Matlab Tutorials —— Vectorization

One of the most special features that difffers Matlab from other program languages is that Matlab is very well optimized for operation involved vectors and matrices. The problems that are usually done in a for loop in other languages, are likely to be done in a vectorization way in Matlab. Vectorization operations are more readable, less error prone and definitely have better performance. Therefore, vectorizing code is a general programming demand in Matlab.

The first simple example is here:

% loop
tic
i = 0;
for t = 0:0.01:10*pi
    i = i + 1;
    y(i) = sin(t);
end
toc

% vectorization
tic
t = 0:0.01:10*pi;
y = sin(t);
toc

Vectoriazed code is about 5 times faster

Another example:

% loop
tic
x = 1:10000;
ylength = (length(x) - mod(length(x),5))/5;
y(1:ylength) = 0;
for n= 5:5:length(x)
    y(n/5) = sum(x(1:n));
end 
toc

% vectorization
tic
x = 1:10000;
xsums = cumsum(x);  % cumulative summation, xsums = Σx
y = xsums(5:5:length(x)); 
toc

This is 35 times faster

Some other examples:

for n = 1:10000
   V(n) = 1/4 * pi * (D(n)^2) * H(n));
end
% vectorization
V = 1/4 * pi * (D.^2) .* H;

A = [97 89 84; 95 82 92; 64 80 99;76 77 67; 88 59 74; 78 66 87; 55 93 85];

mA = mean(A);  % mean by columns
B = zeros(size(A));
for n = 1:size(A,2)
    B(:,n) = A(:,n) - mA(n);
end
% vectorization
devA = A - mean(A);

% like meshgrid, expand matrices automatically
x = (-2:0.2:2)';  % 21-by-1
y = -1.5:0.2:1.5;  % 1-by-16
F = x.*exp(-x.^2-y.^2);  % 21-by-16

D = [-0.2 1.0 1.5 3.0 -1.0 4.2 3.14]; 
Dpositive = D(D > 0);

if all(D < 0)
    display('All values of diameter are negative.');
end

A = repmat(1:3,5,2);  % repeat 1:3 in 5 * 2 matrix form
B = repmat([1 2; 3 4],2,2);  % repeat [1 2; 3 4] in 2 * 2 matrix form

x = [2 1 2 2 3 1 3 2 1 3];
x = sort(x);
difference  = diff([x, max(x)+1]);
notRepeatingX = x(find(difference));  % return x without redundent elements
count = diff(find([1,difference]));  % count the times of appearance of each element

% count NaN and Inf
count_nans = sum(isnan(x(:)));
count_infs = sum(isinf(x(:)));

Useful appendix:

Function	Description
all	Determine if all array elements are nonzero or true
any	Determine if any array elements are nonzero
cumsum	Cumulative sum
diff	Differences and Approximate Derivatives
find	Find indices and values of nonzero elements
ind2sub	Subscripts from linear index
ipermute	Inverse permute dimensions of N-D array
logical	Convert numeric values to logicals
meshgrid	Rectangular grid in 2-D and 3-D space
ndgrid	Rectangular grid in N-D space
permute	Rearrange dimensions of N-D array
prod	Product of array elements
repmat	Repeat copies of array
reshape	Reshape array
shiftdim	Shift dimensions
sort	Sort array elements
squeeze	Remove singleton dimensions
sub2ind	Convert subscripts to linear indices
sum	Sum of array elements