search

wpeterman / ResistanceGA

Optimize resistance surfaces using Genetic Algorithms
36 stars 15 forks source link

multiple surface optimization - value for continous surface #38

Closed colplainwil closed 1 year ago

colplainwil commented 1 year ago

Hello,

I discovered the package and think it will be a great tool in analysing my data. I worked through the vignette (AppendixS2 from the paper) and I have had no problems or errors.

However I am not sure how to retrieve resistance values for continuous surfaces from a multiple surface optimization (MS_optim). From the vignette I gathered, that I can access the resistance values via Multi.Surface_optim.gd$GA.summary@solution and that x1 to xn are my values. In my case I have two categorical resistance surfaces and one continuous surface. The categorical surfaces have 12 and 10 categories respectively. So x1-x12 are the resistance values for the categories of my first surface and x13-x22 are the values for my second surface. Following the vignette, I named the last three values (x23-x25) Transformation (7.46), Shape (0.62) and Max (389.90) and gathered, that they represent the continuous surface.

Now I am wondering how, assumed I interpreted the output correctly, I can find out more about the resistance of the continuous surface. After I did a single surface optimization for the continous surface, I got a graph in the results folder, showing how my original values correspond with the resistance values. Can I create a similar graph based on the output from the multiple surface optimization?

If there is a tutorial/vignette on how to interpret results of a continuous surface in a multiple surface optimization and I have missed it so far, I would be grateful for a link. Otherwise I would be grateful for any input.

Thanks in advance.

colplainwil commented 1 year ago

I just realized, that the function Plot.trans probably does exactly what i want: Plot.trans(PARM = c(0.62, 389.90), Resistance= resist$continous, transformation = 7.46)

I am still not 100% sure I interpreted the output correctly. Maybe it would be possible to give the output of Multi.Surface_optim.gd$GA.summary@solution meaningful labels in the future?

wpeterman commented 1 year ago

Assuming the order of your surfaces is correct, your interpretation of values and use of Plot.trans are correct.