Matlab单元阵列

Question

Currently I have a 435x1 cell array that was formed with importdata. Each element is a string with 17 values separated by commas.

I want to have the strings fill up the columns of each row, rather than have one string per row contained in one element. So basically, I want a 435x17 cell array where each element is one cell array, rather than a 435x1 cell array with each element is a string.

The closest I have gotten so far is:

data = importdata('my_data.data');
for x=1:size(data,1)
    data(x,:) = {strsplit(data{x}, ',')};
end

This will make the string at each element a cell array. With the above code, I still have a 435x1 cell array, but each element is now a 1x17 cell array.

Can anyone tell me to turn this into a 435x17 cell array? Any help is appreciated.

A row from the data file looks like:

republican,n,y,n,y,y,y,n,n,n,y,?,y,y,y,n,y

There are 435 rows of those strings. I want a 435x17 cell array where the columns are the strings delimited by commas.

Answer 1

You can do this using cat . Here's a smaller example:

data = {{1,2,3};{4,5,6};{7,8,9};{10,11,12}};
size(data)
ans =

   4   1

newData = cat(1,data{:});
size(newData)
ans =

   4   3

EDIT: If you have a cell array of strings instead, you can do the following:

data = {{'123'};{'456'};{'789'};{'abc'}};
size(data)
ans =

   4   1

data2 = cat(1, data{:});
data3 = cell2mat(newData);
size(data3)
ans =

   4   3

data3 =

123
456
789
abc

Answer 2

Try something like this:

temp_data = importdata('my_data.data');
data = cell(size(temp_data,1),17);
for x=1:size(temp_data,1)
    data(x,:) = strsplit(temp_data{x}, ',');
end
clear temp_data;

Answer 3

Do you have any sample data? Perhaps just a few rows of the csv file?

Regardless, if you have R2013b you should be able to do this using a table. For the 'data.csv' file:

republican,n,y,n,y,y,y,n,n,n,y,?,y,y,y,n,y
democrat,y,n,y,n,n,n,y,y,y,n,!,n,n,n,y,n

You can then import this into matlab using a table :

>> t = readtable('data.csv','ReadVariableNames', false)

t = 

        Var1        Var2    Var3    Var4    Var5    Var6    Var7    Var8    Var9    Var10    Var11    Var12    Var13    Var14    Var15    Var16    Var17
    ____________    ____    ____    ____    ____    ____    ____    ____    ____    _____    _____    _____    _____    _____    _____    _____    _____

    'republican'    'n'     'y'     'n'     'y'     'y'     'y'     'n'     'n'     'n'      'y'      '?'      'y'      'y'      'y'      'n'      'y'  
    'democrat'      'y'     'n'     'y'     'n'     'n'     'n'     'y'     'y'     'y'      'n'      '!'      'n'      'n'      'n'      'y'      'n' 

That is how you can easily create a table. I would also suggest looking into just using a table to hold this data instead of processing it as a cell array, as it is made to handle data processing tasks and has features that may help with what you are trying to do. However, if you really need it to be a cell array you can certainly just convert the table to a cell:

>> c = table2cell(t)

c = 

    'republican'    'n'    'y'    'n'    'y'    'y'    'y'    'n'    'n'    'n'    'y'    '?'    'y'    'y'    'y'    'n'    'y'
    'democrat'      'y'    'n'    'y'    'n'    'n'    'n'    'y'    'y'    'y'    'n'    '!'    'n'    'n'    'n'    'y'    'n'

>> size(c)

ans =

     2    17

Hope that helps!

Answer 4

If you do not have R2013b, or do not want to leverage table for some reason, you can also do this like so:

cellData = cell(numel(data), 17);  % preallocate
for idx=1:numel(data)
    cellData(idx,:) = strsplit(data{idx}, ',');
end