PCovR-WHODataset takes super long to compute

scikit-learn-contrib / scikit-matter

A collection of scikit-learn compatible utilities that implement methods born out of the materials science and chemistry communities

https://scikit-matter.readthedocs.io/en/v0.2.0/

BSD 3-Clause "New" or "Revised" License

70 stars 18 forks source link

PCovR-WHODataset takes super long to compute #175

Closed PicoCentauri closed 1 year ago

PicoCentauri commented 1 year ago

The example suit of scikit-matter is super nice and well structured. Most of them run also fast, but while I was working on #170 on found that the PCovR-WHODataset examples take ages to compute.

It might be worth revising the example to speed this up. An example that takes several minutes for execution is not very useful. If there is no way to speed this up we at least should add a note.

agoscinski commented 1 year ago

I agree with that. That was also what I planned to do after the publication.

rosecers commented 1 year ago

I'll double-check the kernel CV bottleneck and see if we can significantly cut down there.

rosecers commented 1 year ago

Addressed in #189