VinitaSilaparasetty

install.packages("rpart")
install.packages("dplyr")

library(rpart)
library(dplyr)

data(mtcars)

d.tree = rpart(formula = mtcars$cyl ~ mtcars$mpg, data = mtcars)
rpart.plot::rpart.plot(d.tree)

VinitaSilaparasetty

data(mtcars)
normed.sample <- quantileNormalise(mtcars.sample)
tmp_data <- kMeansBin(normed.sample)

VinitaSilaparasetty

install.packages("classInt")
library(classInt)

data <- c(0, 5, 10, 15, 15, 20, 25, 25, 30)

x <- class_interval(dataset, 5, style = 'quantile')

VinitaSilaparasetty

install.packages("classInt")
library(classInt)

data <- c(0, 5, 10, 15, 15, 20, 25, 25, 30)

x <- class_interval(data,5, style = 'equal')

VinitaSilaparasetty

from sklearn.model_selection import train_test_split
from feature_engine.discretisers import DecisionTreeDiscretiser

treeDisc = DecisionTreeDiscretiser(cv=20, scoring='accuracy',variables=['a', 'b'],regression=False,param_grid={'max_depth': [1,2,3],'min_samples_leaf':[20,4]})

VinitaSilaparasetty

from sklearn.preprocessing import KBinsDiscretizer

discrete_data = KBinsDiscretizer(n_bins=10, encode='ordinal', strategy='kmeans')

VinitaSilaparasetty

from sklearn.preprocessing import KBinsDiscretizer
from feature_engine.discretisers import EqualFrequencyDiscretiser

discrete_data = EqualFrequencyDiscretiser(q=20, variables = ['a', 'b'])

VinitaSilaparasetty

from sklearn.preprocessing import KBinsDiscretizer
from feature_engine.discretisers import EqualWidthDiscretiser

#distretization
discrete_data = KBinsDiscretizer(n_bins=20, encode='ordinal', strategy='uniform')

VinitaSilaparasetty

install.package("mice") #install mice
install.package("lattice")#install lattice
library("mice") #load mice
library("lattice") #load lattice

micedata <- mice(mtcars[, !names(mtcars) %in% "cyl"], method="rf")  # perform mice imputation, based on random forests.
miceOutput <- complete(micedata)  # generate the completed data.

#Check for NAs
anyNA(miceOutput)

VinitaSilaparasetty

install.package("DMwR") #install package
library(DMwR) #load package

#Knn imputation
knnOutput <- knnImputation(mtcars[, !names(mtcars) %in% "cyl"])  # perform knn imputation.

#check for NAs
anyNA(knnOutput)