tsneReductor: t-Distributed Stochastic Neighbor Embedding (t-SNE)

In Evacluster: Evaluation Clustering Methods for Disease Subtypes Diagnosis

t-Distributed Stochastic Neighbor Embedding (t-SNE)

Description

This method is an unsupervised, non-linear technique used for data exploration and visualizing high-dimensional data.This function constructs a low-dimensional embedding of high-dimensional data, distances, or similarities.

Usage

tsneReductor(data = NULL, dim = 2, perplexity = 30, max_iter = 500)

Arguments

data	Data matrix (each row is an observation, each column is a variable)
dim	Integer number; Output dimensional (default=2)
perplexity	numeric; Perplexity parameter (should not be bigger than 3 * perplexity < nrow(X) - 1, default=30)
max_iter	Integer; Number of iterations (default: 500)

Value

tsneY: A Matrix containing the new representations for the observation with selected dimensions by user

Examples

library("mlbench")
data(Sonar)

rndSamples <- sample(nrow(Sonar),150)
trainData <- Sonar[rndSamples,]
testData <- Sonar[-rndSamples,]

tsne_trainData <- tsneReductor(trainData[,1:60],dim = 3,perplexity = 10,max_iter = 1000)

Evacluster documentation built on April 1, 2022, 9:07 a.m.