We recently had several questions on stackoverflow and stats.stackexchange requesting margins, lsmeans, emmeans or similar.
One part that those function in other packages provide are premade grids of values of the explanatory variables, like quantiles of continuous variables, levels of categoricals
open issue and a quick attempt for this
The usage is for predict/get_prediction and for margins/margeff.
I would appreciate it if someone can help with this or knows of code that can be used as reference.
but I'm not really an expert in the pandas part of data handling.
Josef