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
The EvoTreeMLE constructor supports predicitons for both Gaussian and Logistic probabilistic distrbutions (predicitons returns the estimates for each of the 2 parameters of these distributions). Otherwise, there are the deviance maximum likelihood that is used for the gamma and tweedie losses (available losses with EvoTreeRegressor) and poisson (used in EvoTreeCount).
Is there any plan to support probabilistic forecast?
Similar project below:
https://github.com/stanfordmlgroup/ngboost
https://github.com/StatMixedML/XGBoostLSS
https://github.com/StatMixedML/LightGBMLSS
The text was updated successfully, but these errors were encountered: