grid_max_entropy() drops last specified value for qualitative parameters

[rbjanis]

Discovered at Applied Machine Learning Workshop at Rstudio::conf 2020.

When values are specified for qualitative parameters, the last specified value is doesn't appear in the grid produced by grid_max_entropy(). This doesn't seem to happen when grid_regular() is used.

  library(dials)
#> Loading required package: scales
knn_params <- parameters(neighbors(), 
                        weight_func(c('rectangular', 'triangular', 'epanechnikov')), 
                        dist_power()) 
knn_params <- grid_max_entropy(knn_params, size = 50)

unique(knn_params$weight_func)
#> [1] "rectangular" "triangular"
# epanechnikov missing

# Another example
params <- parameters(prune_method(c('backward', 'none', 'exhaustive', 'forward'))) 
params <- grid_max_entropy(params, size = 50)

unique(params$prune_method)
#> [1] "none"       "backward"   "exhaustive"
# forward missing

^{Created on 2020-01-31 by the reprex package (v0.3.0)}

[topepo]

library(dials)
#> Loading required package: scales

knn_params <- parameters(neighbors(), 
                         weight_func(c('rectangular', 'triangular', 'epanechnikov')), 
                         dist_power()) 
knn_params <- grid_max_entropy(knn_params, size = 50)

unique(knn_params$weight_func)
#> [1] "rectangular"  "epanechnikov" "triangular"

# Another example
params <- parameters(prune_method(c('backward', 'none', 'exhaustive', 'forward'))) 
params <- grid_max_entropy(params, size = 50)

unique(params$prune_method)
#> [1] "none"       "exhaustive" "forward"    "backward"

^{Created on 2020-02-16 by the reprex package (v0.3.0)}

[github-actions]

This issue has been automatically locked. If you believe you have found a related problem, please file a new issue (with a reprex: https://reprex.tidyverse.org) and link to this issue.