mlr3's automatic parameter renaming breaks R's naming conventions

[MattJEM]

I have learners that have gone through a few transformations and have ended up with some pretty long parameter names, such as 'encode.classif.xgboost.classif.xgboost.min_child_weight'.

I think these names are interfering with my ability to tune my models. When I try to run benchmark on my learners I get the following error:

Error in .__ParamSet__initialize(self = self, private = private, super = super,  : 
  Assertion on 'ids' failed: Must have names according to R's variable naming conventions, but element 1 does not comply.

Would it be possible to either allow the user to change parameter names in mlr3 or to disable this check?

[larskotthoff]

With mlr3pipelines you can set the names explicitly, see the mlr3 book. That said, the name you posted seems to be a valid R name as far as I can tell. Could you post a minimal example that allows to reproduce the problem please?

[MattJEM]

I can't provide my original data, but this is pretty reliably reproducing my main three errors from the liver data dataset. Sorry for the length--I've added in some extra columns and done an approximate reproduction of my workflow in case something in there is causing the issue.

data('ilpd', package = 'mlr3data')

group_name <- sample(c('red', 'blue', 'yellow'), size = nrow(ilpd), replace = TRUE)
id_name <- 1:nrow(ilpd)
another_factor <- sample(c('factor1', 'factor2', 'factor3'), size = nrow(ilpd), replace = TRUE)

extra <- tibble(group_name, id_name, another_factor)

ilpd2 <- ilpd %>%
  bind_cols(extra) %>%
    mutate(gender = as.numeric(ifelse(gender == "Female", 1, 0)),
           diseased = factor(ifelse(diseased == "yes", 1, 0)),
           alkaline_phosphatase_log = log(alkaline_phosphatase),
           aspartate_transaminase_log = log(aspartate_transaminase),
           direct_bilirubin_log = log(direct_bilirubin),
           group_name = factor(group_name),
           another_factor = factor(another_factor))

ilpd_tsk = TaskClassif$new(id = 'ilpd_tsk',
                           backend = ilpd2,
                           target = 'diseased',
                           positive = '1')

ilpd_tsk$set_col_roles("group_name", roles = "group")
ilpd_tsk$set_col_roles(c('id_name'), roles = 'name')

ilpd_tsk$col_roles$feature = setdiff(ilpd_tsk$col_roles$feature,
                                     c('total_bilirubin',
                                       'direct_bilirubin',
                                       'alkaline_phosphatase',
                                       'aspartate_transaminase',
                                       'alanine_transaminase',
                                       'total_protein',
                                       'albumin_globulin_ratio'
                                       ))

lrn_xgboost_pretreat <- lrn("classif.xgboost", 
                            nrounds = 500,
                            predict_type = 'prob',
                            predict_sets = c('train', 'test'))


fct_encoder = po("encode", method = "one-hot", affect_columns = selector_type("factor"))

the_pipe <- fct_encoder %>>% lrn_xgboost_pretreat

xgboost_piped <- as_learner(the_pipe)

lrnrs <- list(
  lrn_xgboost = xgboost_piped,
  lrn_gbm = lrn('classif.gbm',
                n.trees = 500,
                predict_type = 'prob',
                predict_sets = c('train', 'test'))
)

lrnrs_thresh <- lapply(lrnrs, function(x) {GraphLearner$new(po('learner_cv', x) %>>% po('tunethreshold',
                                                                                        param_vals = list(
                                                                                          measure = 'classif.prauc'))
                                                            )})

tune_params_xgboost <- ps(
  encode.classif.xgboost.classif.xgboost.eta = p_dbl(lower = 0.005, upper = 0.2),
  encode.classif.xgboost.classif.xgboost.min_child_weight = p_dbl(lower = 1, upper = 10),
  encode.classif.xgboost.classif.xgboost.max_depth = p_int(lower = 3, upper = 8)
)

tune_params_gbm <- ps(
  classif.gbm.interaction.depth = p_int(lower = 1, upper = 6),
  classif.gbm.shrinkage = p_dbl(lower = 0.0005, upper = 0.01)
)

lrnrs_thresh$lrn_xgboost <- auto_tuner(
  method = 'random_search',
  learner = lrnrs_thresh$lrn_xgboost,
  resampling = rsmp('repeated_cv', repeats = 5, folds = 5),
  measure = msr('classif.ce'),
  search_space = tune_params_xgboost,
  term_evals = 200
)
lrnrs_thresh$lrn_xgboost$predict_sets = c('train', 'test')

lrnrs_thresh$lrn_gbm <- auto_tuner(
  method = 'random_search',
  learner = lrnrs_thresh$lrn_gbm,
  resampling = rsmp('repeated_cv', repeats = 5, folds = 5),
  measure = msr('classif.ce'),
  search_space = tune_params_gbm,
  term_evals = 200
)
lrnrs_thresh$lrn_gbm$predict_sets = c('train', 'test')

fitting_rsmp <- rsmp('repeated_cv', repeats = 10, folds = 5)

big_grid <- benchmark_grid(
  tasks = ilpd_tsk,
  learners = lrnrs_thresh,
  resamplings = fitting_rsmp
)

bmrs <- benchmark(big_grid, store_models = TRUE)

[spietras]

I had the same issue when trying to train or benchmark GraphLearner with learner_cv -> tunethreshold for classification. In my case, the issue was that my target labels were 0 and 1 which are not valid R names. It worked after renaming values to something like a and b (or anything else more aligned with the task).

[pat-s]

@MattJEM I am getting a different error in your reprex which seems unrelated to the one from your OP. Could you have a look?

library(mlr3verse)
#> Loading required package: mlr3
library(tidyverse)

data("ilpd", package = "mlr3data")

group_name <- sample(c("red", "blue", "yellow"), size = nrow(ilpd), replace = TRUE)
id_name <- 1:nrow(ilpd)
another_factor <- sample(c("factor1", "factor2", "factor3"), size = nrow(ilpd), replace = TRUE)

extra <- tibble(group_name, id_name, another_factor)

ilpd2 <- ilpd %>%
  bind_cols(extra) %>%
  mutate(
    gender = as.numeric(ifelse(gender == "Female", 1, 0)),
    diseased = factor(ifelse(diseased == "yes", 1, 0)),
    alkaline_phosphatase_log = log(alkaline_phosphatase),
    aspartate_transaminase_log = log(aspartate_transaminase),
    direct_bilirubin_log = log(direct_bilirubin),
    group_name = factor(group_name),
    another_factor = factor(another_factor)
  )

ilpd_tsk <- TaskClassif$new(
  id = "ilpd_tsk",
  backend = ilpd2,
  target = "diseased",
  positive = "1"
)

ilpd_tsk$set_col_roles("group_name", roles = "group")
ilpd_tsk$set_col_roles(c("id_name"), roles = "name")

ilpd_tsk$col_roles$feature <- setdiff(
  ilpd_tsk$col_roles$feature,
  c(
    "total_bilirubin",
    "direct_bilirubin",
    "alkaline_phosphatase",
    "aspartate_transaminase",
    "alanine_transaminase",
    "total_protein",
    "albumin_globulin_ratio"
  )
)

lrn_xgboost_pretreat <- lrn("classif.xgboost",
  nrounds = 500,
  predict_type = "prob",
  predict_sets = c("train", "test")
)


fct_encoder <- po("encode", method = "one-hot", affect_columns = selector_type("factor"))

the_pipe <- fct_encoder %>>% lrn_xgboost_pretreat

xgboost_piped <- as_learner(the_pipe)

lrnrs <- list(
  lrn_xgboost = xgboost_piped,
  lrn_gbm = lrn("classif.gbm",
    n.trees = 500,
    predict_type = "prob",
    predict_sets = c("train", "test")
  )
)

lrnrs_thresh <- lapply(lrnrs, function(x) {
  GraphLearner$new(po("learner_cv", x) %>>% po("tunethreshold",
    param_vals = list(
      measure = "classif.prauc"
    )
  ))
})

tune_params_xgboost <- ps(
  encode.classif.xgboost.classif.xgboost.eta = p_dbl(lower = 0.005, upper = 0.2),
  encode.classif.xgboost.classif.xgboost.min_child_weight = p_dbl(lower = 1, upper = 10),
  encode.classif.xgboost.classif.xgboost.max_depth = p_int(lower = 3, upper = 8)
)

tune_params_gbm <- ps(
  classif.gbm.interaction.depth = p_int(lower = 1, upper = 6),
  classif.gbm.shrinkage = p_dbl(lower = 0.0005, upper = 0.01)
)

lrnrs_thresh$lrn_xgboost <- auto_tuner(
  method = "random_search",
  learner = lrnrs_thresh$lrn_xgboost,
  resampling = rsmp("repeated_cv", repeats = 5, folds = 5),
  measure = msr("classif.ce"),
  search_space = tune_params_xgboost,
  term_evals = 200
)
lrnrs_thresh$lrn_xgboost$predict_sets <- c("train", "test")

lrnrs_thresh$lrn_gbm <- auto_tuner(
  method = "random_search",
  learner = lrnrs_thresh$lrn_gbm,
  resampling = rsmp("repeated_cv", repeats = 5, folds = 5),
  measure = msr("classif.ce"),
  search_space = tune_params_gbm,
  term_evals = 200
)
lrnrs_thresh$lrn_gbm$predict_sets <- c("train", "test")

fitting_rsmp <- rsmp("repeated_cv", repeats = 10, folds = 5)

big_grid <- benchmark_grid(
  tasks = ilpd_tsk,
  learners = lrnrs_thresh,
  resamplings = fitting_rsmp
)

bmrs <- benchmark(big_grid, store_models = TRUE)
#> INFO  [16:22:43.879] [mlr3] Running benchmark with 100 resampling iterations 
#> INFO  [16:22:43.915] [mlr3] Applying learner 'encode.classif.xgboost.tunethreshold.tuned' on task 'ilpd_tsk' (iter 23/50) 
#> INFO  [16:22:44.002] [bbotk] Starting to optimize 3 parameter(s) with '<OptimizerRandomSearch>' and '<TerminatorEvals> [n_evals=200, k=0]' 
#> INFO  [16:22:44.010] [bbotk] Evaluating 1 configuration(s) 
#> INFO  [16:22:44.042] [mlr3] Running benchmark with 25 resampling iterations 
#> INFO  [16:22:44.045] [mlr3] Applying learner 'encode.classif.xgboost.tunethreshold' on task 'ilpd_tsk' (iter 24/25) 
#> INFO  [16:22:44.698] [mlr3] Applying learner 'encode.classif.xgboost' on task 'ilpd_tsk' (iter 1/3) 
#> INFO  [16:22:44.829] [mlr3] Applying learner 'encode.classif.xgboost' on task 'ilpd_tsk' (iter 3/3)
#> Error: DataBackend did not return the queried rows correctly: 1 requested, 0 received
#> This happened PipeOp encode's $predict()
#> This happened PipeOp encode.classif.xgboost's $train()

^{Created on 2022-06-03 by the reprex package (v2.0.1)}

Session info

sessioninfo::session_info()
#> ─ Session info ───────────────────────────────────────────────────────────────
#>  setting  value
#>  version  R version 4.2.0 (2022-04-22)
#>  os       macOS Monterey 12.4
#>  system   aarch64, darwin20
#>  ui       X11
#>  language (EN)
#>  collate  en_US.UTF-8
#>  ctype    en_US.UTF-8
#>  tz       Europe/Zurich
#>  date     2022-06-03
#>  pandoc   2.18 @ /Applications/RStudio.app/Contents/Resources/app/quarto/bin/tools/ (via rmarkdown)
#> 
#> ─ Packages ───────────────────────────────────────────────────────────────────
#>  package           * version     date (UTC) lib source
#>  assertthat          0.2.1       2019-03-21 [1] CRAN (R 4.2.0)
#>  backports           1.4.1       2021-12-13 [1] CRAN (R 4.2.0)
#>  bbotk               0.5.3       2022-05-04 [1] CRAN (R 4.2.0)
#>  broom               0.8.0       2022-04-13 [1] CRAN (R 4.2.0)
#>  cellranger          1.1.0       2016-07-27 [1] CRAN (R 4.2.0)
#>  checkmate           2.1.0       2022-04-21 [1] CRAN (R 4.2.0)
#>  cli                 3.3.0       2022-04-25 [1] CRAN (R 4.2.0)
#>  clue                0.3-61      2022-05-30 [1] CRAN (R 4.2.0)
#>  cluster             2.1.3       2022-03-28 [1] CRAN (R 4.2.0)
#>  clusterCrit         1.2.8       2018-07-26 [1] CRAN (R 4.2.0)
#>  codetools           0.2-18      2020-11-04 [3] CRAN (R 4.2.0)
#>  colorspace          2.0-3       2022-02-21 [1] CRAN (R 4.2.0)
#>  crayon              1.5.1       2022-03-26 [1] CRAN (R 4.2.0)
#>  data.table          1.14.2      2021-09-27 [1] CRAN (R 4.2.0)
#>  DBI                 1.1.2       2021-12-20 [1] CRAN (R 4.2.0)
#>  dbplyr              2.1.1       2021-04-06 [1] CRAN (R 4.2.0)
#>  digest              0.6.29      2021-12-01 [1] CRAN (R 4.2.0)
#>  dplyr             * 1.0.9       2022-04-28 [1] CRAN (R 4.2.0)
#>  ellipsis            0.3.2       2021-04-29 [1] CRAN (R 4.2.0)
#>  evaluate            0.15        2022-02-18 [1] CRAN (R 4.2.0)
#>  fansi               1.0.3       2022-03-24 [1] CRAN (R 4.2.0)
#>  fastmap             1.1.0       2021-01-25 [1] CRAN (R 4.2.0)
#>  forcats           * 0.5.1       2021-01-27 [1] CRAN (R 4.2.0)
#>  fs                  1.5.2       2021-12-08 [1] CRAN (R 4.2.0)
#>  future              1.26.1      2022-05-27 [1] CRAN (R 4.2.0)
#>  future.apply        1.9.0       2022-04-25 [1] CRAN (R 4.2.0)
#>  generics            0.1.2       2022-01-31 [1] CRAN (R 4.2.0)
#>  ggplot2           * 3.3.6       2022-05-03 [1] CRAN (R 4.2.0)
#>  globals             0.15.0      2022-05-09 [1] CRAN (R 4.2.0)
#>  glue                1.6.2       2022-02-24 [1] CRAN (R 4.2.0)
#>  gtable              0.3.0       2019-03-25 [1] CRAN (R 4.2.0)
#>  haven               2.5.0       2022-04-15 [1] CRAN (R 4.2.0)
#>  highr               0.9         2021-04-16 [1] CRAN (R 4.2.0)
#>  hms                 1.1.1       2021-09-26 [1] CRAN (R 4.2.0)
#>  htmltools           0.5.2       2021-08-25 [1] CRAN (R 4.2.0)
#>  httr                1.4.3       2022-05-04 [1] CRAN (R 4.2.0)
#>  jsonlite            1.8.0       2022-02-22 [1] CRAN (R 4.2.0)
#>  knitr               1.39        2022-04-26 [1] CRAN (R 4.2.0)
#>  lattice             0.20-45     2021-09-22 [3] CRAN (R 4.2.0)
#>  lgr                 0.4.3       2021-09-16 [1] CRAN (R 4.2.0)
#>  lifecycle           1.0.1       2021-09-24 [1] CRAN (R 4.2.0)
#>  listenv             0.8.0       2019-12-05 [1] CRAN (R 4.2.0)
#>  lubridate           1.8.0       2021-10-07 [1] CRAN (R 4.2.0)
#>  magrittr            2.0.3       2022-03-30 [1] CRAN (R 4.2.0)
#>  Matrix              1.4-1       2022-03-23 [1] CRAN (R 4.2.0)
#>  mlr3              * 0.13.3-9000 2022-05-28 [1] Github (mlr-org/mlr3@7313ae8)
#>  mlr3cluster         0.1.3       2022-04-06 [1] CRAN (R 4.2.0)
#>  mlr3data            0.6.0       2022-03-18 [1] CRAN (R 4.2.0)
#>  mlr3extralearners   0.5.37      2022-05-19 [1] Github (mlr-org/mlr3extralearners@f06f209)
#>  mlr3filters         0.5.0.9000  2022-05-09 [1] Github (mlr-org/mlr3filters@3dde376)
#>  mlr3fselect         0.7.1       2022-05-09 [1] Github (mlr-org/mlr3fselect@c2284af)
#>  mlr3learners        0.5.3       2022-05-25 [1] CRAN (R 4.2.0)
#>  mlr3measures        0.4.1       2022-01-13 [1] CRAN (R 4.2.0)
#>  mlr3misc            0.10.0      2022-01-11 [1] CRAN (R 4.2.0)
#>  mlr3pipelines       0.4.0-9000  2022-05-31 [1] Github (mlr-org/mlr3pipelines@1aae6ba)
#>  mlr3tuning          0.13.1      2022-05-03 [1] CRAN (R 4.2.0)
#>  mlr3tuningspaces    0.2.0       2022-05-30 [1] Github (mlr-org/mlr3tuningspaces@a5a0ca3)
#>  mlr3verse         * 0.2.5       2022-05-18 [1] CRAN (R 4.2.0)
#>  mlr3viz             0.5.9       2022-05-30 [1] local
#>  modelr              0.1.8       2020-05-19 [1] CRAN (R 4.2.0)
#>  munsell             0.5.0       2018-06-12 [1] CRAN (R 4.2.0)
#>  palmerpenguins      0.1.0       2020-07-23 [1] CRAN (R 4.2.0)
#>  paradox             0.9.0.9000  2022-05-09 [1] Github (mlr-org/paradox@d41cc29)
#>  parallelly          1.31.1      2022-04-22 [1] CRAN (R 4.2.0)
#>  pillar              1.7.0       2022-02-01 [1] CRAN (R 4.2.0)
#>  pkgconfig           2.0.3       2019-09-22 [1] CRAN (R 4.2.0)
#>  purrr             * 0.3.4       2020-04-17 [1] CRAN (R 4.2.0)
#>  R.cache             0.15.0      2021-04-30 [1] CRAN (R 4.2.0)
#>  R.methodsS3         1.8.1       2020-08-26 [1] CRAN (R 4.2.0)
#>  R.oo                1.24.0      2020-08-26 [1] CRAN (R 4.2.0)
#>  R.utils             2.11.0      2021-09-26 [1] CRAN (R 4.2.0)
#>  R6                  2.5.1       2021-08-19 [1] CRAN (R 4.2.0)
#>  readr             * 2.1.2       2022-01-30 [1] CRAN (R 4.2.0)
#>  readxl              1.4.0       2022-03-28 [1] CRAN (R 4.2.0)
#>  reprex              2.0.1       2021-08-05 [1] CRAN (R 4.2.0)
#>  rlang               1.0.2       2022-03-04 [1] CRAN (R 4.2.0)
#>  rmarkdown           2.14        2022-04-25 [1] CRAN (R 4.2.0)
#>  rstudioapi          0.13        2020-11-12 [1] CRAN (R 4.2.0)
#>  rvest               1.0.2       2021-10-16 [1] CRAN (R 4.2.0)
#>  scales              1.2.0       2022-04-13 [1] CRAN (R 4.2.0)
#>  sessioninfo         1.2.2       2021-12-06 [1] CRAN (R 4.2.0)
#>  stringi             1.7.6       2021-11-29 [1] CRAN (R 4.2.0)
#>  stringr           * 1.4.0       2019-02-10 [1] CRAN (R 4.2.0)
#>  styler              1.7.0       2022-03-13 [1] CRAN (R 4.2.0)
#>  tibble            * 3.1.7       2022-05-03 [1] CRAN (R 4.2.0)
#>  tidyr             * 1.2.0       2022-02-01 [1] CRAN (R 4.2.0)
#>  tidyselect          1.1.2       2022-02-21 [1] CRAN (R 4.2.0)
#>  tidyverse         * 1.3.1       2021-04-15 [1] CRAN (R 4.2.0)
#>  tzdb                0.3.0       2022-03-28 [1] CRAN (R 4.2.0)
#>  utf8                1.2.2       2021-07-24 [1] CRAN (R 4.2.0)
#>  uuid                1.1-0       2022-04-19 [1] CRAN (R 4.2.0)
#>  vctrs               0.4.1       2022-04-13 [1] CRAN (R 4.2.0)
#>  withr               2.5.0       2022-03-03 [1] CRAN (R 4.2.0)
#>  xfun                0.31        2022-05-10 [1] CRAN (R 4.2.0)
#>  xgboost             1.6.0.1     2022-04-16 [1] CRAN (R 4.2.0)
#>  xml2                1.3.3       2021-11-30 [1] CRAN (R 4.2.0)
#>  yaml                2.3.5       2022-02-21 [1] CRAN (R 4.2.0)
#> 
#>  [1] /Users/pjs/Library/R/arm64/4.2/library
#>  [2] /opt/R/4.2.0-arm64/Resources/site-library
#>  [3] /opt/R/4.2.0-arm64/Resources/library
#> 
#> ──────────────────────────────────────────────────────────────────────────────