Branching feedback (again)

[multimeric]

I finally was able to run through the new grouped example, since #51 was merged!

I think we still have the problem where we're not using the absolute simplest type of branching. The list iteration list was a bit tricky because of the [[1]] needed, but tar_group_by() is also complex because it uses tarchetypes, it is a target factory, and it hides the dplyr::group_by that people know.

How about just pulling out a vector of species names, and branching over that?

list(
  tar_target(penguins_csv_file, path_to_file("penguins_raw.csv"), packages="palmerpenguins"),
  tar_target(penguins_data_raw, read_csv(penguins_csv_file, show_col_types=FALSE)),
  tar_target(penguins_data, clean_penguin_data(penguins_data_raw)),
  tar_target(
    species_name,
    penguins_data |> dplyr::pull(species) |> unique(),
  ),
  tar_target(
    species_summary,
    penguins_data |> dplyr::filter(species == species_name) |> model_glance(),
    pattern = map(species_name)
  )
)

> tar_read(species_name)
[1] "Adelie Penguin (Pygoscelis adeliae)"       "Gentoo penguin (Pygoscelis papua)"        
[3] "Chinstrap penguin (Pygoscelis antarctica)"
> tar_read(species_summary)
# A tibble: 3 × 13
  r.squared adj.r.squared sigma statistic  p.value    df logLik   AIC   BIC deviance df.residual
      <dbl>         <dbl> <dbl>     <dbl>    <dbl> <dbl>  <dbl> <dbl> <dbl>    <dbl>       <int>
1     0.153         0.148 1.12       27.0 6.67e- 7     1 -231.   468.  477.    188.          149
2     0.414         0.409 0.754      85.5 1.02e-15     1 -139.   284.  292.     68.8         121
3     0.427         0.418 0.866      49.2 1.53e- 9     1  -85.7  177.  184.     49.5          66
# ℹ 2 more variables: nobs <int>, species_name <chr>