Loop index is changed on a for loop

Dear all, I've been trying to debug this issue but I m not able to solve it.

After having calculated the Euclidean distance between points I have to check whether or not it is below a certain threshold. If yes I save some statistics and I go for the next row. As soon as the Euclidean distance is higher than 2 I save in another table (c- index that must be different than i) these statistics per each coloumn in the row c :

1- date time associated to the first distance below to the threshold

2- date time associated to the last distance below to the threshold

3- date time = difference between 1 and 2

4- Mean distance

Of course, the index c increases by one in a different manner than i. I paste below my code.

function [report12,d12] = interactions(T1,T2, j, k, stamp1, stamp2)

T1 = array2table(T1); 
T2 = array2table(T2); 
%stamp1 = array2table(stamp1)
stamp2 = array2table(stamp2)
report12 =table;
d12 = array2table(zeros(size(stamp1)));
i = 1; 
counter = 0; 
n = 0; 
c=1;
starter= 0; 
start_time = NaT; 
end_time = NaT; 
duration = NaT; 
mean_distance = 0;
stop_contact=0;
% reshape array size 
T2(1:k-j,:)=[]; %same dim of T1
stamp2(1:k-j,:)=[];
 
for i= 1:height(T1)
    d12{i,1} = (norm(T1{i,3:end} - T2{i,3:end} ));

    if d12{i,1} <= 2.0  
        starter= starter +1;
        counter = counter +d12{i,1};
        n= n+1; 
        if starter == 1 
            start_time= stamp1(i,1)
        end 
        if starter >1
            end_time = stamp1(i,1) 
            duration = end_time - start_time; 
            mean_distance = counter/n; 
        end
    end
    if (d12{i,1} > 2.0) && (starter ~= 1)
   
        report12{c,1} = start_time;
        report12{c,2} = end_time;
        report12{c,3} = duration;
        report12{c,4} = mean_distance;
        c=c+1;
        counter = 0; 
        n=0; 
        starter=0;
        end_time=NaT;
        start_time=NaT; 
        %%duration= NaT;
        mean_distance = 0;
        %stop_contact = 1;
    end
    i= i+1;
end 

end

[Geoff Hayes]

Andrea - so which variable is causing the problem? I see that you do

i= i+1;

which is unnecessary as the for loop will increment i for you.

[Andrea Sbaragli]

This is my output table report12:

At this point I cannot understand why some row have null values since I fill the table only under this condition, setting starter again equal to 0:

if (d12{i,1} > 2.0) && (starter ~= 1)

report12{c,1} = start_time;

report12{c,2} = end_time;

report12{c,3} = duration;

report12{c,4} = mean_distance;

c=c+1;

counter = 0;

n=0;

starter=0;

end_time=NaT;

start_time=NaT;

%%duration= NaT;

mean_distance = 0;

%stop_contact = 1;

end

Answer 1

[Geoff Hayes]

Andrea - when you populate a row in a table, you reset the start and end times

        end_time=NaT;
        start_time=NaT; 

but what guarantee is there on the next iteration of the loop, that the first condition will be satisfied so that you re-initialize these variables

for i= 1:height(T1)
    d12{i,1} = (norm(T1{i,3:end} - T2{i,3:end} ));
    if d12{i,1} <= 2.0  

If the distance is greater than 2 then you will skip this if statement and then go immediately to the other and populate the row with these NaT values. Try confirming this behaviour with the MATAB Debugger.

[Geoff Hayes]

Maybe you can simplify your code to use report12 array to get the end time for the current iteration. In fact, I think that the starter variable is really only necessary until you have populate the array with at least one row/record.

[Andrea Sbaragli]

No starter is necessary in order to keep track of the starting time connected to a bunch of Euclidean distances below 2.

The problem occurs when I have consecutive distance greater than 2.

[Andrea Sbaragli]

Warning: The assignment added rows to the table, but did not assign values to all of the table's existing variables. Those variables are extended with rows containing

default values.

> In tabular/subsasgnParens (line 432)

In tabular/subsasgnBraces (line 156)

In tabular/subsasgn (line 67)

In interactions (line 42)

[Geoff Hayes]

So how should the code handle the case where there are consecutive distances greater than two?

[Image Analyst]

Andrea, you keep forgetting to attach your data. Make it easy for people to help you, not hard.

Here are the guidelines.

TUTORIAL: How to ask a question (on Answers) and get a fast answer

Please attach your variables in a .mat file so we can run your code.

save('answers.mat', 'T1','T2', 'j', 'k', 'stamp1', 'stamp2');

Then attach answers.mat with the paperclip icon.