How to convert cell array into double in table

Hi all, I've a table of the format

   section                   D1      D2        L      sectionID
    ____________________    ____    _____    _____    _________

    1.00000000000000e+00    [ 5]    [  5]    [  6]    [11]     
    2.00000000000000e+00    [12]    [  2]    [ 34]    [11]     
    3.00000000000000e+00    [12]    [312]    [323]    [11]     
    4.00000000000000e+00    [ 3]    [ 45]    [ 23]    [11]     
    5.00000000000000e+00    [ 4]    [  4]    [ 23]    [11]     
    6.00000000000000e+00    [ 2]    [ 43]    [ 23]    [11]     
    7.00000000000000e+00    [ 3]    [  4]    [ 23]    [11]     
    8.00000000000000e+00    [ 4]    [ 23]    [ 32]    [11]

Where Section is a double and rest D1,D2,L and section ID are cell arrays. I want to convert all these too into double arrays. I tried to do using cell2mat

temp_table.D1 = cell2mat(temp_table.D1);

                    temp_table.D2 = cell2mat(temp_table.D2);
                    temp_table.L = cell2mat(temp_table.L);
                    temp_table = cell2mat(temp_table.sectionID);

When I try to use the cell2mat in the above way, I get the following error

Field assignment to a non-structure array object.

    Struct contents reference from a non-struct array object.

    Error in sample_gui_ver2/tb9_Callback (line 1918)
                      temp_table.D1 = cell2mat(temp_table.D1);

    Error while evaluating UIControl Callback

Can anyone help me to fix this issue. Thanks in advance.

Answer 1

[Peter Perkins]

This whole thread begs the question of how the table got like that to begin with, which is worth you figuring out, probably. But I think what you're looking for is

>> t = table([1;2;3;4],{5;6;7;8},{9;10;11;12},[13;14;15;16])
t =
  4×4 table
    Var1    Var2    Var3    Var4
    ____    ____    ____    ____
     1      [5]    [ 9]     13 
     2      [6]    [10]     14 
     3      [7]    [11]     15 
     4      [8]    [12]     16 
>> for i = 1:width(t), if iscell(t.(i)), t.(i) = cell2mat(t.(i)); end, end
>> t
t =
  4×4 table
    Var1    Var2    Var3    Var4
    ____    ____    ____    ____
     1       5        9      13 
     2       6       10      14 
     3       7       11      15 
     4       8       12      16

Alternatively, if all the vars in thetable are cells, use varfun:

>> t = table({1;2;3;4},{5;6;7;8},{9;10;11;12},{13;14;15;16})
t =
  4×4 table
    Var1    Var2    Var3    Var4
    ____    ____    ____    ____
    [1]    [5]    [ 9]    [13]
    [2]    [6]    [10]    [14]
    [3]    [7]    [11]    [15]
    [4]    [8]    [12]    [16]
>> t = varfun(@cell2mat,t)
t =
  4×4 table
    cell2mat_Var1    cell2mat_Var2    cell2mat_Var3    cell2mat_Var4
    _____________    _____________    _____________    _____________
          1                5                9               13      
          2                6               10               14      
          3                7               11               15      
          4                8               12               16

and then patch up the names. varfun is a bit less convenient when only some vars are cells, you'd end up having to interleave two pieces back together.

Answer 2

[Will Wilson]

Try using str2double to get the job done. Something like this:

temp_table.D1 = str2double (temp_table.D1);

[Walter Roberson]

There is no evidence that D1 is a string.

Answer 3

[Walter Roberson]

Notice that the error is

Field assignment to a non-structure array object.

The problem is the destination being assigned to, not the source. At some line above what you show, you have assigned a non-structure to temp_table, such as a numeric array, and then the assignment to temp_table.D1 is trying to treat the existing object as if it were a structure. Instead of just erasing all of the existing temp_table and creating a new structure, MATLAB complains that it is not already a structure.

The problem would not occur if temp_table were a structure (or table), or if temp_table did not exist at that point.

Answer 4

[Azzi Abdelmalek]

Probably one of the columns is not a cell array. Check if all columns are cell array