Styled Marks

We start by generating some data

#let xs = lq.linspace(0, 4 * calc.pi)
#let ys = xs.map(calc.sin)

and making a noisy version of it by adding pseudo-random noise to the $x$-values.

#import "@preview/suiji:0.3.0"
#let rng = suiji.gen-rng(33)

#let (rng, noise) = suiji.normal(
  scale: 0.1, size: xs.len(), rng
)

#let xs-noisy = lq.vec.add(xs, noise)

For this we use the Typst package suiji. Now we create a uniform distribution for coloring the marks.

#let (rng, colors) = suiji.uniform(
  size: xs.len(), rng
)

First we plot the "ideal" sine wave, then we use scatter to plot the "noisy" data with varying colors.

#lq.diagram(
  lq.plot(xs, ys, mark: none), 
  lq.scatter(
    xs-noisy, ys,
    mark: "s",
    color: colors,
    map: color.map.turbo,
    stroke: .5pt + black
  )
)