在MatLab中迭代来自固定总和的值

jam*_*orm 2 matlab loops combinatorics

假设我有大理石,每种都可以是8种颜色中的一种.我想迭代所有可能的大理石颜色组合值.我怎样才能在MatLab中做到这一点?

例如,如果n = 2,那么我想迭代:

2 0 0 0 0 0 0 0

1 1 0 0 0 0 0 0

1 0 1 0 0 0 0 0

1 0 0 1 0 0 0 0

1 0 0 0 1 0 0 0

1 0 0 0 0 1 0 0

1 0 0 0 0 0 1 0

1 0 0 0 0 0 0 1

0 2 0 0 0 0 0 0

0 1 1 0 0 0 0 0

0 1 0 1 0 0 0 0

等等

编辑:

这是一些伪代码的样子,但你可以看到它非常草率.我对这样做更简洁的方式感兴趣,并且不需要if语句......

for i8 = 0:1:n
    for i7 = 0:1:n - i8
        for i6 = 0:1:n - i8 - i7
            for i5 = 0:1:n - i8 - i7 - i6
                for i4 = 0:1:n - i8 - i7 - i6 - i5
                    for i3 = 0:1:n - i8 - i7 - i6 - i5 - i4
                        for i2 = 0:1:n - i8 - i7 - i6 - i5 - i4 - i3
                            for i1 = 0:1:n - i8 - i7 - i6 - i5 - i4 - i3 - i2
                                if i1 + i2 + i3 + i4 + i5 + i6 + i7 + i8 == n
                                    i = [i1, i2, i3, i4, i5, i6, i7, i8]
                                end
                            end
                        end
                    end
                end
            end
        end
    end
end
Run Code Online (Sandbox Code Playgroud)

Jua*_*nPi 5

这是解决方案(代码在GNU/Octave中测试,它也应该在Matlab中工作).此代码派生自Octave-Forge multinom_exp.m

function conf = marbin (nmar,nbin)
%% Returns the position of nmar in nbis, allowing the marbles to be in the same bin.

%% This is standard stars and bars.
 numsymbols = nbin+nmar-1;
 stars = nchoosek (1:numsymbols, nmar);

%% Star labels minus their consecutive position becomes their index
%% position!
 idx = bsxfun (@minus, stars, [0:nmar-1]);

%% Manipulate indices into the proper shape for accumarray.
 nr = size (idx, 1);
 a = repmat ([1:nr], nmar, 1);
 b = idx';
 conf = [a(:), b(:)];
 conf = accumarray (conf, 1);
Run Code Online (Sandbox Code Playgroud)