PerformCV.explore: Perform Monte-Carlo Cross Validation (MCCV)

In xia-lab/MetaboAnalystR: An R Package for Comprehensive Analysis of Metabolomics Data

Perform Monte-Carlo Cross Validation (MCCV)

Description

Classification MCCV, aims to find the best feature subsets using default model parameters

Usage

PerformCV.explore(mSetObj, cls.method, rank.method="auroc", lvNum=2, propTraining=2/3)

Arguments

mSetObj	Input the name of the created mSetObj (see InitDataObjects)
cls.method	Select the classification method, "rf" for random forest classification, "pls" for PLS-DA, and "svm" for support vector machine
rank.method	Select the ranking method, "rf" for random forest mean decrease accuracy, "fisher" for Fisher's univariate ranking based on area under the curve "auroc" for univariate ranking based on area under the curve, "tt" for T-test univariate ranking based on area under the curve, "pls" for partial least squares, and "svm" for support vector machine
lvNum	Input the number of latent variables to include in the analyis, only for PLS-DA classification
propTraining	Input the proportion of samples to use for training

Author(s)

Jeff Xia [e-mail] McGill University, Canada License: GNU GPL (>= 2)

xia-lab/MetaboAnalystR documentation built on Dec. 23, 2024, 3:44 p.m.