Questions tagged [numba]

NumPy-aware optimizing compiler for Python

Filter by
Sorted by
Tagged with
3
votes
0answers
28 views

Unexpected CUDA processing time dependency on thread count [closed]

When calling a kernel function, the number of threads per block should ideally be a multiple of the warp size. This yields more efficient use of resources and lower processing times. However, there ...
13
votes
2answers
6k views

Why does Matlab's integral outperform integrate.quad in Scipy?

I am experiencing some frustration over the way matlab handles numerical integration vs. Scipy. I observe the following differences in my test code below: Matlab's version runs on average 24 times ...