Galapagos does not alter the gene pool input sliders when the solver is run. Works with number sliders though

Grasshopper
1

I am trying to give data to the ‘genome’ of a galapagos component via ‘gene pool’. When the solver is run, it just sets all the input to their minimum value in the ‘gene pool’ sliders and keeps running with no change and a constant straight graph. The fitness value in galapagos is set to ‘minimum’. The ‘gene pool’ has a count of 8 (for the 8 curves set) with min. & max. as -50 & 50 respectively.

image
image1584×631 208 KB

image
image1082×923 236 KB

The same definition works when I provide the input to the ‘genome’ via number sliders.
Why is it not working with ‘gene pool’ input?

Thanks.

P.S. When I manually move the ‘gene pool’ sliders, the curves do move.

2

Can you upload the gh file?

3

Hi David,
Attaching the ‘gh’ files - one with ‘gene pool’ sliders for genome, and another with number sliders. Now the ‘gene pool’ sliders are not set to their minimum upon running the solver, but they don’t move at all, and no change in the geometry.

I have worked further on the algorithm (with number sliders). Can you please also give your views on the fitness functions I am trying? The goal is to have the most compact arrangement.

d_nesting_galapagos_working
d_nesting_galapagos_working1354×738 431 KB

I observed the variable values and set the lower limits and ranges. Though it seems to work better with weights, I would like to avoid using such constants and make the function effective for any magnitude of input.

Thanks.

d_nesting_galapagos3_simplified_genePools.gh (34.8 KB)
d_nesting_galapagos3_simplified_NumSliders.gh (34.4 KB)

5

Anybody?

6

@DavidRutten Did you get the chance to look at the gh file?

Thanks.

7

Indeed genepool doesn’t work on Mac @dan could you fix this please!

8

I see this here. Logged in RH-52149. Thanks for reporting this and providing a definition.