FJLT: Fast-Johnson-Lindenstrauss-Transform (FJLT)

In hemberg-lab/FastSC3: Fast Single-Cell Consensus Clustering

Description

This function calculates the FJLT and project onto d dimensions. The FJLT is faster than standard random pro-jections and just as easy to implement. It is based upon the preconditioning of a sparse projection matrix with a randomized Fourier transform.

Usage

fjlt(x, k = 100)

Arguments

x	input expression matrix
k	number of dimension to reduce to

Details

Ailon, N. and Chazelle, B. Approximate nearest neighbors and the fast Johnson-Lindenstrauss transform. in Proceedings of the thirty-eighth annual ACM symposium on Theory of computing 557–563 (ACM, 2006).

Functions adapted from: http://www.cs.ubc.ca/~jaquesn/MachineLearningTheory.pdf

Value

transformed reduced expression matrix

hemberg-lab/FastSC3 documentation built on May 17, 2019, 3:42 p.m.