Examples not running

[sede-fa]

Hi,

I'm not sure if you're still working on this project- but in case you are:
I'm trying to test the HEN optimiser and the backend_examples models don't seem to be solving. I think this could be an issue with gekko - which I'm not too familiar with. I do see from the diagnostics that the DOF is negative for the examples I've tested, while running an optimisation problem from gekko with better DOF solves perfectly fine. I've also tried difference solvers such as IPOPT but this also does not work.
(this was all after removing the deprecated numpy types)

Would it be possible to know what versions of the necessary libraries were used when testing these examples?

I am using:

• Python 3.11.5
• numpy 1.24.3
• gekko 1.1.1
  (don't think the others matter so much here)

Also, how were you able to get gekko to run locally? for me it always fails so I had to switch it to remote to even see the diagnostics.

Any help would be much appreciated! :)

[PedroSeber]

Hello! Although it has been a while since I last added anything to this project, I still keep an eye on it, and use GEKKO for some other things. I have just checked, and all 6 examples run on my computer without issues.
My environment for ALChemE uses:

• Python 3.10.9
• Numpy 1.23.5
• GEKKO 1.0.6

I would assume the GEKKO version would matter, if anything.
Another important point, especially since you have mentioned that you cannot run a local model: are you using MacOS? I know that GEKKO versions < 1.0.0 do not run locally on Linux or MacOS, but since version >= 1.0.0 ran on Linux, I assumed it also worked on MacOS. The line m = GEKKO(remote = False) should be enough to run GEKKO locally otherwise.

EDIT: although I do not have any WReN examples, the WReN solver is based on SciPy instead of GEKKO, so they should work even on MacOS. Could you try running a simple WReN just to see whether ALChemE is installed properly?