[Numpy-discussion] inversion of large matrices

[David Warde-Farley]

On 2010-08-30, at 5:42 PM, Melissa Mendonça wrote:

> I'm just curious as to why you say "scipy.linalg.solve(), NOT
> numpy.linalg.solve()". Can you explain the reason for this?

Oh, the performance will be similar, provided you've linked against a good BLAS. 

It's just that the NumPy version doesn't have the sym_pos keyword. As for why that is:

http://ask.scipy.org/en/topic/10-what-s-the-difference-between-numpy-linalg-and-scipy-linalg

David