model convergence #333

Activity

1. brandmaier commented on Jul 9, 2026

   

   brandmaier

   on Jul 9, 2026

   ContributorAuthor

   More actions

   I understand that this is optimizer-dependent. Still, couldn't we implement a default way to access the Optim result status for standard models?
2. Maximilian-Stefan-Ernst commented on Jul 9, 2026

   

   Maximilian-Stefan-Ernst

   on Jul 9, 2026

   Collaborator

   More actions

   The SemFit object is currently documented as

   help?> SemFit
   search: SemFit SemFIML Sem edit SemML SemRidge @edit Set SemWLS exit
   
     SemFit
   
     Fitted structural equation model.
   
     Interfaces
     ≡≡≡≡≡≡≡≡≡≡
   
       •  minimum(::SemFit) -> minimum objective value
   
       •  solution(::SemFit) -> parameter estimates
   
       •  start_val(::SemFit) -> starting values
   
       •  model(::SemFit)
   
       •  optimization_result(::SemFit)
   
       •  algorithm_name(::SemFit) -> optimization algorithm
   
       •  n_iterations(::SemFit) -> number of iterations
   
       •  convergence(::SemFit) -> convergence flags
   
       •  converged(::SemFit) -> convergence success

   so the function "converged" is what you are looking for. This relies on the flags of the used backend, and might be different for each backend. Other convergence properties has to be computed manually atm (heywood cases, for example).

   ❤️React with ❤️1Reacted by Andreas Brandmaier
3. brandmaier commented on Jul 9, 2026

   

   brandmaier

   on Jul 9, 2026

   ContributorAuthor

   More actions

   Perfect! Thanks, @Maximilian-Stefan-Ernst
4. brandmaier

   closed this as completedon Jul 9, 2026