In naveedmerchant/BayesScreening: Classical and Bayesian Screening for Binary Classification

knitr::opts_chunk$set(echo = TRUE)
HiggsCSV = "C:\\Users\\Naveed\\Documents\\HIGGS.csv.gz"

KDEs of plots

The following code yields KDEs of col 23 and 29 in the Higgs Boson data set.

library(BSCRN)
fulltrans = read.csv(file = HiggsCSV, colClasses = c(NA,rep("NULL",times = 21),rep(NA,times = 7)), header = FALSE)
dim(fulltrans)

noisedataind = fulltrans[,1]== 0

X = fulltrans[noisedataind,]
#This is "background noise"
Y = fulltrans[!noisedataind,]

fulltrans2 = fulltrans[1:1000000,]

usedsamptrans = fulltrans2[1:20000,]
noisedataind3 = usedsamptrans[,1]== 0
X2 = usedsamptrans[noisedataind3,]
#This is "background noise"
Y2 = usedsamptrans[!noisedataind3,]


plot(density(Y[,8], n = 511, from = 0, to = 4), xlab = "x", main = "")
lines(density(X[,8], n = 511, from = 0, to = 4), col = "blue")

plot(density(Y[,2], n = 511, from = 0, to = 3), xlab = "x", main = "")
lines(density(X[,2], n = 511, from = 0, to = 3), col = "blue")

The following code draws from the predictive posterior of the Higgs Boson data set, and then plots them on a graph.

set.seed(500)
PTcol23 = PolyaTreePriorLikCons(datasetX = X2[,2], c = 1, Ginv = qnorm, leveltot = 9)
PTcol23draws = PolyaTreePredDraws(PTcol23, ndraw = 1000)


set.seed(1000)
trainindX = sample(1:nrow(X2) ,size = 5000)
trainindY = sample(1:nrow(Y2) ,size = 5000)
XT1 <- X2[trainindX,2]
XV1 <- X2[-(trainindX),2]

bwdraws = BSCRN::PredCVBFIndepMHbw(ndraw = 1000, maxIter = 3000, XT1 = XT1, XV1 = XV1)
CVBFpredpost = PredCVBFDens(bwdraws$predbwsamp, XT1 = XT1)

predpostavg2 = CVBFpredpost(seq(from = 0, to = 4, length.out = 10000))

plot(density(fulltrans[noisedataind,2], from = 0, to = 4, n = 10000), lwd = 2)
lines(seq(from = 0, to = 4, length.out = 10000), predpostavg2, col = "purple", lwd = 2)
lines(density(PTcol23draws, from = 0, to = 4, n = 10000), col = "red", lwd = 2)

naveedmerchant/BayesScreening documentation built on June 13, 2024, 7:56 a.m.