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
I don't know much about ESNs beyond the basics, but I was wondering whether there are any barriers to switching from linear models to generalized linear models to allow time series of e.g. count data to be considered? This could involve using MLJGLMInterface.jl, though it doesn't have penalized models, or GLMNet.jl, which does have penalized models but may be more work to link, given the current codebase links to MLJLinearModels.jl.
The text was updated successfully, but these errors were encountered:
From a theoretical point of view I don't see any barriers to implement support for generalized linear models. Giving support for both the packages could be doable regardless of current implementations, I think I will delve into it in the next week. Do you already have some data you wanted to test this on?
I don't know much about ESNs beyond the basics, but I was wondering whether there are any barriers to switching from linear models to generalized linear models to allow time series of e.g. count data to be considered? This could involve using MLJGLMInterface.jl, though it doesn't have penalized models, or GLMNet.jl, which does have penalized models but may be more work to link, given the current codebase links to MLJLinearModels.jl.
The text was updated successfully, but these errors were encountered: