support_vector_machines: Support vector machine (SVM) learning on toy datasets

In wikiselev/clustools: Tools for cluster analysis of single cell high throughput data

Description

Perform SVM learning on either quake, sandberg or linnarsson data sets.

Usage

support_vector_machines(dataset, teach.proportion, kern)

Arguments

dataset	Name of the toy dataset
teach.proportion	Proportion of cells used for teaching, the rest of the cells are used for studying
kern	The kernel used in training and predicting: linear, polynomal, radial, sigmoid. See ?svm from "clue" package for more information.

Value

#'

pred

Predicted cell labels for cells used in study

ari

Adjusted Rand Index - comparison of pred wiht gold standard

model

model parameters used in SVM

training

Training dataset

Examples

res <- support_vector_machines("sandberg", 0.7)

wikiselev/clustools documentation built on May 4, 2019, 5:25 a.m.