R_basic1

what is R? - https://zh.wikipedia.org/wiki/R%E8%AF%AD%E8%A8%80

R vs Python? - https://www.datacamp.com/community/tutorials/r-or-python-for-data-analysis - https://www.kdnuggets.com/2019/05/poll-top-data-science-machine-learning-platforms.html

Kaggle - https://www.kaggle.com/

UCI dataset - https://archive.ics.uci.edu/ml/datasets.html

電子檔講義 - https://www.dropbox.com/s/bhqmpndutktw401/201806%20-%20R%E8%AA%9E%E8%A8%80.pdf?dl=0

R intro

RDemo

#使用範例資料
data(anscombe)
#使用資料中x1,y1變數畫出點散布圖
plot(y1 ~ x1, data = anscombe)
#建立回歸模型並assign到lmfit變數中
lmfit <- lm(y1~x1, data=anscombe) 
#在點散佈圖上加上迴歸線
abline(lmfit, col="red")

Basic type

• numeric: 1,2,1.2
• integer: 1L,2L,3L
• character: “string”
• logical: TRUE,FALSE,T,F
• complex: 1+4i
• date: “2018-05-01”
• posixct, posixlt: “2018-05-01 08:00:00 CST”

Basic Objects

(由相同資料型態組成)

atomic:

• vector
• matrix
• factor

(可以有混合的資料型態)

recursive:

• dataframe
• list

R basic command

#文件查詢
help(package="base")
?base::sum
?sum
help.search("sum")
??sum

#範例演釋
demo()
#使用內建資料集
data()
#看現有變數
ls()

## [1] "anscombe" "lmfit"

#移除變數
rm()

x = c(1,2,3)
#查看資料型態
class(x)

## [1] "numeric"

#查看資料結構
str(x)

##  num [1:3] 1 2 3

Basic computing

3+8

## [1] 11

3-8

## [1] -5

3*8

## [1] 24

3/8

## [1] 0.375

3^2

## [1] 9

11%%2

## [1] 1

3<4

## [1] TRUE

2==5

## [1] FALSE

T == TRUE

## [1] TRUE

Assignment

a = 3
a <- 3 # ( alt + - )
assign("a",3)

a / 2

## [1] 1.5

a = a / 2
a

## [1] 1.5

Vector

• R語言最基本的物件

character(5)  ## character vector of length 5

## [1] "" "" "" "" ""

numeric(5)

## [1] 0 0 0 0 0

logical(5)

## [1] FALSE FALSE FALSE FALSE FALSE

x = c(1,2,3,7)
y= c(2,3,5,1)
x+y

## [1] 3 5 8 8

x*y

## [1]  2  6 15  7

x-y

## [1] -1 -1 -2  6

x/y

## [1] 0.5000000 0.6666667 0.6000000 7.0000000

x = c(1,2,3,7)
x + 10

## [1] 11 12 13 17

x + c(10)

## [1] 11 12 13 17

x + c(1,2)

## [1] 2 4 4 9

x + c(1,2,1,2)

## [1] 2 4 4 9

x == c(1,99,3,4)

## [1]  TRUE FALSE  TRUE FALSE

c(1,2,3)

## [1] 1 2 3

c(2,T,3+0i,"one")

## [1] "2"    "TRUE" "3+0i" "one"

c(2,T,3+0i)

## [1] 2+0i 1+0i 3+0i

c(c(1,2,3,4),c(5))

## [1] 1 2 3 4 5

x = c(1,2,3,4,NA)
sum(x)

## [1] NA

sum(x, na.rm=T)

## [1] 10

x = c(1,2,3,4,NA)
is.na(x)

## [1] FALSE FALSE FALSE FALSE  TRUE

sum(x[!is.na(x)])

## [1] 10

height_vec = c(180,169,173)
height_vec

## [1] 180 169 173

names(height_vec) = c("Brian", "Toby", "Sherry")
height_vec

##  Brian   Toby Sherry 
##    180    169    173

name_vec = c("Brian", "Toby", "Sherry")
names(height_vec) = name_vec
height_vec > 175

##  Brian   Toby Sherry 
##   TRUE  FALSE  FALSE

height_vec / 100

##  Brian   Toby Sherry 
##   1.80   1.69   1.73

height_vec > 175 | height_vec < 170

##  Brian   Toby Sherry 
##   TRUE   TRUE  FALSE

height_vec < 175 & height_vec > 170

##  Brian   Toby Sherry 
##  FALSE  FALSE   TRUE

#R 的index從1開始
height_vec[c(1)] #index

## Brian 
##   180

height_vec['Brian'] #element name

## Brian 
##   180

height_vec[height_vec > 175] #condition (boolean vector)

## Brian 
##   180

p38 example

h = c(180,169,173)
w = c(73,87,43)
bmi = w / ((h/100)^2)
names(bmi) = c("Brian", "Toby", "Sherry")
bmi < 18.5 | bmi >= 24

##  Brian   Toby Sherry 
##  FALSE   TRUE   TRUE

bmi[bmi < 18.5 | bmi >= 24]

##     Toby   Sherry 
## 30.46112 14.36734

seq() & rep() & paste()

1:20

##  [1]  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19 20

seq(1,20)

##  [1]  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19 20

20:1

##  [1] 20 19 18 17 16 15 14 13 12 11 10  9  8  7  6  5  4  3  2  1

?seq
seq(from=1,to=20,by=2)

##  [1]  1  3  5  7  9 11 13 15 17 19

seq(from=1,to=20,length=2)

## [1]  1 20

rep(1,5)

## [1] 1 1 1 1 1

?rep
rep(x=c(1,2), times=5)

##  [1] 1 2 1 2 1 2 1 2 1 2

rep(x=c(1,2), times=c(1,2))

## [1] 1 2 2

rep(x=c(1,2), each=5)

##  [1] 1 1 1 1 1 2 2 2 2 2

rep(x=c(1,2), length=5)

## [1] 1 2 1 2 1

rep_len(x=c(1,2),length.out = 5)

## [1] 1 2 1 2 1

paste("the","big","bang","theory")

## [1] "the big bang theory"

paste("big","bang",sep="-")

## [1] "big-bang"

length(paste("the","big","bang","theory"))

## [1] 1

paste("big","bang",sep="")

## [1] "bigbang"

paste("big","bang",sep=";")

## [1] "big;bang"

paste(c("big","bang"),1:2)

## [1] "big 1"  "bang 2"

paste(c("big","bang"),1:2,collapse = "+" )

## [1] "big 1+bang 2"

length(paste(c("big","bang"),1:4,collapse = "+" ))

## [1] 1

Matrix

matrix(1:9, byrow=TRUE, nrow=3)

##      [,1] [,2] [,3]
## [1,]    1    2    3
## [2,]    4    5    6
## [3,]    7    8    9

matrix(1:9, nrow=3)

##      [,1] [,2] [,3]
## [1,]    1    4    7
## [2,]    2    5    8
## [3,]    3    6    9

kevin = c(85,73)
marry = c(72,64)
jerry = c(59,66)
mat = matrix(c(kevin, marry, jerry), nrow=3, byrow= TRUE)
colnames(mat) = c('first', 'second')
rownames(mat) = c('kevin', 'marry', 'jerry')
mat

##       first second
## kevin    85     73
## marry    72     64
## jerry    59     66

#取得矩陣維度
dim(mat)

## [1] 3 2

#取得矩陣列數
nrow(mat)

## [1] 3

#取得矩陣欄數
ncol(mat)

## [1] 2

#矩陣轉置(transpose)
t(mat)

##        kevin marry jerry
## first     85    72    59
## second    73    64    66

#取第一列
mat[1,]

##  first second 
##     85     73

#取第一行
mat[,1]

## kevin marry jerry 
##    85    72    59

#取第一、二列
mat[1:2,]

##       first second
## kevin    85     73
## marry    72     64

#取kevin和jerry成績
mat[c('kevin','jerry'),]

##       first second
## kevin    85     73
## jerry    59     66

#取kevin和jerry成績的第一次考試成績
mat[c('kevin','jerry'),'first']

## kevin jerry 
##    85    59

#取得第一次考試成績不及格的人
mat[mat[,1] < 60,'first']

## [1] 59

Matrix(續)

#新增列與行
mat2 = rbind(mat, c(78,63))
rownames(mat2)[nrow(mat2)] = 'sam'
mat2

##       first second
## kevin    85     73
## marry    72     64
## jerry    59     66
## sam      78     63

mat3 = cbind(mat2,c(82,77,70,64))
colnames(mat3)[ncol(mat3)] = 'third'
mat3

##       first second third
## kevin    85     73    82
## marry    72     64    77
## jerry    59     66    70
## sam      78     63    64

rowMeans(mat3)

##    kevin    marry    jerry      sam 
## 80.00000 71.00000 65.00000 68.33333

colMeans(mat3)

##  first second  third 
##  73.50  66.50  73.25

# arithmetic
m1 = matrix(1:4, byrow=TRUE, nrow=2)
m2 = matrix(5:8, byrow=TRUE, nrow=2)

m1 + m2

##      [,1] [,2]
## [1,]    6    8
## [2,]   10   12

m1 - m2

##      [,1] [,2]
## [1,]   -4   -4
## [2,]   -4   -4

m1 * m2

##      [,1] [,2]
## [1,]    5   12
## [2,]   21   32

m1 / m2

##           [,1]      [,2]
## [1,] 0.2000000 0.3333333
## [2,] 0.4285714 0.5000000

m1 %*% m2

##      [,1] [,2]
## [1,]   19   22
## [2,]   43   50

Factor

# syntax
weather= c("sunny","rainy", "cloudy", "rainy", "cloudy")
weather_category = factor(weather)
weather_category

## [1] sunny  rainy  cloudy rainy  cloudy
## Levels: cloudy rainy sunny

class(weather)

## [1] "character"

class(weather_category)

## [1] "factor"

levels(weather_category)

## [1] "cloudy" "rainy"  "sunny"

# order
temperature = c("Low", "High", "High", "Medium", "Low", "Medium")
temperature_category = factor(temperature, order = TRUE, levels = c("Low", "Medium", "High"))
temperature_category

## [1] Low    High   High   Medium Low    Medium
## Levels: Low < Medium < High

temperature_category[3] > temperature_category[1]

## [1] TRUE

temperature_category[4] > temperature_category[3]

## [1] FALSE

# change levels name
weather= c("s","r", "c", "r", "c")
weather_factor = factor(weather)
levels(weather_factor) = c("cloudy","rainy","sunny")
weather_factor

## [1] sunny  rainy  cloudy rainy  cloudy
## Levels: cloudy rainy sunny

#type priority
c("string",1+2i,5.5,TRUE)

## [1] "string" "1+2i"   "5.5"    "TRUE"

c(1+2i,5.5,TRUE)

## [1] 1.0+2i 5.5+0i 1.0+0i

c(5.5,TRUE,FALSE)

## [1] 5.5 1.0 0.0