run_benchmark_ranger.R

# License: BSD (3-clause)
# Author: Denis A. Engemann <[email protected]>

library(tidymodels)
library(readr)
library(microbenchmark)

hotels <- 
  read_csv('https://tidymodels.org/start/case-study/hotels.csv') %>%
  mutate_if(is.character, as.factor) 

set.seed(123)
splits <- initial_split(hotels, strata = children)

hotel_other <- training(splits)
hotel_test  <- testing(splits)

set.seed(234)
val_set <- validation_split(hotel_other, 
                            strata = children, 
                            prop = 0.80)

set.seed(345)

rf_recipe <- 
  recipe(children ~ ., data = hotel_other) %>% 
  step_date(arrival_date) %>% 
  step_holiday(arrival_date) %>% 
  step_rm(arrival_date) 

run_rf <- function(cores){
  rf_mod <- 
    rand_forest(mtry = 8, min_n = 7, trees = 100) %>% 
    set_engine("ranger", num.threads = cores, importance = "impurity") %>% 
    set_mode("classification")
  
  rf_workflow <- 
    workflow() %>% 
    add_model(rf_mod) %>% 
    add_recipe(rf_recipe)

  rf_workflow %>%
    last_fit(splits)
}

rf_bench <- microbenchmark(
  cores_1 = run_rf(cores = 1),
  cores_2 = run_rf(cores = 2),
  cores_3 = run_rf(cores = 3),
  cores_4 = run_rf(cores = 4),
  cores_5 = run_rf(cores = 5),
  cores_6 = run_rf(cores = 6),
  cores_7 = run_rf(cores = 7),
  cores_8 = run_rf(cores = 8),
  times = 10
)

rf_bench$arch <- sessionInfo()[[1]][[2]]
rf_bench$os <- sessionInfo()[[4]]
name <- gsub(" ", "-", date(), fixed = T)
name <- paste0("bench_rf_", name, ".csv")

write_csv(rf_bench, name)