mod_eval doesn't work for models of form y ~ 1

[dtkaplan]

> lm(hp ~ 1, data = mtcars) -> mod
> mod_eval(mod)

Error in abs(n) : non-numeric argument to mathematical function

> mod_eval(mod, nlevels = 1)

Error in abs(n) : non-numeric argument to mathematical function

[rpruim]

It looks like you did something to handle these cases.

Should we

(a) drop the null_model_input column before returning the results?
(b) run the output through distinct() so that we never return duplicated rows? [Can that happen now?]

library(mosaicModel)
#> Loading required package: mosaicCore
#> Loading required package: splines
#> Loading required package: magrittr
model <- lm(hp ~ 1, data = mtcars) 
mod_eval(model)
#> # A tibble: 1 x 2
#>   null_model_input model_output
#>   <lgl>                   <dbl>
#> 1 NA                       147.
mod_eval(model, nlevels = 1)
#> # A tibble: 1 x 2
#>   null_model_input model_output
#>   <lgl>                   <dbl>
#> 1 NA                       147.

Created on 2018-06-22 by the reprex package (v0.2.0).