Python实现PCA将数据转化成前K个主成分的伪码大致如下: ''' 减去平均数计算协方差矩阵计算协方差矩阵的特征值和特征向量将特征值从大到小排序保留最大的K个特征(Python PCA data into pseudo code before the K principal components are as follows: the characteristics of 'average minus the covariance matrix to calculate the covariance matrix eigenvalues and eigenvectors. The eigenvalues in descending order retain maximum K features)