说明：  函数实现了lelief算法。（计算样本的每一个特征的分类权重，选出具有最大权重的一组特征。样本的特征就是基因）。 % 对权重较大的一组基因进行两两冗余分析。将相关性弱的和强相关中权重大的基因保留，即为特征基因。 % 已知基因表达谱数据、不同类样本的个数，自行设定分类特征的权重阈值。 % 对原始数据标准化的方法：（x－基因均值）/基因方差。
(Function implements the Lelief algorithm.(Calculate the classification weight of each feature in the sample, and select the group with the largest weight.The characteristics of the sample are the genes). % Perform pairwise redundancy analysis for a group of genes with higher weight.The genes with weak correlation and strong correlation with intermediate weight were retained, namely, the characteristic genes. % the number of known gene expression profile data and samples of different classes, and the weight threshold value of the classification features can be set by oneself. % Method of standardizing raw data :(X-gene mean)/gene variance. % The intra-class and inter-class distance of samples adopts Euclidean distance.)