parse_model
It parses a fitted R model's structure and extracts the components needed to create a dplyr formula for prediction. The function also creates a data frame using an specific format so that other functions in the future can also pass parsed tables to a given formula creating function.
parse_model(model)Arguments
| model | An R model object. It currently supports lm(), glm() and randomForest() models. |
Examples
library(dplyr)
df <- mutate(mtcars, cyl = paste0("cyl", cyl))
model <- lm(mpg ~ wt + cyl * disp, offset = am, data = df)
parse_model(model)#> # A tibble: 13 x 14
#> labels estimate type field_1 field_2 field_3 qr_1 qr_2 qr_3
#> <chr> <dbl> <chr> <chr> <chr> <chr> <dbl> <dbl> <dbl>
#> 1 (Intercep~ 39.4 term <NA> <NA> <NA> - 0.177 - 0.591 - 0.126
#> 2 wt - 1.62 term <NA> <NA> {{:}} 0 0.184 0.0101
#> 3 cylcyl6 -18.4 term cyl6 <NA> <NA> 0 0 0.428
#> 4 cylcyl8 -16.2 term cyl8 <NA> <NA> 0 0 0
#> 5 disp - 0.0930 term <NA> {{:}} <NA> 0 0 0
#> 6 cylcyl6:d~ 0.111 term cyl6 {{:}} <NA> 0 0 0
#> 7 cylcyl8:d~ 0.0880 term cyl8 {{:}} <NA> 0 0 0
#> 8 labels 0 variab~ cyl disp wt NA NA NA
#> 9 model NA variab~ <NA> <NA> <NA> NA NA NA
#> 10 version NA variab~ <NA> <NA> <NA> NA NA NA
#> 11 residual NA variab~ <NA> <NA> <NA> NA NA NA
#> 12 sigma2 NA variab~ <NA> <NA> <NA> NA NA NA
#> 13 offset NA variab~ <NA> <NA> <NA> NA NA NA
#> # ... with 5 more variables: qr_4 <dbl>, qr_5 <dbl>, qr_6 <dbl>, qr_7 <dbl>,
#> # vals <chr>