Some Optimization Algorithms are not working in Tunny

Hello @hiron,

I ran into some problems with Tunny and I hope that you can help me.

I am running Tunny Version 1.1.0 on a Trial License on Windows 11 Pro 10.0.26100 in Rhino Version 8 SR23 (8.23.25251.13001, 2025-09-08).
I am also using the Opossum Plug-In in my file.

I´ve got 13 variables and 1 objective that needs to be minimized.

The following algorithms are working for me:
TPE
GP Optuna
GP BoTorch
CMA-ES
Random
QMC
BruteForce

The following algorithms are not working for me:
AutoSampler-Function
NSGA-II
NSGA-III
HEBO
DE
INGO

When I was trying to use NSGA-II I first got the following error message:

I asked AI how to fix this and it told me to install the latest Git-Version for Windows.
I did that and added the […]\Git\cmd\git.exe to my System-PATH.
But after trying to run the before mentioned algorithms Rhino crashes completely, this time without an error message. :smiley:

Do you know what might be causing this problem and how to fix it?

Thanks a lot and all the best,
Karl

Hi @k01

Version 1.1 changed the specifications of the libraries used, requiring GIT installation.
Please install GIT from the following website:

I have modified the next development version to display this information.

If you have already installed GIT, the most likely issue is that Grasshopper files you’re trying to use with Tunny contain Python 3 components.
This causes a conflict between the Python 3 components and libraries, and we are currently investigating a fix.
You can make it work by rewriting the Python 3 code in C# or other compatible languages.

Thank you for the very fast reply, @hiron. This kind of support is outstanding!

You are right about the Python 3 components.
I am using a lot of them in the file where some of the algorithms keep crashing.

Thanks for pointing out a possible solution as well. :slight_smile:
I hope you will be able to find a fix since rewriting all the code is not an option right now.

Yes.

I’m seeking its solution now, because I also use Python3 component everywhere too and rewrite it is very hard work.

1 Like