Stephen Zhang

This looks like a nice addition. Is there a reason we use `discretemeasure` instead of `DiscreteNonParametric` directly? Perhaps in order to support the use-case as you have, where distances are...

Hi @CaiYitao , Thanks for adding this issue! The link to "Geometric Optimal Transport" takes me to a course webpage. Is there a particular algorithm that you have in mind?...

> Thx for the great package, I have just been toying with it. I noticed there are some newer near-linear time sinkhorn-based algorithms (Greenkhorn and screenkhorn) that are available in...

Edit - how funny. I looked at the last issue and I figured out what was the problem. Would be good if this could be (conspiciously) added to the docs...

@lucianolorenti's package is now archived, and from what I can tell there isn't really another actively maintained and easily useable spectral clustering functionality. In my view it would make sense...

> @zsteve If it is about adding 1-2 source files and little to none new deps (all from JuliaXXX organisations), and there are no licensing issues of reusing that code,...

Hi @Li-shiyue, > however it needs true probability vector whicih needs sum to 1 The Wasserstein distance requires that both inputs have the same sum. In the setting of DiscreteNonParametric,...

@Tchanders It is still possible that similar errors may persist elsewhere in the source, I haven't yet checked beyond Measures.jl

bump @Tchanders

This is still an issue in 2023. Setting `markerstrokewidth = 0` works for me when running in a Jupyter environment, but when using Plots.jl with GR backend this doesn't seem...