demystifying
demystifying copied to clipboard
Replace t-SNE with UMAP as "drop-in replacement"
McInnes et al. suggest using UMAP can speed up some calculations, and is more interpretable than t-SNE. https://pypi.org/project/umap-learn/
Thanks for the suggestion! I have yet to read through up on the method, but do you know if there's an easy way to obtain the weight of each feature in a projection?
I'm actually in the same boat; I've read a few papers that have used it, but haven't played with the method myself! As it has been described as a "drop-in" replacement, it probably is possible to do so.