I assume the size of matrix B in the figure below should be (n, p) instead of (m, p): > For input arrays where A.shape == (m, n) and B.shape == (n, p) then the result shape will be C.shape = (m, p).  URL: https://nyu-cds.github.io/python-numba/05-cuda/
I assume the size of matrix B in the figure below should be (n, p) instead of (m, p):
URL: https://nyu-cds.github.io/python-numba/05-cuda/