python-adaptive / adaptive

:chart_with_upwards_trend: Adaptive: parallel active learning of mathematical functions

Home Page:http://adaptive.readthedocs.io/

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Triangulation and compute volume

rahimentezari opened this issue · comments

Thanks for your great work.
I am working on high dimensional data (8,200K) and I want to sample the convex hull of such points in this space.
I want to use triangulation and then simplex_volume_in_embedding.
I wonder how I can use your code to form triangulation and then simplex_volume_in_embedding? I cannot find an example for learner.triangulation.

Hi!

I don't think methods that rely on locality (like triangulation) work well even with 5-dimensional data due to the curse of dimensionality. Please describe your problem in more detail—maybe there's a better alternative.

Or do you mean that your function values are so high-dimensional?

Overall the triangulation code is relatively straightforward, but it only implements a low level API, and we didn't design it for end users.