Hi, was wondering if there was a way to convert the wasp ‘reset button’ to a ‘seed’ function so as to input a number slider.
You can use Boolean values or 0’s and 1’s instead of the button.
Unfortunately, it is not possible at this stage to fix the seed of an aggregation. I know this is an issue, and will try to address in the next release.
This said, even having a seed will not necessarily help when running optimizations (I suppose that is what you are thinking of…). This because to search the space of a graph (which is what a Wasp aggregation eventually is), you need access to manipulate the whole graph at different points, in order to run an efficient optimization. Flipping the seed to find the best option is a purely random search process, so I doubt it will be very efficient.
Thank you for the detailed reply! I have a feeling that won’t be much of a problem if I can resolve that aspect by working with multiple wasps and a close set of rules for each! (Also the number of components in my case would be quite limited).
True! Not quite sure whether this would be the most efficient way to go about this. But have a feeling that wasp can accomplish this. Any thoughts are appreciated!
I have worked on something similar quite some time ago, developing a system to select the best modules which, no matter how aggregated, would generate functional floorplans. The approach we took was quite different: we first created a very large data-set of room modules, then for each iteration, selected a fixed number of modules, run a large number of combinatorial aggregations with that set, and evaluate the average parameters for the results. Then we selected the best performing sets, recombined them, and re-run the simulation. You can read more here: Self-Assembly Neighbourhoods
However, for such approach to be implemented in Grasshopper (we did it in Processing at the time), you probably need a bit of scripting magic to manage the data accordingly.