To get the best overall performance of the PCA algorithm:
Given a symmetric positive-definite matrix X of size p x p, the problem is to compute the Cholesky decomposition X = LLT, where L is a lower triangular matrix.
To get the best overall performance when Cholesky decomposition:
This mode assumes that data set is split in nblocks blocks across computation nodes.
The SVD algorithm in the distributed processing mode has the following parameters:
To get the best overall performance of singular value decomposition (SVD), for input, output, and auxiliary data, use homogeneous numeric tables of the same type as specified in the algorithmFPType class template parameter.