Skip to contents

Main functions

The core functions to generate the 2D embeddings

tsne()
Rust-based t-SNE
umap()
Rust-based UMAP
phate()
Rust-based PHATE

kNN helpers

Everything and anything around kNN graph generation

generate_knn_graph()
Generate a k-nearest neighbour graph.
generate_nearest_neigbours_class()
Generate a new NearestNeighbours
get_idx_mat()
Get the indices as a matrix
get_idx_flat()
Get the indices as a flat vector
get_dist_mat()
Get the indices as a matrix
get_dist_flat()
Get the distances as a flat vector

Synthetic data

manifold_synthetic_data()
Generate synthetic data for manifold learning

Params

Wrapper functions around core parameters

params_nn()
Wrapper function to generate nearest neighbour parameters
params_umap()
Wrapper function to generate UMAP parameters
params_tsne()
Wrapper function to generate t-SNE parameters
params_phate()
Wrapper function to generate PHATE parameters

Rust wrappers

Everything rusty - only use this if you know what you are doing…

rs_approx_nearest_neighbours()
Wrapper around some nearest neighbour searches integrated into manifold-rs
rs_data_clusters()
Generates clustered data
rs_data_swiss_role()
Generates the SwissRole data
rs_data_trajectory()
Generates tree-like data with branches
rs_phate()
Run PHATE dimensionality reduction
rs_phate_from_knn()
Run PHATE dimensionality reduction from a precomputed kNN graph
rs_tsne()
tSNE implementation
rs_tsne_from_knn()
tSNE implementation
rs_umap()
UMAP implementation
rs_umap_from_knn()
UMAP implementation
rs_check_cluster_separation()
Check cluster separation in an embedding