数据降维的方法: PCA 核PCA 奇异值分解SVD SVD 应用场景:数据降维,图像压缩 from scripy.linalg import svd U,S,V = svd(x_s,full_matrices = False) # 选择最前2个奇异值来近似原始的矩阵 x_t = U[:,:2] https://www.cnblogs.com/endlesscoding/p/10033527.html