EhsanKA/sparsesampler: v1.1.0 - EVR-based Automatic Parameter Selection
Abstract
What's Changed EVR-based heuristic for automatic parameter selection: Both the number of principal components (p) and the Bin Resolution Factor (K) are now automatically derived from the explained variance ratio (EVR) using the formula K = 2/EVR_i, with default EVR index i=12. Simplified API: Removed auto_k and random_projection parameters for a cleaner interface. Dataset-specific EVR index recommendations: 7–20 for flow cytometry, 12–25 for scRNA-seq. Updated README: Comprehensive documentation with EVR-based parameter guidance, usage examples, PCA runtime benchmarks, and preprocessing recommendations. Published to PyPI: https://pypi.org/project/sparsesampler/1.1.0/ Full Changelog: https://github.com/EhsanKA/sparsesampler/compare/v1.0.0...v1.1.0
Fields of Study & Subjects
Single-cell and spatial transcriptomics
Face and Expression Recognition
Statistical Methods and Applications
Publication details
- Journal / source
- Zenodo (CERN European Organization for Nuclear Research)
- Publication date
- 7 February 2026