Util functions
getwd() |
setwd(‘C://file/path’) |
rm(variable_name) |
str(variable_name) |
help.start() |
install.packages("ade4") |
library(ade4) |
detach(package:ade4) |
history() |
DataFrame
d=data.frame(subjectID=1:3,gender=c("M","F","F"),score=c(8,3,6)) |
A list where all elements are the same length. |
rbind(a_data_frame, another_data_frame) |
Bind rows |
cbind(a_data_frame, another_data_frame) |
Bind columns |
Strings
x <- (1:15) ^ 2 |
toString(x) |
toupper("I'm Shouting") |
tolower("I'm Shouting") |
tolower("I'm Shouting") |
strsplit(woodchuck, " ", fixed = TRUE) |
Data.table
library(data.table) |
class(flights) |
head(flights) |
flights[, .(.N), by = .(origin)] |
flights[, head(.SD, 2), by = month] |
flights[1:5,sum(arr_delay,dep_delay),] |
Vectors
t(a) |
transpose |
5 * a |
scalar multiplication |
a+b |
summing vector |
c(1,0) |
unit vectors |
|
|
Matrices
matrix(1:6,2,3) |
m2=matrix(1:3) |
Vectors
y<-c(5,7,7,8,2,5,6,4) |
Numeric vector |
x <- c("one","two","three") |
Character vector |
z <- c(TRUE,TRUE,FALSE) |
Logical vector |
Lists
cars<-list(c("Toyota", "Nissan", "Honda"), c(150,180,50)) |
Collection of elements which can be of different types. |
cars[[1]] |
first row of the list |
Descriptive Statistics
summary(mydat) |
describe(mydat) |
str(mydat) |
names(mydat) |
par(mfrow=c(2,2)) |
plot(density(female_dat$science_score)) |
Functions
hypotenuse(3, 4) |
formalArgs(hypotenuse) |
normalize(c(1, 3, 6, 10, NA)) |
f(sqrt(5)) |
Hypothesis Testing
t.test(x, y) |
t-test - difference between means. |
prop.test |
Test for difference between proportions. |
pairwise.t.test |
t-test for paired data. |
cor.test(sample1,sample2) |
Correlation |
wilcox.test(data3) |
Alternate hypothesis is proved |
chisq.test(marks1) |
Chi square test |
shapiro.test(vnor) |
Distribution is normal |
aov |
ANOVA - Analysis of Variance |
|
|
Arrays & Matrices
(two_d_array <- array( 1:12,
dim = c(4, 3),
dimnames = list(
c("one", "two", "three", "four"),
c("c1", "c2", "c3"))))
dim(two_d_array)
nrow(two_d_array)
ncol(two_d_array)
length(two_d_array)
|
Visualization
barplot(Species) |
ggplot(mydata1,aes(x = subject, fill = subject) ) + geom_bar() |
hist(Sepal.Length) |
plot(Sepal.Width) |
qqnorm(Sepal.Width) |
library(ggplot2) |
pie(table(Species)) |
library(learningr) |
Probability
Uniform |
u <- runif(2000) |
Normal or Gaussian |
u <- rnorm(2000,mean=50,sd=3) |
Exponential |
u <- rexp(2000) |
Binomial Distribution |
mybinom(k,n,p) * 1000 |
Poisson Distribution |
mypois(lambda, 2) |
Matrix Manipulation
det(matrix(c(1,0,0,1),2)) |
Determinant |
solve(m1) %*% m1 |
Inverse |
library(MASS) |
ginv(m1) |
Statistics - Algorithms
predict(model3,mydat) |
Regression |
table(predict.glm(modelg,newdata=mydat,type="response")>0.5) |
Classification |
cl$cluster |
Clustering |
|
