How to calculate the average with "for loop"

Hello. I'm beginner at Matlab.

I want to calculate the average of DTW distances using rows.

For example, I have two datasets, data11(19200 x 25), data22(19200 x 25).

After computing the DTW distance between two rows, I wanna get the average value of all dtw distances.

Please let me know how to get the average value using "For loop" code.

clear all
data11 = csvread('person01_boxing_d1_uncomp.csv');
data22 = csvread('person01_handclapping_d1_uncomp.csv');
x = data11(1,:);     % 1 ~ 19200
y = data22(19199,:); % 1 ~ 19200
dtw(x,y); 

% I get one dtw distance.
% I want to compute the average value of all kk, kk1, kk3, .....
x = data11(1,:);
y = data22(1,:);
kk = dtw(x,y);

x = data11(1,:);
y = data22(2,:);
kk1 = dtw(x,y);

x = data11(1,:);
y = data22(19199,:);
kk3 = dtw(x,y);

average = mean(kk, kk1, kk3) % ??

Answer 1

[KALYAN ACHARJYA]

How to calculate the average with "for loop"

data11 = csvread('person01_boxing_d1_uncomp.csv');
data22 = csvread('person01_handclapping_d1_uncomp.csv');
x = data11(1,:);     % 1 ~ 19200
kk=zeros(1,size(data22,1)-1);
for i=1:length(kk)
y = data22(i,:);
kk(i)= dtw(x,y);
end
mean(kk)

[Kong]

Thank you so much!

Could I change also the x vector from 1 to 19119?

And I want to compute all dtw' between x and y.

x = data11(i,:); % instand of x = data11(1,:);

[KALYAN ACHARJYA]

Yes

data11 = csvread('person01_boxing_d1_uncomp.csv');
data22 = csvread('person01_handclapping_d1_uncomp.csv');
kk=zeros(1,size(data22,1)-1);
for i=1:length(kk)
x = data11(i,:);    
y = data22(i,:);
kk(i)= dtw(x,y);
end
mean(kk)

In this case it would be like

x = data11(1,:);
y = data22(1,:);
kk = dtw(x,y);
x = data11(2,:);
y = data22(2,:);
kk1 = dtw(x,y);
.......
x = data11(19199,:);
y = data22(19199,:);
kk19199 = dtw(x,y);

Then mean

[Kong]

Thank you so much!

This code will compute the same rows of x and y.

Can I compute all distances?

x = data11(1,:);
y = data22(i,:);  % i = 1 ~ 19200

x = data11(2,:);
y = data22(i,:);  % i = 1 ~ 19200

x = data11(3,:);
y = data22(i,:); % % i = 1 ~ 19200

..............
x = data11(19200,:);
y = data22(i,:); % % i = 1 ~ 19200
    

[KALYAN ACHARJYA]

Yes, in such case you may have to use two nested loop.

[Kong]

Hello. How about this code for all rows?

clear all

data11 = csvread('person01_boxing_d1_uncomp.csv');
data22 = csvread('person01_handclapping_d1_uncomp.csv');
kk=zeros(1,size(data22,1)-1);
for i=1:length(kk)
    x = data11(i,:); 
    for j=1:length(kk)
    y = data22(j,:);
    kk(j)= dtw(x,y);
    end
end
mean(kk)

[KALYAN ACHARJYA]

Seem so

data11 = csvread('person01_boxing_d1_uncomp.csv');
data22 = csvread('person01_handclapping_d1_uncomp.csv');
iter=size(data11,1)-1;
kk=zeros(iter,iter);
for i=1:iter
     x = data11(i,:); 
     for j=1:iter
     y = data22(j,:);
     kk(i,j)=dtw(x,y);
     end
end
mean(kk(:))

You may look for avoid loop (if possible), since it takes a long time due to the large number of repetitions. You may check arrayfun, cellfun or other similar questions in the fourmn. More, suggested to avoid clear all

https://in.mathworks.com/matlabcentral/answers/272064-difference-between-clear-and-clear-all