Specifying the iteration range for 'for' loops
显示 更早的评论
Say I have a matrix of size 5-by-5. I want to sum the elements in the rows, but I don't want to sum from the first column all the time, I want to choose from which column to sum from for any row. So say first row I want to sum 2:5, 2nd row 1:5, 3rd row 4:5, etc. How do I do this?
采纳的回答
There is no need to use a loop:
>> M = randi(9,5,5) % random 5x5 matrix
M =
7 1 5 6 6
7 4 8 2 5
7 4 4 1 7
7 3 7 6 4
8 8 6 9 1
>> V = [2,1,4,4,3]; % first column
>> X = bsxfun(@ge,1:5,V(:))
X =
0 1 1 1 1
1 1 1 1 1
0 0 0 1 1
0 0 0 1 1
0 0 1 1 1
>> sum(M.*X,2)
ans =
18
26
8
10
16
4 个评论
Guillaume
2019-2-1
Since R2016b:
X = (1:size(M, 2)) >= V(:); %no need for bsxfun
Sam Smith
2019-2-1
"How do I perform an operation, say an assignment, where I only want the assignment to take place where the X matrix is a 1. "
Possibly you can just X as an index. If that does not do what you want, please show a simple example with all input and output matrices, and explain in more detail what you are trying to achieve.
Sam Smith
2019-2-1
更多回答（1 个）
Cris LaPierre
2019-2-1
0 个投票
Are those arbiltrary choices of the columns to sum or is there a pattern?
If doing this with a for loop, ideally you are looping through each row, but the exact same code is executed for each row. You'd have to code the pattern into the code that executes. If you can't be more precise on how the columns are selected, I'm afraid you'd either have to preprocess your data by zeroing the data you don't want to sum, or you'd have to manually sum each row.
1 个评论
类别
在 帮助中心 和 File Exchange 中查找有关 Loops and Conditional Statements 的更多信息
Translated by 
