How to convert a matrix to a structure array?

Question

Hannah 2018-7-18

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/410942-how-to-convert-a-matrix-to-a-structure-array

评论： Hannah 2018-7-19

I am working with data that is in a matrix but would like to easily convert it to a structure array. At the moment I am using the "table" and "table2struct" commands as per the simple example below but I feel that there should be a more efficient method that I'm missing as I will likely have far more than three variables and the current method I'm using seems inefficient. Any assistance would be appreciated!

M=randn(5,3);
T=table(M(:,1),M(:,2),M(:,3),'VariableNames',{'Var1' 'Var2' 'Var3'});
S=table2struct(T);

2 个评论
显示无隐藏无

Steven Lord 2018-7-18

Why do you need to convert the matrix into a struct array? How are you planning to use the fields in the struct later in your code?

Personally most of the time I'd just leave it as M and use indexing to extract the columns as I need them.

Hannah 2018-7-19

I normally work with the data as M as you suggest but I would like to transform the data to a struct array as it is more rigorous in terms ensuring that there is no confusion about which variable is which when making data publically accessible.

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Guillaume 2018-7-19

1
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/410942-how-to-convert-a-matrix-to-a-structure-array#answer_329430

在 MATLAB Online 中打开

I'm with Steven, it would make more sense to keep it as is, or if the columns indeed represent different things, to change it to a table. A simpler version of your conversion would be:

T = array2table(M, 'VariableNames', compose('Var%d', 1:size(M, 2)))

tables are a lot more flexible than structures as you can access data by row or by column. With a structure you're putting the emphasis on the columns at the expense of rows.

If you really want a structure, then, for a matrix with an arbitrary number of column:

S = cell2struct(num2cell(M, 1), compose('Var%d', 1:size(M, 2)), 2)