If ROCR package is not installed on your machine, please install it before running the following functions.
Parameters
Parameters
- pred - Predicted probability column of event (interesting class)
- depvar - Dependent / Target / Outcome Variable
Training Dataset Prediction Column
rfPred <- predict(tuned, type ="prob")
Validation Dataset Prediction Column
rfPred <- predict(tuned, val, type ="prob")R Functions for AUC and KS Statistics
AUC <- function(pred,depvar){
require("ROCR")
p <- prediction(as.numeric(pred),depvar)
auc <- performance(p,"auc")
auc <- unlist(slot(auc, "y.values"))
return(auc)
}
KS <- function(pred,depvar){
require("ROCR")
p <- prediction(as.numeric(pred),depvar)
perf <- performance(p, "tpr", "fpr")
ks <- max(attr(perf, "y.values")[[1]] - (attr(perf, "x.values")[[1]]))
return(ks)
}
KS(rfPred[,2], dev[,1])
AUC(rfPred[,2], dev[,1])
About Author:
Deepanshu founded ListenData with a simple objective - Make analytics easy to understand and follow. He has over 10 years of experience in data science. During his tenure, he has worked with global clients in various domains like Banking, Insurance, Private Equity, Telecom and Human Resource.
Post a Comment