How to count sum for values corresponding to repeated numbers in matrixes.

Dear experts,

I have a 2 column ascii file with numbers as it looks at the first 2 columns from the left.

Shortly:

EventID   Hits   Sum
----------------------------
9170    1        1
9443    2       14
9443    4       14
9443    8       14
15872    4      12
15872    8      12
16077    4      12
16077    8      12
22374    1       1
23801    4      12

How I could get the 3rd column as a product? For example, the number 9443 is repeated 3 times and I would like to place the sum 2+4+8=14 on the place fo 9443 etc

I tried by counting unique numbers like:

fid = fopen('200Mevents_file4_geant.txt', 'r');
data = fscanf(fid, '%f');
nRows = data(1);
data = reshape(data(1:end), 106695, 1).';
c = unique(data);
for i = 1:length(data)
   counts(i,1) = sum(data==data(i)); % number of times each unique value is repeated
 end

Could you help me?

Georgios

Answer 1

[KSSV]

clc; clear all ;
data = [9170    1        1
9443    2       14
9443    4       14
9443    8       14
15872    4      12
15872    8      12
16077    4      12
16077    8      12
22374    1       1
23801    4      12] ; 

[c,ia,ib] = unique(data(:,1)) ; 
n = length(c) ; 
iwant = zeros(n,3) ;
for i = 1:n
    iwant(i,:) = [c(i) sum(data(ib==i,2)) data(ia(i),3)] ;
end

[Georgios Tsiledakis]

Thanks so much! It works at first sight calculating the sums BUT it does not reproduce the 3rd column.

I used it as follows:

load uprm.dat

[c,ia,ib] = unique(uprm(:,1)) ;

n = length(c) ;

iwant = zeros(n,1) ;

for i = 1:n

%for i = 1:length(uprm)

iwant(i) = sum(uprm(ib==i,2)) ;

end

where uprm.dat is the two column file:

9170 1

9443 2

9443 4

9443 8

15872 4

15872 8

16077 4

16077 8

22374 1

23801 4

23801 8

23804 2

31177 2

43616 1

43616 2

43616 4

43616 8

54402 2

54402 4

54402 8

62843 1

62843 2

66301 2

66301 4

66301 8

73971 1

I would like to have a same dimension product as a result:

1

14

14

14

12

12

12

12

1

12

12

2

2

15

15

15

15

14

14

14

3

3

14

14

14

1

Any help is appreciated.

[KSSV]

Edited the code....

[Georgios Tsiledakis]

Hi,

Sorry for botherig again but unfortunately it is not correct yet.

I run the updated code and the product is iwant is 6x3 instead of 10x1.

9170 1 1

9443 14 14

15872 12 12

16077 12 12

22374 1 1

23801 4 12

I would like to have as a result the 3rd colum of the 10x3 array.

Thus,

1

14

14

14

12

12

12

12

1

12

12

2

I also forgot 2 more elements in the 3x12 array.

Here is the right one:

9170 1 1

9443 2 14

9443 4 14

9443 8 14

15872 4 12

15872 8 12

16077 4 12

16077 8 12

22374 1 1

23801 4 12

23801 8 12

23804 2 2

Answer 2

[Georgios Tsiledakis]

Hello,

I manage to get what I wanted thanks to your help.

I provide the code below.

load data.dat

[c,ia,ib] = unique(data(:,1)) ;

n = length(c) ;

n2 = length(data) ;

iwant = zeros(n,1) ;

iwant2 = zeros(n,1) ;

iwant3 = zeros(n,1) ;

%%%%%%%%%%%%%%%%%%%%%%%%

for i = 1:n

iwant(i) = sum(data(ib==i,2)) ;

end

%%%%%%%%%%%%%%%%%%%%%%%

for i = 1:length(data)

counts(i,1) = sum(ib==ib(i));

end

%%%%%%%%%%%%%%%%%%%%%%%

for i = 1:n2

iwant2(i) = sum(data(ib==ib(i),2)) ;

end

%%%%%%%%%%%%%%%%%%%%%%%%

for i = 1:n2

iwant3(i) = sum(data(ib==i,2)) ;

end