superevolution

[MichielStock]

No description provided.

[MichielStock]

• please rename your notebook to something more simple
• you can showcase your implementation of the superformula
• mathematical variables in markdown should be in math mode, e.g. $a$
• for the plots, it might make sense to put the aspect ratio to equal
• "The Blind Watchmaker" is not from 2015, it is other than you (or even me, if I am not mistaken)
• for the objective you best use fourier descriptors: just perform a FFT on the radius in polar coordinates and take the MSE between target and model.

Looks already impressive, good work!

[yvanlaere]

I don't know if this is the right place to do it, but here is my peer review.

I have to use Pluto.reset_notebook_environment(path) every time i try to use the notebook, otherwise i get an error about failing to precompile Plots. This may be something on my end though.

In the "Exploring what the Superformula can do" section, maybe don't allow the first two sliders (a and b) to be set to 0, that makes the figure disappear.

I don't know if it's possible, but maybe move the legend in the applet, so that it doesn't cover the figures.

At first, i couldn't get part 4: Fitting Supershapes to images of diatoms, flowers and starfish using simulated annealing and MAP-Elites to work. In the range() function, "start" isn't a keyword argument in my julia version. After changing that, it worked for me.

Apart from those things, it's a very extensive and well documented notebook with a lot of interaction. The applet is really nice.

[Triana-Forment]

I just have a super minor remark, when using ellipsis "...", there is no space between the last word and ellipsis.
For the rest, I just had the error running SA and MAP-Elites, but between you and the feedback of yvanlaere, it was fixed,
It's an amazing project that I've enjoyed a lot.

[vandeveldet]

Great notebook, nice to see the evolution! You were able to make a really clear walkthrough. Some notes:

• Was it explained somewhere what a 'mutation' is, i.e. adding rand()?
• I had an error running SA and MAP-Elites, but this was solved when update to julia 1.7.1.
• When running the plots on all mutants, I would implement that as a function, maybe use a loop for creating mutants + plots
• In the SA plot, it says 'ME approximation'. Maybe zoom in a bit to make the figure more relevant?
• Are there certain limits for the parameters?