Prev: Installing SimPowerSystems
Next: fminsearch
From: Shar Teel on 25 Jun 2010 10:04
Dear all,
I am developing a test framework for categorization. It includes automatic checking for the optimal parameters. This means, I generate a large number of variations. I want to process a single matrix with all possible variations without using a for loop.

I have a matrix, representing movements of a human face across time. Each column corresponds to a facial articulator. The rows represent the locations of those across time:

% head eyes mouth
% 1 2 4
% 1 3 5

To discretize each column, I apply a filter with 5 distinct values. So, each articulator has its own unique set ot 5 values. I want to apply each set of 5 values to its corresponding column.

Question 1: is there any way to avoid the for loop here? Right now it looks awkward:
----------------
for i=1:size(matrix,2),
matrix(:,i) = simplifySingle(matrix(:,i),incrBound(i),...
decrBound(i), numFirstFrames(i), numLastFrames(i), smoothSpan(i));
end
----------------
where simplifySingle is the actual function with a singleton parameter.

Thank you in advance for your consideration
 | 
Pages: 1
Prev: Installing SimPowerSystems
Next: fminsearch