Improve machine learning vignette

vincentarelbundock / marginaleffects

R package to compute and plot predictions, slopes, marginal means, and comparisons (contrasts, risk ratios, odds, etc.) for over 100 classes of statistical and ML models. Conduct linear and non-linear hypothesis tests, or equivalence tests. Calculate uncertainty estimates using the delta method, bootstrapping, or simulation-based inference

https://marginaleffects.com

Other

462 stars 47 forks source link

Improve machine learning vignette #920

Closed vincentarelbundock closed 11 months ago

vincentarelbundock commented 1 year ago

[ ] Explicit and clear distinction between ALE and PDP, with examples
[x] Classification task. May not work for mlr3 or tidymodels when the underlying engine is not supported as a standalone model.
[x] Better error message when bad type for tidymodels

etiennebacher commented 1 year ago

Hi, there are also two typos in the last plot of the vignette:

Celcius -> Celsius
title is "Black: random forest predictions. Green: LOESS smoother" but the LOESS is actually in orange

vincentarelbundock commented 1 year ago

library(tidymodels)
library(marginaleffects)

# broken
workflow(factor(gear) ~ hp * mpg, multinom_reg()) |>
  fit(data = mtcars)  |>
  get_vcov()

# works
set_engine(multinom_reg(), "nnet") |>
  fit(factor(gear) ~ mpg * hp, data = mtcars) |>
  get_vcov()

vincentarelbundock commented 1 year ago

Thanks @etiennebacher Should be fixed now