zadania5

# 0
# txt
tapply(bikes$cnt, bikes$yr, sum)
# gra
barplot(tapply(bikes$cnt, bikes$yr, sum),
        main = "Ogólna liczba wypożyczeń w poszczególnych latach",
        space = 1)

# 1
# txt
bikes11 <- bikes[bikes$yr==2011,]
bikes115 <- bikes11[bikes11$mnth==5,]
tapply(bikes115$cnt, bikes115$weekday, mean)
# gra
bikes2001 <- bikes[bikes$yr==2011,]
bikes5 <- bikes2001[bikes2001$mnth==5,]
bikes2001
bikes5
barplot(tapply(bikes5$cnt, bikes5$weekday, mean) ,
        main = "Średnia liczba wypożyczeń w poszczególnych dniach tygodnia w maju 2011", space = 1)
# 2
# txt
bikes <- read.csv2("./data/bikes.csv")
bikes$season <- factor(bikes$season, levels = 1:4,labels = c("winter", "spring", "summer", "fall"))
bikes$yr <- factor(bikes$yr, levels = 0:1, labels = c("2011", "2012"))
bikes$weathersit <- factor(bikes$weathersit, levels = 1:3, labels = c("clear", "mist", "light snow or rain"))
bikes1 <- bikes[bikes$yr=="2011",]
tapply(bikes1$cnt, bikes1$mnth, mad)
# gra
bikes <- read.csv2("./data/bikes.csv")
bikes$season <- factor(bikes$season, levels = 1:4,labels = c("winter", "spring", "summer", "fall"))
bikes$yr <- factor(bikes$yr, levels = 0:1, labels = c("2011", "2012"))
bikes$weathersit <- factor(bikes$weathersit, levels = 1:3, labels = c("clear", "mist", "light snow or rain"))
bikes1 <- bikes[bikes$yr=="2011",]
boxplot(bikes1$cnt~bikes1$mnth, main = "Zróżnicowanie wypożyczeń w roku 2011",space = 1)

# 3
# txt
bikes2011 <- bikes[bikes$yr==2011,]
bikes2011kw <- bikes2011[bikes2011$mnth==4,]
tapply(bikes2011kw$cnt, bikes2011kw$weathersit, sum)
# gra
bikes11 <- bikes[bikes$yr==2011,]
bikes114 <- bikes11[bikes11$mnth==4,]
boxplot(bikes114$cnt~bikes114$weathersit)

# 4
# txt
bikes2011 <- bikes[bikes$yr=="2011",]
bikes2011
spring <- bikes2011[bikes2011$season=="spring",]
spring
tapply(sort(spring$cnt), spring$atemp, sum)

# gra
bikes2011 <- bikes[bikes$yr=="2011",]
bikes2011s <- bikes2011[bikes2011$season=="spring",]
plot(bikes2011s$cnt~bikes2011s$atemp)

# 5
# txt
bikes <- read.csv2("./data/bikes.csv")
bikes$season <- factor(bikes$season, levels = 1:4,labels = c("winter", "spring", "summer", "fall"))
bikes$yr <- factor(bikes$yr, levels = 0:1, labels = c("2011", "2012"))
bikes1 <- bikes[bikes$yr=="2012",]
tapply(bikes1$casual, bikes1$season, mean)
tapply(bikes1$registered, bikes1$season, mean)

# gra
bikes <- read.csv2("./data/bikes.csv")
bikes$season <- factor(bikes$season, levels = 1:4,labels = c("winter", "spring", "summer", "fall"))
bikes$yr <- factor(bikes$yr, levels = 0:1, labels = c("2011", "2012"))
bikes1 <- bikes[bikes$yr=="2012",]

boxplot(bikes1$casual~bikes1$season, main = "Liczba wypożyczeń w 2012 roku wśród niezarejestrowanych użytkowników ze względu na porę roku", xlab="pora roku",ylab="wypożyczenia wśród niezarejestrowanych")

boxplot(bikes1$registered~bikes1$season, main = "Liczba wypożyczeń w 2012 roku wśród zarejestrowanych użytkowników ze względu na porę roku", xlab="pora roku",ylab="wypożyczenia wśród zarejestrowanych")

# 6
vgs <- read.csv("./data/vgs.csv")
aggregate(vgs$Global_Sales, by=list(platform=vgs$Platform), FUN=sum)

# 7

sum(grepl("Mario ", vgs$Name, ignore.case = TRUE) | grepl("Mario$", vgs$Name, ignore.case = TRUE) )

# 8
# txt
VGS1 <- vgs[, c("Name", "Global_Sales", "Platform")]
subset(VGS1, VGS1$Name == "The Witcher 3: Wild Hunt")

# gra

tmp <- vgs[vgs$Name=='The Witcher 3: Wild Hunt',  ]
tmp$Platform <- factor(tmp$Platform, levels = c("PS4", "XOne","PC"), labels = c("PS4", "XOne","PC"))
plot(tmp$Global_Sales ~ tmp$Platform)


# 9
vgs<- read.csv("./data/vgs.csv")
vgs1 <- subset(vgs$Platform, vgs$Genre == "Platform")
vgsx <- vgs1=="XB" | vgs1== "X360" | vgs1 == "XOne"
sum(vgs1=="XB")/sum(vgsx)*100
sum(vgs1=="X360")/sum(vgsx)*100
sum(vgs1=="XOne")/sum(vgsx)*100

# 10
plot(table(vgs$Year)[1:37],type='l',ylab = 'Liczba gier')

# 11

boxplot(vgs$Global_Sales~vgs$Genre, outline=FALSE,  xlab = "", ylab="", main= "rozpiętość sprzedaży - poszczególnych gatunków gier")

# 12

vgs<- read.csv("./data/vgs.csv")
vgssale <- sum(vgs$Global_Sales)
sum(vgs$JP_Sales)/vgssale*100
sum(vgs$EU_Sales)/vgssale*100
sum(vgs$NA_Sales)/vgssale*100
sum(vgs$Other_Sales)/vgssale*100

# 13

Ameryka<- tapply(vgs$NA_Sales, vgs$Genre, sum)
UE<- tapply(vgs$EU_Sales, vgs$Genre, sum)
Japonia<- tapply(vgs$JP_Sales, vgs$Genre, sum)
vgss<- data.frame (Ameryka,UE,Japonia)
vgss

# 14


plot(levels(vgs$Year),
     tapply(vgs$Global_Sales[vgs$Platform=="PS"], vgs$Year[vgs$Platform=="PS"], sum), type = 'b', col="red",
     ylim=c(0, 250),
     xlim=c(1990,2018), xlab = "Rok", ylab = 'Sprzedaż (w milionach kopii)',
     main='Sprzedaż gier na generacje PlayStation')
lines(levels(vgs$Year),
      tapply(vgs$Global_Sales[vgs$Platform=="PS2"], vgs$Year[vgs$Platform=="PS2"], sum),
      type = 'b', col="blue")
lines(levels(vgs$Year),
      tapply(vgs$Global_Sales[vgs$Platform=="PS3"], vgs$Year[vgs$Platform=="PS3"], sum),
      type = 'b', col="green")
lines(levels(vgs$Year),
      tapply(vgs$Global_Sales[vgs$Platform=="PS4"], vgs$Year[vgs$Platform=="PS4"], sum),
      type = 'b', col="violet")
legend("topright", legend = c("ps", 'ps2', 'ps3', 'ps4'), col=c('red', 'blue', 'green', 'violet'), lty = 1)

# 15
library(ggplot2)
table(diamonds$color)
?diamonds
# 16

hist(diamonds$price,breaks=30,xlim=c(0,20000),ylim=c(0,14000),main="Przedział cen diamentów")

# 17
diamonds$carat2 <- cut(diamonds$carat,breaks=c(seq(0,5,0.5),Inf))
tapply(diamonds$price,diamonds$carat2,mean)

# 18
boxplot(diamonds$price~diamonds$clarity,outline=FALSE,xlab="Klasa przejrzystości",ylab="Cena ($)")

# 19
boxplot(diamonds$price~diamonds$cut,outline=FALSE,xlab="Jakość szlifu",ylab="Cena ($)")


# 20
premdmd<- diamonds[diamonds$cut=="Premium" & diamonds$carat>2, ]
table(cut(premdmd$price, breaks = 5, labels = c('5-8tys.','8-11tys.',

                                                '11-14tys.','14-17tys.','17-20tys.')), cut(premdmd$carat, breaks = 5))

table(cut(premdmd$price, breaks = 5, labels = c('5-8tys.','8-11tys.',

                                                '11-14tys.','14-17tys.','17-20tys.')), cut(premdmd$y, breaks = 5))

table(cut(premdmd$price, breaks = 5, labels = c('5-8tys.','8-11tys.',

                                                '11-14tys.','14-17tys.','17-20tys.')), cut(premdmd$x, breaks = 5))
# 21

round(mean(diamonds$price[diamonds$cut=='Premium' & diamonds$clarity=='SI2'])-mean(diamonds$price[diamonds$cut=='Premium' & diamonds$clarity=='IF']),0)