How to concatenate vectors of 10x2 and 11x1 in size?

1 次查看（过去 30 天）

Johannes Sundqvist 2020-2-5

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/503874-how-to-concatenate-vectors-of-10x2-and-11x1-in-size

评论： Steven Lord 2020-2-6

Hi,

I'm currently working on a project, where the original data consists of a vector that is 10x2, and another vector that is 11x1 in size. I'm now wondering if there is a "realistic" way to concatenate these two vectors (for example, insert the data from one vector from one position, insert the data from the second vector from another position, then once it has to be used, define which position we want to work "from"? If that makes any sense?) - Or if the only way to get it to work without doing a huge work-around would be to hold them in two different variables?

I realise this might be a bit messy as far as the explanation goes, so feel free to let me know if anything needs further explanation!

Thanks in advance.

5 个评论
显示 3更早的评论隐藏 3更早的评论

Steven Lord 2020-2-5

编辑：Steven Lord 2020-2-5

在 MATLAB Online 中打开

As you've written it in the comment above, z is not a valid array in MATLAB.

>> z = [1 2; 3 4; 5 6; 7 8; 9 10; 11 12 13 14 15 16]
Error using vertcat
Dimensions of arrays being concatenated are not consistent.

Don't worry about what commands you'd need to write to create your desired z, type out what you want explicitly. Once we understand what you want we can determine how to build what you want.

One option would be to drop the extra elements of y.

x = [1 2; 3 4; 5 6; 7 8; 9 10]
y = (11:16).'
z = [1 2 11; 3 4 12; 5 6 13; 7 8 14; 9 10 15] % 16 dropped

Another would be to pad x with an extra row of NaN values to match the height of y.

x = [1 2; 3 4; 5 6; 7 8; 9 10]
y = (11:16).'
z = [1 2 11; 3 4 12; 5 6 13; 7 8 14; 9 10 15; NaN NaN 16] % extra row added

You could pad y with NaN values then reshape the padded array so it has the same number of rows as x.

x = [1 2; 3 4; 5 6; 7 8; 9 10]
y = (11:16).'
z = [1 2 11 16; 3 4 12 NaN; 5 6 13 NaN; 7 8 14 NaN; 9 10 15 NaN]

You mention 'how we separately could select the x and y "subsets"' -- do you have fixed x and y data and you want to find some x that "contains" them both, or do you have a desired z array and you want to figure out which x and y arrays you'd need to start with to somehow create z?

Johannes Sundqvist 2020-2-6

I've thought about the issue, looked at what you suggested and have been experimenting around a little bit (though not with much success), but I'm starting to realise a lot more than I initially thought will probably have to be changed ("down the line") if these vectors are concatenated, and as such I might put the whole thing on ice for right now. Either way, it's not like you haven't helped/made me learn new things, so thanks a bunch!