getMetrics: Compute the machine learning evaluation metrics

In transbioZI/BioMMex: BioMM: Biological-informed Multi-stage Machine learning framework for phenotype prediction using omics data

Compute the machine learning evaluation metrics

Description

Compute the evaluation metrics in the classification setting: area under curve (AUC), the area under the Precision-Recall curve, classification accuracy (ACC) and the pseudo R square (R2).

Usage

getMetrics(dataY, predY)

Arguments

dataY	The observed outcome.
predY	The predicted outcome.

Details

If all samples are predicted into one class, then R2 is 0.

Value

A set of metrics for model evaluation: AUC, AUCPR, ACC and R2.

Author(s)

Junfang Chen

Examples

 
## Load data  
methylfile <- system.file('extdata', 'methylData.rds', package='BioMM')  
methylData <- readRDS(methylfile)   
dataY <- methylData[,1]
methylSub <- data.frame(label=dataY, methylData[,c(2:1001)])  
library(ranger) 
library(precrec)
library(rms)
library(BiocParallel) 
param1 <- MulticoreParam(workers = 1) 
param2 <- MulticoreParam(workers = 10)  
predY <- predByCV(methylSub, repeats=1, nfolds=10,   
                  FSmethod=NULL, cutP=0.1, 
                  fdr=NULL, FScore=param1, 
                  classifier='randForest',
                  predMode='classification', 
                  paramlist=list(ntree=300, nthreads=20),
                  innerCore=param2)   
metrics <- getMetrics(dataY=dataY, predY=predY)
print(metrics)  

transbioZI/BioMMex documentation built on Jan. 27, 2023, 4:14 a.m.