Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Small feature requests regarding parameter name outputs #262

Open
Cole-Monnahan-NOAA opened this issue Jul 22, 2022 · 0 comments
Open

Small feature requests regarding parameter name outputs #262

Cole-Monnahan-NOAA opened this issue Jul 22, 2022 · 0 comments

Comments

@Cole-Monnahan-NOAA
Copy link
Contributor

I'd like to request two small additions. First is that during optimization, there is inconsistency in whether parameter numbers or names are reported.

For instance if there's a problem inverting the Hessian, the parameter number is printed here. Likewise when printing the sdreport variables a similar approach is taken here. It would be a big improvement to print the parameter names instead. The machinery for doing this for parameters is already developed as can be seen here in par_name. The sdreport variables will be slightly different. It's done here but doesn't include the vector index '[i]'. For instance output looks like this for a model:

Warning: Non-positive variance of sdreport variables: 423, 424, 425, 442, 452

where 423, 424, and 425 are the first three elements from a vector called slctfsh so should be outputted as slctfsh[1], slctfsh[2], slctfsh[3]. Note the value 423 comes from a vector of the parameters concatenated with the sdreport variables.

The second request is related to the parameter names as well. I need the active parameter names written to file even if the model fails to converge and produce a covariance matrix. These do get written to the .std file, but only if the Hessian is invertible. In cases when it's not, I still want these parameter names. I already modified admodel.hes to write the values here, could we also add the names on there as well? The same syntax as above, using [i] to denote vector element i, would be ideal.

So in short, could someone help me incorporate these minor changes in before the 13.0 release?

Thanks,
Cole

@johnoel johnoel modified the milestones: ADMB-13.0, ADMB-13.1 Aug 8, 2022
@johnoel johnoel removed this from the ADMB-13.1 milestone Dec 20, 2022
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants