求大神帮我看看这个GMM代码！

我是MATLAB新手，在论坛上找了一个GMM的代码，但是我放到matlab里面显示“此上下文中不支持函数定义。请在代码文件中创建函数”，请问我需要创建什么样的函数？实在是搞不懂该怎么弄了，求各位大神帮忙看一看！谢谢！！

这是那位大佬的高斯混合模型的代码，我现在有一个1×365的excel数据，我想要得到模型的估计参数值，谢谢大家了！

 

function varargout = gmm(X, K_or_centroids)

% ============================================================

%转载自http://blog.pluskid.org/?p=39

% Expectation-Maximization iteration implementation of

% Gaussian Mixture Model.

%

% PX = GMM(X, K_OR_CENTROIDS)

% [PX MODEL] = GMM(X, K_OR_CENTROIDS)

%

%  - X: N-by-D data matrix.%需要注意的是这里的X包括了全部

%  - K_OR_CENTROIDS: either K indicating the number of

%       components or a K-by-D matrix indicating the

%       choosing of the initial K centroids.

%

%  - PX: N-by-K matrix indicating the probability of each

%       component generating each point.

%  - MODEL: a structure containing the parameters for a GMM:

%       MODEL.Miu: a K-by-D matrix.

%       MODEL.Sigma: a D-by-D-by-K matrix.

%       MODEL.Pi: a 1-by-K vector.

% ============================================================
    threshold = 1e-15;

    [N, D] = size(X);

 

    if
isscalar(K_or_centroids)

        K = K_or_centroids;

        % randomly pick centroids

        rndp = randperm(N);

        centroids = X(rndp(1:K),:);

    else

        K = size(K_or_centroids, 1);

        centroids = K_or_centroids;

    end

 

    % initial values

    [pMiu pPi pSigma] = init_params();

 

    Lprev = -inf;

    while
true

        Px = calc_prob();%计算N(x|mu,sigma)

 

        % new value for pGamma

        pGamma = Px .* repmat(pPi, N, 1);%估计 gamma 是个N*K的矩阵

        pGamma = pGamma ./ repmat(sum(pGamma, 2), 1, K);%对矩阵的理解真是出神入化,

 

        % new value for parameters of each Component

        Nk = sum(pGamma, 1);%N_K

        pMiu = diag(1./Nk) * pGamma' * X; % 数字 *( K-by-N * N-by-D)加个括号有助理解

        pPi = Nk/N;

        for
kk = 1:K

            Xshift = X-repmat(pMiu(kk, : ), N, 1);%x-u

            pSigma(:, :, kk) = (Xshift' * ...

                (diag(pGamma(:, kk)) * Xshift)) / Nk(kk);%更新sigma

        end

 

        % check for convergence

        L = sum(log(Px*pPi'));

        if
L-Lprev < threshold

            break;

        end

        Lprev = L;

    end

 

    if
nargout == 1

        varargout = {Px};

    else

        model = [];

        model.Miu = pMiu;

        model.Sigma = pSigma;

        model.Pi = pPi;

        varargout = {pGamma, model};%注意！！！！！这里和大神代码不同，他返回的是px，而我是 pGamma

    end

 

    function
[pMiu pPi pSigma] = init_params()%初始化参数

        pMiu = centroids;% K-by-D matrix

        pPi = zeros(1, K);%1-by-K matrix

        pSigma = zeros(D, D, K);%

 

        % hard assign x to each centroids

        distmat = repmat(sum(X.*X, 2), 1, K) + ... % X is a N-by-D data matrix.

            repmat(sum(pMiu.*pMiu, 2)', N, 1) - ...% X->K列 U->N行 XU^T is N-by-K

            2*X*pMiu';%计算每个点到K个中心的距离

        [~, labels] = min(distmat, [], 2);%找到离X最近的pMiu，[C,I] labels代表这个最小值是从那列选出来的

 

        for
k=1:K

            Xk = X(labels == k, : );% Xk是所有被归到K类的X向量构成的矩阵

            pPi(k) = size(Xk, 1)/N;% 数一数几个归到K类的

            pSigma(:, :, k) = cov(Xk); %计算协方差矩阵，D-by-D matrix,最小方差无偏估计

        end

    end

 

    function
Px = calc_prob()

        Px = zeros(N, K);

        for
k = 1:K

            Xshift = X-repmat(pMiu(k, : ), N, 1);%x-u

            lemda=1e-5;

            conv=pSigma(:, :, k)+lemda*diag(diag(ones(D)));%这里处理singular问题，为协方差矩阵加上一个很小lemda*I

            inv_pSigma = inv(conv);%协方差的逆

            tmp = sum((Xshift*inv_pSigma) .* Xshift, 2);%(X-U_k)sigma.*(X-U_k),tmp是个N*1的向量

            coef = (2*pi)^(-D/2) * sqrt(det(inv_pSigma));%前面的参数

            Px(:, k) = coef * exp(-0.5*tmp);%把数据点 x 带入到 Gaussian model 里得到的值

        end

    end

end