mat pca uts.r

mat pca uts.r

library(ggplot2)

cr = function(d){
    print(c(
        nrow(d), ncol(d)
    ))
}

ple = function(d){
    print(length(d))
}

printx = function(d){
    print(d)
    quit()
}

hr = function(d){
    print(paste("================",d,"================"))
}

data = matrix(
    c(
        2.5, 2.4,
        0.5, 0.7,
        2.2, 2.9,
        1.9, 2.2,
        3.1, 3,
        2.3, 2.7,
        2, 1.6,
        1, 1.1,
        1.5, 1.6,
        1.1, 0.9
    ),
    ncol=2,
    byrow=T
)

pca = function(data=NULL){
    obj = list(
        data=data,
        x=NULL,
        adjustData=NULL,
        e=NULL,
        rowFeatVector = NULL,
        rowZeroMeanData = NULL,
        finalData = NULL,
        rowOriginalData = NULL
    )

    
    getcovars = function(m){
        n = ncol(m)
        covars = c()
        for (i in 1: n){
            for ( j in 1:n) {
                # print(cov(m[,i], m[,j]))
                covars = append(covars, cov(m[,i], m[,j]))
            }
        }
        covars
    }

    adjustData = function(m){
        n = ncol(m)
        means = c()
        for (i in 1:n){
            me = mean(m[,i])
            means = append(means, me)
            m[,i] = m[,i] - me
        }

        list(
            m=m,
            means=means
        )
    }

    covars = getcovars(data)

    x = matrix(
        covars,
        # c(
        #     0.616555556, 0.615444444,
        #     0.615444444, 0.716555556
        # ),
        nrow=ncol(data),
        byrow=T
    )
    obj$x = x
    e = eigen(x, T)
    obj$e = e
    hr("vector")
    print(e$vector)
    print(t(e$vector))
    objad = adjustData(data)
    adjData = objad$m
    # hr("adjdata")
    # print(adjData)
    # print(t(adjData))
    obj$adjustData = adjData
    obj$rowFeatVector = t(adjData)
    obj$rowZeroMeanData = t(e$vector)
    # finalData = rowFeatVector %*% rowZeroMeanData
    obj$finalData = obj$rowZeroMeanData %*% obj$rowFeatVector
    print(obj$finalData[,1])
    obj$rowOriginalData = t(obj$rowFeatVector) %*% obj$finalData + objad$means
    objad$means

    # finalData = t(finalData)
    print(x)
    print(e)
    print(data)
    plot(obj$finalData[1,], obj$finalData[1,])
    print(t(obj$finalData))
    print(obj$rowOriginalData)

    obj
}

# pca1 = pca(data)
# pca1

fname = "/media/data1/_S2/grad-notes/1.3-mtk-stat/dummy/phising.csv"
# fname = "./dummy/phising.csv"
df = read.csv(fname)
x = as.matrix( subset(df, select=having_IP_Address:Statistical_report) )
x = x[1:30,]
head(x)

doPCA1 = function(x){
    # https://www.rdocumentation.org/packages/stats/versions/3.6.2/topics/prcomp
    pcaresult = prcomp(x)
    plot(pcaresult)
    summary(pcaresult)
}
doPCA1(x)

# pca(x)