348 0.006 2024-07-24 11:01:51 2 0 0 -#> 5: 0.03478261 0.005 2024-07-24 11:01:51 2 0 0 -#> 6: 0.03508772 0.005 2024-07-24 11:01:51 2 0 0 -#> 7: 0.03488372 0.006 2024-07-24 11:01:51 3 0 0 -#> 8: 0.03779070 0.005 2024-07-24 11:01:51 4 0 0 +#> 1: 0.08695652 0.006 2024-07-24 12:01:17 1 0 0 +#> 2: 0.03478261 0.005 2024-07-24 12:01:17 1 0 0 +#> 3: 0.03508772 0.004 2024-07-24 12:01:17 1 0 0 +#> 4: 0.11304348 0.006 2024-07-24 12:01:17 2 0 0 +#> 5: 0.03478261 0.005 2024-07-24 12:01:17 2 0 0 +#> 6: 0.03508772 0.004 2024-07-24 12:01:17 2 0 0 +#> 7: 0.03488372 0.006 2024-07-24 12:01:17 3 0 0 +#> 8: 0.03779070 0.005 2024-07-24 12:01:17 4 0 0 #> importance iteration #> <list> <int> #> 1: 95.543823,86.523123,86.157289,83.431536,77.058416, 7.495822,... 1 diff --git a/dev/reference/mlr_fselectors_sequential.html b/dev/reference/mlr_fselectors_sequential.html index da185fb3..e0e960d0 100644 --- a/dev/reference/mlr_fselectors_sequential.html +++ b/dev/reference/mlr_fselectors_sequential.html @@ -208,19 +208,19 @@

Examples#> 13: FALSE FALSE FALSE TRUE FALSE FALSE TRUE #> classif.ce runtime_learners timestamp batch_nr warnings errors #> <num> <num> <POSc> <int> <int> <int> -#> 1: 0.26086957 0.005 2024-07-24 11:01:52 1 0 0 -#> 2: 0.26086957 0.005 2024-07-24 11:01:52 1 0 0 -#> 3: 0.30434783 0.004 2024-07-24 11:01:52 1 0 0 -#> 4: 0.17391304 0.024 2024-07-24 11:01:52 1 0 0 -#> 5: 0.26086957 0.007 2024-07-24 11:01:52 1 0 0 -#> 6: 0.58260870 0.005 2024-07-24 11:01:52 1 0 0 -#> 7: 0.58260870 0.005 2024-07-24 11:01:52 1 0 0 -#> 8: 0.19130435 0.005 2024-07-24 11:01:52 2 0 0 -#> 9: 0.05217391 0.005 2024-07-24 11:01:52 2 0 0 -#> 10: 0.15652174 0.005 2024-07-24 11:01:52 2 0 0 -#> 11: 0.11304348 0.005 2024-07-24 11:01:52 2 0 0 -#> 12: 0.16521739 0.004 2024-07-24 11:01:52 2 0 0 -#> 13: 0.16521739 0.004 2024-07-24 11:01:52 2 0 0 +#> 1: 0.26086957 0.005 2024-07-24 12:01:18 1 0 0 +#> 2: 0.26086957 0.003 2024-07-24 12:01:18 1 0 0 +#> 3: 0.30434783 0.003 2024-07-24 12:01:18 1 0 0 +#> 4: 0.17391304 0.006 2024-07-24 12:01:18 1 0 0 +#> 5: 0.26086957 0.006 2024-07-24 12:01:18 1 0 0 +#> 6: 0.58260870 0.005 2024-07-24 12:01:18 1 0 0 +#> 7: 0.58260870 0.004 2024-07-24 12:01:18 1 0 0 +#> 8: 0.19130435 0.004 2024-07-24 12:01:18 2 0 0 +#> 9: 0.05217391 0.003 2024-07-24 12:01:18 2 0 0 +#> 10: 0.15652174 0.003 2024-07-24 12:01:18 2 0 0 +#> 11: 0.11304348 0.004 2024-07-24 12:01:18 2 0 0 +#> 12: 0.16521739 0.004 2024-07-24 12:01:18 2 0 0 +#> 13: 0.16521739 0.004 2024-07-24 12:01:18 2 0 0 #> features n_features resample_result #> <list> <list> <list> #> 1: bill_depth 1 <ResampleResult> diff --git a/dev/reference/mlr_fselectors_shadow_variable_search.html b/dev/reference/mlr_fselectors_shadow_variable_search.html index e16e38af..6f22b628 100644 --- a/dev/reference/mlr_fselectors_shadow_variable_search.html +++ b/dev/reference/mlr_fselectors_shadow_variable_search.html @@ -242,56 +242,56 @@

Examples#> bill_depth bill_length body_mass flipper_length island sex year #> classif.ce runtime_learners timestamp batch_nr #> <num> <num> <POSc> <int> -#> 1: 0.32173913 0.014 2024-07-24 11:01:53 1 -#> 2: 0.24347826 0.013 2024-07-24 11:01:53 1 -#> 3: 0.31304348 0.012 2024-07-24 11:01:53 1 -#> 4: 0.19130435 0.012 2024-07-24 11:01:53 1 -#> 5: 0.28695652 0.013 2024-07-24 11:01:53 1 -#> 6: 0.55652174 0.012 2024-07-24 11:01:53 1 -#> 7: 0.55652174 0.011 2024-07-24 11:01:53 1 -#> 8: 0.60000000 0.010 2024-07-24 11:01:53 1 -#> 9: 0.61739130 0.010 2024-07-24 11:01:53 1 -#> 10: 0.60000000 0.009 2024-07-24 11:01:53 1 -#> 11: 0.57391304 0.011 2024-07-24 11:01:53 1 -#> 12: 0.58260870 0.009 2024-07-24 11:01:53 1 -#> 13: 0.55652174 0.010 2024-07-24 11:01:53 1 -#> 14: 0.55652174 0.010 2024-07-24 11:01:53 1 -#> 15: 0.21739130 0.014 2024-07-24 11:01:54 2 -#> 16: 0.07826087 0.012 2024-07-24 11:01:54 2 -#> 17: 0.19130435 0.012 2024-07-24 11:01:54 2 -#> 18: 0.13043478 0.012 2024-07-24 11:01:54 2 -#> 19: 0.25217391 0.012 2024-07-24 11:01:54 2 -#> 20: 0.19130435 0.012 2024-07-24 11:01:54 2 -#> 21: 0.19130435 0.013 2024-07-24 11:01:54 2 -#> 22: 0.18260870 0.013 2024-07-24 11:01:54 2 -#> 23: 0.20000000 0.012 2024-07-24 11:01:54 2 -#> 24: 0.20869565 0.012 2024-07-24 11:01:54 2 -#> 25: 0.19130435 0.012 2024-07-24 11:01:54 2 -#> 26: 0.19130435 0.012 2024-07-24 11:01:54 2 -#> 27: 0.19130435 0.013 2024-07-24 11:01:54 2 -#> 28: 0.07826087 0.013 2024-07-24 11:01:54 3 -#> 29: 0.07826087 0.012 2024-07-24 11:01:54 3 -#> 30: 0.06956522 0.012 2024-07-24 11:01:54 3 -#> 31: 0.07826087 0.012 2024-07-24 11:01:54 3 -#> 32: 0.07826087 0.012 2024-07-24 11:01:54 3 -#> 33: 0.07826087 0.012 2024-07-24 11:01:54 3 -#> 34: 0.07826087 0.012 2024-07-24 11:01:54 3 -#> 35: 0.07826087 0.012 2024-07-24 11:01:54 3 -#> 36: 0.07826087 0.012 2024-07-24 11:01:54 3 -#> 37: 0.07826087 0.013 2024-07-24 11:01:54 3 -#> 38: 0.07826087 0.012 2024-07-24 11:01:54 3 -#> 39: 0.07826087 0.018 2024-07-24 11:01:54 3 -#> 40: 0.06956522 0.013 2024-07-24 11:01:54 4 -#> 41: 0.06956522 0.013 2024-07-24 11:01:54 4 -#> 42: 0.06956522 0.013 2024-07-24 11:01:54 4 -#> 43: 0.06956522 0.012 2024-07-24 11:01:54 4 -#> 44: 0.06956522 0.013 2024-07-24 11:01:54 4 -#> 45: 0.06956522 0.012 2024-07-24 11:01:54 4 -#> 46: 0.06956522 0.012 2024-07-24 11:01:54 4 -#> 47: 0.06956522 0.012 2024-07-24 11:01:54 4 -#> 48: 0.06956522 0.013 2024-07-24 11:01:54 4 -#> 49: 0.06956522 0.036 2024-07-24 11:01:54 4 -#> 50: 0.06956522 0.018 2024-07-24 11:01:54 4 +#> 1: 0.32173913 0.012 2024-07-24 12:01:19 1 +#> 2: 0.24347826 0.012 2024-07-24 12:01:19 1 +#> 3: 0.31304348 0.012 2024-07-24 12:01:19 1 +#> 4: 0.19130435 0.012 2024-07-24 12:01:19 1 +#> 5: 0.28695652 0.011 2024-07-24 12:01:19 1 +#> 6: 0.55652174 0.010 2024-07-24 12:01:19 1 +#> 7: 0.55652174 0.012 2024-07-24 12:01:19 1 +#> 8: 0.60000000 0.010 2024-07-24 12:01:19 1 +#> 9: 0.61739130 0.009 2024-07-24 12:01:19 1 +#> 10: 0.60000000 0.010 2024-07-24 12:01:19 1 +#> 11: 0.57391304 0.010 2024-07-24 12:01:19 1 +#> 12: 0.58260870 0.011 2024-07-24 12:01:19 1 +#> 13: 0.55652174 0.010 2024-07-24 12:01:19 1 +#> 14: 0.55652174 0.010 2024-07-24 12:01:19 1 +#> 15: 0.21739130 0.013 2024-07-24 12:01:19 2 +#> 16: 0.07826087 0.012 2024-07-24 12:01:19 2 +#> 17: 0.19130435 0.011 2024-07-24 12:01:19 2 +#> 18: 0.13043478 0.012 2024-07-24 12:01:19 2 +#> 19: 0.25217391 0.012 2024-07-24 12:01:19 2 +#> 20: 0.19130435 0.012 2024-07-24 12:01:19 2 +#> 21: 0.19130435 0.012 2024-07-24 12:01:19 2 +#> 22: 0.18260870 0.012 2024-07-24 12:01:19 2 +#> 23: 0.20000000 0.012 2024-07-24 12:01:19 2 +#> 24: 0.20869565 0.012 2024-07-24 12:01:19 2 +#> 25: 0.19130435 0.012 2024-07-24 12:01:19 2 +#> 26: 0.19130435 0.012 2024-07-24 12:01:19 2 +#> 27: 0.19130435 0.010 2024-07-24 12:01:19 2 +#> 28: 0.07826087 0.014 2024-07-24 12:01:20 3 +#> 29: 0.07826087 0.011 2024-07-24 12:01:20 3 +#> 30: 0.06956522 0.012 2024-07-24 12:01:20 3 +#> 31: 0.07826087 0.012 2024-07-24 12:01:20 3 +#> 32: 0.07826087 0.012 2024-07-24 12:01:20 3 +#> 33: 0.07826087 0.010 2024-07-24 12:01:20 3 +#> 34: 0.07826087 0.011 2024-07-24 12:01:20 3 +#> 35: 0.07826087 0.011 2024-07-24 12:01:20 3 +#> 36: 0.07826087 0.012 2024-07-24 12:01:20 3 +#> 37: 0.07826087 0.012 2024-07-24 12:01:20 3 +#> 38: 0.07826087 0.012 2024-07-24 12:01:20 3 +#> 39: 0.07826087 0.017 2024-07-24 12:01:20 3 +#> 40: 0.06956522 0.012 2024-07-24 12:01:20 4 +#> 41: 0.06956522 0.014 2024-07-24 12:01:20 4 +#> 42: 0.06956522 0.013 2024-07-24 12:01:20 4 +#> 43: 0.06956522 0.012 2024-07-24 12:01:20 4 +#> 44: 0.06956522 0.012 2024-07-24 12:01:20 4 +#> 45: 0.06956522 0.012 2024-07-24 12:01:20 4 +#> 46: 0.06956522 0.013 2024-07-24 12:01:20 4 +#> 47: 0.06956522 0.012 2024-07-24 12:01:20 4 +#> 48: 0.06956522 0.012 2024-07-24 12:01:20 4 +#> 49: 0.06956522 0.034 2024-07-24 12:01:20 4 +#> 50: 0.06956522 0.017 2024-07-24 12:01:20 4 #> classif.ce runtime_learners timestamp batch_nr #> permuted__bill_depth permuted__bill_length permuted__body_mass #> <lgcl> <lgcl> <lgcl> diff --git a/dev/search.json b/dev/search.json index d1303986..071148ca 100644 --- a/dev/search.json +++ b/dev/search.json @@ -1 +1 @@ -[{"path":"https://mlr3fselect.mlr-org.com/dev/authors.html","id":null,"dir":"","previous_headings":"","what":"Authors","title":"Authors and Citation","text":"Marc Becker. Author, maintainer. Patrick Schratz. Author. Michel Lang. Author. Bernd Bischl. Author. John Zobolas. Author.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/authors.html","id":"citation","dir":"","previous_headings":"","what":"Citation","title":"Authors and Citation","text":"Becker M, Schratz P, Lang M, Bischl B, Zobolas J (2024). mlr3fselect: Feature Selection 'mlr3'. R package version 1.0.0.9000, https://github.com/mlr-org/mlr3fselect, https://mlr3fselect.mlr-org.com.","code":"@Manual{, title = {mlr3fselect: Feature Selection for 'mlr3'}, author = {Marc Becker and Patrick Schratz and Michel Lang and Bernd Bischl and John Zobolas}, year = {2024}, note = {R package version 1.0.0.9000, https://github.com/mlr-org/mlr3fselect}, url = {https://mlr3fselect.mlr-org.com}, }"},{"path":"https://mlr3fselect.mlr-org.com/dev/index.html","id":"mlr3fselect-","dir":"","previous_headings":"","what":"Feature Selection for mlr3","title":"Feature Selection for mlr3","text":"Package website: release | dev mlr3fselect feature selection package mlr3 ecosystem. selects optimal feature set mlr3 learner. package works several optimization algorithms e.g. Random Search, Recursive Feature Elimination, Genetic Search. Moreover, can automatically optimize learners estimate performance optimized feature sets nested resampling. package built optimization framework bbotk.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/index.html","id":"resources","dir":"","previous_headings":"","what":"Resources","title":"Feature Selection for mlr3","text":"several section feature selection mlr3book. Getting started wrapper feature selection. sequential forward selection Palmer Penguins data set. Optimize multiple performance measures. Estimate Model Performance nested resampling. gallery features collection case studies demos optimization. Utilize built-feature importance models Recursive Feature Elimination. Run feature selection Shadow Variable Search. Feature Selection Titanic data set. cheatsheet summarizes important functions mlr3fselect.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/index.html","id":"installation","dir":"","previous_headings":"","what":"Installation","title":"Feature Selection for mlr3","text":"Install last release CRAN: Install development version GitHub:","code":"install.packages(\"mlr3fselect\") remotes::install_github(\"mlr-org/mlr3fselect\")"},{"path":"https://mlr3fselect.mlr-org.com/dev/index.html","id":"example","dir":"","previous_headings":"","what":"Example","title":"Feature Selection for mlr3","text":"run feature selection support vector machine Spam data set. construct instance fsi() function. instance describes optimization problem. select simple random search optimization algorithm. start feature selection, simply pass instance fselector. fselector writes best hyperparameter configuration instance. corresponding measured performance. archive contains evaluated hyperparameter configurations. fit final model optimized feature set make predictions new data.","code":"library(\"mlr3verse\") tsk(\"spam\") ## (4601 x 58): HP Spam Detection ## * Target: type ## * Properties: twoclass ## * Features (57): ## - dbl (57): address, addresses, all, business, capitalAve, capitalLong, capitalTotal, ## charDollar, charExclamation, charHash, charRoundbracket, charSemicolon, ## charSquarebracket, conference, credit, cs, data, direct, edu, email, font, free, ## george, hp, hpl, internet, lab, labs, mail, make, meeting, money, num000, num1999, ## num3d, num415, num650, num85, num857, order, original, our, over, parts, people, pm, ## project, re, receive, remove, report, table, technology, telnet, will, you, your instance = fsi( task = tsk(\"spam\"), learner = lrn(\"classif.svm\", type = \"C-classification\"), resampling = rsmp(\"cv\", folds = 3), measures = msr(\"classif.ce\"), terminator = trm(\"evals\", n_evals = 20) ) instance ## ## * State: Not optimized ## * Objective: ## * Terminator: fselector = fs(\"random_search\", batch_size = 5) fselector ## : Random Search ## * Parameters: batch_size=5 ## * Properties: single-crit, multi-crit ## * Packages: mlr3fselect fselector$optimize(instance) instance$result_feature_set ## [1] \"address\" \"addresses\" \"all\" \"business\" ## [5] \"capitalAve\" \"capitalLong\" \"capitalTotal\" \"charDollar\" ## [9] \"charExclamation\" \"charHash\" \"charRoundbracket\" \"charSemicolon\" ## [13] \"charSquarebracket\" \"conference\" \"credit\" \"cs\" ## [17] \"data\" \"direct\" \"edu\" \"email\" ## [21] \"font\" \"free\" \"george\" \"hp\" ## [25] \"internet\" \"lab\" \"labs\" \"mail\" ## [29] \"make\" \"meeting\" \"money\" \"num000\" ## [33] \"num1999\" \"num3d\" \"num415\" \"num650\" ## [37] \"num85\" \"num857\" \"order\" \"our\" ## [41] \"parts\" \"people\" \"pm\" \"project\" ## [45] \"re\" \"receive\" \"remove\" \"report\" ## [49] \"table\" \"technology\" \"telnet\" \"will\" ## [53] \"you\" \"your\" instance$result_y ## classif.ce ## 0.07042005 as.data.table(instance$archive) ## address addresses all business capitalAve capitalLong capitalTotal charDollar charExclamation ## 1: TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE ## 2: TRUE TRUE TRUE FALSE FALSE TRUE TRUE TRUE TRUE ## 3: TRUE TRUE FALSE FALSE TRUE TRUE TRUE TRUE TRUE ## 4: TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE ## 5: FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE ## --- ## 16: FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE ## 17: FALSE FALSE FALSE TRUE TRUE TRUE FALSE FALSE TRUE ## 18: FALSE FALSE TRUE TRUE FALSE FALSE FALSE TRUE FALSE ## 19: TRUE TRUE TRUE TRUE FALSE TRUE TRUE TRUE TRUE ## 20: TRUE FALSE TRUE FALSE FALSE TRUE FALSE TRUE FALSE ## 56 variables not shown: [charHash, charRoundbracket, charSemicolon, charSquarebracket, conference, credit, cs, data, direct, edu, ...] task = tsk(\"spam\") learner = lrn(\"classif.svm\", type = \"C-classification\") task$select(instance$result_feature_set) learner$train(task)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":null,"dir":"Reference","previous_headings":"","what":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"ArchiveBatchFSelect stores evaluated feature sets performance scores.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"ArchiveBatchFSelect container around data.table::data.table(). row corresponds single evaluation feature set. See section Data Structure information. archive stores additionally mlr3::BenchmarkResult ($benchmark_result) records resampling experiments. experiment corresponds single evaluation feature set. table ($data) benchmark result ($benchmark_result) linked uhash column. archive passed .data.table(), joined automatically.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"data-structure","dir":"Reference","previous_headings":"","what":"Data structure","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"table ($data) following columns: One column feature task ($search_space). One column performance measure ($codomain). runtime_learners (numeric(1)) Sum training predict times logged learners per mlr3::ResampleResult / evaluation. include potential overhead time. timestamp (POSIXct) Time stamp evaluation logged archive. batch_nr (integer(1)) Feature sets evaluated batches. batch unique batch number. uhash (character(1)) Connects feature set resampling experiment stored mlr3::BenchmarkResult.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"analysis","dir":"Reference","previous_headings":"","what":"Analysis","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"analyzing feature selection results, recommended pass archive .data.table(). returned data table joined benchmark result adds mlr3::ResampleResult feature set. archive provides various getters (e.g. $learners()) ease access. getters extract position () unique hash (uhash). complete list getters see methods section. benchmark result ($benchmark_result) allows score feature sets different measure. Alternatively, measures can supplied .data.table().","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"s-methods","dir":"Reference","previous_headings":"","what":"S3 Methods","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":".data.table.ArchiveBatchFSelect(x, exclude_columns = \"uhash\", measures = NULL) Returns tabular view evaluated feature sets. ArchiveBatchFSelect -> data.table::data.table() x (ArchiveBatchFSelect) exclude_columns (character()) Exclude columns table. Set NULL column excluded. measures (list mlr3::Measure) Score feature sets additional measures.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"super-classes","dir":"Reference","previous_headings":"","what":"Super classes","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"bbotk::Archive -> bbotk::ArchiveBatch -> ArchiveBatchFSelect","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"public-fields","dir":"Reference","previous_headings":"","what":"Public fields","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"benchmark_result (mlr3::BenchmarkResult) Benchmark result.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"active-bindings","dir":"Reference","previous_headings":"","what":"Active bindings","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"ties_method (character(1)) Method handle ties.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"bbotk::Archive$format() bbotk::Archive$help() bbotk::ArchiveBatch$add_evals() bbotk::ArchiveBatch$clear() bbotk::ArchiveBatch$nds_selection()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"public-methods","dir":"Reference","previous_headings":"","what":"Public methods","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"ArchiveBatchFSelect$new() ArchiveBatchFSelect$learner() ArchiveBatchFSelect$learners() ArchiveBatchFSelect$predictions() ArchiveBatchFSelect$resample_result() ArchiveBatchFSelect$print() ArchiveBatchFSelect$best() ArchiveBatchFSelect$clone()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"method-new-","dir":"Reference","previous_headings":"","what":"Method new()","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"Creates new instance R6 class.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"","code":"ArchiveBatchFSelect$new( search_space, codomain, check_values = TRUE, ties_method = \"least_features\" )"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"search_space (paradox::ParamSet) Search space. Internally created provided mlr3::Task instance. codomain (bbotk::Codomain) Specifies codomain objective function .e. set performance measures. Internally created provided mlr3::Measures instance. check_values (logical(1)) TRUE (default), hyperparameter configurations check validity. ties_method (character(1)) method break ties selecting sets optimizing selecting best set. Can \"least_features\" \"random\". option \"least_features\" (default) selects feature set least features. multiple best feature sets number features, one selected randomly. random method returns random feature set best feature sets. Ignored multiple measures used.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"method-learner-","dir":"Reference","previous_headings":"","what":"Method learner()","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"Retrieve mlr3::Learner -th evaluation, position unique hash uhash. uhash mutually exclusive. Learner contain model. Use $learners() get learners models.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"usage-1","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"","code":"ArchiveBatchFSelect$learner(i = NULL, uhash = NULL)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"arguments-1","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"(integer(1)) iteration value filter . uhash (logical(1)) uhash value filter .","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"method-learners-","dir":"Reference","previous_headings":"","what":"Method learners()","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"Retrieve list trained mlr3::Learner objects -th evaluation, position unique hash uhash. uhash mutually exclusive.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"usage-2","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"","code":"ArchiveBatchFSelect$learners(i = NULL, uhash = NULL)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"arguments-2","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"(integer(1)) iteration value filter . uhash (logical(1)) uhash value filter .","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"method-predictions-","dir":"Reference","previous_headings":"","what":"Method predictions()","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"Retrieve list mlr3::Prediction objects -th evaluation, position unique hash uhash. uhash mutually exclusive.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"usage-3","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"","code":"ArchiveBatchFSelect$predictions(i = NULL, uhash = NULL)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"arguments-3","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"(integer(1)) iteration value filter . uhash (logical(1)) uhash value filter .","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"method-resample-result-","dir":"Reference","previous_headings":"","what":"Method resample_result()","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"Retrieve mlr3::ResampleResult -th evaluation, position unique hash uhash. uhash mutually exclusive.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"usage-4","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"","code":"ArchiveBatchFSelect$resample_result(i = NULL, uhash = NULL)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"arguments-4","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"(integer(1)) iteration value filter . uhash (logical(1)) uhash value filter .","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"method-print-","dir":"Reference","previous_headings":"","what":"Method print()","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"Printer.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"usage-5","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"","code":"ArchiveBatchFSelect$print()"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"arguments-5","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"... (ignored).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"method-best-","dir":"Reference","previous_headings":"","what":"Method best()","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"Returns best scoring feature sets.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"usage-6","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"","code":"ArchiveBatchFSelect$best(batch = NULL, ties_method = NULL)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"arguments-6","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"batch (integer()) batch number(s) limit best results . Default batches. ties_method (character(1)) Method handle ties. NULL (default), global ties method set initialization used. default global ties method least_features selects feature set least features. multiple best feature sets number features, one selected randomly. random method returns random feature set best feature sets.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"returns","dir":"Reference","previous_headings":"","what":"Returns","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"data.table::data.table()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"method-clone-","dir":"Reference","previous_headings":"","what":"Method clone()","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"objects class cloneable method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"usage-7","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"","code":"ArchiveBatchFSelect$clone(deep = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"arguments-7","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"deep Whether make deep clone.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":null,"dir":"Reference","previous_headings":"","what":"Class for Automatic Feature Selection — AutoFSelector","title":"Class for Automatic Feature Selection — AutoFSelector","text":"AutoFSelector wraps mlr3::Learner augments automatic feature selection. auto_fselector() function creates AutoFSelector object.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Class for Automatic Feature Selection — AutoFSelector","text":"AutoFSelector mlr3::Learner wraps another mlr3::Learner performs following steps $train(): wrapped (inner) learner trained feature subsets via resampling. feature selection can specified providing FSelector, bbotk::Terminator, mlr3::Resampling mlr3::Measure. final model fit complete training data best-found feature subset. $predict() AutoFSelector just calls predict method wrapped (inner) learner.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"resources","dir":"Reference","previous_headings":"","what":"Resources","title":"Class for Automatic Feature Selection — AutoFSelector","text":"several sections feature selection mlr3book. Estimate Model Performance nested resampling. gallery features collection case studies demos optimization.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"nested-resampling","dir":"Reference","previous_headings":"","what":"Nested Resampling","title":"Class for Automatic Feature Selection — AutoFSelector","text":"Nested resampling can performed passing AutoFSelector object mlr3::resample() mlr3::benchmark(). access inner resampling results, set store_fselect_instance = TRUE execute mlr3::resample() mlr3::benchmark() store_models = TRUE (see examples). mlr3::Resampling passed AutoFSelector meant inner resampling, operating training set arbitrary outer resampling. reason feasible pass instantiated mlr3::Resampling .","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"super-class","dir":"Reference","previous_headings":"","what":"Super class","title":"Class for Automatic Feature Selection — AutoFSelector","text":"mlr3::Learner -> AutoFSelector","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"public-fields","dir":"Reference","previous_headings":"","what":"Public fields","title":"Class for Automatic Feature Selection — AutoFSelector","text":"instance_args (list()) arguments construction create FSelectInstanceBatchSingleCrit. fselector (FSelector) Optimization algorithm.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"active-bindings","dir":"Reference","previous_headings":"","what":"Active bindings","title":"Class for Automatic Feature Selection — AutoFSelector","text":"archive ([ArchiveBatchFSelect) Returns FSelectInstanceBatchSingleCrit archive. learner (mlr3::Learner) Trained learner. fselect_instance (FSelectInstanceBatchSingleCrit) Internally created feature selection instance intermediate results. fselect_result (data.table::data.table) Short-cut $result FSelectInstanceBatchSingleCrit. predict_type (character(1)) Stores currently active predict type, e.g. \"response\". Must element $predict_types. hash (character(1)) Hash (unique identifier) object. phash (character(1)) Hash (unique identifier) partial object, excluding components varied systematically tuning (parameter values) feature selection (feature names).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Class for Automatic Feature Selection — AutoFSelector","text":"mlr3::Learner$format() mlr3::Learner$help() mlr3::Learner$predict() mlr3::Learner$predict_newdata() mlr3::Learner$reset() mlr3::Learner$train()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"public-methods","dir":"Reference","previous_headings":"","what":"Public methods","title":"Class for Automatic Feature Selection — AutoFSelector","text":"AutoFSelector$new() AutoFSelector$base_learner() AutoFSelector$importance() AutoFSelector$selected_features() AutoFSelector$oob_error() AutoFSelector$loglik() AutoFSelector$print() AutoFSelector$clone()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"method-new-","dir":"Reference","previous_headings":"","what":"Method new()","title":"Class for Automatic Feature Selection — AutoFSelector","text":"Creates new instance R6 class.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Automatic Feature Selection — AutoFSelector","text":"","code":"AutoFSelector$new( fselector, learner, resampling, measure = NULL, terminator, store_fselect_instance = TRUE, store_benchmark_result = TRUE, store_models = FALSE, check_values = FALSE, callbacks = NULL, ties_method = \"least_features\" )"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Automatic Feature Selection — AutoFSelector","text":"fselector (FSelector) Optimization algorithm. learner (mlr3::Learner) Learner optimize feature subset . resampling (mlr3::Resampling) Resampling used evaluated performance feature subsets. Uninstantiated resamplings instantiated construction feature subsets evaluated data splits. Already instantiated resamplings kept unchanged. measure (mlr3::Measure) Measure optimize. NULL, default measure used. terminator (bbotk::Terminator) Stop criterion feature selection. store_fselect_instance (logical(1)) TRUE (default), stores internally created FSelectInstanceBatchSingleCrit intermediate results slot $fselect_instance. set TRUE, store_models = TRUE store_benchmark_result (logical(1)) Store benchmark result archive? store_models (logical(1)). Store models benchmark result? check_values (logical(1)) Check parameters evaluation results validity? callbacks (list CallbackBatchFSelect) List callbacks. ties_method (character(1)) method break ties selecting sets optimizing selecting best set. Can \"least_features\" \"random\". option \"least_features\" (default) selects feature set least features. multiple best feature sets number features, one selected randomly. random method returns random feature set best feature sets. Ignored multiple measures used.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"method-base-learner-","dir":"Reference","previous_headings":"","what":"Method base_learner()","title":"Class for Automatic Feature Selection — AutoFSelector","text":"Extracts base learner nested learner objects like GraphLearner mlr3pipelines. recursive = 0, (tuned) learner returned.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"usage-1","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Automatic Feature Selection — AutoFSelector","text":"","code":"AutoFSelector$base_learner(recursive = Inf)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"arguments-1","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Automatic Feature Selection — AutoFSelector","text":"recursive (integer(1)) Depth recursion multiple nested objects.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"returns","dir":"Reference","previous_headings":"","what":"Returns","title":"Class for Automatic Feature Selection — AutoFSelector","text":"mlr3::Learner.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"method-importance-","dir":"Reference","previous_headings":"","what":"Method importance()","title":"Class for Automatic Feature Selection — AutoFSelector","text":"importance scores final model.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"usage-2","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Automatic Feature Selection — AutoFSelector","text":"","code":"AutoFSelector$importance()"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"returns-1","dir":"Reference","previous_headings":"","what":"Returns","title":"Class for Automatic Feature Selection — AutoFSelector","text":"Named numeric().","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"method-selected-features-","dir":"Reference","previous_headings":"","what":"Method selected_features()","title":"Class for Automatic Feature Selection — AutoFSelector","text":"selected features final model. features selected internally learner.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"usage-3","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Automatic Feature Selection — AutoFSelector","text":"","code":"AutoFSelector$selected_features()"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"returns-2","dir":"Reference","previous_headings":"","what":"Returns","title":"Class for Automatic Feature Selection — AutoFSelector","text":"character().","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"method-oob-error-","dir":"Reference","previous_headings":"","what":"Method oob_error()","title":"Class for Automatic Feature Selection — AutoFSelector","text":"--bag error final model.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"usage-4","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Automatic Feature Selection — AutoFSelector","text":"","code":"AutoFSelector$oob_error()"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"returns-3","dir":"Reference","previous_headings":"","what":"Returns","title":"Class for Automatic Feature Selection — AutoFSelector","text":"numeric(1).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"method-loglik-","dir":"Reference","previous_headings":"","what":"Method loglik()","title":"Class for Automatic Feature Selection — AutoFSelector","text":"log-likelihood final model.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"usage-5","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Automatic Feature Selection — AutoFSelector","text":"","code":"AutoFSelector$loglik()"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"returns-4","dir":"Reference","previous_headings":"","what":"Returns","title":"Class for Automatic Feature Selection — AutoFSelector","text":"logLik. Printer.","code":""},{"path":[]},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"usage-6","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Automatic Feature Selection — AutoFSelector","text":"","code":"AutoFSelector$print()"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"arguments-2","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Automatic Feature Selection — AutoFSelector","text":"... (ignored).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"method-clone-","dir":"Reference","previous_headings":"","what":"Method clone()","title":"Class for Automatic Feature Selection — AutoFSelector","text":"objects class cloneable method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"usage-7","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Automatic Feature Selection — AutoFSelector","text":"","code":"AutoFSelector$clone(deep = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"arguments-3","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Automatic Feature Selection — AutoFSelector","text":"deep Whether make deep clone.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Class for Automatic Feature Selection — AutoFSelector","text":"","code":"# Automatic Feature Selection # \\donttest{ # split to train and external set task = tsk(\"penguins\") split = partition(task, ratio = 0.8) # create auto fselector afs = auto_fselector( fselector = fs(\"random_search\"), learner = lrn(\"classif.rpart\"), resampling = rsmp (\"holdout\"), measure = msr(\"classif.ce\"), term_evals = 4) # optimize feature subset and fit final model afs$train(task, row_ids = split$train) # predict with final model afs$predict(task, row_ids = split$test) #> for 69 observations: #> row_ids truth response #> 15 Adelie Adelie #> 16 Adelie Adelie #> 20 Adelie Chinstrap #> --- #> 338 Chinstrap Chinstrap #> 340 Chinstrap Gentoo #> 344 Chinstrap Chinstrap # show result afs$fselect_result #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: FALSE TRUE FALSE TRUE FALSE TRUE FALSE #> features n_features classif.ce #> #> 1: bill_length,flipper_length,sex 3 0.05434783 # model slot contains trained learner and fselect instance afs$model #> $learner #> : Classification Tree #> * Model: rpart #> * Parameters: xval=0 #> * Packages: mlr3, rpart #> * Predict Types: [response], prob #> * Feature Types: logical, integer, numeric, factor, ordered #> * Properties: importance, missings, multiclass, selected_features, #> twoclass, weights #> #> $features #> [1] \"bill_length\" \"flipper_length\" \"sex\" #> #> $fselect_instance #> #> * State: Optimized #> * Objective: #> * Terminator: #> * Result: #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: FALSE TRUE FALSE TRUE FALSE TRUE FALSE #> classif.ce #> #> 1: 0.05434783 #> * Archive: #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> 2: FALSE TRUE FALSE TRUE FALSE TRUE FALSE #> 3: TRUE FALSE FALSE TRUE FALSE FALSE TRUE #> 4: FALSE FALSE FALSE TRUE TRUE FALSE FALSE #> 5: TRUE TRUE TRUE FALSE TRUE TRUE TRUE #> 6: FALSE FALSE FALSE TRUE TRUE TRUE TRUE #> 7: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> 8: FALSE TRUE TRUE TRUE FALSE TRUE FALSE #> 9: TRUE TRUE TRUE TRUE TRUE TRUE FALSE #> 10: FALSE TRUE FALSE FALSE TRUE FALSE TRUE #> classif.ce #> #> 1: 0.05434783 #> 2: 0.05434783 #> 3: 0.18478261 #> 4: 0.11956522 #> 5: 0.08695652 #> 6: 0.13043478 #> 7: 0.05434783 #> 8: 0.05434783 #> 9: 0.05434783 #> 10: 0.08695652 #> # shortcut trained learner afs$learner #> : Classification Tree #> * Model: rpart #> * Parameters: xval=0 #> * Packages: mlr3, rpart #> * Predict Types: [response], prob #> * Feature Types: logical, integer, numeric, factor, ordered #> * Properties: importance, missings, multiclass, selected_features, #> twoclass, weights # shortcut fselect instance afs$fselect_instance #> #> * State: Optimized #> * Objective: #> * Terminator: #> * Result: #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: FALSE TRUE FALSE TRUE FALSE TRUE FALSE #> classif.ce #> #> 1: 0.05434783 #> * Archive: #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> 2: FALSE TRUE FALSE TRUE FALSE TRUE FALSE #> 3: TRUE FALSE FALSE TRUE FALSE FALSE TRUE #> 4: FALSE FALSE FALSE TRUE TRUE FALSE FALSE #> 5: TRUE TRUE TRUE FALSE TRUE TRUE TRUE #> 6: FALSE FALSE FALSE TRUE TRUE TRUE TRUE #> 7: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> 8: FALSE TRUE TRUE TRUE FALSE TRUE FALSE #> 9: TRUE TRUE TRUE TRUE TRUE TRUE FALSE #> 10: FALSE TRUE FALSE FALSE TRUE FALSE TRUE #> classif.ce #> #> 1: 0.05434783 #> 2: 0.05434783 #> 3: 0.18478261 #> 4: 0.11956522 #> 5: 0.08695652 #> 6: 0.13043478 #> 7: 0.05434783 #> 8: 0.05434783 #> 9: 0.05434783 #> 10: 0.08695652 # Nested Resampling afs = auto_fselector( fselector = fs(\"random_search\"), learner = lrn(\"classif.rpart\"), resampling = rsmp (\"holdout\"), measure = msr(\"classif.ce\"), term_evals = 4) resampling_outer = rsmp(\"cv\", folds = 3) rr = resample(task, afs, resampling_outer, store_models = TRUE) # retrieve inner feature selection results. extract_inner_fselect_results(rr) #> iteration bill_depth bill_length body_mass flipper_length island sex #> #> 1: 1 TRUE TRUE FALSE FALSE TRUE FALSE #> 2: 2 FALSE TRUE FALSE TRUE TRUE FALSE #> 3: 3 TRUE TRUE TRUE TRUE FALSE FALSE #> year classif.ce features n_features #> #> 1: FALSE 0.03947368 bill_depth,bill_length,island 3 #> 2: TRUE 0.07894737 bill_length,flipper_length,island,year 4 #> 3: FALSE 0.05194805 bill_depth,bill_length,body_mass,flipper_length 4 #> task_id learner_id resampling_id #> #> 1: penguins classif.rpart.fselector cv #> 2: penguins classif.rpart.fselector cv #> 3: penguins classif.rpart.fselector cv # performance scores estimated on the outer resampling rr$score() #> task_id learner_id resampling_id iteration classif.ce #> #> 1: penguins classif.rpart.fselector cv 1 0.10434783 #> 2: penguins classif.rpart.fselector cv 2 0.04347826 #> 3: penguins classif.rpart.fselector cv 3 0.05263158 #> Hidden columns: task, learner, resampling, prediction # unbiased performance of the final model trained on the full data set rr$aggregate() #> classif.ce #> 0.06681922 # }"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/CallbackBatchFSelect.html","id":null,"dir":"Reference","previous_headings":"","what":"Create Feature Selection Callback — CallbackBatchFSelect","title":"Create Feature Selection Callback — CallbackBatchFSelect","text":"Specialized bbotk::CallbackBatch feature selection. Callbacks allow customizing behavior processes mlr3fselect. callback_batch_fselect() function creates CallbackBatchFSelect. Predefined callbacks stored dictionary mlr_callbacks can retrieved clbk(). information callbacks see callback_batch_fselect().","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/CallbackBatchFSelect.html","id":"super-classes","dir":"Reference","previous_headings":"","what":"Super classes","title":"Create Feature Selection Callback — CallbackBatchFSelect","text":"mlr3misc::Callback -> bbotk::CallbackBatch -> CallbackBatchFSelect","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/CallbackBatchFSelect.html","id":"public-fields","dir":"Reference","previous_headings":"","what":"Public fields","title":"Create Feature Selection Callback — CallbackBatchFSelect","text":"on_eval_after_design (function()) Stage called design created. Called ObjectiveFSelectBatch$eval_many(). on_eval_after_benchmark (function()) Stage called feature sets evaluated. Called ObjectiveFSelectBatch$eval_many(). on_eval_before_archive (function()) Stage called performance values written archive. Called ObjectiveFSelectBatch$eval_many().","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/CallbackBatchFSelect.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Create Feature Selection Callback — CallbackBatchFSelect","text":"mlr3misc::Callback$call() mlr3misc::Callback$format() mlr3misc::Callback$help() mlr3misc::Callback$initialize() mlr3misc::Callback$print()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/CallbackBatchFSelect.html","id":"public-methods","dir":"Reference","previous_headings":"","what":"Public methods","title":"Create Feature Selection Callback — CallbackBatchFSelect","text":"CallbackBatchFSelect$clone()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/CallbackBatchFSelect.html","id":"method-clone-","dir":"Reference","previous_headings":"","what":"Method clone()","title":"Create Feature Selection Callback — CallbackBatchFSelect","text":"objects class cloneable method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/CallbackBatchFSelect.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Create Feature Selection Callback — CallbackBatchFSelect","text":"","code":"CallbackBatchFSelect$clone(deep = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/CallbackBatchFSelect.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Create Feature Selection Callback — CallbackBatchFSelect","text":"deep Whether make deep clone.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/CallbackBatchFSelect.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Create Feature Selection Callback — CallbackBatchFSelect","text":"","code":"# Write archive to disk callback_batch_fselect(\"mlr3fselect.backup\", on_optimization_end = function(callback, context) { saveRDS(context$instance$archive, \"archive.rds\") } ) #> #> * Active Stages: on_optimization_end"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ContextBatchFSelect.html","id":null,"dir":"Reference","previous_headings":"","what":"Evaluation Context — ContextBatchFSelect","title":"Evaluation Context — ContextBatchFSelect","text":"ContextBatchFSelect allows CallbackBatchFSelects access modify data batch feature sets evaluated. See section active bindings list modifiable objects. See callback_batch_fselect() list stages access ContextBatchFSelect.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ContextBatchFSelect.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Evaluation Context — ContextBatchFSelect","text":"context re-created time new batch feature sets evaluated. Changes $objective_fselect, $design $benchmark_result discarded function finished. Modification data table $aggregated_performance written archive. number columns can added.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ContextBatchFSelect.html","id":"super-classes","dir":"Reference","previous_headings":"","what":"Super classes","title":"Evaluation Context — ContextBatchFSelect","text":"mlr3misc::Context -> bbotk::ContextBatch -> ContextBatchFSelect","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ContextBatchFSelect.html","id":"active-bindings","dir":"Reference","previous_headings":"","what":"Active bindings","title":"Evaluation Context — ContextBatchFSelect","text":"xss (list()) feature sets latest batch. design (data.table::data.table) benchmark design latest batch. benchmark_result (mlr3::BenchmarkResult) benchmark result latest batch. aggregated_performance (data.table::data.table) Aggregated performance scores training time latest batch. data table passed archive. callback can add additional columns also written archive.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ContextBatchFSelect.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Evaluation Context — ContextBatchFSelect","text":"mlr3misc::Context$format() mlr3misc::Context$print() bbotk::ContextBatch$initialize()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ContextBatchFSelect.html","id":"public-methods","dir":"Reference","previous_headings":"","what":"Public methods","title":"Evaluation Context — ContextBatchFSelect","text":"ContextBatchFSelect$clone()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ContextBatchFSelect.html","id":"method-clone-","dir":"Reference","previous_headings":"","what":"Method clone()","title":"Evaluation Context — ContextBatchFSelect","text":"objects class cloneable method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ContextBatchFSelect.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Evaluation Context — ContextBatchFSelect","text":"","code":"ContextBatchFSelect$clone(deep = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ContextBatchFSelect.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Evaluation Context — ContextBatchFSelect","text":"deep Whether make deep clone.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchMultiCrit.html","id":null,"dir":"Reference","previous_headings":"","what":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","title":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","text":"FSelectInstanceBatchMultiCrit specifies feature selection problem FSelector. function fsi() creates FSelectInstanceBatchMultiCrit function fselect() creates instance internally.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchMultiCrit.html","id":"resources","dir":"Reference","previous_headings":"","what":"Resources","title":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","text":"several sections feature selection mlr3book. Learn multi-objective optimization. gallery features collection case studies demos optimization.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchMultiCrit.html","id":"analysis","dir":"Reference","previous_headings":"","what":"Analysis","title":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","text":"analyzing feature selection results, recommended pass archive .data.table(). returned data table joined benchmark result adds mlr3::ResampleResult feature set. archive provides various getters (e.g. $learners()) ease access. getters extract position () unique hash (uhash). complete list getters see methods section. benchmark result ($benchmark_result) allows score feature sets different measure. Alternatively, measures can supplied .data.table().","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchMultiCrit.html","id":"super-classes","dir":"Reference","previous_headings":"","what":"Super classes","title":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","text":"bbotk::OptimInstance -> bbotk::OptimInstanceBatch -> bbotk::OptimInstanceBatchMultiCrit -> FSelectInstanceBatchMultiCrit","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchMultiCrit.html","id":"active-bindings","dir":"Reference","previous_headings":"","what":"Active bindings","title":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","text":"result_feature_set (list character()) Feature sets task subsetting.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchMultiCrit.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","text":"bbotk::OptimInstance$clear() bbotk::OptimInstance$format() bbotk::OptimInstanceBatch$eval_batch() bbotk::OptimInstanceBatch$objective_function()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchMultiCrit.html","id":"public-methods","dir":"Reference","previous_headings":"","what":"Public methods","title":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","text":"FSelectInstanceBatchMultiCrit$new() FSelectInstanceBatchMultiCrit$assign_result() FSelectInstanceBatchMultiCrit$print() FSelectInstanceBatchMultiCrit$clone()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchMultiCrit.html","id":"method-new-","dir":"Reference","previous_headings":"","what":"Method new()","title":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","text":"Creates new instance R6 class.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchMultiCrit.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","text":"","code":"FSelectInstanceBatchMultiCrit$new( task, learner, resampling, measures, terminator, store_benchmark_result = TRUE, store_models = FALSE, check_values = FALSE, callbacks = NULL )"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchMultiCrit.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","text":"task (mlr3::Task) Task operate . learner (mlr3::Learner) Learner optimize feature subset . resampling (mlr3::Resampling) Resampling used evaluated performance feature subsets. Uninstantiated resamplings instantiated construction feature subsets evaluated data splits. Already instantiated resamplings kept unchanged. measures (list mlr3::Measure) Measures optimize. NULL, mlr3's default measure used. terminator (bbotk::Terminator) Stop criterion feature selection. store_benchmark_result (logical(1)) Store benchmark result archive? store_models (logical(1)). Store models benchmark result? check_values (logical(1)) Check parameters evaluation results validity? callbacks (list CallbackBatchFSelect) List callbacks.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchMultiCrit.html","id":"method-assign-result-","dir":"Reference","previous_headings":"","what":"Method assign_result()","title":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","text":"FSelector object writes best found feature subsets estimated performance values . internal use.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchMultiCrit.html","id":"usage-1","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","text":"","code":"FSelectInstanceBatchMultiCrit$assign_result(xdt, ydt)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchMultiCrit.html","id":"arguments-1","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","text":"xdt (data.table::data.table()) x values data.table. row one point. Contains value search space FSelectInstanceBatchMultiCrit object. Can contain additional columns extra information. ydt (data.table::data.table()) Optimal outcomes, e.g. Pareto front.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchMultiCrit.html","id":"method-print-","dir":"Reference","previous_headings":"","what":"Method print()","title":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","text":"Printer.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchMultiCrit.html","id":"usage-2","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","text":"","code":"FSelectInstanceBatchMultiCrit$print(...)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchMultiCrit.html","id":"arguments-2","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","text":"... (ignored).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchMultiCrit.html","id":"method-clone-","dir":"Reference","previous_headings":"","what":"Method clone()","title":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","text":"objects class cloneable method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchMultiCrit.html","id":"usage-3","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","text":"","code":"FSelectInstanceBatchMultiCrit$clone(deep = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchMultiCrit.html","id":"arguments-3","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","text":"deep Whether make deep clone.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchMultiCrit.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","text":"","code":"# Feature selection on Palmer Penguins data set # \\donttest{ task = tsk(\"penguins\") # Construct feature selection instance instance = fsi( task = task, learner = lrn(\"classif.rpart\"), resampling = rsmp(\"cv\", folds = 3), measures = msrs(c(\"classif.ce\", \"time_train\")), terminator = trm(\"evals\", n_evals = 4) ) # Choose optimization algorithm fselector = fs(\"random_search\", batch_size = 2) # Run feature selection fselector$optimize(instance) #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE FALSE TRUE TRUE FALSE TRUE TRUE #> 2: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> features n_features #> #> 1: bill_depth,body_mass,flipper_length,sex,year 5 #> 2: bill_depth,bill_length,body_mass,flipper_length,island,sex,... 5 #> classif.ce time_train #> #> 1: 0.2035850 0.002666667 #> 2: 0.0698449 0.003000000 # Optimal feature sets instance$result_feature_set #> [[1]] #> [1] \"bill_depth\" \"body_mass\" \"flipper_length\" \"sex\" #> [5] \"year\" #> #> [[2]] #> [1] \"bill_depth\" \"bill_length\" \"body_mass\" \"flipper_length\" #> [5] \"island\" \"sex\" \"year\" #> # Inspect all evaluated sets as.data.table(instance$archive) #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE TRUE TRUE TRUE FALSE TRUE TRUE #> 2: TRUE FALSE TRUE TRUE FALSE TRUE TRUE #> 3: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> 4: FALSE TRUE TRUE TRUE FALSE FALSE FALSE #> classif.ce time_train runtime_learners timestamp batch_nr #> #> 1: 0.08151538 0.003000000 0.016 2024-07-24 11:01:20 1 #> 2: 0.20358505 0.002666667 0.014 2024-07-24 11:01:20 1 #> 3: 0.06984490 0.003000000 0.017 2024-07-24 11:01:20 2 #> 4: 0.08151538 0.007666667 0.029 2024-07-24 11:01:20 2 #> warnings errors #> #> 1: 0 0 #> 2: 0 0 #> 3: 0 0 #> 4: 0 0 #> features n_features #> #> 1: bill_depth,bill_length,body_mass,flipper_length,sex,year 6 #> 2: bill_depth,body_mass,flipper_length,sex,year 5 #> 3: bill_depth,bill_length,body_mass,flipper_length,island,sex,... 7 #> 4: bill_length,body_mass,flipper_length 3 #> resample_result #> #> 1: #> 2: #> 3: #> 4: # }"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":null,"dir":"Reference","previous_headings":"","what":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"FSelectInstanceBatchSingleCrit specifies feature selection problem FSelector. function fsi() creates FSelectInstanceBatchSingleCrit function fselect() creates instance internally. instance contains ObjectiveFSelectBatch object encodes black box objective function FSelector optimize. instance allows basic operations querying objective design points ($eval_batch()). operation usually done FSelector. Evaluations feature subsets performed batches calling mlr3::benchmark() internally. evaluated feature subsets stored Archive ($archive). batch evaluated, bbotk::Terminator queried remaining budget. available budget exhausted, exception raised, evaluations can performed point . FSelector also supposed store final result, consisting selected feature subset associated estimated performance values, calling method instance$assign_result().","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":"default-measures","dir":"Reference","previous_headings":"","what":"Default Measures","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"measure passed, default measure used. default measure depends task type.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":"resources","dir":"Reference","previous_headings":"","what":"Resources","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"several sections feature selection mlr3book. Getting started wrapper feature selection. sequential forward selection Palmer Penguins data set. gallery features collection case studies demos optimization. Utilize built-feature importance models Recursive Feature Elimination. Run feature selection Shadow Variable Search. Feature Selection Titanic data set.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":"analysis","dir":"Reference","previous_headings":"","what":"Analysis","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"analyzing feature selection results, recommended pass archive .data.table(). returned data table joined benchmark result adds mlr3::ResampleResult feature set. archive provides various getters (e.g. $learners()) ease access. getters extract position () unique hash (uhash). complete list getters see methods section. benchmark result ($benchmark_result) allows score feature sets different measure. Alternatively, measures can supplied .data.table().","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":"super-classes","dir":"Reference","previous_headings":"","what":"Super classes","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"bbotk::OptimInstance -> bbotk::OptimInstanceBatch -> bbotk::OptimInstanceBatchSingleCrit -> FSelectInstanceBatchSingleCrit","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":"active-bindings","dir":"Reference","previous_headings":"","what":"Active bindings","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"result_feature_set (character()) Feature set task subsetting.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"bbotk::OptimInstance$clear() bbotk::OptimInstance$format() bbotk::OptimInstanceBatch$eval_batch() bbotk::OptimInstanceBatch$objective_function()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":"public-methods","dir":"Reference","previous_headings":"","what":"Public methods","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"FSelectInstanceBatchSingleCrit$new() FSelectInstanceBatchSingleCrit$assign_result() FSelectInstanceBatchSingleCrit$print() FSelectInstanceBatchSingleCrit$clone()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":"method-new-","dir":"Reference","previous_headings":"","what":"Method new()","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"Creates new instance R6 class.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"","code":"FSelectInstanceBatchSingleCrit$new( task, learner, resampling, measure, terminator, store_benchmark_result = TRUE, store_models = FALSE, check_values = FALSE, callbacks = NULL, ties_method = \"least_features\" )"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"task (mlr3::Task) Task operate . learner (mlr3::Learner) Learner optimize feature subset . resampling (mlr3::Resampling) Resampling used evaluated performance feature subsets. Uninstantiated resamplings instantiated construction feature subsets evaluated data splits. Already instantiated resamplings kept unchanged. measure (mlr3::Measure) Measure optimize. NULL, default measure used. terminator (bbotk::Terminator) Stop criterion feature selection. store_benchmark_result (logical(1)) Store benchmark result archive? store_models (logical(1)). Store models benchmark result? check_values (logical(1)) Check parameters evaluation results validity? callbacks (list CallbackBatchFSelect) List callbacks. ties_method (character(1)) method break ties selecting sets optimizing selecting best set. Can \"least_features\" \"random\". option \"least_features\" (default) selects feature set least features. multiple best feature sets number features, one selected randomly. random method returns random feature set best feature sets. Ignored multiple measures used.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":"method-assign-result-","dir":"Reference","previous_headings":"","what":"Method assign_result()","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"FSelector writes best found feature subset estimated performance value . internal use.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":"usage-1","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"","code":"FSelectInstanceBatchSingleCrit$assign_result(xdt, y)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":"arguments-1","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"xdt (data.table::data.table()) x values data.table. row one point. Contains value search space FSelectInstanceBatchMultiCrit object. Can contain additional columns extra information. y (numeric(1)) Optimal outcome.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":"method-print-","dir":"Reference","previous_headings":"","what":"Method print()","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"Printer.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":"usage-2","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"","code":"FSelectInstanceBatchSingleCrit$print(...)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":"arguments-2","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"... (ignored).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":"method-clone-","dir":"Reference","previous_headings":"","what":"Method clone()","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"objects class cloneable method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":"usage-3","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"","code":"FSelectInstanceBatchSingleCrit$clone(deep = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":"arguments-3","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"deep Whether make deep clone.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"","code":"# Feature selection on Palmer Penguins data set # \\donttest{ task = tsk(\"penguins\") learner = lrn(\"classif.rpart\") # Construct feature selection instance instance = fsi( task = task, learner = learner, resampling = rsmp(\"cv\", folds = 3), measures = msr(\"classif.ce\"), terminator = trm(\"evals\", n_evals = 4) ) # Choose optimization algorithm fselector = fs(\"random_search\", batch_size = 2) # Run feature selection fselector$optimize(instance) #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> features n_features #> #> 1: bill_depth,bill_length,body_mass,flipper_length,island,sex,... 7 #> classif.ce #> #> 1: 0.04652937 # Subset task to optimal feature set task$select(instance$result_feature_set) # Train the learner with optimal feature set on the full data set learner$train(task) # Inspect all evaluated sets as.data.table(instance$archive) #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: FALSE TRUE TRUE FALSE FALSE FALSE FALSE #> 2: TRUE TRUE TRUE FALSE TRUE TRUE TRUE #> 3: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> 4: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> classif.ce runtime_learners timestamp batch_nr warnings errors #> #> 1: 0.11929825 0.013 2024-07-24 11:01:21 1 0 0 #> 2: 0.06117468 0.015 2024-07-24 11:01:21 1 0 0 #> 3: 0.04652937 0.015 2024-07-24 11:01:21 2 0 0 #> 4: 0.04652937 0.033 2024-07-24 11:01:21 2 0 0 #> features n_features #> #> 1: bill_length,body_mass 2 #> 2: bill_depth,bill_length,body_mass,island,sex,year 6 #> 3: bill_depth,bill_length,body_mass,flipper_length,island,sex,... 7 #> 4: bill_depth,bill_length,body_mass,flipper_length,island,sex,... 7 #> resample_result #> #> 1: #> 2: #> 3: #> 4: # }"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":null,"dir":"Reference","previous_headings":"","what":"FSelector — FSelector","title":"FSelector — FSelector","text":"`FSelector“ implements optimization algorithm.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"FSelector — FSelector","text":"FSelector abstract base class implements base functionality fselector must provide.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":"resources","dir":"Reference","previous_headings":"","what":"Resources","title":"FSelector — FSelector","text":"several sections feature selection mlr3book. Learn fselectors. gallery features collection case studies demos optimization. Utilize built-feature importance models Recursive Feature Elimination. Run feature selection Shadow Variable Search.","code":""},{"path":[]},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":"public-fields","dir":"Reference","previous_headings":"","what":"Public fields","title":"FSelector — FSelector","text":"id (character(1)) Identifier object. Used tables, plot text output.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":"active-bindings","dir":"Reference","previous_headings":"","what":"Active bindings","title":"FSelector — FSelector","text":"param_set paradox::ParamSet Set control parameters. properties (character()) Set properties fselector. Must subset mlr_reflections$fselect_properties. packages (character()) Set required packages. Note packages loaded via requireNamespace(), attached. label (character(1)) Label object. Can used tables, plot text output instead ID. man (character(1)) String format [pkg]::[topic] pointing manual page object. referenced help package can opened via method $help().","code":""},{"path":[]},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":"public-methods","dir":"Reference","previous_headings":"","what":"Public methods","title":"FSelector — FSelector","text":"FSelector$new() FSelector$format() FSelector$print() FSelector$help() FSelector$clone()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":"method-new-","dir":"Reference","previous_headings":"","what":"Method new()","title":"FSelector — FSelector","text":"Creates new instance R6 class.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"FSelector — FSelector","text":"","code":"FSelector$new( id = \"fselector\", param_set, properties, packages = character(), label = NA_character_, man = NA_character_ )"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"FSelector — FSelector","text":"id (character(1)) Identifier new instance. param_set paradox::ParamSet Set control parameters. properties (character()) Set properties fselector. Must subset mlr_reflections$fselect_properties. packages (character()) Set required packages. Note packages loaded via requireNamespace(), attached. label (character(1)) Label object. Can used tables, plot text output instead ID. man (character(1)) String format [pkg]::[topic] pointing manual page object. referenced help package can opened via method $help().","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":"method-format-","dir":"Reference","previous_headings":"","what":"Method format()","title":"FSelector — FSelector","text":"Helper print outputs.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":"usage-1","dir":"Reference","previous_headings":"","what":"Usage","title":"FSelector — FSelector","text":"","code":"FSelector$format(...)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":"arguments-1","dir":"Reference","previous_headings":"","what":"Arguments","title":"FSelector — FSelector","text":"... (ignored).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":"returns","dir":"Reference","previous_headings":"","what":"Returns","title":"FSelector — FSelector","text":"(character()).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":"method-print-","dir":"Reference","previous_headings":"","what":"Method print()","title":"FSelector — FSelector","text":"Print method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":"usage-2","dir":"Reference","previous_headings":"","what":"Usage","title":"FSelector — FSelector","text":"","code":"FSelector$print()"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":"returns-1","dir":"Reference","previous_headings":"","what":"Returns","title":"FSelector — FSelector","text":"(character()).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":"method-help-","dir":"Reference","previous_headings":"","what":"Method help()","title":"FSelector — FSelector","text":"Opens corresponding help page referenced field $man.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":"usage-3","dir":"Reference","previous_headings":"","what":"Usage","title":"FSelector — FSelector","text":"","code":"FSelector$help()"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":"method-clone-","dir":"Reference","previous_headings":"","what":"Method clone()","title":"FSelector — FSelector","text":"objects class cloneable method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":"usage-4","dir":"Reference","previous_headings":"","what":"Usage","title":"FSelector — FSelector","text":"","code":"FSelector$clone(deep = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":"arguments-2","dir":"Reference","previous_headings":"","what":"Arguments","title":"FSelector — FSelector","text":"deep Whether make deep clone.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatch.html","id":null,"dir":"Reference","previous_headings":"","what":"Class for Batch Feature Selection Algorithms — FSelectorBatch","title":"Class for Batch Feature Selection Algorithms — FSelectorBatch","text":"FSelectorBatch implements optimization algorithm.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatch.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Class for Batch Feature Selection Algorithms — FSelectorBatch","text":"FSelectorBatch abstract base class implements base functionality fselector must provide. subclass implemented following way: Inherit FSelectorBatch. Specify private abstract method $.optimize() use call optimizer. need call instance$eval_batch() evaluate design points. batch evaluation requested FSelectInstanceBatchSingleCrit/FSelectInstanceBatchMultiCrit object instance, batch possibly executed parallel via mlr3::benchmark(), evaluations stored inside instance$archive. batch evaluation, bbotk::Terminator checked, positive, exception class \"terminated_error\" generated. latter case current batch evaluations still stored instance, numeric scores sent back handling optimizer lost execution control. exception caught select best set instance$archive return . Note therefore points specified bbotk::Terminator may evaluated, Terminator checked batch evaluation, -evaluation batch. many depends setting batch size. Overwrite private super-method .assign_result() want decide estimate final set instance estimated performance. default behavior : pick best resample experiment, regarding given measure, assign set aggregated performance instance.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatch.html","id":"private-methods","dir":"Reference","previous_headings":"","what":"Private Methods","title":"Class for Batch Feature Selection Algorithms — FSelectorBatch","text":".optimize(instance) -> NULL Abstract base method. Implement specify feature selection subclass. See technical details sections. .assign_result(instance) -> NULL Abstract base method. Implement specify final feature subset selected. See technical details sections.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatch.html","id":"resources","dir":"Reference","previous_headings":"","what":"Resources","title":"Class for Batch Feature Selection Algorithms — FSelectorBatch","text":"several sections feature selection mlr3book. Learn fselectors. gallery features collection case studies demos optimization. Utilize built-feature importance models Recursive Feature Elimination. Run feature selection Shadow Variable Search.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatch.html","id":"super-class","dir":"Reference","previous_headings":"","what":"Super class","title":"Class for Batch Feature Selection Algorithms — FSelectorBatch","text":"mlr3fselect::FSelector -> FSelectorBatch","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatch.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Class for Batch Feature Selection Algorithms — FSelectorBatch","text":"mlr3fselect::FSelector$format() mlr3fselect::FSelector$help() mlr3fselect::FSelector$print()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatch.html","id":"public-methods","dir":"Reference","previous_headings":"","what":"Public methods","title":"Class for Batch Feature Selection Algorithms — FSelectorBatch","text":"FSelectorBatch$new() FSelectorBatch$optimize() FSelectorBatch$clone()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatch.html","id":"method-new-","dir":"Reference","previous_headings":"","what":"Method new()","title":"Class for Batch Feature Selection Algorithms — FSelectorBatch","text":"Creates new instance R6 class.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatch.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Batch Feature Selection Algorithms — FSelectorBatch","text":"","code":"FSelectorBatch$new( id = \"fselector_batch\", param_set, properties, packages = character(), label = NA_character_, man = NA_character_ )"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatch.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Batch Feature Selection Algorithms — FSelectorBatch","text":"id (character(1)) Identifier new instance. param_set paradox::ParamSet Set control parameters. properties (character()) Set properties fselector. Must subset mlr_reflections$fselect_properties. packages (character()) Set required packages. Note packages loaded via requireNamespace(), attached. label (character(1)) Label object. Can used tables, plot text output instead ID. man (character(1)) String format [pkg]::[topic] pointing manual page object. referenced help package can opened via method $help().","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatch.html","id":"method-optimize-","dir":"Reference","previous_headings":"","what":"Method optimize()","title":"Class for Batch Feature Selection Algorithms — FSelectorBatch","text":"Performs feature selection FSelectInstanceBatchSingleCrit FSelectInstanceBatchMultiCrit termination. single evaluations written ArchiveBatchFSelect resides FSelectInstanceBatchSingleCrit / FSelectInstanceBatchMultiCrit. result written instance object.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatch.html","id":"usage-1","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Batch Feature Selection Algorithms — FSelectorBatch","text":"","code":"FSelectorBatch$optimize(inst)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatch.html","id":"arguments-1","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Batch Feature Selection Algorithms — FSelectorBatch","text":"inst (FSelectInstanceBatchSingleCrit | FSelectInstanceBatchMultiCrit).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatch.html","id":"returns","dir":"Reference","previous_headings":"","what":"Returns","title":"Class for Batch Feature Selection Algorithms — FSelectorBatch","text":"data.table::data.table().","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatch.html","id":"method-clone-","dir":"Reference","previous_headings":"","what":"Method clone()","title":"Class for Batch Feature Selection Algorithms — FSelectorBatch","text":"objects class cloneable method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatch.html","id":"usage-2","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Batch Feature Selection Algorithms — FSelectorBatch","text":"","code":"FSelectorBatch$clone(deep = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatch.html","id":"arguments-2","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Batch Feature Selection Algorithms — FSelectorBatch","text":"deep Whether make deep clone.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatchFromOptimizerBatch.html","id":null,"dir":"Reference","previous_headings":"","what":"FSelectorBatchFromOptimizerBatch — FSelectorBatchFromOptimizerBatch","title":"FSelectorBatchFromOptimizerBatch — FSelectorBatchFromOptimizerBatch","text":"Internally used transform bbotk::Optimizer FSelector.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatchFromOptimizerBatch.html","id":"super-classes","dir":"Reference","previous_headings":"","what":"Super classes","title":"FSelectorBatchFromOptimizerBatch — FSelectorBatchFromOptimizerBatch","text":"mlr3fselect::FSelector -> mlr3fselect::FSelectorBatch -> FSelectorBatchFromOptimizerBatch","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatchFromOptimizerBatch.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"FSelectorBatchFromOptimizerBatch — FSelectorBatchFromOptimizerBatch","text":"mlr3fselect::FSelector$format() mlr3fselect::FSelector$help() mlr3fselect::FSelector$print()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatchFromOptimizerBatch.html","id":"public-methods","dir":"Reference","previous_headings":"","what":"Public methods","title":"FSelectorBatchFromOptimizerBatch — FSelectorBatchFromOptimizerBatch","text":"FSelectorBatchFromOptimizerBatch$new() FSelectorBatchFromOptimizerBatch$optimize() FSelectorBatchFromOptimizerBatch$clone()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatchFromOptimizerBatch.html","id":"method-new-","dir":"Reference","previous_headings":"","what":"Method new()","title":"FSelectorBatchFromOptimizerBatch — FSelectorBatchFromOptimizerBatch","text":"Creates new instance R6 class.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatchFromOptimizerBatch.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"FSelectorBatchFromOptimizerBatch — FSelectorBatchFromOptimizerBatch","text":"","code":"FSelectorBatchFromOptimizerBatch$new(optimizer, man = NA_character_)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatchFromOptimizerBatch.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"FSelectorBatchFromOptimizerBatch — FSelectorBatchFromOptimizerBatch","text":"optimizer bbotk::Optimizer Optimizer called. man (character(1)) String format [pkg]::[topic] pointing manual page object. referenced help package can opened via method $help().","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatchFromOptimizerBatch.html","id":"method-optimize-","dir":"Reference","previous_headings":"","what":"Method optimize()","title":"FSelectorBatchFromOptimizerBatch — FSelectorBatchFromOptimizerBatch","text":"Performs feature selection FSelectInstanceBatchSingleCrit / FSelectInstanceBatchMultiCrit termination.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatchFromOptimizerBatch.html","id":"usage-1","dir":"Reference","previous_headings":"","what":"Usage","title":"FSelectorBatchFromOptimizerBatch — FSelectorBatchFromOptimizerBatch","text":"","code":"FSelectorBatchFromOptimizerBatch$optimize(inst)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatchFromOptimizerBatch.html","id":"arguments-1","dir":"Reference","previous_headings":"","what":"Arguments","title":"FSelectorBatchFromOptimizerBatch — FSelectorBatchFromOptimizerBatch","text":"inst (FSelectInstanceBatchSingleCrit | FSelectInstanceBatchMultiCrit).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatchFromOptimizerBatch.html","id":"returns","dir":"Reference","previous_headings":"","what":"Returns","title":"FSelectorBatchFromOptimizerBatch — FSelectorBatchFromOptimizerBatch","text":"data.table::data.table.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatchFromOptimizerBatch.html","id":"method-clone-","dir":"Reference","previous_headings":"","what":"Method clone()","title":"FSelectorBatchFromOptimizerBatch — FSelectorBatchFromOptimizerBatch","text":"objects class cloneable method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatchFromOptimizerBatch.html","id":"usage-2","dir":"Reference","previous_headings":"","what":"Usage","title":"FSelectorBatchFromOptimizerBatch — FSelectorBatchFromOptimizerBatch","text":"","code":"FSelectorBatchFromOptimizerBatch$clone(deep = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatchFromOptimizerBatch.html","id":"arguments-2","dir":"Reference","previous_headings":"","what":"Arguments","title":"FSelectorBatchFromOptimizerBatch — FSelectorBatchFromOptimizerBatch","text":"deep Whether make deep clone.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelect.html","id":null,"dir":"Reference","previous_headings":"","what":"Class for Feature Selection Objective — ObjectiveFSelect","title":"Class for Feature Selection Objective — ObjectiveFSelect","text":"Stores objective function estimates performance feature subsets. class usually constructed internally FSelectInstanceBatchSingleCrit / FSelectInstanceBatchMultiCrit.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelect.html","id":"super-class","dir":"Reference","previous_headings":"","what":"Super class","title":"Class for Feature Selection Objective — ObjectiveFSelect","text":"bbotk::Objective -> ObjectiveFSelect","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelect.html","id":"public-fields","dir":"Reference","previous_headings":"","what":"Public fields","title":"Class for Feature Selection Objective — ObjectiveFSelect","text":"task (mlr3::Task). learner (mlr3::Learner). resampling (mlr3::Resampling). measures (list mlr3::Measure). store_models (logical(1)). store_benchmark_result (logical(1)). callbacks (List CallbackBatchFSelects).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelect.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Class for Feature Selection Objective — ObjectiveFSelect","text":"bbotk::Objective$eval() bbotk::Objective$eval_dt() bbotk::Objective$eval_many() bbotk::Objective$format() bbotk::Objective$help() bbotk::Objective$print()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelect.html","id":"public-methods","dir":"Reference","previous_headings":"","what":"Public methods","title":"Class for Feature Selection Objective — ObjectiveFSelect","text":"ObjectiveFSelect$new() ObjectiveFSelect$clone()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelect.html","id":"method-new-","dir":"Reference","previous_headings":"","what":"Method new()","title":"Class for Feature Selection Objective — ObjectiveFSelect","text":"Creates new instance R6 class.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelect.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Feature Selection Objective — ObjectiveFSelect","text":"","code":"ObjectiveFSelect$new( task, learner, resampling, measures, check_values = TRUE, store_benchmark_result = TRUE, store_models = FALSE, callbacks = NULL )"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelect.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Feature Selection Objective — ObjectiveFSelect","text":"task (mlr3::Task) Task operate . learner (mlr3::Learner) Learner optimize feature subset . resampling (mlr3::Resampling) Resampling used evaluated performance feature subsets. Uninstantiated resamplings instantiated construction feature subsets evaluated data splits. Already instantiated resamplings kept unchanged. measures (list mlr3::Measure) Measures optimize. NULL, mlr3's default measure used. check_values (logical(1)) Check parameters evaluation results validity? store_benchmark_result (logical(1)) Store benchmark result archive? store_models (logical(1)). Store models benchmark result? callbacks (list CallbackBatchFSelect) List callbacks.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelect.html","id":"method-clone-","dir":"Reference","previous_headings":"","what":"Method clone()","title":"Class for Feature Selection Objective — ObjectiveFSelect","text":"objects class cloneable method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelect.html","id":"usage-1","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Feature Selection Objective — ObjectiveFSelect","text":"","code":"ObjectiveFSelect$clone(deep = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelect.html","id":"arguments-1","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Feature Selection Objective — ObjectiveFSelect","text":"deep Whether make deep clone.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelectBatch.html","id":null,"dir":"Reference","previous_headings":"","what":"Class for Feature Selection Objective — ObjectiveFSelectBatch","title":"Class for Feature Selection Objective — ObjectiveFSelectBatch","text":"Stores objective function estimates performance feature subsets. class usually constructed internally FSelectInstanceBatchSingleCrit / FSelectInstanceBatchMultiCrit.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelectBatch.html","id":"super-classes","dir":"Reference","previous_headings":"","what":"Super classes","title":"Class for Feature Selection Objective — ObjectiveFSelectBatch","text":"bbotk::Objective -> mlr3fselect::ObjectiveFSelect -> ObjectiveFSelectBatch","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelectBatch.html","id":"public-fields","dir":"Reference","previous_headings":"","what":"Public fields","title":"Class for Feature Selection Objective — ObjectiveFSelectBatch","text":"archive (ArchiveBatchFSelect).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelectBatch.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Class for Feature Selection Objective — ObjectiveFSelectBatch","text":"bbotk::Objective$eval() bbotk::Objective$eval_dt() bbotk::Objective$eval_many() bbotk::Objective$format() bbotk::Objective$help() bbotk::Objective$print()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelectBatch.html","id":"public-methods","dir":"Reference","previous_headings":"","what":"Public methods","title":"Class for Feature Selection Objective — ObjectiveFSelectBatch","text":"ObjectiveFSelectBatch$new() ObjectiveFSelectBatch$clone()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelectBatch.html","id":"method-new-","dir":"Reference","previous_headings":"","what":"Method new()","title":"Class for Feature Selection Objective — ObjectiveFSelectBatch","text":"Creates new instance R6 class.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelectBatch.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Feature Selection Objective — ObjectiveFSelectBatch","text":"","code":"ObjectiveFSelectBatch$new( task, learner, resampling, measures, check_values = TRUE, store_benchmark_result = TRUE, store_models = FALSE, archive = NULL, callbacks = NULL )"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelectBatch.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Feature Selection Objective — ObjectiveFSelectBatch","text":"task (mlr3::Task) Task operate . learner (mlr3::Learner) Learner optimize feature subset . resampling (mlr3::Resampling) Resampling used evaluated performance feature subsets. Uninstantiated resamplings instantiated construction feature subsets evaluated data splits. Already instantiated resamplings kept unchanged. measures (list mlr3::Measure) Measures optimize. NULL, mlr3's default measure used. check_values (logical(1)) Check parameters evaluation results validity? store_benchmark_result (logical(1)) Store benchmark result archive? store_models (logical(1)). Store models benchmark result? archive (ArchiveBatchFSelect) Reference archive FSelectInstanceBatchSingleCrit | FSelectInstanceBatchMultiCrit. NULL (default), benchmark result models stored. callbacks (list CallbackBatchFSelect) List callbacks.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelectBatch.html","id":"method-clone-","dir":"Reference","previous_headings":"","what":"Method clone()","title":"Class for Feature Selection Objective — ObjectiveFSelectBatch","text":"objects class cloneable method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelectBatch.html","id":"usage-1","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Feature Selection Objective — ObjectiveFSelectBatch","text":"","code":"ObjectiveFSelectBatch$clone(deep = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelectBatch.html","id":"arguments-1","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Feature Selection Objective — ObjectiveFSelectBatch","text":"deep Whether make deep clone.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/auto_fselector.html","id":null,"dir":"Reference","previous_headings":"","what":"Function for Automatic Feature Selection — auto_fselector","title":"Function for Automatic Feature Selection — auto_fselector","text":"AutoFSelector wraps mlr3::Learner augments automatic feature selection. auto_fselector() function creates AutoFSelector object.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/auto_fselector.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Function for Automatic Feature Selection — auto_fselector","text":"","code":"auto_fselector( fselector, learner, resampling, measure = NULL, term_evals = NULL, term_time = NULL, terminator = NULL, store_fselect_instance = TRUE, store_benchmark_result = TRUE, store_models = FALSE, check_values = FALSE, callbacks = NULL, ties_method = \"least_features\" )"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/auto_fselector.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Function for Automatic Feature Selection — auto_fselector","text":"fselector (FSelector) Optimization algorithm. learner (mlr3::Learner) Learner optimize feature subset . resampling (mlr3::Resampling) Resampling used evaluated performance feature subsets. Uninstantiated resamplings instantiated construction feature subsets evaluated data splits. Already instantiated resamplings kept unchanged. measure (mlr3::Measure) Measure optimize. NULL, default measure used. term_evals (integer(1)) Number allowed evaluations. Ignored terminator passed. term_time (integer(1)) Maximum allowed time seconds. Ignored terminator passed. terminator (bbotk::Terminator) Stop criterion feature selection. store_fselect_instance (logical(1)) TRUE (default), stores internally created FSelectInstanceBatchSingleCrit intermediate results slot $fselect_instance. set TRUE, store_models = TRUE store_benchmark_result (logical(1)) Store benchmark result archive? store_models (logical(1)). Store models benchmark result? check_values (logical(1)) Check parameters evaluation results validity? callbacks (list CallbackBatchFSelect) List callbacks. ties_method (character(1)) method break ties selecting sets optimizing selecting best set. Can \"least_features\" \"random\". option \"least_features\" (default) selects feature set least features. multiple best feature sets number features, one selected randomly. random method returns random feature set best feature sets. Ignored multiple measures used.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/auto_fselector.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Function for Automatic Feature Selection — auto_fselector","text":"AutoFSelector.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/auto_fselector.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Function for Automatic Feature Selection — auto_fselector","text":"AutoFSelector mlr3::Learner wraps another mlr3::Learner performs following steps $train(): wrapped (inner) learner trained feature subsets via resampling. feature selection can specified providing FSelector, bbotk::Terminator, mlr3::Resampling mlr3::Measure. final model fit complete training data best-found feature subset. $predict() AutoFSelector just calls predict method wrapped (inner) learner.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/auto_fselector.html","id":"resources","dir":"Reference","previous_headings":"","what":"Resources","title":"Function for Automatic Feature Selection — auto_fselector","text":"several sections feature selection mlr3book. Estimate Model Performance nested resampling. gallery features collection case studies demos optimization.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/auto_fselector.html","id":"nested-resampling","dir":"Reference","previous_headings":"","what":"Nested Resampling","title":"Function for Automatic Feature Selection — auto_fselector","text":"Nested resampling can performed passing AutoFSelector object mlr3::resample() mlr3::benchmark(). access inner resampling results, set store_fselect_instance = TRUE execute mlr3::resample() mlr3::benchmark() store_models = TRUE (see examples). mlr3::Resampling passed AutoFSelector meant inner resampling, operating training set arbitrary outer resampling. reason feasible pass instantiated mlr3::Resampling .","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/auto_fselector.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Function for Automatic Feature Selection — auto_fselector","text":"","code":"# Automatic Feature Selection # \\donttest{ # split to train and external set task = tsk(\"penguins\") split = partition(task, ratio = 0.8) # create auto fselector afs = auto_fselector( fselector = fs(\"random_search\"), learner = lrn(\"classif.rpart\"), resampling = rsmp (\"holdout\"), measure = msr(\"classif.ce\"), term_evals = 4) # optimize feature subset and fit final model afs$train(task, row_ids = split$train) # predict with final model afs$predict(task, row_ids = split$test) #> for 69 observations: #> row_ids truth response #> 5 Adelie Adelie #> 11 Adelie Adelie #> 12 Adelie Adelie #> --- #> 338 Chinstrap Chinstrap #> 339 Chinstrap Chinstrap #> 340 Chinstrap Chinstrap # show result afs$fselect_result #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE TRUE TRUE TRUE FALSE TRUE FALSE #> features n_features classif.ce #> #> 1: bill_depth,bill_length,body_mass,flipper_length,sex 5 0.06521739 # model slot contains trained learner and fselect instance afs$model #> $learner #> : Classification Tree #> * Model: rpart #> * Parameters: xval=0 #> * Packages: mlr3, rpart #> * Predict Types: [response], prob #> * Feature Types: logical, integer, numeric, factor, ordered #> * Properties: importance, missings, multiclass, selected_features, #> twoclass, weights #> #> $features #> [1] \"bill_depth\" \"bill_length\" \"body_mass\" \"flipper_length\" #> [5] \"sex\" #> #> $fselect_instance #> #> * State: Optimized #> * Objective: #> * Terminator: #> * Result: #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE TRUE TRUE TRUE FALSE TRUE FALSE #> classif.ce #> #> 1: 0.06521739 #> * Archive: #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE FALSE FALSE TRUE TRUE TRUE TRUE #> 2: TRUE TRUE TRUE FALSE TRUE TRUE TRUE #> 3: FALSE TRUE TRUE FALSE FALSE FALSE FALSE #> 4: TRUE TRUE TRUE FALSE TRUE TRUE TRUE #> 5: TRUE TRUE TRUE TRUE FALSE TRUE TRUE #> 6: TRUE TRUE TRUE TRUE FALSE TRUE FALSE #> 7: FALSE FALSE FALSE FALSE TRUE FALSE FALSE #> 8: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> 9: FALSE FALSE FALSE TRUE TRUE FALSE TRUE #> 10: TRUE TRUE TRUE FALSE TRUE TRUE TRUE #> classif.ce #> #> 1: 0.16304348 #> 2: 0.07608696 #> 3: 0.13043478 #> 4: 0.07608696 #> 5: 0.06521739 #> 6: 0.06521739 #> 7: 0.31521739 #> 8: 0.06521739 #> 9: 0.16304348 #> 10: 0.07608696 #> # shortcut trained learner afs$learner #> : Classification Tree #> * Model: rpart #> * Parameters: xval=0 #> * Packages: mlr3, rpart #> * Predict Types: [response], prob #> * Feature Types: logical, integer, numeric, factor, ordered #> * Properties: importance, missings, multiclass, selected_features, #> twoclass, weights # shortcut fselect instance afs$fselect_instance #> #> * State: Optimized #> * Objective: #> * Terminator: #> * Result: #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE TRUE TRUE TRUE FALSE TRUE FALSE #> classif.ce #> #> 1: 0.06521739 #> * Archive: #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE FALSE FALSE TRUE TRUE TRUE TRUE #> 2: TRUE TRUE TRUE FALSE TRUE TRUE TRUE #> 3: FALSE TRUE TRUE FALSE FALSE FALSE FALSE #> 4: TRUE TRUE TRUE FALSE TRUE TRUE TRUE #> 5: TRUE TRUE TRUE TRUE FALSE TRUE TRUE #> 6: TRUE TRUE TRUE TRUE FALSE TRUE FALSE #> 7: FALSE FALSE FALSE FALSE TRUE FALSE FALSE #> 8: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> 9: FALSE FALSE FALSE TRUE TRUE FALSE TRUE #> 10: TRUE TRUE TRUE FALSE TRUE TRUE TRUE #> classif.ce #> #> 1: 0.16304348 #> 2: 0.07608696 #> 3: 0.13043478 #> 4: 0.07608696 #> 5: 0.06521739 #> 6: 0.06521739 #> 7: 0.31521739 #> 8: 0.06521739 #> 9: 0.16304348 #> 10: 0.07608696 # Nested Resampling afs = auto_fselector( fselector = fs(\"random_search\"), learner = lrn(\"classif.rpart\"), resampling = rsmp (\"holdout\"), measure = msr(\"classif.ce\"), term_evals = 4) resampling_outer = rsmp(\"cv\", folds = 3) rr = resample(task, afs, resampling_outer, store_models = TRUE) # retrieve inner feature selection results. extract_inner_fselect_results(rr) #> iteration bill_depth bill_length body_mass flipper_length island sex #> #> 1: 1 TRUE TRUE TRUE TRUE TRUE FALSE #> 2: 2 TRUE TRUE TRUE FALSE TRUE FALSE #> 3: 3 TRUE TRUE TRUE TRUE TRUE TRUE #> year classif.ce #> #> 1: FALSE 0.06578947 #> 2: TRUE 0.03947368 #> 3: TRUE 0.06493506 #> features n_features #> #> 1: bill_depth,bill_length,body_mass,flipper_length,island 5 #> 2: bill_depth,bill_length,body_mass,island,year 5 #> 3: bill_depth,bill_length,body_mass,flipper_length,island,sex,... 7 #> task_id learner_id resampling_id #> #> 1: penguins classif.rpart.fselector cv #> 2: penguins classif.rpart.fselector cv #> 3: penguins classif.rpart.fselector cv # performance scores estimated on the outer resampling rr$score() #> task_id learner_id resampling_id iteration classif.ce #> #> 1: penguins classif.rpart.fselector cv 1 0.05217391 #> 2: penguins classif.rpart.fselector cv 2 0.05217391 #> 3: penguins classif.rpart.fselector cv 3 0.07017544 #> Hidden columns: task, learner, resampling, prediction # unbiased performance of the final model trained on the full data set rr$aggregate() #> classif.ce #> 0.05817442 # }"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/callback_batch_fselect.html","id":null,"dir":"Reference","previous_headings":"","what":"Create Feature Selection Callback — callback_batch_fselect","title":"Create Feature Selection Callback — callback_batch_fselect","text":"Function create CallbackBatchFSelect. Predefined callbacks stored dictionary mlr_callbacks can retrieved clbk(). Feature selection callbacks can called different stages feature selection. stages prefixed on_*. See also section parameters information stages. feature selection callback works bbotk::ContextBatch ContextBatchFSelect.","code":"Start Feature Selection - on_optimization_begin Start FSelect Batch - on_optimizer_before_eval Start Evaluation - on_eval_after_design - on_eval_after_benchmark - on_eval_before_archive End Evaluation - on_optimizer_after_eval End FSelect Batch - on_result - on_optimization_end End Feature Selection"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/callback_batch_fselect.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Create Feature Selection Callback — callback_batch_fselect","text":"","code":"callback_batch_fselect( id, label = NA_character_, man = NA_character_, on_optimization_begin = NULL, on_optimizer_before_eval = NULL, on_eval_after_design = NULL, on_eval_after_benchmark = NULL, on_eval_before_archive = NULL, on_optimizer_after_eval = NULL, on_result = NULL, on_optimization_end = NULL )"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/callback_batch_fselect.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Create Feature Selection Callback — callback_batch_fselect","text":"id (character(1)) Identifier new instance. label (character(1)) Label new instance. man (character(1)) String format [pkg]::[topic] pointing manual page object. referenced help package can opened via method $help(). on_optimization_begin (function()) Stage called beginning optimization. Called Optimizer$optimize(). on_optimizer_before_eval (function()) Stage called optimizer proposes points. Called OptimInstance$eval_batch(). on_eval_after_design (function()) Stage called design created. Called ObjectiveFSelectBatch$eval_many(). on_eval_after_benchmark (function()) Stage called feature sets evaluated. Called ObjectiveFSelectBatch$eval_many(). on_eval_before_archive (function()) Stage called performance values written archive. Called ObjectiveFSelectBatch$eval_many(). on_optimizer_after_eval (function()) Stage called points evaluated. Called OptimInstance$eval_batch(). on_result (function()) Stage called result written. Called OptimInstance$assign_result(). on_optimization_end (function()) Stage called end optimization. Called Optimizer$optimize().","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/callback_batch_fselect.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Create Feature Selection Callback — callback_batch_fselect","text":"implementing callback, function must two arguments named callback context. callback can write data state ($state), e.g. settings affect callback . Avoid writing large data state. can slow feature selection evaluation configurations parallelized. Feature selection callbacks access two different contexts depending stage. stages on_eval_after_design, on_eval_after_benchmark, on_eval_before_archive access ContextBatchFSelect. context can used customize evaluation batch feature sets. Changes state callback lost evaluation batch changes fselect instance fselector possible. Persistent data written archive via $aggregated_performance (see ContextBatchFSelect). stages access bbotk::ContextBatch. context can used modify fselect instance, archive, fselector final result. two different contexts evaluation can parallelized .e. multiple instances ContextBatchFSelect exists different workers time.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/callback_batch_fselect.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Create Feature Selection Callback — callback_batch_fselect","text":"","code":"# Write archive to disk callback_batch_fselect(\"mlr3fselect.backup\", on_optimization_end = function(callback, context) { saveRDS(context$instance$archive, \"archive.rds\") } ) #> #> * Active Stages: on_optimization_end"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":null,"dir":"Reference","previous_headings":"","what":"Ensemble Feature Selection Result — ensemble_fs_result","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"EnsembleFSResult stores results ensemble feature selection. includes methods evaluating stability feature selection process ranking selected features among others. function ensemble_fselect() returns object class.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"s-methods","dir":"Reference","previous_headings":"","what":"S3 Methods","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":".data.table.EnsembleFSResult(x, benchmark_result = TRUE) Returns tabular view ensemble feature selection. EnsembleFSResult -> data.table::data.table() x (EnsembleFSResult) benchmark_result (logical(1)) Whether add learner, task resampling information benchmark result.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"Das, (1999). “characterizing 'knee' Pareto curve based normal-boundary intersection.” Structural Optimization, 18(1-2), 107–115. ISSN 09344373.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"public-fields","dir":"Reference","previous_headings":"","what":"Public fields","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"benchmark_result (mlr3::BenchmarkResult) benchmark result. man (character(1)) Manual page object.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"active-bindings","dir":"Reference","previous_headings":"","what":"Active bindings","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"result (data.table::data.table) Returns result ensemble feature selection. n_learners (numeric(1)) Returns number learners used ensemble feature selection. measure (character(1)) Returns measure id used ensemble feature selection.","code":""},{"path":[]},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"public-methods","dir":"Reference","previous_headings":"","what":"Public methods","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"EnsembleFSResult$new() EnsembleFSResult$format() EnsembleFSResult$print() EnsembleFSResult$help() EnsembleFSResult$feature_ranking() EnsembleFSResult$stability() EnsembleFSResult$pareto_front() EnsembleFSResult$knee_points() EnsembleFSResult$clone()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"method-new-","dir":"Reference","previous_headings":"","what":"Method new()","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"Creates new instance R6 class.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"","code":"EnsembleFSResult$new( result, features, benchmark_result = NULL, measure_id, minimize = TRUE )"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"result (data.table::data.table) result ensemble feature selection. Column names include \"resampling_iteration\", \"learner_id\", \"features\" \"n_features\". features (character()) vector features task used ensemble feature selection. benchmark_result (mlr3::BenchmarkResult) benchmark result object. measure_id (character(1)) Column name \"result\" corresponds measure used. minimize (logical(1)) TRUE (default), lower values measure correspond higher performance.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"method-format-","dir":"Reference","previous_headings":"","what":"Method format()","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"Helper print outputs.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"usage-1","dir":"Reference","previous_headings":"","what":"Usage","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"","code":"EnsembleFSResult$format(...)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"arguments-1","dir":"Reference","previous_headings":"","what":"Arguments","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"... (ignored).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"method-print-","dir":"Reference","previous_headings":"","what":"Method print()","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"Printer.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"usage-2","dir":"Reference","previous_headings":"","what":"Usage","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"","code":"EnsembleFSResult$print(...)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"arguments-2","dir":"Reference","previous_headings":"","what":"Arguments","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"... (ignored).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"method-help-","dir":"Reference","previous_headings":"","what":"Method help()","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"Opens corresponding help page referenced field $man.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"usage-3","dir":"Reference","previous_headings":"","what":"Usage","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"","code":"EnsembleFSResult$help()"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"method-feature-ranking-","dir":"Reference","previous_headings":"","what":"Method feature_ranking()","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"Calculates feature ranking.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"usage-4","dir":"Reference","previous_headings":"","what":"Usage","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"","code":"EnsembleFSResult$feature_ranking(method = \"approval_voting\")"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"arguments-3","dir":"Reference","previous_headings":"","what":"Arguments","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"method (character(1)) method calculate feature ranking.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"feature ranking process built following framework: models act voters, features act candidates, voters select certain candidates (features). primary objective compile selections consensus ranked list features, effectively forming committee. Currently, \"approval_voting\" method supported, selects candidates/features highest approval score selection frequency, .e. appear often.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"returns","dir":"Reference","previous_headings":"","what":"Returns","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"data.table::data.table listing features, ordered decreasing inclusion probability scores (depending method)","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"method-stability-","dir":"Reference","previous_headings":"","what":"Method stability()","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"Calculates stability selected features stabm package. results cached. stability measure requested different arguments, cache must reset.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"usage-5","dir":"Reference","previous_headings":"","what":"Usage","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"","code":"EnsembleFSResult$stability( stability_measure = \"jaccard\", stability_args = NULL, global = TRUE, reset_cache = FALSE )"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"arguments-4","dir":"Reference","previous_headings":"","what":"Arguments","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"stability_measure (character(1)) stability measure used. One measures returned stabm::listStabilityMeasures() lower case. Default \"jaccard\". stability_args (list) Additional arguments passed stability measure function. global (logical(1)) Whether calculate stability globally learner. reset_cache (logical(1)) TRUE, cached results ignored.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"returns-1","dir":"Reference","previous_headings":"","what":"Returns","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"numeric() value representing stability selected features. numeric() vector stability selected features learner.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"method-pareto-front-","dir":"Reference","previous_headings":"","what":"Method pareto_front()","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"function identifies Pareto front ensemble feature selection process, .e., set points represent trade-number features performance (e.g. classification error).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"usage-6","dir":"Reference","previous_headings":"","what":"Usage","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"","code":"EnsembleFSResult$pareto_front(type = \"empirical\")"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"arguments-5","dir":"Reference","previous_headings":"","what":"Arguments","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"type (character(1)) Specifies type Pareto front return. See details.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"details-1","dir":"Reference","previous_headings":"","what":"Details","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"Two options available Pareto front: \"empirical\" (default): returns empirical Pareto front. \"estimated\": Pareto front points estimated fitting linear model inversed number features (\\(1/x\\)) input associated performance scores output. method useful Pareto points sparse front assumes convex shape better performance corresponds lower measure values (e.g. classification error), concave shape otherwise (e.g. classification accuracy). estimated Pareto front include points number features ranging 1 maximum number found empirical Pareto front.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"returns-2","dir":"Reference","previous_headings":"","what":"Returns","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"data.table::data.table columns number features performance together form Pareto front.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"method-knee-points-","dir":"Reference","previous_headings":"","what":"Method knee_points()","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"function implements various knee point identification (KPI) methods, select points Pareto front, optimal trade-performance number features achieved. cases, one point returned.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"usage-7","dir":"Reference","previous_headings":"","what":"Usage","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"","code":"EnsembleFSResult$knee_points(method = \"NBI\", type = \"empirical\")"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"arguments-6","dir":"Reference","previous_headings":"","what":"Arguments","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"method (character(1)) Type method use identify knee point. See details. type (character(1)) Specifies type Pareto front use identification knee point. See pareto_front() method details.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"details-2","dir":"Reference","previous_headings":"","what":"Details","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"available KPI methods : \"NBI\" (default): Normal-Boundary Intersection method geometry-based method calculates perpendicular distance point line connecting first last points Pareto front. knee point determined Pareto point maximum distance line, see Das (1999).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"returns-3","dir":"Reference","previous_headings":"","what":"Returns","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"data.table::data.table knee point(s) Pareto front.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"method-clone-","dir":"Reference","previous_headings":"","what":"Method clone()","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"objects class cloneable method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"usage-8","dir":"Reference","previous_headings":"","what":"Usage","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"","code":"EnsembleFSResult$clone(deep = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"arguments-7","dir":"Reference","previous_headings":"","what":"Arguments","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"deep Whether make deep clone.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"","code":"# \\donttest{ efsr = ensemble_fselect( fselector = fs(\"rfe\", n_features = 2, feature_fraction = 0.8), task = tsk(\"sonar\"), learners = lrns(c(\"classif.rpart\", \"classif.featureless\")), init_resampling = rsmp(\"subsampling\", repeats = 2), inner_resampling = rsmp(\"cv\", folds = 3), measure = msr(\"classif.ce\"), terminator = trm(\"none\") ) # contains the benchmark result efsr$benchmark_result #> of 4 rows with 4 resampling runs #> nr task_id learner_id resampling_id iters warnings errors #> 1 sonar classif.rpart.fselector insample 1 0 0 #> 2 sonar classif.featureless.fselector insample 1 0 0 #> 3 sonar classif.rpart.fselector insample 1 0 0 #> 4 sonar classif.featureless.fselector insample 1 0 0 # contains the selected features for each iteration efsr$result #> resampling_iteration learner_id features #> #> 1: 1 classif.rpart V1,V10,V11,V12,V13,V14,... #> 2: 1 classif.featureless V14,V9 #> 3: 2 classif.rpart V10,V11,V12,V13,V15,V44,... #> 4: 2 classif.featureless V20,V22 #> n_features classif.ce #> #> 1: 60 0.2158495 #> 2: 2 0.4750231 #> 3: 12 0.2810669 #> 4: 2 0.5610546 #> importance #> #> 1: 60.00000,58.33333,58.00000,57.66667,55.66667,49.00000,... #> 2: 1.666667,1.333333 #> 3: 11.333333,11.333333, 9.666667, 8.666667, 8.000000, 6.666667,... #> 4: 1.666667,1.333333 #> task learner #> #> 1: #> 2: #> 3: #> 4: #> resampling #> #> 1: #> 2: #> 3: #> 4: # returns the stability of the selected features efsr$stability(stability_measure = \"jaccard\") #> [1] 0.05726496 # returns a ranking of all features head(efsr$feature_ranking()) #> feature inclusion_probability #> #> 1: V9 0.75 #> 2: V10 0.50 #> 3: V11 0.50 #> 4: V12 0.50 #> 5: V13 0.50 #> 6: V14 0.50 # returns the empirical pareto front (nfeatures vs error) efsr$pareto_front() #> n_features classif.ce #> #> 1: 2 0.5610546 #> 2: 2 0.4750231 #> 3: 12 0.2810669 #> 4: 60 0.2158495 # }"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fselect.html","id":null,"dir":"Reference","previous_headings":"","what":"Ensemble Feature Selection — ensemble_fselect","title":"Ensemble Feature Selection — ensemble_fselect","text":"Ensemble feature selection using multiple learners. ensemble feature selection method designed identify informative features given dataset leveraging multiple machine learning models resampling techniques. Returns EnsembleFSResult.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fselect.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Ensemble Feature Selection — ensemble_fselect","text":"","code":"ensemble_fselect( fselector, task, learners, init_resampling, inner_resampling, measure, terminator, callbacks = NULL, store_benchmark_result = TRUE, store_models = TRUE )"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fselect.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Ensemble Feature Selection — ensemble_fselect","text":"Saeys, Yvan, Abeel, Thomas, Van De Peer, Yves (2008). “Robust feature selection using ensemble feature selection techniques.” Machine Learning Knowledge Discovery Databases, 5212 LNAI, 313–325. doi:10.1007/978-3-540-87481-2_21 . Abeel, Thomas, Helleputte, Thibault, Van de Peer, Yves, Dupont, Pierre, Saeys, Yvan (2010). “Robust biomarker identification cancer diagnosis ensemble feature selection methods.” Bioinformatics, 26, 392–398. ISSN 1367-4803, doi:10.1093/BIOINFORMATICS/BTP630 . Pes, Barbara (2020). “Ensemble feature selection high-dimensional data: stability analysis across multiple domains.” Neural Computing Applications, 32(10), 5951–5973. ISSN 14333058, doi:10.1007/s00521-019-04082-3 .","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fselect.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Ensemble Feature Selection — ensemble_fselect","text":"fselector (FSelector) Optimization algorithm. task (mlr3::Task) Task operate . learners (list mlr3::Learner) learners used feature selection. init_resampling (mlr3::Resampling) initial resampling strategy data, train set passed learners. Can mlr3::ResamplingSubsampling mlr3::ResamplingBootstrap. inner_resampling (mlr3::Resampling) inner resampling strategy used FSelector. measure (mlr3::Measure) Measure optimize. NULL, default measure used. terminator (bbotk::Terminator) Stop criterion feature selection. callbacks (list lists CallbackBatchFSelect) Callbacks used learner. lists must length number learners. store_benchmark_result (logical(1)) Whether store benchmark result EnsembleFSResult . store_models (logical(1)) Whether store models auto_fselector .","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fselect.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Ensemble Feature Selection — ensemble_fselect","text":"EnsembleFSResult object.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fselect.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Ensemble Feature Selection — ensemble_fselect","text":"method begins applying initial resampling technique specified user, create multiple subsamples original dataset. resampling process helps generating diverse subsets data robust feature selection. subsample generated previous step, method performs wrapped-based feature selection (auto_fselector) using provided learner, given inner resampling method, performance measure optimization algorithm. process generates best feature subset combination subsample learner. Results stored EnsembleFSResult.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fselect.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Ensemble Feature Selection — ensemble_fselect","text":"","code":"# \\donttest{ efsr = ensemble_fselect( fselector = fs(\"random_search\"), task = tsk(\"sonar\"), learners = lrns(c(\"classif.rpart\", \"classif.featureless\")), init_resampling = rsmp(\"subsampling\", repeats = 2), inner_resampling = rsmp(\"cv\", folds = 3), measure = msr(\"classif.ce\"), terminator = trm(\"evals\", n_evals = 10) ) efsr #> #> resampling_iteration learner_id n_features #> #> 1: 1 classif.rpart 23 #> 2: 1 classif.featureless 2 #> 3: 2 classif.rpart 34 #> 4: 2 classif.featureless 15 # }"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/extract_inner_fselect_archives.html","id":null,"dir":"Reference","previous_headings":"","what":"Extract Inner Feature Selection Archives — extract_inner_fselect_archives","title":"Extract Inner Feature Selection Archives — extract_inner_fselect_archives","text":"Extract inner feature selection archives nested resampling. Implemented mlr3::ResampleResult mlr3::BenchmarkResult. function iterates AutoFSelector objects binds archives data.table::data.table(). AutoFSelector must initialized store_fselect_instance = TRUE resample() benchmark() must called store_models = TRUE.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/extract_inner_fselect_archives.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Extract Inner Feature Selection Archives — extract_inner_fselect_archives","text":"","code":"extract_inner_fselect_archives(x, exclude_columns = \"uhash\")"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/extract_inner_fselect_archives.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Extract Inner Feature Selection Archives — extract_inner_fselect_archives","text":"x (mlr3::ResampleResult | mlr3::BenchmarkResult). exclude_columns (character()) Exclude columns result table. Set NULL column excluded.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/extract_inner_fselect_archives.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Extract Inner Feature Selection Archives — extract_inner_fselect_archives","text":"data.table::data.table().","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/extract_inner_fselect_archives.html","id":"data-structure","dir":"Reference","previous_headings":"","what":"Data structure","title":"Extract Inner Feature Selection Archives — extract_inner_fselect_archives","text":"returned data table following columns: experiment (integer(1)) Index, giving according row number original benchmark grid. iteration (integer(1)) Iteration outer resampling. One column feature task. One column performance measure. runtime_learners (numeric(1)) Sum training predict times logged learners per mlr3::ResampleResult / evaluation. include potential overhead time. timestamp (POSIXct) Time stamp evaluation logged archive. batch_nr (integer(1)) Feature sets evaluated batches. batch unique batch number. resample_result (mlr3::ResampleResult) Resample result inner resampling. task_id (character(1)). learner_id (character(1)). resampling_id (character(1)).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/extract_inner_fselect_archives.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Extract Inner Feature Selection Archives — extract_inner_fselect_archives","text":"","code":"# Nested Resampling on Palmer Penguins Data Set # create auto fselector at = auto_fselector( fselector = fs(\"random_search\"), learner = lrn(\"classif.rpart\"), resampling = rsmp (\"holdout\"), measure = msr(\"classif.ce\"), term_evals = 4) resampling_outer = rsmp(\"cv\", folds = 2) rr = resample(tsk(\"penguins\"), at, resampling_outer, store_models = TRUE) # extract inner archives extract_inner_fselect_archives(rr) #> iteration bill_depth bill_length body_mass flipper_length island sex #> #> 1: 1 TRUE TRUE FALSE TRUE TRUE TRUE #> 2: 1 FALSE TRUE FALSE FALSE FALSE FALSE #> 3: 1 TRUE TRUE TRUE FALSE TRUE TRUE #> 4: 1 TRUE TRUE TRUE TRUE FALSE TRUE #> 5: 1 FALSE FALSE FALSE FALSE TRUE FALSE #> 6: 1 TRUE TRUE FALSE TRUE TRUE TRUE #> 7: 1 FALSE TRUE FALSE FALSE TRUE FALSE #> 8: 1 TRUE FALSE TRUE TRUE TRUE FALSE #> 9: 1 TRUE FALSE FALSE TRUE FALSE TRUE #> 10: 1 FALSE FALSE TRUE TRUE FALSE FALSE #> 11: 2 FALSE FALSE FALSE FALSE FALSE FALSE #> 12: 2 FALSE FALSE FALSE TRUE FALSE FALSE #> 13: 2 FALSE FALSE TRUE FALSE FALSE FALSE #> 14: 2 TRUE TRUE TRUE TRUE TRUE TRUE #> 15: 2 TRUE FALSE TRUE TRUE TRUE FALSE #> 16: 2 TRUE TRUE FALSE TRUE TRUE TRUE #> 17: 2 FALSE TRUE FALSE FALSE FALSE FALSE #> 18: 2 TRUE TRUE TRUE FALSE TRUE TRUE #> 19: 2 FALSE FALSE FALSE TRUE TRUE FALSE #> 20: 2 FALSE FALSE TRUE FALSE FALSE TRUE #> iteration bill_depth bill_length body_mass flipper_length island sex #> year classif.ce runtime_learners timestamp batch_nr warnings #> #> 1: TRUE 0.03508772 0.006 2024-07-24 11:01:37 1 0 #> 2: TRUE 0.19298246 0.005 2024-07-24 11:01:37 1 0 #> 3: FALSE 0.03508772 0.005 2024-07-24 11:01:37 1 0 #> 4: FALSE 0.03508772 0.006 2024-07-24 11:01:37 1 0 #> 5: TRUE 0.29824561 0.004 2024-07-24 11:01:37 1 0 #> 6: TRUE 0.03508772 0.005 2024-07-24 11:01:37 1 0 #> 7: FALSE 0.03508772 0.005 2024-07-24 11:01:37 1 0 #> 8: FALSE 0.10526316 0.005 2024-07-24 11:01:37 1 0 #> 9: FALSE 0.15789474 0.005 2024-07-24 11:01:37 1 0 #> 10: FALSE 0.17543860 0.004 2024-07-24 11:01:37 1 0 #> 11: TRUE 0.57894737 0.004 2024-07-24 11:01:37 1 0 #> 12: FALSE 0.19298246 0.004 2024-07-24 11:01:37 1 0 #> 13: FALSE 0.28070175 0.005 2024-07-24 11:01:37 1 0 #> 14: TRUE 0.14035088 0.005 2024-07-24 11:01:37 1 0 #> 15: FALSE 0.21052632 0.005 2024-07-24 11:01:37 1 0 #> 16: TRUE 0.08771930 0.005 2024-07-24 11:01:37 1 0 #> 17: FALSE 0.33333333 0.003 2024-07-24 11:01:37 1 0 #> 18: FALSE 0.14035088 0.005 2024-07-24 11:01:37 1 0 #> 19: TRUE 0.12280702 0.005 2024-07-24 11:01:37 1 0 #> 20: FALSE 0.31578947 0.004 2024-07-24 11:01:37 1 0 #> year classif.ce runtime_learners timestamp batch_nr warnings #> errors features #> #> 1: 0 bill_depth,bill_length,flipper_length,island,sex,year #> 2: 0 bill_length,year #> 3: 0 bill_depth,bill_length,body_mass,island,sex #> 4: 0 bill_depth,bill_length,body_mass,flipper_length,sex #> 5: 0 island,year #> 6: 0 bill_depth,bill_length,flipper_length,island,sex,year #> 7: 0 bill_length,island #> 8: 0 bill_depth,body_mass,flipper_length,island #> 9: 0 bill_depth,flipper_length,sex #> 10: 0 body_mass,flipper_length #> 11: 0 year #> 12: 0 flipper_length #> 13: 0 body_mass #> 14: 0 bill_depth,bill_length,body_mass,flipper_length,island,sex,... #> 15: 0 bill_depth,body_mass,flipper_length,island #> 16: 0 bill_depth,bill_length,flipper_length,island,sex,year #> 17: 0 bill_length #> 18: 0 bill_depth,bill_length,body_mass,island,sex #> 19: 0 flipper_length,island,year #> 20: 0 body_mass,sex #> errors features #> n_features resample_result task_id learner_id resampling_id #> #> 1: 6 penguins classif.rpart.fselector cv #> 2: 2 penguins classif.rpart.fselector cv #> 3: 5 penguins classif.rpart.fselector cv #> 4: 5 penguins classif.rpart.fselector cv #> 5: 2 penguins classif.rpart.fselector cv #> 6: 6 penguins classif.rpart.fselector cv #> 7: 2 penguins classif.rpart.fselector cv #> 8: 4 penguins classif.rpart.fselector cv #> 9: 3 penguins classif.rpart.fselector cv #> 10: 2 penguins classif.rpart.fselector cv #> 11: 1 penguins classif.rpart.fselector cv #> 12: 1 penguins classif.rpart.fselector cv #> 13: 1 penguins classif.rpart.fselector cv #> 14: 7 penguins classif.rpart.fselector cv #> 15: 4 penguins classif.rpart.fselector cv #> 16: 6 penguins classif.rpart.fselector cv #> 17: 1 penguins classif.rpart.fselector cv #> 18: 5 penguins classif.rpart.fselector cv #> 19: 3 penguins classif.rpart.fselector cv #> 20: 2 penguins classif.rpart.fselector cv #> n_features resample_result task_id learner_id resampling_id"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/extract_inner_fselect_results.html","id":null,"dir":"Reference","previous_headings":"","what":"Extract Inner Feature Selection Results — extract_inner_fselect_results","title":"Extract Inner Feature Selection Results — extract_inner_fselect_results","text":"Extract inner feature selection results nested resampling. Implemented mlr3::ResampleResult mlr3::BenchmarkResult.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/extract_inner_fselect_results.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Extract Inner Feature Selection Results — extract_inner_fselect_results","text":"","code":"extract_inner_fselect_results(x, fselect_instance, ...)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/extract_inner_fselect_results.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Extract Inner Feature Selection Results — extract_inner_fselect_results","text":"x (mlr3::ResampleResult | mlr3::BenchmarkResult). fselect_instance (logical(1)) TRUE, instances added table. ... () Additional arguments.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/extract_inner_fselect_results.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Extract Inner Feature Selection Results — extract_inner_fselect_results","text":"data.table::data.table().","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/extract_inner_fselect_results.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Extract Inner Feature Selection Results — extract_inner_fselect_results","text":"function iterates AutoFSelector objects binds feature selection results data.table::data.table(). AutoFSelector must initialized store_fselect_instance = TRUE resample() benchmark() must called store_models = TRUE. Optionally, instance can added iteration.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/extract_inner_fselect_results.html","id":"data-structure","dir":"Reference","previous_headings":"","what":"Data structure","title":"Extract Inner Feature Selection Results — extract_inner_fselect_results","text":"returned data table following columns: experiment (integer(1)) Index, giving according row number original benchmark grid. iteration (integer(1)) Iteration outer resampling. One column feature task. One column performance measure. features (character()) Vector selected feature set. task_id (character(1)). learner_id (character(1)). resampling_id (character(1)).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/extract_inner_fselect_results.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Extract Inner Feature Selection Results — extract_inner_fselect_results","text":"","code":"# Nested Resampling on Palmer Penguins Data Set # create auto fselector at = auto_fselector( fselector = fs(\"random_search\"), learner = lrn(\"classif.rpart\"), resampling = rsmp (\"holdout\"), measure = msr(\"classif.ce\"), term_evals = 4) resampling_outer = rsmp(\"cv\", folds = 2) rr = resample(tsk(\"iris\"), at, resampling_outer, store_models = TRUE) # extract inner results extract_inner_fselect_results(rr) #> iteration Petal.Length Petal.Width Sepal.Length Sepal.Width classif.ce #> #> 1: 1 FALSE TRUE TRUE FALSE 0.04 #> 2: 2 TRUE FALSE FALSE FALSE 0.08 #> features n_features task_id learner_id #> #> 1: Petal.Width,Sepal.Length 2 iris classif.rpart.fselector #> 2: Petal.Length 1 iris classif.rpart.fselector #> resampling_id #> #> 1: cv #> 2: cv"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fs.html","id":null,"dir":"Reference","previous_headings":"","what":"Syntactic Sugar for FSelect Construction — fs","title":"Syntactic Sugar for FSelect Construction — fs","text":"Functions retrieve objects, set parameters assign fields one go. Relies mlr3misc::dictionary_sugar_get() extract objects respective mlr3misc::Dictionary: fs() FSelector mlr_fselectors. fss() list FSelector mlr_fselectors. trm() bbotk::Terminator mlr_terminators. trms() list Terminators mlr_terminators.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fs.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Syntactic Sugar for FSelect Construction — fs","text":"","code":"fs(.key, ...) fss(.keys, ...)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fs.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Syntactic Sugar for FSelect Construction — fs","text":".key (character(1)) Key passed respective dictionary retrieve object. ... () Additional arguments. .keys (character()) Keys passed respective dictionary retrieve multiple objects.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fs.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Syntactic Sugar for FSelect Construction — fs","text":"R6::R6Class object respective type, list R6::R6Class objects plural versions.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fs.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Syntactic Sugar for FSelect Construction — fs","text":"","code":"# random search with batch size of 5 fs(\"random_search\", batch_size = 5) #> : Random Search #> * Parameters: batch_size=5 #> * Properties: single-crit, multi-crit #> * Packages: mlr3fselect # run time terminator with 20 seconds trm(\"run_time\", secs = 20) #> : Run Time #> * Parameters: secs=20"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fselect.html","id":null,"dir":"Reference","previous_headings":"","what":"Function for Feature Selection — fselect","title":"Function for Feature Selection — fselect","text":"Function optimize features mlr3::Learner. function internally creates FSelectInstanceBatchSingleCrit FSelectInstanceBatchMultiCrit describes feature selection problem. executes feature selection FSelector (method) returns result fselect instance ($result). ArchiveBatchFSelect ($archive) stores evaluated hyperparameter configurations performance scores.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fselect.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Function for Feature Selection — fselect","text":"","code":"fselect( fselector, task, learner, resampling, measures = NULL, term_evals = NULL, term_time = NULL, terminator = NULL, store_benchmark_result = TRUE, store_models = FALSE, check_values = FALSE, callbacks = NULL, ties_method = \"least_features\" )"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fselect.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Function for Feature Selection — fselect","text":"fselector (FSelector) Optimization algorithm. task (mlr3::Task) Task operate . learner (mlr3::Learner) Learner optimize feature subset . resampling (mlr3::Resampling) Resampling used evaluated performance feature subsets. Uninstantiated resamplings instantiated construction feature subsets evaluated data splits. Already instantiated resamplings kept unchanged. measures (mlr3::Measure list mlr3::Measure) single measure creates FSelectInstanceBatchSingleCrit multiple measures FSelectInstanceBatchMultiCrit. NULL, default measure used. term_evals (integer(1)) Number allowed evaluations. Ignored terminator passed. term_time (integer(1)) Maximum allowed time seconds. Ignored terminator passed. terminator (bbotk::Terminator) Stop criterion feature selection. store_benchmark_result (logical(1)) Store benchmark result archive? store_models (logical(1)). Store models benchmark result? check_values (logical(1)) Check parameters evaluation results validity? callbacks (list CallbackBatchFSelect) List callbacks. ties_method (character(1)) method break ties selecting sets optimizing selecting best set. Can \"least_features\" \"random\". option \"least_features\" (default) selects feature set least features. multiple best feature sets number features, one selected randomly. random method returns random feature set best feature sets. Ignored multiple measures used.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fselect.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Function for Feature Selection — fselect","text":"FSelectInstanceBatchSingleCrit | FSelectInstanceBatchMultiCrit","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fselect.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Function for Feature Selection — fselect","text":"mlr3::Task, mlr3::Learner, mlr3::Resampling, mlr3::Measure bbotk::Terminator used construct FSelectInstanceBatchSingleCrit. multiple performance Measures supplied, FSelectInstanceBatchMultiCrit created. parameter term_evals term_time shortcuts create bbotk::Terminator. parameters passed, bbotk::TerminatorCombo constructed. Terminators, pass one terminator. termination criterion needed, set term_evals, term_time terminator NULL.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fselect.html","id":"resources","dir":"Reference","previous_headings":"","what":"Resources","title":"Function for Feature Selection — fselect","text":"several sections feature selection mlr3book. Getting started wrapper feature selection. sequential forward selection Palmer Penguins data set. gallery features collection case studies demos optimization. Utilize built-feature importance models Recursive Feature Elimination. Run feature selection Shadow Variable Search. Feature Selection Titanic data set.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fselect.html","id":"analysis","dir":"Reference","previous_headings":"","what":"Analysis","title":"Function for Feature Selection — fselect","text":"analyzing feature selection results, recommended pass archive .data.table(). returned data table joined benchmark result adds mlr3::ResampleResult feature set. archive provides various getters (e.g. $learners()) ease access. getters extract position () unique hash (uhash). complete list getters see methods section. benchmark result ($benchmark_result) allows score feature sets different measure. Alternatively, measures can supplied .data.table().","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fselect.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Function for Feature Selection — fselect","text":"","code":"# Feature selection on the Palmer Penguins data set task = tsk(\"pima\") learner = lrn(\"classif.rpart\") # Run feature selection instance = fselect( fselector = fs(\"random_search\"), task = task, learner = learner, resampling = rsmp (\"holdout\"), measures = msr(\"classif.ce\"), term_evals = 4) # Subset task to optimized feature set task$select(instance$result_feature_set) # Train the learner with optimal feature set on the full data set learner$train(task) # Inspect all evaluated configurations as.data.table(instance$archive) #> age glucose insulin mass pedigree pregnant pressure triceps classif.ce #> #> 1: TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE 0.2500000 #> 2: FALSE FALSE TRUE FALSE TRUE TRUE FALSE FALSE 0.2929688 #> 3: FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE 0.3203125 #> 4: FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE 0.3281250 #> 5: TRUE TRUE TRUE FALSE TRUE FALSE TRUE TRUE 0.2851562 #> 6: TRUE TRUE FALSE FALSE TRUE FALSE TRUE TRUE 0.2851562 #> 7: TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE 0.2500000 #> 8: TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE 0.2500000 #> 9: TRUE TRUE TRUE FALSE TRUE TRUE FALSE FALSE 0.2734375 #> 10: TRUE TRUE TRUE TRUE TRUE FALSE TRUE FALSE 0.2734375 #> runtime_learners timestamp batch_nr warnings errors #> #> 1: 0.009 2024-07-24 11:01:39 1 0 0 #> 2: 0.007 2024-07-24 11:01:39 1 0 0 #> 3: 0.007 2024-07-24 11:01:39 1 0 0 #> 4: 0.007 2024-07-24 11:01:39 1 0 0 #> 5: 0.008 2024-07-24 11:01:39 1 0 0 #> 6: 0.008 2024-07-24 11:01:39 1 0 0 #> 7: 0.009 2024-07-24 11:01:39 1 0 0 #> 8: 0.009 2024-07-24 11:01:39 1 0 0 #> 9: 0.008 2024-07-24 11:01:39 1 0 0 #> 10: 0.007 2024-07-24 11:01:39 1 0 0 #> features n_features resample_result #> #> 1: age,glucose,insulin,mass,pedigree,pregnant,... 8 #> 2: insulin,pedigree,pregnant 3 #> 3: pedigree 1 #> 4: pregnant 1 #> 5: age,glucose,insulin,pedigree,pressure,triceps 6 #> 6: age,glucose,pedigree,pressure,triceps 5 #> 7: age,glucose,insulin,mass,pedigree,pregnant,... 8 #> 8: age,glucose,insulin,mass,pedigree,pregnant,... 8 #> 9: age,glucose,insulin,pedigree,pregnant 5 #> 10: age,glucose,insulin,mass,pedigree,pressure 6 "},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fselect_nested.html","id":null,"dir":"Reference","previous_headings":"","what":"Function for Nested Resampling — fselect_nested","title":"Function for Nested Resampling — fselect_nested","text":"Function conduct nested resampling.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fselect_nested.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Function for Nested Resampling — fselect_nested","text":"","code":"fselect_nested( fselector, task, learner, inner_resampling, outer_resampling, measure = NULL, term_evals = NULL, term_time = NULL, terminator = NULL, store_fselect_instance = TRUE, store_benchmark_result = TRUE, store_models = FALSE, check_values = FALSE, callbacks = NULL, ties_method = \"least_features\" )"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fselect_nested.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Function for Nested Resampling — fselect_nested","text":"fselector (FSelector) Optimization algorithm. task (mlr3::Task) Task operate . learner (mlr3::Learner) Learner optimize feature subset . inner_resampling (mlr3::Resampling) Resampling used inner loop. outer_resampling mlr3::Resampling) Resampling used outer loop. measure (mlr3::Measure) Measure optimize. NULL, default measure used. term_evals (integer(1)) Number allowed evaluations. Ignored terminator passed. term_time (integer(1)) Maximum allowed time seconds. Ignored terminator passed. terminator (bbotk::Terminator) Stop criterion feature selection. store_fselect_instance (logical(1)) TRUE (default), stores internally created FSelectInstanceBatchSingleCrit intermediate results slot $fselect_instance. set TRUE, store_models = TRUE store_benchmark_result (logical(1)) Store benchmark result archive? store_models (logical(1)). Store models benchmark result? check_values (logical(1)) Check parameters evaluation results validity? callbacks (list CallbackBatchFSelect) List callbacks. ties_method (character(1)) method break ties selecting sets optimizing selecting best set. Can \"least_features\" \"random\". option \"least_features\" (default) selects feature set least features. multiple best feature sets number features, one selected randomly. random method returns random feature set best feature sets. Ignored multiple measures used.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fselect_nested.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Function for Nested Resampling — fselect_nested","text":"mlr3::ResampleResult","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fselect_nested.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Function for Nested Resampling — fselect_nested","text":"","code":"# Nested resampling on Palmer Penguins data set rr = fselect_nested( fselector = fs(\"random_search\"), task = tsk(\"penguins\"), learner = lrn(\"classif.rpart\"), inner_resampling = rsmp (\"holdout\"), outer_resampling = rsmp(\"cv\", folds = 2), measure = msr(\"classif.ce\"), term_evals = 4) # Performance scores estimated on the outer resampling rr$score() #> task_id learner_id resampling_id iteration classif.ce #> #> 1: penguins classif.rpart.fselector cv 1 0.05813953 #> 2: penguins classif.rpart.fselector cv 2 0.08139535 #> Hidden columns: task, learner, resampling, prediction # Unbiased performance of the final model trained on the full data set rr$aggregate() #> classif.ce #> 0.06976744"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fsi.html","id":null,"dir":"Reference","previous_headings":"","what":"Syntactic Sugar for Instance Construction — fsi","title":"Syntactic Sugar for Instance Construction — fsi","text":"Function construct FSelectInstanceBatchSingleCrit FSelectInstanceBatchMultiCrit.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fsi.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Syntactic Sugar for Instance Construction — fsi","text":"","code":"fsi( task, learner, resampling, measures = NULL, terminator, store_benchmark_result = TRUE, store_models = FALSE, check_values = FALSE, callbacks = NULL, ties_method = \"least_features\" )"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fsi.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Syntactic Sugar for Instance Construction — fsi","text":"task (mlr3::Task) Task operate . learner (mlr3::Learner) Learner optimize feature subset . resampling (mlr3::Resampling) Resampling used evaluated performance feature subsets. Uninstantiated resamplings instantiated construction feature subsets evaluated data splits. Already instantiated resamplings kept unchanged. measures (mlr3::Measure list mlr3::Measure) single measure creates FSelectInstanceBatchSingleCrit multiple measures FSelectInstanceBatchMultiCrit. NULL, default measure used. terminator (bbotk::Terminator) Stop criterion feature selection. store_benchmark_result (logical(1)) Store benchmark result archive? store_models (logical(1)). Store models benchmark result? check_values (logical(1)) Check parameters evaluation results validity? callbacks (list CallbackBatchFSelect) List callbacks. ties_method (character(1)) method break ties selecting sets optimizing selecting best set. Can \"least_features\" \"random\". option \"least_features\" (default) selects feature set least features. multiple best feature sets number features, one selected randomly. random method returns random feature set best feature sets. Ignored multiple measures used.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fsi.html","id":"resources","dir":"Reference","previous_headings":"","what":"Resources","title":"Syntactic Sugar for Instance Construction — fsi","text":"several sections feature selection mlr3book. Getting started wrapper feature selection. sequential forward selection Palmer Penguins data set. gallery features collection case studies demos optimization. Utilize built-feature importance models Recursive Feature Elimination. Run feature selection Shadow Variable Search. Feature Selection Titanic data set.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fsi.html","id":"default-measures","dir":"Reference","previous_headings":"","what":"Default Measures","title":"Syntactic Sugar for Instance Construction — fsi","text":"measure passed, default measure used. default measure depends task type.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fsi.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Syntactic Sugar for Instance Construction — fsi","text":"","code":"# Feature selection on Palmer Penguins data set # \\donttest{ task = tsk(\"penguins\") learner = lrn(\"classif.rpart\") # Construct feature selection instance instance = fsi( task = task, learner = learner, resampling = rsmp(\"cv\", folds = 3), measures = msr(\"classif.ce\"), terminator = trm(\"evals\", n_evals = 4) ) # Choose optimization algorithm fselector = fs(\"random_search\", batch_size = 2) # Run feature selection fselector$optimize(instance) #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE TRUE FALSE FALSE TRUE TRUE TRUE #> features n_features classif.ce #> #> 1: bill_depth,bill_length,island,sex,year 5 0.06114925 # Subset task to optimal feature set task$select(instance$result_feature_set) # Train the learner with optimal feature set on the full data set learner$train(task) # Inspect all evaluated sets as.data.table(instance$archive) #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE TRUE TRUE FALSE TRUE TRUE FALSE #> 2: FALSE FALSE FALSE TRUE FALSE TRUE FALSE #> 3: FALSE TRUE FALSE TRUE TRUE TRUE TRUE #> 4: TRUE TRUE FALSE FALSE TRUE TRUE TRUE #> classif.ce runtime_learners timestamp batch_nr warnings errors #> #> 1: 0.06114925 0.015 2024-07-24 11:01:41 1 0 0 #> 2: 0.19471142 0.015 2024-07-24 11:01:41 1 0 0 #> 3: 0.06687007 0.016 2024-07-24 11:01:41 2 0 0 #> 4: 0.06114925 0.017 2024-07-24 11:01:41 2 0 0 #> features n_features resample_result #> #> 1: bill_depth,bill_length,body_mass,island,sex 5 #> 2: flipper_length,sex 2 #> 3: bill_length,flipper_length,island,sex,year 5 #> 4: bill_depth,bill_length,island,sex,year 5 # }"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr3fselect-package.html","id":null,"dir":"Reference","previous_headings":"","what":"mlr3fselect: Feature Selection for 'mlr3' — mlr3fselect-package","title":"mlr3fselect: Feature Selection for 'mlr3' — mlr3fselect-package","text":"Feature selection package 'mlr3' ecosystem. selects optimal feature set 'mlr3' learner. package works several optimization algorithms e.g. Random Search, Recursive Feature Elimination, Genetic Search. Moreover, can automatically optimize learners estimate performance optimized feature sets nested resampling.","code":""},{"path":[]},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr3fselect-package.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"mlr3fselect: Feature Selection for 'mlr3' — mlr3fselect-package","text":"Maintainer: Marc Becker marcbecker@posteo.de (ORCID) Authors: Patrick Schratz patrick.schratz@gmail.com (ORCID) Michel Lang michellang@gmail.com (ORCID) Bernd Bischl bernd_bischl@gmx.net (ORCID) John Zobolas bblodfon@gmail.com (ORCID)","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr3fselect.backup.html","id":null,"dir":"Reference","previous_headings":"","what":"Backup Benchmark Result Callback — mlr3fselect.backup","title":"Backup Benchmark Result Callback — mlr3fselect.backup","text":"CallbackBatchFSelect writes mlr3::BenchmarkResult batch disk.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr3fselect.backup.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Backup Benchmark Result Callback — mlr3fselect.backup","text":"","code":"clbk(\"mlr3fselect.backup\", path = \"backup.rds\") #> : Backup Benchmark Result Callback #> * Active Stages: on_optimizer_after_eval, on_optimization_begin # Run feature selection on the Palmer Penguins data set instance = fselect( fselector = fs(\"random_search\"), task = tsk(\"pima\"), learner = lrn(\"classif.rpart\"), resampling = rsmp (\"holdout\"), measures = msr(\"classif.ce\"), term_evals = 4, callbacks = clbk(\"mlr3fselect.backup\", path = tempfile(fileext = \".rds\")))"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr3fselect.one_se_rule.html","id":null,"dir":"Reference","previous_headings":"","what":"One Standard Error Rule Callback — mlr3fselect.one_se_rule","title":"One Standard Error Rule Callback — mlr3fselect.one_se_rule","text":"Selects smallest feature set within one standard error best result. multiple feature sets number features, first one selected. sets exactly performance different number features, one smallest number features selected.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr3fselect.one_se_rule.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"One Standard Error Rule Callback — mlr3fselect.one_se_rule","text":"Kuhn, Max, Johnson, Kjell (2013). “Applied Predictive Modeling.” chapter -Fitting Model Tuning, 61–92. Springer New York, New York, NY. ISBN 978-1-4614-6849-3.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr3fselect.one_se_rule.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"One Standard Error Rule Callback — mlr3fselect.one_se_rule","text":"","code":"clbk(\"mlr3fselect.one_se_rule\") #> : One Standard Error Rule Callback #> * Active Stages: on_result # Run feature selection on the pima data set with the callback instance = fselect( fselector = fs(\"random_search\"), task = tsk(\"pima\"), learner = lrn(\"classif.rpart\"), resampling = rsmp (\"cv\", folds = 3), measures = msr(\"classif.ce\"), term_evals = 10, callbacks = clbk(\"mlr3fselect.one_se_rule\")) # Smallest feature set within one standard error of the best instance$result #> age glucose insulin mass pedigree pregnant pressure triceps #> #> 1: TRUE TRUE TRUE TRUE FALSE FALSE TRUE FALSE #> features n_features classif.ce #> #> 1: age,glucose,insulin,mass,pressure 5 0.2578125"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr3fselect.svm_rfe.html","id":null,"dir":"Reference","previous_headings":"","what":"SVM-RFE Callback — mlr3fselect.svm_rfe","title":"SVM-RFE Callback — mlr3fselect.svm_rfe","text":"Runs recursive feature elimination mlr3learners::LearnerClassifSVM. SVM must configured type = \"C-classification\" kernel = \"linear\".","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr3fselect.svm_rfe.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"SVM-RFE Callback — mlr3fselect.svm_rfe","text":"Guyon , Weston J, Barnhill S, Vapnik V (2002). “Gene Selection Cancer Classification using Support Vector Machines.” Machine Learning, 46(1), 389–422. ISSN 1573-0565, doi:10.1023/:1012487302797 .","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr3fselect.svm_rfe.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"SVM-RFE Callback — mlr3fselect.svm_rfe","text":"","code":"clbk(\"mlr3fselect.svm_rfe\") #> : SVM-RFE Callback #> * Active Stages: on_optimization_begin library(mlr3learners) # Create instance with classification svm with linear kernel instance = fsi( task = tsk(\"sonar\"), learner = lrn(\"classif.svm\", type = \"C-classification\", kernel = \"linear\"), resampling = rsmp(\"cv\", folds = 3), measures = msr(\"classif.ce\"), terminator = trm(\"none\"), callbacks = clbk(\"mlr3fselect.svm_rfe\"), store_models = TRUE ) fselector = fs(\"rfe\", feature_number = 5, n_features = 10) # Run recursive feature elimination on the Sonar data set fselector$optimize(instance) #> V1 V10 V11 V12 V13 V14 V15 V16 V17 V18 V19 #> #> 1: TRUE FALSE TRUE TRUE FALSE TRUE FALSE TRUE FALSE FALSE FALSE #> V2 V20 V21 V22 V23 V24 V25 V26 V27 V28 V29 #> #> 1: FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE #> V3 V30 V31 V32 V33 V34 V35 V36 V37 V38 V39 #> #> 1: TRUE TRUE TRUE TRUE FALSE FALSE FALSE TRUE TRUE TRUE FALSE #> V4 V40 V41 V42 V43 V44 V45 V46 V47 V48 V49 #> #> 1: TRUE TRUE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE #> V5 V50 V51 V52 V53 V54 V55 V56 V57 V58 V59 #> #> 1: FALSE TRUE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE #> V6 V60 V7 V8 V9 #> #> 1: FALSE FALSE TRUE FALSE FALSE #> importance #> #> 1: 20.00000,17.33333,16.33333,15.66667,15.00000,13.00000,... #> features n_features classif.ce #> #> 1: V1,V11,V12,V14,V16,V23,... 20 0.1538992"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr3fselect_assertions.html","id":null,"dir":"Reference","previous_headings":"","what":"Assertion for mlr3fselect objects — mlr3fselect_assertions","title":"Assertion for mlr3fselect objects — mlr3fselect_assertions","text":"assertion functions ensure right class attribute, optionally additional properties.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr3fselect_assertions.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Assertion for mlr3fselect objects — mlr3fselect_assertions","text":"","code":"assert_fselectors(fselectors) assert_fselector_async(fselector) assert_fselector_batch(fselector) assert_fselect_instance(inst) assert_fselect_instance_async(inst) assert_fselect_instance_batch(inst)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr3fselect_assertions.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Assertion for mlr3fselect objects — mlr3fselect_assertions","text":"fselectors (list FSelector). fselector (FSelectorBatch). inst (FSelectInstanceBatchSingleCrit | FSelectInstanceBatchMultiCrit).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors.html","id":null,"dir":"Reference","previous_headings":"","what":"Dictionary of FSelectors — mlr_fselectors","title":"Dictionary of FSelectors — mlr_fselectors","text":"mlr3misc::Dictionary storing objects class FSelector. fselector associated help page, see mlr_fselectors_[id]. convenient way retrieve construct fselectors, see fs()/fss().","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Dictionary of FSelectors — mlr_fselectors","text":"R6::R6Class object inheriting mlr3misc::Dictionary.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Dictionary of FSelectors — mlr_fselectors","text":"See mlr3misc::Dictionary.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors.html","id":"s-methods","dir":"Reference","previous_headings":"","what":"S3 methods","title":"Dictionary of FSelectors — mlr_fselectors","text":".data.table(dict, ..., objects = FALSE)mlr3misc::Dictionary -> data.table::data.table() Returns data.table::data.table() fields \"key\", \"label\", \"properties\" \"packages\" columns. objects set TRUE, constructed objects returned list column named object.","code":""},{"path":[]},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Dictionary of FSelectors — mlr_fselectors","text":"","code":"as.data.table(mlr_fselectors) #> Key: #> key label #> #> 1: design_points Design Points #> 2: exhaustive_search Exhaustive Search #> 3: genetic_search Genetic Search #> 4: random_search Random Search #> 5: rfe Recursive Feature Elimination #> 6: rfecv Recursive Feature Elimination #> 7: sequential Sequential Search #> 8: shadow_variable_search Shadow Variable Search #> properties packages #> #> 1: dependencies,single-crit,multi-crit mlr3fselect,bbotk #> 2: single-crit,multi-crit mlr3fselect #> 3: single-crit mlr3fselect,genalg #> 4: single-crit,multi-crit mlr3fselect #> 5: single-crit,requires_model mlr3fselect #> 6: single-crit,requires_model mlr3fselect #> 7: single-crit mlr3fselect #> 8: single-crit mlr3fselect mlr_fselectors$get(\"random_search\") #> : Random Search #> * Parameters: batch_size=10 #> * Properties: single-crit, multi-crit #> * Packages: mlr3fselect fs(\"random_search\") #> : Random Search #> * Parameters: batch_size=10 #> * Properties: single-crit, multi-crit #> * Packages: mlr3fselect"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_design_points.html","id":null,"dir":"Reference","previous_headings":"","what":"Feature Selection with Design Points — mlr_fselectors_design_points","title":"Feature Selection with Design Points — mlr_fselectors_design_points","text":"Feature selection using user-defined feature sets.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_design_points.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Feature Selection with Design Points — mlr_fselectors_design_points","text":"feature sets evaluated order given. feature selection terminates feature sets evaluated. necessary set termination criterion.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_design_points.html","id":"dictionary","dir":"Reference","previous_headings":"","what":"Dictionary","title":"Feature Selection with Design Points — mlr_fselectors_design_points","text":"FSelector can instantiated associated sugar function fs():","code":"fs(\"design_points\")"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_design_points.html","id":"parameters","dir":"Reference","previous_headings":"","what":"Parameters","title":"Feature Selection with Design Points — mlr_fselectors_design_points","text":"batch_size integer(1) Maximum number configurations try batch. design data.table::data.table Design points try search, one per row.","code":""},{"path":[]},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_design_points.html","id":"super-classes","dir":"Reference","previous_headings":"","what":"Super classes","title":"Feature Selection with Design Points — mlr_fselectors_design_points","text":"mlr3fselect::FSelector -> mlr3fselect::FSelectorBatch -> mlr3fselect::FSelectorBatchFromOptimizerBatch -> FSelectorBatchDesignPoints","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_design_points.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Feature Selection with Design Points — mlr_fselectors_design_points","text":"mlr3fselect::FSelector$format() mlr3fselect::FSelector$help() mlr3fselect::FSelector$print() mlr3fselect::FSelectorBatchFromOptimizerBatch$optimize()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_design_points.html","id":"public-methods","dir":"Reference","previous_headings":"","what":"Public methods","title":"Feature Selection with Design Points — mlr_fselectors_design_points","text":"FSelectorBatchDesignPoints$new() FSelectorBatchDesignPoints$clone()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_design_points.html","id":"method-new-","dir":"Reference","previous_headings":"","what":"Method new()","title":"Feature Selection with Design Points — mlr_fselectors_design_points","text":"Creates new instance R6 class.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_design_points.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Feature Selection with Design Points — mlr_fselectors_design_points","text":"","code":"FSelectorBatchDesignPoints$new()"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_design_points.html","id":"method-clone-","dir":"Reference","previous_headings":"","what":"Method clone()","title":"Feature Selection with Design Points — mlr_fselectors_design_points","text":"objects class cloneable method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_design_points.html","id":"usage-1","dir":"Reference","previous_headings":"","what":"Usage","title":"Feature Selection with Design Points — mlr_fselectors_design_points","text":"","code":"FSelectorBatchDesignPoints$clone(deep = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_design_points.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Feature Selection with Design Points — mlr_fselectors_design_points","text":"deep Whether make deep clone.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_design_points.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Feature Selection with Design Points — mlr_fselectors_design_points","text":"","code":"# Feature Selection # \\donttest{ # retrieve task and load learner task = tsk(\"pima\") learner = lrn(\"classif.rpart\") # create design design = mlr3misc::rowwise_table( ~age, ~glucose, ~insulin, ~mass, ~pedigree, ~pregnant, ~pressure, ~triceps, TRUE, FALSE, TRUE, TRUE, FALSE, TRUE, FALSE, TRUE, TRUE, TRUE, FALSE, TRUE, FALSE, TRUE, FALSE, FALSE, TRUE, FALSE, TRUE, TRUE, FALSE, TRUE, FALSE, FALSE, TRUE, FALSE, TRUE, TRUE, FALSE, TRUE, TRUE, TRUE ) # run feature selection on the Pima Indians diabetes data set instance = fselect( fselector = fs(\"design_points\", design = design), task = task, learner = learner, resampling = rsmp(\"holdout\"), measure = msr(\"classif.ce\") ) # best performing feature set instance$result #> age glucose insulin mass pedigree pregnant pressure triceps #> #> 1: TRUE FALSE TRUE TRUE FALSE TRUE FALSE TRUE #> features n_features classif.ce #> #> 1: age,insulin,mass,pregnant,triceps 5 0.2617188 # all evaluated feature sets as.data.table(instance$archive) #> age glucose insulin mass pedigree pregnant pressure triceps classif.ce #> #> 1: TRUE FALSE TRUE TRUE FALSE TRUE FALSE TRUE 0.2617188 #> 2: TRUE TRUE FALSE TRUE FALSE TRUE FALSE FALSE 0.2734375 #> 3: TRUE FALSE TRUE TRUE FALSE TRUE FALSE FALSE 0.2734375 #> 4: TRUE FALSE TRUE TRUE FALSE TRUE TRUE TRUE 0.2617188 #> runtime_learners timestamp batch_nr warnings errors #> #> 1: 0.007 2024-07-24 11:01:46 1 0 0 #> 2: 0.008 2024-07-24 11:01:46 2 0 0 #> 3: 0.008 2024-07-24 11:01:46 3 0 0 #> 4: 0.008 2024-07-24 11:01:46 4 0 0 #> features n_features resample_result #> #> 1: age,insulin,mass,pregnant,triceps 5 #> 2: age,glucose,mass,pregnant 4 #> 3: age,insulin,mass,pregnant 4 #> 4: age,insulin,mass,pregnant,pressure,triceps 6 # subset the task and fit the final model task$select(instance$result_feature_set) learner$train(task) # }"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_exhaustive_search.html","id":null,"dir":"Reference","previous_headings":"","what":"Feature Selection with Exhaustive Search — mlr_fselectors_exhaustive_search","title":"Feature Selection with Exhaustive Search — mlr_fselectors_exhaustive_search","text":"Feature Selection using Exhaustive Search Algorithm. Exhaustive Search generates possible feature sets.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_exhaustive_search.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Feature Selection with Exhaustive Search — mlr_fselectors_exhaustive_search","text":"feature selection terminates feature sets evaluated. necessary set termination criterion.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_exhaustive_search.html","id":"dictionary","dir":"Reference","previous_headings":"","what":"Dictionary","title":"Feature Selection with Exhaustive Search — mlr_fselectors_exhaustive_search","text":"FSelector can instantiated associated sugar function fs():","code":"fs(\"exhaustive_search\")"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_exhaustive_search.html","id":"control-parameters","dir":"Reference","previous_headings":"","what":"Control Parameters","title":"Feature Selection with Exhaustive Search — mlr_fselectors_exhaustive_search","text":"max_features integer(1) Maximum number features. default, number features mlr3::Task.","code":""},{"path":[]},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_exhaustive_search.html","id":"super-classes","dir":"Reference","previous_headings":"","what":"Super classes","title":"Feature Selection with Exhaustive Search — mlr_fselectors_exhaustive_search","text":"mlr3fselect::FSelector -> mlr3fselect::FSelectorBatch -> FSelectorBatchExhaustiveSearch","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_exhaustive_search.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Feature Selection with Exhaustive Search — mlr_fselectors_exhaustive_search","text":"mlr3fselect::FSelector$format() mlr3fselect::FSelector$help() mlr3fselect::FSelector$print() mlr3fselect::FSelectorBatch$optimize()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_exhaustive_search.html","id":"public-methods","dir":"Reference","previous_headings":"","what":"Public methods","title":"Feature Selection with Exhaustive Search — mlr_fselectors_exhaustive_search","text":"FSelectorBatchExhaustiveSearch$new() FSelectorBatchExhaustiveSearch$clone()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_exhaustive_search.html","id":"method-new-","dir":"Reference","previous_headings":"","what":"Method new()","title":"Feature Selection with Exhaustive Search — mlr_fselectors_exhaustive_search","text":"Creates new instance R6 class.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_exhaustive_search.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Feature Selection with Exhaustive Search — mlr_fselectors_exhaustive_search","text":"","code":"FSelectorBatchExhaustiveSearch$new()"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_exhaustive_search.html","id":"method-clone-","dir":"Reference","previous_headings":"","what":"Method clone()","title":"Feature Selection with Exhaustive Search — mlr_fselectors_exhaustive_search","text":"objects class cloneable method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_exhaustive_search.html","id":"usage-1","dir":"Reference","previous_headings":"","what":"Usage","title":"Feature Selection with Exhaustive Search — mlr_fselectors_exhaustive_search","text":"","code":"FSelectorBatchExhaustiveSearch$clone(deep = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_exhaustive_search.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Feature Selection with Exhaustive Search — mlr_fselectors_exhaustive_search","text":"deep Whether make deep clone.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_exhaustive_search.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Feature Selection with Exhaustive Search — mlr_fselectors_exhaustive_search","text":"","code":"# Feature Selection # \\donttest{ # retrieve task and load learner task = tsk(\"penguins\") learner = lrn(\"classif.rpart\") # run feature selection on the Palmer Penguins data set instance = fselect( fselector = fs(\"exhaustive_search\"), task = task, learner = learner, resampling = rsmp(\"holdout\"), measure = msr(\"classif.ce\"), term_evals = 10 ) # best performing feature set instance$result #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE TRUE FALSE FALSE FALSE FALSE FALSE #> features n_features classif.ce #> #> 1: bill_depth,bill_length 2 0.09565217 # all evaluated feature sets as.data.table(instance$archive) #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE FALSE FALSE FALSE FALSE FALSE FALSE #> 2: FALSE TRUE FALSE FALSE FALSE FALSE FALSE #> 3: FALSE FALSE TRUE FALSE FALSE FALSE FALSE #> 4: FALSE FALSE FALSE TRUE FALSE FALSE FALSE #> 5: FALSE FALSE FALSE FALSE TRUE FALSE FALSE #> 6: FALSE FALSE FALSE FALSE FALSE TRUE FALSE #> 7: FALSE FALSE FALSE FALSE FALSE FALSE TRUE #> 8: TRUE TRUE FALSE FALSE FALSE FALSE FALSE #> 9: TRUE FALSE TRUE FALSE FALSE FALSE FALSE #> 10: TRUE FALSE FALSE TRUE FALSE FALSE FALSE #> classif.ce runtime_learners timestamp batch_nr warnings errors #> #> 1: 0.26086957 0.004 2024-07-24 11:01:47 1 0 0 #> 2: 0.21739130 0.005 2024-07-24 11:01:47 1 0 0 #> 3: 0.29565217 0.004 2024-07-24 11:01:47 1 0 0 #> 4: 0.15652174 0.005 2024-07-24 11:01:47 1 0 0 #> 5: 0.27826087 0.004 2024-07-24 11:01:47 1 0 0 #> 6: 0.57391304 0.005 2024-07-24 11:01:47 1 0 0 #> 7: 0.57391304 0.003 2024-07-24 11:01:47 1 0 0 #> 8: 0.09565217 0.004 2024-07-24 11:01:47 1 0 0 #> 9: 0.25217391 0.004 2024-07-24 11:01:47 1 0 0 #> 10: 0.15652174 0.004 2024-07-24 11:01:47 1 0 0 #> features n_features resample_result #> #> 1: bill_depth 1 #> 2: bill_length 1 #> 3: body_mass 1 #> 4: flipper_length 1 #> 5: island 1 #> 6: sex 1 #> 7: year 1 #> 8: bill_depth,bill_length 2 #> 9: bill_depth,body_mass 2 #> 10: bill_depth,flipper_length 2 # subset the task and fit the final model task$select(instance$result_feature_set) learner$train(task) # }"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_genetic_search.html","id":null,"dir":"Reference","previous_headings":"","what":"Feature Selection with Genetic Search — mlr_fselectors_genetic_search","title":"Feature Selection with Genetic Search — mlr_fselectors_genetic_search","text":"Feature selection using Genetic Algorithm package genalg.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_genetic_search.html","id":"dictionary","dir":"Reference","previous_headings":"","what":"Dictionary","title":"Feature Selection with Genetic Search — mlr_fselectors_genetic_search","text":"FSelector can instantiated associated sugar function fs():","code":"fs(\"genetic_search\")"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_genetic_search.html","id":"control-parameters","dir":"Reference","previous_headings":"","what":"Control Parameters","title":"Feature Selection with Genetic Search — mlr_fselectors_genetic_search","text":"meaning control parameters, see genalg::rbga.bin(). genalg::rbga.bin() internally terminates iters iteration. set ìters = 100000 allow termination via terminators. iterations needed, set ìters higher value parameter set.","code":""},{"path":[]},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_genetic_search.html","id":"super-classes","dir":"Reference","previous_headings":"","what":"Super classes","title":"Feature Selection with Genetic Search — mlr_fselectors_genetic_search","text":"mlr3fselect::FSelector -> mlr3fselect::FSelectorBatch -> FSelectorBatchGeneticSearch","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_genetic_search.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Feature Selection with Genetic Search — mlr_fselectors_genetic_search","text":"mlr3fselect::FSelector$format() mlr3fselect::FSelector$help() mlr3fselect::FSelector$print() mlr3fselect::FSelectorBatch$optimize()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_genetic_search.html","id":"public-methods","dir":"Reference","previous_headings":"","what":"Public methods","title":"Feature Selection with Genetic Search — mlr_fselectors_genetic_search","text":"FSelectorBatchGeneticSearch$new() FSelectorBatchGeneticSearch$clone()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_genetic_search.html","id":"method-new-","dir":"Reference","previous_headings":"","what":"Method new()","title":"Feature Selection with Genetic Search — mlr_fselectors_genetic_search","text":"Creates new instance R6 class.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_genetic_search.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Feature Selection with Genetic Search — mlr_fselectors_genetic_search","text":"","code":"FSelectorBatchGeneticSearch$new()"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_genetic_search.html","id":"method-clone-","dir":"Reference","previous_headings":"","what":"Method clone()","title":"Feature Selection with Genetic Search — mlr_fselectors_genetic_search","text":"objects class cloneable method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_genetic_search.html","id":"usage-1","dir":"Reference","previous_headings":"","what":"Usage","title":"Feature Selection with Genetic Search — mlr_fselectors_genetic_search","text":"","code":"FSelectorBatchGeneticSearch$clone(deep = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_genetic_search.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Feature Selection with Genetic Search — mlr_fselectors_genetic_search","text":"deep Whether make deep clone.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_genetic_search.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Feature Selection with Genetic Search — mlr_fselectors_genetic_search","text":"","code":"# Feature Selection # \\donttest{ # retrieve task and load learner task = tsk(\"penguins\") learner = lrn(\"classif.rpart\") # run feature selection on the Palmer Penguins data set instance = fselect( fselector = fs(\"genetic_search\"), task = task, learner = learner, resampling = rsmp(\"holdout\"), measure = msr(\"classif.ce\"), term_evals = 10 ) # best performing feature set instance$result #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: FALSE TRUE TRUE FALSE FALSE FALSE FALSE #> features n_features classif.ce #> #> 1: bill_length,body_mass 2 0.03478261 # all evaluated feature sets as.data.table(instance$archive) #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE FALSE FALSE FALSE FALSE FALSE FALSE #> 2: FALSE FALSE TRUE FALSE FALSE FALSE FALSE #> 3: TRUE FALSE FALSE TRUE FALSE FALSE FALSE #> 4: FALSE TRUE FALSE FALSE FALSE FALSE FALSE #> 5: FALSE FALSE FALSE TRUE FALSE FALSE FALSE #> 6: TRUE FALSE FALSE FALSE FALSE FALSE FALSE #> 7: FALSE TRUE TRUE FALSE FALSE FALSE FALSE #> 8: FALSE TRUE FALSE FALSE FALSE FALSE FALSE #> 9: FALSE FALSE FALSE FALSE TRUE FALSE FALSE #> 10: FALSE TRUE FALSE TRUE TRUE FALSE FALSE #> classif.ce runtime_learners timestamp batch_nr warnings errors #> #> 1: 0.30434783 0.004 2024-07-24 11:01:48 1 0 0 #> 2: 0.31304348 0.005 2024-07-24 11:01:48 2 0 0 #> 3: 0.24347826 0.003 2024-07-24 11:01:48 3 0 0 #> 4: 0.24347826 0.003 2024-07-24 11:01:48 4 0 0 #> 5: 0.22608696 0.004 2024-07-24 11:01:48 5 0 0 #> 6: 0.30434783 0.004 2024-07-24 11:01:48 6 0 0 #> 7: 0.03478261 0.005 2024-07-24 11:01:48 7 0 0 #> 8: 0.24347826 0.008 2024-07-24 11:01:48 8 0 0 #> 9: 0.37391304 0.004 2024-07-24 11:01:48 9 0 0 #> 10: 0.06086957 0.006 2024-07-24 11:01:48 10 0 0 #> features n_features resample_result #> #> 1: bill_depth 1 #> 2: body_mass 1 #> 3: bill_depth,flipper_length 2 #> 4: bill_length 1 #> 5: flipper_length 1 #> 6: bill_depth 1 #> 7: bill_length,body_mass 2 #> 8: bill_length 1 #> 9: island 1 #> 10: bill_length,flipper_length,island 3 # subset the task and fit the final model task$select(instance$result_feature_set) learner$train(task) # }"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_random_search.html","id":null,"dir":"Reference","previous_headings":"","what":"Feature Selection with Random Search — mlr_fselectors_random_search","title":"Feature Selection with Random Search — mlr_fselectors_random_search","text":"Feature selection using Random Search Algorithm.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_random_search.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Feature Selection with Random Search — mlr_fselectors_random_search","text":"Bergstra J, Bengio Y (2012). “Random Search Hyper-Parameter Optimization.” Journal Machine Learning Research, 13(10), 281–305. https://jmlr.csail.mit.edu/papers/v13/bergstra12a.html.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_random_search.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Feature Selection with Random Search — mlr_fselectors_random_search","text":"feature sets randomly drawn. sets evaluated batches size batch_size. Larger batches mean can parallelize , smaller batches imply fine-grained checking termination criteria.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_random_search.html","id":"dictionary","dir":"Reference","previous_headings":"","what":"Dictionary","title":"Feature Selection with Random Search — mlr_fselectors_random_search","text":"FSelector can instantiated associated sugar function fs():","code":"fs(\"random_search\")"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_random_search.html","id":"control-parameters","dir":"Reference","previous_headings":"","what":"Control Parameters","title":"Feature Selection with Random Search — mlr_fselectors_random_search","text":"max_features integer(1) Maximum number features. default, number features mlr3::Task. batch_size integer(1) Maximum number feature sets try batch.","code":""},{"path":[]},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_random_search.html","id":"super-classes","dir":"Reference","previous_headings":"","what":"Super classes","title":"Feature Selection with Random Search — mlr_fselectors_random_search","text":"mlr3fselect::FSelector -> mlr3fselect::FSelectorBatch -> FSelectorBatchRandomSearch","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_random_search.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Feature Selection with Random Search — mlr_fselectors_random_search","text":"mlr3fselect::FSelector$format() mlr3fselect::FSelector$help() mlr3fselect::FSelector$print() mlr3fselect::FSelectorBatch$optimize()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_random_search.html","id":"public-methods","dir":"Reference","previous_headings":"","what":"Public methods","title":"Feature Selection with Random Search — mlr_fselectors_random_search","text":"FSelectorBatchRandomSearch$new() FSelectorBatchRandomSearch$clone()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_random_search.html","id":"method-new-","dir":"Reference","previous_headings":"","what":"Method new()","title":"Feature Selection with Random Search — mlr_fselectors_random_search","text":"Creates new instance R6 class.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_random_search.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Feature Selection with Random Search — mlr_fselectors_random_search","text":"","code":"FSelectorBatchRandomSearch$new()"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_random_search.html","id":"method-clone-","dir":"Reference","previous_headings":"","what":"Method clone()","title":"Feature Selection with Random Search — mlr_fselectors_random_search","text":"objects class cloneable method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_random_search.html","id":"usage-1","dir":"Reference","previous_headings":"","what":"Usage","title":"Feature Selection with Random Search — mlr_fselectors_random_search","text":"","code":"FSelectorBatchRandomSearch$clone(deep = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_random_search.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Feature Selection with Random Search — mlr_fselectors_random_search","text":"deep Whether make deep clone.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_random_search.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Feature Selection with Random Search — mlr_fselectors_random_search","text":"","code":"# Feature Selection # \\donttest{ # retrieve task and load learner task = tsk(\"penguins\") learner = lrn(\"classif.rpart\") # run feature selection on the Palmer Penguins data set instance = fselect( fselector = fs(\"random_search\"), task = task, learner = learner, resampling = rsmp(\"holdout\"), measure = msr(\"classif.ce\"), term_evals = 10 ) # best performing feature subset instance$result #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: FALSE TRUE FALSE TRUE TRUE FALSE FALSE #> features n_features classif.ce #> #> 1: bill_length,flipper_length,island 3 0.08695652 # all evaluated feature subsets as.data.table(instance$archive) #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> 2: FALSE FALSE TRUE FALSE FALSE TRUE TRUE #> 3: FALSE TRUE FALSE TRUE TRUE FALSE FALSE #> 4: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> 5: TRUE FALSE FALSE FALSE TRUE TRUE FALSE #> 6: TRUE TRUE TRUE TRUE TRUE TRUE FALSE #> 7: FALSE FALSE FALSE TRUE FALSE FALSE FALSE #> 8: FALSE TRUE FALSE FALSE FALSE FALSE FALSE #> 9: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> 10: FALSE FALSE FALSE TRUE FALSE FALSE FALSE #> classif.ce runtime_learners timestamp batch_nr warnings errors #> #> 1: 0.08695652 0.006 2024-07-24 11:01:49 1 0 0 #> 2: 0.27826087 0.006 2024-07-24 11:01:49 1 0 0 #> 3: 0.08695652 0.005 2024-07-24 11:01:49 1 0 0 #> 4: 0.08695652 0.006 2024-07-24 11:01:49 1 0 0 #> 5: 0.21739130 0.005 2024-07-24 11:01:49 1 0 0 #> 6: 0.08695652 0.005 2024-07-24 11:01:49 1 0 0 #> 7: 0.23478261 0.005 2024-07-24 11:01:49 1 0 0 #> 8: 0.23478261 0.005 2024-07-24 11:01:49 1 0 0 #> 9: 0.08695652 0.005 2024-07-24 11:01:49 1 0 0 #> 10: 0.23478261 0.004 2024-07-24 11:01:49 1 0 0 #> features n_features #> #> 1: bill_depth,bill_length,body_mass,flipper_length,island,sex,... 7 #> 2: body_mass,sex,year 3 #> 3: bill_length,flipper_length,island 3 #> 4: bill_depth,bill_length,body_mass,flipper_length,island,sex,... 7 #> 5: bill_depth,island,sex 3 #> 6: bill_depth,bill_length,body_mass,flipper_length,island,sex 6 #> 7: flipper_length 1 #> 8: bill_length 1 #> 9: bill_depth,bill_length,body_mass,flipper_length,island,sex,... 7 #> 10: flipper_length 1 #> resample_result #> #> 1: #> 2: #> 3: #> 4: #> 5: #> 6: #> 7: #> 8: #> 9: #> 10: # subset the task and fit the final model task$select(instance$result_feature_set) learner$train(task) # }"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfe.html","id":null,"dir":"Reference","previous_headings":"","what":"Feature Selection with Recursive Feature Elimination — mlr_fselectors_rfe","title":"Feature Selection with Recursive Feature Elimination — mlr_fselectors_rfe","text":"Feature selection using Recursive Feature Elimination (RFE) algorithm. Recursive feature elimination iteratively removes features low importance score. works mlr3::Learners can calculate importance scores (see section optional extractors mlr3::Learner).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfe.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Feature Selection with Recursive Feature Elimination — mlr_fselectors_rfe","text":"Guyon , Weston J, Barnhill S, Vapnik V (2002). “Gene Selection Cancer Classification using Support Vector Machines.” Machine Learning, 46(1), 389–422. ISSN 1573-0565, doi:10.1023/:1012487302797 .","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfe.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Feature Selection with Recursive Feature Elimination — mlr_fselectors_rfe","text":"learner trained features start importance scores calculated feature. least important feature removed learner trained reduced feature set. importance scores calculated procedure repeated desired number features reached. non-recursive option (recursive = FALSE) uses importance scores calculated first iteration. feature selection terminates n_features reached. necessary set termination criterion. using cross-validation resampling strategy, importance scores resampling iterations aggregated. parameter aggregation determines importance scores aggregated. default (\"rank\"), importance score vector fold ranked feature lowest average rank removed. option \"mean\" averages score feature across resampling iterations removes feature lowest average score. Averaging scores appropriate importance measures.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfe.html","id":"archive","dir":"Reference","previous_headings":"","what":"Archive","title":"Feature Selection with Recursive Feature Elimination — mlr_fselectors_rfe","text":"ArchiveBatchFSelect holds following additional columns: \"importance\" (numeric()) importance score vector feature subset.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfe.html","id":"resources","dir":"Reference","previous_headings":"","what":"Resources","title":"Feature Selection with Recursive Feature Elimination — mlr_fselectors_rfe","text":"gallery features collection case studies demos optimization. Utilize built-feature importance models Recursive Feature Elimination.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfe.html","id":"dictionary","dir":"Reference","previous_headings":"","what":"Dictionary","title":"Feature Selection with Recursive Feature Elimination — mlr_fselectors_rfe","text":"FSelector can instantiated associated sugar function fs():","code":"fs(\"rfe\")"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfe.html","id":"control-parameters","dir":"Reference","previous_headings":"","what":"Control Parameters","title":"Feature Selection with Recursive Feature Elimination — mlr_fselectors_rfe","text":"n_features integer(1) minimum number features select, default half features. feature_fraction double(1) Fraction features retain iteration. default 0.5 retains half features. feature_number integer(1) Number features remove iteration. subset_sizes integer() Vector number features retain iteration. Must sorted decreasing order. recursive logical(1) TRUE (default), feature importance calculated iteration. aggregation character(1) aggregation method importance scores resampling iterations. See details. parameter feature_fraction, feature_number subset_sizes mutually exclusive.","code":""},{"path":[]},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfe.html","id":"super-classes","dir":"Reference","previous_headings":"","what":"Super classes","title":"Feature Selection with Recursive Feature Elimination — mlr_fselectors_rfe","text":"mlr3fselect::FSelector -> mlr3fselect::FSelectorBatch -> FSelectorBatchRFE","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfe.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Feature Selection with Recursive Feature Elimination — mlr_fselectors_rfe","text":"mlr3fselect::FSelector$format() mlr3fselect::FSelector$help() mlr3fselect::FSelector$print() mlr3fselect::FSelectorBatch$optimize()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfe.html","id":"public-methods","dir":"Reference","previous_headings":"","what":"Public methods","title":"Feature Selection with Recursive Feature Elimination — mlr_fselectors_rfe","text":"FSelectorBatchRFE$new() FSelectorBatchRFE$clone()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfe.html","id":"method-new-","dir":"Reference","previous_headings":"","what":"Method new()","title":"Feature Selection with Recursive Feature Elimination — mlr_fselectors_rfe","text":"Creates new instance R6 class.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfe.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Feature Selection with Recursive Feature Elimination — mlr_fselectors_rfe","text":"","code":"FSelectorBatchRFE$new()"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfe.html","id":"method-clone-","dir":"Reference","previous_headings":"","what":"Method clone()","title":"Feature Selection with Recursive Feature Elimination — mlr_fselectors_rfe","text":"objects class cloneable method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfe.html","id":"usage-1","dir":"Reference","previous_headings":"","what":"Usage","title":"Feature Selection with Recursive Feature Elimination — mlr_fselectors_rfe","text":"","code":"FSelectorBatchRFE$clone(deep = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfe.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Feature Selection with Recursive Feature Elimination — mlr_fselectors_rfe","text":"deep Whether make deep clone.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfe.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Feature Selection with Recursive Feature Elimination — mlr_fselectors_rfe","text":"","code":"# Feature Selection # \\donttest{ # retrieve task and load learner task = tsk(\"penguins\") learner = lrn(\"classif.rpart\") # run feature selection on the Palmer Penguins data set instance = fselect( fselector = fs(\"rfe\"), task = task, learner = learner, resampling = rsmp(\"holdout\"), measure = msr(\"classif.ce\"), store_models = TRUE ) # best performing feature subset instance$result #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> importance #> #> 1: 7,6,5,4,3,2,... #> features n_features #> #> 1: bill_depth,bill_length,body_mass,flipper_length,island,sex,... 7 #> classif.ce #> #> 1: 0.08695652 # all evaluated feature subsets as.data.table(instance$archive) #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> 2: TRUE TRUE FALSE TRUE FALSE FALSE FALSE #> classif.ce runtime_learners timestamp batch_nr warnings errors #> #> 1: 0.08695652 0.006 2024-07-24 11:01:50 1 0 0 #> 2: 0.10434783 0.006 2024-07-24 11:01:50 2 0 0 #> importance #> #> 1: 7,6,5,4,3,2,... #> 2: 3,2,1 #> features n_features #> #> 1: bill_depth,bill_length,body_mass,flipper_length,island,sex,... 7 #> 2: bill_depth,bill_length,flipper_length 3 #> resample_result #> #> 1: #> 2: # subset the task and fit the final model task$select(instance$result_feature_set) learner$train(task) # }"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfecv.html","id":null,"dir":"Reference","previous_headings":"","what":"Feature Selection with Recursive Feature Elimination with Cross Validation — mlr_fselectors_rfecv","title":"Feature Selection with Recursive Feature Elimination with Cross Validation — mlr_fselectors_rfecv","text":"Feature selection using Recursive Feature Elimination Cross-Validation (RFE-CV) algorithm. See FSelectorBatchRFE description base algorithm. RFE-CV runs recursive feature elimination iteration cross-validation determine optimal number features. recursive feature elimination run complete dataset optimal number features final feature set size. performance optimal feature set calculated complete data set reported performance final model. works mlr3::Learners can calculate importance scores (see section optional extractors mlr3::Learner).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfecv.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Feature Selection with Recursive Feature Elimination with Cross Validation — mlr_fselectors_rfecv","text":"resampling strategy changed feature selection. resampling strategy passed instance (resampling) used determine optimal number features. Usually, cross-validation strategy used recursive feature elimination run iteration cross-validation. Internally, mlr3::ResamplingCustom used emulate part algorithm. final recursive feature elimination run resampling strategy changed mlr3::ResamplingInsample .e. complete data set used training testing. feature selection terminates optimal number features reached. necessary set termination criterion.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfecv.html","id":"archive","dir":"Reference","previous_headings":"","what":"Archive","title":"Feature Selection with Recursive Feature Elimination with Cross Validation — mlr_fselectors_rfecv","text":"ArchiveBatchFSelect holds following additional columns: \"iteration\" (integer(1)) resampling iteration feature subset evaluated. \"importance\" (numeric()) importance score vector feature subset.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfecv.html","id":"resources","dir":"Reference","previous_headings":"","what":"Resources","title":"Feature Selection with Recursive Feature Elimination with Cross Validation — mlr_fselectors_rfecv","text":"gallery features collection case studies demos optimization. Utilize built-feature importance models Recursive Feature Elimination.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfecv.html","id":"dictionary","dir":"Reference","previous_headings":"","what":"Dictionary","title":"Feature Selection with Recursive Feature Elimination with Cross Validation — mlr_fselectors_rfecv","text":"FSelector can instantiated associated sugar function fs():","code":"fs(\"rfe\")"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfecv.html","id":"control-parameters","dir":"Reference","previous_headings":"","what":"Control Parameters","title":"Feature Selection with Recursive Feature Elimination with Cross Validation — mlr_fselectors_rfecv","text":"n_features integer(1) number features select. default half features selected. feature_fraction double(1) Fraction features retain iteration. default 0.5 retrains half features. feature_number integer(1) Number features remove iteration. subset_sizes integer() Vector number features retain iteration. Must sorted decreasing order. recursive logical(1) TRUE (default), feature importance calculated iteration. parameter feature_fraction, feature_number subset_sizes mutually exclusive.","code":""},{"path":[]},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfecv.html","id":"super-classes","dir":"Reference","previous_headings":"","what":"Super classes","title":"Feature Selection with Recursive Feature Elimination with Cross Validation — mlr_fselectors_rfecv","text":"mlr3fselect::FSelector -> mlr3fselect::FSelectorBatch -> FSelectorBatchRFECV","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfecv.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Feature Selection with Recursive Feature Elimination with Cross Validation — mlr_fselectors_rfecv","text":"mlr3fselect::FSelector$format() mlr3fselect::FSelector$help() mlr3fselect::FSelector$print() mlr3fselect::FSelectorBatch$optimize()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfecv.html","id":"public-methods","dir":"Reference","previous_headings":"","what":"Public methods","title":"Feature Selection with Recursive Feature Elimination with Cross Validation — mlr_fselectors_rfecv","text":"FSelectorBatchRFECV$new() FSelectorBatchRFECV$clone()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfecv.html","id":"method-new-","dir":"Reference","previous_headings":"","what":"Method new()","title":"Feature Selection with Recursive Feature Elimination with Cross Validation — mlr_fselectors_rfecv","text":"Creates new instance R6 class.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfecv.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Feature Selection with Recursive Feature Elimination with Cross Validation — mlr_fselectors_rfecv","text":"","code":"FSelectorBatchRFECV$new()"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfecv.html","id":"method-clone-","dir":"Reference","previous_headings":"","what":"Method clone()","title":"Feature Selection with Recursive Feature Elimination with Cross Validation — mlr_fselectors_rfecv","text":"objects class cloneable method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfecv.html","id":"usage-1","dir":"Reference","previous_headings":"","what":"Usage","title":"Feature Selection with Recursive Feature Elimination with Cross Validation — mlr_fselectors_rfecv","text":"","code":"FSelectorBatchRFECV$clone(deep = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfecv.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Feature Selection with Recursive Feature Elimination with Cross Validation — mlr_fselectors_rfecv","text":"deep Whether make deep clone.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfecv.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Feature Selection with Recursive Feature Elimination with Cross Validation — mlr_fselectors_rfecv","text":"","code":"# Feature Selection # \\donttest{ # retrieve task and load learner task = tsk(\"penguins\") learner = lrn(\"classif.rpart\") # run feature selection on the Palmer Penguins data set instance = fselect( fselector = fs(\"rfecv\"), task = task, learner = learner, resampling = rsmp(\"cv\", folds = 3), measure = msr(\"classif.ce\"), store_models = TRUE ) # best performing feature subset instance$result #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE TRUE FALSE TRUE FALSE FALSE FALSE #> features n_features classif.ce #> #> 1: bill_depth,bill_length,flipper_length 3 0.0377907 # all evaluated feature subsets as.data.table(instance$archive) #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> 2: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> 3: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> 4: FALSE TRUE TRUE TRUE FALSE FALSE FALSE #> 5: TRUE TRUE FALSE TRUE FALSE FALSE FALSE #> 6: TRUE TRUE FALSE TRUE FALSE FALSE FALSE #> 7: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> 8: TRUE TRUE FALSE TRUE FALSE FALSE FALSE #> classif.ce runtime_learners timestamp batch_nr warnings errors #> #> 1: 0.08695652 0.006 2024-07-24 11:01:51 1 0 0 #> 2: 0.03478261 0.005 2024-07-24 11:01:51 1 0 0 #> 3: 0.03508772 0.005 2024-07-24 11:01:51 1 0 0 #> 4: 0.11304348 0.006 2024-07-24 11:01:51 2 0 0 #> 5: 0.03478261 0.005 2024-07-24 11:01:51 2 0 0 #> 6: 0.03508772 0.005 2024-07-24 11:01:51 2 0 0 #> 7: 0.03488372 0.006 2024-07-24 11:01:51 3 0 0 #> 8: 0.03779070 0.005 2024-07-24 11:01:51 4 0 0 #> importance iteration #> #> 1: 95.543823,86.523123,86.157289,83.431536,77.058416, 7.495822,... 1 #> 2: 86.78056,77.56622,62.14872,59.65407,57.93443, 0.00000,... 2 #> 3: 82.17703,79.53680,71.89359,61.50646,48.19125, 0.00000,... 3 #> 4: 94.16064,83.77806,76.82767 1 #> 5: 86.78056,77.56622,62.14872 2 #> 6: 82.17703,79.53680,71.89359 3 #> 7: 124.20793,121.52400,102.74919, 87.26186, 78.61700, 0.00000,... NA #> 8: 124.2079,121.5240,104.2507 NA #> features n_features #> #> 1: bill_depth,bill_length,body_mass,flipper_length,island,sex,... 7 #> 2: bill_depth,bill_length,body_mass,flipper_length,island,sex,... 7 #> 3: bill_depth,bill_length,body_mass,flipper_length,island,sex,... 7 #> 4: bill_length,body_mass,flipper_length 3 #> 5: bill_depth,bill_length,flipper_length 3 #> 6: bill_depth,bill_length,flipper_length 3 #> 7: bill_depth,bill_length,body_mass,flipper_length,island,sex,... 7 #> 8: bill_depth,bill_length,flipper_length 3 #> resample_result #> #> 1: #> 2: #> 3: #> 4: #> 5: #> 6: #> 7: #> 8: # subset the task and fit the final model task$select(instance$result_feature_set) learner$train(task) # }"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_sequential.html","id":null,"dir":"Reference","previous_headings":"","what":"Feature Selection with Sequential Search — mlr_fselectors_sequential","title":"Feature Selection with Sequential Search — mlr_fselectors_sequential","text":"Feature selection using Sequential Search Algorithm.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_sequential.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Feature Selection with Sequential Search — mlr_fselectors_sequential","text":"Sequential forward selection (strategy = fsf) extends feature set iteration feature increases model's performance . Sequential backward selection (strategy = fsb) follows idea starts features removes features set. feature selection terminates min_features max_features reached. necessary set termination criterion.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_sequential.html","id":"dictionary","dir":"Reference","previous_headings":"","what":"Dictionary","title":"Feature Selection with Sequential Search — mlr_fselectors_sequential","text":"FSelector can instantiated associated sugar function fs():","code":"fs(\"sequential\")"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_sequential.html","id":"control-parameters","dir":"Reference","previous_headings":"","what":"Control Parameters","title":"Feature Selection with Sequential Search — mlr_fselectors_sequential","text":"min_features integer(1) Minimum number features. default, 1. max_features integer(1) Maximum number features. default, number features mlr3::Task. strategy character(1) Search method sfs (forward search) sbs (backward search).","code":""},{"path":[]},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_sequential.html","id":"super-classes","dir":"Reference","previous_headings":"","what":"Super classes","title":"Feature Selection with Sequential Search — mlr_fselectors_sequential","text":"mlr3fselect::FSelector -> mlr3fselect::FSelectorBatch -> FSelectorBatchSequential","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_sequential.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Feature Selection with Sequential Search — mlr_fselectors_sequential","text":"mlr3fselect::FSelector$format() mlr3fselect::FSelector$help() mlr3fselect::FSelector$print() mlr3fselect::FSelectorBatch$optimize()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_sequential.html","id":"public-methods","dir":"Reference","previous_headings":"","what":"Public methods","title":"Feature Selection with Sequential Search — mlr_fselectors_sequential","text":"FSelectorBatchSequential$new() FSelectorBatchSequential$optimization_path() FSelectorBatchSequential$clone()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_sequential.html","id":"method-new-","dir":"Reference","previous_headings":"","what":"Method new()","title":"Feature Selection with Sequential Search — mlr_fselectors_sequential","text":"Creates new instance R6 class.`","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_sequential.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Feature Selection with Sequential Search — mlr_fselectors_sequential","text":"","code":"FSelectorBatchSequential$new()"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_sequential.html","id":"method-optimization-path-","dir":"Reference","previous_headings":"","what":"Method optimization_path()","title":"Feature Selection with Sequential Search — mlr_fselectors_sequential","text":"Returns optimization path.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_sequential.html","id":"usage-1","dir":"Reference","previous_headings":"","what":"Usage","title":"Feature Selection with Sequential Search — mlr_fselectors_sequential","text":"","code":"FSelectorBatchSequential$optimization_path(inst, include_uhash = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_sequential.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Feature Selection with Sequential Search — mlr_fselectors_sequential","text":"inst (FSelectInstanceBatchSingleCrit) Instance optimized FSelectorBatchSequential. include_uhash (logical(1)) Include uhash column?","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_sequential.html","id":"returns","dir":"Reference","previous_headings":"","what":"Returns","title":"Feature Selection with Sequential Search — mlr_fselectors_sequential","text":"data.table::data.table()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_sequential.html","id":"method-clone-","dir":"Reference","previous_headings":"","what":"Method clone()","title":"Feature Selection with Sequential Search — mlr_fselectors_sequential","text":"objects class cloneable method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_sequential.html","id":"usage-2","dir":"Reference","previous_headings":"","what":"Usage","title":"Feature Selection with Sequential Search — mlr_fselectors_sequential","text":"","code":"FSelectorBatchSequential$clone(deep = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_sequential.html","id":"arguments-1","dir":"Reference","previous_headings":"","what":"Arguments","title":"Feature Selection with Sequential Search — mlr_fselectors_sequential","text":"deep Whether make deep clone.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_sequential.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Feature Selection with Sequential Search — mlr_fselectors_sequential","text":"","code":"# Feature Selection # \\donttest{ # retrieve task and load learner task = tsk(\"penguins\") learner = lrn(\"classif.rpart\") # run feature selection on the Palmer Penguins data set instance = fselect( fselector = fs(\"sequential\"), task = task, learner = learner, resampling = rsmp(\"holdout\"), measure = msr(\"classif.ce\"), term_evals = 10 ) # best performing feature set instance$result #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: FALSE TRUE FALSE TRUE FALSE FALSE FALSE #> features n_features classif.ce #> #> 1: bill_length,flipper_length 2 0.05217391 # all evaluated feature sets as.data.table(instance$archive) #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE FALSE FALSE FALSE FALSE FALSE FALSE #> 2: FALSE TRUE FALSE FALSE FALSE FALSE FALSE #> 3: FALSE FALSE TRUE FALSE FALSE FALSE FALSE #> 4: FALSE FALSE FALSE TRUE FALSE FALSE FALSE #> 5: FALSE FALSE FALSE FALSE TRUE FALSE FALSE #> 6: FALSE FALSE FALSE FALSE FALSE TRUE FALSE #> 7: FALSE FALSE FALSE FALSE FALSE FALSE TRUE #> 8: TRUE FALSE FALSE TRUE FALSE FALSE FALSE #> 9: FALSE TRUE FALSE TRUE FALSE FALSE FALSE #> 10: FALSE FALSE TRUE TRUE FALSE FALSE FALSE #> 11: FALSE FALSE FALSE TRUE TRUE FALSE FALSE #> 12: FALSE FALSE FALSE TRUE FALSE TRUE FALSE #> 13: FALSE FALSE FALSE TRUE FALSE FALSE TRUE #> classif.ce runtime_learners timestamp batch_nr warnings errors #> #> 1: 0.26086957 0.005 2024-07-24 11:01:52 1 0 0 #> 2: 0.26086957 0.005 2024-07-24 11:01:52 1 0 0 #> 3: 0.30434783 0.004 2024-07-24 11:01:52 1 0 0 #> 4: 0.17391304 0.024 2024-07-24 11:01:52 1 0 0 #> 5: 0.26086957 0.007 2024-07-24 11:01:52 1 0 0 #> 6: 0.58260870 0.005 2024-07-24 11:01:52 1 0 0 #> 7: 0.58260870 0.005 2024-07-24 11:01:52 1 0 0 #> 8: 0.19130435 0.005 2024-07-24 11:01:52 2 0 0 #> 9: 0.05217391 0.005 2024-07-24 11:01:52 2 0 0 #> 10: 0.15652174 0.005 2024-07-24 11:01:52 2 0 0 #> 11: 0.11304348 0.005 2024-07-24 11:01:52 2 0 0 #> 12: 0.16521739 0.004 2024-07-24 11:01:52 2 0 0 #> 13: 0.16521739 0.004 2024-07-24 11:01:52 2 0 0 #> features n_features resample_result #> #> 1: bill_depth 1 #> 2: bill_length 1 #> 3: body_mass 1 #> 4: flipper_length 1 #> 5: island 1 #> 6: sex 1 #> 7: year 1 #> 8: bill_depth,flipper_length 2 #> 9: bill_length,flipper_length 2 #> 10: body_mass,flipper_length 2 #> 11: flipper_length,island 2 #> 12: flipper_length,sex 2 #> 13: flipper_length,year 2 # subset the task and fit the final model task$select(instance$result_feature_set) learner$train(task) # }"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_shadow_variable_search.html","id":null,"dir":"Reference","previous_headings":"","what":"Feature Selection with Shadow Variable Search — mlr_fselectors_shadow_variable_search","title":"Feature Selection with Shadow Variable Search — mlr_fselectors_shadow_variable_search","text":"Feature selection using Shadow Variable Search Algorithm. Shadow variable search creates feature permutated copy stops one selected.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_shadow_variable_search.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Feature Selection with Shadow Variable Search — mlr_fselectors_shadow_variable_search","text":"Thomas J, Hepp T, Mayr , Bischl B (2017). “Probing Sparse Fast Variable Selection Model-Based Boosting.” Computational Mathematical Methods Medicine, 2017, 1–8. doi:10.1155/2017/1421409 . Wu Y, Boos DD, Stefanski LA (2007). “Controlling Variable Selection Addition Pseudovariables.” Journal American Statistical Association, 102(477), 235–243. doi:10.1198/016214506000000843 .","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_shadow_variable_search.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Feature Selection with Shadow Variable Search — mlr_fselectors_shadow_variable_search","text":"feature selection terminates first shadow variable selected. necessary set termination criterion.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_shadow_variable_search.html","id":"resources","dir":"Reference","previous_headings":"","what":"Resources","title":"Feature Selection with Shadow Variable Search — mlr_fselectors_shadow_variable_search","text":"gallery features collection case studies demos optimization. Run feature selection Shadow Variable Search.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_shadow_variable_search.html","id":"dictionary","dir":"Reference","previous_headings":"","what":"Dictionary","title":"Feature Selection with Shadow Variable Search — mlr_fselectors_shadow_variable_search","text":"FSelector can instantiated associated sugar function fs():","code":"fs(\"shadow_variable_search\")"},{"path":[]},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_shadow_variable_search.html","id":"super-classes","dir":"Reference","previous_headings":"","what":"Super classes","title":"Feature Selection with Shadow Variable Search — mlr_fselectors_shadow_variable_search","text":"mlr3fselect::FSelector -> mlr3fselect::FSelectorBatch -> FSelectorBatchShadowVariableSearch","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_shadow_variable_search.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Feature Selection with Shadow Variable Search — mlr_fselectors_shadow_variable_search","text":"mlr3fselect::FSelector$format() mlr3fselect::FSelector$help() mlr3fselect::FSelector$print() mlr3fselect::FSelectorBatch$optimize()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_shadow_variable_search.html","id":"public-methods","dir":"Reference","previous_headings":"","what":"Public methods","title":"Feature Selection with Shadow Variable Search — mlr_fselectors_shadow_variable_search","text":"FSelectorBatchShadowVariableSearch$new() FSelectorBatchShadowVariableSearch$optimization_path() FSelectorBatchShadowVariableSearch$clone()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_shadow_variable_search.html","id":"method-new-","dir":"Reference","previous_headings":"","what":"Method new()","title":"Feature Selection with Shadow Variable Search — mlr_fselectors_shadow_variable_search","text":"Creates new instance R6 class.`","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_shadow_variable_search.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Feature Selection with Shadow Variable Search — mlr_fselectors_shadow_variable_search","text":"","code":"FSelectorBatchShadowVariableSearch$new()"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_shadow_variable_search.html","id":"method-optimization-path-","dir":"Reference","previous_headings":"","what":"Method optimization_path()","title":"Feature Selection with Shadow Variable Search — mlr_fselectors_shadow_variable_search","text":"Returns optimization path.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_shadow_variable_search.html","id":"usage-1","dir":"Reference","previous_headings":"","what":"Usage","title":"Feature Selection with Shadow Variable Search — mlr_fselectors_shadow_variable_search","text":"","code":"FSelectorBatchShadowVariableSearch$optimization_path(inst)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_shadow_variable_search.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Feature Selection with Shadow Variable Search — mlr_fselectors_shadow_variable_search","text":"inst (FSelectInstanceBatchSingleCrit) Instance optimized FSelectorBatchShadowVariableSearch.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_shadow_variable_search.html","id":"returns","dir":"Reference","previous_headings":"","what":"Returns","title":"Feature Selection with Shadow Variable Search — mlr_fselectors_shadow_variable_search","text":"data.table::data.table","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_shadow_variable_search.html","id":"method-clone-","dir":"Reference","previous_headings":"","what":"Method clone()","title":"Feature Selection with Shadow Variable Search — mlr_fselectors_shadow_variable_search","text":"objects class cloneable method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_shadow_variable_search.html","id":"usage-2","dir":"Reference","previous_headings":"","what":"Usage","title":"Feature Selection with Shadow Variable Search — mlr_fselectors_shadow_variable_search","text":"","code":"FSelectorBatchShadowVariableSearch$clone(deep = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_shadow_variable_search.html","id":"arguments-1","dir":"Reference","previous_headings":"","what":"Arguments","title":"Feature Selection with Shadow Variable Search — mlr_fselectors_shadow_variable_search","text":"deep Whether make deep clone.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_shadow_variable_search.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Feature Selection with Shadow Variable Search — mlr_fselectors_shadow_variable_search","text":"","code":"# Feature Selection # \\donttest{ # retrieve task and load learner task = tsk(\"penguins\") learner = lrn(\"classif.rpart\") # run feature selection on the Palmer Penguins data set instance = fselect( fselector = fs(\"shadow_variable_search\"), task = task, learner = learner, resampling = rsmp(\"holdout\"), measure = msr(\"classif.ce\"), ) # best performing feature subset instance$result #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: FALSE TRUE FALSE TRUE TRUE FALSE FALSE #> features n_features classif.ce #> #> 1: bill_length,flipper_length,island 3 0.06956522 # all evaluated feature subsets as.data.table(instance$archive) #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE FALSE FALSE FALSE FALSE FALSE FALSE #> 2: FALSE TRUE FALSE FALSE FALSE FALSE FALSE #> 3: FALSE FALSE TRUE FALSE FALSE FALSE FALSE #> 4: FALSE FALSE FALSE TRUE FALSE FALSE FALSE #> 5: FALSE FALSE FALSE FALSE TRUE FALSE FALSE #> 6: FALSE FALSE FALSE FALSE FALSE TRUE FALSE #> 7: FALSE FALSE FALSE FALSE FALSE FALSE TRUE #> 8: FALSE FALSE FALSE FALSE FALSE FALSE FALSE #> 9: FALSE FALSE FALSE FALSE FALSE FALSE FALSE #> 10: FALSE FALSE FALSE FALSE FALSE FALSE FALSE #> 11: FALSE FALSE FALSE FALSE FALSE FALSE FALSE #> 12: FALSE FALSE FALSE FALSE FALSE FALSE FALSE #> 13: FALSE FALSE FALSE FALSE FALSE FALSE FALSE #> 14: FALSE FALSE FALSE FALSE FALSE FALSE FALSE #> 15: TRUE FALSE FALSE TRUE FALSE FALSE FALSE #> 16: FALSE TRUE FALSE TRUE FALSE FALSE FALSE #> 17: FALSE FALSE TRUE TRUE FALSE FALSE FALSE #> 18: FALSE FALSE FALSE TRUE TRUE FALSE FALSE #> 19: FALSE FALSE FALSE TRUE FALSE TRUE FALSE #> 20: FALSE FALSE FALSE TRUE FALSE FALSE TRUE #> 21: FALSE FALSE FALSE TRUE FALSE FALSE FALSE #> 22: FALSE FALSE FALSE TRUE FALSE FALSE FALSE #> 23: FALSE FALSE FALSE TRUE FALSE FALSE FALSE #> 24: FALSE FALSE FALSE TRUE FALSE FALSE FALSE #> 25: FALSE FALSE FALSE TRUE FALSE FALSE FALSE #> 26: FALSE FALSE FALSE TRUE FALSE FALSE FALSE #> 27: FALSE FALSE FALSE TRUE FALSE FALSE FALSE #> 28: TRUE TRUE FALSE TRUE FALSE FALSE FALSE #> 29: FALSE TRUE TRUE TRUE FALSE FALSE FALSE #> 30: FALSE TRUE FALSE TRUE TRUE FALSE FALSE #> 31: FALSE TRUE FALSE TRUE FALSE TRUE FALSE #> 32: FALSE TRUE FALSE TRUE FALSE FALSE TRUE #> 33: FALSE TRUE FALSE TRUE FALSE FALSE FALSE #> 34: FALSE TRUE FALSE TRUE FALSE FALSE FALSE #> 35: FALSE TRUE FALSE TRUE FALSE FALSE FALSE #> 36: FALSE TRUE FALSE TRUE FALSE FALSE FALSE #> 37: FALSE TRUE FALSE TRUE FALSE FALSE FALSE #> 38: FALSE TRUE FALSE TRUE FALSE FALSE FALSE #> 39: FALSE TRUE FALSE TRUE FALSE FALSE FALSE #> 40: TRUE TRUE FALSE TRUE TRUE FALSE FALSE #> 41: FALSE TRUE TRUE TRUE TRUE FALSE FALSE #> 42: FALSE TRUE FALSE TRUE TRUE TRUE FALSE #> 43: FALSE TRUE FALSE TRUE TRUE FALSE TRUE #> 44: FALSE TRUE FALSE TRUE TRUE FALSE FALSE #> 45: FALSE TRUE FALSE TRUE TRUE FALSE FALSE #> 46: FALSE TRUE FALSE TRUE TRUE FALSE FALSE #> 47: FALSE TRUE FALSE TRUE TRUE FALSE FALSE #> 48: FALSE TRUE FALSE TRUE TRUE FALSE FALSE #> 49: FALSE TRUE FALSE TRUE TRUE FALSE FALSE #> 50: FALSE TRUE FALSE TRUE TRUE FALSE FALSE #> bill_depth bill_length body_mass flipper_length island sex year #> classif.ce runtime_learners timestamp batch_nr #> #> 1: 0.32173913 0.014 2024-07-24 11:01:53 1 #> 2: 0.24347826 0.013 2024-07-24 11:01:53 1 #> 3: 0.31304348 0.012 2024-07-24 11:01:53 1 #> 4: 0.19130435 0.012 2024-07-24 11:01:53 1 #> 5: 0.28695652 0.013 2024-07-24 11:01:53 1 #> 6: 0.55652174 0.012 2024-07-24 11:01:53 1 #> 7: 0.55652174 0.011 2024-07-24 11:01:53 1 #> 8: 0.60000000 0.010 2024-07-24 11:01:53 1 #> 9: 0.61739130 0.010 2024-07-24 11:01:53 1 #> 10: 0.60000000 0.009 2024-07-24 11:01:53 1 #> 11: 0.57391304 0.011 2024-07-24 11:01:53 1 #> 12: 0.58260870 0.009 2024-07-24 11:01:53 1 #> 13: 0.55652174 0.010 2024-07-24 11:01:53 1 #> 14: 0.55652174 0.010 2024-07-24 11:01:53 1 #> 15: 0.21739130 0.014 2024-07-24 11:01:54 2 #> 16: 0.07826087 0.012 2024-07-24 11:01:54 2 #> 17: 0.19130435 0.012 2024-07-24 11:01:54 2 #> 18: 0.13043478 0.012 2024-07-24 11:01:54 2 #> 19: 0.25217391 0.012 2024-07-24 11:01:54 2 #> 20: 0.19130435 0.012 2024-07-24 11:01:54 2 #> 21: 0.19130435 0.013 2024-07-24 11:01:54 2 #> 22: 0.18260870 0.013 2024-07-24 11:01:54 2 #> 23: 0.20000000 0.012 2024-07-24 11:01:54 2 #> 24: 0.20869565 0.012 2024-07-24 11:01:54 2 #> 25: 0.19130435 0.012 2024-07-24 11:01:54 2 #> 26: 0.19130435 0.012 2024-07-24 11:01:54 2 #> 27: 0.19130435 0.013 2024-07-24 11:01:54 2 #> 28: 0.07826087 0.013 2024-07-24 11:01:54 3 #> 29: 0.07826087 0.012 2024-07-24 11:01:54 3 #> 30: 0.06956522 0.012 2024-07-24 11:01:54 3 #> 31: 0.07826087 0.012 2024-07-24 11:01:54 3 #> 32: 0.07826087 0.012 2024-07-24 11:01:54 3 #> 33: 0.07826087 0.012 2024-07-24 11:01:54 3 #> 34: 0.07826087 0.012 2024-07-24 11:01:54 3 #> 35: 0.07826087 0.012 2024-07-24 11:01:54 3 #> 36: 0.07826087 0.012 2024-07-24 11:01:54 3 #> 37: 0.07826087 0.013 2024-07-24 11:01:54 3 #> 38: 0.07826087 0.012 2024-07-24 11:01:54 3 #> 39: 0.07826087 0.018 2024-07-24 11:01:54 3 #> 40: 0.06956522 0.013 2024-07-24 11:01:54 4 #> 41: 0.06956522 0.013 2024-07-24 11:01:54 4 #> 42: 0.06956522 0.013 2024-07-24 11:01:54 4 #> 43: 0.06956522 0.012 2024-07-24 11:01:54 4 #> 44: 0.06956522 0.013 2024-07-24 11:01:54 4 #> 45: 0.06956522 0.012 2024-07-24 11:01:54 4 #> 46: 0.06956522 0.012 2024-07-24 11:01:54 4 #> 47: 0.06956522 0.012 2024-07-24 11:01:54 4 #> 48: 0.06956522 0.013 2024-07-24 11:01:54 4 #> 49: 0.06956522 0.036 2024-07-24 11:01:54 4 #> 50: 0.06956522 0.018 2024-07-24 11:01:54 4 #> classif.ce runtime_learners timestamp batch_nr #> permuted__bill_depth permuted__bill_length permuted__body_mass #> #> 1: FALSE FALSE FALSE #> 2: FALSE FALSE FALSE #> 3: FALSE FALSE FALSE #> 4: FALSE FALSE FALSE #> 5: FALSE FALSE FALSE #> 6: FALSE FALSE FALSE #> 7: FALSE FALSE FALSE #> 8: TRUE FALSE FALSE #> 9: FALSE TRUE FALSE #> 10: FALSE FALSE TRUE #> 11: FALSE FALSE FALSE #> 12: FALSE FALSE FALSE #> 13: FALSE FALSE FALSE #> 14: FALSE FALSE FALSE #> 15: FALSE FALSE FALSE #> 16: FALSE FALSE FALSE #> 17: FALSE FALSE FALSE #> 18: FALSE FALSE FALSE #> 19: FALSE FALSE FALSE #> 20: FALSE FALSE FALSE #> 21: TRUE FALSE FALSE #> 22: FALSE TRUE FALSE #> 23: FALSE FALSE TRUE #> 24: FALSE FALSE FALSE #> 25: FALSE FALSE FALSE #> 26: FALSE FALSE FALSE #> 27: FALSE FALSE FALSE #> 28: FALSE FALSE FALSE #> 29: FALSE FALSE FALSE #> 30: FALSE FALSE FALSE #> 31: FALSE FALSE FALSE #> 32: FALSE FALSE FALSE #> 33: TRUE FALSE FALSE #> 34: FALSE TRUE FALSE #> 35: FALSE FALSE TRUE #> 36: FALSE FALSE FALSE #> 37: FALSE FALSE FALSE #> 38: FALSE FALSE FALSE #> 39: FALSE FALSE FALSE #> 40: FALSE FALSE FALSE #> 41: FALSE FALSE FALSE #> 42: FALSE FALSE FALSE #> 43: FALSE FALSE FALSE #> 44: TRUE FALSE FALSE #> 45: FALSE TRUE FALSE #> 46: FALSE FALSE TRUE #> 47: FALSE FALSE FALSE #> 48: FALSE FALSE FALSE #> 49: FALSE FALSE FALSE #> 50: FALSE FALSE FALSE #> permuted__bill_depth permuted__bill_length permuted__body_mass #> permuted__flipper_length permuted__island permuted__sex permuted__year #> #> 1: FALSE FALSE FALSE FALSE #> 2: FALSE FALSE FALSE FALSE #> 3: FALSE FALSE FALSE FALSE #> 4: FALSE FALSE FALSE FALSE #> 5: FALSE FALSE FALSE FALSE #> 6: FALSE FALSE FALSE FALSE #> 7: FALSE FALSE FALSE FALSE #> 8: FALSE FALSE FALSE FALSE #> 9: FALSE FALSE FALSE FALSE #> 10: FALSE FALSE FALSE FALSE #> 11: TRUE FALSE FALSE FALSE #> 12: FALSE TRUE FALSE FALSE #> 13: FALSE FALSE TRUE FALSE #> 14: FALSE FALSE FALSE TRUE #> 15: FALSE FALSE FALSE FALSE #> 16: FALSE FALSE FALSE FALSE #> 17: FALSE FALSE FALSE FALSE #> 18: FALSE FALSE FALSE FALSE #> 19: FALSE FALSE FALSE FALSE #> 20: FALSE FALSE FALSE FALSE #> 21: FALSE FALSE FALSE FALSE #> 22: FALSE FALSE FALSE FALSE #> 23: FALSE FALSE FALSE FALSE #> 24: TRUE FALSE FALSE FALSE #> 25: FALSE TRUE FALSE FALSE #> 26: FALSE FALSE TRUE FALSE #> 27: FALSE FALSE FALSE TRUE #> 28: FALSE FALSE FALSE FALSE #> 29: FALSE FALSE FALSE FALSE #> 30: FALSE FALSE FALSE FALSE #> 31: FALSE FALSE FALSE FALSE #> 32: FALSE FALSE FALSE FALSE #> 33: FALSE FALSE FALSE FALSE #> 34: FALSE FALSE FALSE FALSE #> 35: FALSE FALSE FALSE FALSE #> 36: TRUE FALSE FALSE FALSE #> 37: FALSE TRUE FALSE FALSE #> 38: FALSE FALSE TRUE FALSE #> 39: FALSE FALSE FALSE TRUE #> 40: FALSE FALSE FALSE FALSE #> 41: FALSE FALSE FALSE FALSE #> 42: FALSE FALSE FALSE FALSE #> 43: FALSE FALSE FALSE FALSE #> 44: FALSE FALSE FALSE FALSE #> 45: FALSE FALSE FALSE FALSE #> 46: FALSE FALSE FALSE FALSE #> 47: TRUE FALSE FALSE FALSE #> 48: FALSE TRUE FALSE FALSE #> 49: FALSE FALSE TRUE FALSE #> 50: FALSE FALSE FALSE TRUE #> permuted__flipper_length permuted__island permuted__sex permuted__year #> warnings errors features n_features #> #> 1: 0 0 bill_depth 1 #> 2: 0 0 bill_length 1 #> 3: 0 0 body_mass 1 #> 4: 0 0 flipper_length 1 #> 5: 0 0 island 1 #> 6: 0 0 sex 1 #> 7: 0 0 year 1 #> 8: 0 0 0 #> 9: 0 0 0 #> 10: 0 0 0 #> 11: 0 0 0 #> 12: 0 0 0 #> 13: 0 0 0 #> 14: 0 0 0 #> 15: 0 0 bill_depth,flipper_length 2 #> 16: 0 0 bill_length,flipper_length 2 #> 17: 0 0 body_mass,flipper_length 2 #> 18: 0 0 flipper_length,island 2 #> 19: 0 0 flipper_length,sex 2 #> 20: 0 0 flipper_length,year 2 #> 21: 0 0 flipper_length 1 #> 22: 0 0 flipper_length 1 #> 23: 0 0 flipper_length 1 #> 24: 0 0 flipper_length 1 #> 25: 0 0 flipper_length 1 #> 26: 0 0 flipper_length 1 #> 27: 0 0 flipper_length 1 #> 28: 0 0 bill_depth,bill_length,flipper_length 3 #> 29: 0 0 bill_length,body_mass,flipper_length 3 #> 30: 0 0 bill_length,flipper_length,island 3 #> 31: 0 0 bill_length,flipper_length,sex 3 #> 32: 0 0 bill_length,flipper_length,year 3 #> 33: 0 0 bill_length,flipper_length 2 #> 34: 0 0 bill_length,flipper_length 2 #> 35: 0 0 bill_length,flipper_length 2 #> 36: 0 0 bill_length,flipper_length 2 #> 37: 0 0 bill_length,flipper_length 2 #> 38: 0 0 bill_length,flipper_length 2 #> 39: 0 0 bill_length,flipper_length 2 #> 40: 0 0 bill_depth,bill_length,flipper_length,island 4 #> 41: 0 0 bill_length,body_mass,flipper_length,island 4 #> 42: 0 0 bill_length,flipper_length,island,sex 4 #> 43: 0 0 bill_length,flipper_length,island,year 4 #> 44: 0 0 bill_length,flipper_length,island 3 #> 45: 0 0 bill_length,flipper_length,island 3 #> 46: 0 0 bill_length,flipper_length,island 3 #> 47: 0 0 bill_length,flipper_length,island 3 #> 48: 0 0 bill_length,flipper_length,island 3 #> 49: 0 0 bill_length,flipper_length,island 3 #> 50: 0 0 bill_length,flipper_length,island 3 #> warnings errors features n_features #> resample_result #> #> 1: #> 2: #> 3: #> 4: #> 5: #> 6: #> 7: #> 8: #> 9: #> 10: #> 11: #> 12: #> 13: #> 14: #> 15: #> 16: #> 17: #> 18: #> 19: #> 20: #> 21: #> 22: #> 23: #> 24: #> 25: #> 26: #> 27: #> 28: #> 29: #> 30: #> 31: #> 32: #> 33: #> 34: #> 35: #> 36: #> 37: #> 38: #> 39: #> 40: #> 41: #> 42: #> 43: #> 44: #> 45: #> 46: #> 47: #> 48: #> 49: #> 50: #> resample_result # subset the task and fit the final model task$select(instance$result_feature_set) learner$train(task) # }"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/reexports.html","id":null,"dir":"Reference","previous_headings":"","what":"Objects exported from other packages — reexports","title":"Objects exported from other packages — reexports","text":"objects imported packages. Follow links see documentation. bbotk mlr_terminators, trm, trms mlr3misc clbk, clbks, mlr_callbacks","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/news/index.html","id":"mlr3fselect-development-version","dir":"Changelog","previous_headings":"","what":"mlr3fselect (development version)","title":"mlr3fselect (development version)","text":"fix: Delete intermediate BenchmarkResult ObjectiveFSelectBatch optimization.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/news/index.html","id":"mlr3fselect-100","dir":"Changelog","previous_headings":"","what":"mlr3fselect 1.0.0","title":"mlr3fselect 1.0.0","text":"CRAN release: 2024-06-29 feat: Add ensemble feature selection function ensemble_fselect(). BREAKING CHANGE: FSelector class FSelectorBatch now. BREAKING CHANGE: FSelectInstanceSingleCrit FSelectInstanceMultiCrit classes FSelectInstanceBatchSingleCrit FSelectInstanceBatchMultiCrit now. BREAKING CHANGE: CallbackFSelect class CallbackBatchFSelect now. BREAKING CHANGE: ContextEval class ContextBatchFSelect now.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/news/index.html","id":"mlr3fselect-0120","dir":"Changelog","previous_headings":"","what":"mlr3fselect 0.12.0","title":"mlr3fselect 0.12.0","text":"CRAN release: 2024-03-09 feat: Add number features instance$result. feat: Add ties_method options \"least_features\" \"random\" ArchiveBatchFSelect$best(). refactor: Optimize runtime ArchiveBatchFSelect$best() method. feat: Add importance scores result FSelectorRFE. feat: Add number features .data.table.ArchiveBatchFSelect(). feat: Features can always included always_include column role. fix: Add $phash() method AutoFSelector. fix: Include FSelector hash AutoFSelector. refactor: Change default batch size FSelectorBatchRandomSearch 10. feat: Add batch_size parameter FSelectorBatchExhaustiveSearch reduce memory consumption. compatibility: Work new paradox version 1.0.0","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/news/index.html","id":"mlr3fselect-0110","dir":"Changelog","previous_headings":"","what":"mlr3fselect 0.11.0","title":"mlr3fselect 0.11.0","text":"CRAN release: 2023-03-02 BREAKING CHANGE: method parameter fselect(), fselect_nested() auto_fselector() renamed fselector. FSelector objects accepted now. Arguments fselector passed ... anymore. BREAKING CHANGE: fselect parameter FSelector moved first position achieve consistency functions. docs: Update resources sections. docs: Add list default measures.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/news/index.html","id":"mlr3fselect-0100","dir":"Changelog","previous_headings":"","what":"mlr3fselect 0.10.0","title":"mlr3fselect 0.10.0","text":"CRAN release: 2023-02-21 feat: Add callback mlr3fselect.svm_rfe run recursive feature elimination linear support vector machines. refactor: importance scores FSelectorRFE now aggregated rank instead averaging . feat: Add FSelectorRFECV optimizer run recursive feature elimination cross-validation. refactor: FSelectorRFE works without store_models = TRUE now. feat: .data.table.ArchiveBatchFSelect() function additionally returns character vector selected features row. refactor: Add callbacks argument fsi() function.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/news/index.html","id":"mlr3fselect-091","dir":"Changelog","previous_headings":"","what":"mlr3fselect 0.9.1","title":"mlr3fselect 0.9.1","text":"CRAN release: 2023-01-26 refactor: Remove internal use mlr3pipelines. fix: Feature selection measures require importance oob error works now.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/news/index.html","id":"mlr3fselect-090","dir":"Changelog","previous_headings":"","what":"mlr3fselect 0.9.0","title":"mlr3fselect 0.9.0","text":"CRAN release: 2022-12-21 fix: Add genalg required packages FSelectorBatchGeneticSearch. feat: Add new callback backups benchmark result disk batch. feat: Create custom callbacks callback_batch_fselect() function.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/news/index.html","id":"mlr3fselect-080","dir":"Changelog","previous_headings":"","what":"mlr3fselect 0.8.0","title":"mlr3fselect 0.8.0","text":"CRAN release: 2022-11-16 refactor: FSelectorRFE throws error learner support $importance() method. refactor: AutoFSelector stores instance benchmark result store_models = TRUE. refactor: AutoFSelector stores instance store_benchmark_result = TRUE. feat: Add missing parameters AutoFSelector auto_fselect(). feat: Add fsi() function create FSelectInstanceBatchSingleCrit FSelectInstanceBatchMultiCrit. refactor: Remove unnest option .data.table.ArchiveBatchFSelect() function.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/news/index.html","id":"mlr3fselect-072","dir":"Changelog","previous_headings":"","what":"mlr3fselect 0.7.2","title":"mlr3fselect 0.7.2","text":"CRAN release: 2022-08-25 docs: Re-generate rd files valid html.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/news/index.html","id":"mlr3fselect-071","dir":"Changelog","previous_headings":"","what":"mlr3fselect 0.7.1","title":"mlr3fselect 0.7.1","text":"CRAN release: 2022-05-03 feat: FSelector objects field $id now.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/news/index.html","id":"mlr3fselect-070","dir":"Changelog","previous_headings":"","what":"mlr3fselect 0.7.0","title":"mlr3fselect 0.7.0","text":"CRAN release: 2022-04-08 feat: Allow pass FSelector objects method fselect() auto_fselector(). feat: Added $label FSelectors. docs: New examples fselect() function. feat: $help() method opens manual page FSelector. feat: Added .data.table.DictionaryFSelector function. feat: Added min_features parameter FSelectorBatchSequential.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/news/index.html","id":"mlr3fselect-061","dir":"Changelog","previous_headings":"","what":"mlr3fselect 0.6.1","title":"mlr3fselect 0.6.1","text":"CRAN release: 2022-01-20 Add store_models flag fselect(). Remove store_x_domain flag.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/news/index.html","id":"mlr3fselect-060","dir":"Changelog","previous_headings":"","what":"mlr3fselect 0.6.0","title":"mlr3fselect 0.6.0","text":"CRAN release: 2021-09-13 Adds AutoFSelector$base_learner() method extract base learner nested learner objects. Adds fselect(), auto_fselector() fselect_nested() sugar functions. Adds extract_inner_fselect_results() extract_inner_fselect_archives() helper function extract inner feature selection results archives.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/news/index.html","id":"mlr3fselect-051","dir":"Changelog","previous_headings":"","what":"mlr3fselect 0.5.1","title":"mlr3fselect 0.5.1","text":"CRAN release: 2021-03-09 Remove x_domain column archive.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/news/index.html","id":"mlr3fselect-050","dir":"Changelog","previous_headings":"","what":"mlr3fselect 0.5.0","title":"mlr3fselect 0.5.0","text":"CRAN release: 2021-01-24 FSelectorRFE stores importance values evaluated feature set archive. ArchiveBatchFSelect$data public field now.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/news/index.html","id":"mlr3fselect-041","dir":"Changelog","previous_headings":"","what":"mlr3fselect 0.4.1","title":"mlr3fselect 0.4.1","text":"CRAN release: 2020-10-30 Fix bug AutoFSelector$predict()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/news/index.html","id":"mlr3fselect-040","dir":"Changelog","previous_headings":"","what":"mlr3fselect 0.4.0","title":"mlr3fselect 0.4.0","text":"CRAN release: 2020-10-22 Compact -memory representation R6 objects save space saving mlr3 objects via saveRDS(), serialize() etc. FSelectorRFE supports fraction features retain iteration (feature_fraction), number features remove iteration (feature_number) vector number features retain iteration (subset_sizes). AutoFSelect renamed AutoFSelector. retrieve inner feature selection results nested resampling, .data.table(rr)$learner[[1]]$fselect_result must used now. Option control store_benchmark_result, store_models check_values AutoFSelector. store_fselect_instance must set parameter initialization. Adds FSelectorBatchGeneticSearch. Fixes check_values flag FSelectInstanceBatchSingleCrit FSelectInstanceBatchMultiCrit. Removed dependency orphaned package bibtex. PipeOpSelect internally used task subsetting.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/news/index.html","id":"mlr3fselect-030","dir":"Changelog","previous_headings":"","what":"mlr3fselect 0.3.0","title":"mlr3fselect 0.3.0","text":"CRAN release: 2020-09-22 Archive ArchiveBatchFSelect now stores benchmark result $benchmark_result. change removed resample results archive can still accessed via benchmark result.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/news/index.html","id":"mlr3fselect-021","dir":"Changelog","previous_headings":"","what":"mlr3fselect 0.2.1","title":"mlr3fselect 0.2.1","text":"CRAN release: 2020-09-10 Warning message external package feature selection installed.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/news/index.html","id":"mlr3fselect-020","dir":"Changelog","previous_headings":"","what":"mlr3fselect 0.2.0","title":"mlr3fselect 0.2.0","text":"CRAN release: 2020-08-23 Initial CRAN release.","code":""}] +[{"path":"https://mlr3fselect.mlr-org.com/dev/authors.html","id":null,"dir":"","previous_headings":"","what":"Authors","title":"Authors and Citation","text":"Marc Becker. Author, maintainer. Patrick Schratz. Author. Michel Lang. Author. Bernd Bischl. Author. John Zobolas. Author.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/authors.html","id":"citation","dir":"","previous_headings":"","what":"Citation","title":"Authors and Citation","text":"Becker M, Schratz P, Lang M, Bischl B, Zobolas J (2024). mlr3fselect: Feature Selection 'mlr3'. R package version 1.0.0.9000, https://github.com/mlr-org/mlr3fselect, https://mlr3fselect.mlr-org.com.","code":"@Manual{, title = {mlr3fselect: Feature Selection for 'mlr3'}, author = {Marc Becker and Patrick Schratz and Michel Lang and Bernd Bischl and John Zobolas}, year = {2024}, note = {R package version 1.0.0.9000, https://github.com/mlr-org/mlr3fselect}, url = {https://mlr3fselect.mlr-org.com}, }"},{"path":"https://mlr3fselect.mlr-org.com/dev/index.html","id":"mlr3fselect-","dir":"","previous_headings":"","what":"Feature Selection for mlr3","title":"Feature Selection for mlr3","text":"Package website: release | dev mlr3fselect feature selection package mlr3 ecosystem. selects optimal feature set mlr3 learner. package works several optimization algorithms e.g. Random Search, Recursive Feature Elimination, Genetic Search. Moreover, can automatically optimize learners estimate performance optimized feature sets nested resampling. package built optimization framework bbotk.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/index.html","id":"resources","dir":"","previous_headings":"","what":"Resources","title":"Feature Selection for mlr3","text":"several section feature selection mlr3book. Getting started wrapper feature selection. sequential forward selection Palmer Penguins data set. Optimize multiple performance measures. Estimate Model Performance nested resampling. gallery features collection case studies demos optimization. Utilize built-feature importance models Recursive Feature Elimination. Run feature selection Shadow Variable Search. Feature Selection Titanic data set. cheatsheet summarizes important functions mlr3fselect.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/index.html","id":"installation","dir":"","previous_headings":"","what":"Installation","title":"Feature Selection for mlr3","text":"Install last release CRAN: Install development version GitHub:","code":"install.packages(\"mlr3fselect\") remotes::install_github(\"mlr-org/mlr3fselect\")"},{"path":"https://mlr3fselect.mlr-org.com/dev/index.html","id":"example","dir":"","previous_headings":"","what":"Example","title":"Feature Selection for mlr3","text":"run feature selection support vector machine Spam data set. construct instance fsi() function. instance describes optimization problem. select simple random search optimization algorithm. start feature selection, simply pass instance fselector. fselector writes best hyperparameter configuration instance. corresponding measured performance. archive contains evaluated hyperparameter configurations. fit final model optimized feature set make predictions new data.","code":"library(\"mlr3verse\") tsk(\"spam\") ## (4601 x 58): HP Spam Detection ## * Target: type ## * Properties: twoclass ## * Features (57): ## - dbl (57): address, addresses, all, business, capitalAve, capitalLong, capitalTotal, ## charDollar, charExclamation, charHash, charRoundbracket, charSemicolon, ## charSquarebracket, conference, credit, cs, data, direct, edu, email, font, free, ## george, hp, hpl, internet, lab, labs, mail, make, meeting, money, num000, num1999, ## num3d, num415, num650, num85, num857, order, original, our, over, parts, people, pm, ## project, re, receive, remove, report, table, technology, telnet, will, you, your instance = fsi( task = tsk(\"spam\"), learner = lrn(\"classif.svm\", type = \"C-classification\"), resampling = rsmp(\"cv\", folds = 3), measures = msr(\"classif.ce\"), terminator = trm(\"evals\", n_evals = 20) ) instance ## ## * State: Not optimized ## * Objective: ## * Terminator: fselector = fs(\"random_search\", batch_size = 5) fselector ## : Random Search ## * Parameters: batch_size=5 ## * Properties: single-crit, multi-crit ## * Packages: mlr3fselect fselector$optimize(instance) instance$result_feature_set ## [1] \"address\" \"addresses\" \"all\" \"business\" ## [5] \"capitalAve\" \"capitalLong\" \"capitalTotal\" \"charDollar\" ## [9] \"charExclamation\" \"charHash\" \"charRoundbracket\" \"charSemicolon\" ## [13] \"charSquarebracket\" \"conference\" \"credit\" \"cs\" ## [17] \"data\" \"direct\" \"edu\" \"email\" ## [21] \"font\" \"free\" \"george\" \"hp\" ## [25] \"internet\" \"lab\" \"labs\" \"mail\" ## [29] \"make\" \"meeting\" \"money\" \"num000\" ## [33] \"num1999\" \"num3d\" \"num415\" \"num650\" ## [37] \"num85\" \"num857\" \"order\" \"our\" ## [41] \"parts\" \"people\" \"pm\" \"project\" ## [45] \"re\" \"receive\" \"remove\" \"report\" ## [49] \"table\" \"technology\" \"telnet\" \"will\" ## [53] \"you\" \"your\" instance$result_y ## classif.ce ## 0.07042005 as.data.table(instance$archive) ## address addresses all business capitalAve capitalLong capitalTotal charDollar charExclamation ## 1: TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE ## 2: TRUE TRUE TRUE FALSE FALSE TRUE TRUE TRUE TRUE ## 3: TRUE TRUE FALSE FALSE TRUE TRUE TRUE TRUE TRUE ## 4: TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE ## 5: FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE ## --- ## 16: FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE ## 17: FALSE FALSE FALSE TRUE TRUE TRUE FALSE FALSE TRUE ## 18: FALSE FALSE TRUE TRUE FALSE FALSE FALSE TRUE FALSE ## 19: TRUE TRUE TRUE TRUE FALSE TRUE TRUE TRUE TRUE ## 20: TRUE FALSE TRUE FALSE FALSE TRUE FALSE TRUE FALSE ## 56 variables not shown: [charHash, charRoundbracket, charSemicolon, charSquarebracket, conference, credit, cs, data, direct, edu, ...] task = tsk(\"spam\") learner = lrn(\"classif.svm\", type = \"C-classification\") task$select(instance$result_feature_set) learner$train(task)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":null,"dir":"Reference","previous_headings":"","what":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"ArchiveBatchFSelect stores evaluated feature sets performance scores.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"ArchiveBatchFSelect container around data.table::data.table(). row corresponds single evaluation feature set. See section Data Structure information. archive stores additionally mlr3::BenchmarkResult ($benchmark_result) records resampling experiments. experiment corresponds single evaluation feature set. table ($data) benchmark result ($benchmark_result) linked uhash column. archive passed .data.table(), joined automatically.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"data-structure","dir":"Reference","previous_headings":"","what":"Data structure","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"table ($data) following columns: One column feature task ($search_space). One column performance measure ($codomain). runtime_learners (numeric(1)) Sum training predict times logged learners per mlr3::ResampleResult / evaluation. include potential overhead time. timestamp (POSIXct) Time stamp evaluation logged archive. batch_nr (integer(1)) Feature sets evaluated batches. batch unique batch number. uhash (character(1)) Connects feature set resampling experiment stored mlr3::BenchmarkResult.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"analysis","dir":"Reference","previous_headings":"","what":"Analysis","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"analyzing feature selection results, recommended pass archive .data.table(). returned data table joined benchmark result adds mlr3::ResampleResult feature set. archive provides various getters (e.g. $learners()) ease access. getters extract position () unique hash (uhash). complete list getters see methods section. benchmark result ($benchmark_result) allows score feature sets different measure. Alternatively, measures can supplied .data.table().","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"s-methods","dir":"Reference","previous_headings":"","what":"S3 Methods","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":".data.table.ArchiveBatchFSelect(x, exclude_columns = \"uhash\", measures = NULL) Returns tabular view evaluated feature sets. ArchiveBatchFSelect -> data.table::data.table() x (ArchiveBatchFSelect) exclude_columns (character()) Exclude columns table. Set NULL column excluded. measures (list mlr3::Measure) Score feature sets additional measures.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"super-classes","dir":"Reference","previous_headings":"","what":"Super classes","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"bbotk::Archive -> bbotk::ArchiveBatch -> ArchiveBatchFSelect","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"public-fields","dir":"Reference","previous_headings":"","what":"Public fields","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"benchmark_result (mlr3::BenchmarkResult) Benchmark result.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"active-bindings","dir":"Reference","previous_headings":"","what":"Active bindings","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"ties_method (character(1)) Method handle ties.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"bbotk::Archive$format() bbotk::Archive$help() bbotk::ArchiveBatch$add_evals() bbotk::ArchiveBatch$clear() bbotk::ArchiveBatch$nds_selection()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"public-methods","dir":"Reference","previous_headings":"","what":"Public methods","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"ArchiveBatchFSelect$new() ArchiveBatchFSelect$learner() ArchiveBatchFSelect$learners() ArchiveBatchFSelect$predictions() ArchiveBatchFSelect$resample_result() ArchiveBatchFSelect$print() ArchiveBatchFSelect$best() ArchiveBatchFSelect$clone()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"method-new-","dir":"Reference","previous_headings":"","what":"Method new()","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"Creates new instance R6 class.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"","code":"ArchiveBatchFSelect$new( search_space, codomain, check_values = TRUE, ties_method = \"least_features\" )"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"search_space (paradox::ParamSet) Search space. Internally created provided mlr3::Task instance. codomain (bbotk::Codomain) Specifies codomain objective function .e. set performance measures. Internally created provided mlr3::Measures instance. check_values (logical(1)) TRUE (default), hyperparameter configurations check validity. ties_method (character(1)) method break ties selecting sets optimizing selecting best set. Can \"least_features\" \"random\". option \"least_features\" (default) selects feature set least features. multiple best feature sets number features, one selected randomly. random method returns random feature set best feature sets. Ignored multiple measures used.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"method-learner-","dir":"Reference","previous_headings":"","what":"Method learner()","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"Retrieve mlr3::Learner -th evaluation, position unique hash uhash. uhash mutually exclusive. Learner contain model. Use $learners() get learners models.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"usage-1","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"","code":"ArchiveBatchFSelect$learner(i = NULL, uhash = NULL)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"arguments-1","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"(integer(1)) iteration value filter . uhash (logical(1)) uhash value filter .","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"method-learners-","dir":"Reference","previous_headings":"","what":"Method learners()","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"Retrieve list trained mlr3::Learner objects -th evaluation, position unique hash uhash. uhash mutually exclusive.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"usage-2","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"","code":"ArchiveBatchFSelect$learners(i = NULL, uhash = NULL)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"arguments-2","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"(integer(1)) iteration value filter . uhash (logical(1)) uhash value filter .","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"method-predictions-","dir":"Reference","previous_headings":"","what":"Method predictions()","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"Retrieve list mlr3::Prediction objects -th evaluation, position unique hash uhash. uhash mutually exclusive.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"usage-3","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"","code":"ArchiveBatchFSelect$predictions(i = NULL, uhash = NULL)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"arguments-3","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"(integer(1)) iteration value filter . uhash (logical(1)) uhash value filter .","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"method-resample-result-","dir":"Reference","previous_headings":"","what":"Method resample_result()","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"Retrieve mlr3::ResampleResult -th evaluation, position unique hash uhash. uhash mutually exclusive.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"usage-4","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"","code":"ArchiveBatchFSelect$resample_result(i = NULL, uhash = NULL)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"arguments-4","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"(integer(1)) iteration value filter . uhash (logical(1)) uhash value filter .","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"method-print-","dir":"Reference","previous_headings":"","what":"Method print()","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"Printer.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"usage-5","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"","code":"ArchiveBatchFSelect$print()"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"arguments-5","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"... (ignored).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"method-best-","dir":"Reference","previous_headings":"","what":"Method best()","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"Returns best scoring feature sets.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"usage-6","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"","code":"ArchiveBatchFSelect$best(batch = NULL, ties_method = NULL)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"arguments-6","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"batch (integer()) batch number(s) limit best results . Default batches. ties_method (character(1)) Method handle ties. NULL (default), global ties method set initialization used. default global ties method least_features selects feature set least features. multiple best feature sets number features, one selected randomly. random method returns random feature set best feature sets.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"returns","dir":"Reference","previous_headings":"","what":"Returns","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"data.table::data.table()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"method-clone-","dir":"Reference","previous_headings":"","what":"Method clone()","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"objects class cloneable method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"usage-7","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"","code":"ArchiveBatchFSelect$clone(deep = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ArchiveBatchFSelect.html","id":"arguments-7","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Logging Evaluated Feature Sets — ArchiveBatchFSelect","text":"deep Whether make deep clone.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":null,"dir":"Reference","previous_headings":"","what":"Class for Automatic Feature Selection — AutoFSelector","title":"Class for Automatic Feature Selection — AutoFSelector","text":"AutoFSelector wraps mlr3::Learner augments automatic feature selection. auto_fselector() function creates AutoFSelector object.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Class for Automatic Feature Selection — AutoFSelector","text":"AutoFSelector mlr3::Learner wraps another mlr3::Learner performs following steps $train(): wrapped (inner) learner trained feature subsets via resampling. feature selection can specified providing FSelector, bbotk::Terminator, mlr3::Resampling mlr3::Measure. final model fit complete training data best-found feature subset. $predict() AutoFSelector just calls predict method wrapped (inner) learner.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"resources","dir":"Reference","previous_headings":"","what":"Resources","title":"Class for Automatic Feature Selection — AutoFSelector","text":"several sections feature selection mlr3book. Estimate Model Performance nested resampling. gallery features collection case studies demos optimization.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"nested-resampling","dir":"Reference","previous_headings":"","what":"Nested Resampling","title":"Class for Automatic Feature Selection — AutoFSelector","text":"Nested resampling can performed passing AutoFSelector object mlr3::resample() mlr3::benchmark(). access inner resampling results, set store_fselect_instance = TRUE execute mlr3::resample() mlr3::benchmark() store_models = TRUE (see examples). mlr3::Resampling passed AutoFSelector meant inner resampling, operating training set arbitrary outer resampling. reason feasible pass instantiated mlr3::Resampling .","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"super-class","dir":"Reference","previous_headings":"","what":"Super class","title":"Class for Automatic Feature Selection — AutoFSelector","text":"mlr3::Learner -> AutoFSelector","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"public-fields","dir":"Reference","previous_headings":"","what":"Public fields","title":"Class for Automatic Feature Selection — AutoFSelector","text":"instance_args (list()) arguments construction create FSelectInstanceBatchSingleCrit. fselector (FSelector) Optimization algorithm.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"active-bindings","dir":"Reference","previous_headings":"","what":"Active bindings","title":"Class for Automatic Feature Selection — AutoFSelector","text":"archive ([ArchiveBatchFSelect) Returns FSelectInstanceBatchSingleCrit archive. learner (mlr3::Learner) Trained learner. fselect_instance (FSelectInstanceBatchSingleCrit) Internally created feature selection instance intermediate results. fselect_result (data.table::data.table) Short-cut $result FSelectInstanceBatchSingleCrit. predict_type (character(1)) Stores currently active predict type, e.g. \"response\". Must element $predict_types. hash (character(1)) Hash (unique identifier) object. phash (character(1)) Hash (unique identifier) partial object, excluding components varied systematically tuning (parameter values) feature selection (feature names).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Class for Automatic Feature Selection — AutoFSelector","text":"mlr3::Learner$format() mlr3::Learner$help() mlr3::Learner$predict() mlr3::Learner$predict_newdata() mlr3::Learner$reset() mlr3::Learner$train()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"public-methods","dir":"Reference","previous_headings":"","what":"Public methods","title":"Class for Automatic Feature Selection — AutoFSelector","text":"AutoFSelector$new() AutoFSelector$base_learner() AutoFSelector$importance() AutoFSelector$selected_features() AutoFSelector$oob_error() AutoFSelector$loglik() AutoFSelector$print() AutoFSelector$clone()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"method-new-","dir":"Reference","previous_headings":"","what":"Method new()","title":"Class for Automatic Feature Selection — AutoFSelector","text":"Creates new instance R6 class.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Automatic Feature Selection — AutoFSelector","text":"","code":"AutoFSelector$new( fselector, learner, resampling, measure = NULL, terminator, store_fselect_instance = TRUE, store_benchmark_result = TRUE, store_models = FALSE, check_values = FALSE, callbacks = NULL, ties_method = \"least_features\" )"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Automatic Feature Selection — AutoFSelector","text":"fselector (FSelector) Optimization algorithm. learner (mlr3::Learner) Learner optimize feature subset . resampling (mlr3::Resampling) Resampling used evaluated performance feature subsets. Uninstantiated resamplings instantiated construction feature subsets evaluated data splits. Already instantiated resamplings kept unchanged. measure (mlr3::Measure) Measure optimize. NULL, default measure used. terminator (bbotk::Terminator) Stop criterion feature selection. store_fselect_instance (logical(1)) TRUE (default), stores internally created FSelectInstanceBatchSingleCrit intermediate results slot $fselect_instance. set TRUE, store_models = TRUE store_benchmark_result (logical(1)) Store benchmark result archive? store_models (logical(1)). Store models benchmark result? check_values (logical(1)) Check parameters evaluation results validity? callbacks (list CallbackBatchFSelect) List callbacks. ties_method (character(1)) method break ties selecting sets optimizing selecting best set. Can \"least_features\" \"random\". option \"least_features\" (default) selects feature set least features. multiple best feature sets number features, one selected randomly. random method returns random feature set best feature sets. Ignored multiple measures used.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"method-base-learner-","dir":"Reference","previous_headings":"","what":"Method base_learner()","title":"Class for Automatic Feature Selection — AutoFSelector","text":"Extracts base learner nested learner objects like GraphLearner mlr3pipelines. recursive = 0, (tuned) learner returned.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"usage-1","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Automatic Feature Selection — AutoFSelector","text":"","code":"AutoFSelector$base_learner(recursive = Inf)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"arguments-1","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Automatic Feature Selection — AutoFSelector","text":"recursive (integer(1)) Depth recursion multiple nested objects.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"returns","dir":"Reference","previous_headings":"","what":"Returns","title":"Class for Automatic Feature Selection — AutoFSelector","text":"mlr3::Learner.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"method-importance-","dir":"Reference","previous_headings":"","what":"Method importance()","title":"Class for Automatic Feature Selection — AutoFSelector","text":"importance scores final model.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"usage-2","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Automatic Feature Selection — AutoFSelector","text":"","code":"AutoFSelector$importance()"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"returns-1","dir":"Reference","previous_headings":"","what":"Returns","title":"Class for Automatic Feature Selection — AutoFSelector","text":"Named numeric().","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"method-selected-features-","dir":"Reference","previous_headings":"","what":"Method selected_features()","title":"Class for Automatic Feature Selection — AutoFSelector","text":"selected features final model. features selected internally learner.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"usage-3","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Automatic Feature Selection — AutoFSelector","text":"","code":"AutoFSelector$selected_features()"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"returns-2","dir":"Reference","previous_headings":"","what":"Returns","title":"Class for Automatic Feature Selection — AutoFSelector","text":"character().","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"method-oob-error-","dir":"Reference","previous_headings":"","what":"Method oob_error()","title":"Class for Automatic Feature Selection — AutoFSelector","text":"--bag error final model.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"usage-4","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Automatic Feature Selection — AutoFSelector","text":"","code":"AutoFSelector$oob_error()"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"returns-3","dir":"Reference","previous_headings":"","what":"Returns","title":"Class for Automatic Feature Selection — AutoFSelector","text":"numeric(1).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"method-loglik-","dir":"Reference","previous_headings":"","what":"Method loglik()","title":"Class for Automatic Feature Selection — AutoFSelector","text":"log-likelihood final model.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"usage-5","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Automatic Feature Selection — AutoFSelector","text":"","code":"AutoFSelector$loglik()"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"returns-4","dir":"Reference","previous_headings":"","what":"Returns","title":"Class for Automatic Feature Selection — AutoFSelector","text":"logLik. Printer.","code":""},{"path":[]},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"usage-6","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Automatic Feature Selection — AutoFSelector","text":"","code":"AutoFSelector$print()"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"arguments-2","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Automatic Feature Selection — AutoFSelector","text":"... (ignored).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"method-clone-","dir":"Reference","previous_headings":"","what":"Method clone()","title":"Class for Automatic Feature Selection — AutoFSelector","text":"objects class cloneable method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"usage-7","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Automatic Feature Selection — AutoFSelector","text":"","code":"AutoFSelector$clone(deep = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"arguments-3","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Automatic Feature Selection — AutoFSelector","text":"deep Whether make deep clone.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/AutoFSelector.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Class for Automatic Feature Selection — AutoFSelector","text":"","code":"# Automatic Feature Selection # \\donttest{ # split to train and external set task = tsk(\"penguins\") split = partition(task, ratio = 0.8) # create auto fselector afs = auto_fselector( fselector = fs(\"random_search\"), learner = lrn(\"classif.rpart\"), resampling = rsmp (\"holdout\"), measure = msr(\"classif.ce\"), term_evals = 4) # optimize feature subset and fit final model afs$train(task, row_ids = split$train) # predict with final model afs$predict(task, row_ids = split$test) #> for 69 observations: #> row_ids truth response #> 15 Adelie Adelie #> 16 Adelie Adelie #> 20 Adelie Chinstrap #> --- #> 338 Chinstrap Chinstrap #> 340 Chinstrap Gentoo #> 344 Chinstrap Chinstrap # show result afs$fselect_result #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: FALSE TRUE FALSE TRUE FALSE TRUE FALSE #> features n_features classif.ce #> #> 1: bill_length,flipper_length,sex 3 0.05434783 # model slot contains trained learner and fselect instance afs$model #> $learner #> : Classification Tree #> * Model: rpart #> * Parameters: xval=0 #> * Packages: mlr3, rpart #> * Predict Types: [response], prob #> * Feature Types: logical, integer, numeric, factor, ordered #> * Properties: importance, missings, multiclass, selected_features, #> twoclass, weights #> #> $features #> [1] \"bill_length\" \"flipper_length\" \"sex\" #> #> $fselect_instance #> #> * State: Optimized #> * Objective: #> * Terminator: #> * Result: #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: FALSE TRUE FALSE TRUE FALSE TRUE FALSE #> classif.ce #> #> 1: 0.05434783 #> * Archive: #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> 2: FALSE TRUE FALSE TRUE FALSE TRUE FALSE #> 3: TRUE FALSE FALSE TRUE FALSE FALSE TRUE #> 4: FALSE FALSE FALSE TRUE TRUE FALSE FALSE #> 5: TRUE TRUE TRUE FALSE TRUE TRUE TRUE #> 6: FALSE FALSE FALSE TRUE TRUE TRUE TRUE #> 7: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> 8: FALSE TRUE TRUE TRUE FALSE TRUE FALSE #> 9: TRUE TRUE TRUE TRUE TRUE TRUE FALSE #> 10: FALSE TRUE FALSE FALSE TRUE FALSE TRUE #> classif.ce #> #> 1: 0.05434783 #> 2: 0.05434783 #> 3: 0.18478261 #> 4: 0.11956522 #> 5: 0.08695652 #> 6: 0.13043478 #> 7: 0.05434783 #> 8: 0.05434783 #> 9: 0.05434783 #> 10: 0.08695652 #> # shortcut trained learner afs$learner #> : Classification Tree #> * Model: rpart #> * Parameters: xval=0 #> * Packages: mlr3, rpart #> * Predict Types: [response], prob #> * Feature Types: logical, integer, numeric, factor, ordered #> * Properties: importance, missings, multiclass, selected_features, #> twoclass, weights # shortcut fselect instance afs$fselect_instance #> #> * State: Optimized #> * Objective: #> * Terminator: #> * Result: #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: FALSE TRUE FALSE TRUE FALSE TRUE FALSE #> classif.ce #> #> 1: 0.05434783 #> * Archive: #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> 2: FALSE TRUE FALSE TRUE FALSE TRUE FALSE #> 3: TRUE FALSE FALSE TRUE FALSE FALSE TRUE #> 4: FALSE FALSE FALSE TRUE TRUE FALSE FALSE #> 5: TRUE TRUE TRUE FALSE TRUE TRUE TRUE #> 6: FALSE FALSE FALSE TRUE TRUE TRUE TRUE #> 7: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> 8: FALSE TRUE TRUE TRUE FALSE TRUE FALSE #> 9: TRUE TRUE TRUE TRUE TRUE TRUE FALSE #> 10: FALSE TRUE FALSE FALSE TRUE FALSE TRUE #> classif.ce #> #> 1: 0.05434783 #> 2: 0.05434783 #> 3: 0.18478261 #> 4: 0.11956522 #> 5: 0.08695652 #> 6: 0.13043478 #> 7: 0.05434783 #> 8: 0.05434783 #> 9: 0.05434783 #> 10: 0.08695652 # Nested Resampling afs = auto_fselector( fselector = fs(\"random_search\"), learner = lrn(\"classif.rpart\"), resampling = rsmp (\"holdout\"), measure = msr(\"classif.ce\"), term_evals = 4) resampling_outer = rsmp(\"cv\", folds = 3) rr = resample(task, afs, resampling_outer, store_models = TRUE) # retrieve inner feature selection results. extract_inner_fselect_results(rr) #> iteration bill_depth bill_length body_mass flipper_length island sex #> #> 1: 1 TRUE TRUE FALSE FALSE TRUE FALSE #> 2: 2 FALSE TRUE FALSE TRUE TRUE FALSE #> 3: 3 TRUE TRUE TRUE TRUE FALSE FALSE #> year classif.ce features n_features #> #> 1: FALSE 0.03947368 bill_depth,bill_length,island 3 #> 2: TRUE 0.07894737 bill_length,flipper_length,island,year 4 #> 3: FALSE 0.05194805 bill_depth,bill_length,body_mass,flipper_length 4 #> task_id learner_id resampling_id #> #> 1: penguins classif.rpart.fselector cv #> 2: penguins classif.rpart.fselector cv #> 3: penguins classif.rpart.fselector cv # performance scores estimated on the outer resampling rr$score() #> task_id learner_id resampling_id iteration classif.ce #> #> 1: penguins classif.rpart.fselector cv 1 0.10434783 #> 2: penguins classif.rpart.fselector cv 2 0.04347826 #> 3: penguins classif.rpart.fselector cv 3 0.05263158 #> Hidden columns: task, learner, resampling, prediction # unbiased performance of the final model trained on the full data set rr$aggregate() #> classif.ce #> 0.06681922 # }"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/CallbackBatchFSelect.html","id":null,"dir":"Reference","previous_headings":"","what":"Create Feature Selection Callback — CallbackBatchFSelect","title":"Create Feature Selection Callback — CallbackBatchFSelect","text":"Specialized bbotk::CallbackBatch feature selection. Callbacks allow customizing behavior processes mlr3fselect. callback_batch_fselect() function creates CallbackBatchFSelect. Predefined callbacks stored dictionary mlr_callbacks can retrieved clbk(). information callbacks see callback_batch_fselect().","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/CallbackBatchFSelect.html","id":"super-classes","dir":"Reference","previous_headings":"","what":"Super classes","title":"Create Feature Selection Callback — CallbackBatchFSelect","text":"mlr3misc::Callback -> bbotk::CallbackBatch -> CallbackBatchFSelect","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/CallbackBatchFSelect.html","id":"public-fields","dir":"Reference","previous_headings":"","what":"Public fields","title":"Create Feature Selection Callback — CallbackBatchFSelect","text":"on_eval_after_design (function()) Stage called design created. Called ObjectiveFSelectBatch$eval_many(). on_eval_after_benchmark (function()) Stage called feature sets evaluated. Called ObjectiveFSelectBatch$eval_many(). on_eval_before_archive (function()) Stage called performance values written archive. Called ObjectiveFSelectBatch$eval_many().","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/CallbackBatchFSelect.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Create Feature Selection Callback — CallbackBatchFSelect","text":"mlr3misc::Callback$call() mlr3misc::Callback$format() mlr3misc::Callback$help() mlr3misc::Callback$initialize() mlr3misc::Callback$print()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/CallbackBatchFSelect.html","id":"public-methods","dir":"Reference","previous_headings":"","what":"Public methods","title":"Create Feature Selection Callback — CallbackBatchFSelect","text":"CallbackBatchFSelect$clone()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/CallbackBatchFSelect.html","id":"method-clone-","dir":"Reference","previous_headings":"","what":"Method clone()","title":"Create Feature Selection Callback — CallbackBatchFSelect","text":"objects class cloneable method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/CallbackBatchFSelect.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Create Feature Selection Callback — CallbackBatchFSelect","text":"","code":"CallbackBatchFSelect$clone(deep = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/CallbackBatchFSelect.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Create Feature Selection Callback — CallbackBatchFSelect","text":"deep Whether make deep clone.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/CallbackBatchFSelect.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Create Feature Selection Callback — CallbackBatchFSelect","text":"","code":"# Write archive to disk callback_batch_fselect(\"mlr3fselect.backup\", on_optimization_end = function(callback, context) { saveRDS(context$instance$archive, \"archive.rds\") } ) #> #> * Active Stages: on_optimization_end"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ContextBatchFSelect.html","id":null,"dir":"Reference","previous_headings":"","what":"Evaluation Context — ContextBatchFSelect","title":"Evaluation Context — ContextBatchFSelect","text":"ContextBatchFSelect allows CallbackBatchFSelects access modify data batch feature sets evaluated. See section active bindings list modifiable objects. See callback_batch_fselect() list stages access ContextBatchFSelect.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ContextBatchFSelect.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Evaluation Context — ContextBatchFSelect","text":"context re-created time new batch feature sets evaluated. Changes $objective_fselect, $design $benchmark_result discarded function finished. Modification data table $aggregated_performance written archive. number columns can added.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ContextBatchFSelect.html","id":"super-classes","dir":"Reference","previous_headings":"","what":"Super classes","title":"Evaluation Context — ContextBatchFSelect","text":"mlr3misc::Context -> bbotk::ContextBatch -> ContextBatchFSelect","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ContextBatchFSelect.html","id":"active-bindings","dir":"Reference","previous_headings":"","what":"Active bindings","title":"Evaluation Context — ContextBatchFSelect","text":"xss (list()) feature sets latest batch. design (data.table::data.table) benchmark design latest batch. benchmark_result (mlr3::BenchmarkResult) benchmark result latest batch. aggregated_performance (data.table::data.table) Aggregated performance scores training time latest batch. data table passed archive. callback can add additional columns also written archive.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ContextBatchFSelect.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Evaluation Context — ContextBatchFSelect","text":"mlr3misc::Context$format() mlr3misc::Context$print() bbotk::ContextBatch$initialize()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ContextBatchFSelect.html","id":"public-methods","dir":"Reference","previous_headings":"","what":"Public methods","title":"Evaluation Context — ContextBatchFSelect","text":"ContextBatchFSelect$clone()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ContextBatchFSelect.html","id":"method-clone-","dir":"Reference","previous_headings":"","what":"Method clone()","title":"Evaluation Context — ContextBatchFSelect","text":"objects class cloneable method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ContextBatchFSelect.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Evaluation Context — ContextBatchFSelect","text":"","code":"ContextBatchFSelect$clone(deep = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ContextBatchFSelect.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Evaluation Context — ContextBatchFSelect","text":"deep Whether make deep clone.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchMultiCrit.html","id":null,"dir":"Reference","previous_headings":"","what":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","title":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","text":"FSelectInstanceBatchMultiCrit specifies feature selection problem FSelector. function fsi() creates FSelectInstanceBatchMultiCrit function fselect() creates instance internally.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchMultiCrit.html","id":"resources","dir":"Reference","previous_headings":"","what":"Resources","title":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","text":"several sections feature selection mlr3book. Learn multi-objective optimization. gallery features collection case studies demos optimization.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchMultiCrit.html","id":"analysis","dir":"Reference","previous_headings":"","what":"Analysis","title":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","text":"analyzing feature selection results, recommended pass archive .data.table(). returned data table joined benchmark result adds mlr3::ResampleResult feature set. archive provides various getters (e.g. $learners()) ease access. getters extract position () unique hash (uhash). complete list getters see methods section. benchmark result ($benchmark_result) allows score feature sets different measure. Alternatively, measures can supplied .data.table().","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchMultiCrit.html","id":"super-classes","dir":"Reference","previous_headings":"","what":"Super classes","title":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","text":"bbotk::OptimInstance -> bbotk::OptimInstanceBatch -> bbotk::OptimInstanceBatchMultiCrit -> FSelectInstanceBatchMultiCrit","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchMultiCrit.html","id":"active-bindings","dir":"Reference","previous_headings":"","what":"Active bindings","title":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","text":"result_feature_set (list character()) Feature sets task subsetting.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchMultiCrit.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","text":"bbotk::OptimInstance$clear() bbotk::OptimInstance$format() bbotk::OptimInstanceBatch$eval_batch() bbotk::OptimInstanceBatch$objective_function()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchMultiCrit.html","id":"public-methods","dir":"Reference","previous_headings":"","what":"Public methods","title":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","text":"FSelectInstanceBatchMultiCrit$new() FSelectInstanceBatchMultiCrit$assign_result() FSelectInstanceBatchMultiCrit$print() FSelectInstanceBatchMultiCrit$clone()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchMultiCrit.html","id":"method-new-","dir":"Reference","previous_headings":"","what":"Method new()","title":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","text":"Creates new instance R6 class.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchMultiCrit.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","text":"","code":"FSelectInstanceBatchMultiCrit$new( task, learner, resampling, measures, terminator, store_benchmark_result = TRUE, store_models = FALSE, check_values = FALSE, callbacks = NULL )"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchMultiCrit.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","text":"task (mlr3::Task) Task operate . learner (mlr3::Learner) Learner optimize feature subset . resampling (mlr3::Resampling) Resampling used evaluated performance feature subsets. Uninstantiated resamplings instantiated construction feature subsets evaluated data splits. Already instantiated resamplings kept unchanged. measures (list mlr3::Measure) Measures optimize. NULL, mlr3's default measure used. terminator (bbotk::Terminator) Stop criterion feature selection. store_benchmark_result (logical(1)) Store benchmark result archive? store_models (logical(1)). Store models benchmark result? check_values (logical(1)) Check parameters evaluation results validity? callbacks (list CallbackBatchFSelect) List callbacks.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchMultiCrit.html","id":"method-assign-result-","dir":"Reference","previous_headings":"","what":"Method assign_result()","title":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","text":"FSelector object writes best found feature subsets estimated performance values . internal use.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchMultiCrit.html","id":"usage-1","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","text":"","code":"FSelectInstanceBatchMultiCrit$assign_result(xdt, ydt)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchMultiCrit.html","id":"arguments-1","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","text":"xdt (data.table::data.table()) x values data.table. row one point. Contains value search space FSelectInstanceBatchMultiCrit object. Can contain additional columns extra information. ydt (data.table::data.table()) Optimal outcomes, e.g. Pareto front.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchMultiCrit.html","id":"method-print-","dir":"Reference","previous_headings":"","what":"Method print()","title":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","text":"Printer.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchMultiCrit.html","id":"usage-2","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","text":"","code":"FSelectInstanceBatchMultiCrit$print(...)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchMultiCrit.html","id":"arguments-2","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","text":"... (ignored).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchMultiCrit.html","id":"method-clone-","dir":"Reference","previous_headings":"","what":"Method clone()","title":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","text":"objects class cloneable method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchMultiCrit.html","id":"usage-3","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","text":"","code":"FSelectInstanceBatchMultiCrit$clone(deep = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchMultiCrit.html","id":"arguments-3","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","text":"deep Whether make deep clone.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchMultiCrit.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Class for Multi Criteria Feature Selection — FSelectInstanceBatchMultiCrit","text":"","code":"# Feature selection on Palmer Penguins data set # \\donttest{ task = tsk(\"penguins\") # Construct feature selection instance instance = fsi( task = task, learner = lrn(\"classif.rpart\"), resampling = rsmp(\"cv\", folds = 3), measures = msrs(c(\"classif.ce\", \"time_train\")), terminator = trm(\"evals\", n_evals = 4) ) # Choose optimization algorithm fselector = fs(\"random_search\", batch_size = 2) # Run feature selection fselector$optimize(instance) #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> 2: FALSE TRUE TRUE TRUE FALSE FALSE FALSE #> features n_features #> #> 1: bill_depth,bill_length,body_mass,flipper_length,island,sex,... 7 #> 2: bill_length,body_mass,flipper_length 7 #> classif.ce time_train #> #> 1: 0.06984490 0.003 #> 2: 0.08151538 0.002 # Optimal feature sets instance$result_feature_set #> [[1]] #> [1] \"bill_depth\" \"bill_length\" \"body_mass\" \"flipper_length\" #> [5] \"island\" \"sex\" \"year\" #> #> [[2]] #> [1] \"bill_length\" \"body_mass\" \"flipper_length\" #> # Inspect all evaluated sets as.data.table(instance$archive) #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE TRUE TRUE TRUE FALSE TRUE TRUE #> 2: TRUE FALSE TRUE TRUE FALSE TRUE TRUE #> 3: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> 4: FALSE TRUE TRUE TRUE FALSE FALSE FALSE #> classif.ce time_train runtime_learners timestamp batch_nr #> #> 1: 0.08151538 0.003333333 0.016 2024-07-24 12:00:46 1 #> 2: 0.20358505 0.002666667 0.014 2024-07-24 12:00:46 1 #> 3: 0.06984490 0.003000000 0.016 2024-07-24 12:00:46 2 #> 4: 0.08151538 0.002000000 0.013 2024-07-24 12:00:46 2 #> warnings errors #> #> 1: 0 0 #> 2: 0 0 #> 3: 0 0 #> 4: 0 0 #> features n_features #> #> 1: bill_depth,bill_length,body_mass,flipper_length,sex,year 6 #> 2: bill_depth,body_mass,flipper_length,sex,year 5 #> 3: bill_depth,bill_length,body_mass,flipper_length,island,sex,... 7 #> 4: bill_length,body_mass,flipper_length 3 #> resample_result #> #> 1: #> 2: #> 3: #> 4: # }"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":null,"dir":"Reference","previous_headings":"","what":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"FSelectInstanceBatchSingleCrit specifies feature selection problem FSelector. function fsi() creates FSelectInstanceBatchSingleCrit function fselect() creates instance internally. instance contains ObjectiveFSelectBatch object encodes black box objective function FSelector optimize. instance allows basic operations querying objective design points ($eval_batch()). operation usually done FSelector. Evaluations feature subsets performed batches calling mlr3::benchmark() internally. evaluated feature subsets stored Archive ($archive). batch evaluated, bbotk::Terminator queried remaining budget. available budget exhausted, exception raised, evaluations can performed point . FSelector also supposed store final result, consisting selected feature subset associated estimated performance values, calling method instance$assign_result().","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":"default-measures","dir":"Reference","previous_headings":"","what":"Default Measures","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"measure passed, default measure used. default measure depends task type.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":"resources","dir":"Reference","previous_headings":"","what":"Resources","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"several sections feature selection mlr3book. Getting started wrapper feature selection. sequential forward selection Palmer Penguins data set. gallery features collection case studies demos optimization. Utilize built-feature importance models Recursive Feature Elimination. Run feature selection Shadow Variable Search. Feature Selection Titanic data set.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":"analysis","dir":"Reference","previous_headings":"","what":"Analysis","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"analyzing feature selection results, recommended pass archive .data.table(). returned data table joined benchmark result adds mlr3::ResampleResult feature set. archive provides various getters (e.g. $learners()) ease access. getters extract position () unique hash (uhash). complete list getters see methods section. benchmark result ($benchmark_result) allows score feature sets different measure. Alternatively, measures can supplied .data.table().","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":"super-classes","dir":"Reference","previous_headings":"","what":"Super classes","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"bbotk::OptimInstance -> bbotk::OptimInstanceBatch -> bbotk::OptimInstanceBatchSingleCrit -> FSelectInstanceBatchSingleCrit","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":"active-bindings","dir":"Reference","previous_headings":"","what":"Active bindings","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"result_feature_set (character()) Feature set task subsetting.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"bbotk::OptimInstance$clear() bbotk::OptimInstance$format() bbotk::OptimInstanceBatch$eval_batch() bbotk::OptimInstanceBatch$objective_function()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":"public-methods","dir":"Reference","previous_headings":"","what":"Public methods","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"FSelectInstanceBatchSingleCrit$new() FSelectInstanceBatchSingleCrit$assign_result() FSelectInstanceBatchSingleCrit$print() FSelectInstanceBatchSingleCrit$clone()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":"method-new-","dir":"Reference","previous_headings":"","what":"Method new()","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"Creates new instance R6 class.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"","code":"FSelectInstanceBatchSingleCrit$new( task, learner, resampling, measure, terminator, store_benchmark_result = TRUE, store_models = FALSE, check_values = FALSE, callbacks = NULL, ties_method = \"least_features\" )"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"task (mlr3::Task) Task operate . learner (mlr3::Learner) Learner optimize feature subset . resampling (mlr3::Resampling) Resampling used evaluated performance feature subsets. Uninstantiated resamplings instantiated construction feature subsets evaluated data splits. Already instantiated resamplings kept unchanged. measure (mlr3::Measure) Measure optimize. NULL, default measure used. terminator (bbotk::Terminator) Stop criterion feature selection. store_benchmark_result (logical(1)) Store benchmark result archive? store_models (logical(1)). Store models benchmark result? check_values (logical(1)) Check parameters evaluation results validity? callbacks (list CallbackBatchFSelect) List callbacks. ties_method (character(1)) method break ties selecting sets optimizing selecting best set. Can \"least_features\" \"random\". option \"least_features\" (default) selects feature set least features. multiple best feature sets number features, one selected randomly. random method returns random feature set best feature sets. Ignored multiple measures used.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":"method-assign-result-","dir":"Reference","previous_headings":"","what":"Method assign_result()","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"FSelector writes best found feature subset estimated performance value . internal use.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":"usage-1","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"","code":"FSelectInstanceBatchSingleCrit$assign_result(xdt, y)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":"arguments-1","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"xdt (data.table::data.table()) x values data.table. row one point. Contains value search space FSelectInstanceBatchMultiCrit object. Can contain additional columns extra information. y (numeric(1)) Optimal outcome.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":"method-print-","dir":"Reference","previous_headings":"","what":"Method print()","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"Printer.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":"usage-2","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"","code":"FSelectInstanceBatchSingleCrit$print(...)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":"arguments-2","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"... (ignored).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":"method-clone-","dir":"Reference","previous_headings":"","what":"Method clone()","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"objects class cloneable method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":"usage-3","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"","code":"FSelectInstanceBatchSingleCrit$clone(deep = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":"arguments-3","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"deep Whether make deep clone.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectInstanceBatchSingleCrit.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Class for Single Criterion Feature Selection — FSelectInstanceBatchSingleCrit","text":"","code":"# Feature selection on Palmer Penguins data set # \\donttest{ task = tsk(\"penguins\") learner = lrn(\"classif.rpart\") # Construct feature selection instance instance = fsi( task = task, learner = learner, resampling = rsmp(\"cv\", folds = 3), measures = msr(\"classif.ce\"), terminator = trm(\"evals\", n_evals = 4) ) # Choose optimization algorithm fselector = fs(\"random_search\", batch_size = 2) # Run feature selection fselector$optimize(instance) #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> features n_features #> #> 1: bill_depth,bill_length,body_mass,flipper_length,island,sex,... 7 #> classif.ce #> #> 1: 0.04652937 # Subset task to optimal feature set task$select(instance$result_feature_set) # Train the learner with optimal feature set on the full data set learner$train(task) # Inspect all evaluated sets as.data.table(instance$archive) #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: FALSE TRUE TRUE FALSE FALSE FALSE FALSE #> 2: TRUE TRUE TRUE FALSE TRUE TRUE TRUE #> 3: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> 4: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> classif.ce runtime_learners timestamp batch_nr warnings errors #> #> 1: 0.11929825 0.010 2024-07-24 12:00:47 1 0 0 #> 2: 0.06117468 0.016 2024-07-24 12:00:47 1 0 0 #> 3: 0.04652937 0.014 2024-07-24 12:00:47 2 0 0 #> 4: 0.04652937 0.033 2024-07-24 12:00:47 2 0 0 #> features n_features #> #> 1: bill_length,body_mass 2 #> 2: bill_depth,bill_length,body_mass,island,sex,year 6 #> 3: bill_depth,bill_length,body_mass,flipper_length,island,sex,... 7 #> 4: bill_depth,bill_length,body_mass,flipper_length,island,sex,... 7 #> resample_result #> #> 1: #> 2: #> 3: #> 4: # }"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":null,"dir":"Reference","previous_headings":"","what":"FSelector — FSelector","title":"FSelector — FSelector","text":"`FSelector“ implements optimization algorithm.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"FSelector — FSelector","text":"FSelector abstract base class implements base functionality fselector must provide.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":"resources","dir":"Reference","previous_headings":"","what":"Resources","title":"FSelector — FSelector","text":"several sections feature selection mlr3book. Learn fselectors. gallery features collection case studies demos optimization. Utilize built-feature importance models Recursive Feature Elimination. Run feature selection Shadow Variable Search.","code":""},{"path":[]},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":"public-fields","dir":"Reference","previous_headings":"","what":"Public fields","title":"FSelector — FSelector","text":"id (character(1)) Identifier object. Used tables, plot text output.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":"active-bindings","dir":"Reference","previous_headings":"","what":"Active bindings","title":"FSelector — FSelector","text":"param_set paradox::ParamSet Set control parameters. properties (character()) Set properties fselector. Must subset mlr_reflections$fselect_properties. packages (character()) Set required packages. Note packages loaded via requireNamespace(), attached. label (character(1)) Label object. Can used tables, plot text output instead ID. man (character(1)) String format [pkg]::[topic] pointing manual page object. referenced help package can opened via method $help().","code":""},{"path":[]},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":"public-methods","dir":"Reference","previous_headings":"","what":"Public methods","title":"FSelector — FSelector","text":"FSelector$new() FSelector$format() FSelector$print() FSelector$help() FSelector$clone()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":"method-new-","dir":"Reference","previous_headings":"","what":"Method new()","title":"FSelector — FSelector","text":"Creates new instance R6 class.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"FSelector — FSelector","text":"","code":"FSelector$new( id = \"fselector\", param_set, properties, packages = character(), label = NA_character_, man = NA_character_ )"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"FSelector — FSelector","text":"id (character(1)) Identifier new instance. param_set paradox::ParamSet Set control parameters. properties (character()) Set properties fselector. Must subset mlr_reflections$fselect_properties. packages (character()) Set required packages. Note packages loaded via requireNamespace(), attached. label (character(1)) Label object. Can used tables, plot text output instead ID. man (character(1)) String format [pkg]::[topic] pointing manual page object. referenced help package can opened via method $help().","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":"method-format-","dir":"Reference","previous_headings":"","what":"Method format()","title":"FSelector — FSelector","text":"Helper print outputs.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":"usage-1","dir":"Reference","previous_headings":"","what":"Usage","title":"FSelector — FSelector","text":"","code":"FSelector$format(...)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":"arguments-1","dir":"Reference","previous_headings":"","what":"Arguments","title":"FSelector — FSelector","text":"... (ignored).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":"returns","dir":"Reference","previous_headings":"","what":"Returns","title":"FSelector — FSelector","text":"(character()).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":"method-print-","dir":"Reference","previous_headings":"","what":"Method print()","title":"FSelector — FSelector","text":"Print method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":"usage-2","dir":"Reference","previous_headings":"","what":"Usage","title":"FSelector — FSelector","text":"","code":"FSelector$print()"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":"returns-1","dir":"Reference","previous_headings":"","what":"Returns","title":"FSelector — FSelector","text":"(character()).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":"method-help-","dir":"Reference","previous_headings":"","what":"Method help()","title":"FSelector — FSelector","text":"Opens corresponding help page referenced field $man.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":"usage-3","dir":"Reference","previous_headings":"","what":"Usage","title":"FSelector — FSelector","text":"","code":"FSelector$help()"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":"method-clone-","dir":"Reference","previous_headings":"","what":"Method clone()","title":"FSelector — FSelector","text":"objects class cloneable method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":"usage-4","dir":"Reference","previous_headings":"","what":"Usage","title":"FSelector — FSelector","text":"","code":"FSelector$clone(deep = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelector.html","id":"arguments-2","dir":"Reference","previous_headings":"","what":"Arguments","title":"FSelector — FSelector","text":"deep Whether make deep clone.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatch.html","id":null,"dir":"Reference","previous_headings":"","what":"Class for Batch Feature Selection Algorithms — FSelectorBatch","title":"Class for Batch Feature Selection Algorithms — FSelectorBatch","text":"FSelectorBatch implements optimization algorithm.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatch.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Class for Batch Feature Selection Algorithms — FSelectorBatch","text":"FSelectorBatch abstract base class implements base functionality fselector must provide. subclass implemented following way: Inherit FSelectorBatch. Specify private abstract method $.optimize() use call optimizer. need call instance$eval_batch() evaluate design points. batch evaluation requested FSelectInstanceBatchSingleCrit/FSelectInstanceBatchMultiCrit object instance, batch possibly executed parallel via mlr3::benchmark(), evaluations stored inside instance$archive. batch evaluation, bbotk::Terminator checked, positive, exception class \"terminated_error\" generated. latter case current batch evaluations still stored instance, numeric scores sent back handling optimizer lost execution control. exception caught select best set instance$archive return . Note therefore points specified bbotk::Terminator may evaluated, Terminator checked batch evaluation, -evaluation batch. many depends setting batch size. Overwrite private super-method .assign_result() want decide estimate final set instance estimated performance. default behavior : pick best resample experiment, regarding given measure, assign set aggregated performance instance.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatch.html","id":"private-methods","dir":"Reference","previous_headings":"","what":"Private Methods","title":"Class for Batch Feature Selection Algorithms — FSelectorBatch","text":".optimize(instance) -> NULL Abstract base method. Implement specify feature selection subclass. See technical details sections. .assign_result(instance) -> NULL Abstract base method. Implement specify final feature subset selected. See technical details sections.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatch.html","id":"resources","dir":"Reference","previous_headings":"","what":"Resources","title":"Class for Batch Feature Selection Algorithms — FSelectorBatch","text":"several sections feature selection mlr3book. Learn fselectors. gallery features collection case studies demos optimization. Utilize built-feature importance models Recursive Feature Elimination. Run feature selection Shadow Variable Search.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatch.html","id":"super-class","dir":"Reference","previous_headings":"","what":"Super class","title":"Class for Batch Feature Selection Algorithms — FSelectorBatch","text":"mlr3fselect::FSelector -> FSelectorBatch","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatch.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Class for Batch Feature Selection Algorithms — FSelectorBatch","text":"mlr3fselect::FSelector$format() mlr3fselect::FSelector$help() mlr3fselect::FSelector$print()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatch.html","id":"public-methods","dir":"Reference","previous_headings":"","what":"Public methods","title":"Class for Batch Feature Selection Algorithms — FSelectorBatch","text":"FSelectorBatch$new() FSelectorBatch$optimize() FSelectorBatch$clone()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatch.html","id":"method-new-","dir":"Reference","previous_headings":"","what":"Method new()","title":"Class for Batch Feature Selection Algorithms — FSelectorBatch","text":"Creates new instance R6 class.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatch.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Batch Feature Selection Algorithms — FSelectorBatch","text":"","code":"FSelectorBatch$new( id = \"fselector_batch\", param_set, properties, packages = character(), label = NA_character_, man = NA_character_ )"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatch.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Batch Feature Selection Algorithms — FSelectorBatch","text":"id (character(1)) Identifier new instance. param_set paradox::ParamSet Set control parameters. properties (character()) Set properties fselector. Must subset mlr_reflections$fselect_properties. packages (character()) Set required packages. Note packages loaded via requireNamespace(), attached. label (character(1)) Label object. Can used tables, plot text output instead ID. man (character(1)) String format [pkg]::[topic] pointing manual page object. referenced help package can opened via method $help().","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatch.html","id":"method-optimize-","dir":"Reference","previous_headings":"","what":"Method optimize()","title":"Class for Batch Feature Selection Algorithms — FSelectorBatch","text":"Performs feature selection FSelectInstanceBatchSingleCrit FSelectInstanceBatchMultiCrit termination. single evaluations written ArchiveBatchFSelect resides FSelectInstanceBatchSingleCrit / FSelectInstanceBatchMultiCrit. result written instance object.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatch.html","id":"usage-1","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Batch Feature Selection Algorithms — FSelectorBatch","text":"","code":"FSelectorBatch$optimize(inst)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatch.html","id":"arguments-1","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Batch Feature Selection Algorithms — FSelectorBatch","text":"inst (FSelectInstanceBatchSingleCrit | FSelectInstanceBatchMultiCrit).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatch.html","id":"returns","dir":"Reference","previous_headings":"","what":"Returns","title":"Class for Batch Feature Selection Algorithms — FSelectorBatch","text":"data.table::data.table().","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatch.html","id":"method-clone-","dir":"Reference","previous_headings":"","what":"Method clone()","title":"Class for Batch Feature Selection Algorithms — FSelectorBatch","text":"objects class cloneable method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatch.html","id":"usage-2","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Batch Feature Selection Algorithms — FSelectorBatch","text":"","code":"FSelectorBatch$clone(deep = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatch.html","id":"arguments-2","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Batch Feature Selection Algorithms — FSelectorBatch","text":"deep Whether make deep clone.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatchFromOptimizerBatch.html","id":null,"dir":"Reference","previous_headings":"","what":"FSelectorBatchFromOptimizerBatch — FSelectorBatchFromOptimizerBatch","title":"FSelectorBatchFromOptimizerBatch — FSelectorBatchFromOptimizerBatch","text":"Internally used transform bbotk::Optimizer FSelector.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatchFromOptimizerBatch.html","id":"super-classes","dir":"Reference","previous_headings":"","what":"Super classes","title":"FSelectorBatchFromOptimizerBatch — FSelectorBatchFromOptimizerBatch","text":"mlr3fselect::FSelector -> mlr3fselect::FSelectorBatch -> FSelectorBatchFromOptimizerBatch","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatchFromOptimizerBatch.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"FSelectorBatchFromOptimizerBatch — FSelectorBatchFromOptimizerBatch","text":"mlr3fselect::FSelector$format() mlr3fselect::FSelector$help() mlr3fselect::FSelector$print()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatchFromOptimizerBatch.html","id":"public-methods","dir":"Reference","previous_headings":"","what":"Public methods","title":"FSelectorBatchFromOptimizerBatch — FSelectorBatchFromOptimizerBatch","text":"FSelectorBatchFromOptimizerBatch$new() FSelectorBatchFromOptimizerBatch$optimize() FSelectorBatchFromOptimizerBatch$clone()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatchFromOptimizerBatch.html","id":"method-new-","dir":"Reference","previous_headings":"","what":"Method new()","title":"FSelectorBatchFromOptimizerBatch — FSelectorBatchFromOptimizerBatch","text":"Creates new instance R6 class.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatchFromOptimizerBatch.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"FSelectorBatchFromOptimizerBatch — FSelectorBatchFromOptimizerBatch","text":"","code":"FSelectorBatchFromOptimizerBatch$new(optimizer, man = NA_character_)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatchFromOptimizerBatch.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"FSelectorBatchFromOptimizerBatch — FSelectorBatchFromOptimizerBatch","text":"optimizer bbotk::Optimizer Optimizer called. man (character(1)) String format [pkg]::[topic] pointing manual page object. referenced help package can opened via method $help().","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatchFromOptimizerBatch.html","id":"method-optimize-","dir":"Reference","previous_headings":"","what":"Method optimize()","title":"FSelectorBatchFromOptimizerBatch — FSelectorBatchFromOptimizerBatch","text":"Performs feature selection FSelectInstanceBatchSingleCrit / FSelectInstanceBatchMultiCrit termination.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatchFromOptimizerBatch.html","id":"usage-1","dir":"Reference","previous_headings":"","what":"Usage","title":"FSelectorBatchFromOptimizerBatch — FSelectorBatchFromOptimizerBatch","text":"","code":"FSelectorBatchFromOptimizerBatch$optimize(inst)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatchFromOptimizerBatch.html","id":"arguments-1","dir":"Reference","previous_headings":"","what":"Arguments","title":"FSelectorBatchFromOptimizerBatch — FSelectorBatchFromOptimizerBatch","text":"inst (FSelectInstanceBatchSingleCrit | FSelectInstanceBatchMultiCrit).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatchFromOptimizerBatch.html","id":"returns","dir":"Reference","previous_headings":"","what":"Returns","title":"FSelectorBatchFromOptimizerBatch — FSelectorBatchFromOptimizerBatch","text":"data.table::data.table.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatchFromOptimizerBatch.html","id":"method-clone-","dir":"Reference","previous_headings":"","what":"Method clone()","title":"FSelectorBatchFromOptimizerBatch — FSelectorBatchFromOptimizerBatch","text":"objects class cloneable method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatchFromOptimizerBatch.html","id":"usage-2","dir":"Reference","previous_headings":"","what":"Usage","title":"FSelectorBatchFromOptimizerBatch — FSelectorBatchFromOptimizerBatch","text":"","code":"FSelectorBatchFromOptimizerBatch$clone(deep = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/FSelectorBatchFromOptimizerBatch.html","id":"arguments-2","dir":"Reference","previous_headings":"","what":"Arguments","title":"FSelectorBatchFromOptimizerBatch — FSelectorBatchFromOptimizerBatch","text":"deep Whether make deep clone.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelect.html","id":null,"dir":"Reference","previous_headings":"","what":"Class for Feature Selection Objective — ObjectiveFSelect","title":"Class for Feature Selection Objective — ObjectiveFSelect","text":"Stores objective function estimates performance feature subsets. class usually constructed internally FSelectInstanceBatchSingleCrit / FSelectInstanceBatchMultiCrit.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelect.html","id":"super-class","dir":"Reference","previous_headings":"","what":"Super class","title":"Class for Feature Selection Objective — ObjectiveFSelect","text":"bbotk::Objective -> ObjectiveFSelect","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelect.html","id":"public-fields","dir":"Reference","previous_headings":"","what":"Public fields","title":"Class for Feature Selection Objective — ObjectiveFSelect","text":"task (mlr3::Task). learner (mlr3::Learner). resampling (mlr3::Resampling). measures (list mlr3::Measure). store_models (logical(1)). store_benchmark_result (logical(1)). callbacks (List CallbackBatchFSelects).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelect.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Class for Feature Selection Objective — ObjectiveFSelect","text":"bbotk::Objective$eval() bbotk::Objective$eval_dt() bbotk::Objective$eval_many() bbotk::Objective$format() bbotk::Objective$help() bbotk::Objective$print()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelect.html","id":"public-methods","dir":"Reference","previous_headings":"","what":"Public methods","title":"Class for Feature Selection Objective — ObjectiveFSelect","text":"ObjectiveFSelect$new() ObjectiveFSelect$clone()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelect.html","id":"method-new-","dir":"Reference","previous_headings":"","what":"Method new()","title":"Class for Feature Selection Objective — ObjectiveFSelect","text":"Creates new instance R6 class.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelect.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Feature Selection Objective — ObjectiveFSelect","text":"","code":"ObjectiveFSelect$new( task, learner, resampling, measures, check_values = TRUE, store_benchmark_result = TRUE, store_models = FALSE, callbacks = NULL )"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelect.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Feature Selection Objective — ObjectiveFSelect","text":"task (mlr3::Task) Task operate . learner (mlr3::Learner) Learner optimize feature subset . resampling (mlr3::Resampling) Resampling used evaluated performance feature subsets. Uninstantiated resamplings instantiated construction feature subsets evaluated data splits. Already instantiated resamplings kept unchanged. measures (list mlr3::Measure) Measures optimize. NULL, mlr3's default measure used. check_values (logical(1)) Check parameters evaluation results validity? store_benchmark_result (logical(1)) Store benchmark result archive? store_models (logical(1)). Store models benchmark result? callbacks (list CallbackBatchFSelect) List callbacks.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelect.html","id":"method-clone-","dir":"Reference","previous_headings":"","what":"Method clone()","title":"Class for Feature Selection Objective — ObjectiveFSelect","text":"objects class cloneable method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelect.html","id":"usage-1","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Feature Selection Objective — ObjectiveFSelect","text":"","code":"ObjectiveFSelect$clone(deep = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelect.html","id":"arguments-1","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Feature Selection Objective — ObjectiveFSelect","text":"deep Whether make deep clone.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelectBatch.html","id":null,"dir":"Reference","previous_headings":"","what":"Class for Feature Selection Objective — ObjectiveFSelectBatch","title":"Class for Feature Selection Objective — ObjectiveFSelectBatch","text":"Stores objective function estimates performance feature subsets. class usually constructed internally FSelectInstanceBatchSingleCrit / FSelectInstanceBatchMultiCrit.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelectBatch.html","id":"super-classes","dir":"Reference","previous_headings":"","what":"Super classes","title":"Class for Feature Selection Objective — ObjectiveFSelectBatch","text":"bbotk::Objective -> mlr3fselect::ObjectiveFSelect -> ObjectiveFSelectBatch","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelectBatch.html","id":"public-fields","dir":"Reference","previous_headings":"","what":"Public fields","title":"Class for Feature Selection Objective — ObjectiveFSelectBatch","text":"archive (ArchiveBatchFSelect).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelectBatch.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Class for Feature Selection Objective — ObjectiveFSelectBatch","text":"bbotk::Objective$eval() bbotk::Objective$eval_dt() bbotk::Objective$eval_many() bbotk::Objective$format() bbotk::Objective$help() bbotk::Objective$print()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelectBatch.html","id":"public-methods","dir":"Reference","previous_headings":"","what":"Public methods","title":"Class for Feature Selection Objective — ObjectiveFSelectBatch","text":"ObjectiveFSelectBatch$new() ObjectiveFSelectBatch$clone()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelectBatch.html","id":"method-new-","dir":"Reference","previous_headings":"","what":"Method new()","title":"Class for Feature Selection Objective — ObjectiveFSelectBatch","text":"Creates new instance R6 class.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelectBatch.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Feature Selection Objective — ObjectiveFSelectBatch","text":"","code":"ObjectiveFSelectBatch$new( task, learner, resampling, measures, check_values = TRUE, store_benchmark_result = TRUE, store_models = FALSE, archive = NULL, callbacks = NULL )"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelectBatch.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Feature Selection Objective — ObjectiveFSelectBatch","text":"task (mlr3::Task) Task operate . learner (mlr3::Learner) Learner optimize feature subset . resampling (mlr3::Resampling) Resampling used evaluated performance feature subsets. Uninstantiated resamplings instantiated construction feature subsets evaluated data splits. Already instantiated resamplings kept unchanged. measures (list mlr3::Measure) Measures optimize. NULL, mlr3's default measure used. check_values (logical(1)) Check parameters evaluation results validity? store_benchmark_result (logical(1)) Store benchmark result archive? store_models (logical(1)). Store models benchmark result? archive (ArchiveBatchFSelect) Reference archive FSelectInstanceBatchSingleCrit | FSelectInstanceBatchMultiCrit. NULL (default), benchmark result models stored. callbacks (list CallbackBatchFSelect) List callbacks.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelectBatch.html","id":"method-clone-","dir":"Reference","previous_headings":"","what":"Method clone()","title":"Class for Feature Selection Objective — ObjectiveFSelectBatch","text":"objects class cloneable method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelectBatch.html","id":"usage-1","dir":"Reference","previous_headings":"","what":"Usage","title":"Class for Feature Selection Objective — ObjectiveFSelectBatch","text":"","code":"ObjectiveFSelectBatch$clone(deep = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ObjectiveFSelectBatch.html","id":"arguments-1","dir":"Reference","previous_headings":"","what":"Arguments","title":"Class for Feature Selection Objective — ObjectiveFSelectBatch","text":"deep Whether make deep clone.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/auto_fselector.html","id":null,"dir":"Reference","previous_headings":"","what":"Function for Automatic Feature Selection — auto_fselector","title":"Function for Automatic Feature Selection — auto_fselector","text":"AutoFSelector wraps mlr3::Learner augments automatic feature selection. auto_fselector() function creates AutoFSelector object.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/auto_fselector.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Function for Automatic Feature Selection — auto_fselector","text":"","code":"auto_fselector( fselector, learner, resampling, measure = NULL, term_evals = NULL, term_time = NULL, terminator = NULL, store_fselect_instance = TRUE, store_benchmark_result = TRUE, store_models = FALSE, check_values = FALSE, callbacks = NULL, ties_method = \"least_features\" )"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/auto_fselector.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Function for Automatic Feature Selection — auto_fselector","text":"fselector (FSelector) Optimization algorithm. learner (mlr3::Learner) Learner optimize feature subset . resampling (mlr3::Resampling) Resampling used evaluated performance feature subsets. Uninstantiated resamplings instantiated construction feature subsets evaluated data splits. Already instantiated resamplings kept unchanged. measure (mlr3::Measure) Measure optimize. NULL, default measure used. term_evals (integer(1)) Number allowed evaluations. Ignored terminator passed. term_time (integer(1)) Maximum allowed time seconds. Ignored terminator passed. terminator (bbotk::Terminator) Stop criterion feature selection. store_fselect_instance (logical(1)) TRUE (default), stores internally created FSelectInstanceBatchSingleCrit intermediate results slot $fselect_instance. set TRUE, store_models = TRUE store_benchmark_result (logical(1)) Store benchmark result archive? store_models (logical(1)). Store models benchmark result? check_values (logical(1)) Check parameters evaluation results validity? callbacks (list CallbackBatchFSelect) List callbacks. ties_method (character(1)) method break ties selecting sets optimizing selecting best set. Can \"least_features\" \"random\". option \"least_features\" (default) selects feature set least features. multiple best feature sets number features, one selected randomly. random method returns random feature set best feature sets. Ignored multiple measures used.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/auto_fselector.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Function for Automatic Feature Selection — auto_fselector","text":"AutoFSelector.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/auto_fselector.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Function for Automatic Feature Selection — auto_fselector","text":"AutoFSelector mlr3::Learner wraps another mlr3::Learner performs following steps $train(): wrapped (inner) learner trained feature subsets via resampling. feature selection can specified providing FSelector, bbotk::Terminator, mlr3::Resampling mlr3::Measure. final model fit complete training data best-found feature subset. $predict() AutoFSelector just calls predict method wrapped (inner) learner.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/auto_fselector.html","id":"resources","dir":"Reference","previous_headings":"","what":"Resources","title":"Function for Automatic Feature Selection — auto_fselector","text":"several sections feature selection mlr3book. Estimate Model Performance nested resampling. gallery features collection case studies demos optimization.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/auto_fselector.html","id":"nested-resampling","dir":"Reference","previous_headings":"","what":"Nested Resampling","title":"Function for Automatic Feature Selection — auto_fselector","text":"Nested resampling can performed passing AutoFSelector object mlr3::resample() mlr3::benchmark(). access inner resampling results, set store_fselect_instance = TRUE execute mlr3::resample() mlr3::benchmark() store_models = TRUE (see examples). mlr3::Resampling passed AutoFSelector meant inner resampling, operating training set arbitrary outer resampling. reason feasible pass instantiated mlr3::Resampling .","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/auto_fselector.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Function for Automatic Feature Selection — auto_fselector","text":"","code":"# Automatic Feature Selection # \\donttest{ # split to train and external set task = tsk(\"penguins\") split = partition(task, ratio = 0.8) # create auto fselector afs = auto_fselector( fselector = fs(\"random_search\"), learner = lrn(\"classif.rpart\"), resampling = rsmp (\"holdout\"), measure = msr(\"classif.ce\"), term_evals = 4) # optimize feature subset and fit final model afs$train(task, row_ids = split$train) # predict with final model afs$predict(task, row_ids = split$test) #> for 69 observations: #> row_ids truth response #> 5 Adelie Adelie #> 11 Adelie Adelie #> 12 Adelie Adelie #> --- #> 338 Chinstrap Chinstrap #> 339 Chinstrap Chinstrap #> 340 Chinstrap Chinstrap # show result afs$fselect_result #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE TRUE TRUE TRUE FALSE TRUE FALSE #> features n_features classif.ce #> #> 1: bill_depth,bill_length,body_mass,flipper_length,sex 5 0.06521739 # model slot contains trained learner and fselect instance afs$model #> $learner #> : Classification Tree #> * Model: rpart #> * Parameters: xval=0 #> * Packages: mlr3, rpart #> * Predict Types: [response], prob #> * Feature Types: logical, integer, numeric, factor, ordered #> * Properties: importance, missings, multiclass, selected_features, #> twoclass, weights #> #> $features #> [1] \"bill_depth\" \"bill_length\" \"body_mass\" \"flipper_length\" #> [5] \"sex\" #> #> $fselect_instance #> #> * State: Optimized #> * Objective: #> * Terminator: #> * Result: #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE TRUE TRUE TRUE FALSE TRUE FALSE #> classif.ce #> #> 1: 0.06521739 #> * Archive: #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE FALSE FALSE TRUE TRUE TRUE TRUE #> 2: TRUE TRUE TRUE FALSE TRUE TRUE TRUE #> 3: FALSE TRUE TRUE FALSE FALSE FALSE FALSE #> 4: TRUE TRUE TRUE FALSE TRUE TRUE TRUE #> 5: TRUE TRUE TRUE TRUE FALSE TRUE TRUE #> 6: TRUE TRUE TRUE TRUE FALSE TRUE FALSE #> 7: FALSE FALSE FALSE FALSE TRUE FALSE FALSE #> 8: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> 9: FALSE FALSE FALSE TRUE TRUE FALSE TRUE #> 10: TRUE TRUE TRUE FALSE TRUE TRUE TRUE #> classif.ce #> #> 1: 0.16304348 #> 2: 0.07608696 #> 3: 0.13043478 #> 4: 0.07608696 #> 5: 0.06521739 #> 6: 0.06521739 #> 7: 0.31521739 #> 8: 0.06521739 #> 9: 0.16304348 #> 10: 0.07608696 #> # shortcut trained learner afs$learner #> : Classification Tree #> * Model: rpart #> * Parameters: xval=0 #> * Packages: mlr3, rpart #> * Predict Types: [response], prob #> * Feature Types: logical, integer, numeric, factor, ordered #> * Properties: importance, missings, multiclass, selected_features, #> twoclass, weights # shortcut fselect instance afs$fselect_instance #> #> * State: Optimized #> * Objective: #> * Terminator: #> * Result: #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE TRUE TRUE TRUE FALSE TRUE FALSE #> classif.ce #> #> 1: 0.06521739 #> * Archive: #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE FALSE FALSE TRUE TRUE TRUE TRUE #> 2: TRUE TRUE TRUE FALSE TRUE TRUE TRUE #> 3: FALSE TRUE TRUE FALSE FALSE FALSE FALSE #> 4: TRUE TRUE TRUE FALSE TRUE TRUE TRUE #> 5: TRUE TRUE TRUE TRUE FALSE TRUE TRUE #> 6: TRUE TRUE TRUE TRUE FALSE TRUE FALSE #> 7: FALSE FALSE FALSE FALSE TRUE FALSE FALSE #> 8: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> 9: FALSE FALSE FALSE TRUE TRUE FALSE TRUE #> 10: TRUE TRUE TRUE FALSE TRUE TRUE TRUE #> classif.ce #> #> 1: 0.16304348 #> 2: 0.07608696 #> 3: 0.13043478 #> 4: 0.07608696 #> 5: 0.06521739 #> 6: 0.06521739 #> 7: 0.31521739 #> 8: 0.06521739 #> 9: 0.16304348 #> 10: 0.07608696 # Nested Resampling afs = auto_fselector( fselector = fs(\"random_search\"), learner = lrn(\"classif.rpart\"), resampling = rsmp (\"holdout\"), measure = msr(\"classif.ce\"), term_evals = 4) resampling_outer = rsmp(\"cv\", folds = 3) rr = resample(task, afs, resampling_outer, store_models = TRUE) # retrieve inner feature selection results. extract_inner_fselect_results(rr) #> iteration bill_depth bill_length body_mass flipper_length island sex #> #> 1: 1 TRUE TRUE TRUE TRUE TRUE FALSE #> 2: 2 TRUE TRUE TRUE FALSE TRUE FALSE #> 3: 3 TRUE TRUE TRUE TRUE TRUE TRUE #> year classif.ce #> #> 1: FALSE 0.06578947 #> 2: TRUE 0.03947368 #> 3: TRUE 0.06493506 #> features n_features #> #> 1: bill_depth,bill_length,body_mass,flipper_length,island 5 #> 2: bill_depth,bill_length,body_mass,island,year 5 #> 3: bill_depth,bill_length,body_mass,flipper_length,island,sex,... 7 #> task_id learner_id resampling_id #> #> 1: penguins classif.rpart.fselector cv #> 2: penguins classif.rpart.fselector cv #> 3: penguins classif.rpart.fselector cv # performance scores estimated on the outer resampling rr$score() #> task_id learner_id resampling_id iteration classif.ce #> #> 1: penguins classif.rpart.fselector cv 1 0.05217391 #> 2: penguins classif.rpart.fselector cv 2 0.05217391 #> 3: penguins classif.rpart.fselector cv 3 0.07017544 #> Hidden columns: task, learner, resampling, prediction # unbiased performance of the final model trained on the full data set rr$aggregate() #> classif.ce #> 0.05817442 # }"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/callback_batch_fselect.html","id":null,"dir":"Reference","previous_headings":"","what":"Create Feature Selection Callback — callback_batch_fselect","title":"Create Feature Selection Callback — callback_batch_fselect","text":"Function create CallbackBatchFSelect. Predefined callbacks stored dictionary mlr_callbacks can retrieved clbk(). Feature selection callbacks can called different stages feature selection. stages prefixed on_*. See also section parameters information stages. feature selection callback works bbotk::ContextBatch ContextBatchFSelect.","code":"Start Feature Selection - on_optimization_begin Start FSelect Batch - on_optimizer_before_eval Start Evaluation - on_eval_after_design - on_eval_after_benchmark - on_eval_before_archive End Evaluation - on_optimizer_after_eval End FSelect Batch - on_result - on_optimization_end End Feature Selection"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/callback_batch_fselect.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Create Feature Selection Callback — callback_batch_fselect","text":"","code":"callback_batch_fselect( id, label = NA_character_, man = NA_character_, on_optimization_begin = NULL, on_optimizer_before_eval = NULL, on_eval_after_design = NULL, on_eval_after_benchmark = NULL, on_eval_before_archive = NULL, on_optimizer_after_eval = NULL, on_result = NULL, on_optimization_end = NULL )"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/callback_batch_fselect.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Create Feature Selection Callback — callback_batch_fselect","text":"id (character(1)) Identifier new instance. label (character(1)) Label new instance. man (character(1)) String format [pkg]::[topic] pointing manual page object. referenced help package can opened via method $help(). on_optimization_begin (function()) Stage called beginning optimization. Called Optimizer$optimize(). on_optimizer_before_eval (function()) Stage called optimizer proposes points. Called OptimInstance$eval_batch(). on_eval_after_design (function()) Stage called design created. Called ObjectiveFSelectBatch$eval_many(). on_eval_after_benchmark (function()) Stage called feature sets evaluated. Called ObjectiveFSelectBatch$eval_many(). on_eval_before_archive (function()) Stage called performance values written archive. Called ObjectiveFSelectBatch$eval_many(). on_optimizer_after_eval (function()) Stage called points evaluated. Called OptimInstance$eval_batch(). on_result (function()) Stage called result written. Called OptimInstance$assign_result(). on_optimization_end (function()) Stage called end optimization. Called Optimizer$optimize().","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/callback_batch_fselect.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Create Feature Selection Callback — callback_batch_fselect","text":"implementing callback, function must two arguments named callback context. callback can write data state ($state), e.g. settings affect callback . Avoid writing large data state. can slow feature selection evaluation configurations parallelized. Feature selection callbacks access two different contexts depending stage. stages on_eval_after_design, on_eval_after_benchmark, on_eval_before_archive access ContextBatchFSelect. context can used customize evaluation batch feature sets. Changes state callback lost evaluation batch changes fselect instance fselector possible. Persistent data written archive via $aggregated_performance (see ContextBatchFSelect). stages access bbotk::ContextBatch. context can used modify fselect instance, archive, fselector final result. two different contexts evaluation can parallelized .e. multiple instances ContextBatchFSelect exists different workers time.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/callback_batch_fselect.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Create Feature Selection Callback — callback_batch_fselect","text":"","code":"# Write archive to disk callback_batch_fselect(\"mlr3fselect.backup\", on_optimization_end = function(callback, context) { saveRDS(context$instance$archive, \"archive.rds\") } ) #> #> * Active Stages: on_optimization_end"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":null,"dir":"Reference","previous_headings":"","what":"Ensemble Feature Selection Result — ensemble_fs_result","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"EnsembleFSResult stores results ensemble feature selection. includes methods evaluating stability feature selection process ranking selected features among others. function ensemble_fselect() returns object class.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"s-methods","dir":"Reference","previous_headings":"","what":"S3 Methods","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":".data.table.EnsembleFSResult(x, benchmark_result = TRUE) Returns tabular view ensemble feature selection. EnsembleFSResult -> data.table::data.table() x (EnsembleFSResult) benchmark_result (logical(1)) Whether add learner, task resampling information benchmark result.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"Das, (1999). “characterizing 'knee' Pareto curve based normal-boundary intersection.” Structural Optimization, 18(1-2), 107–115. ISSN 09344373.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"public-fields","dir":"Reference","previous_headings":"","what":"Public fields","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"benchmark_result (mlr3::BenchmarkResult) benchmark result. man (character(1)) Manual page object.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"active-bindings","dir":"Reference","previous_headings":"","what":"Active bindings","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"result (data.table::data.table) Returns result ensemble feature selection. n_learners (numeric(1)) Returns number learners used ensemble feature selection. measure (character(1)) Returns measure id used ensemble feature selection.","code":""},{"path":[]},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"public-methods","dir":"Reference","previous_headings":"","what":"Public methods","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"EnsembleFSResult$new() EnsembleFSResult$format() EnsembleFSResult$print() EnsembleFSResult$help() EnsembleFSResult$feature_ranking() EnsembleFSResult$stability() EnsembleFSResult$pareto_front() EnsembleFSResult$knee_points() EnsembleFSResult$clone()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"method-new-","dir":"Reference","previous_headings":"","what":"Method new()","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"Creates new instance R6 class.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"","code":"EnsembleFSResult$new( result, features, benchmark_result = NULL, measure_id, minimize = TRUE )"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"result (data.table::data.table) result ensemble feature selection. Column names include \"resampling_iteration\", \"learner_id\", \"features\" \"n_features\". features (character()) vector features task used ensemble feature selection. benchmark_result (mlr3::BenchmarkResult) benchmark result object. measure_id (character(1)) Column name \"result\" corresponds measure used. minimize (logical(1)) TRUE (default), lower values measure correspond higher performance.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"method-format-","dir":"Reference","previous_headings":"","what":"Method format()","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"Helper print outputs.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"usage-1","dir":"Reference","previous_headings":"","what":"Usage","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"","code":"EnsembleFSResult$format(...)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"arguments-1","dir":"Reference","previous_headings":"","what":"Arguments","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"... (ignored).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"method-print-","dir":"Reference","previous_headings":"","what":"Method print()","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"Printer.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"usage-2","dir":"Reference","previous_headings":"","what":"Usage","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"","code":"EnsembleFSResult$print(...)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"arguments-2","dir":"Reference","previous_headings":"","what":"Arguments","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"... (ignored).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"method-help-","dir":"Reference","previous_headings":"","what":"Method help()","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"Opens corresponding help page referenced field $man.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"usage-3","dir":"Reference","previous_headings":"","what":"Usage","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"","code":"EnsembleFSResult$help()"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"method-feature-ranking-","dir":"Reference","previous_headings":"","what":"Method feature_ranking()","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"Calculates feature ranking.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"usage-4","dir":"Reference","previous_headings":"","what":"Usage","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"","code":"EnsembleFSResult$feature_ranking(method = \"approval_voting\")"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"arguments-3","dir":"Reference","previous_headings":"","what":"Arguments","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"method (character(1)) method calculate feature ranking.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"feature ranking process built following framework: models act voters, features act candidates, voters select certain candidates (features). primary objective compile selections consensus ranked list features, effectively forming committee. Currently, \"approval_voting\" method supported, selects candidates/features highest approval score selection frequency, .e. appear often.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"returns","dir":"Reference","previous_headings":"","what":"Returns","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"data.table::data.table listing features, ordered decreasing inclusion probability scores (depending method)","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"method-stability-","dir":"Reference","previous_headings":"","what":"Method stability()","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"Calculates stability selected features stabm package. results cached. stability measure requested different arguments, cache must reset.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"usage-5","dir":"Reference","previous_headings":"","what":"Usage","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"","code":"EnsembleFSResult$stability( stability_measure = \"jaccard\", stability_args = NULL, global = TRUE, reset_cache = FALSE )"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"arguments-4","dir":"Reference","previous_headings":"","what":"Arguments","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"stability_measure (character(1)) stability measure used. One measures returned stabm::listStabilityMeasures() lower case. Default \"jaccard\". stability_args (list) Additional arguments passed stability measure function. global (logical(1)) Whether calculate stability globally learner. reset_cache (logical(1)) TRUE, cached results ignored.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"returns-1","dir":"Reference","previous_headings":"","what":"Returns","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"numeric() value representing stability selected features. numeric() vector stability selected features learner.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"method-pareto-front-","dir":"Reference","previous_headings":"","what":"Method pareto_front()","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"function identifies Pareto front ensemble feature selection process, .e., set points represent trade-number features performance (e.g. classification error).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"usage-6","dir":"Reference","previous_headings":"","what":"Usage","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"","code":"EnsembleFSResult$pareto_front(type = \"empirical\")"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"arguments-5","dir":"Reference","previous_headings":"","what":"Arguments","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"type (character(1)) Specifies type Pareto front return. See details.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"details-1","dir":"Reference","previous_headings":"","what":"Details","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"Two options available Pareto front: \"empirical\" (default): returns empirical Pareto front. \"estimated\": Pareto front points estimated fitting linear model inversed number features (\\(1/x\\)) input associated performance scores output. method useful Pareto points sparse front assumes convex shape better performance corresponds lower measure values (e.g. classification error), concave shape otherwise (e.g. classification accuracy). estimated Pareto front include points number features ranging 1 maximum number found empirical Pareto front.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"returns-2","dir":"Reference","previous_headings":"","what":"Returns","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"data.table::data.table columns number features performance together form Pareto front.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"method-knee-points-","dir":"Reference","previous_headings":"","what":"Method knee_points()","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"function implements various knee point identification (KPI) methods, select points Pareto front, optimal trade-performance number features achieved. cases, one point returned.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"usage-7","dir":"Reference","previous_headings":"","what":"Usage","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"","code":"EnsembleFSResult$knee_points(method = \"NBI\", type = \"empirical\")"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"arguments-6","dir":"Reference","previous_headings":"","what":"Arguments","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"method (character(1)) Type method use identify knee point. See details. type (character(1)) Specifies type Pareto front use identification knee point. See pareto_front() method details.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"details-2","dir":"Reference","previous_headings":"","what":"Details","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"available KPI methods : \"NBI\" (default): Normal-Boundary Intersection method geometry-based method calculates perpendicular distance point line connecting first last points Pareto front. knee point determined Pareto point maximum distance line, see Das (1999).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"returns-3","dir":"Reference","previous_headings":"","what":"Returns","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"data.table::data.table knee point(s) Pareto front.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"method-clone-","dir":"Reference","previous_headings":"","what":"Method clone()","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"objects class cloneable method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"usage-8","dir":"Reference","previous_headings":"","what":"Usage","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"","code":"EnsembleFSResult$clone(deep = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"arguments-7","dir":"Reference","previous_headings":"","what":"Arguments","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"deep Whether make deep clone.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fs_result.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Ensemble Feature Selection Result — ensemble_fs_result","text":"","code":"# \\donttest{ efsr = ensemble_fselect( fselector = fs(\"rfe\", n_features = 2, feature_fraction = 0.8), task = tsk(\"sonar\"), learners = lrns(c(\"classif.rpart\", \"classif.featureless\")), init_resampling = rsmp(\"subsampling\", repeats = 2), inner_resampling = rsmp(\"cv\", folds = 3), measure = msr(\"classif.ce\"), terminator = trm(\"none\") ) # contains the benchmark result efsr$benchmark_result #> of 4 rows with 4 resampling runs #> nr task_id learner_id resampling_id iters warnings errors #> 1 sonar classif.rpart.fselector insample 1 0 0 #> 2 sonar classif.featureless.fselector insample 1 0 0 #> 3 sonar classif.rpart.fselector insample 1 0 0 #> 4 sonar classif.featureless.fselector insample 1 0 0 # contains the selected features for each iteration efsr$result #> resampling_iteration learner_id features #> #> 1: 1 classif.rpart V1,V10,V11,V12,V13,V14,... #> 2: 1 classif.featureless V14,V9 #> 3: 2 classif.rpart V10,V11,V12,V13,V15,V44,... #> 4: 2 classif.featureless V20,V22 #> n_features classif.ce #> #> 1: 60 0.2158495 #> 2: 2 0.4750231 #> 3: 12 0.2810669 #> 4: 2 0.5610546 #> importance #> #> 1: 60.00000,58.33333,58.00000,57.66667,55.66667,49.00000,... #> 2: 1.666667,1.333333 #> 3: 11.333333,11.333333, 9.666667, 8.666667, 8.000000, 6.666667,... #> 4: 1.666667,1.333333 #> task learner #> #> 1: #> 2: #> 3: #> 4: #> resampling #> #> 1: #> 2: #> 3: #> 4: # returns the stability of the selected features efsr$stability(stability_measure = \"jaccard\") #> [1] 0.05726496 # returns a ranking of all features head(efsr$feature_ranking()) #> feature inclusion_probability #> #> 1: V9 0.75 #> 2: V10 0.50 #> 3: V11 0.50 #> 4: V12 0.50 #> 5: V13 0.50 #> 6: V14 0.50 # returns the empirical pareto front (nfeatures vs error) efsr$pareto_front() #> n_features classif.ce #> #> 1: 2 0.5610546 #> 2: 2 0.4750231 #> 3: 12 0.2810669 #> 4: 60 0.2158495 # }"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fselect.html","id":null,"dir":"Reference","previous_headings":"","what":"Ensemble Feature Selection — ensemble_fselect","title":"Ensemble Feature Selection — ensemble_fselect","text":"Ensemble feature selection using multiple learners. ensemble feature selection method designed identify informative features given dataset leveraging multiple machine learning models resampling techniques. Returns EnsembleFSResult.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fselect.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Ensemble Feature Selection — ensemble_fselect","text":"","code":"ensemble_fselect( fselector, task, learners, init_resampling, inner_resampling, measure, terminator, callbacks = NULL, store_benchmark_result = TRUE, store_models = TRUE )"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fselect.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Ensemble Feature Selection — ensemble_fselect","text":"Saeys, Yvan, Abeel, Thomas, Van De Peer, Yves (2008). “Robust feature selection using ensemble feature selection techniques.” Machine Learning Knowledge Discovery Databases, 5212 LNAI, 313–325. doi:10.1007/978-3-540-87481-2_21 . Abeel, Thomas, Helleputte, Thibault, Van de Peer, Yves, Dupont, Pierre, Saeys, Yvan (2010). “Robust biomarker identification cancer diagnosis ensemble feature selection methods.” Bioinformatics, 26, 392–398. ISSN 1367-4803, doi:10.1093/BIOINFORMATICS/BTP630 . Pes, Barbara (2020). “Ensemble feature selection high-dimensional data: stability analysis across multiple domains.” Neural Computing Applications, 32(10), 5951–5973. ISSN 14333058, doi:10.1007/s00521-019-04082-3 .","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fselect.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Ensemble Feature Selection — ensemble_fselect","text":"fselector (FSelector) Optimization algorithm. task (mlr3::Task) Task operate . learners (list mlr3::Learner) learners used feature selection. init_resampling (mlr3::Resampling) initial resampling strategy data, train set passed learners. Can mlr3::ResamplingSubsampling mlr3::ResamplingBootstrap. inner_resampling (mlr3::Resampling) inner resampling strategy used FSelector. measure (mlr3::Measure) Measure optimize. NULL, default measure used. terminator (bbotk::Terminator) Stop criterion feature selection. callbacks (list lists CallbackBatchFSelect) Callbacks used learner. lists must length number learners. store_benchmark_result (logical(1)) Whether store benchmark result EnsembleFSResult . store_models (logical(1)) Whether store models auto_fselector .","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fselect.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Ensemble Feature Selection — ensemble_fselect","text":"EnsembleFSResult object.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fselect.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Ensemble Feature Selection — ensemble_fselect","text":"method begins applying initial resampling technique specified user, create multiple subsamples original dataset. resampling process helps generating diverse subsets data robust feature selection. subsample generated previous step, method performs wrapped-based feature selection (auto_fselector) using provided learner, given inner resampling method, performance measure optimization algorithm. process generates best feature subset combination subsample learner. Results stored EnsembleFSResult.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/ensemble_fselect.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Ensemble Feature Selection — ensemble_fselect","text":"","code":"# \\donttest{ efsr = ensemble_fselect( fselector = fs(\"random_search\"), task = tsk(\"sonar\"), learners = lrns(c(\"classif.rpart\", \"classif.featureless\")), init_resampling = rsmp(\"subsampling\", repeats = 2), inner_resampling = rsmp(\"cv\", folds = 3), measure = msr(\"classif.ce\"), terminator = trm(\"evals\", n_evals = 10) ) efsr #> #> resampling_iteration learner_id n_features #> #> 1: 1 classif.rpart 23 #> 2: 1 classif.featureless 2 #> 3: 2 classif.rpart 34 #> 4: 2 classif.featureless 15 # }"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/extract_inner_fselect_archives.html","id":null,"dir":"Reference","previous_headings":"","what":"Extract Inner Feature Selection Archives — extract_inner_fselect_archives","title":"Extract Inner Feature Selection Archives — extract_inner_fselect_archives","text":"Extract inner feature selection archives nested resampling. Implemented mlr3::ResampleResult mlr3::BenchmarkResult. function iterates AutoFSelector objects binds archives data.table::data.table(). AutoFSelector must initialized store_fselect_instance = TRUE resample() benchmark() must called store_models = TRUE.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/extract_inner_fselect_archives.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Extract Inner Feature Selection Archives — extract_inner_fselect_archives","text":"","code":"extract_inner_fselect_archives(x, exclude_columns = \"uhash\")"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/extract_inner_fselect_archives.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Extract Inner Feature Selection Archives — extract_inner_fselect_archives","text":"x (mlr3::ResampleResult | mlr3::BenchmarkResult). exclude_columns (character()) Exclude columns result table. Set NULL column excluded.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/extract_inner_fselect_archives.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Extract Inner Feature Selection Archives — extract_inner_fselect_archives","text":"data.table::data.table().","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/extract_inner_fselect_archives.html","id":"data-structure","dir":"Reference","previous_headings":"","what":"Data structure","title":"Extract Inner Feature Selection Archives — extract_inner_fselect_archives","text":"returned data table following columns: experiment (integer(1)) Index, giving according row number original benchmark grid. iteration (integer(1)) Iteration outer resampling. One column feature task. One column performance measure. runtime_learners (numeric(1)) Sum training predict times logged learners per mlr3::ResampleResult / evaluation. include potential overhead time. timestamp (POSIXct) Time stamp evaluation logged archive. batch_nr (integer(1)) Feature sets evaluated batches. batch unique batch number. resample_result (mlr3::ResampleResult) Resample result inner resampling. task_id (character(1)). learner_id (character(1)). resampling_id (character(1)).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/extract_inner_fselect_archives.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Extract Inner Feature Selection Archives — extract_inner_fselect_archives","text":"","code":"# Nested Resampling on Palmer Penguins Data Set # create auto fselector at = auto_fselector( fselector = fs(\"random_search\"), learner = lrn(\"classif.rpart\"), resampling = rsmp (\"holdout\"), measure = msr(\"classif.ce\"), term_evals = 4) resampling_outer = rsmp(\"cv\", folds = 2) rr = resample(tsk(\"penguins\"), at, resampling_outer, store_models = TRUE) # extract inner archives extract_inner_fselect_archives(rr) #> iteration bill_depth bill_length body_mass flipper_length island sex #> #> 1: 1 TRUE TRUE FALSE TRUE TRUE TRUE #> 2: 1 FALSE TRUE FALSE FALSE FALSE FALSE #> 3: 1 TRUE TRUE TRUE FALSE TRUE TRUE #> 4: 1 TRUE TRUE TRUE TRUE FALSE TRUE #> 5: 1 FALSE FALSE FALSE FALSE TRUE FALSE #> 6: 1 TRUE TRUE FALSE TRUE TRUE TRUE #> 7: 1 FALSE TRUE FALSE FALSE TRUE FALSE #> 8: 1 TRUE FALSE TRUE TRUE TRUE FALSE #> 9: 1 TRUE FALSE FALSE TRUE FALSE TRUE #> 10: 1 FALSE FALSE TRUE TRUE FALSE FALSE #> 11: 2 FALSE FALSE FALSE FALSE FALSE FALSE #> 12: 2 FALSE FALSE FALSE TRUE FALSE FALSE #> 13: 2 FALSE FALSE TRUE FALSE FALSE FALSE #> 14: 2 TRUE TRUE TRUE TRUE TRUE TRUE #> 15: 2 TRUE FALSE TRUE TRUE TRUE FALSE #> 16: 2 TRUE TRUE FALSE TRUE TRUE TRUE #> 17: 2 FALSE TRUE FALSE FALSE FALSE FALSE #> 18: 2 TRUE TRUE TRUE FALSE TRUE TRUE #> 19: 2 FALSE FALSE FALSE TRUE TRUE FALSE #> 20: 2 FALSE FALSE TRUE FALSE FALSE TRUE #> iteration bill_depth bill_length body_mass flipper_length island sex #> year classif.ce runtime_learners timestamp batch_nr warnings #> #> 1: TRUE 0.03508772 0.007 2024-07-24 12:01:03 1 0 #> 2: TRUE 0.19298246 0.005 2024-07-24 12:01:03 1 0 #> 3: FALSE 0.03508772 0.005 2024-07-24 12:01:03 1 0 #> 4: FALSE 0.03508772 0.005 2024-07-24 12:01:03 1 0 #> 5: TRUE 0.29824561 0.004 2024-07-24 12:01:03 1 0 #> 6: TRUE 0.03508772 0.005 2024-07-24 12:01:03 1 0 #> 7: FALSE 0.03508772 0.005 2024-07-24 12:01:03 1 0 #> 8: FALSE 0.10526316 0.005 2024-07-24 12:01:03 1 0 #> 9: FALSE 0.15789474 0.005 2024-07-24 12:01:03 1 0 #> 10: FALSE 0.17543860 0.004 2024-07-24 12:01:03 1 0 #> 11: TRUE 0.57894737 0.005 2024-07-24 12:01:03 1 0 #> 12: FALSE 0.19298246 0.004 2024-07-24 12:01:03 1 0 #> 13: FALSE 0.28070175 0.005 2024-07-24 12:01:03 1 0 #> 14: TRUE 0.14035088 0.005 2024-07-24 12:01:03 1 0 #> 15: FALSE 0.21052632 0.005 2024-07-24 12:01:03 1 0 #> 16: TRUE 0.08771930 0.005 2024-07-24 12:01:03 1 0 #> 17: FALSE 0.33333333 0.005 2024-07-24 12:01:03 1 0 #> 18: FALSE 0.14035088 0.005 2024-07-24 12:01:03 1 0 #> 19: TRUE 0.12280702 0.004 2024-07-24 12:01:03 1 0 #> 20: FALSE 0.31578947 0.004 2024-07-24 12:01:03 1 0 #> year classif.ce runtime_learners timestamp batch_nr warnings #> errors features #> #> 1: 0 bill_depth,bill_length,flipper_length,island,sex,year #> 2: 0 bill_length,year #> 3: 0 bill_depth,bill_length,body_mass,island,sex #> 4: 0 bill_depth,bill_length,body_mass,flipper_length,sex #> 5: 0 island,year #> 6: 0 bill_depth,bill_length,flipper_length,island,sex,year #> 7: 0 bill_length,island #> 8: 0 bill_depth,body_mass,flipper_length,island #> 9: 0 bill_depth,flipper_length,sex #> 10: 0 body_mass,flipper_length #> 11: 0 year #> 12: 0 flipper_length #> 13: 0 body_mass #> 14: 0 bill_depth,bill_length,body_mass,flipper_length,island,sex,... #> 15: 0 bill_depth,body_mass,flipper_length,island #> 16: 0 bill_depth,bill_length,flipper_length,island,sex,year #> 17: 0 bill_length #> 18: 0 bill_depth,bill_length,body_mass,island,sex #> 19: 0 flipper_length,island,year #> 20: 0 body_mass,sex #> errors features #> n_features resample_result task_id learner_id resampling_id #> #> 1: 6 penguins classif.rpart.fselector cv #> 2: 2 penguins classif.rpart.fselector cv #> 3: 5 penguins classif.rpart.fselector cv #> 4: 5 penguins classif.rpart.fselector cv #> 5: 2 penguins classif.rpart.fselector cv #> 6: 6 penguins classif.rpart.fselector cv #> 7: 2 penguins classif.rpart.fselector cv #> 8: 4 penguins classif.rpart.fselector cv #> 9: 3 penguins classif.rpart.fselector cv #> 10: 2 penguins classif.rpart.fselector cv #> 11: 1 penguins classif.rpart.fselector cv #> 12: 1 penguins classif.rpart.fselector cv #> 13: 1 penguins classif.rpart.fselector cv #> 14: 7 penguins classif.rpart.fselector cv #> 15: 4 penguins classif.rpart.fselector cv #> 16: 6 penguins classif.rpart.fselector cv #> 17: 1 penguins classif.rpart.fselector cv #> 18: 5 penguins classif.rpart.fselector cv #> 19: 3 penguins classif.rpart.fselector cv #> 20: 2 penguins classif.rpart.fselector cv #> n_features resample_result task_id learner_id resampling_id"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/extract_inner_fselect_results.html","id":null,"dir":"Reference","previous_headings":"","what":"Extract Inner Feature Selection Results — extract_inner_fselect_results","title":"Extract Inner Feature Selection Results — extract_inner_fselect_results","text":"Extract inner feature selection results nested resampling. Implemented mlr3::ResampleResult mlr3::BenchmarkResult.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/extract_inner_fselect_results.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Extract Inner Feature Selection Results — extract_inner_fselect_results","text":"","code":"extract_inner_fselect_results(x, fselect_instance, ...)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/extract_inner_fselect_results.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Extract Inner Feature Selection Results — extract_inner_fselect_results","text":"x (mlr3::ResampleResult | mlr3::BenchmarkResult). fselect_instance (logical(1)) TRUE, instances added table. ... () Additional arguments.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/extract_inner_fselect_results.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Extract Inner Feature Selection Results — extract_inner_fselect_results","text":"data.table::data.table().","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/extract_inner_fselect_results.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Extract Inner Feature Selection Results — extract_inner_fselect_results","text":"function iterates AutoFSelector objects binds feature selection results data.table::data.table(). AutoFSelector must initialized store_fselect_instance = TRUE resample() benchmark() must called store_models = TRUE. Optionally, instance can added iteration.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/extract_inner_fselect_results.html","id":"data-structure","dir":"Reference","previous_headings":"","what":"Data structure","title":"Extract Inner Feature Selection Results — extract_inner_fselect_results","text":"returned data table following columns: experiment (integer(1)) Index, giving according row number original benchmark grid. iteration (integer(1)) Iteration outer resampling. One column feature task. One column performance measure. features (character()) Vector selected feature set. task_id (character(1)). learner_id (character(1)). resampling_id (character(1)).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/extract_inner_fselect_results.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Extract Inner Feature Selection Results — extract_inner_fselect_results","text":"","code":"# Nested Resampling on Palmer Penguins Data Set # create auto fselector at = auto_fselector( fselector = fs(\"random_search\"), learner = lrn(\"classif.rpart\"), resampling = rsmp (\"holdout\"), measure = msr(\"classif.ce\"), term_evals = 4) resampling_outer = rsmp(\"cv\", folds = 2) rr = resample(tsk(\"iris\"), at, resampling_outer, store_models = TRUE) # extract inner results extract_inner_fselect_results(rr) #> iteration Petal.Length Petal.Width Sepal.Length Sepal.Width classif.ce #> #> 1: 1 FALSE TRUE TRUE FALSE 0.04 #> 2: 2 TRUE FALSE FALSE FALSE 0.08 #> features n_features task_id learner_id #> #> 1: Petal.Width,Sepal.Length 2 iris classif.rpart.fselector #> 2: Petal.Length 1 iris classif.rpart.fselector #> resampling_id #> #> 1: cv #> 2: cv"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fs.html","id":null,"dir":"Reference","previous_headings":"","what":"Syntactic Sugar for FSelect Construction — fs","title":"Syntactic Sugar for FSelect Construction — fs","text":"Functions retrieve objects, set parameters assign fields one go. Relies mlr3misc::dictionary_sugar_get() extract objects respective mlr3misc::Dictionary: fs() FSelector mlr_fselectors. fss() list FSelector mlr_fselectors. trm() bbotk::Terminator mlr_terminators. trms() list Terminators mlr_terminators.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fs.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Syntactic Sugar for FSelect Construction — fs","text":"","code":"fs(.key, ...) fss(.keys, ...)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fs.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Syntactic Sugar for FSelect Construction — fs","text":".key (character(1)) Key passed respective dictionary retrieve object. ... () Additional arguments. .keys (character()) Keys passed respective dictionary retrieve multiple objects.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fs.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Syntactic Sugar for FSelect Construction — fs","text":"R6::R6Class object respective type, list R6::R6Class objects plural versions.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fs.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Syntactic Sugar for FSelect Construction — fs","text":"","code":"# random search with batch size of 5 fs(\"random_search\", batch_size = 5) #> : Random Search #> * Parameters: batch_size=5 #> * Properties: single-crit, multi-crit #> * Packages: mlr3fselect # run time terminator with 20 seconds trm(\"run_time\", secs = 20) #> : Run Time #> * Parameters: secs=20"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fselect.html","id":null,"dir":"Reference","previous_headings":"","what":"Function for Feature Selection — fselect","title":"Function for Feature Selection — fselect","text":"Function optimize features mlr3::Learner. function internally creates FSelectInstanceBatchSingleCrit FSelectInstanceBatchMultiCrit describes feature selection problem. executes feature selection FSelector (method) returns result fselect instance ($result). ArchiveBatchFSelect ($archive) stores evaluated hyperparameter configurations performance scores.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fselect.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Function for Feature Selection — fselect","text":"","code":"fselect( fselector, task, learner, resampling, measures = NULL, term_evals = NULL, term_time = NULL, terminator = NULL, store_benchmark_result = TRUE, store_models = FALSE, check_values = FALSE, callbacks = NULL, ties_method = \"least_features\" )"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fselect.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Function for Feature Selection — fselect","text":"fselector (FSelector) Optimization algorithm. task (mlr3::Task) Task operate . learner (mlr3::Learner) Learner optimize feature subset . resampling (mlr3::Resampling) Resampling used evaluated performance feature subsets. Uninstantiated resamplings instantiated construction feature subsets evaluated data splits. Already instantiated resamplings kept unchanged. measures (mlr3::Measure list mlr3::Measure) single measure creates FSelectInstanceBatchSingleCrit multiple measures FSelectInstanceBatchMultiCrit. NULL, default measure used. term_evals (integer(1)) Number allowed evaluations. Ignored terminator passed. term_time (integer(1)) Maximum allowed time seconds. Ignored terminator passed. terminator (bbotk::Terminator) Stop criterion feature selection. store_benchmark_result (logical(1)) Store benchmark result archive? store_models (logical(1)). Store models benchmark result? check_values (logical(1)) Check parameters evaluation results validity? callbacks (list CallbackBatchFSelect) List callbacks. ties_method (character(1)) method break ties selecting sets optimizing selecting best set. Can \"least_features\" \"random\". option \"least_features\" (default) selects feature set least features. multiple best feature sets number features, one selected randomly. random method returns random feature set best feature sets. Ignored multiple measures used.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fselect.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Function for Feature Selection — fselect","text":"FSelectInstanceBatchSingleCrit | FSelectInstanceBatchMultiCrit","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fselect.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Function for Feature Selection — fselect","text":"mlr3::Task, mlr3::Learner, mlr3::Resampling, mlr3::Measure bbotk::Terminator used construct FSelectInstanceBatchSingleCrit. multiple performance Measures supplied, FSelectInstanceBatchMultiCrit created. parameter term_evals term_time shortcuts create bbotk::Terminator. parameters passed, bbotk::TerminatorCombo constructed. Terminators, pass one terminator. termination criterion needed, set term_evals, term_time terminator NULL.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fselect.html","id":"resources","dir":"Reference","previous_headings":"","what":"Resources","title":"Function for Feature Selection — fselect","text":"several sections feature selection mlr3book. Getting started wrapper feature selection. sequential forward selection Palmer Penguins data set. gallery features collection case studies demos optimization. Utilize built-feature importance models Recursive Feature Elimination. Run feature selection Shadow Variable Search. Feature Selection Titanic data set.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fselect.html","id":"analysis","dir":"Reference","previous_headings":"","what":"Analysis","title":"Function for Feature Selection — fselect","text":"analyzing feature selection results, recommended pass archive .data.table(). returned data table joined benchmark result adds mlr3::ResampleResult feature set. archive provides various getters (e.g. $learners()) ease access. getters extract position () unique hash (uhash). complete list getters see methods section. benchmark result ($benchmark_result) allows score feature sets different measure. Alternatively, measures can supplied .data.table().","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fselect.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Function for Feature Selection — fselect","text":"","code":"# Feature selection on the Palmer Penguins data set task = tsk(\"pima\") learner = lrn(\"classif.rpart\") # Run feature selection instance = fselect( fselector = fs(\"random_search\"), task = task, learner = learner, resampling = rsmp (\"holdout\"), measures = msr(\"classif.ce\"), term_evals = 4) # Subset task to optimized feature set task$select(instance$result_feature_set) # Train the learner with optimal feature set on the full data set learner$train(task) # Inspect all evaluated configurations as.data.table(instance$archive) #> age glucose insulin mass pedigree pregnant pressure triceps classif.ce #> #> 1: TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE 0.2500000 #> 2: FALSE FALSE TRUE FALSE TRUE TRUE FALSE FALSE 0.2929688 #> 3: FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE 0.3203125 #> 4: FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE 0.3281250 #> 5: TRUE TRUE TRUE FALSE TRUE FALSE TRUE TRUE 0.2851562 #> 6: TRUE TRUE FALSE FALSE TRUE FALSE TRUE TRUE 0.2851562 #> 7: TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE 0.2500000 #> 8: TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE 0.2500000 #> 9: TRUE TRUE TRUE FALSE TRUE TRUE FALSE FALSE 0.2734375 #> 10: TRUE TRUE TRUE TRUE TRUE FALSE TRUE FALSE 0.2734375 #> runtime_learners timestamp batch_nr warnings errors #> #> 1: 0.009 2024-07-24 12:01:05 1 0 0 #> 2: 0.009 2024-07-24 12:01:05 1 0 0 #> 3: 0.006 2024-07-24 12:01:05 1 0 0 #> 4: 0.006 2024-07-24 12:01:05 1 0 0 #> 5: 0.008 2024-07-24 12:01:05 1 0 0 #> 6: 0.007 2024-07-24 12:01:05 1 0 0 #> 7: 0.009 2024-07-24 12:01:05 1 0 0 #> 8: 0.008 2024-07-24 12:01:05 1 0 0 #> 9: 0.007 2024-07-24 12:01:05 1 0 0 #> 10: 0.008 2024-07-24 12:01:05 1 0 0 #> features n_features resample_result #> #> 1: age,glucose,insulin,mass,pedigree,pregnant,... 8 #> 2: insulin,pedigree,pregnant 3 #> 3: pedigree 1 #> 4: pregnant 1 #> 5: age,glucose,insulin,pedigree,pressure,triceps 6 #> 6: age,glucose,pedigree,pressure,triceps 5 #> 7: age,glucose,insulin,mass,pedigree,pregnant,... 8 #> 8: age,glucose,insulin,mass,pedigree,pregnant,... 8 #> 9: age,glucose,insulin,pedigree,pregnant 5 #> 10: age,glucose,insulin,mass,pedigree,pressure 6 "},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fselect_nested.html","id":null,"dir":"Reference","previous_headings":"","what":"Function for Nested Resampling — fselect_nested","title":"Function for Nested Resampling — fselect_nested","text":"Function conduct nested resampling.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fselect_nested.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Function for Nested Resampling — fselect_nested","text":"","code":"fselect_nested( fselector, task, learner, inner_resampling, outer_resampling, measure = NULL, term_evals = NULL, term_time = NULL, terminator = NULL, store_fselect_instance = TRUE, store_benchmark_result = TRUE, store_models = FALSE, check_values = FALSE, callbacks = NULL, ties_method = \"least_features\" )"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fselect_nested.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Function for Nested Resampling — fselect_nested","text":"fselector (FSelector) Optimization algorithm. task (mlr3::Task) Task operate . learner (mlr3::Learner) Learner optimize feature subset . inner_resampling (mlr3::Resampling) Resampling used inner loop. outer_resampling mlr3::Resampling) Resampling used outer loop. measure (mlr3::Measure) Measure optimize. NULL, default measure used. term_evals (integer(1)) Number allowed evaluations. Ignored terminator passed. term_time (integer(1)) Maximum allowed time seconds. Ignored terminator passed. terminator (bbotk::Terminator) Stop criterion feature selection. store_fselect_instance (logical(1)) TRUE (default), stores internally created FSelectInstanceBatchSingleCrit intermediate results slot $fselect_instance. set TRUE, store_models = TRUE store_benchmark_result (logical(1)) Store benchmark result archive? store_models (logical(1)). Store models benchmark result? check_values (logical(1)) Check parameters evaluation results validity? callbacks (list CallbackBatchFSelect) List callbacks. ties_method (character(1)) method break ties selecting sets optimizing selecting best set. Can \"least_features\" \"random\". option \"least_features\" (default) selects feature set least features. multiple best feature sets number features, one selected randomly. random method returns random feature set best feature sets. Ignored multiple measures used.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fselect_nested.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Function for Nested Resampling — fselect_nested","text":"mlr3::ResampleResult","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fselect_nested.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Function for Nested Resampling — fselect_nested","text":"","code":"# Nested resampling on Palmer Penguins data set rr = fselect_nested( fselector = fs(\"random_search\"), task = tsk(\"penguins\"), learner = lrn(\"classif.rpart\"), inner_resampling = rsmp (\"holdout\"), outer_resampling = rsmp(\"cv\", folds = 2), measure = msr(\"classif.ce\"), term_evals = 4) # Performance scores estimated on the outer resampling rr$score() #> task_id learner_id resampling_id iteration classif.ce #> #> 1: penguins classif.rpart.fselector cv 1 0.05813953 #> 2: penguins classif.rpart.fselector cv 2 0.08139535 #> Hidden columns: task, learner, resampling, prediction # Unbiased performance of the final model trained on the full data set rr$aggregate() #> classif.ce #> 0.06976744"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fsi.html","id":null,"dir":"Reference","previous_headings":"","what":"Syntactic Sugar for Instance Construction — fsi","title":"Syntactic Sugar for Instance Construction — fsi","text":"Function construct FSelectInstanceBatchSingleCrit FSelectInstanceBatchMultiCrit.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fsi.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Syntactic Sugar for Instance Construction — fsi","text":"","code":"fsi( task, learner, resampling, measures = NULL, terminator, store_benchmark_result = TRUE, store_models = FALSE, check_values = FALSE, callbacks = NULL, ties_method = \"least_features\" )"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fsi.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Syntactic Sugar for Instance Construction — fsi","text":"task (mlr3::Task) Task operate . learner (mlr3::Learner) Learner optimize feature subset . resampling (mlr3::Resampling) Resampling used evaluated performance feature subsets. Uninstantiated resamplings instantiated construction feature subsets evaluated data splits. Already instantiated resamplings kept unchanged. measures (mlr3::Measure list mlr3::Measure) single measure creates FSelectInstanceBatchSingleCrit multiple measures FSelectInstanceBatchMultiCrit. NULL, default measure used. terminator (bbotk::Terminator) Stop criterion feature selection. store_benchmark_result (logical(1)) Store benchmark result archive? store_models (logical(1)). Store models benchmark result? check_values (logical(1)) Check parameters evaluation results validity? callbacks (list CallbackBatchFSelect) List callbacks. ties_method (character(1)) method break ties selecting sets optimizing selecting best set. Can \"least_features\" \"random\". option \"least_features\" (default) selects feature set least features. multiple best feature sets number features, one selected randomly. random method returns random feature set best feature sets. Ignored multiple measures used.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fsi.html","id":"resources","dir":"Reference","previous_headings":"","what":"Resources","title":"Syntactic Sugar for Instance Construction — fsi","text":"several sections feature selection mlr3book. Getting started wrapper feature selection. sequential forward selection Palmer Penguins data set. gallery features collection case studies demos optimization. Utilize built-feature importance models Recursive Feature Elimination. Run feature selection Shadow Variable Search. Feature Selection Titanic data set.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fsi.html","id":"default-measures","dir":"Reference","previous_headings":"","what":"Default Measures","title":"Syntactic Sugar for Instance Construction — fsi","text":"measure passed, default measure used. default measure depends task type.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/fsi.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Syntactic Sugar for Instance Construction — fsi","text":"","code":"# Feature selection on Palmer Penguins data set # \\donttest{ task = tsk(\"penguins\") learner = lrn(\"classif.rpart\") # Construct feature selection instance instance = fsi( task = task, learner = learner, resampling = rsmp(\"cv\", folds = 3), measures = msr(\"classif.ce\"), terminator = trm(\"evals\", n_evals = 4) ) # Choose optimization algorithm fselector = fs(\"random_search\", batch_size = 2) # Run feature selection fselector$optimize(instance) #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE TRUE FALSE FALSE TRUE TRUE TRUE #> features n_features classif.ce #> #> 1: bill_depth,bill_length,island,sex,year 5 0.06114925 # Subset task to optimal feature set task$select(instance$result_feature_set) # Train the learner with optimal feature set on the full data set learner$train(task) # Inspect all evaluated sets as.data.table(instance$archive) #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE TRUE TRUE FALSE TRUE TRUE FALSE #> 2: FALSE FALSE FALSE TRUE FALSE TRUE FALSE #> 3: FALSE TRUE FALSE TRUE TRUE TRUE TRUE #> 4: TRUE TRUE FALSE FALSE TRUE TRUE TRUE #> classif.ce runtime_learners timestamp batch_nr warnings errors #> #> 1: 0.06114925 0.015 2024-07-24 12:01:07 1 0 0 #> 2: 0.19471142 0.014 2024-07-24 12:01:07 1 0 0 #> 3: 0.06687007 0.015 2024-07-24 12:01:07 2 0 0 #> 4: 0.06114925 0.015 2024-07-24 12:01:07 2 0 0 #> features n_features resample_result #> #> 1: bill_depth,bill_length,body_mass,island,sex 5 #> 2: flipper_length,sex 2 #> 3: bill_length,flipper_length,island,sex,year 5 #> 4: bill_depth,bill_length,island,sex,year 5 # }"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr3fselect-package.html","id":null,"dir":"Reference","previous_headings":"","what":"mlr3fselect: Feature Selection for 'mlr3' — mlr3fselect-package","title":"mlr3fselect: Feature Selection for 'mlr3' — mlr3fselect-package","text":"Feature selection package 'mlr3' ecosystem. selects optimal feature set 'mlr3' learner. package works several optimization algorithms e.g. Random Search, Recursive Feature Elimination, Genetic Search. Moreover, can automatically optimize learners estimate performance optimized feature sets nested resampling.","code":""},{"path":[]},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr3fselect-package.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"mlr3fselect: Feature Selection for 'mlr3' — mlr3fselect-package","text":"Maintainer: Marc Becker marcbecker@posteo.de (ORCID) Authors: Patrick Schratz patrick.schratz@gmail.com (ORCID) Michel Lang michellang@gmail.com (ORCID) Bernd Bischl bernd_bischl@gmx.net (ORCID) John Zobolas bblodfon@gmail.com (ORCID)","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr3fselect.backup.html","id":null,"dir":"Reference","previous_headings":"","what":"Backup Benchmark Result Callback — mlr3fselect.backup","title":"Backup Benchmark Result Callback — mlr3fselect.backup","text":"CallbackBatchFSelect writes mlr3::BenchmarkResult batch disk.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr3fselect.backup.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Backup Benchmark Result Callback — mlr3fselect.backup","text":"","code":"clbk(\"mlr3fselect.backup\", path = \"backup.rds\") #> : Backup Benchmark Result Callback #> * Active Stages: on_optimizer_after_eval, on_optimization_begin # Run feature selection on the Palmer Penguins data set instance = fselect( fselector = fs(\"random_search\"), task = tsk(\"pima\"), learner = lrn(\"classif.rpart\"), resampling = rsmp (\"holdout\"), measures = msr(\"classif.ce\"), term_evals = 4, callbacks = clbk(\"mlr3fselect.backup\", path = tempfile(fileext = \".rds\")))"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr3fselect.one_se_rule.html","id":null,"dir":"Reference","previous_headings":"","what":"One Standard Error Rule Callback — mlr3fselect.one_se_rule","title":"One Standard Error Rule Callback — mlr3fselect.one_se_rule","text":"Selects smallest feature set within one standard error best result. multiple feature sets number features, first one selected. sets exactly performance different number features, one smallest number features selected.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr3fselect.one_se_rule.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"One Standard Error Rule Callback — mlr3fselect.one_se_rule","text":"Kuhn, Max, Johnson, Kjell (2013). “Applied Predictive Modeling.” chapter -Fitting Model Tuning, 61–92. Springer New York, New York, NY. ISBN 978-1-4614-6849-3.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr3fselect.one_se_rule.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"One Standard Error Rule Callback — mlr3fselect.one_se_rule","text":"","code":"clbk(\"mlr3fselect.one_se_rule\") #> : One Standard Error Rule Callback #> * Active Stages: on_result # Run feature selection on the pima data set with the callback instance = fselect( fselector = fs(\"random_search\"), task = tsk(\"pima\"), learner = lrn(\"classif.rpart\"), resampling = rsmp (\"cv\", folds = 3), measures = msr(\"classif.ce\"), term_evals = 10, callbacks = clbk(\"mlr3fselect.one_se_rule\")) # Smallest feature set within one standard error of the best instance$result #> age glucose insulin mass pedigree pregnant pressure triceps #> #> 1: TRUE TRUE TRUE TRUE FALSE FALSE TRUE FALSE #> features n_features classif.ce #> #> 1: age,glucose,insulin,mass,pressure 5 0.2578125"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr3fselect.svm_rfe.html","id":null,"dir":"Reference","previous_headings":"","what":"SVM-RFE Callback — mlr3fselect.svm_rfe","title":"SVM-RFE Callback — mlr3fselect.svm_rfe","text":"Runs recursive feature elimination mlr3learners::LearnerClassifSVM. SVM must configured type = \"C-classification\" kernel = \"linear\".","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr3fselect.svm_rfe.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"SVM-RFE Callback — mlr3fselect.svm_rfe","text":"Guyon , Weston J, Barnhill S, Vapnik V (2002). “Gene Selection Cancer Classification using Support Vector Machines.” Machine Learning, 46(1), 389–422. ISSN 1573-0565, doi:10.1023/:1012487302797 .","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr3fselect.svm_rfe.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"SVM-RFE Callback — mlr3fselect.svm_rfe","text":"","code":"clbk(\"mlr3fselect.svm_rfe\") #> : SVM-RFE Callback #> * Active Stages: on_optimization_begin library(mlr3learners) # Create instance with classification svm with linear kernel instance = fsi( task = tsk(\"sonar\"), learner = lrn(\"classif.svm\", type = \"C-classification\", kernel = \"linear\"), resampling = rsmp(\"cv\", folds = 3), measures = msr(\"classif.ce\"), terminator = trm(\"none\"), callbacks = clbk(\"mlr3fselect.svm_rfe\"), store_models = TRUE ) fselector = fs(\"rfe\", feature_number = 5, n_features = 10) # Run recursive feature elimination on the Sonar data set fselector$optimize(instance) #> V1 V10 V11 V12 V13 V14 V15 V16 V17 V18 V19 #> #> 1: TRUE FALSE TRUE TRUE FALSE TRUE FALSE TRUE FALSE FALSE FALSE #> V2 V20 V21 V22 V23 V24 V25 V26 V27 V28 V29 #> #> 1: FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE #> V3 V30 V31 V32 V33 V34 V35 V36 V37 V38 V39 #> #> 1: TRUE TRUE TRUE TRUE FALSE FALSE FALSE TRUE TRUE TRUE FALSE #> V4 V40 V41 V42 V43 V44 V45 V46 V47 V48 V49 #> #> 1: TRUE TRUE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE #> V5 V50 V51 V52 V53 V54 V55 V56 V57 V58 V59 #> #> 1: FALSE TRUE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE #> V6 V60 V7 V8 V9 #> #> 1: FALSE FALSE TRUE FALSE FALSE #> importance #> #> 1: 20.00000,17.33333,16.33333,15.66667,15.00000,13.00000,... #> features n_features classif.ce #> #> 1: V1,V11,V12,V14,V16,V23,... 20 0.1538992"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr3fselect_assertions.html","id":null,"dir":"Reference","previous_headings":"","what":"Assertion for mlr3fselect objects — mlr3fselect_assertions","title":"Assertion for mlr3fselect objects — mlr3fselect_assertions","text":"assertion functions ensure right class attribute, optionally additional properties.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr3fselect_assertions.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Assertion for mlr3fselect objects — mlr3fselect_assertions","text":"","code":"assert_fselectors(fselectors) assert_fselector_async(fselector) assert_fselector_batch(fselector) assert_fselect_instance(inst) assert_fselect_instance_async(inst) assert_fselect_instance_batch(inst)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr3fselect_assertions.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Assertion for mlr3fselect objects — mlr3fselect_assertions","text":"fselectors (list FSelector). fselector (FSelectorBatch). inst (FSelectInstanceBatchSingleCrit | FSelectInstanceBatchMultiCrit).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors.html","id":null,"dir":"Reference","previous_headings":"","what":"Dictionary of FSelectors — mlr_fselectors","title":"Dictionary of FSelectors — mlr_fselectors","text":"mlr3misc::Dictionary storing objects class FSelector. fselector associated help page, see mlr_fselectors_[id]. convenient way retrieve construct fselectors, see fs()/fss().","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Dictionary of FSelectors — mlr_fselectors","text":"R6::R6Class object inheriting mlr3misc::Dictionary.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Dictionary of FSelectors — mlr_fselectors","text":"See mlr3misc::Dictionary.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors.html","id":"s-methods","dir":"Reference","previous_headings":"","what":"S3 methods","title":"Dictionary of FSelectors — mlr_fselectors","text":".data.table(dict, ..., objects = FALSE)mlr3misc::Dictionary -> data.table::data.table() Returns data.table::data.table() fields \"key\", \"label\", \"properties\" \"packages\" columns. objects set TRUE, constructed objects returned list column named object.","code":""},{"path":[]},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Dictionary of FSelectors — mlr_fselectors","text":"","code":"as.data.table(mlr_fselectors) #> Key: #> key label #> #> 1: design_points Design Points #> 2: exhaustive_search Exhaustive Search #> 3: genetic_search Genetic Search #> 4: random_search Random Search #> 5: rfe Recursive Feature Elimination #> 6: rfecv Recursive Feature Elimination #> 7: sequential Sequential Search #> 8: shadow_variable_search Shadow Variable Search #> properties packages #> #> 1: dependencies,single-crit,multi-crit mlr3fselect,bbotk #> 2: single-crit,multi-crit mlr3fselect #> 3: single-crit mlr3fselect,genalg #> 4: single-crit,multi-crit mlr3fselect #> 5: single-crit,requires_model mlr3fselect #> 6: single-crit,requires_model mlr3fselect #> 7: single-crit mlr3fselect #> 8: single-crit mlr3fselect mlr_fselectors$get(\"random_search\") #> : Random Search #> * Parameters: batch_size=10 #> * Properties: single-crit, multi-crit #> * Packages: mlr3fselect fs(\"random_search\") #> : Random Search #> * Parameters: batch_size=10 #> * Properties: single-crit, multi-crit #> * Packages: mlr3fselect"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_design_points.html","id":null,"dir":"Reference","previous_headings":"","what":"Feature Selection with Design Points — mlr_fselectors_design_points","title":"Feature Selection with Design Points — mlr_fselectors_design_points","text":"Feature selection using user-defined feature sets.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_design_points.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Feature Selection with Design Points — mlr_fselectors_design_points","text":"feature sets evaluated order given. feature selection terminates feature sets evaluated. necessary set termination criterion.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_design_points.html","id":"dictionary","dir":"Reference","previous_headings":"","what":"Dictionary","title":"Feature Selection with Design Points — mlr_fselectors_design_points","text":"FSelector can instantiated associated sugar function fs():","code":"fs(\"design_points\")"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_design_points.html","id":"parameters","dir":"Reference","previous_headings":"","what":"Parameters","title":"Feature Selection with Design Points — mlr_fselectors_design_points","text":"batch_size integer(1) Maximum number configurations try batch. design data.table::data.table Design points try search, one per row.","code":""},{"path":[]},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_design_points.html","id":"super-classes","dir":"Reference","previous_headings":"","what":"Super classes","title":"Feature Selection with Design Points — mlr_fselectors_design_points","text":"mlr3fselect::FSelector -> mlr3fselect::FSelectorBatch -> mlr3fselect::FSelectorBatchFromOptimizerBatch -> FSelectorBatchDesignPoints","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_design_points.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Feature Selection with Design Points — mlr_fselectors_design_points","text":"mlr3fselect::FSelector$format() mlr3fselect::FSelector$help() mlr3fselect::FSelector$print() mlr3fselect::FSelectorBatchFromOptimizerBatch$optimize()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_design_points.html","id":"public-methods","dir":"Reference","previous_headings":"","what":"Public methods","title":"Feature Selection with Design Points — mlr_fselectors_design_points","text":"FSelectorBatchDesignPoints$new() FSelectorBatchDesignPoints$clone()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_design_points.html","id":"method-new-","dir":"Reference","previous_headings":"","what":"Method new()","title":"Feature Selection with Design Points — mlr_fselectors_design_points","text":"Creates new instance R6 class.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_design_points.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Feature Selection with Design Points — mlr_fselectors_design_points","text":"","code":"FSelectorBatchDesignPoints$new()"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_design_points.html","id":"method-clone-","dir":"Reference","previous_headings":"","what":"Method clone()","title":"Feature Selection with Design Points — mlr_fselectors_design_points","text":"objects class cloneable method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_design_points.html","id":"usage-1","dir":"Reference","previous_headings":"","what":"Usage","title":"Feature Selection with Design Points — mlr_fselectors_design_points","text":"","code":"FSelectorBatchDesignPoints$clone(deep = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_design_points.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Feature Selection with Design Points — mlr_fselectors_design_points","text":"deep Whether make deep clone.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_design_points.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Feature Selection with Design Points — mlr_fselectors_design_points","text":"","code":"# Feature Selection # \\donttest{ # retrieve task and load learner task = tsk(\"pima\") learner = lrn(\"classif.rpart\") # create design design = mlr3misc::rowwise_table( ~age, ~glucose, ~insulin, ~mass, ~pedigree, ~pregnant, ~pressure, ~triceps, TRUE, FALSE, TRUE, TRUE, FALSE, TRUE, FALSE, TRUE, TRUE, TRUE, FALSE, TRUE, FALSE, TRUE, FALSE, FALSE, TRUE, FALSE, TRUE, TRUE, FALSE, TRUE, FALSE, FALSE, TRUE, FALSE, TRUE, TRUE, FALSE, TRUE, TRUE, TRUE ) # run feature selection on the Pima Indians diabetes data set instance = fselect( fselector = fs(\"design_points\", design = design), task = task, learner = learner, resampling = rsmp(\"holdout\"), measure = msr(\"classif.ce\") ) # best performing feature set instance$result #> age glucose insulin mass pedigree pregnant pressure triceps #> #> 1: TRUE FALSE TRUE TRUE FALSE TRUE FALSE TRUE #> features n_features classif.ce #> #> 1: age,insulin,mass,pregnant,triceps 5 0.2617188 # all evaluated feature sets as.data.table(instance$archive) #> age glucose insulin mass pedigree pregnant pressure triceps classif.ce #> #> 1: TRUE FALSE TRUE TRUE FALSE TRUE FALSE TRUE 0.2617188 #> 2: TRUE TRUE FALSE TRUE FALSE TRUE FALSE FALSE 0.2734375 #> 3: TRUE FALSE TRUE TRUE FALSE TRUE FALSE FALSE 0.2734375 #> 4: TRUE FALSE TRUE TRUE FALSE TRUE TRUE TRUE 0.2617188 #> runtime_learners timestamp batch_nr warnings errors #> #> 1: 0.009 2024-07-24 12:01:12 1 0 0 #> 2: 0.008 2024-07-24 12:01:12 2 0 0 #> 3: 0.008 2024-07-24 12:01:12 3 0 0 #> 4: 0.008 2024-07-24 12:01:12 4 0 0 #> features n_features resample_result #> #> 1: age,insulin,mass,pregnant,triceps 5 #> 2: age,glucose,mass,pregnant 4 #> 3: age,insulin,mass,pregnant 4 #> 4: age,insulin,mass,pregnant,pressure,triceps 6 # subset the task and fit the final model task$select(instance$result_feature_set) learner$train(task) # }"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_exhaustive_search.html","id":null,"dir":"Reference","previous_headings":"","what":"Feature Selection with Exhaustive Search — mlr_fselectors_exhaustive_search","title":"Feature Selection with Exhaustive Search — mlr_fselectors_exhaustive_search","text":"Feature Selection using Exhaustive Search Algorithm. Exhaustive Search generates possible feature sets.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_exhaustive_search.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Feature Selection with Exhaustive Search — mlr_fselectors_exhaustive_search","text":"feature selection terminates feature sets evaluated. necessary set termination criterion.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_exhaustive_search.html","id":"dictionary","dir":"Reference","previous_headings":"","what":"Dictionary","title":"Feature Selection with Exhaustive Search — mlr_fselectors_exhaustive_search","text":"FSelector can instantiated associated sugar function fs():","code":"fs(\"exhaustive_search\")"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_exhaustive_search.html","id":"control-parameters","dir":"Reference","previous_headings":"","what":"Control Parameters","title":"Feature Selection with Exhaustive Search — mlr_fselectors_exhaustive_search","text":"max_features integer(1) Maximum number features. default, number features mlr3::Task.","code":""},{"path":[]},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_exhaustive_search.html","id":"super-classes","dir":"Reference","previous_headings":"","what":"Super classes","title":"Feature Selection with Exhaustive Search — mlr_fselectors_exhaustive_search","text":"mlr3fselect::FSelector -> mlr3fselect::FSelectorBatch -> FSelectorBatchExhaustiveSearch","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_exhaustive_search.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Feature Selection with Exhaustive Search — mlr_fselectors_exhaustive_search","text":"mlr3fselect::FSelector$format() mlr3fselect::FSelector$help() mlr3fselect::FSelector$print() mlr3fselect::FSelectorBatch$optimize()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_exhaustive_search.html","id":"public-methods","dir":"Reference","previous_headings":"","what":"Public methods","title":"Feature Selection with Exhaustive Search — mlr_fselectors_exhaustive_search","text":"FSelectorBatchExhaustiveSearch$new() FSelectorBatchExhaustiveSearch$clone()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_exhaustive_search.html","id":"method-new-","dir":"Reference","previous_headings":"","what":"Method new()","title":"Feature Selection with Exhaustive Search — mlr_fselectors_exhaustive_search","text":"Creates new instance R6 class.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_exhaustive_search.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Feature Selection with Exhaustive Search — mlr_fselectors_exhaustive_search","text":"","code":"FSelectorBatchExhaustiveSearch$new()"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_exhaustive_search.html","id":"method-clone-","dir":"Reference","previous_headings":"","what":"Method clone()","title":"Feature Selection with Exhaustive Search — mlr_fselectors_exhaustive_search","text":"objects class cloneable method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_exhaustive_search.html","id":"usage-1","dir":"Reference","previous_headings":"","what":"Usage","title":"Feature Selection with Exhaustive Search — mlr_fselectors_exhaustive_search","text":"","code":"FSelectorBatchExhaustiveSearch$clone(deep = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_exhaustive_search.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Feature Selection with Exhaustive Search — mlr_fselectors_exhaustive_search","text":"deep Whether make deep clone.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_exhaustive_search.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Feature Selection with Exhaustive Search — mlr_fselectors_exhaustive_search","text":"","code":"# Feature Selection # \\donttest{ # retrieve task and load learner task = tsk(\"penguins\") learner = lrn(\"classif.rpart\") # run feature selection on the Palmer Penguins data set instance = fselect( fselector = fs(\"exhaustive_search\"), task = task, learner = learner, resampling = rsmp(\"holdout\"), measure = msr(\"classif.ce\"), term_evals = 10 ) # best performing feature set instance$result #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE TRUE FALSE FALSE FALSE FALSE FALSE #> features n_features classif.ce #> #> 1: bill_depth,bill_length 2 0.09565217 # all evaluated feature sets as.data.table(instance$archive) #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE FALSE FALSE FALSE FALSE FALSE FALSE #> 2: FALSE TRUE FALSE FALSE FALSE FALSE FALSE #> 3: FALSE FALSE TRUE FALSE FALSE FALSE FALSE #> 4: FALSE FALSE FALSE TRUE FALSE FALSE FALSE #> 5: FALSE FALSE FALSE FALSE TRUE FALSE FALSE #> 6: FALSE FALSE FALSE FALSE FALSE TRUE FALSE #> 7: FALSE FALSE FALSE FALSE FALSE FALSE TRUE #> 8: TRUE TRUE FALSE FALSE FALSE FALSE FALSE #> 9: TRUE FALSE TRUE FALSE FALSE FALSE FALSE #> 10: TRUE FALSE FALSE TRUE FALSE FALSE FALSE #> classif.ce runtime_learners timestamp batch_nr warnings errors #> #> 1: 0.26086957 0.005 2024-07-24 12:01:13 1 0 0 #> 2: 0.21739130 0.003 2024-07-24 12:01:13 1 0 0 #> 3: 0.29565217 0.005 2024-07-24 12:01:13 1 0 0 #> 4: 0.15652174 0.005 2024-07-24 12:01:13 1 0 0 #> 5: 0.27826087 0.004 2024-07-24 12:01:13 1 0 0 #> 6: 0.57391304 0.004 2024-07-24 12:01:13 1 0 0 #> 7: 0.57391304 0.004 2024-07-24 12:01:13 1 0 0 #> 8: 0.09565217 0.005 2024-07-24 12:01:13 1 0 0 #> 9: 0.25217391 0.004 2024-07-24 12:01:13 1 0 0 #> 10: 0.15652174 0.003 2024-07-24 12:01:13 1 0 0 #> features n_features resample_result #> #> 1: bill_depth 1 #> 2: bill_length 1 #> 3: body_mass 1 #> 4: flipper_length 1 #> 5: island 1 #> 6: sex 1 #> 7: year 1 #> 8: bill_depth,bill_length 2 #> 9: bill_depth,body_mass 2 #> 10: bill_depth,flipper_length 2 # subset the task and fit the final model task$select(instance$result_feature_set) learner$train(task) # }"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_genetic_search.html","id":null,"dir":"Reference","previous_headings":"","what":"Feature Selection with Genetic Search — mlr_fselectors_genetic_search","title":"Feature Selection with Genetic Search — mlr_fselectors_genetic_search","text":"Feature selection using Genetic Algorithm package genalg.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_genetic_search.html","id":"dictionary","dir":"Reference","previous_headings":"","what":"Dictionary","title":"Feature Selection with Genetic Search — mlr_fselectors_genetic_search","text":"FSelector can instantiated associated sugar function fs():","code":"fs(\"genetic_search\")"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_genetic_search.html","id":"control-parameters","dir":"Reference","previous_headings":"","what":"Control Parameters","title":"Feature Selection with Genetic Search — mlr_fselectors_genetic_search","text":"meaning control parameters, see genalg::rbga.bin(). genalg::rbga.bin() internally terminates iters iteration. set ìters = 100000 allow termination via terminators. iterations needed, set ìters higher value parameter set.","code":""},{"path":[]},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_genetic_search.html","id":"super-classes","dir":"Reference","previous_headings":"","what":"Super classes","title":"Feature Selection with Genetic Search — mlr_fselectors_genetic_search","text":"mlr3fselect::FSelector -> mlr3fselect::FSelectorBatch -> FSelectorBatchGeneticSearch","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_genetic_search.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Feature Selection with Genetic Search — mlr_fselectors_genetic_search","text":"mlr3fselect::FSelector$format() mlr3fselect::FSelector$help() mlr3fselect::FSelector$print() mlr3fselect::FSelectorBatch$optimize()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_genetic_search.html","id":"public-methods","dir":"Reference","previous_headings":"","what":"Public methods","title":"Feature Selection with Genetic Search — mlr_fselectors_genetic_search","text":"FSelectorBatchGeneticSearch$new() FSelectorBatchGeneticSearch$clone()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_genetic_search.html","id":"method-new-","dir":"Reference","previous_headings":"","what":"Method new()","title":"Feature Selection with Genetic Search — mlr_fselectors_genetic_search","text":"Creates new instance R6 class.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_genetic_search.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Feature Selection with Genetic Search — mlr_fselectors_genetic_search","text":"","code":"FSelectorBatchGeneticSearch$new()"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_genetic_search.html","id":"method-clone-","dir":"Reference","previous_headings":"","what":"Method clone()","title":"Feature Selection with Genetic Search — mlr_fselectors_genetic_search","text":"objects class cloneable method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_genetic_search.html","id":"usage-1","dir":"Reference","previous_headings":"","what":"Usage","title":"Feature Selection with Genetic Search — mlr_fselectors_genetic_search","text":"","code":"FSelectorBatchGeneticSearch$clone(deep = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_genetic_search.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Feature Selection with Genetic Search — mlr_fselectors_genetic_search","text":"deep Whether make deep clone.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_genetic_search.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Feature Selection with Genetic Search — mlr_fselectors_genetic_search","text":"","code":"# Feature Selection # \\donttest{ # retrieve task and load learner task = tsk(\"penguins\") learner = lrn(\"classif.rpart\") # run feature selection on the Palmer Penguins data set instance = fselect( fselector = fs(\"genetic_search\"), task = task, learner = learner, resampling = rsmp(\"holdout\"), measure = msr(\"classif.ce\"), term_evals = 10 ) # best performing feature set instance$result #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: FALSE TRUE TRUE FALSE FALSE FALSE FALSE #> features n_features classif.ce #> #> 1: bill_length,body_mass 2 0.03478261 # all evaluated feature sets as.data.table(instance$archive) #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE FALSE FALSE FALSE FALSE FALSE FALSE #> 2: FALSE FALSE TRUE FALSE FALSE FALSE FALSE #> 3: TRUE FALSE FALSE TRUE FALSE FALSE FALSE #> 4: FALSE TRUE FALSE FALSE FALSE FALSE FALSE #> 5: FALSE FALSE FALSE TRUE FALSE FALSE FALSE #> 6: TRUE FALSE FALSE FALSE FALSE FALSE FALSE #> 7: FALSE TRUE TRUE FALSE FALSE FALSE FALSE #> 8: FALSE TRUE FALSE FALSE FALSE FALSE FALSE #> 9: FALSE FALSE FALSE FALSE TRUE FALSE FALSE #> 10: FALSE TRUE FALSE TRUE TRUE FALSE FALSE #> classif.ce runtime_learners timestamp batch_nr warnings errors #> #> 1: 0.30434783 0.005 2024-07-24 12:01:14 1 0 0 #> 2: 0.31304348 0.004 2024-07-24 12:01:14 2 0 0 #> 3: 0.24347826 0.005 2024-07-24 12:01:14 3 0 0 #> 4: 0.24347826 0.004 2024-07-24 12:01:14 4 0 0 #> 5: 0.22608696 0.004 2024-07-24 12:01:14 5 0 0 #> 6: 0.30434783 0.005 2024-07-24 12:01:14 6 0 0 #> 7: 0.03478261 0.005 2024-07-24 12:01:14 7 0 0 #> 8: 0.24347826 0.005 2024-07-24 12:01:14 8 0 0 #> 9: 0.37391304 0.005 2024-07-24 12:01:14 9 0 0 #> 10: 0.06086957 0.005 2024-07-24 12:01:14 10 0 0 #> features n_features resample_result #> #> 1: bill_depth 1 #> 2: body_mass 1 #> 3: bill_depth,flipper_length 2 #> 4: bill_length 1 #> 5: flipper_length 1 #> 6: bill_depth 1 #> 7: bill_length,body_mass 2 #> 8: bill_length 1 #> 9: island 1 #> 10: bill_length,flipper_length,island 3 # subset the task and fit the final model task$select(instance$result_feature_set) learner$train(task) # }"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_random_search.html","id":null,"dir":"Reference","previous_headings":"","what":"Feature Selection with Random Search — mlr_fselectors_random_search","title":"Feature Selection with Random Search — mlr_fselectors_random_search","text":"Feature selection using Random Search Algorithm.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_random_search.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Feature Selection with Random Search — mlr_fselectors_random_search","text":"Bergstra J, Bengio Y (2012). “Random Search Hyper-Parameter Optimization.” Journal Machine Learning Research, 13(10), 281–305. https://jmlr.csail.mit.edu/papers/v13/bergstra12a.html.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_random_search.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Feature Selection with Random Search — mlr_fselectors_random_search","text":"feature sets randomly drawn. sets evaluated batches size batch_size. Larger batches mean can parallelize , smaller batches imply fine-grained checking termination criteria.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_random_search.html","id":"dictionary","dir":"Reference","previous_headings":"","what":"Dictionary","title":"Feature Selection with Random Search — mlr_fselectors_random_search","text":"FSelector can instantiated associated sugar function fs():","code":"fs(\"random_search\")"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_random_search.html","id":"control-parameters","dir":"Reference","previous_headings":"","what":"Control Parameters","title":"Feature Selection with Random Search — mlr_fselectors_random_search","text":"max_features integer(1) Maximum number features. default, number features mlr3::Task. batch_size integer(1) Maximum number feature sets try batch.","code":""},{"path":[]},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_random_search.html","id":"super-classes","dir":"Reference","previous_headings":"","what":"Super classes","title":"Feature Selection with Random Search — mlr_fselectors_random_search","text":"mlr3fselect::FSelector -> mlr3fselect::FSelectorBatch -> FSelectorBatchRandomSearch","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_random_search.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Feature Selection with Random Search — mlr_fselectors_random_search","text":"mlr3fselect::FSelector$format() mlr3fselect::FSelector$help() mlr3fselect::FSelector$print() mlr3fselect::FSelectorBatch$optimize()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_random_search.html","id":"public-methods","dir":"Reference","previous_headings":"","what":"Public methods","title":"Feature Selection with Random Search — mlr_fselectors_random_search","text":"FSelectorBatchRandomSearch$new() FSelectorBatchRandomSearch$clone()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_random_search.html","id":"method-new-","dir":"Reference","previous_headings":"","what":"Method new()","title":"Feature Selection with Random Search — mlr_fselectors_random_search","text":"Creates new instance R6 class.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_random_search.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Feature Selection with Random Search — mlr_fselectors_random_search","text":"","code":"FSelectorBatchRandomSearch$new()"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_random_search.html","id":"method-clone-","dir":"Reference","previous_headings":"","what":"Method clone()","title":"Feature Selection with Random Search — mlr_fselectors_random_search","text":"objects class cloneable method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_random_search.html","id":"usage-1","dir":"Reference","previous_headings":"","what":"Usage","title":"Feature Selection with Random Search — mlr_fselectors_random_search","text":"","code":"FSelectorBatchRandomSearch$clone(deep = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_random_search.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Feature Selection with Random Search — mlr_fselectors_random_search","text":"deep Whether make deep clone.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_random_search.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Feature Selection with Random Search — mlr_fselectors_random_search","text":"","code":"# Feature Selection # \\donttest{ # retrieve task and load learner task = tsk(\"penguins\") learner = lrn(\"classif.rpart\") # run feature selection on the Palmer Penguins data set instance = fselect( fselector = fs(\"random_search\"), task = task, learner = learner, resampling = rsmp(\"holdout\"), measure = msr(\"classif.ce\"), term_evals = 10 ) # best performing feature subset instance$result #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: FALSE TRUE FALSE TRUE TRUE FALSE FALSE #> features n_features classif.ce #> #> 1: bill_length,flipper_length,island 3 0.08695652 # all evaluated feature subsets as.data.table(instance$archive) #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> 2: FALSE FALSE TRUE FALSE FALSE TRUE TRUE #> 3: FALSE TRUE FALSE TRUE TRUE FALSE FALSE #> 4: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> 5: TRUE FALSE FALSE FALSE TRUE TRUE FALSE #> 6: TRUE TRUE TRUE TRUE TRUE TRUE FALSE #> 7: FALSE FALSE FALSE TRUE FALSE FALSE FALSE #> 8: FALSE TRUE FALSE FALSE FALSE FALSE FALSE #> 9: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> 10: FALSE FALSE FALSE TRUE FALSE FALSE FALSE #> classif.ce runtime_learners timestamp batch_nr warnings errors #> #> 1: 0.08695652 0.006 2024-07-24 12:01:15 1 0 0 #> 2: 0.27826087 0.004 2024-07-24 12:01:15 1 0 0 #> 3: 0.08695652 0.005 2024-07-24 12:01:15 1 0 0 #> 4: 0.08695652 0.005 2024-07-24 12:01:15 1 0 0 #> 5: 0.21739130 0.005 2024-07-24 12:01:15 1 0 0 #> 6: 0.08695652 0.005 2024-07-24 12:01:15 1 0 0 #> 7: 0.23478261 0.003 2024-07-24 12:01:15 1 0 0 #> 8: 0.23478261 0.004 2024-07-24 12:01:15 1 0 0 #> 9: 0.08695652 0.006 2024-07-24 12:01:15 1 0 0 #> 10: 0.23478261 0.004 2024-07-24 12:01:15 1 0 0 #> features n_features #> #> 1: bill_depth,bill_length,body_mass,flipper_length,island,sex,... 7 #> 2: body_mass,sex,year 3 #> 3: bill_length,flipper_length,island 3 #> 4: bill_depth,bill_length,body_mass,flipper_length,island,sex,... 7 #> 5: bill_depth,island,sex 3 #> 6: bill_depth,bill_length,body_mass,flipper_length,island,sex 6 #> 7: flipper_length 1 #> 8: bill_length 1 #> 9: bill_depth,bill_length,body_mass,flipper_length,island,sex,... 7 #> 10: flipper_length 1 #> resample_result #> #> 1: #> 2: #> 3: #> 4: #> 5: #> 6: #> 7: #> 8: #> 9: #> 10: # subset the task and fit the final model task$select(instance$result_feature_set) learner$train(task) # }"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfe.html","id":null,"dir":"Reference","previous_headings":"","what":"Feature Selection with Recursive Feature Elimination — mlr_fselectors_rfe","title":"Feature Selection with Recursive Feature Elimination — mlr_fselectors_rfe","text":"Feature selection using Recursive Feature Elimination (RFE) algorithm. Recursive feature elimination iteratively removes features low importance score. works mlr3::Learners can calculate importance scores (see section optional extractors mlr3::Learner).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfe.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Feature Selection with Recursive Feature Elimination — mlr_fselectors_rfe","text":"Guyon , Weston J, Barnhill S, Vapnik V (2002). “Gene Selection Cancer Classification using Support Vector Machines.” Machine Learning, 46(1), 389–422. ISSN 1573-0565, doi:10.1023/:1012487302797 .","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfe.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Feature Selection with Recursive Feature Elimination — mlr_fselectors_rfe","text":"learner trained features start importance scores calculated feature. least important feature removed learner trained reduced feature set. importance scores calculated procedure repeated desired number features reached. non-recursive option (recursive = FALSE) uses importance scores calculated first iteration. feature selection terminates n_features reached. necessary set termination criterion. using cross-validation resampling strategy, importance scores resampling iterations aggregated. parameter aggregation determines importance scores aggregated. default (\"rank\"), importance score vector fold ranked feature lowest average rank removed. option \"mean\" averages score feature across resampling iterations removes feature lowest average score. Averaging scores appropriate importance measures.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfe.html","id":"archive","dir":"Reference","previous_headings":"","what":"Archive","title":"Feature Selection with Recursive Feature Elimination — mlr_fselectors_rfe","text":"ArchiveBatchFSelect holds following additional columns: \"importance\" (numeric()) importance score vector feature subset.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfe.html","id":"resources","dir":"Reference","previous_headings":"","what":"Resources","title":"Feature Selection with Recursive Feature Elimination — mlr_fselectors_rfe","text":"gallery features collection case studies demos optimization. Utilize built-feature importance models Recursive Feature Elimination.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfe.html","id":"dictionary","dir":"Reference","previous_headings":"","what":"Dictionary","title":"Feature Selection with Recursive Feature Elimination — mlr_fselectors_rfe","text":"FSelector can instantiated associated sugar function fs():","code":"fs(\"rfe\")"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfe.html","id":"control-parameters","dir":"Reference","previous_headings":"","what":"Control Parameters","title":"Feature Selection with Recursive Feature Elimination — mlr_fselectors_rfe","text":"n_features integer(1) minimum number features select, default half features. feature_fraction double(1) Fraction features retain iteration. default 0.5 retains half features. feature_number integer(1) Number features remove iteration. subset_sizes integer() Vector number features retain iteration. Must sorted decreasing order. recursive logical(1) TRUE (default), feature importance calculated iteration. aggregation character(1) aggregation method importance scores resampling iterations. See details. parameter feature_fraction, feature_number subset_sizes mutually exclusive.","code":""},{"path":[]},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfe.html","id":"super-classes","dir":"Reference","previous_headings":"","what":"Super classes","title":"Feature Selection with Recursive Feature Elimination — mlr_fselectors_rfe","text":"mlr3fselect::FSelector -> mlr3fselect::FSelectorBatch -> FSelectorBatchRFE","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfe.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Feature Selection with Recursive Feature Elimination — mlr_fselectors_rfe","text":"mlr3fselect::FSelector$format() mlr3fselect::FSelector$help() mlr3fselect::FSelector$print() mlr3fselect::FSelectorBatch$optimize()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfe.html","id":"public-methods","dir":"Reference","previous_headings":"","what":"Public methods","title":"Feature Selection with Recursive Feature Elimination — mlr_fselectors_rfe","text":"FSelectorBatchRFE$new() FSelectorBatchRFE$clone()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfe.html","id":"method-new-","dir":"Reference","previous_headings":"","what":"Method new()","title":"Feature Selection with Recursive Feature Elimination — mlr_fselectors_rfe","text":"Creates new instance R6 class.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfe.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Feature Selection with Recursive Feature Elimination — mlr_fselectors_rfe","text":"","code":"FSelectorBatchRFE$new()"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfe.html","id":"method-clone-","dir":"Reference","previous_headings":"","what":"Method clone()","title":"Feature Selection with Recursive Feature Elimination — mlr_fselectors_rfe","text":"objects class cloneable method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfe.html","id":"usage-1","dir":"Reference","previous_headings":"","what":"Usage","title":"Feature Selection with Recursive Feature Elimination — mlr_fselectors_rfe","text":"","code":"FSelectorBatchRFE$clone(deep = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfe.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Feature Selection with Recursive Feature Elimination — mlr_fselectors_rfe","text":"deep Whether make deep clone.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfe.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Feature Selection with Recursive Feature Elimination — mlr_fselectors_rfe","text":"","code":"# Feature Selection # \\donttest{ # retrieve task and load learner task = tsk(\"penguins\") learner = lrn(\"classif.rpart\") # run feature selection on the Palmer Penguins data set instance = fselect( fselector = fs(\"rfe\"), task = task, learner = learner, resampling = rsmp(\"holdout\"), measure = msr(\"classif.ce\"), store_models = TRUE ) # best performing feature subset instance$result #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> importance #> #> 1: 7,6,5,4,3,2,... #> features n_features #> #> 1: bill_depth,bill_length,body_mass,flipper_length,island,sex,... 7 #> classif.ce #> #> 1: 0.08695652 # all evaluated feature subsets as.data.table(instance$archive) #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> 2: TRUE TRUE FALSE TRUE FALSE FALSE FALSE #> classif.ce runtime_learners timestamp batch_nr warnings errors #> #> 1: 0.08695652 0.006 2024-07-24 12:01:16 1 0 0 #> 2: 0.10434783 0.005 2024-07-24 12:01:16 2 0 0 #> importance #> #> 1: 7,6,5,4,3,2,... #> 2: 3,2,1 #> features n_features #> #> 1: bill_depth,bill_length,body_mass,flipper_length,island,sex,... 7 #> 2: bill_depth,bill_length,flipper_length 3 #> resample_result #> #> 1: #> 2: # subset the task and fit the final model task$select(instance$result_feature_set) learner$train(task) # }"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfecv.html","id":null,"dir":"Reference","previous_headings":"","what":"Feature Selection with Recursive Feature Elimination with Cross Validation — mlr_fselectors_rfecv","title":"Feature Selection with Recursive Feature Elimination with Cross Validation — mlr_fselectors_rfecv","text":"Feature selection using Recursive Feature Elimination Cross-Validation (RFE-CV) algorithm. See FSelectorBatchRFE description base algorithm. RFE-CV runs recursive feature elimination iteration cross-validation determine optimal number features. recursive feature elimination run complete dataset optimal number features final feature set size. performance optimal feature set calculated complete data set reported performance final model. works mlr3::Learners can calculate importance scores (see section optional extractors mlr3::Learner).","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfecv.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Feature Selection with Recursive Feature Elimination with Cross Validation — mlr_fselectors_rfecv","text":"resampling strategy changed feature selection. resampling strategy passed instance (resampling) used determine optimal number features. Usually, cross-validation strategy used recursive feature elimination run iteration cross-validation. Internally, mlr3::ResamplingCustom used emulate part algorithm. final recursive feature elimination run resampling strategy changed mlr3::ResamplingInsample .e. complete data set used training testing. feature selection terminates optimal number features reached. necessary set termination criterion.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfecv.html","id":"archive","dir":"Reference","previous_headings":"","what":"Archive","title":"Feature Selection with Recursive Feature Elimination with Cross Validation — mlr_fselectors_rfecv","text":"ArchiveBatchFSelect holds following additional columns: \"iteration\" (integer(1)) resampling iteration feature subset evaluated. \"importance\" (numeric()) importance score vector feature subset.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfecv.html","id":"resources","dir":"Reference","previous_headings":"","what":"Resources","title":"Feature Selection with Recursive Feature Elimination with Cross Validation — mlr_fselectors_rfecv","text":"gallery features collection case studies demos optimization. Utilize built-feature importance models Recursive Feature Elimination.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfecv.html","id":"dictionary","dir":"Reference","previous_headings":"","what":"Dictionary","title":"Feature Selection with Recursive Feature Elimination with Cross Validation — mlr_fselectors_rfecv","text":"FSelector can instantiated associated sugar function fs():","code":"fs(\"rfe\")"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfecv.html","id":"control-parameters","dir":"Reference","previous_headings":"","what":"Control Parameters","title":"Feature Selection with Recursive Feature Elimination with Cross Validation — mlr_fselectors_rfecv","text":"n_features integer(1) number features select. default half features selected. feature_fraction double(1) Fraction features retain iteration. default 0.5 retrains half features. feature_number integer(1) Number features remove iteration. subset_sizes integer() Vector number features retain iteration. Must sorted decreasing order. recursive logical(1) TRUE (default), feature importance calculated iteration. parameter feature_fraction, feature_number subset_sizes mutually exclusive.","code":""},{"path":[]},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfecv.html","id":"super-classes","dir":"Reference","previous_headings":"","what":"Super classes","title":"Feature Selection with Recursive Feature Elimination with Cross Validation — mlr_fselectors_rfecv","text":"mlr3fselect::FSelector -> mlr3fselect::FSelectorBatch -> FSelectorBatchRFECV","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfecv.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Feature Selection with Recursive Feature Elimination with Cross Validation — mlr_fselectors_rfecv","text":"mlr3fselect::FSelector$format() mlr3fselect::FSelector$help() mlr3fselect::FSelector$print() mlr3fselect::FSelectorBatch$optimize()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfecv.html","id":"public-methods","dir":"Reference","previous_headings":"","what":"Public methods","title":"Feature Selection with Recursive Feature Elimination with Cross Validation — mlr_fselectors_rfecv","text":"FSelectorBatchRFECV$new() FSelectorBatchRFECV$clone()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfecv.html","id":"method-new-","dir":"Reference","previous_headings":"","what":"Method new()","title":"Feature Selection with Recursive Feature Elimination with Cross Validation — mlr_fselectors_rfecv","text":"Creates new instance R6 class.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfecv.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Feature Selection with Recursive Feature Elimination with Cross Validation — mlr_fselectors_rfecv","text":"","code":"FSelectorBatchRFECV$new()"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfecv.html","id":"method-clone-","dir":"Reference","previous_headings":"","what":"Method clone()","title":"Feature Selection with Recursive Feature Elimination with Cross Validation — mlr_fselectors_rfecv","text":"objects class cloneable method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfecv.html","id":"usage-1","dir":"Reference","previous_headings":"","what":"Usage","title":"Feature Selection with Recursive Feature Elimination with Cross Validation — mlr_fselectors_rfecv","text":"","code":"FSelectorBatchRFECV$clone(deep = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfecv.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Feature Selection with Recursive Feature Elimination with Cross Validation — mlr_fselectors_rfecv","text":"deep Whether make deep clone.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_rfecv.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Feature Selection with Recursive Feature Elimination with Cross Validation — mlr_fselectors_rfecv","text":"","code":"# Feature Selection # \\donttest{ # retrieve task and load learner task = tsk(\"penguins\") learner = lrn(\"classif.rpart\") # run feature selection on the Palmer Penguins data set instance = fselect( fselector = fs(\"rfecv\"), task = task, learner = learner, resampling = rsmp(\"cv\", folds = 3), measure = msr(\"classif.ce\"), store_models = TRUE ) # best performing feature subset instance$result #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE TRUE FALSE TRUE FALSE FALSE FALSE #> features n_features classif.ce #> #> 1: bill_depth,bill_length,flipper_length 3 0.0377907 # all evaluated feature subsets as.data.table(instance$archive) #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> 2: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> 3: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> 4: FALSE TRUE TRUE TRUE FALSE FALSE FALSE #> 5: TRUE TRUE FALSE TRUE FALSE FALSE FALSE #> 6: TRUE TRUE FALSE TRUE FALSE FALSE FALSE #> 7: TRUE TRUE TRUE TRUE TRUE TRUE TRUE #> 8: TRUE TRUE FALSE TRUE FALSE FALSE FALSE #> classif.ce runtime_learners timestamp batch_nr warnings errors #> #> 1: 0.08695652 0.006 2024-07-24 12:01:17 1 0 0 #> 2: 0.03478261 0.005 2024-07-24 12:01:17 1 0 0 #> 3: 0.03508772 0.004 2024-07-24 12:01:17 1 0 0 #> 4: 0.11304348 0.006 2024-07-24 12:01:17 2 0 0 #> 5: 0.03478261 0.005 2024-07-24 12:01:17 2 0 0 #> 6: 0.03508772 0.004 2024-07-24 12:01:17 2 0 0 #> 7: 0.03488372 0.006 2024-07-24 12:01:17 3 0 0 #> 8: 0.03779070 0.005 2024-07-24 12:01:17 4 0 0 #> importance iteration #> #> 1: 95.543823,86.523123,86.157289,83.431536,77.058416, 7.495822,... 1 #> 2: 86.78056,77.56622,62.14872,59.65407,57.93443, 0.00000,... 2 #> 3: 82.17703,79.53680,71.89359,61.50646,48.19125, 0.00000,... 3 #> 4: 94.16064,83.77806,76.82767 1 #> 5: 86.78056,77.56622,62.14872 2 #> 6: 82.17703,79.53680,71.89359 3 #> 7: 124.20793,121.52400,102.74919, 87.26186, 78.61700, 0.00000,... NA #> 8: 124.2079,121.5240,104.2507 NA #> features n_features #> #> 1: bill_depth,bill_length,body_mass,flipper_length,island,sex,... 7 #> 2: bill_depth,bill_length,body_mass,flipper_length,island,sex,... 7 #> 3: bill_depth,bill_length,body_mass,flipper_length,island,sex,... 7 #> 4: bill_length,body_mass,flipper_length 3 #> 5: bill_depth,bill_length,flipper_length 3 #> 6: bill_depth,bill_length,flipper_length 3 #> 7: bill_depth,bill_length,body_mass,flipper_length,island,sex,... 7 #> 8: bill_depth,bill_length,flipper_length 3 #> resample_result #> #> 1: #> 2: #> 3: #> 4: #> 5: #> 6: #> 7: #> 8: # subset the task and fit the final model task$select(instance$result_feature_set) learner$train(task) # }"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_sequential.html","id":null,"dir":"Reference","previous_headings":"","what":"Feature Selection with Sequential Search — mlr_fselectors_sequential","title":"Feature Selection with Sequential Search — mlr_fselectors_sequential","text":"Feature selection using Sequential Search Algorithm.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_sequential.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Feature Selection with Sequential Search — mlr_fselectors_sequential","text":"Sequential forward selection (strategy = fsf) extends feature set iteration feature increases model's performance . Sequential backward selection (strategy = fsb) follows idea starts features removes features set. feature selection terminates min_features max_features reached. necessary set termination criterion.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_sequential.html","id":"dictionary","dir":"Reference","previous_headings":"","what":"Dictionary","title":"Feature Selection with Sequential Search — mlr_fselectors_sequential","text":"FSelector can instantiated associated sugar function fs():","code":"fs(\"sequential\")"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_sequential.html","id":"control-parameters","dir":"Reference","previous_headings":"","what":"Control Parameters","title":"Feature Selection with Sequential Search — mlr_fselectors_sequential","text":"min_features integer(1) Minimum number features. default, 1. max_features integer(1) Maximum number features. default, number features mlr3::Task. strategy character(1) Search method sfs (forward search) sbs (backward search).","code":""},{"path":[]},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_sequential.html","id":"super-classes","dir":"Reference","previous_headings":"","what":"Super classes","title":"Feature Selection with Sequential Search — mlr_fselectors_sequential","text":"mlr3fselect::FSelector -> mlr3fselect::FSelectorBatch -> FSelectorBatchSequential","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_sequential.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Feature Selection with Sequential Search — mlr_fselectors_sequential","text":"mlr3fselect::FSelector$format() mlr3fselect::FSelector$help() mlr3fselect::FSelector$print() mlr3fselect::FSelectorBatch$optimize()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_sequential.html","id":"public-methods","dir":"Reference","previous_headings":"","what":"Public methods","title":"Feature Selection with Sequential Search — mlr_fselectors_sequential","text":"FSelectorBatchSequential$new() FSelectorBatchSequential$optimization_path() FSelectorBatchSequential$clone()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_sequential.html","id":"method-new-","dir":"Reference","previous_headings":"","what":"Method new()","title":"Feature Selection with Sequential Search — mlr_fselectors_sequential","text":"Creates new instance R6 class.`","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_sequential.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Feature Selection with Sequential Search — mlr_fselectors_sequential","text":"","code":"FSelectorBatchSequential$new()"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_sequential.html","id":"method-optimization-path-","dir":"Reference","previous_headings":"","what":"Method optimization_path()","title":"Feature Selection with Sequential Search — mlr_fselectors_sequential","text":"Returns optimization path.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_sequential.html","id":"usage-1","dir":"Reference","previous_headings":"","what":"Usage","title":"Feature Selection with Sequential Search — mlr_fselectors_sequential","text":"","code":"FSelectorBatchSequential$optimization_path(inst, include_uhash = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_sequential.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Feature Selection with Sequential Search — mlr_fselectors_sequential","text":"inst (FSelectInstanceBatchSingleCrit) Instance optimized FSelectorBatchSequential. include_uhash (logical(1)) Include uhash column?","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_sequential.html","id":"returns","dir":"Reference","previous_headings":"","what":"Returns","title":"Feature Selection with Sequential Search — mlr_fselectors_sequential","text":"data.table::data.table()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_sequential.html","id":"method-clone-","dir":"Reference","previous_headings":"","what":"Method clone()","title":"Feature Selection with Sequential Search — mlr_fselectors_sequential","text":"objects class cloneable method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_sequential.html","id":"usage-2","dir":"Reference","previous_headings":"","what":"Usage","title":"Feature Selection with Sequential Search — mlr_fselectors_sequential","text":"","code":"FSelectorBatchSequential$clone(deep = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_sequential.html","id":"arguments-1","dir":"Reference","previous_headings":"","what":"Arguments","title":"Feature Selection with Sequential Search — mlr_fselectors_sequential","text":"deep Whether make deep clone.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_sequential.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Feature Selection with Sequential Search — mlr_fselectors_sequential","text":"","code":"# Feature Selection # \\donttest{ # retrieve task and load learner task = tsk(\"penguins\") learner = lrn(\"classif.rpart\") # run feature selection on the Palmer Penguins data set instance = fselect( fselector = fs(\"sequential\"), task = task, learner = learner, resampling = rsmp(\"holdout\"), measure = msr(\"classif.ce\"), term_evals = 10 ) # best performing feature set instance$result #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: FALSE TRUE FALSE TRUE FALSE FALSE FALSE #> features n_features classif.ce #> #> 1: bill_length,flipper_length 2 0.05217391 # all evaluated feature sets as.data.table(instance$archive) #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE FALSE FALSE FALSE FALSE FALSE FALSE #> 2: FALSE TRUE FALSE FALSE FALSE FALSE FALSE #> 3: FALSE FALSE TRUE FALSE FALSE FALSE FALSE #> 4: FALSE FALSE FALSE TRUE FALSE FALSE FALSE #> 5: FALSE FALSE FALSE FALSE TRUE FALSE FALSE #> 6: FALSE FALSE FALSE FALSE FALSE TRUE FALSE #> 7: FALSE FALSE FALSE FALSE FALSE FALSE TRUE #> 8: TRUE FALSE FALSE TRUE FALSE FALSE FALSE #> 9: FALSE TRUE FALSE TRUE FALSE FALSE FALSE #> 10: FALSE FALSE TRUE TRUE FALSE FALSE FALSE #> 11: FALSE FALSE FALSE TRUE TRUE FALSE FALSE #> 12: FALSE FALSE FALSE TRUE FALSE TRUE FALSE #> 13: FALSE FALSE FALSE TRUE FALSE FALSE TRUE #> classif.ce runtime_learners timestamp batch_nr warnings errors #> #> 1: 0.26086957 0.005 2024-07-24 12:01:18 1 0 0 #> 2: 0.26086957 0.003 2024-07-24 12:01:18 1 0 0 #> 3: 0.30434783 0.003 2024-07-24 12:01:18 1 0 0 #> 4: 0.17391304 0.006 2024-07-24 12:01:18 1 0 0 #> 5: 0.26086957 0.006 2024-07-24 12:01:18 1 0 0 #> 6: 0.58260870 0.005 2024-07-24 12:01:18 1 0 0 #> 7: 0.58260870 0.004 2024-07-24 12:01:18 1 0 0 #> 8: 0.19130435 0.004 2024-07-24 12:01:18 2 0 0 #> 9: 0.05217391 0.003 2024-07-24 12:01:18 2 0 0 #> 10: 0.15652174 0.003 2024-07-24 12:01:18 2 0 0 #> 11: 0.11304348 0.004 2024-07-24 12:01:18 2 0 0 #> 12: 0.16521739 0.004 2024-07-24 12:01:18 2 0 0 #> 13: 0.16521739 0.004 2024-07-24 12:01:18 2 0 0 #> features n_features resample_result #> #> 1: bill_depth 1 #> 2: bill_length 1 #> 3: body_mass 1 #> 4: flipper_length 1 #> 5: island 1 #> 6: sex 1 #> 7: year 1 #> 8: bill_depth,flipper_length 2 #> 9: bill_length,flipper_length 2 #> 10: body_mass,flipper_length 2 #> 11: flipper_length,island 2 #> 12: flipper_length,sex 2 #> 13: flipper_length,year 2 # subset the task and fit the final model task$select(instance$result_feature_set) learner$train(task) # }"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_shadow_variable_search.html","id":null,"dir":"Reference","previous_headings":"","what":"Feature Selection with Shadow Variable Search — mlr_fselectors_shadow_variable_search","title":"Feature Selection with Shadow Variable Search — mlr_fselectors_shadow_variable_search","text":"Feature selection using Shadow Variable Search Algorithm. Shadow variable search creates feature permutated copy stops one selected.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_shadow_variable_search.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Feature Selection with Shadow Variable Search — mlr_fselectors_shadow_variable_search","text":"Thomas J, Hepp T, Mayr , Bischl B (2017). “Probing Sparse Fast Variable Selection Model-Based Boosting.” Computational Mathematical Methods Medicine, 2017, 1–8. doi:10.1155/2017/1421409 . Wu Y, Boos DD, Stefanski LA (2007). “Controlling Variable Selection Addition Pseudovariables.” Journal American Statistical Association, 102(477), 235–243. doi:10.1198/016214506000000843 .","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_shadow_variable_search.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Feature Selection with Shadow Variable Search — mlr_fselectors_shadow_variable_search","text":"feature selection terminates first shadow variable selected. necessary set termination criterion.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_shadow_variable_search.html","id":"resources","dir":"Reference","previous_headings":"","what":"Resources","title":"Feature Selection with Shadow Variable Search — mlr_fselectors_shadow_variable_search","text":"gallery features collection case studies demos optimization. Run feature selection Shadow Variable Search.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_shadow_variable_search.html","id":"dictionary","dir":"Reference","previous_headings":"","what":"Dictionary","title":"Feature Selection with Shadow Variable Search — mlr_fselectors_shadow_variable_search","text":"FSelector can instantiated associated sugar function fs():","code":"fs(\"shadow_variable_search\")"},{"path":[]},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_shadow_variable_search.html","id":"super-classes","dir":"Reference","previous_headings":"","what":"Super classes","title":"Feature Selection with Shadow Variable Search — mlr_fselectors_shadow_variable_search","text":"mlr3fselect::FSelector -> mlr3fselect::FSelectorBatch -> FSelectorBatchShadowVariableSearch","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_shadow_variable_search.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Feature Selection with Shadow Variable Search — mlr_fselectors_shadow_variable_search","text":"mlr3fselect::FSelector$format() mlr3fselect::FSelector$help() mlr3fselect::FSelector$print() mlr3fselect::FSelectorBatch$optimize()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_shadow_variable_search.html","id":"public-methods","dir":"Reference","previous_headings":"","what":"Public methods","title":"Feature Selection with Shadow Variable Search — mlr_fselectors_shadow_variable_search","text":"FSelectorBatchShadowVariableSearch$new() FSelectorBatchShadowVariableSearch$optimization_path() FSelectorBatchShadowVariableSearch$clone()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_shadow_variable_search.html","id":"method-new-","dir":"Reference","previous_headings":"","what":"Method new()","title":"Feature Selection with Shadow Variable Search — mlr_fselectors_shadow_variable_search","text":"Creates new instance R6 class.`","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_shadow_variable_search.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Feature Selection with Shadow Variable Search — mlr_fselectors_shadow_variable_search","text":"","code":"FSelectorBatchShadowVariableSearch$new()"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_shadow_variable_search.html","id":"method-optimization-path-","dir":"Reference","previous_headings":"","what":"Method optimization_path()","title":"Feature Selection with Shadow Variable Search — mlr_fselectors_shadow_variable_search","text":"Returns optimization path.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_shadow_variable_search.html","id":"usage-1","dir":"Reference","previous_headings":"","what":"Usage","title":"Feature Selection with Shadow Variable Search — mlr_fselectors_shadow_variable_search","text":"","code":"FSelectorBatchShadowVariableSearch$optimization_path(inst)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_shadow_variable_search.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Feature Selection with Shadow Variable Search — mlr_fselectors_shadow_variable_search","text":"inst (FSelectInstanceBatchSingleCrit) Instance optimized FSelectorBatchShadowVariableSearch.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_shadow_variable_search.html","id":"returns","dir":"Reference","previous_headings":"","what":"Returns","title":"Feature Selection with Shadow Variable Search — mlr_fselectors_shadow_variable_search","text":"data.table::data.table","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_shadow_variable_search.html","id":"method-clone-","dir":"Reference","previous_headings":"","what":"Method clone()","title":"Feature Selection with Shadow Variable Search — mlr_fselectors_shadow_variable_search","text":"objects class cloneable method.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_shadow_variable_search.html","id":"usage-2","dir":"Reference","previous_headings":"","what":"Usage","title":"Feature Selection with Shadow Variable Search — mlr_fselectors_shadow_variable_search","text":"","code":"FSelectorBatchShadowVariableSearch$clone(deep = FALSE)"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_shadow_variable_search.html","id":"arguments-1","dir":"Reference","previous_headings":"","what":"Arguments","title":"Feature Selection with Shadow Variable Search — mlr_fselectors_shadow_variable_search","text":"deep Whether make deep clone.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/mlr_fselectors_shadow_variable_search.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Feature Selection with Shadow Variable Search — mlr_fselectors_shadow_variable_search","text":"","code":"# Feature Selection # \\donttest{ # retrieve task and load learner task = tsk(\"penguins\") learner = lrn(\"classif.rpart\") # run feature selection on the Palmer Penguins data set instance = fselect( fselector = fs(\"shadow_variable_search\"), task = task, learner = learner, resampling = rsmp(\"holdout\"), measure = msr(\"classif.ce\"), ) # best performing feature subset instance$result #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: FALSE TRUE FALSE TRUE TRUE FALSE FALSE #> features n_features classif.ce #> #> 1: bill_length,flipper_length,island 3 0.06956522 # all evaluated feature subsets as.data.table(instance$archive) #> bill_depth bill_length body_mass flipper_length island sex year #> #> 1: TRUE FALSE FALSE FALSE FALSE FALSE FALSE #> 2: FALSE TRUE FALSE FALSE FALSE FALSE FALSE #> 3: FALSE FALSE TRUE FALSE FALSE FALSE FALSE #> 4: FALSE FALSE FALSE TRUE FALSE FALSE FALSE #> 5: FALSE FALSE FALSE FALSE TRUE FALSE FALSE #> 6: FALSE FALSE FALSE FALSE FALSE TRUE FALSE #> 7: FALSE FALSE FALSE FALSE FALSE FALSE TRUE #> 8: FALSE FALSE FALSE FALSE FALSE FALSE FALSE #> 9: FALSE FALSE FALSE FALSE FALSE FALSE FALSE #> 10: FALSE FALSE FALSE FALSE FALSE FALSE FALSE #> 11: FALSE FALSE FALSE FALSE FALSE FALSE FALSE #> 12: FALSE FALSE FALSE FALSE FALSE FALSE FALSE #> 13: FALSE FALSE FALSE FALSE FALSE FALSE FALSE #> 14: FALSE FALSE FALSE FALSE FALSE FALSE FALSE #> 15: TRUE FALSE FALSE TRUE FALSE FALSE FALSE #> 16: FALSE TRUE FALSE TRUE FALSE FALSE FALSE #> 17: FALSE FALSE TRUE TRUE FALSE FALSE FALSE #> 18: FALSE FALSE FALSE TRUE TRUE FALSE FALSE #> 19: FALSE FALSE FALSE TRUE FALSE TRUE FALSE #> 20: FALSE FALSE FALSE TRUE FALSE FALSE TRUE #> 21: FALSE FALSE FALSE TRUE FALSE FALSE FALSE #> 22: FALSE FALSE FALSE TRUE FALSE FALSE FALSE #> 23: FALSE FALSE FALSE TRUE FALSE FALSE FALSE #> 24: FALSE FALSE FALSE TRUE FALSE FALSE FALSE #> 25: FALSE FALSE FALSE TRUE FALSE FALSE FALSE #> 26: FALSE FALSE FALSE TRUE FALSE FALSE FALSE #> 27: FALSE FALSE FALSE TRUE FALSE FALSE FALSE #> 28: TRUE TRUE FALSE TRUE FALSE FALSE FALSE #> 29: FALSE TRUE TRUE TRUE FALSE FALSE FALSE #> 30: FALSE TRUE FALSE TRUE TRUE FALSE FALSE #> 31: FALSE TRUE FALSE TRUE FALSE TRUE FALSE #> 32: FALSE TRUE FALSE TRUE FALSE FALSE TRUE #> 33: FALSE TRUE FALSE TRUE FALSE FALSE FALSE #> 34: FALSE TRUE FALSE TRUE FALSE FALSE FALSE #> 35: FALSE TRUE FALSE TRUE FALSE FALSE FALSE #> 36: FALSE TRUE FALSE TRUE FALSE FALSE FALSE #> 37: FALSE TRUE FALSE TRUE FALSE FALSE FALSE #> 38: FALSE TRUE FALSE TRUE FALSE FALSE FALSE #> 39: FALSE TRUE FALSE TRUE FALSE FALSE FALSE #> 40: TRUE TRUE FALSE TRUE TRUE FALSE FALSE #> 41: FALSE TRUE TRUE TRUE TRUE FALSE FALSE #> 42: FALSE TRUE FALSE TRUE TRUE TRUE FALSE #> 43: FALSE TRUE FALSE TRUE TRUE FALSE TRUE #> 44: FALSE TRUE FALSE TRUE TRUE FALSE FALSE #> 45: FALSE TRUE FALSE TRUE TRUE FALSE FALSE #> 46: FALSE TRUE FALSE TRUE TRUE FALSE FALSE #> 47: FALSE TRUE FALSE TRUE TRUE FALSE FALSE #> 48: FALSE TRUE FALSE TRUE TRUE FALSE FALSE #> 49: FALSE TRUE FALSE TRUE TRUE FALSE FALSE #> 50: FALSE TRUE FALSE TRUE TRUE FALSE FALSE #> bill_depth bill_length body_mass flipper_length island sex year #> classif.ce runtime_learners timestamp batch_nr #> #> 1: 0.32173913 0.012 2024-07-24 12:01:19 1 #> 2: 0.24347826 0.012 2024-07-24 12:01:19 1 #> 3: 0.31304348 0.012 2024-07-24 12:01:19 1 #> 4: 0.19130435 0.012 2024-07-24 12:01:19 1 #> 5: 0.28695652 0.011 2024-07-24 12:01:19 1 #> 6: 0.55652174 0.010 2024-07-24 12:01:19 1 #> 7: 0.55652174 0.012 2024-07-24 12:01:19 1 #> 8: 0.60000000 0.010 2024-07-24 12:01:19 1 #> 9: 0.61739130 0.009 2024-07-24 12:01:19 1 #> 10: 0.60000000 0.010 2024-07-24 12:01:19 1 #> 11: 0.57391304 0.010 2024-07-24 12:01:19 1 #> 12: 0.58260870 0.011 2024-07-24 12:01:19 1 #> 13: 0.55652174 0.010 2024-07-24 12:01:19 1 #> 14: 0.55652174 0.010 2024-07-24 12:01:19 1 #> 15: 0.21739130 0.013 2024-07-24 12:01:19 2 #> 16: 0.07826087 0.012 2024-07-24 12:01:19 2 #> 17: 0.19130435 0.011 2024-07-24 12:01:19 2 #> 18: 0.13043478 0.012 2024-07-24 12:01:19 2 #> 19: 0.25217391 0.012 2024-07-24 12:01:19 2 #> 20: 0.19130435 0.012 2024-07-24 12:01:19 2 #> 21: 0.19130435 0.012 2024-07-24 12:01:19 2 #> 22: 0.18260870 0.012 2024-07-24 12:01:19 2 #> 23: 0.20000000 0.012 2024-07-24 12:01:19 2 #> 24: 0.20869565 0.012 2024-07-24 12:01:19 2 #> 25: 0.19130435 0.012 2024-07-24 12:01:19 2 #> 26: 0.19130435 0.012 2024-07-24 12:01:19 2 #> 27: 0.19130435 0.010 2024-07-24 12:01:19 2 #> 28: 0.07826087 0.014 2024-07-24 12:01:20 3 #> 29: 0.07826087 0.011 2024-07-24 12:01:20 3 #> 30: 0.06956522 0.012 2024-07-24 12:01:20 3 #> 31: 0.07826087 0.012 2024-07-24 12:01:20 3 #> 32: 0.07826087 0.012 2024-07-24 12:01:20 3 #> 33: 0.07826087 0.010 2024-07-24 12:01:20 3 #> 34: 0.07826087 0.011 2024-07-24 12:01:20 3 #> 35: 0.07826087 0.011 2024-07-24 12:01:20 3 #> 36: 0.07826087 0.012 2024-07-24 12:01:20 3 #> 37: 0.07826087 0.012 2024-07-24 12:01:20 3 #> 38: 0.07826087 0.012 2024-07-24 12:01:20 3 #> 39: 0.07826087 0.017 2024-07-24 12:01:20 3 #> 40: 0.06956522 0.012 2024-07-24 12:01:20 4 #> 41: 0.06956522 0.014 2024-07-24 12:01:20 4 #> 42: 0.06956522 0.013 2024-07-24 12:01:20 4 #> 43: 0.06956522 0.012 2024-07-24 12:01:20 4 #> 44: 0.06956522 0.012 2024-07-24 12:01:20 4 #> 45: 0.06956522 0.012 2024-07-24 12:01:20 4 #> 46: 0.06956522 0.013 2024-07-24 12:01:20 4 #> 47: 0.06956522 0.012 2024-07-24 12:01:20 4 #> 48: 0.06956522 0.012 2024-07-24 12:01:20 4 #> 49: 0.06956522 0.034 2024-07-24 12:01:20 4 #> 50: 0.06956522 0.017 2024-07-24 12:01:20 4 #> classif.ce runtime_learners timestamp batch_nr #> permuted__bill_depth permuted__bill_length permuted__body_mass #> #> 1: FALSE FALSE FALSE #> 2: FALSE FALSE FALSE #> 3: FALSE FALSE FALSE #> 4: FALSE FALSE FALSE #> 5: FALSE FALSE FALSE #> 6: FALSE FALSE FALSE #> 7: FALSE FALSE FALSE #> 8: TRUE FALSE FALSE #> 9: FALSE TRUE FALSE #> 10: FALSE FALSE TRUE #> 11: FALSE FALSE FALSE #> 12: FALSE FALSE FALSE #> 13: FALSE FALSE FALSE #> 14: FALSE FALSE FALSE #> 15: FALSE FALSE FALSE #> 16: FALSE FALSE FALSE #> 17: FALSE FALSE FALSE #> 18: FALSE FALSE FALSE #> 19: FALSE FALSE FALSE #> 20: FALSE FALSE FALSE #> 21: TRUE FALSE FALSE #> 22: FALSE TRUE FALSE #> 23: FALSE FALSE TRUE #> 24: FALSE FALSE FALSE #> 25: FALSE FALSE FALSE #> 26: FALSE FALSE FALSE #> 27: FALSE FALSE FALSE #> 28: FALSE FALSE FALSE #> 29: FALSE FALSE FALSE #> 30: FALSE FALSE FALSE #> 31: FALSE FALSE FALSE #> 32: FALSE FALSE FALSE #> 33: TRUE FALSE FALSE #> 34: FALSE TRUE FALSE #> 35: FALSE FALSE TRUE #> 36: FALSE FALSE FALSE #> 37: FALSE FALSE FALSE #> 38: FALSE FALSE FALSE #> 39: FALSE FALSE FALSE #> 40: FALSE FALSE FALSE #> 41: FALSE FALSE FALSE #> 42: FALSE FALSE FALSE #> 43: FALSE FALSE FALSE #> 44: TRUE FALSE FALSE #> 45: FALSE TRUE FALSE #> 46: FALSE FALSE TRUE #> 47: FALSE FALSE FALSE #> 48: FALSE FALSE FALSE #> 49: FALSE FALSE FALSE #> 50: FALSE FALSE FALSE #> permuted__bill_depth permuted__bill_length permuted__body_mass #> permuted__flipper_length permuted__island permuted__sex permuted__year #> #> 1: FALSE FALSE FALSE FALSE #> 2: FALSE FALSE FALSE FALSE #> 3: FALSE FALSE FALSE FALSE #> 4: FALSE FALSE FALSE FALSE #> 5: FALSE FALSE FALSE FALSE #> 6: FALSE FALSE FALSE FALSE #> 7: FALSE FALSE FALSE FALSE #> 8: FALSE FALSE FALSE FALSE #> 9: FALSE FALSE FALSE FALSE #> 10: FALSE FALSE FALSE FALSE #> 11: TRUE FALSE FALSE FALSE #> 12: FALSE TRUE FALSE FALSE #> 13: FALSE FALSE TRUE FALSE #> 14: FALSE FALSE FALSE TRUE #> 15: FALSE FALSE FALSE FALSE #> 16: FALSE FALSE FALSE FALSE #> 17: FALSE FALSE FALSE FALSE #> 18: FALSE FALSE FALSE FALSE #> 19: FALSE FALSE FALSE FALSE #> 20: FALSE FALSE FALSE FALSE #> 21: FALSE FALSE FALSE FALSE #> 22: FALSE FALSE FALSE FALSE #> 23: FALSE FALSE FALSE FALSE #> 24: TRUE FALSE FALSE FALSE #> 25: FALSE TRUE FALSE FALSE #> 26: FALSE FALSE TRUE FALSE #> 27: FALSE FALSE FALSE TRUE #> 28: FALSE FALSE FALSE FALSE #> 29: FALSE FALSE FALSE FALSE #> 30: FALSE FALSE FALSE FALSE #> 31: FALSE FALSE FALSE FALSE #> 32: FALSE FALSE FALSE FALSE #> 33: FALSE FALSE FALSE FALSE #> 34: FALSE FALSE FALSE FALSE #> 35: FALSE FALSE FALSE FALSE #> 36: TRUE FALSE FALSE FALSE #> 37: FALSE TRUE FALSE FALSE #> 38: FALSE FALSE TRUE FALSE #> 39: FALSE FALSE FALSE TRUE #> 40: FALSE FALSE FALSE FALSE #> 41: FALSE FALSE FALSE FALSE #> 42: FALSE FALSE FALSE FALSE #> 43: FALSE FALSE FALSE FALSE #> 44: FALSE FALSE FALSE FALSE #> 45: FALSE FALSE FALSE FALSE #> 46: FALSE FALSE FALSE FALSE #> 47: TRUE FALSE FALSE FALSE #> 48: FALSE TRUE FALSE FALSE #> 49: FALSE FALSE TRUE FALSE #> 50: FALSE FALSE FALSE TRUE #> permuted__flipper_length permuted__island permuted__sex permuted__year #> warnings errors features n_features #> #> 1: 0 0 bill_depth 1 #> 2: 0 0 bill_length 1 #> 3: 0 0 body_mass 1 #> 4: 0 0 flipper_length 1 #> 5: 0 0 island 1 #> 6: 0 0 sex 1 #> 7: 0 0 year 1 #> 8: 0 0 0 #> 9: 0 0 0 #> 10: 0 0 0 #> 11: 0 0 0 #> 12: 0 0 0 #> 13: 0 0 0 #> 14: 0 0 0 #> 15: 0 0 bill_depth,flipper_length 2 #> 16: 0 0 bill_length,flipper_length 2 #> 17: 0 0 body_mass,flipper_length 2 #> 18: 0 0 flipper_length,island 2 #> 19: 0 0 flipper_length,sex 2 #> 20: 0 0 flipper_length,year 2 #> 21: 0 0 flipper_length 1 #> 22: 0 0 flipper_length 1 #> 23: 0 0 flipper_length 1 #> 24: 0 0 flipper_length 1 #> 25: 0 0 flipper_length 1 #> 26: 0 0 flipper_length 1 #> 27: 0 0 flipper_length 1 #> 28: 0 0 bill_depth,bill_length,flipper_length 3 #> 29: 0 0 bill_length,body_mass,flipper_length 3 #> 30: 0 0 bill_length,flipper_length,island 3 #> 31: 0 0 bill_length,flipper_length,sex 3 #> 32: 0 0 bill_length,flipper_length,year 3 #> 33: 0 0 bill_length,flipper_length 2 #> 34: 0 0 bill_length,flipper_length 2 #> 35: 0 0 bill_length,flipper_length 2 #> 36: 0 0 bill_length,flipper_length 2 #> 37: 0 0 bill_length,flipper_length 2 #> 38: 0 0 bill_length,flipper_length 2 #> 39: 0 0 bill_length,flipper_length 2 #> 40: 0 0 bill_depth,bill_length,flipper_length,island 4 #> 41: 0 0 bill_length,body_mass,flipper_length,island 4 #> 42: 0 0 bill_length,flipper_length,island,sex 4 #> 43: 0 0 bill_length,flipper_length,island,year 4 #> 44: 0 0 bill_length,flipper_length,island 3 #> 45: 0 0 bill_length,flipper_length,island 3 #> 46: 0 0 bill_length,flipper_length,island 3 #> 47: 0 0 bill_length,flipper_length,island 3 #> 48: 0 0 bill_length,flipper_length,island 3 #> 49: 0 0 bill_length,flipper_length,island 3 #> 50: 0 0 bill_length,flipper_length,island 3 #> warnings errors features n_features #> resample_result #> #> 1: #> 2: #> 3: #> 4: #> 5: #> 6: #> 7: #> 8: #> 9: #> 10: #> 11: #> 12: #> 13: #> 14: #> 15: #> 16: #> 17: #> 18: #> 19: #> 20: #> 21: #> 22: #> 23: #> 24: #> 25: #> 26: #> 27: #> 28: #> 29: #> 30: #> 31: #> 32: #> 33: #> 34: #> 35: #> 36: #> 37: #> 38: #> 39: #> 40: #> 41: #> 42: #> 43: #> 44: #> 45: #> 46: #> 47: #> 48: #> 49: #> 50: #> resample_result # subset the task and fit the final model task$select(instance$result_feature_set) learner$train(task) # }"},{"path":"https://mlr3fselect.mlr-org.com/dev/reference/reexports.html","id":null,"dir":"Reference","previous_headings":"","what":"Objects exported from other packages — reexports","title":"Objects exported from other packages — reexports","text":"objects imported packages. Follow links see documentation. bbotk mlr_terminators, trm, trms mlr3misc clbk, clbks, mlr_callbacks","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/news/index.html","id":"mlr3fselect-development-version","dir":"Changelog","previous_headings":"","what":"mlr3fselect (development version)","title":"mlr3fselect (development version)","text":"fix: Delete intermediate BenchmarkResult ObjectiveFSelectBatch optimization.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/news/index.html","id":"mlr3fselect-100","dir":"Changelog","previous_headings":"","what":"mlr3fselect 1.0.0","title":"mlr3fselect 1.0.0","text":"CRAN release: 2024-06-29 feat: Add ensemble feature selection function ensemble_fselect(). BREAKING CHANGE: FSelector class FSelectorBatch now. BREAKING CHANGE: FSelectInstanceSingleCrit FSelectInstanceMultiCrit classes FSelectInstanceBatchSingleCrit FSelectInstanceBatchMultiCrit now. BREAKING CHANGE: CallbackFSelect class CallbackBatchFSelect now. BREAKING CHANGE: ContextEval class ContextBatchFSelect now.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/news/index.html","id":"mlr3fselect-0120","dir":"Changelog","previous_headings":"","what":"mlr3fselect 0.12.0","title":"mlr3fselect 0.12.0","text":"CRAN release: 2024-03-09 feat: Add number features instance$result. feat: Add ties_method options \"least_features\" \"random\" ArchiveBatchFSelect$best(). refactor: Optimize runtime ArchiveBatchFSelect$best() method. feat: Add importance scores result FSelectorRFE. feat: Add number features .data.table.ArchiveBatchFSelect(). feat: Features can always included always_include column role. fix: Add $phash() method AutoFSelector. fix: Include FSelector hash AutoFSelector. refactor: Change default batch size FSelectorBatchRandomSearch 10. feat: Add batch_size parameter FSelectorBatchExhaustiveSearch reduce memory consumption. compatibility: Work new paradox version 1.0.0","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/news/index.html","id":"mlr3fselect-0110","dir":"Changelog","previous_headings":"","what":"mlr3fselect 0.11.0","title":"mlr3fselect 0.11.0","text":"CRAN release: 2023-03-02 BREAKING CHANGE: method parameter fselect(), fselect_nested() auto_fselector() renamed fselector. FSelector objects accepted now. Arguments fselector passed ... anymore. BREAKING CHANGE: fselect parameter FSelector moved first position achieve consistency functions. docs: Update resources sections. docs: Add list default measures.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/news/index.html","id":"mlr3fselect-0100","dir":"Changelog","previous_headings":"","what":"mlr3fselect 0.10.0","title":"mlr3fselect 0.10.0","text":"CRAN release: 2023-02-21 feat: Add callback mlr3fselect.svm_rfe run recursive feature elimination linear support vector machines. refactor: importance scores FSelectorRFE now aggregated rank instead averaging . feat: Add FSelectorRFECV optimizer run recursive feature elimination cross-validation. refactor: FSelectorRFE works without store_models = TRUE now. feat: .data.table.ArchiveBatchFSelect() function additionally returns character vector selected features row. refactor: Add callbacks argument fsi() function.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/news/index.html","id":"mlr3fselect-091","dir":"Changelog","previous_headings":"","what":"mlr3fselect 0.9.1","title":"mlr3fselect 0.9.1","text":"CRAN release: 2023-01-26 refactor: Remove internal use mlr3pipelines. fix: Feature selection measures require importance oob error works now.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/news/index.html","id":"mlr3fselect-090","dir":"Changelog","previous_headings":"","what":"mlr3fselect 0.9.0","title":"mlr3fselect 0.9.0","text":"CRAN release: 2022-12-21 fix: Add genalg required packages FSelectorBatchGeneticSearch. feat: Add new callback backups benchmark result disk batch. feat: Create custom callbacks callback_batch_fselect() function.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/news/index.html","id":"mlr3fselect-080","dir":"Changelog","previous_headings":"","what":"mlr3fselect 0.8.0","title":"mlr3fselect 0.8.0","text":"CRAN release: 2022-11-16 refactor: FSelectorRFE throws error learner support $importance() method. refactor: AutoFSelector stores instance benchmark result store_models = TRUE. refactor: AutoFSelector stores instance store_benchmark_result = TRUE. feat: Add missing parameters AutoFSelector auto_fselect(). feat: Add fsi() function create FSelectInstanceBatchSingleCrit FSelectInstanceBatchMultiCrit. refactor: Remove unnest option .data.table.ArchiveBatchFSelect() function.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/news/index.html","id":"mlr3fselect-072","dir":"Changelog","previous_headings":"","what":"mlr3fselect 0.7.2","title":"mlr3fselect 0.7.2","text":"CRAN release: 2022-08-25 docs: Re-generate rd files valid html.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/news/index.html","id":"mlr3fselect-071","dir":"Changelog","previous_headings":"","what":"mlr3fselect 0.7.1","title":"mlr3fselect 0.7.1","text":"CRAN release: 2022-05-03 feat: FSelector objects field $id now.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/news/index.html","id":"mlr3fselect-070","dir":"Changelog","previous_headings":"","what":"mlr3fselect 0.7.0","title":"mlr3fselect 0.7.0","text":"CRAN release: 2022-04-08 feat: Allow pass FSelector objects method fselect() auto_fselector(). feat: Added $label FSelectors. docs: New examples fselect() function. feat: $help() method opens manual page FSelector. feat: Added .data.table.DictionaryFSelector function. feat: Added min_features parameter FSelectorBatchSequential.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/news/index.html","id":"mlr3fselect-061","dir":"Changelog","previous_headings":"","what":"mlr3fselect 0.6.1","title":"mlr3fselect 0.6.1","text":"CRAN release: 2022-01-20 Add store_models flag fselect(). Remove store_x_domain flag.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/news/index.html","id":"mlr3fselect-060","dir":"Changelog","previous_headings":"","what":"mlr3fselect 0.6.0","title":"mlr3fselect 0.6.0","text":"CRAN release: 2021-09-13 Adds AutoFSelector$base_learner() method extract base learner nested learner objects. Adds fselect(), auto_fselector() fselect_nested() sugar functions. Adds extract_inner_fselect_results() extract_inner_fselect_archives() helper function extract inner feature selection results archives.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/news/index.html","id":"mlr3fselect-051","dir":"Changelog","previous_headings":"","what":"mlr3fselect 0.5.1","title":"mlr3fselect 0.5.1","text":"CRAN release: 2021-03-09 Remove x_domain column archive.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/news/index.html","id":"mlr3fselect-050","dir":"Changelog","previous_headings":"","what":"mlr3fselect 0.5.0","title":"mlr3fselect 0.5.0","text":"CRAN release: 2021-01-24 FSelectorRFE stores importance values evaluated feature set archive. ArchiveBatchFSelect$data public field now.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/news/index.html","id":"mlr3fselect-041","dir":"Changelog","previous_headings":"","what":"mlr3fselect 0.4.1","title":"mlr3fselect 0.4.1","text":"CRAN release: 2020-10-30 Fix bug AutoFSelector$predict()","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/news/index.html","id":"mlr3fselect-040","dir":"Changelog","previous_headings":"","what":"mlr3fselect 0.4.0","title":"mlr3fselect 0.4.0","text":"CRAN release: 2020-10-22 Compact -memory representation R6 objects save space saving mlr3 objects via saveRDS(), serialize() etc. FSelectorRFE supports fraction features retain iteration (feature_fraction), number features remove iteration (feature_number) vector number features retain iteration (subset_sizes). AutoFSelect renamed AutoFSelector. retrieve inner feature selection results nested resampling, .data.table(rr)$learner[[1]]$fselect_result must used now. Option control store_benchmark_result, store_models check_values AutoFSelector. store_fselect_instance must set parameter initialization. Adds FSelectorBatchGeneticSearch. Fixes check_values flag FSelectInstanceBatchSingleCrit FSelectInstanceBatchMultiCrit. Removed dependency orphaned package bibtex. PipeOpSelect internally used task subsetting.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/news/index.html","id":"mlr3fselect-030","dir":"Changelog","previous_headings":"","what":"mlr3fselect 0.3.0","title":"mlr3fselect 0.3.0","text":"CRAN release: 2020-09-22 Archive ArchiveBatchFSelect now stores benchmark result $benchmark_result. change removed resample results archive can still accessed via benchmark result.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/news/index.html","id":"mlr3fselect-021","dir":"Changelog","previous_headings":"","what":"mlr3fselect 0.2.1","title":"mlr3fselect 0.2.1","text":"CRAN release: 2020-09-10 Warning message external package feature selection installed.","code":""},{"path":"https://mlr3fselect.mlr-org.com/dev/news/index.html","id":"mlr3fselect-020","dir":"Changelog","previous_headings":"","what":"mlr3fselect 0.2.0","title":"mlr3fselect 0.2.0","text":"CRAN release: 2020-08-23 Initial CRAN release.","code":""}]