MATLAB and Simulink resources for Arduino, LEGO, and Raspberry Pi

Learn moreOpportunities for recent engineering grads.

Apply Today**New to MATLAB?**

Asked by Niraj Poudel
on 29 Dec 2012

Hi there,

I have imported data from Excel and now have a 240X3 cell array called 'num'.

Once sorted it looks something like this:

16 1232 56

16 3455 56

16 5443 76

16 1232 88

17 4545 65

17 3432 87

All I want to do is for to write a for loop to select all rows of the data which start with 16's and then save it as another array say num_1, then take another chunk that starts with 17's and save it as num_2..etc.

How can I compose a for loop to take care of that. I have tried different measures based on other questions here in matlab central but to no avail.

Any help will be appreciated.

Sincerely yours; Nj

Answer by Walter Roberson
on 29 Dec 2012

unum = unique(num(:,1)); for K = 1 : length(unum) these_num = num(num == unum(K), :); num_{K} = these_num; end

This will create num_{1}, num_{2} and so on. Creating actual new variables is not a good idea; see http://matlab.wikia.com/wiki/FAQ#How_can_I_create_variables_A1.2C_A2.2C....2CA10_in_a_loop.3F

Niraj Poudel
on 29 Dec 2012

Hi Walter, Thank you for your prompt response. However when I run the code you just sent me, it keeps telling me that the index exceeds matrix dimensions, plus I do not see anywhere in the code where I can specify the rows starting with 16 and 17 and so forth?

Am I missing something? Any help will be appreciated. Nj

Answer by Andrei Bobrov
on 29 Dec 2012

Edited by Andrei Bobrov
on 29 Dec 2012

a0 = {16 1232 56 16 3455 56 16 5443 76 16 1232 88 17 4545 65 17 3432 87}; a = cell2mat(a0);

[b,~,c] = unique(a(a(:,1)>=16,1)); num = ... [num2cell(b),cellfun(@(x)a(x,2:3),accumarray(c,(1:numel(c))',[],@(x){x}),'un',0)];

Niraj Poudel
on 29 Dec 2012

For some reason this does not work either. In this code there does not seem anywhere that rows starting with 16 should be saved as a different array? And 17 and 18 and so forth? I might be wrong. please correct me if I am so. Thank you for your response.

## 0 Comments