Grasshopper

algorithmic modeling for Rhino

Wallacei Team,

Needless to say...great improvement over Galapagos. Is there a way for the Genome Decoder to provide an ordered list of the gene titles? I have 13 genes running, and when Wallacei auto-selects the "wlc_" genes, I don't know which order they are in to replicate any particular solution. Would be great if it could provide a separate output with "headers" in the same order that the genes are offered from the Gene Decoder component.

Unless there's already a way to do this and I'm missing something?

Thanks!

Views: 460

Replies to This Discussion

Hi Jeremy,

Have a look at this video:

Wallacei X - Order of Genes when Connecting to Wallacei X:
https://www.youtube.com/watch?time_continue=3&v=xQW8mGQagN0
There are many other short tutorials and tips you can find on wallacei website:

https://www.wallacei.com/learn

Thanks! The video is helpful to see how Wallacei prioritizes gene input and sorting. I can add a number prefix to each slider in order to order it in a predictable fashion.

If at all possible, I do still think it would be useful to provide a "header list" in the output. My issue is that I'm developing a few tools for GH users who are not incredibly experienced, and any textual reference is incredibly valuable for them.

Cheers,

Hello again Jeremy,

In addition to the video tutorials; the recommended 'best practice' when connecting genes into the WallaceiX component, especially for the purposes of decoding the genomes of the evolved population for further analysis, is by right clicking the WallaceiX component and clicking ‘select all sliders’.

When doing so (connecting genes using the ‘select all sliders’ feature), please note the following:


1. For WallaceiX to automatically select sliders/genepools, make sure you rename the sliders/genepools with the following prefix (ignore the quotation marks) “wlc_” (please note, do not using capital letters, and make sure you include the underscore).

2. The order that WallaceiX will select the sliders is either numerically or alphabetically (numerical first then alphabetical). If using numerical numbering, please follow the windows format of 00, 01, 02, etc…. 

3. Despite the numerical or alphabetical naming above, number sliders will always be ordered before genepools. Here below is an example of the above:

Additionally, if you are interested in ‘recreating’ a single solution to analyse it, you can also export the solution’s phenotype in Wallacei’s User interface (by adding the solution in the second tab of the user interface and clicking ‘export’ in the third tab of the user interface). When you export only one solution, it reinstates its genome (i.e. its slider values) in the Grasshopper canvas. Thus allowing you to conduct further analysis on that specific solution. You can watch is a short video tutorial for re-instating a solution here:

https://www.youtube.com/watch?v=DDsiyyj8ob8

I hope the above makes things clearer! 

Hey Mohammed,

Thanks for the notes. I've been following the best practices as noted (which works just great, btw). I'll name all my sliders "wlc_01 (descriptive title)" as a standard practice. Exporting a single solution to re-instate the gh sliders is a nifty solution.

What I'm trying to achieve is an interface where multiple solutions can be exported, and as the user slides through the options, the gene values with titles would display nicely in the canvas:

I can make the naming convention work. Maybe something for a future release if others find it useful.

One other thing I just noticed...the tree structure after the phenotype decoder gets reset. I don't know if others find it useful downstream, but would it be possible to maintain the tree structure {gen,indv} after the phenotype decoder?

And if it wasn't already clear...you guys are rockstars! This is so valuable as a tool and interface.

Thanks for your feedback, we are working on this function. hopefully, the decoder will have this output in the next update.

Cheers,
Tao
Hi Jeremy,

The data ouputted by the 'decode phenotype' component maintains the path numbers of each phenotype (i.e. maintains the generation number and solution number).. please see below. (note that when using the decode phenotype component, and empty path is added to each branch, this is easily removed by simplifying the data tree).

https://www.dropbox.com/s/2vwcnzaupl6szln/Decode%20Phenotype%20-%20...

Also something to note, one of the great features of the decode phenotype component is that it maintains the tree structure of the original phenotype you inputted into the Wallacei X component. The image below should help clarify what i mean.

https://www.dropbox.com/s/xibq5h2j1207qbo/Decode%20Phenotype%20-%20...

(apologies for sending the images through dropbox, the buttons for inserting images to this post have gone missing from my webpage for some reason!!)

I am just adding Mohammed's images to the discussion:

RSS

About

Translate

Search

Photos

  • Add Photos
  • View All

© 2021   Created by Scott Davidson.   Powered by

Badges  |  Report an Issue  |  Terms of Service