DIFM-Brain / ofpetrial

GNU General Public License v3.0
0 stars 1 forks source link

Two by two design error #7

Closed brittanikedge closed 4 months ago

brittanikedge commented 5 months ago

When both inputs have two rates, the resulting designs are both perfect checkerboards resulting in only two combinations of the input rates

tmieno2 commented 5 months ago

This happens with "ls" option?

brittanikedge commented 5 months ago

@tmieno2 Let me check, I've been working on adapting the scripts for the website to work with the package, and I tested the new assign_rates() function on this case with two on/off trials. But I don't remember all of the options I used.

brittanikedge commented 5 months ago

Yes, with "ls

tmieno2 commented 5 months ago

Cool. Thanks. Let me try that.

tmieno2 commented 5 months ago

@brittanikedge, please check the 2 by 2 case and close this if the code works fine.

brittanikedge commented 5 months ago
Screenshot 2024-04-10 at 1 01 19 PM

@tmieno2 Here is the result for the 2 by 2 case. I was expecting a checkerboard but with two identical rates side-by-side, but the result creates horizontal stripes across the field.

tmieno2 commented 5 months ago

Yeah, actually, I don't exactly know which one is better. This or the one you expected. This is what it looks like with longer plots. But, it is possible to have a ditch that happens to follow one of the stripes exactly.

Screenshot 2024-04-10 at 1 43 06 PM
brittanikedge commented 5 months ago

Yeah, I'm not sure either, although somehow I feel like with the stripes at a slight diagonal, it is more likely that a field characteristic could follow the same spatial pattern. But that is just an intuition.