search

LLNL / MuyGPyS

A fast, pure python implementation of the MuyGPs Gaussian process realization and training algorithm.
Other
25 stars 11 forks source link

numpy parallelism is slow #20

Open bwpriest opened 2 years ago

bwpriest commented 2 years ago

The solve operations actually speed up when numpy is limited to use only one core, i.e.

import os
os.environ["OMP_NUM_THREADS"] = "1"
import numpy as np

It is not obvious why this is the case, nor how to fix it.