[SciPy-User] flattened index for Sparse Matrix?

Lutz Maibaum lutz.maibaum@gmail....
Thu Nov 4 10:39:52 CDT 2010


On Thu, Nov 4, 2010 at 6:44 AM, braingateway <braingateway@gmail.com> wrote:
> Thanks a lot! Then what about the first question? How could I index the
> sparse matrix as flattened version?

If you want to index only the non-zero matrix elements, you can use
the (for CSR matrices) the data array directly. If you want to index
all matrix elements, I can't think of a better way than to convert an
index i to the two-dimensional matrix coordinates:

  a[i / a.shape[1], i % a.shape[1]]

Hope this helps,

  Lutz


More information about the SciPy-User mailing list