Not possible to one hot encode in a timetable column

I am trying to do a simple one hot encoding in one of the column of a time table but it's becoming an impossible task.

I have tried to convert that column to a table and even to an array buty nothing works.

months = timetable2table(df_tmp(:, {'month'}), 'ConvertRowTimes', false);
nehotencode(months);

The error states: "First argument must contain a single label for each observation."

It shouldn't be that difficult but I do not manage to convert a single column to a one hot enconding.

Thanks in advance,

Rafa

Answer 1

[Ayush Modi]

Hi Rafael,

I am assuming you are referring to onehotencode function in MATLAB.

The error is because onehotencode function expects a table of categorical data. Otherwise, a class for each variable needs to be provided to the function.

Here is the example code to demonstrate how to achieve the same with the categorical data:

% I am using sample data for the example.
dt = datetime(2021, 1, 1) + calmonths(0:11); % Monthly data for one year
months = {'Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun', 'Jul', 'Aug', 'Sep', 'Oct', 'Nov', 'Dec'};
df_table = timetable(dt', months', 'VariableNames', {'Months'});

disp(df_table);

       Time        Months 
    ___________    _______

    01-Jan-2021    {'Jan'}
    01-Feb-2021    {'Feb'}
    01-Mar-2021    {'Mar'}
    01-Apr-2021    {'Apr'}
    01-May-2021    {'May'}
    01-Jun-2021    {'Jun'}
    01-Jul-2021    {'Jul'}
    01-Aug-2021    {'Aug'}
    01-Sep-2021    {'Sep'}
    01-Oct-2021    {'Oct'}
    01-Nov-2021    {'Nov'}
    01-Dec-2021    {'Dec'}

% Converting the data into categorical data
categoricalData = categorical(df_table.Months)

categoricalData = 12x1 categorical array
     Jan 
     Feb 
     Mar 
     Apr 
     May 
     Jun 
     Jul 
     Aug 
     Sep 
     Oct 
     Nov 
     Dec 

% Converting to a table
tblA = table(categoricalData)

tblA = 12x1 table
    categoricalData
    _______________

          Jan      
          Feb      
          Mar      
          Apr      
          May      
          Jun      
          Jul      
          Aug      
          Sep      
          Oct      
          Nov      
          Dec      

encodedData = onehotencode(tblA);
disp(encodedData);

    Apr    Aug    Dec    Feb    Jan    Jul    Jun    Mar    May    Nov    Oct    Sep
    ___    ___    ___    ___    ___    ___    ___    ___    ___    ___    ___    ___

     0      0      0      0      1      0      0      0      0      0      0      0 
     0      0      0      1      0      0      0      0      0      0      0      0 
     0      0      0      0      0      0      0      1      0      0      0      0 
     1      0      0      0      0      0      0      0      0      0      0      0 
     0      0      0      0      0      0      0      0      1      0      0      0 
     0      0      0      0      0      0      1      0      0      0      0      0 
     0      0      0      0      0      1      0      0      0      0      0      0 
     0      1      0      0      0      0      0      0      0      0      0      0 
     0      0      0      0      0      0      0      0      0      0      0      1 
     0      0      0      0      0      0      0      0      0      0      1      0 
     0      0      0      0      0      0      0      0      0      1      0      0 
     0      0      1      0      0      0      0      0      0      0      0      0 

Refer to the following MathWorks documentation for more information on:

• categorical arrays - https://www.mathworks.com/help/matlab/categorical-arrays.html
• onehotencode - https://www.mathworks.com/help/deeplearning/ref/onehotencode.html#mw_1f9b6836-0899-493e-82f8-476ba325a92e

[Rafael]

Thanks a lot Ayush!

That solved the problem!