Limits of the UMAP manifold

[chrisfoulon]

Hi,

I have started using UMAP and have been playing with some examples and I was wondering whether it was possible to know the boundaries of the values of the output of UMAP.
For example, I generate a 2D representation of my high dimensional data, is there a limit to the values I can have in the 2D space? Is it knowable before running UMAP? If not, can I calculate a limit to the values after I trained the UMAP (in the case I will want to add new data point in the same space after the training)?

Maybe I missed something or misunderstood how it works and it's a dumb question, I don't know! :)

Anyway, thank you for this amazing tool!

Cheers,
Chris