Error in overscope_eval_next(overscope, expr) : object 'G' not found

[taqtiqa-mark]

Thank you for all your efforts in making these libraries available. Much appreciated!
I believe I have found a minimal(ish) reproducible example of an issue I am also encountering.

library(tidyverse)
library(modelr)

smaller <- diamonds %>% filter(carat < 3)
mod_diamond2 <- lm(price ~ carat + color + cut + clarity, data = smaller)
grid <- smaller %>% 
  data_grid(cut, .model = mod_diamond2) %>% 
  add_predictions(mod_diamond2)
#> Error in eval_tidy(xs[[i]], unique_output): object 'G' not found

Created on 2018-05-10 by the reprex package (v0.2.0).

[hughchipman]

FWIW, a specific fix to the R for Data Science book is to replace

grid <- diamonds2 %>%
data_grid(cut, .model = mod_diamond2) %>%
add_predictions(mod_diamond2)

with this:

grid <- diamonds2 %>%
data_grid(cut, lcarat = -0.515, color = "G", clarity = "SI1") %>%
add_predictions(mod_diamond2)

Basically explicitly specifying the levels of all factors that are not to be gridded.

[hadley]

Minimal reprex:

library(modelr)

df <- tibble::tibble(x = rep(c("a", "b"), each = 5), y = rep(c("a", "b"), 5), z = rnorm(10))
mod <- lm(z ~ x + y, data = df)
data_grid(df, .model = mod)
#> Error in eval_tidy(xs[[i]], unique_output): object 'a' not found

Created on 2018-05-10 by the reprex package (v0.2.0).