Tidymodels, Error in finalize_workflow, object not found

Machine Learning and Modeling
1

I'm receiving the error "in finalize_workflow(., lr_best): object 'lr_wf' not found
I'm running:
R version 4.0.5,
R Studio Version 1.4.1106,
and tidymodels with the developer versions of broom, recipes, and workflows.

Assistance is greatly appreciated, reprex is as follows:

library(tidyverse)
library(tidymodels)

dfin=read_csv(file = "C:/Users/vanwag3/OneDrive - UW/Documents/UW_PhD/PhD_Research/Code_from_Drought_Drive/Drought_Analysis_RCode/LR_NAIP_wCE.csv")
#> 
#> -- Column specification --------------------------------------------------------
#> cols(
#>   .default = col_double(),
#>   neon2017 = col_character()
#> )
#> i Use `spec()` for the full column specifications.

df=dfin %>% mutate(neon2017=factor(neon2017))

set.seed(111)
s_split2=initial_split(df, strata = neon2017) 
train_data2=training(s_split2)
test_data2=testing(s_split2)

library(themis)
#> Registered S3 methods overwritten by 'themis':
#>   method                  from   
#>   bake.step_downsample    recipes
#>   bake.step_upsample      recipes
#>   prep.step_downsample    recipes
#>   prep.step_upsample      recipes
#>   tidy.step_downsample    recipes
#>   tidy.step_upsample      recipes
#>   tunable.step_downsample recipes
#>   tunable.step_upsample   recipes
#> 
#> Attaching package: 'themis'
#> The following objects are masked from 'package:recipes':
#> 
#>     step_downsample, step_upsample
train_rec2=recipe(neon2017 ~ ., data = train_data2) %>% 
  update_role(basin_id, new_role="Id")%>% 
  step_naomit(all_predictors(), skip = TRUE) %>% 
  step_normalize(all_predictors(), -all_outcomes()) %>% 
  step_smote(neon2017) 

tree_prep2=prep(train_rec2)
juiced2=juice(tree_prep2)

#Cross Validation Folds/Resamples
lr_cv=vfold_cv(train_data2, repeats = 10)

#Specify the Logistic model with the GLMNET engine
lr_mod=
  logistic_reg(penalty = tune(), mixture = 1) %>% 
  set_engine("glmnet") %>% 
  set_mode("classification")

lr_wflow=
  workflow() %>% 
  add_model(lr_mod) %>% 
  add_recipe(train_rec2)

#Set the grid for the l.r.
lr_reg_grid <- tibble(penalty = 10^seq(-5,-1, length.out = 30))

#and train the model and tune the parameters
lr_best =
  lr_wflow %>% 
  tune_grid(resamples=lr_cv,
            grid = lr_reg_grid,
            control = control_grid(save_pred = TRUE),
            metrics = metric_set(accuracy, kap,
                                 mcc, roc_auc
                                 )) %>% 
  select_best("roc_auc")

lr_best
#> # A tibble: 1 x 2
#>   penalty .config              
#>     <dbl> <chr>                
#> 1 0.00117 Preprocessor1_Model16

library(workflows)
library(broom)
lr_wf %>% 
  finalize_workflow(lr_best) %>%
  fit(train_data2) %>%
  pull_workflow_fit() %>%
  tidy()
#> Error in finalize_workflow(., lr_best): object 'lr_wf' not found
3

It looks like your tuneable workflow is called lr_wflow, not lr_wf. What if you try this instead?

lr_wflow %>% 
  finalize_workflow(lr_best)
1 Like
4

Thank you it worked! My kiddo being on spring break is definitely conducive to more silly errors than usual.

5

Definitely been there! :grin:

6

This topic was automatically closed 7 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.