Error in tune_race_anova

Machine Learning and Modeling
,
1

Hi guys, I have follow Julia blog to learn how to use tidymodel, in predict board games topic I have used tune_race_anova to fit model but it run out error.
I try to find the reason by type in show_notes(.Last.tune.result) and it said: Error in map(., str_replace_all, " ", "_"): could not find function "map" which is quite weird because I have install purrr and stringr package from the beginning.
This is the first time I try to create a reprex and ask for help in R community so if I make the reprex a little bit ugly or wrong some how please kindly advice me further. Thanks :slight_smile:

#import data
library(tidymodels)
library(tidyverse)

ratings <- read_csv("https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2022/2022-01-25/ratings.csv")
#> Rows: 21831 Columns: 10
#> ── Column specification ────────────────────────────────────────────────────────
#> Delimiter: ","
#> chr (3): name, url, thumbnail
#> dbl (7): num, id, year, rank, average, bayes_average, users_rated
#> 
#> ℹ Use `spec()` to retrieve the full column specification for this data.
#> ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
details <- read_csv("https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2022/2022-01-25/details.csv")
#> Rows: 21631 Columns: 23
#> ── Column specification ────────────────────────────────────────────────────────
#> Delimiter: ","
#> chr (10): primary, description, boardgamecategory, boardgamemechanic, boardg...
#> dbl (13): num, id, yearpublished, minplayers, maxplayers, playingtime, minpl...
#> 
#> ℹ Use `spec()` to retrieve the full column specification for this data.
#> ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.

ratings_joined <-
  ratings %>%
  left_join(details, by = "id")

#Fitting model

set.seed(123)
game_split <-
  ratings_joined %>%
  select(name, average, matches("min|max"), boardgamecategory) %>%
  na.omit() %>%
  initial_split(strata = average)
game_train <- training(game_split)
game_test <- testing(game_split)

set.seed(234)
game_folds <- vfold_cv(game_train, strata = average)
game_folds
#> #  10-fold cross-validation using stratification 
#> # A tibble: 10 × 2
#>    splits               id    
#>    <list>               <chr> 
#>  1 <split [14407/1602]> Fold01
#>  2 <split [14407/1602]> Fold02
#>  3 <split [14407/1602]> Fold03
#>  4 <split [14408/1601]> Fold04
#>  5 <split [14408/1601]> Fold05
#>  6 <split [14408/1601]> Fold06
#>  7 <split [14408/1601]> Fold07
#>  8 <split [14408/1601]> Fold08
#>  9 <split [14410/1599]> Fold09
#> 10 <split [14410/1599]> Fold10


library(textrecipes)

split_category <- function(x) {
  x %>%
    str_split(", ") %>%
    map(str_remove_all, "[:punct:]") %>%
    map(str_squish) %>%
    map(str_to_lower) %>%
    map(str_replace_all, " ", "_")
}

game_rec <-
  recipe(average ~ ., data = game_train) %>%
  update_role(name, new_role = "id") %>%
  step_tokenize(boardgamecategory, custom_token = split_category) %>%
  step_tokenfilter(boardgamecategory, max_tokens = 30) %>%
  step_tf(boardgamecategory)


game_prep <- prep(game_rec)
bake(game_prep, new_data = NULL) %>% skimr::skim()
Name Piped data
Number of rows 16009
Number of columns 37
_______________________
Column type frequency:
factor 1
numeric 36
________________________
Group variables None

Data summary

Variable type: factor

skim_variable n_missing complete_rate ordered n_unique top_counts
name 0 1 FALSE 15781 Rob: 6, Cha: 4, Ali: 3, Aro: 3

Variable type: numeric

skim_variable n_missing complete_rate mean sd p0 p25 p50 p75 p100 hist
minplayers 0 1 2.01 0.69 0.00 2.00 2.00 2.00 10.00 ▇▁▁▁▁
maxplayers 0 1 5.73 16.43 0.00 4.00 4.00 6.00 999.00 ▇▁▁▁▁
minplaytime 0 1 66.25 517.90 0.00 20.00 30.00 60.00 60000.00 ▇▁▁▁▁
maxplaytime 0 1 93.98 609.41 0.00 25.00 45.00 90.00 60000.00 ▇▁▁▁▁
minage 0 1 9.64 3.65 0.00 8.00 10.00 12.00 25.00 ▂▇▆▁▁
average 0 1 6.42 0.93 1.31 5.83 6.45 7.04 9.57 ▁▁▆▇▁
tf_boardgamecategory_abstract_strategy 0 1 0.07 0.26 0.00 0.00 0.00 0.00 1.00 ▇▁▁▁▁
tf_boardgamecategory_action_dexterity 0 1 0.05 0.22 0.00 0.00 0.00 0.00 1.00 ▇▁▁▁▁
tf_boardgamecategory_adventure 0 1 0.06 0.23 0.00 0.00 0.00 0.00 1.00 ▇▁▁▁▁
tf_boardgamecategory_ancient 0 1 0.03 0.18 0.00 0.00 0.00 0.00 1.00 ▇▁▁▁▁
tf_boardgamecategory_animals 0 1 0.06 0.24 0.00 0.00 0.00 0.00 1.00 ▇▁▁▁▁
tf_boardgamecategory_bluffing 0 1 0.06 0.23 0.00 0.00 0.00 0.00 1.00 ▇▁▁▁▁
tf_boardgamecategory_card_game 0 1 0.30 0.46 0.00 0.00 0.00 1.00 1.00 ▇▁▁▁▃
tf_boardgamecategory_childrens_game 0 1 0.08 0.26 0.00 0.00 0.00 0.00 1.00 ▇▁▁▁▁
tf_boardgamecategory_deduction 0 1 0.05 0.22 0.00 0.00 0.00 0.00 1.00 ▇▁▁▁▁
tf_boardgamecategory_dice 0 1 0.09 0.28 0.00 0.00 0.00 0.00 1.00 ▇▁▁▁▁
tf_boardgamecategory_economic 0 1 0.07 0.26 0.00 0.00 0.00 0.00 1.00 ▇▁▁▁▁
tf_boardgamecategory_exploration 0 1 0.04 0.20 0.00 0.00 0.00 0.00 1.00 ▇▁▁▁▁
tf_boardgamecategory_fantasy 0 1 0.13 0.33 0.00 0.00 0.00 0.00 1.00 ▇▁▁▁▁
tf_boardgamecategory_fighting 0 1 0.08 0.27 0.00 0.00 0.00 0.00 1.00 ▇▁▁▁▁
tf_boardgamecategory_horror 0 1 0.03 0.18 0.00 0.00 0.00 0.00 1.00 ▇▁▁▁▁
tf_boardgamecategory_humor 0 1 0.06 0.23 0.00 0.00 0.00 0.00 1.00 ▇▁▁▁▁
tf_boardgamecategory_medieval 0 1 0.05 0.21 0.00 0.00 0.00 0.00 1.00 ▇▁▁▁▁
tf_boardgamecategory_miniatures 0 1 0.05 0.22 0.00 0.00 0.00 0.00 1.00 ▇▁▁▁▁
tf_boardgamecategory_movies_tv_radio_theme 0 1 0.05 0.22 0.00 0.00 0.00 0.00 1.00 ▇▁▁▁▁
tf_boardgamecategory_nautical 0 1 0.03 0.17 0.00 0.00 0.00 0.00 1.00 ▇▁▁▁▁
tf_boardgamecategory_negotiation 0 1 0.03 0.17 0.00 0.00 0.00 0.00 1.00 ▇▁▁▁▁
tf_boardgamecategory_party_game 0 1 0.09 0.29 0.00 0.00 0.00 0.00 1.00 ▇▁▁▁▁
tf_boardgamecategory_print_play 0 1 0.03 0.16 0.00 0.00 0.00 0.00 1.00 ▇▁▁▁▁
tf_boardgamecategory_puzzle 0 1 0.03 0.17 0.00 0.00 0.00 0.00 1.00 ▇▁▁▁▁
tf_boardgamecategory_racing 0 1 0.03 0.17 0.00 0.00 0.00 0.00 1.00 ▇▁▁▁▁
tf_boardgamecategory_realtime 0 1 0.04 0.18 0.00 0.00 0.00 0.00 1.00 ▇▁▁▁▁
tf_boardgamecategory_science_fiction 0 1 0.08 0.27 0.00 0.00 0.00 0.00 1.00 ▇▁▁▁▁
tf_boardgamecategory_trivia 0 1 0.03 0.16 0.00 0.00 0.00 0.00 1.00 ▇▁▁▁▁
tf_boardgamecategory_wargame 0 1 0.18 0.38 0.00 0.00 0.00 0.00 1.00 ▇▁▁▁▂
tf_boardgamecategory_world_war_ii 0 1 0.06 0.23 0.00 0.00 0.00 0.00 1.00 ▇▁▁▁▁ 

xgb_spec <-
  boost_tree(
    trees = tune(),
    mtry = tune(),
    min_n = tune(),
    learn_rate = 0.01
  ) %>%
  set_engine("xgboost") %>%
  set_mode("regression")

xgb_wf <- workflow(game_rec, xgb_spec)
xgb_wf
#> ══ Workflow ════════════════════════════════════════════════════════════════════
#> Preprocessor: Recipe
#> Model: boost_tree()
#> 
#> ── Preprocessor ────────────────────────────────────────────────────────────────
#> 3 Recipe Steps
#> 
#> • step_tokenize()
#> • step_tokenfilter()
#> • step_tf()
#> 
#> ── Model ───────────────────────────────────────────────────────────────────────
#> Boosted Tree Model Specification (regression)
#> 
#> Main Arguments:
#>   mtry = tune()
#>   trees = tune()
#>   min_n = tune()
#>   learn_rate = 0.01
#> 
#> Computational engine: xgboost


library(finetune)
doParallel::registerDoParallel()

set.seed(234)
xgb_game_rs <-
  tune_race_anova(
    xgb_wf,
    game_folds,
    grid = 20
  )
#> i Creating pre-processing data to finalize unknown parameter: mtry
#> Warning: All models failed. Run `show_notes(.Last.tune.result)` for more
#> information.
#> Error in `test_parameters_gls()`:
#> ! There were no valid metrics for the ANOVA model.

xgb_game_rs
#> Error in eval(expr, envir, enclos): object 'xgb_game_rs' not found

Created on 2022-07-27 by the reprex package (v2.0.1)

2

Thanks for the post, @conlelevn! Welcome to RStudio Community. :slight_smile: Your reprex looks great.

I was able to replicate your issue! (For posterity, others can do so by finishing out the doParallel setup by specifying a backend.)

The errors you're seeing arise from your parallel processing setup. The environments that doParallel sets up to run your computations in parallel are a bit different than a typical function environment, and thus trip up when trying to find where map comes from.

I was able to run your code successfully by specifying the namespaces of stringr and purrr explicitly in the split_category function definition:

split_category <- function(x) {
  x %>%
    stringr::str_split(", ") %>%
    purrr::map(stringr::str_remove_all, "[:punct:]") %>%
    purrr::map(stringr::str_squish) %>%
    purrr::map(stringr::str_to_lower) %>%
    purrr::map(stringr::str_replace_all, " ", "_")
}

Hopefully this does the trick for you!

3

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.