Open bosscha opened 5 years ago
Ok, thanks for the update. I will try it and close the issue. For now i am using dbscan quite successfully..
unfortunately, I did not have success with AP in Julia. The result is not consistent with python sklearn which was supposed to be more stable. https://github.com/JuliaStats/Clustering.jl/issues/131
my comparison code is not correct. python start from 0, so I should add 1 for python labels. Julia implementation seems correct.
actually, the result do have some difference.
it turns out just some label value difference, the rand index is 1.0, so it is correct!
The affinity propagation clustering method would gain with some docs on the use.