
Run PHATE dimensionality reduction from a precomputed kNN graph
Source:R/extendr-wrappers.R
rs_phate_from_knn.Rd
Leverages the PHATE implementation in manifolds-rs - a very fast Rust-based
implementation. This version uses a pre-computed kNN graph, please see
new_nearest_neighbour().
Arguments
- embd
Numerical matrix. The data to embed of shape samples x features.
- knn_data
NearestNeighboursclass from R.- n_dim
Integer. Number of PHATE dimensions to return. Currently only
2Lis supported.- k
Integer. Number of nearest neighbours that set the kernel bandwidth. Every neighbour in
knn_datacan join the kernel while its affinity clearsthresh, so build it6 * k + 5wide to match the internal search.- phate_params
Named list. Contains all key parameters for PHATE, see
params_phate()andparams_nn().- seed
Integer. Seed for reproducibility.
- verbose
Integer. If
0L-> silent or1Lfor normal verbosity;2Lfor detailed verbosity.