## R basics

x <- 5

y.1 <- 10


## Data types

x1 <- "Hello everyone!"
x2 <- 5.4
x3 <- 20L
x4 <- 1 + 1i
x5 <- TRUE

# print content of a variable
x1
# or
print(x1)
# get class of the variable:
class(x1)


x <- 5
y <- c(1,1,2,3,5,8)
p <- c("one", "two", "three")
r <- c(T, F, TRUE, FALSE)
r


z1 <- 1:10
z1 <- seq(1, 10, 1)  # from, to, by
z1

z2 <- 5:-5
z2 <- seq(5, -5, -1)
z2

length(z2)

a <- numeric(0)
length(a)
b <- numeric(5)
b

list1 <- list("a", FALSE, 4) 
list2 <- list(1, list1)  # nested list
list2

m1 <- matrix(1:8, nrow = 2, ncol = 4)
m1

m2 <- matrix(0, nrow = 2, ncol = 2)
m2

first = 1:3
second = c("a", "b", "cd")
d <- data.frame(first, second)
d

char.vector <- c("male", "female", "female", "male")
f <- factor(char.vector)
f
levels(f)


## Operators

1 + 1
5 - 4
5 * 4
5 / 3  # real division
5 %/% 3  # integer division
5 %% 3  # modulo
5 ^ 4  # power
c(1,1,2) - c(3,5,8)
c(1,1,2) * 3
c(1,1,2) * c(3,5,8)


paste(c("You", "are", "welcome!"), collapse = " ")

c(2,3,5) %*% c(1,2,3)

a <- matrix(1:4, 2, 2); a
a %*% a

y <- c(1,1,2,3,5,8)
y > 1
!(y > 1)
y > 1 & y < 8   # AND
y == 2 | y == 5   # OR

y <- c(1,1,2,3,5,8)
y %in% 2:5

## Control Structures

x <- 2
if (x > 4) {
    print("YES")
} else {
    print("NO")
}

for(i in c("a", "b", "c")) {
    print(i)
}

x <- 1
while(x < 5) {
    print(x)
    x <- x+1
}

x <- 1
repeat {
    if (x == 3) {
        x <- x+1
        next    # continue
    }
    if (x > 5) {
        break
    }
    print(x)
    x <- x+1
}

myFunc <- function(x, y = 3) {
    x + y
}
myFunc(2)
myFunc(2, 5)


myFunc2 <- function(x, y = 3) {
    return(x + y)
}
myFunc2(2)
myFunc2(x = 2, y = 5)


## Data indexing

y <- c(1,1,2,3,5,8)
y[3]
y[c(F, F, T, T, F, F)]
y[2:4]
y[-c(2,3)]


y <- c(1,1,2,3,5,8)
y[numeric(0)]
y[-numeric(0)]

y <- c(1,1,2,3,5,8)
y[y >= 2]   # returns values
which(y >= 2)  # returns indices
which(c(T,F,T))

d <- data.frame(first = 4:6, second = c("a", "b", "c"))
d 
d[2,2]
d[2, 1:2] # also: d[2, ]

d <- data.frame(first = 1:3, second = c("a", "b", "c"))
d$first
d[[1]]
d[2, "first"]

d[1]

## File reading and writing
weather <- read.csv("weather.csv")
weather[1:3, ]

weather <- read.csv("weather.csv") 
head(weather)  # see also tail() 
tail(weather)

summary(weather)

table(weather$outlook)
dim(weather)  # dimension
nrow(weather)  # number of rows
ncol(weather)  # number of columns


## Missing values
a <- c(1,2,3,NA,5,NA)

is.na(a)


## Plots
plot(x = weather$temperature, y = weather$humidity)

plot(x = weather$temperature, y = weather$humidity,
     col = weather$play, pch = 20, cex = 2,
     xlab = "temperature", ylab = "humidity",
     main = "Weather data")
legend("topright", levels(weather$play),
       col = 1:2, pch = 20)

random_values <- rnorm(1000, mean = 25, sd = 10)
head(random_values)
hist(random_values)

plot(1:20, random_values[1:20], type = "l")


## R example - once again

library(RWeka)
library(partykit)
weather <- read.csv("weather.csv")
model <- J48(play ~ ., data = weather)
plot(model)


## -------------------------------------------------
## EXTRA - not in slides

## indexing lists
l <- list(1:3, 30:35, 3:8)
l[1]
l[[1]]
class(l[1])
class(l[[1]])

l[2:3]
l[[3]][[1]]
l[[c(3,1)]]



## repetition of values
rep(c(1,2,3), 5)
rep(c(1,2,3), c(3,1,3))
rep(c(1,2,3), each = 5)

## combine data frames:
df1 <- data.frame(a = 1:3, b = c("a", "b", "c"))
df2 <- data.frame(a = 4:5, b = c("d", "e"))
df3 <- data.frame(x = c(T,T,F), y = 10:12)
rbind(df1, df2)  ## by rows
cbind(df1, df3)  ## by columns
cbind(df1, m = c("x", "y", "z"))  ## append column

## data conversion
a <- 4:10
a
as.character(a)

f <- as.factor(a)
f
as.integer(f)  ## returns integer values used for internal representation !!!
unclass(f)  ## same values

as.numeric(as.character(f))  ## this will return original values

## computing summary statistics
a <- 1:10
sum(a)
max(a)
min(a)
mean(a)
sd(a)
median(a)
quantile(a)

m <- matrix(rep(0:3, each = 4), 8, 2)
m
rowSums(m)
colSums(m)
rowMeans(m)
colMeans(m)

## LAPPLY, SAPPLY
lapply(1:4, sqrt)

sapply(1:4, sqrt)

sapply(1:4, function(x) {
    sqrt(x)
})

sapply(1:4, function(x) {
    -x
})

l <- list(1:3, 30:35, 3:8)
l
sapply(l, mean)

## MAPPLY
a <- 2:4
b <- 3:5
a
b
mapply(function(x, y) {
    x ^ y
}, 2:4, 3:5)

## APPLY
m <- matrix(rep(0:3, each = 4), 8, 2)
m
apply(m, 1, sum)
apply(m, 2, sum)