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
Which functions should be externally accessible. I propose, similar to glmnet, two functions:
hdcd, similar to currently BinarySegmentation
cv.hdcd, similar to CrossValidation
I do not like the idea of having one function that does cv if one parameter is missing or similar. Maybe throw a warning, choose lambda according to asymptotic theory, delta = 0.1 and return a list of cps corresponding to different gammas if no params are given. Choose a default method (e.g. glasso) and a default optimizer.
The text was updated successfully, but these errors were encountered:
Which functions should be externally accessible. I propose, similar to glmnet, two functions:
I do not like the idea of having one function that does cv if one parameter is missing or similar. Maybe throw a warning, choose lambda according to asymptotic theory, delta = 0.1 and return a list of cps corresponding to different gammas if no params are given. Choose a default method (e.g. glasso) and a default optimizer.
The text was updated successfully, but these errors were encountered: