我有以下程序：
m = 4;
N = 3;
a = [ones(N+m-1,1)' zeros(m,1)']; % creates an array with specified number of 0's and 1's
b = perms(a);
c = unique(b,'rows'); % with these last two lines, I find all possible combinations
% the rest is probably not relevant for the question
d = [ones(length(c),1) c ones(length(c),1)];
a_powers = zeros(length(d(:,1)),1);
for i = 1:length(d(:,1))
a_powers(i) = nnz(diff(d(i,:))==0);
end
[n_terms,which_power]=hist(a_powers',unique(a_powers'));
但是，当我尝试m = 5和N = 2时，我的计算机内存不足，并出现以下错误：
Out of memory. Type HELP MEMORY for your options.
Error in perms>permsr (line 53)
P = V(P);
Error in perms (line 26)
P = permsr(V);
我以为我可以使用nchoosek（）或combnk（），但是它们并没有满足我的要求（在数组中获得给定数目的1和0的所有可能的不同组合）。
我该怎么做才能优化我的程序？
谢谢！
nchoosek是您要寻找的。对于结果中的每一行，nchoosek将给出其中的列号。该行的其余列将为零。
m = 4;
N = 3;
numberOnes = N+m-1; % This is `k`
patternLength = numberOnes + m; % This is `n`
patternCount = nchoosek(patternLength, numberOnes); % Total number of combinations
bitPatterns = zeros(patternCount, patternLength); % Preallocate output matrix
patterns = nchoosek(1:patternLength, numberOnes); % Gives us the column numbers of the 1's
bitLocations = bsxfun(@(r,c) sub2ind(size(bitPatterns), r, c), % Convert the column numbers in to array indices
[1:patternCount]', patterns); % for each row in `patterns`
bitPatterns(bitLocations) = 1; % Set all of the bitLocations indices to 1
本文收集自互联网，转载请注明来源。
如有侵权，请联系 [email protected] 删除。
我来说两句