Closed mkborregaard closed 3 years ago
This should be good to be reviewed. Try and check some of the choices I made e.g. in terms of giving the labels random numbers.
Here's how blend
now works with a cluster
wave = rand(WaveSurface(0, 2), 200, 200)
sources = unique(rand(1:40000, 10))
clust = NeutralLandscapes.classify!(rand(DistanceGradient(sources), (200, 200)), [0.5, 1, 1, 0.5])
res = blend(clust, wave)
plot(
heatmap(wave),
heatmap(clust),
heatmap(res),
size = (1400, 350), layout = (1,3)
)
And this is the label function in action itself
heatmap(label(clust)[1], fillcolor = :picasso)
I really struggled with debugging this one - fun challenge but pretty happy to have the PR made. label(clust)
in the example runs in 51 ms on my laptop.
Codecov Report
7.49% <0.00%> (-0.85%)
Flags with carried forward coverage won't be shown. Click here to find out more.
100.00% <ø> (ø)
0.00% <0.00%> (ø)
Continue to review full report at Codecov.