Skip to content

Random Forest, using Ranger

Source: vignettes/ranger.Rmd
ranger.Rmd
Function Works
tidypredict_fit(), tidypredict_sql(), parse_model() ✔
tidypredict_to_column() ✗
tidypredict_test() ✔
tidypredict_interval(), tidypredict_sql_interval() ✗
parsnip ✔

How it works

Here is a simple ranger() model using the mtcars dataset:

library(dplyr)
library(tidypredict)
library(ranger)

model <- ranger(mpg ~ ., data = mtcars, num.trees = 5, max.depth = 2)

Under the hood

The parser is based on the output from the ranger::treeInfo() function. It will return as many decision paths as there are non-NA rows in the prediction field.

treeInfo(model) %>%
  head()
#>   nodeID leftChild rightChild splitvarID splitvarName splitval
#> 1      0         1          2          8         gear     3.50
#> 2      1         3          4          2           hp   192.50
#> 3      2         5          6          4           wt     2.26
#> 4      3        NA         NA         NA         <NA>       NA
#> 5      4        NA         NA         NA         <NA>       NA
#> 6      5        NA         NA         NA         <NA>       NA
#>   terminal prediction
#> 1    FALSE         NA
#> 2    FALSE         NA
#> 3    FALSE         NA
#> 4     TRUE    17.4000
#> 5     TRUE    12.9000
#> 6     TRUE    29.4375

The output from parse_model() is transformed into a dplyr, a.k.a Tidy Eval, formula. Each decision tree becomes one dplyr::case_when() statement, which are then combined.

tidypredict_fit(model)
#> (case_when(gear <= 3.5 ~ case_when(hp <= 192.5 ~ 17.4, .default = 12.9), 
#>     .default = case_when(wt <= 2.26 ~ 29.4375, .default = 21.3625)) + 
#>     case_when(wt <= 2.455 ~ case_when(vs <= 0.5 ~ 26, .default = 31.8), 
#>         .default = case_when(gear <= 3.5 ~ 14.25, .default = 19.0666666666667)) + 
#>     case_when(disp <= 120.65 ~ case_when(hp <= 65.5 ~ 31.275, 
#>         .default = 26.6333333333333), .default = case_when(wt <= 
#>         3.505 ~ 19.9533333333333, .default = 14.6857142857143)) + 
#>     case_when(cyl <= 5 ~ case_when(disp <= 93.5 ~ 30.625, .default = 22.32), 
#>         .default = case_when(cyl <= 7 ~ 18.78, .default = 15.7769230769231)) + 
#>     case_when(cyl <= 5 ~ case_when(disp <= 107.6 ~ 31.1333333333333, 
#>         .default = 23.6625), .default = case_when(hp <= 192.5 ~ 
#>         18.65, .default = 12.8625)))/5

From there, the Tidy Eval formula can be used anywhere where it can be operated. tidypredict provides three paths:

  • Use directly inside dplyr, mutate(iris, !! tidypredict_fit(model))
  • Use tidypredict_to_column(model) to a piped command set
  • Use tidypredict_to_sql(model) to retrieve the SQL statement

parsnip

tidypredict also supports ranger model objects fitted via the parsnip package.

library(parsnip)

parsnip_model <- rand_forest(mode = "regression", trees = 5) %>%
  set_engine("ranger", max.depth = 2) %>%
  fit(mpg ~ ., data = mtcars)

tidypredict_fit(parsnip_model)
#> (case_when(vs <= 0.5 ~ case_when(disp <= 450 ~ 16.9684210526316, 
#>     .default = 10.4), .default = case_when(drat <= 4 ~ 24.2, 
#>     .default = 30.6857142857143)) + case_when(hp <= 131.5 ~ case_when(wt <= 
#>     2.3325 ~ 31.6666666666667, .default = 21.7571428571429), 
#>     .default = case_when(drat <= 3.035 ~ 12.1, .default = 16.3153846153846)) + 
#>     case_when(cyl <= 5 ~ case_when(hp <= 78.5 ~ 31.28, .default = 24.4), 
#>         .default = case_when(wt <= 4.747 ~ 17.52, .default = 10.4)) + 
#>     case_when(drat <= 3.615 ~ case_when(carb <= 2.5 ~ 17.5875, 
#>         .default = 13.4), .default = case_when(wt <= 2.23 ~ 28.8857142857143, 
#>         .default = 20.4285714285714)) + case_when(disp <= 266.9 ~ 
#>     case_when(wt <= 2.2775 ~ 31.375, .default = 22.4133333333333), 
#>     .default = case_when(wt <= 4.747 ~ 16.5818181818182, .default = 10.4)))/5