Search
TEST https://goo.gl/forms/foQgBffnLz8JYq2p1
Do svého pracovního adresáře uložte soubor iris.csv
# načtení dat setwd('/home/user/...') data <- read.csv("iris.csv")
library(datasets) data <- iris
# popis dat dim(data) names(data) data[1:5,] str(data) summary(data) dsum <- apply(data[,1:4],2,mean) dsum <- tapply(data$Sepal.Length, data$Species, mean) dsum <- by(data[1:4],data$Species,function(x) {apply(x,2,mean)}) # vrací list - není moc praktické pro další manipulaci, lze převést na data frame pomocí do.call('rbind',dsum) # pokročilou manipulaci s daty nabízí knihovna plyr library(plyr) dsum <- ddply(data,.(Species),function(x){apply(x[,1:4],2,mean)}) # užitečné funkce - mean, sd, median, is.na, is.nan, is.finite, ...
Klasické bodové a čarové grafy
plot(data$Sepal.Length) plot(data$Sepal.Length,type='l') plot(data$Sepal.Length,type='b') plot(data$Sepal.Length,data$Sepal.Width) plot(data$Sepal.Length,data$Sepal.Width,col=data$Species) plot(data$Sepal.Length,data$Sepal.Width,col=data$Species,pch=19) plot(data$Sepal.Length, data$Sepal.Width, col=data$Species, pch=19, main='Edgar Anderson\'s Iris data', xlab='Sepal Length', ylab='Sepal Width') # legenda legend('topright', legend=levels(data$Species), pch=19, col=1:length(levels(data$Species))) # vykreslit body avgSepal.Length <- tapply(data$Sepal.Length,data$Species,mean) avgSepal.Width <- tapply(data$Sepal.Width,data$Species,mean) points(avgSepal.Length,avgSepal.Width,pch='X',cex=2,col=4) # vykreslit čáry lines(lowess(data$Sepal.Length[data$Species=='setosa'],data$Sepal.Width[data$Species=='setosa'])) lines(lowess(data$Sepal.Length[data$Species=='versicolor'],data$Sepal.Width[data$Species=='versicolor']),col=2) lines(lowess(data$Sepal.Length[data$Species=='virginica'],data$Sepal.Width[data$Species=='virginica']),col=3) # další elementy - line(), abline(),
boxplot(data$Sepal.Length ~ data$Species)
hist(data$Sepal.Length) hist(data$Sepal.Length,100)
hist(data$Sepal.Length[data$Species == 'setosa'], 50, xlim=c(min(data$Sepal.Length), max(data$Sepal.Length)), col=2, main='Edgar Anderson\'s Iris data') hist(data$Sepal.Length[data$Species == 'versicolor'],50,col=3,add=T) hist(data$Sepal.Length[data$Species == 'virginica'],50,col=4,add=T)
pairs(data)
library(ggplot2)
Mapování dat → vizualizace
ggplot(data, aes(x=Sepal.Length, y=Sepal.Width))+geom_point() ggplot(data, aes(x=Sepal.Length, y=Sepal.Width))+geom_point()+geom_line() ggplot(data, aes(x=Sepal.Length, y=Sepal.Width))+geom_point(aes(color=Species))+geom_line() ggplot(data, aes(x=Sepal.Length, y=Sepal.Width))+geom_point()+geom_line(aes(color=Species)) ggplot(data, aes(x=Sepal.Length, y=Sepal.Width, color=Species))+geom_point()+geom_line()
Příklady grafů
ggplot(data, aes(x=Species, y=Petal.Width))+geom_boxplot() ggplot(data, aes(x=Sepal.Length))+geom_histogram() ggplot(data, aes(x=Sepal.Length))+geom_histogram(binwidth=0.1) ggplot(data, aes(x=Sepal.Length))+geom_histogram(binwidth=0.1, aes(colour=Species)) ggplot(data, aes(x=Sepal.Length))+geom_histogram(binwidth=0.1, aes(fill=Species)) ggplot(data, aes(x=Sepal.Length))+geom_histogram(binwidth=0.1, alpha=0.4, aes(fill=Species)) ggplot(data, aes(x=Sepal.Length))+geom_histogram(binwidth=0.1, position="identity", alpha=0.4, aes(fill=Species))
Použijte data "ADULT" z UCI Machine Learning Repository uložená v souboru adult.csv
bude doplněn po cvičení