The CPP implementation of RPCA (Robust Principal Component Analysis) based on Armadillo, Truncated SVD, and Eigen libraries.
-
- The RPCA function based on Armaidillo is not very stable, where the program may not converge in some certain cases.
-
- The RPCA function based on Truncated SVD has a risk of out of memory (OOM) when is called many times, since the created matrices are not released properly.