suppressPackageStartupMessages( require(tidyverse) )
suppressPackageStartupMessages( require(pryr) )
suppressPackageStartupMessages( require('recipes') )

caret offers unified modelling syntax for a variety of modelling packages and it offers to be compatible with recipes, rsample.

1 Tree

data = rsample::attrition
formula = Attrition ~ JobSatisfaction + Gender + MonthlyIncome

grid = expand.grid( cp = 2^c(-10:10) )

car = caret::train( formula
                    , data = data
                    , method = 'rpart'
                    , tuneGrid = grid
                     )

car

## CART 
## 
## 1470 samples
##    3 predictor
##    2 classes: 'No', 'Yes' 
## 
## No pre-processing
## Resampling: Bootstrapped (25 reps) 
## Summary of sample sizes: 1470, 1470, 1470, 1470, 1470, 1470, ... 
## Resampling results across tuning parameters:
## 
##   cp            Accuracy   Kappa     
##   9.765625e-04  0.7889875  0.06750448
##   1.953125e-03  0.7958955  0.07026647
##   3.906250e-03  0.8099298  0.07156791
##   7.812500e-03  0.8269818  0.07289258
##   1.562500e-02  0.8360317  0.01851284
##   3.125000e-02  0.8384850  0.00000000
##   6.250000e-02  0.8384850  0.00000000
##   1.250000e-01  0.8384850  0.00000000
##   2.500000e-01  0.8384850  0.00000000
##   5.000000e-01  0.8384850  0.00000000
##   1.000000e+00  0.8384850  0.00000000
##   2.000000e+00  0.8384850  0.00000000
##   4.000000e+00  0.8384850  0.00000000
##   8.000000e+00  0.8384850  0.00000000
##   1.600000e+01  0.8384850  0.00000000
##   3.200000e+01  0.8384850  0.00000000
##   6.400000e+01  0.8384850  0.00000000
##   1.280000e+02  0.8384850  0.00000000
##   2.560000e+02  0.8384850  0.00000000
##   5.120000e+02  0.8384850  0.00000000
##   1.024000e+03  0.8384850  0.00000000
## 
## Accuracy was used to select the optimal model using the largest value.
## The final value used for the model was cp = 1024.

2 Tree + `recipes`

The recipe can be used to replace the formula in the caret syntax

rec = recipes::recipe(formula, data) %>%
  recipes::step_center( all_numeric() )

car = caret::train( rec
                    , data
                    , method = 'rpart'
                    , tuneGrid = grid
                    )
car

## CART 
## 
## 1470 samples
##   30 predictor
##    2 classes: 'No', 'Yes' 
## 
## Recipe steps: center 
## Resampling: Bootstrapped (25 reps) 
## Summary of sample sizes: 1470, 1470, 1470, 1470, 1470, 1470, ... 
## Resampling results across tuning parameters:
## 
##   cp            Accuracy   Kappa      
##   9.765625e-04  0.7855593  0.062057222
##   1.953125e-03  0.7935167  0.065357391
##   3.906250e-03  0.8095335  0.069442569
##   7.812500e-03  0.8276583  0.074730075
##   1.562500e-02  0.8356320  0.040784077
##   3.125000e-02  0.8387912  0.009979790
##   6.250000e-02  0.8405723  0.001571582
##   1.250000e-01  0.8410226  0.000000000
##   2.500000e-01  0.8410226  0.000000000
##   5.000000e-01  0.8410226  0.000000000
##   1.000000e+00  0.8410226  0.000000000
##   2.000000e+00  0.8410226  0.000000000
##   4.000000e+00  0.8410226  0.000000000
##   8.000000e+00  0.8410226  0.000000000
##   1.600000e+01  0.8410226  0.000000000
##   3.200000e+01  0.8410226  0.000000000
##   6.400000e+01  0.8410226  0.000000000
##   1.280000e+02  0.8410226  0.000000000
##   2.560000e+02  0.8410226  0.000000000
##   5.120000e+02  0.8410226  0.000000000
##   1.024000e+03  0.8410226  0.000000000
## 
## Accuracy was used to select the optimal model using the largest value.
## The final value used for the model was cp = 1024.

3 Tree + `rsample`

rs = rsample::vfold_cv( data, v = 10, repeats = 10) %>%
  rsample::rsample2caret()

car = caret::train( formula
                    , data = data
                    , method = 'rpart'
                    , tuneGrid = grid
                    , trControl = caret::trainControl(index = rs$index
                                                      , indexOut = rs$indexOut
                                                      , method = 'cv'
                                                      , verboseIter = T
                                                      , savePredictions = T)
                     )

## + Repeat01.Fold01: cp=0.0009766 
## - Repeat01.Fold01: cp=0.0009766 
## + Repeat01.Fold02: cp=0.0009766 
## - Repeat01.Fold02: cp=0.0009766 
## + Repeat01.Fold03: cp=0.0009766 
## - Repeat01.Fold03: cp=0.0009766 
## + Repeat01.Fold04: cp=0.0009766 
## - Repeat01.Fold04: cp=0.0009766 
## + Repeat01.Fold05: cp=0.0009766 
## - Repeat01.Fold05: cp=0.0009766 
## + Repeat01.Fold06: cp=0.0009766 
## - Repeat01.Fold06: cp=0.0009766 
## + Repeat01.Fold07: cp=0.0009766 
## - Repeat01.Fold07: cp=0.0009766 
## + Repeat01.Fold08: cp=0.0009766 
## - Repeat01.Fold08: cp=0.0009766 
## + Repeat01.Fold09: cp=0.0009766 
## - Repeat01.Fold09: cp=0.0009766 
## + Repeat01.Fold10: cp=0.0009766 
## - Repeat01.Fold10: cp=0.0009766 
## + Repeat02.Fold01: cp=0.0009766 
## - Repeat02.Fold01: cp=0.0009766 
## + Repeat02.Fold02: cp=0.0009766 
## - Repeat02.Fold02: cp=0.0009766 
## + Repeat02.Fold03: cp=0.0009766 
## - Repeat02.Fold03: cp=0.0009766 
## + Repeat02.Fold04: cp=0.0009766 
## - Repeat02.Fold04: cp=0.0009766 
## + Repeat02.Fold05: cp=0.0009766 
## - Repeat02.Fold05: cp=0.0009766 
## + Repeat02.Fold06: cp=0.0009766 
## - Repeat02.Fold06: cp=0.0009766 
## + Repeat02.Fold07: cp=0.0009766 
## - Repeat02.Fold07: cp=0.0009766 
## + Repeat02.Fold08: cp=0.0009766 
## - Repeat02.Fold08: cp=0.0009766 
## + Repeat02.Fold09: cp=0.0009766 
## - Repeat02.Fold09: cp=0.0009766 
## + Repeat02.Fold10: cp=0.0009766 
## - Repeat02.Fold10: cp=0.0009766 
## + Repeat03.Fold01: cp=0.0009766 
## - Repeat03.Fold01: cp=0.0009766 
## + Repeat03.Fold02: cp=0.0009766 
## - Repeat03.Fold02: cp=0.0009766 
## + Repeat03.Fold03: cp=0.0009766 
## - Repeat03.Fold03: cp=0.0009766 
## + Repeat03.Fold04: cp=0.0009766 
## - Repeat03.Fold04: cp=0.0009766 
## + Repeat03.Fold05: cp=0.0009766 
## - Repeat03.Fold05: cp=0.0009766 
## + Repeat03.Fold06: cp=0.0009766 
## - Repeat03.Fold06: cp=0.0009766 
## + Repeat03.Fold07: cp=0.0009766 
## - Repeat03.Fold07: cp=0.0009766 
## + Repeat03.Fold08: cp=0.0009766 
## - Repeat03.Fold08: cp=0.0009766 
## + Repeat03.Fold09: cp=0.0009766 
## - Repeat03.Fold09: cp=0.0009766 
## + Repeat03.Fold10: cp=0.0009766 
## - Repeat03.Fold10: cp=0.0009766 
## + Repeat04.Fold01: cp=0.0009766 
## - Repeat04.Fold01: cp=0.0009766 
## + Repeat04.Fold02: cp=0.0009766 
## - Repeat04.Fold02: cp=0.0009766 
## + Repeat04.Fold03: cp=0.0009766 
## - Repeat04.Fold03: cp=0.0009766 
## + Repeat04.Fold04: cp=0.0009766 
## - Repeat04.Fold04: cp=0.0009766 
## + Repeat04.Fold05: cp=0.0009766 
## - Repeat04.Fold05: cp=0.0009766 
## + Repeat04.Fold06: cp=0.0009766 
## - Repeat04.Fold06: cp=0.0009766 
## + Repeat04.Fold07: cp=0.0009766 
## - Repeat04.Fold07: cp=0.0009766 
## + Repeat04.Fold08: cp=0.0009766 
## - Repeat04.Fold08: cp=0.0009766 
## + Repeat04.Fold09: cp=0.0009766 
## - Repeat04.Fold09: cp=0.0009766 
## + Repeat04.Fold10: cp=0.0009766 
## - Repeat04.Fold10: cp=0.0009766 
## + Repeat05.Fold01: cp=0.0009766 
## - Repeat05.Fold01: cp=0.0009766 
## + Repeat05.Fold02: cp=0.0009766 
## - Repeat05.Fold02: cp=0.0009766 
## + Repeat05.Fold03: cp=0.0009766 
## - Repeat05.Fold03: cp=0.0009766 
## + Repeat05.Fold04: cp=0.0009766 
## - Repeat05.Fold04: cp=0.0009766 
## + Repeat05.Fold05: cp=0.0009766 
## - Repeat05.Fold05: cp=0.0009766 
## + Repeat05.Fold06: cp=0.0009766 
## - Repeat05.Fold06: cp=0.0009766 
## + Repeat05.Fold07: cp=0.0009766 
## - Repeat05.Fold07: cp=0.0009766 
## + Repeat05.Fold08: cp=0.0009766 
## - Repeat05.Fold08: cp=0.0009766 
## + Repeat05.Fold09: cp=0.0009766 
## - Repeat05.Fold09: cp=0.0009766 
## + Repeat05.Fold10: cp=0.0009766 
## - Repeat05.Fold10: cp=0.0009766 
## + Repeat06.Fold01: cp=0.0009766 
## - Repeat06.Fold01: cp=0.0009766 
## + Repeat06.Fold02: cp=0.0009766 
## - Repeat06.Fold02: cp=0.0009766 
## + Repeat06.Fold03: cp=0.0009766 
## - Repeat06.Fold03: cp=0.0009766 
## + Repeat06.Fold04: cp=0.0009766 
## - Repeat06.Fold04: cp=0.0009766 
## + Repeat06.Fold05: cp=0.0009766 
## - Repeat06.Fold05: cp=0.0009766 
## + Repeat06.Fold06: cp=0.0009766 
## - Repeat06.Fold06: cp=0.0009766 
## + Repeat06.Fold07: cp=0.0009766 
## - Repeat06.Fold07: cp=0.0009766 
## + Repeat06.Fold08: cp=0.0009766 
## - Repeat06.Fold08: cp=0.0009766 
## + Repeat06.Fold09: cp=0.0009766 
## - Repeat06.Fold09: cp=0.0009766 
## + Repeat06.Fold10: cp=0.0009766 
## - Repeat06.Fold10: cp=0.0009766 
## + Repeat07.Fold01: cp=0.0009766 
## - Repeat07.Fold01: cp=0.0009766 
## + Repeat07.Fold02: cp=0.0009766 
## - Repeat07.Fold02: cp=0.0009766 
## + Repeat07.Fold03: cp=0.0009766 
## - Repeat07.Fold03: cp=0.0009766 
## + Repeat07.Fold04: cp=0.0009766 
## - Repeat07.Fold04: cp=0.0009766 
## + Repeat07.Fold05: cp=0.0009766 
## - Repeat07.Fold05: cp=0.0009766 
## + Repeat07.Fold06: cp=0.0009766 
## - Repeat07.Fold06: cp=0.0009766 
## + Repeat07.Fold07: cp=0.0009766 
## - Repeat07.Fold07: cp=0.0009766 
## + Repeat07.Fold08: cp=0.0009766 
## - Repeat07.Fold08: cp=0.0009766 
## + Repeat07.Fold09: cp=0.0009766 
## - Repeat07.Fold09: cp=0.0009766 
## + Repeat07.Fold10: cp=0.0009766 
## - Repeat07.Fold10: cp=0.0009766 
## + Repeat08.Fold01: cp=0.0009766 
## - Repeat08.Fold01: cp=0.0009766 
## + Repeat08.Fold02: cp=0.0009766 
## - Repeat08.Fold02: cp=0.0009766 
## + Repeat08.Fold03: cp=0.0009766 
## - Repeat08.Fold03: cp=0.0009766 
## + Repeat08.Fold04: cp=0.0009766 
## - Repeat08.Fold04: cp=0.0009766 
## + Repeat08.Fold05: cp=0.0009766 
## - Repeat08.Fold05: cp=0.0009766 
## + Repeat08.Fold06: cp=0.0009766 
## - Repeat08.Fold06: cp=0.0009766 
## + Repeat08.Fold07: cp=0.0009766 
## - Repeat08.Fold07: cp=0.0009766 
## + Repeat08.Fold08: cp=0.0009766 
## - Repeat08.Fold08: cp=0.0009766 
## + Repeat08.Fold09: cp=0.0009766 
## - Repeat08.Fold09: cp=0.0009766 
## + Repeat08.Fold10: cp=0.0009766 
## - Repeat08.Fold10: cp=0.0009766 
## + Repeat09.Fold01: cp=0.0009766 
## - Repeat09.Fold01: cp=0.0009766 
## + Repeat09.Fold02: cp=0.0009766 
## - Repeat09.Fold02: cp=0.0009766 
## + Repeat09.Fold03: cp=0.0009766 
## - Repeat09.Fold03: cp=0.0009766 
## + Repeat09.Fold04: cp=0.0009766 
## - Repeat09.Fold04: cp=0.0009766 
## + Repeat09.Fold05: cp=0.0009766 
## - Repeat09.Fold05: cp=0.0009766 
## + Repeat09.Fold06: cp=0.0009766 
## - Repeat09.Fold06: cp=0.0009766 
## + Repeat09.Fold07: cp=0.0009766 
## - Repeat09.Fold07: cp=0.0009766 
## + Repeat09.Fold08: cp=0.0009766 
## - Repeat09.Fold08: cp=0.0009766 
## + Repeat09.Fold09: cp=0.0009766 
## - Repeat09.Fold09: cp=0.0009766 
## + Repeat09.Fold10: cp=0.0009766 
## - Repeat09.Fold10: cp=0.0009766 
## + Repeat10.Fold01: cp=0.0009766 
## - Repeat10.Fold01: cp=0.0009766 
## + Repeat10.Fold02: cp=0.0009766 
## - Repeat10.Fold02: cp=0.0009766 
## + Repeat10.Fold03: cp=0.0009766 
## - Repeat10.Fold03: cp=0.0009766 
## + Repeat10.Fold04: cp=0.0009766 
## - Repeat10.Fold04: cp=0.0009766 
## + Repeat10.Fold05: cp=0.0009766 
## - Repeat10.Fold05: cp=0.0009766 
## + Repeat10.Fold06: cp=0.0009766 
## - Repeat10.Fold06: cp=0.0009766 
## + Repeat10.Fold07: cp=0.0009766 
## - Repeat10.Fold07: cp=0.0009766 
## + Repeat10.Fold08: cp=0.0009766 
## - Repeat10.Fold08: cp=0.0009766 
## + Repeat10.Fold09: cp=0.0009766 
## - Repeat10.Fold09: cp=0.0009766 
## + Repeat10.Fold10: cp=0.0009766 
## - Repeat10.Fold10: cp=0.0009766 
## Aggregating results
## Selecting tuning parameters
## Fitting cp = 1024 on full training set

unique(car$pred$cp)

##  [1] 9.765625e-04 1.953125e-03 3.906250e-03 7.812500e-03 1.562500e-02
##  [6] 3.125000e-02 6.250000e-02 1.250000e-01 2.500000e-01 5.000000e-01
## [11] 1.000000e+00 2.000000e+00 4.000000e+00 8.000000e+00 1.600000e+01
## [16] 3.200000e+01 6.400000e+01 1.280000e+02 2.560000e+02 5.120000e+02
## [21] 1.024000e+03

unique(car$pred$Resample)

##   [1] "Repeat01.Fold01" "Repeat01.Fold02" "Repeat01.Fold03"
##   [4] "Repeat01.Fold04" "Repeat01.Fold05" "Repeat01.Fold06"
##   [7] "Repeat01.Fold07" "Repeat01.Fold08" "Repeat01.Fold09"
##  [10] "Repeat01.Fold10" "Repeat02.Fold01" "Repeat02.Fold02"
##  [13] "Repeat02.Fold03" "Repeat02.Fold04" "Repeat02.Fold05"
##  [16] "Repeat02.Fold06" "Repeat02.Fold07" "Repeat02.Fold08"
##  [19] "Repeat02.Fold09" "Repeat02.Fold10" "Repeat03.Fold01"
##  [22] "Repeat03.Fold02" "Repeat03.Fold03" "Repeat03.Fold04"
##  [25] "Repeat03.Fold05" "Repeat03.Fold06" "Repeat03.Fold07"
##  [28] "Repeat03.Fold08" "Repeat03.Fold09" "Repeat03.Fold10"
##  [31] "Repeat04.Fold01" "Repeat04.Fold02" "Repeat04.Fold03"
##  [34] "Repeat04.Fold04" "Repeat04.Fold05" "Repeat04.Fold06"
##  [37] "Repeat04.Fold07" "Repeat04.Fold08" "Repeat04.Fold09"
##  [40] "Repeat04.Fold10" "Repeat05.Fold01" "Repeat05.Fold02"
##  [43] "Repeat05.Fold03" "Repeat05.Fold04" "Repeat05.Fold05"
##  [46] "Repeat05.Fold06" "Repeat05.Fold07" "Repeat05.Fold08"
##  [49] "Repeat05.Fold09" "Repeat05.Fold10" "Repeat06.Fold01"
##  [52] "Repeat06.Fold02" "Repeat06.Fold03" "Repeat06.Fold04"
##  [55] "Repeat06.Fold05" "Repeat06.Fold06" "Repeat06.Fold07"
##  [58] "Repeat06.Fold08" "Repeat06.Fold09" "Repeat06.Fold10"
##  [61] "Repeat07.Fold01" "Repeat07.Fold02" "Repeat07.Fold03"
##  [64] "Repeat07.Fold04" "Repeat07.Fold05" "Repeat07.Fold06"
##  [67] "Repeat07.Fold07" "Repeat07.Fold08" "Repeat07.Fold09"
##  [70] "Repeat07.Fold10" "Repeat08.Fold01" "Repeat08.Fold02"
##  [73] "Repeat08.Fold03" "Repeat08.Fold04" "Repeat08.Fold05"
##  [76] "Repeat08.Fold06" "Repeat08.Fold07" "Repeat08.Fold08"
##  [79] "Repeat08.Fold09" "Repeat08.Fold10" "Repeat09.Fold01"
##  [82] "Repeat09.Fold02" "Repeat09.Fold03" "Repeat09.Fold04"
##  [85] "Repeat09.Fold05" "Repeat09.Fold06" "Repeat09.Fold07"
##  [88] "Repeat09.Fold08" "Repeat09.Fold09" "Repeat09.Fold10"
##  [91] "Repeat10.Fold01" "Repeat10.Fold02" "Repeat10.Fold03"
##  [94] "Repeat10.Fold04" "Repeat10.Fold05" "Repeat10.Fold06"
##  [97] "Repeat10.Fold07" "Repeat10.Fold08" "Repeat10.Fold09"
## [100] "Repeat10.Fold10"

car$pred %>%
  as_tibble() %>%
  dplyr::group_by(rowIndex) %>%
  dplyr::count() %>%
  head()

## # A tibble: 6 x 2
## # Groups:   rowIndex [6]
##   rowIndex     n
##      <int> <int>
## 1        1   210
## 2        2   210
## 3        3   210
## 4        4   210
## 5        5   210
## 6        6   210

nrow(data)

## [1] 1470

nrow(grid)

## [1] 21

car$pred %>%
  as_tibble() %>%
  dplyr::group_by(rowIndex) %>%
  dplyr::count() %>%
  nrow()

## [1] 1470

pryr::object_size(car)

## 10.7 MB

pryr::object_size(data)

## 275 kB

4 Bringing it all together

using all three packages and fitting two different kinds of models using two different formulas in a modelling dataframe

require(recipes)

data = rsample::attrition

formula1 = Attrition ~ JobSatisfaction + Gender + MonthlyIncome
formula2 = Attrition ~ JobSatisfaction + Gender + MonthlyIncome + DistanceFromHome

grid_tree = expand.grid( cp = 2^c(-10:10), method = 'rpart', stringsAsFactors = FALSE) %>%
  as_tibble() %>%
  nest( cp, .key = 'grid' ) %>%
  mutate()

grid_svm  = expand.grid( mtry = c(1:3), method = 'rf', stringsAsFactors = FALSE )  %>%
  as_tibble() %>%
  nest( mtry, .key = 'grid' )

df = expand.grid( formula = list(formula1, formula2)
             , grid = list( grid_tree, grid_svm  ) 
             ) %>%
  unnest(grid) %>%
  as.tibble()

df = df %>%
  mutate(  rec = map( formula, recipe, data)
          , rec = map( rec, step_scale, all_numeric() )
          , rec = map( rec, step_center, all_numeric() )
          , cv = list(rsample::vfold_cv(data, v = 5, repeats = 1) )
          , cv_rs = map( cv, rsample::rsample2caret )
          ) 

object_size(df)

## 4.26 MB

object_size(data)

## 275 kB

5 Wrapper for `caret::train()`

car = function( recipe, rsample, method, grid, data){
  
  grid = as.data.frame(grid)
  
  car = caret::train( recipe
                      , data
                      , method = method
                      , tuneGrid = grid
                      , trControl = caret::trainControl(index = rsample$index
                                                        , indexOut = rsample$indexOut
                                                        , method = 'cv'
                                                        , verboseIter = T
                                                        , savePredictions = T
                                                        , classProbs = T )
                       )
  return( as.tibble(car$pred) )
}

6 Apply wrapper

df = df %>%
  mutate( preds = pmap(list(rec, cv_rs, method, grid), car, data = data) )

## Preparing recipe
## + Fold1: cp=0.0009766 
## - Fold1: cp=0.0009766 
## + Fold2: cp=0.0009766 
## - Fold2: cp=0.0009766 
## + Fold3: cp=0.0009766 
## - Fold3: cp=0.0009766 
## + Fold4: cp=0.0009766 
## - Fold4: cp=0.0009766 
## + Fold5: cp=0.0009766 
## - Fold5: cp=0.0009766 
## Aggregating results
## Selecting tuning parameters
## Fitting cp = 1024 on full training set
## Preparing recipe
## + Fold1: cp=0.0009766 
## - Fold1: cp=0.0009766 
## + Fold2: cp=0.0009766 
## - Fold2: cp=0.0009766 
## + Fold3: cp=0.0009766 
## - Fold3: cp=0.0009766 
## + Fold4: cp=0.0009766 
## - Fold4: cp=0.0009766 
## + Fold5: cp=0.0009766 
## - Fold5: cp=0.0009766 
## Aggregating results
## Selecting tuning parameters
## Fitting cp = 1024 on full training set
## Preparing recipe
## + Fold1: mtry=1 
## - Fold1: mtry=1 
## + Fold1: mtry=2 
## - Fold1: mtry=2 
## + Fold1: mtry=3 
## - Fold1: mtry=3 
## + Fold2: mtry=1 
## - Fold2: mtry=1 
## + Fold2: mtry=2 
## - Fold2: mtry=2 
## + Fold2: mtry=3 
## - Fold2: mtry=3 
## + Fold3: mtry=1 
## - Fold3: mtry=1 
## + Fold3: mtry=2 
## - Fold3: mtry=2 
## + Fold3: mtry=3 
## - Fold3: mtry=3 
## + Fold4: mtry=1 
## - Fold4: mtry=1 
## + Fold4: mtry=2 
## - Fold4: mtry=2 
## + Fold4: mtry=3 
## - Fold4: mtry=3 
## + Fold5: mtry=1 
## - Fold5: mtry=1 
## + Fold5: mtry=2 
## - Fold5: mtry=2 
## + Fold5: mtry=3 
## - Fold5: mtry=3 
## Aggregating results
## Selecting tuning parameters
## Fitting mtry = 1 on full training set
## Preparing recipe
## + Fold1: mtry=1 
## - Fold1: mtry=1 
## + Fold1: mtry=2 
## - Fold1: mtry=2 
## + Fold1: mtry=3 
## - Fold1: mtry=3 
## + Fold2: mtry=1 
## - Fold2: mtry=1 
## + Fold2: mtry=2 
## - Fold2: mtry=2 
## + Fold2: mtry=3 
## - Fold2: mtry=3 
## + Fold3: mtry=1 
## - Fold3: mtry=1 
## + Fold3: mtry=2 
## - Fold3: mtry=2 
## + Fold3: mtry=3 
## - Fold3: mtry=3 
## + Fold4: mtry=1 
## - Fold4: mtry=1 
## + Fold4: mtry=2 
## - Fold4: mtry=2 
## + Fold4: mtry=3 
## - Fold4: mtry=3 
## + Fold5: mtry=1 
## - Fold5: mtry=1 
## + Fold5: mtry=2 
## - Fold5: mtry=2 
## + Fold5: mtry=3 
## - Fold5: mtry=3 
## Aggregating results
## Selecting tuning parameters
## Fitting mtry = 1 on full training set

df

## # A tibble: 4 x 7
##         formula method              grid          rec               cv
##          <list>  <chr>            <list>       <list>           <list>
## 1 <S3: formula>  rpart <tibble [21 x 1]> <S3: recipe> <tibble [5 x 2]>
## 2 <S3: formula>  rpart <tibble [21 x 1]> <S3: recipe> <tibble [5 x 2]>
## 3 <S3: formula>     rf  <tibble [3 x 1]> <S3: recipe> <tibble [5 x 2]>
## 4 <S3: formula>     rf  <tibble [3 x 1]> <S3: recipe> <tibble [5 x 2]>
## # ... with 2 more variables: cv_rs <list>, preds <list>

caret

OEB

December 21, 2017

1 Tree

2 Tree + `recipes`

3 Tree + `rsample`

4 Bringing it all together

5 Wrapper for `caret::train()`

6 Apply wrapper

caret

OEB

December 21, 2017

1 Tree

2 Tree + recipes

3 Tree + rsample

4 Bringing it all together

5 Wrapper for caret::train()

6 Apply wrapper

2 Tree + `recipes`

3 Tree + `rsample`

5 Wrapper for `caret::train()`