Working Octopus Explicit example, Rhino 7?

Grasshopper
,
1

Hi guys.

I’ve been looking at building a replacement for Galapagos in my code from Octopus Explicit components.

I need to be able to start, monitor and stop the solver process via HumanUI, which is not available either via Galapagos or the Standard Octopus component.

The only example I have found using the Explicit components is in the Octopus 0.4 install file. This example shows a bunch of errors on opening then crashes in Rhino 5/6/7.

I have managed to open it in R5 with grasshopper solver disabled, and followed the logic to rebuild in R7/GH1.

So I can get the Octopus full solver to run and do find good solutions on my fitness function.

But replacing this with the explicit parts I only get nulls returned on a population test of 100.

I can’t post my current code, I might try and create a test problem that replicates the issue.

Does anyone have a modern working example using these components that I can follow?

Cheers

DK

2 Likes
2

hi there, trying to do a similar thing. do you have any updates or code you could share?

3

Octopus hasn’t been maintained since 2018, so I can’t say whether or not it will play nice with modern plugins, or even work correctly in Rhino 8. As an alternative you could try using Wallacei as a multi objective optimizer instead? I don’t know if it’s compatible with Human UI.

4

From memory I did manage to get the Octopus components working with a HumanUI control.

BUT

Then I made the decision to rebuild my code to run cross platform and ended up writing my own replacements for both Octopus and HumanUI.

Sorry I’m not able to share that code at this current time.

However using Anemone for looping it’s not impossible to build a genetic solver in grasshopper alone, no python/C# required.

Cheers

DK

5

thanks for the feedback! glad you got it working. When using Anemone, did you end up using Octopus or evolutionary solver modules within the loop or was it more if a brute force type of thing with native components until you found a solution you thought worked? thanks for the help.

6

100% brute force, I just looked at a few YouTube videos describing how to write genetic solvers and worked out the details using Anemone.

It is in fact two loops you need, an inner loop over the current set of results and an outer loop to feed the latest set of genomes back into the system.

Its not simple but also not beyond a bit of r+d.

Cheers

DK

1 Like