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
When I run optimize, it sometimes exits with the error message
Optimization terminated with error:
Line search failed to achieve a sufficient decrease, no more progress can be made
returning the error code 1 to the OS.
However, this may happen after a good fit is found and written and actually indicates convergence and the failure is likely due to numerical weirdness near the optimum.
So I am not sure whether this should even be treated as an error. If yes, I would like to propose for cmdstan to return an error code different than 1 to the OS, so the shell calling it may decide to ignore it easily (where now parsing stderr would be required). Now, every error seems to simply return 1.
In fact, returning a different error code for each different possible problem would perhaps be nice, but this one particular case strikes me as actually not-always-a-problem.
The text was updated successfully, but these errors were encountered:
When I run optimize, it sometimes exits with the error message
returning the error code 1 to the OS.
However, this may happen after a good fit is found and written and actually indicates convergence and the failure is likely due to numerical weirdness near the optimum.
So I am not sure whether this should even be treated as an error. If yes, I would like to propose for cmdstan to return an error code different than 1 to the OS, so the shell calling it may decide to ignore it easily (where now parsing stderr would be required). Now, every error seems to simply return 1.
In fact, returning a different error code for each different possible problem would perhaps be nice, but this one particular case strikes me as actually not-always-a-problem.
The text was updated successfully, but these errors were encountered: