通过向量访问具有索引的矩阵 [英] Access to a matrix with indexes from a vector

问题描述

我有一个矩阵a(16x3)和一个向量b(16x1). b显示了矩阵a中每行有多少个有效值.

I have a matrix a(16x3) and a vector b(16x1). b shows how many of the values in the matrix a are valid per row.

a = magic(3)

a =

     8     1     6
     3     5     7
     4     9     2

b = [1;3;2]

b =

     1
     3
     2

我正在尝试将无效值设置为NaN:

What I'm trying to do is setting the invalid values to NaN:

a(:,b+1:end)=NaN

结果是:

a =

     8   NaN   NaN
     3   NaN   NaN
     4   NaN   NaN

但是我期望的是:

a =

 8   NaN   NaN
 3     5     7
 4     9   NaN

这是怎么了?

推荐答案

bsxfun 和@gt创建这些元素的逻辑掩码，然后

Perfect setup for bsxfun with @gt to create a logical mask of those elements and then logically index into a to set them as NaNs -

a(bsxfun(@gt,1:size(a,2),b(:))) = NaN

针对一般m x n案例的示例运行-

Sample run for a generic m x n case -

a =
     2     9     7     2     9
     5     7     2     9     5
     7     5     1     3     1
     8     1     6     2     2
b =
     1
     4
     3
     2
a =
     2   NaN   NaN   NaN   NaN
     5     7     2     9   NaN
     7     5     1   NaN   NaN
     8     1   NaN   NaN   NaN

在这里，logical mask是-

>> bsxfun(@gt,1:size(a,2),b(:))
ans =
     0     1     1     1     1
     0     0     0     0     1
     0     0     0     1     1
     0     0     1     1     1

因此，1s用于选择 to be 设置为NaNs的元素，其余元素被触摸 not to be 或改变了.

So, the 1s were used to select the elements that were to be set as NaNs and rest of the elements were not to be touched or changed.

您也可以使用a(~bsxfun(@le,1:size(a,2),b(:))) = NaN来达到相同的效果.

You can also use a(~bsxfun(@le,1:size(a,2),b(:))) = NaN for the same effect.

这篇关于通过向量访问具有索引的矩阵的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！