Reshape large sparse matrices

7 次查看（过去 30 天）
Matteo
Matteo 2011-7-6
Hi, I need to reshape large sparse matrices into vectors by concatenating the rows. Large in my case means square matrices up to Nrows=20000, the density of such matrices is very small, with only about Nrows non zero elements. However, when I try to use reshape to build the corresponding vector (predefined to be sparse also), I don't get any error but the calculation take very long, which seems to me strange since in fact only a few values are non-zero. The only explanation I have for this is that reshape does not treat sparse matrices as sparse. So, is there any way to do the same faster? I have a workaround but it is not so fast and more importantly it is not at all elegant.
  1 个评论
Jan
Jan 2011-7-6
Please post a code snippet, which reproduces the problem.

请先登录，再进行评论。

请先登录，再回答此问题。

回答（1 个）

the cyclist
the cyclist 2011-7-6
The following code, which reshapes a large sparse array, is very fast on my machine:
N = 20000;
s = speye(N);
rs = reshape(s,[N*N,1]);
I second Jan's suggestion to supply a code snippet, if you can.

类别

在 Help Center 和 File Exchange 中查找有关 Creating and Concatenating Matrices 的更多信息

标签

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by