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

pestpp-glm wfinit ignored? #138

Open
michal686 opened this issue Oct 14, 2021 · 2 comments
Open

pestpp-glm wfinit ignored? #138

michal686 opened this issue Oct 14, 2021 · 2 comments

Comments

@michal686
Copy link

michal686 commented Oct 14, 2021

Hi,
I am not sure if it is or not, however it seems like a bug to me. Here is part of the record file:

             pestpp-glm 5.1.1

    by The PEST++ Development Team



version: 5.1.1
binary compiled on Sep 21 2021 at 16:02:06

...

Regularization information:
    phimlim = 1e-10
    fracphim = 0.1
    phimaccept = 1.05e-10
    wfinit = 1

...

    ---  Solving for regularization weight factor   ---   
    Starting regularization weight factor      : 0
    Starting measurement objective function    : 600
    Starting regularization objective function : 0

    Target measurement objective function      : 60

      updated regularization  weight factor            : 1e-10
      FRACPHIM adjusted measurement objective function : 1.79769e+308
      discrepancy                                      : inf%

    ---  optimal regularization weight factor found  ---   
      updated regularization  weight factor            : 1e-10
      FRACPHIM adjusted measurement objective function : 1.79769e+308
      discrepancy                                      : inf%

It looks like if the routine responsible for regularization weight factor calculation did not received wfinit value correctly, instead it was zero initialized. What do you think, is this normal?

Best wishes
Michal

@jtwhite79
Copy link
Collaborator

@michal686 thanks for reporting this. The phimlim solver in pestpp-glm is admittedly weak and needs work - pest_hp has this functionality dialed. I will look into this issue tho...

J

@michal686
Copy link
Author

@jtwhite79 thank you for looking into it. I am not sure if it could harm the process or not. It was the 1.79e+308 value for measuremt objective function that got me worried. However, the parameter upgrade went well, so it is probably not a critical issue. I used pest_hp before, but there was a problem of agents disconnecting from manager, when it was busy for a long time period. So I decided to give pestpp a try :)

Best wishes
Michal

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