gaussianKernel: Computes gaussian kernel matrix

In scAlign: An alignment and integration method for single cell genomics

Description

Tensorflow implementation of tSNE's gaussian kernel.

Usage

gaussianKernel(data, data_shape, labels = NULL, method = NULL,
  perplexity = 30, diag = "zero")

Arguments

data	cell x feature data matrix
data_shape	number of features for data
labels	cell x 1 annotation (label) vector
method	Kernel to compute pairwise cell similarity
perplexity	neighborhood parameter for gaussian kernel
diag	indicator for self similarity

Value

Tensorflow op

Examples

## Input data, 100 cells x 10 features
data = matrix(sample.int(1000, 100*10, TRUE), 100, 10)
rownames(data) = paste0("cell", seq_len(100))
colnames(data) = paste0("gene", seq_len(10))

result = gaussianKernel(data, nrow(data))

scAlign documentation built on April 28, 2020, 6:10 p.m.