gtsummary
gtsummary copied to clipboard
Published
20 hours ago •
ddsjoberg
ddsjoberg
Feature Request: Improve output for glmnet models
Do not use this form to ask a question, or ask for assistance. Instead, ask on https://stackoverflow.com/ using the gtsummary tag. Questions about a function's use will be closed without a response.
If you have found a bug, please briefly describe your problem and what output you expect.
INCLUDE a minimal reproducible example (AKA a reprex). If you've never heard of a reprex before, start by reading https://www.tidyverse.org/help/#reprex. It'll take minutes to master creating a reprex. ISSUES WITHOUT A REPRODUCIBLE EXAMPLE WILL LIKELY BE CLOSED WITHOUT A RESPONSE.
Hi @ddsjoberg, when passing a tidymodels object to tbl_regression I get the below error and output. Julia from the `tidymodels team is confident it's not an error on their end. Would you know how to sort this? Thanks!
``` r
library(tidymodels)
data(penguins)
my_split <- penguins %>%
na.omit() %>%
initial_split()
rec <- recipe(sex ~ species + bill_length_mm + bill_depth_mm,
data = penguins
) %>%
step_dummy(species)
glmnet_spec <- logistic_reg(penalty = 0.1, mixture = 1) %>%
set_engine("glmnet")
glmnet_fit <-
workflow(rec, glmnet_spec) %>%
last_fit(my_split)
extract_workflow(glmnet_fit) %>%
gtsummary::tbl_regression(exponentiate = TRUE) %>%
gtsummary::as_kable()
#> Extracting {workflows} model fit with `workflows::extract_fit_parsnip(x) %>% tbl_regression(...)`
#> Extracting {parsnip} model fit with `tbl_regression(x = x$fit, ...)`
#> x Unable to identify the list of variables.
#>
#> This is usually due to an error calling `stats::model.frame(x)`or `stats::model.matrix(x)`.
#> It could be the case if that type of model does not implement these methods.
#> Rarely, this error may occur if the model object was created within
#> a functional programming framework (e.g. using `lappy()`, `purrr::map()`, etc.).
| Characteristic | exp(Beta) |
|---|---|
| (Intercept) | |
| (Intercept) | -0.09 |
| (Intercept) | -0.69 |
| (Intercept) | -1.77 |
| (Intercept) | -2.97 |
| (Intercept) | -4.09 |
| (Intercept) | -5.15 |
| (Intercept) | -6.16 |
| (Intercept) | -7.11 |
| (Intercept) | -8.02 |
| (Intercept) | -9.37 |
| (Intercept) | -11.1 |
| (Intercept) | -12.8 |
| (Intercept) | -14.4 |
| (Intercept) | -16.0 |
| (Intercept) | -17.7 |
| (Intercept) | -19.2 |
| (Intercept) | -20.8 |
| (Intercept) | -22.4 |
| (Intercept) | -23.9 |
| (Intercept) | -25.4 |
| (Intercept) | -27.0 |
| (Intercept) | -28.5 |
| (Intercept) | -29.9 |
| (Intercept) | -31.4 |
| (Intercept) | -32.8 |
| (Intercept) | -34.2 |
| (Intercept) | -35.6 |
| (Intercept) | -37.0 |
| (Intercept) | -38.3 |
| (Intercept) | -39.6 |
| (Intercept) | -40.9 |
| (Intercept) | -42.1 |
| (Intercept) | -43.4 |
| (Intercept) | -44.6 |
| (Intercept) | -45.7 |
| (Intercept) | -46.8 |
| (Intercept) | -47.9 |
| (Intercept) | -48.9 |
| (Intercept) | -49.9 |
| (Intercept) | -50.8 |
| (Intercept) | -51.7 |
| (Intercept) | -52.6 |
| (Intercept) | -53.4 |
| (Intercept) | -54.1 |
| (Intercept) | -54.8 |
| (Intercept) | -55.5 |
| (Intercept) | -56.1 |
| (Intercept) | -56.6 |
| (Intercept) | -57.2 |
| (Intercept) | -57.7 |
| (Intercept) | -58.2 |
| (Intercept) | -58.6 |
| (Intercept) | -59.0 |
| (Intercept) | -59.4 |
| (Intercept) | -59.7 |
| (Intercept) | -60.0 |
| (Intercept) | -60.3 |
| (Intercept) | -60.6 |
| (Intercept) | -60.8 |
| (Intercept) | -61.0 |
| (Intercept) | -61.3 |
| (Intercept) | -61.4 |
| (Intercept) | -61.6 |
| (Intercept) | -61.8 |
| (Intercept) | -61.9 |
| (Intercept) | -62.1 |
| (Intercept) | -62.2 |
| (Intercept) | -62.3 |
| (Intercept) | -62.4 |
| (Intercept) | -62.5 |
| bill_length_mm | |
| bill_length_mm | 0.01 |
| bill_length_mm | 0.02 |
| bill_length_mm | 0.04 |
| bill_length_mm | 0.05 |
| bill_length_mm | 0.06 |
| bill_length_mm | 0.07 |
| bill_length_mm | 0.08 |
| bill_length_mm | 0.09 |
| bill_length_mm | 0.11 |
| bill_length_mm | 0.13 |
| bill_length_mm | 0.15 |
| bill_length_mm | 0.17 |
| bill_length_mm | 0.19 |
| bill_length_mm | 0.21 |
| bill_length_mm | 0.23 |
| bill_length_mm | 0.25 |
| bill_length_mm | 0.26 |
| bill_length_mm | 0.28 |
| bill_length_mm | 0.30 |
| bill_length_mm | 0.32 |
| bill_length_mm | 0.33 |
| bill_length_mm | 0.35 |
| bill_length_mm | 0.37 |
| bill_length_mm | 0.38 |
| bill_length_mm | 0.40 |
| bill_length_mm | 0.42 |
| bill_length_mm | 0.43 |
| bill_length_mm | 0.45 |
| bill_length_mm | 0.46 |
| bill_length_mm | 0.47 |
| bill_length_mm | 0.48 |
| bill_length_mm | 0.49 |
| bill_length_mm | 0.50 |
| bill_length_mm | 0.51 |
| bill_length_mm | 0.51 |
| bill_length_mm | 0.52 |
| bill_length_mm | 0.53 |
| bill_length_mm | 0.54 |
| bill_length_mm | 0.54 |
| bill_length_mm | 0.55 |
| bill_length_mm | 0.55 |
| bill_length_mm | 0.56 |
| bill_length_mm | 0.56 |
| bill_length_mm | 0.57 |
| bill_length_mm | 0.57 |
| bill_length_mm | 0.58 |
| bill_length_mm | 0.58 |
| bill_length_mm | 0.59 |
| bill_length_mm | 0.59 |
| bill_length_mm | 0.59 |
| bill_length_mm | 0.60 |
| bill_length_mm | 0.60 |
| bill_length_mm | 0.60 |
| bill_length_mm | 0.60 |
| bill_length_mm | 0.61 |
| bill_length_mm | 0.61 |
| bill_length_mm | 0.61 |
| bill_length_mm | 0.61 |
| bill_length_mm | 0.61 |
| bill_length_mm | 0.61 |
| bill_length_mm | 0.62 |
| bill_length_mm | 0.62 |
| bill_length_mm | 0.62 |
| bill_length_mm | 0.62 |
| bill_length_mm | 0.62 |
| bill_length_mm | 0.62 |
| bill_length_mm | 0.62 |
| bill_length_mm | 0.62 |
| bill_depth_mm | |
| bill_depth_mm | 0.04 |
| bill_depth_mm | 0.07 |
| bill_depth_mm | 0.11 |
| bill_depth_mm | 0.14 |
| bill_depth_mm | 0.18 |
| bill_depth_mm | 0.21 |
| bill_depth_mm | 0.23 |
| bill_depth_mm | 0.26 |
| bill_depth_mm | 0.30 |
| bill_depth_mm | 0.35 |
| bill_depth_mm | 0.40 |
| bill_depth_mm | 0.45 |
| bill_depth_mm | 0.50 |
| bill_depth_mm | 0.55 |
| bill_depth_mm | 0.60 |
| bill_depth_mm | 0.65 |
| bill_depth_mm | 0.69 |
| bill_depth_mm | 0.74 |
| bill_depth_mm | 0.79 |
| bill_depth_mm | 0.83 |
| bill_depth_mm | 0.88 |
| bill_depth_mm | 0.92 |
| bill_depth_mm | 0.96 |
| bill_depth_mm | 1.01 |
| bill_depth_mm | 1.05 |
| bill_depth_mm | 1.09 |
| bill_depth_mm | 1.13 |
| bill_depth_mm | 1.17 |
| bill_depth_mm | 1.21 |
| bill_depth_mm | 1.25 |
| bill_depth_mm | 1.30 |
| bill_depth_mm | 1.35 |
| bill_depth_mm | 1.39 |
| bill_depth_mm | 1.44 |
| bill_depth_mm | 1.48 |
| bill_depth_mm | 1.52 |
| bill_depth_mm | 1.56 |
| bill_depth_mm | 1.60 |
| bill_depth_mm | 1.64 |
| bill_depth_mm | 1.67 |
| bill_depth_mm | 1.71 |
| bill_depth_mm | 1.74 |
| bill_depth_mm | 1.77 |
| bill_depth_mm | 1.79 |
| bill_depth_mm | 1.82 |
| bill_depth_mm | 1.84 |
| bill_depth_mm | 1.86 |
| bill_depth_mm | 1.88 |
| bill_depth_mm | 1.90 |
| bill_depth_mm | 1.92 |
| bill_depth_mm | 1.94 |
| bill_depth_mm | 1.95 |
| bill_depth_mm | 1.97 |
| bill_depth_mm | 1.98 |
| bill_depth_mm | 1.99 |
| bill_depth_mm | 2.01 |
| bill_depth_mm | 2.02 |
| bill_depth_mm | 2.03 |
| bill_depth_mm | 2.03 |
| bill_depth_mm | 2.04 |
| bill_depth_mm | 2.05 |
| bill_depth_mm | 2.06 |
| bill_depth_mm | 2.06 |
| bill_depth_mm | 2.07 |
| bill_depth_mm | 2.07 |
| bill_depth_mm | 2.08 |
| bill_depth_mm | 2.08 |
| bill_depth_mm | 2.09 |
| bill_depth_mm | 2.09 |
| species_Chinstrap | |
| species_Chinstrap | -0.15 |
| species_Chinstrap | -0.42 |
| species_Chinstrap | -0.68 |
| species_Chinstrap | -0.94 |
| species_Chinstrap | -1.19 |
| species_Chinstrap | -1.43 |
| species_Chinstrap | -1.67 |
| species_Chinstrap | -1.90 |
| species_Chinstrap | -2.13 |
| species_Chinstrap | -2.36 |
| species_Chinstrap | -2.59 |
| species_Chinstrap | -2.81 |
| species_Chinstrap | -3.03 |
| species_Chinstrap | -3.25 |
| species_Chinstrap | -3.46 |
| species_Chinstrap | -3.67 |
| species_Chinstrap | -3.88 |
| species_Chinstrap | -4.08 |
| species_Chinstrap | -4.28 |
| species_Chinstrap | -4.48 |
| species_Chinstrap | -4.67 |
| species_Chinstrap | -4.85 |
| species_Chinstrap | -5.00 |
| species_Chinstrap | -5.12 |
| species_Chinstrap | -5.24 |
| species_Chinstrap | -5.36 |
| species_Chinstrap | -5.47 |
| species_Chinstrap | -5.58 |
| species_Chinstrap | -5.68 |
| species_Chinstrap | -5.78 |
| species_Chinstrap | -5.87 |
| species_Chinstrap | -5.96 |
| species_Chinstrap | -6.04 |
| species_Chinstrap | -6.13 |
| species_Chinstrap | -6.20 |
| species_Chinstrap | -6.27 |
| species_Chinstrap | -6.34 |
| species_Chinstrap | -6.40 |
| species_Chinstrap | -6.46 |
| species_Chinstrap | -6.51 |
| species_Chinstrap | -6.56 |
| species_Chinstrap | -6.61 |
| species_Chinstrap | -6.66 |
| species_Chinstrap | -6.70 |
| species_Chinstrap | -6.73 |
| species_Chinstrap | -6.77 |
| species_Chinstrap | -6.80 |
| species_Chinstrap | -6.83 |
| species_Chinstrap | -6.86 |
| species_Chinstrap | -6.88 |
| species_Chinstrap | -6.91 |
| species_Chinstrap | -6.93 |
| species_Chinstrap | -6.95 |
| species_Chinstrap | -6.97 |
| species_Chinstrap | -6.98 |
| species_Chinstrap | -7.00 |
| species_Chinstrap | -7.01 |
| species_Chinstrap | -7.02 |
| species_Chinstrap | -7.03 |
| species_Chinstrap | -7.04 |
| species_Chinstrap | -7.05 |
| species_Gentoo | |
| species_Gentoo | 0.06 |
| species_Gentoo | 0.16 |
| species_Gentoo | 0.25 |
| species_Gentoo | 0.34 |
| species_Gentoo | 0.42 |
| species_Gentoo | 0.50 |
| species_Gentoo | 0.58 |
| species_Gentoo | 0.65 |
| species_Gentoo | 0.72 |
| species_Gentoo | 0.79 |
| species_Gentoo | 0.85 |
| species_Gentoo | 0.91 |
| species_Gentoo | 0.96 |
| species_Gentoo | 1.01 |
| species_Gentoo | 1.06 |
| species_Gentoo | 1.10 |
| species_Gentoo | 1.14 |
| species_Gentoo | 1.18 |
| species_Gentoo | 1.22 |
| species_Gentoo | 1.25 |
| species_Gentoo | 1.28 |
| species_Gentoo | 1.31 |
| species_Gentoo | 1.34 |
| species_Gentoo | 1.36 |
| species_Gentoo | 1.38 |
| species_Gentoo | 1.40 |
| species_Gentoo | 1.42 |
| species_Gentoo | 1.44 |
| species_Gentoo | 1.46 |
| species_Gentoo | 1.47 |
| species_Gentoo | 1.49 |
| species_Gentoo | 1.50 |
| species_Gentoo | 1.51 |
| species_Gentoo | 1.52 |
| species_Gentoo | 1.53 |
| species_Gentoo | 1.54 |
| species_Gentoo | 1.55 |
| species_Gentoo | 1.55 |
| species_Gentoo | 1.56 |
Created on 2022-06-18 by the reprex package (v2.0.1)
Session info
sessioninfo::session_info()
#> - Session info ---------------------------------------------------------------
#> setting value
#> version R version 4.1.3 (2022-03-10)
#> os Windows 10 x64 (build 22000)
#> system x86_64, mingw32
#> ui RTerm
#> language (EN)
#> collate English_United Kingdom.1252
#> ctype English_United Kingdom.1252
#> tz Europe/London
#> date 2022-06-18
#> pandoc 2.18 @ C:/PROGRA~3/chocolatey/bin/ (via rmarkdown)
#>
#> - Packages -------------------------------------------------------------------
#> package * version date (UTC) lib source
#> assertthat 0.2.1 2019-03-21 [1] CRAN (R 4.1.2)
#> backports 1.4.1 2021-12-13 [1] CRAN (R 4.1.2)
#> broom * 0.8.0 2022-04-13 [1] CRAN (R 4.1.3)
#> broom.helpers 1.6.0 2022-01-12 [1] CRAN (R 4.1.2)
#> class 7.3-20 2022-01-13 [1] CRAN (R 4.1.2)
#> cli 3.2.0 2022-02-14 [1] CRAN (R 4.1.2)
#> codetools 0.2-18 2020-11-04 [2] CRAN (R 4.1.3)
#> colorspace 2.0-3 2022-02-21 [1] CRAN (R 4.1.3)
#> crayon 1.5.1 2022-03-26 [1] CRAN (R 4.1.3)
#> DBI 1.1.2 2021-12-20 [1] CRAN (R 4.1.2)
#> dials * 0.1.0 2022-01-31 [1] CRAN (R 4.1.2)
#> DiceDesign 1.9 2021-02-13 [1] CRAN (R 4.1.2)
#> digest 0.6.29 2021-12-01 [1] CRAN (R 4.1.2)
#> dplyr * 1.0.8 2022-02-08 [1] CRAN (R 4.1.2)
#> ellipsis 0.3.2 2021-04-29 [1] CRAN (R 4.1.2)
#> evaluate 0.15 2022-02-18 [1] CRAN (R 4.1.2)
#> fansi 1.0.3 2022-03-24 [1] CRAN (R 4.1.3)
#> fastmap 1.1.0 2021-01-25 [1] CRAN (R 4.1.2)
#> foreach 1.5.2 2022-02-02 [1] CRAN (R 4.1.2)
#> fs 1.5.2 2021-12-08 [1] CRAN (R 4.1.2)
#> furrr 0.2.3 2021-06-25 [1] CRAN (R 4.1.2)
#> future 1.24.0 2022-02-19 [1] CRAN (R 4.1.3)
#> future.apply 1.8.1 2021-08-10 [1] CRAN (R 4.1.2)
#> generics 0.1.2 2022-01-31 [1] CRAN (R 4.1.2)
#> ggplot2 * 3.3.5 2021-06-25 [1] CRAN (R 4.1.2)
#> glmnet * 4.1-4 2022-04-15 [1] CRAN (R 4.1.3)
#> globals 0.14.0 2020-11-22 [1] CRAN (R 4.1.1)
#> glue 1.6.2 2022-02-24 [1] CRAN (R 4.1.3)
#> gower 1.0.0 2022-02-03 [1] CRAN (R 4.1.2)
#> GPfit 1.0-8 2019-02-08 [1] CRAN (R 4.1.2)
#> gt 0.6.0 2022-05-24 [1] CRAN (R 4.1.3)
#> gtable 0.3.0 2019-03-25 [1] CRAN (R 4.1.2)
#> gtsummary 1.5.2 2022-01-29 [1] CRAN (R 4.1.2)
#> hardhat 0.2.0 2022-01-24 [1] CRAN (R 4.1.2)
#> highr 0.9 2021-04-16 [1] CRAN (R 4.1.2)
#> htmltools 0.5.2 2021-08-25 [1] CRAN (R 4.1.2)
#> infer * 1.0.0 2021-08-13 [1] CRAN (R 4.1.2)
#> ipred 0.9-12 2021-09-15 [1] CRAN (R 4.1.2)
#> iterators 1.0.14 2022-02-05 [1] CRAN (R 4.1.2)
#> knitr 1.39 2022-04-26 [1] CRAN (R 4.1.3)
#> lattice 0.20-45 2021-09-22 [2] CRAN (R 4.1.3)
#> lava 1.6.10 2021-09-02 [1] CRAN (R 4.1.2)
#> lhs 1.1.5 2022-03-22 [1] CRAN (R 4.1.3)
#> lifecycle 1.0.1 2021-09-24 [1] CRAN (R 4.1.2)
#> listenv 0.8.0 2019-12-05 [1] CRAN (R 4.1.2)
#> lubridate 1.8.0 2021-10-07 [1] CRAN (R 4.1.2)
#> magrittr 2.0.3 2022-03-30 [1] CRAN (R 4.1.3)
#> MASS 7.3-55 2022-01-13 [1] CRAN (R 4.1.2)
#> Matrix * 1.4-0 2021-12-08 [2] CRAN (R 4.1.3)
#> modeldata * 0.1.1 2021-07-14 [1] CRAN (R 4.1.2)
#> munsell 0.5.0 2018-06-12 [1] CRAN (R 4.1.2)
#> nnet 7.3-17 2022-01-13 [1] CRAN (R 4.1.2)
#> parallelly 1.30.0 2021-12-17 [1] CRAN (R 4.1.2)
#> parsnip * 0.2.1.9000 2022-03-29 [1] Github (tidymodels/parsnip@9ce41c8)
#> pillar 1.7.0 2022-02-01 [1] CRAN (R 4.1.2)
#> pkgconfig 2.0.3 2019-09-22 [1] CRAN (R 4.1.2)
#> plyr 1.8.7 2022-03-24 [1] CRAN (R 4.1.3)
#> pROC 1.18.0 2021-09-03 [1] CRAN (R 4.1.2)
#> prodlim 2019.11.13 2019-11-17 [1] CRAN (R 4.1.2)
#> purrr * 0.3.4 2020-04-17 [1] CRAN (R 4.1.2)
#> R.cache 0.15.0 2021-04-30 [1] CRAN (R 4.1.2)
#> R.methodsS3 1.8.1 2020-08-26 [1] CRAN (R 4.1.1)
#> R.oo 1.24.0 2020-08-26 [1] CRAN (R 4.1.1)
#> R.utils 2.11.0 2021-09-26 [1] CRAN (R 4.1.2)
#> R6 2.5.1 2021-08-19 [1] CRAN (R 4.1.2)
#> Rcpp 1.0.8.3 2022-03-17 [1] CRAN (R 4.1.3)
#> recipes * 0.2.0 2022-02-18 [1] CRAN (R 4.1.2)
#> reprex 2.0.1 2021-08-05 [1] CRAN (R 4.1.2)
#> rlang 1.0.2 2022-03-04 [1] CRAN (R 4.1.3)
#> rmarkdown 2.14 2022-04-25 [1] CRAN (R 4.1.3)
#> rpart 4.1.16 2022-01-24 [1] CRAN (R 4.1.2)
#> rsample * 0.1.1 2021-11-08 [1] CRAN (R 4.1.2)
#> rstudioapi 0.13 2020-11-12 [1] CRAN (R 4.1.2)
#> scales * 1.2.0 2022-04-13 [1] CRAN (R 4.1.3)
#> sessioninfo 1.2.2 2021-12-06 [1] CRAN (R 4.1.2)
#> shape 1.4.6 2021-05-19 [1] CRAN (R 4.1.1)
#> stringi 1.7.6 2021-11-29 [1] CRAN (R 4.1.2)
#> stringr 1.4.0 2019-02-10 [1] CRAN (R 4.1.2)
#> styler 1.7.0 2022-03-13 [1] CRAN (R 4.1.3)
#> survival 3.2-13 2021-08-24 [2] CRAN (R 4.1.3)
#> tibble * 3.1.6 2021-11-07 [1] CRAN (R 4.1.2)
#> tidymodels * 0.2.0 2022-03-19 [1] CRAN (R 4.1.3)
#> tidyr * 1.2.0 2022-02-01 [1] CRAN (R 4.1.2)
#> tidyselect 1.1.2 2022-02-21 [1] CRAN (R 4.1.3)
#> timeDate 3043.102 2018-02-21 [1] CRAN (R 4.1.1)
#> tune * 0.2.0.9000 2022-03-29 [1] Github (tidymodels/tune@6ed30a4)
#> utf8 1.2.2 2021-07-24 [1] CRAN (R 4.1.2)
#> vctrs 0.4.1 2022-04-13 [1] CRAN (R 4.1.3)
#> withr 2.5.0 2022-03-03 [1] CRAN (R 4.1.3)
#> workflows * 0.2.6 2022-03-18 [1] CRAN (R 4.1.3)
#> workflowsets * 0.2.1 2022-03-15 [1] CRAN (R 4.1.3)
#> xfun 0.30 2022-03-02 [1] CRAN (R 4.1.3)
#> yaml 2.3.5 2022-02-21 [1] CRAN (R 4.1.2)
#> yardstick * 0.0.9 2021-11-22 [1] CRAN (R 4.1.2)
#>
#> [1] C:/Users/rmkja/Documents/R/win-library/4.1
#> [2] C:/Program Files/R/R-4.1.3/library
#>
#> ------------------------------------------------------------------------------
@themichjam what is it you'd like to see from this model summary?
library(tidymodels)
data(penguins)
my_split <- penguins %>%
na.omit() %>%
initial_split()
rec <- recipe(sex ~ species + bill_length_mm + bill_depth_mm,
data = penguins) %>%
step_dummy(species)
glmnet_spec <-
logistic_reg(penalty = 0.1, mixture = 1) %>%
set_engine("glmnet")
glmnet_fit <-
workflow(rec, glmnet_spec) %>%
last_fit(my_split)
mod <-
extract_workflow(glmnet_fit) %>%
workflows::extract_fit_parsnip() %>%
purrr::pluck("fit")
class(mod)
#> [1] "lognet" "glmnet"
# THIS IS WHAT TBL_REGRESSION IS PRINTING
df_broom <- mod %>%
broom::tidy()
df_broom
#> # A tibble: 314 × 5
#> term step estimate lambda dev.ratio
#> <chr> <dbl> <dbl> <dbl> <dbl>
#> 1 (Intercept) 1 -0.0241 0.211 -3.04e-15
#> 2 (Intercept) 2 -0.686 0.192 2.18e- 2
#> 3 (Intercept) 3 -1.73 0.175 5.43e- 2
#> 4 (Intercept) 4 -3.09 0.160 9.36e- 2
#> 5 (Intercept) 5 -4.36 0.145 1.27e- 1
#> 6 (Intercept) 6 -5.57 0.133 1.56e- 1
#> 7 (Intercept) 7 -6.71 0.121 1.81e- 1
#> 8 (Intercept) 8 -7.81 0.110 2.03e- 1
#> 9 (Intercept) 9 -8.86 0.100 2.22e- 1
#> 10 (Intercept) 10 -9.87 0.0914 2.39e- 1
#> # … with 304 more rows
# IS THIS WHAT YOU WANT TO PRINT?
df_broom %>%
dplyr::filter(step == max(step))
#> # A tibble: 5 × 5
#> term step estimate lambda dev.ratio
#> <chr> <dbl> <dbl> <dbl> <dbl>
#> 1 (Intercept) 71 -64.2 0.000313 0.597
#> 2 bill_length_mm 71 0.621 0.000313 0.597
#> 3 bill_depth_mm 71 2.18 0.000313 0.597
#> 4 species_Chinstrap 71 -6.45 0.000313 0.597
#> 5 species_Gentoo 71 1.82 0.000313 0.597
Created on 2022-06-18 by the reprex package (v2.0.1)
@ddsjoberg I was wondering if it was possible to get the same output with the tidymodels model as you would get passing a traditional glm through your package's tble_regression()?
The model returns many coef estimates. Can you show me which you want to display?
NOTE: We have a check for workflow objects to print an informative message about the default dummy variable creation that occurs in worflows/parsnip. When a workflows object creates dummy variables for the categorical variables, they do it in a way such that we cannot identify the larger variable name, place the variable header, etc.: each dummy variable is treated as a separate variable in the summary.
tbl_regression.workflow <- function(x, ...) {
assert_package("workflows", "tbl_regression.workflow()")
if (isTRUE(!x$pre$actions$formula$blueprint$indicators %in% "none")) {
paste("To take full advantage of model formatting, e.g. grouping categorical",
"variables, please add the following argument to the `workflows::add_model()` call:") %>%
stringr::str_wrap() %>%
paste("`blueprint = hardhat::default_formula_blueprint(indicators = 'none')`", sep = "\n") %>%
paste("\n") %>%
rlang::inform()
}
BUT it seems that the internal structure may have changed, and this was not being triggered in the examples above.
Also, it seems that we just need a tbl_regression.glmnet() method to better handle the results of these models.
@larmarange It depends on how we seek to resolve the issue, but I think the best solutions would likely involve broom.helpers. I am no expert on workflows/parsnip, but here what I understand.
- the returned model objects do not have the typical terms object; tidymodels has alternative versions of terms,
model_frame(), etc. - The current implementation we have in gtsummary simply extracts the underlying model fit, and re-passes the object (and the original arguments) to
tbl_regression(). - A more robust solution would be to utilize tidiers built for these objects. However, workflows can be many many many different types of models. Will the unified structure of the workflows/parsnip object allow us a single tidier for all model types, or are we then going to need to have special handling of various types of models built with tidymodels?
@themichjam the issue you've raised has two parts, 1. Improving the tbl_regression() output for glmnet models, and 2. better handling for workflows/parsnip models.
We can track the first in this issue, and the second will be handled here https://github.com/larmarange/broom.helpers/issues/160 .
@themichjam to address the first part, it would be helpful to have your input on what exactly you want to report from tbl_regression(). The glmnet tidier returns many many many versions of the coefs that are returned, and the tidier is used in the background to construct the tbl_regression() tables.
cc @larmarange
Dear @ddsjoberg look at mod2 in the example below
library(tidymodels)
data(penguins)
my_split <- penguins %>%
na.omit() %>%
initial_split()
rec <- recipe(sex ~ species + bill_length_mm + bill_depth_mm,
data = penguins) %>%
step_dummy(species)
glmnet_spec <-
logistic_reg(penalty = 0.1, mixture = 1) %>%
set_engine("glmnet")
glmnet_fit <-
workflow(rec, glmnet_spec) %>%
last_fit(my_split)
mod <-
extract_workflow(glmnet_fit) %>%
workflows::extract_fit_parsnip() %>%
purrr::pluck("fit")
class(mod)
#> [1] "lognet" "glmnet"
mod %>% tidy()
#> # A tibble: 304 x 5
#> term step estimate lambda dev.ratio
#> <chr> <dbl> <dbl> <dbl> <dbl>
#> 1 (Intercept) 1 -0.0402 0.190 3.69e-15
#> 2 (Intercept) 2 -0.631 0.173 1.78e- 2
#> 3 (Intercept) 3 -1.78 0.158 4.95e- 2
#> 4 (Intercept) 4 -3.00 0.144 8.06e- 2
#> 5 (Intercept) 5 -4.15 0.131 1.07e- 1
#> 6 (Intercept) 6 -5.23 0.119 1.30e- 1
#> 7 (Intercept) 7 -6.25 0.109 1.50e- 1
#> 8 (Intercept) 8 -7.22 0.0992 1.67e- 1
#> 9 (Intercept) 9 -8.41 0.0904 1.90e- 1
#> 10 (Intercept) 10 -10.3 0.0824 2.33e- 1
#> # ... with 294 more rows
mod2 <-
extract_workflow(glmnet_fit) %>%
workflows::extract_fit_parsnip()
class(mod2)
#> [1] "_lognet" "model_fit"
mod2 %>% tidy()
#> # A tibble: 5 x 3
#> term estimate penalty
#> <chr> <dbl> <dbl>
#> 1 (Intercept) -7.14 0.1
#> 2 bill_length_mm 0.0708 0.1
#> 3 bill_depth_mm 0.232 0.1
#> 4 species_Chinstrap 0 0.1
#> 5 species_Gentoo 0 0.1
Created on 2022-06-20 by the reprex package (v2.0.1)
However, this tidier works only on a model_fit object and not on the underlying glmnet
An option would be to support natively model_fit objects in broom.helpers and to remove tbl_regression.model_fit() from gtsummary?
Very first try of native support of model_fit in broom.helpers (see https://github.com/larmarange/broom.helpers/pull/161 )
With this, you can apply tbl_regression.default() directly on a model_fit object. It will use dedicated tidy methods for model_fit while broom.helpers will consider, where relevant, the model$fit object.
library(tidymodels)
library(gtsummary)
#>
#> Attachement du package : 'gtsummary'
#> L'objet suivant est masqué depuis 'package:recipes':
#>
#> all_numeric
library(broom.helpers)
#>
#> Attachement du package : 'broom.helpers'
#> Les objets suivants sont masqués depuis 'package:gtsummary':
#>
#> all_continuous, all_contrasts
data(penguins)
my_split <- penguins %>%
na.omit() %>%
initial_split()
rec <- recipe(sex ~ species + bill_length_mm + bill_depth_mm,
data = penguins
) %>%
step_dummy(species)
glmnet_spec <- logistic_reg(penalty = 0.1, mixture = 1) %>%
set_engine("glmnet")
glmnet_fit <-
workflow(rec, glmnet_spec) %>%
last_fit(my_split)
f <- extract_fit_parsnip(glmnet_fit)
tidy(f)
#> # A tibble: 5 x 3
#> term estimate penalty
#> <chr> <dbl> <dbl>
#> 1 (Intercept) -8.50 0.1
#> 2 bill_length_mm 0.0912 0.1
#> 3 bill_depth_mm 0.258 0.1
#> 4 species_Chinstrap 0 0.1
#> 5 species_Gentoo 0 0.1
tidy_plus_plus(f)
#> x Unable to identify the list of variables.
#>
#> This is usually due to an error calling `stats::model.frame(x)`or `stats::model.matrix(x)`.
#> It could be the case if that type of model does not implement these methods.
#> Rarely, this error may occur if the model object was created within
#> a functional programming framework (e.g. using `lappy()`, `purrr::map()`, etc.).
#> # A tibble: 5 x 13
#> term variable var_label var_class var_type var_nlevels contrasts
#> <chr> <chr> <chr> <int> <chr> <int> <chr>
#> 1 (Intercept) (Interce~ (Interce~ NA unknown NA <NA>
#> 2 bill_length_mm bill_len~ bill_len~ NA unknown NA <NA>
#> 3 bill_depth_mm bill_dep~ bill_dep~ NA unknown NA <NA>
#> 4 species_Chinstrap species_~ species_~ NA unknown NA <NA>
#> 5 species_Gentoo species_~ species_~ NA unknown NA <NA>
#> # ... with 6 more variables: contrasts_type <chr>, reference_row <lgl>,
#> # label <chr>, estimate <dbl>, penalty <dbl>, n <dbl>
gtsummary:::tbl_regression.default(f) %>%
as_kable()
#> x Unable to identify the list of variables.
#>
#> This is usually due to an error calling `stats::model.frame(x)`or `stats::model.matrix(x)`.
#> It could be the case if that type of model does not implement these methods.
#> Rarely, this error may occur if the model object was created within
#> a functional programming framework (e.g. using `lappy()`, `purrr::map()`, etc.).
| Characteristic | Beta |
|---|---|
| (Intercept) | -8.5 |
| bill_length_mm | 0.09 |
| bill_depth_mm | 0.26 |
| species_Chinstrap | 0.00 |
| species_Gentoo | 0.00 |
Created on 2022-06-20 by the reprex package (v2.0.1)
@larmarange looks great! and you brought up a great point about the species being treated as individual variables because glmnet requires a matrix with summary variable coding and does not allow for the factor/character variables to be passed
Progress can be followed here: https://github.com/larmarange/broom.helpers/issues/162