Issue with parsnip survival_reg() funcion

[davestr2]

I am trying to use the new survival_reg() function in parsnip 0.1.7. When I use this, I get an error when trying to do a fit on a workflow that uses this function. The older version surv_reg(), that is still available in 0.1.7, works as desired.

If this is a duplicate issue, then sorry about that !

# Trying to run code found in Chapter 7 Section 7.4.1 of the
#     Tidy Modeling with R book (https://www.tmwr.org/workflows.html#special-model-formulas)

# Load the libraries

library(tidymodels)
#> Registered S3 method overwritten by 'tune':
#>   method                   from   
#>   required_pkgs.model_spec parsnip
library(survival)

# Create  model for use with survival package:


# Use the "old" pre parsnip 0.1.6 function  -----   surv_reg()
#     As is done in the manual

parametric_model_Old_func <- 
  surv_reg() %>% 
  set_engine("survival")
#> Warning: `surv_reg()` was deprecated in parsnip 0.1.6.
#> Please use `survival_reg()` instead.

# Notice the warning about surv_reg() being depreciated and to use survival_reg()
#       See this  function being used below


# Run the output from above thru translate() function

surv_reg() %>% set_engine("survival") %>% translate()
#> Warning: `surv_reg()` was deprecated in parsnip 0.1.6.
#> Please use `survival_reg()` instead.
#> Parametric Survival Regression Model Specification (regression)
#> 
#> Computational engine: survival 
#> 
#> Model fit template:
#> survival::survreg(formula = missing_arg(), data = missing_arg(), 
#>     weights = missing_arg(), model = TRUE)

# Create a workflow

parametric_workflow_Old_func <- 
  workflow() %>% 
  # Pass the data along as-is: 
  add_variables(outcome = c(fustat, futime), predictors = c(age, rx)) %>% 
  add_model(parametric_model_Old_func, 
            # This formula is given to the model
            formula = Surv(futime, fustat) ~ age + strata(rx))

# fit the workflow

parametric_fit_Old_func <- fit(parametric_workflow_Old_func, data = ovarian)

# Take a look at the fitted model

parametric_fit_Old_func
#> ══ Workflow [trained] ══════════════════════════════════════════════════════════
#> Preprocessor: Variables
#> Model: surv_reg()
#> 
#> ── Preprocessor ────────────────────────────────────────────────────────────────
#> Outcomes: c(fustat, futime)
#> Predictors: c(age, rx)
#> 
#> ── Model ───────────────────────────────────────────────────────────────────────
#> Call:
#> survival::survreg(formula = Surv(futime, fustat) ~ age + strata(rx), 
#>     data = data, model = TRUE)
#> 
#> Coefficients:
#> (Intercept)         age 
#>  12.8734120  -0.1033569 
#> 
#> Scale:
#>      rx=1      rx=2 
#> 0.7695509 0.4703602 
#> 
#> Loglik(model)= -89.4   Loglik(intercept only)= -97.1
#>  Chisq= 15.36 on 1 degrees of freedom, p= 8.88e-05 
#> n= 26



# Use the "new" 0.1.6 function  -----   survival_reg()

parametric_model_New_func <- 
  survival_reg() %>% 
  set_engine("survival")

# Run the output from above thru translate() function
# Here the "issue" raising its head, maybe

survival_reg() %>% set_engine("survival") %>% translate()
#> Parametric Survival Regression Model Specification (censored regression)
#> 
#> Computational engine: survival 
#> 
#> Model fit template:
#> Error: Can't create call to non-callable object

# Notice the Error above

# Create a workflow 
# No error or warning when creating workflow !!!!

parametric_workflow_New_func <- 
  workflow() %>% 
  # Pass the data along as-is: 
  add_variables(outcome = c(fustat, futime), predictors = c(age, rx)) %>% 
  add_model(parametric_model_New_func, 
            # This formula is given to the model
            formula = Surv(futime, fustat) ~ age + strata(rx))

# Fit the workflow
# Bang here

parametric_fit_New_func <- fit(parametric_workflow_New_func, data = ovarian)
#> Error: formula_ is unknown.

# Notice error above

#parametric_fit_New_func

rlang::last_error()
#> <error/rlang_error>
#> formula_ is unknown.
#> Backtrace:
#>  1. generics::fit(parametric_workflow_New_func, data = ovarian)
#>  2. workflows:::fit.workflow(parametric_workflow_New_func, data = ovarian)
#>  3. workflows::.fit_model(workflow, control)
#>  5. workflows:::fit.action_model(...)
#>  6. workflows:::fit_from_formula(spec, mold, control_parsnip, formula)
#>  8. parsnip::fit.model_spec(...)
#> Run `rlang::last_trace()` to see the full context.

rlang::last_trace()
#> <error/rlang_error>
#> formula_ is unknown.
#> Backtrace:
#>     █
#>  1. ├─generics::fit(parametric_workflow_New_func, data = ovarian)
#>  2. └─workflows:::fit.workflow(parametric_workflow_New_func, data = ovarian)
#>  3.   └─workflows::.fit_model(workflow, control)
#>  4.     ├─generics::fit(action_model, workflow = workflow, control = control)
#>  5.     └─workflows:::fit.action_model(...)
#>  6.       └─workflows:::fit_from_formula(spec, mold, control_parsnip, formula)
#>  7.         ├─generics::fit(spec, formula = formula, data = data, control = control_parsnip)
#>  8.         └─parsnip::fit.model_spec(...)

sessioninfo::session_info()
#> ─ Session info ───────────────────────────────────────────────────────────────
#>  setting  value                       
#>  version  R version 4.1.1 (2021-08-10)
#>  os       Ubuntu 20.04.2 LTS          
#>  system   x86_64, linux-gnu           
#>  ui       X11                         
#>  language (EN)                        
#>  collate  en_US.UTF-8                 
#>  ctype    en_US.UTF-8                 
#>  tz       America/New_York            
#>  date     2021-08-19                  
#> 
#> ─ Packages ───────────────────────────────────────────────────────────────────
#>  package      * version    date       lib
#>  assertthat     0.2.1      2019-03-21 [2]
#>  backports      1.2.1      2020-12-09 [2]
#>  broom        * 0.7.9      2021-07-27 [2]
#>  class          7.3-19     2021-05-03 [4]
#>  cli            3.0.1      2021-07-17 [2]
#>  codetools      0.2-18     2020-11-04 [4]
#>  colorspace     2.0-2      2021-06-24 [2]
#>  crayon         1.4.1      2021-02-08 [2]
#>  DBI            1.1.1      2021-01-15 [2]
#>  dials        * 0.0.9      2020-09-16 [2]
#>  DiceDesign     1.9        2021-02-13 [2]
#>  digest         0.6.27     2020-10-24 [2]
#>  dplyr        * 1.0.7      2021-06-18 [2]
#>  ellipsis       0.3.2      2021-04-29 [2]
#>  evaluate       0.14       2019-05-28 [2]
#>  fansi          0.5.0      2021-05-25 [2]
#>  foreach        1.5.1      2020-10-15 [2]
#>  fs             1.5.0      2020-07-31 [2]
#>  furrr          0.2.3      2021-06-25 [2]
#>  future         1.21.0     2020-12-10 [2]
#>  generics       0.1.0      2020-10-31 [2]
#>  ggplot2      * 3.3.5      2021-06-25 [2]
#>  globals        0.14.0     2020-11-22 [2]
#>  glue           1.4.2      2020-08-27 [2]
#>  gower          0.2.2      2020-06-23 [2]
#>  GPfit          1.0-8      2019-02-08 [2]
#>  gtable         0.3.0      2019-03-25 [2]
#>  hardhat        0.1.6      2021-07-14 [2]
#>  highr          0.9        2021-04-16 [2]
#>  htmltools      0.5.1.1    2021-01-22 [2]
#>  infer        * 0.5.4      2021-01-13 [2]
#>  ipred          0.9-11     2021-03-12 [2]
#>  iterators      1.0.13     2020-10-15 [2]
#>  knitr          1.33       2021-04-24 [2]
#>  lattice        0.20-44    2021-05-02 [4]
#>  lava           1.6.9      2021-03-11 [2]
#>  lhs            1.1.1      2020-10-05 [2]
#>  lifecycle      1.0.0      2021-02-15 [2]
#>  listenv        0.8.0      2019-12-05 [2]
#>  lubridate      1.7.10     2021-02-26 [2]
#>  magrittr       2.0.1      2020-11-17 [2]
#>  MASS           7.3-54     2021-05-03 [4]
#>  Matrix         1.3-4      2021-06-01 [4]
#>  modeldata    * 0.1.1      2021-07-14 [2]
#>  munsell        0.5.0      2018-06-12 [2]
#>  nnet           7.3-16     2021-05-03 [4]
#>  parallelly     1.27.0     2021-07-19 [2]
#>  parsnip      * 0.1.7      2021-07-21 [2]
#>  pillar         1.6.2      2021-07-29 [2]
#>  pkgconfig      2.0.3      2019-09-22 [2]
#>  plyr           1.8.6      2020-03-03 [2]
#>  pROC           1.17.0.1   2021-01-13 [2]
#>  prodlim        2019.11.13 2019-11-17 [2]
#>  purrr        * 0.3.4      2020-04-17 [2]
#>  R6             2.5.0      2020-10-28 [2]
#>  Rcpp           1.0.7      2021-07-07 [2]
#>  recipes      * 0.1.16     2021-04-16 [2]
#>  reprex         2.0.0      2021-04-02 [2]
#>  rlang          0.4.11     2021-04-30 [2]
#>  rmarkdown      2.9        2021-06-15 [2]
#>  rpart          4.1-15     2019-04-12 [4]
#>  rsample      * 0.1.0      2021-05-08 [2]
#>  rstudioapi     0.13       2020-11-12 [2]
#>  scales       * 1.1.1      2020-05-11 [2]
#>  sessioninfo    1.1.1      2018-11-05 [2]
#>  stringi        1.6.2      2021-05-17 [2]
#>  stringr        1.4.0      2019-02-10 [2]
#>  survival     * 3.2-11     2021-04-26 [4]
#>  tibble       * 3.1.3      2021-07-23 [2]
#>  tidymodels   * 0.1.3      2021-04-19 [2]
#>  tidyr        * 1.1.3      2021-03-03 [2]
#>  tidyselect     1.1.1      2021-04-30 [2]
#>  timeDate       3043.102   2018-02-21 [2]
#>  tune         * 0.1.6      2021-07-21 [2]
#>  utf8           1.2.2      2021-07-24 [2]
#>  vctrs          0.3.8      2021-04-29 [2]
#>  withr          2.4.2      2021-04-18 [2]
#>  workflows    * 0.2.3      2021-07-16 [2]
#>  workflowsets * 0.1.0.9000 2021-08-15 [2]
#>  xfun           0.24       2021-06-15 [2]
#>  yaml           2.2.1      2020-02-01 [2]
#>  yardstick    * 0.0.8      2021-03-28 [2]
#>  source                                  
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.1)                          
#>  CRAN (R 4.0.5)                          
#>  CRAN (R 4.1.1)                          
#>  CRAN (R 4.0.3)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.1)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.1)                          
#>  CRAN (R 4.1.1)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.1)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.1)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.0.5)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.1)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.0.5)                          
#>  CRAN (R 4.1.1)                          
#>  CRAN (R 4.1.1)                          
#>  CRAN (R 4.1.1)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.0.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.1)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.0.5)                          
#>  CRAN (R 4.1.1)                          
#>  CRAN (R 4.1.1)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.1)                          
#>  CRAN (R 4.1.1)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.1)                          
#>  Github (tidymodels/workflowsets@57006c0)
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.0)                          
#>  CRAN (R 4.1.1)                          
#> 
#> [1] /home/dave/R/x86_64-pc-linux-gnu-library/4.1
#> [2] /usr/local/lib/R/site-library
#> [3] /usr/lib/R/site-library
#> [4] /usr/lib/R/library

^{Created on 2021-08-19 by the reprex package (v2.0.0)}

[EmilHvitfeldt]

The engine of survival_reg() is located in censored. Loading {censored} should make it work.

library(tidymodels)
library(survival)
library(censored)

parametric_model_New_func <- 
  survival_reg() %>% 
  set_engine("survival")

survival_reg() %>% set_engine("survival") %>% translate()
#> Parametric Survival Regression Model Specification (censored regression)
#> 
#> Computational engine: survival 
#> 
#> Model fit template:
#> survival::survreg(formula = missing_arg(), data = missing_arg(), 
#>     weights = missing_arg(), model = TRUE)

parametric_workflow_New_func <- 
  workflow() %>% 
  add_variables(outcome = c(fustat, futime), predictors = c(age, rx)) %>% 
  add_model(parametric_model_New_func, 
            formula = Surv(futime, fustat) ~ age + strata(rx))

parametric_fit_New_func <- fit(parametric_workflow_New_func, data = ovarian)

parametric_fit_New_func
#> ══ Workflow [trained] ══════════════════════════════════════════════════════════
#> Preprocessor: Variables
#> Model: survival_reg()
#> 
#> ── Preprocessor ────────────────────────────────────────────────────────────────
#> Outcomes: c(fustat, futime)
#> Predictors: c(age, rx)
#> 
#> ── Model ───────────────────────────────────────────────────────────────────────
#> Call:
#> survival::survreg(formula = Surv(futime, fustat) ~ age + strata(rx), 
#>     data = data, model = TRUE)
#> 
#> Coefficients:
#> (Intercept)         age 
#>  12.8734120  -0.1033569 
#> 
#> Scale:
#>      rx=1      rx=2 
#> 0.7695509 0.4703602 
#> 
#> Loglik(model)= -89.4   Loglik(intercept only)= -97.1
#>  Chisq= 15.36 on 1 degrees of freedom, p= 8.88e-05 
#> n= 26

^{Created on 2021-08-19 by the reprex package (v2.0.1)}

This issue should properly still stay open since the error message was quite unclear as to what the problem was. Smaller reprex:

library(parsnip)
library(survival)

parametric_model_New_func <- 
  survival_reg() %>% 
  set_engine("survival")

fit(parametric_model_New_func, 
    Surv(futime, fustat) ~ age + strata(rx), 
    data = ovarian)
#> Error: formula_ is unknown.

^{Created on 2021-08-19 by the reprex package (v2.0.1)}

[davestr2]

Wow, thank you for the very fast response! I see that the censored package is in early development. I guess I will stay with the pre-0.1.6 code for now.

Once again greatly appreciate the quick turn-around and you folks at RStudio / tidymodels / tidyberse are the best!

[hfrick]

thanks for the feedback @davestr2 ! let us know if you run into anything else survival related.

[github-actions]

This issue has been automatically locked. If you believe you have found a related problem, please file a new issue (with a reprex: https://reprex.tidyverse.org) and link to this issue.