evaluateAlgorithmOutput.Rd

% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/evaluate.R
\name{evaluateAlgorithmOutput}
\alias{evaluateAlgorithmOutput}
\title{Evaluate AlgorithmOutput}
\usage{
evaluateAlgorithmOutput(modality, algorithmOutput, validationData,
  maxDifference = 10)
}
\arguments{
\item{modality}{numerical, for which modality output is evaluated, 1 for unimodal,
2 for bimodal,...}

\item{algorithmOutput}{The formatted output of the algorithm that is now
evaluated}

\item{validationData}{the validationData data.frame}

\item{maxDifference}{the maximum percentage difference between the estimated
value and the validation data value with which the estimated is classified
as TRUE (correctly estimated)}
}
\value{
Returns a data.frame where for each as TP classified gene it was
evaluated whether it was estimated approximately correct or not (TRUE or
FALSE)
}
\description{
Evaluate AlgorithmOutput
}
\details{
The estimated values of mean1, mean2, sd1, sd2, proportion1 and proportion2
of each as TP classified gene are compared to those values of the validationData.
If the difference is greater than 10 %, the estimation is classified as
FALSE (not correctly estimated) for this value. If the difference is less
or equal to 10%, the estimation is classified as TRUE (correctly estimated)
}
\examples{
\dontrun{
params <- newParams(nGenes = list("1"=10,"2"=list("gauss"= 10,"gamma"= 10)))
simulation <- simulateExpression(params= params,verbose=FALSE)
expression <- simulation$expressionData
validationData <- simulation$validationData
output <- useMclust(expression, verbose=FALSE)
 # TRUE if less than +- 10 \% difference to value of validationData
evaluation <- evaluateAlgorithmOutput(modality = 2,algorithmOutput = output,
validationData = validationData)
# TRUE if less than +- 20 \% difference to value of validationData
evaluation <- evaluateAlgorithmOutput(modality = 2,algorithmOutput = output,
validationData = validationData,maxDifference = 20)
}
}