Is your feature request related to a problem? Please describe.
I often need to extract dominant patterns of variability from data sets. However, standard PCA is hard to interpret due to its dense solutions. Rotated PCA has been used traditionally but is ad-hoc. Modern methods add penalties or constraints (L0 or L1 norms) to induce sparsity in the components, improving interpretability. Sparse PCA by Erichson et al. (2020) is one such approach, but there are many others.
Describe the solution you'd like
Include Sparse PCA via regularized SVD (sPCA-rSVD) by Shen and Huang (2008)
Describe alternatives you've considered
Many other approaches exist, but they are often computationally intensive.
Is your feature request related to a problem? Please describe. I often need to extract dominant patterns of variability from data sets. However, standard PCA is hard to interpret due to its dense solutions. Rotated PCA has been used traditionally but is ad-hoc. Modern methods add penalties or constraints (L0 or L1 norms) to induce sparsity in the components, improving interpretability. Sparse PCA by Erichson et al. (2020) is one such approach, but there are many others.
Describe the solution you'd like Include Sparse PCA via regularized SVD (sPCA-rSVD) by Shen and Huang (2008)
Describe alternatives you've considered Many other approaches exist, but they are often computationally intensive.
Additional context None.