-
Notifications
You must be signed in to change notification settings - Fork 1
/
csum.m
37 lines (34 loc) · 1.05 KB
/
csum.m
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
function s=csum(x)
% CSUM Sum of elements using a compensated summation algorithm
%
% For large vectors, the native sum command in Matlab does not appear to
% use a compensated summation algorithm which can cause significant round
% off errors.
%
% This code implements a variant of Kahan's compensated summation algorithm
% which often takes about twice as long, but produces more accurate sums
% when the number of elements is large.
%
% See also SUM
%
% Example:
% v=rand(1e7,1);
% sum1 = sum(v);
% sum2 = csum(v);
% fprintf('sum1 = %18.16e\nsum2 = %18.16e\n', sum1, sum2);
% David Gleich
% Copyright, Stanford University, 2008-2009
% 2008-05-01: Initial version based on other codes.
% 2009-06-03: Integrated version to work on big and small problems efficiently
s=0; e=0; temp=0; y=0; i=1; numelx= numel(x);
if numelx<2^31
% for loops are a bit faster, when they work
for i=1:numel(x)
temp=s; y=x(i)+e; s=temp+y; e=(temp-s)+y;
end
else
while i<=numelx
temp=s; y=x(i)+e; s=temp+y; e=(temp-s)+y;
i=i+1;
end
end