dist_sne: Calculate similarity matrices for a t-SNE analysis

In opisthokonta/tsnemicrobiota: t-distributed stochastic neighbor embedding for microbiota data

Description

t-SNE uses a similarity measure that can be interpreted as the proability that two points are nearest neighbors. These functions calculate these probabilities.

Usage

dist_sne(xdata, perplexity = 25)

dist_to_sne(dist_, perplexity = 25)

Arguments

xdata	A matrix or data frame.
perplexity	The perplexity. Usually a number between 5 and 50. Default is 25.
dist_	A matrix of distances or dist object.

Details

dist_sne computes the SNE similarities using euclidean distances, as in the originla t-SNE formulation.

dist_to_sne converts an already computed dissimilarity matrix to a matrix of SNE similarities.

Examples

data(iris)
iris_distances <- dist(iris[,1:4], method = 'manhattan')
iris_sne_distances <- dist_to_sne(iris_distances)

opisthokonta/tsnemicrobiota documentation built on June 1, 2019, 9:44 a.m.