Labeling the columns in an array

If we import a .csv or Excel file using readtable, and then convert the table to an array as the following:

Data=table2array(readtable('Filename.csv'));

We get an output like the following. This syntax process removed all the original labels of each column. Is it possible to label the columns, such as Serial number, Time, Signal 1, Signal 2?

Although one can read and separately isolate the columns, does Matlab allow retaining the labels? Thanks

[John Doe]

Unless you are doing an operation that does not support table data sets, leave your data as a table.

Alternatively, save all your variables to cell array using:

varNames = t.Properties.VariableNames;

Then when you want to output your data can recreate the table afterwards.

Don't split your data in to individual variables.

Answer 1

[Peter Jarosi]

Do not convert your table into an array! Just use table object!

Data=readtable('Filename.csv');

Later, if you want to refer it as a matrix you can use the following syntax:

Data{:,:}

If you want to refer your columns as column vectors use:

Data{:,{'Serial number'}}
Data{:,{'Time'}}
etc.

If you want your signals in a n by two matrix:

Data{:,{'Signal 1','Signal 2'}}

[FW]

Thanks all, I want to apply moving average smoothing on column 3. Couple of other operations don't work in the table after smoothing. Currently, I read the columns from the array. May be it would be worthwhile to keep the table for reference and then use the array for operations.

% Hamming filter

L=13; % size of Hamming windows

H=hamming(L);

Signal1=Data(1:382, 3); data in column in 3

Signal1_Hamming = conv(Signal1, H./sum(H), 'same') ;

[Peter Jarosi]

You're welcome!

You can still use your data in a table format, just change your 3rd line to:

Signal1=Data{1:382, {'Signal 1'}}; % data in column in 3