Sparse Array - Special Command?

1 次查看（过去 30 天）

显示更早的评论

Martin 2013-11-12

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/105762-sparse-array-special-command

评论： Martin 2013-11-18

采纳的回答： Walter Roberson

在 MATLAB Online 中打开

Hi,

I'm dealing with a sparse array. Given 2 arrays "col" and "val", I want to create a one-rowed sparse array of value val(i) in the column col(i), for all i.

What I did is:

if true
% val, col given
leng = length(val);
row = ones(1, leng);
result = sparse(row, col, val, 1, leng);
end

Is there a better way to do, without creating an array "row" in particular? Because getting the length and creating the row take a bit of time in my problem.

Thanks!

1 个评论
显示 -1更早的评论隐藏 -1更早的评论

Martin 2013-11-12

No ideas? :)

请先登录，再进行评论。

请先登录，再回答此问题。

采纳的回答

Walter Roberson 2013-11-13

1
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/105762-sparse-array-special-command#answer_115053

在 MATLAB Online 中打开

result = sparse(1, col, val);

Note: this is not exactly equivalent to what you wrote. What you wrote only allocates a 1 x leng array, but leng is the number of elements in col, not the maximum col.

For example, sparse(1, 10, 3, 1, 1) would try to create a 1 x 1 sparse matrix, but it needs a 1 x 10 sparse matrix.

4 个评论
显示 2更早的评论隐藏 2更早的评论

Walter Roberson 2013-11-15

Much faster or just "faster" ?

Martin 2013-11-18

In my program, the time necessary to create the matrix itself is similar, but I saved the time to create the "row" array. In particular, calculating the length of this array took time.

I'd say I saved about 25% of time but I reckon this is very problem-specific.

请先登录，再进行评论。