Averaging in a cell array
2 visualizaciones (últimos 30 días)
Mostrar comentarios más antiguos
avantika
el 11 de Sept. de 2014
Comentada: Andrei Bobrov
el 12 de Sept. de 2014
Hi!
I have a cell array named data (4856x24). The data looks as follows
I want to write a function that averages out the values in each row for positive emotions (Happy,Romantic,Devotional,Calm) and negative emotions(Angry,Longing,Sad,Tensed).The averaged out value column names should be positive and negative respectively. Can anyone pls help. Thanks in advance.
2 comentarios
Adam
el 11 de Sept. de 2014
I'm not sure what you mean by averaging out when the inputs are strings.
I would suggest you use a struct or table (if you are using Matlab 2014) though for your data as having column headings as the first row of a cell array is a lot less easy to work with than having them as fields of a struct or table headings.
Respuesta aceptada
Andrei Bobrov
el 11 de Sept. de 2014
t1 = ~cellfun('isempty',regexp(data(1,:),'Happy|Romantic|Devotional|Calm'));
t2 = ~cellfun('isempty',regexp(data(1,:),'Angry|Longing|Sad|Tensed'));
out = [data(1,:);cell(1,size(data,2))];
out(2,t1) = num2cell(mean(cell2mat(data(2:end,t1))));
out(2,t2) = num2cell(-mean(cell2mat(data(2:end,t2))));
3 comentarios
Andrei Bobrov
el 12 de Sept. de 2014
Hi Avantika! I beg to fulfill the request by Sean, he will show us a more better solution with table array.
Más respuestas (1)
Sean de Wolski
el 11 de Sept. de 2014
Editada: Sean de Wolski
el 11 de Sept. de 2014
If you convert that to a table (cell2table) and covert the feelings to a categorical, you can then use grpstats and varfun to do this for you, grouping by whatever you want.
If you post the mat file (as a zip) I can help you through these steps, it'll be much simpler and easier to understand than regular expressions/cellfun.
0 comentarios
Ver también
Categorías
Más información sobre Numeric Types en Help Center y File Exchange.
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!