I’d love to show off some outstanding Makie examples at JuliaCon - and in general work towards collecting examples to show off Makies capabilities and get an overview of what people are up to.
Code would be great for future integration into some Makie Gallery, but isn’t strictly necessary for the current purpose.
I want to retweet the examples (https://twitter.com/MakiePlots), since twitter offers a nice way to collect examples for a wider audience.
Don’t be shy, some of my favorite examples came from people who didn’t even consider their examples to be worth anything!
It would be nice to stick to the following template:
- short description + involved authors
- picture / gif
- link to code [optional]
- twitter handle / name for attribution, if you’re ok with retweets [optional]
Beautiful Makie is of course already on my radar, and we may want to add the examples there as well
Please only start discussions in new topics or slack, so that the thread can stay noise free
Here’s a collection of animations I made for a lecture on computational physics.
Some quick notes about each plot:
The band structure plot uses data from http://www.thp.uni-koeln.de/trebst/pubs/ChaoticTransmons.pdf. There are about 1000 individual lines in this, which takes a bit to get going but runs smoothly afterwards.
The correct and incorrect sampling plots relate to sampling points on a unit sphere. The plots are made with GitHub - ffreyer/SphereSurfaceHistogram.jl: A Histogram of unit vectors, showing sampling of theta and phi (incorrect) vs sampling of z and phi (correct).
The entropy plot shows the entropy of a classical Ising model on a triangular lattice. The antiferromagnetic version is a classical spin liquid at low temperatures, i.e. it has many different configuration fighting for dominance and therefore some residual entropy.
The Fermi surface plot shows an energy dispersion of -2t (cos(kx) + cos(ky)). The line is just a contour3d! plot with one level.
The Kullback Leibler plot shows a continuous Gaussian vs values sampled with
randn. The top plot shows the distributions, the bottom plot a measure for the difference between them.
For the percolation plot each cell is occupied with a probability p. Cells that are directly connected count as a cluster and get the same color. Around a critical value of p clusters will become percolating, i.e. connect end-to-end regardless of many cells you have. This is an example for universality.
The Wang Landau plot illustrates the related Monte-Carlo algorithm. The left side shows the logarithm of the density of states of a square lattice, the right side the histogram that the algorithm flattens. Each run is a refinement step.