You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Requiring separate arguments when specifying upper and lower bounds can lead to redundant computation. By taking a single tuple containing the upper/lower bounds, models can reuse quantities as much as possible
Example
This would be great for copula with discrete marginals, where the linear predictor term is used to specify bounds
Current Version:
v4.9.0
The text was updated successfully, but these errors were encountered:
I’d want to see an example, but if my first guess is correct it seems reasonable to have a lower_upper=(,) keyword
Brill! An alternative would also be to allow using transformed parameters in parameter bounds, but I have no idea how complex that might be on the backend
Description
Requiring separate arguments when specifying upper and lower bounds can lead to redundant computation. By taking a single tuple containing the upper/lower bounds, models can reuse quantities as much as possible
Example
This would be great for copula with discrete marginals, where the linear predictor term is used to specify bounds
Current Version:
v4.9.0
The text was updated successfully, but these errors were encountered: