This session is to practice apply family functions
There are several functions in apply family function, but we will focus on four main methods
Before we jump in applying the family function, we will create a dataset (matrix called mt)
set.seed(122)
mt<-matrix(rnorm(100),ncol = 10,nrow = 10,byrow = T)
head(mt)## [,1] [,2] [,3] [,4] [,5] [,6]
## [1,] 1.3107009 -0.8758531 0.1995237 0.4659544 -1.8020568 1.44874346
## [2,] -0.1321288 -0.8416347 0.2742522 -2.4625904 -2.1533960 1.15917575
## [3,] -0.5378254 -0.5226487 -1.0529219 0.1015172 1.8635011 -0.01960909
## [4,] 1.4113547 0.6013653 -0.9139976 1.5804259 -0.4699156 1.60183889
## [5,] -0.6580097 -1.6085024 -1.3440913 -0.1966739 1.3756627 0.17331500
## [6,] 2.5567545 -0.3615805 -0.4385265 -1.0592134 1.0109921 0.05498478
## [,7] [,8] [,9] [,10]
## [1,] 0.2988538 0.36181165 -1.0069404 -0.2569991
## [2,] 0.6897401 0.88025132 -0.9433702 0.7190729
## [3,] -1.0783075 0.17260428 0.5330541 1.9522645
## [4,] 0.3727399 0.03387786 0.1041034 -2.0183743
## [5,] -0.4392776 -1.59532290 -0.7442776 1.8152081
## [6,] 0.4523293 -0.17415183 0.8398331 0.8454829We should give the matrix column and row names, should not we?
# Label matrix columns
colnames(mt)<-paste("Col",1:ncol(mt),sep="_")
# Label matrix rows
rownames(mt)<-paste("Row",1:nrow(mt),sep="_")
# Check the new labeled matrix
head(mt)## Col_1 Col_2 Col_3 Col_4 Col_5 Col_6
## Row_1 1.3107009 -0.8758531 0.1995237 0.4659544 -1.8020568 1.44874346
## Row_2 -0.1321288 -0.8416347 0.2742522 -2.4625904 -2.1533960 1.15917575
## Row_3 -0.5378254 -0.5226487 -1.0529219 0.1015172 1.8635011 -0.01960909
## Row_4 1.4113547 0.6013653 -0.9139976 1.5804259 -0.4699156 1.60183889
## Row_5 -0.6580097 -1.6085024 -1.3440913 -0.1966739 1.3756627 0.17331500
## Row_6 2.5567545 -0.3615805 -0.4385265 -1.0592134 1.0109921 0.05498478
## Col_7 Col_8 Col_9 Col_10
## Row_1 0.2988538 0.36181165 -1.0069404 -0.2569991
## Row_2 0.6897401 0.88025132 -0.9433702 0.7190729
## Row_3 -1.0783075 0.17260428 0.5330541 1.9522645
## Row_4 0.3727399 0.03387786 0.1041034 -2.0183743
## Row_5 -0.4392776 -1.59532290 -0.7442776 1.8152081
## Row_6 0.4523293 -0.17415183 0.8398331 0.8454829We will apply the first function apply
# apply function takes three main arguments mt (matrix), margin (row "1" or colum "2", fun (what methods))
# Calculating the mean of all columns
apply(mt,2,mean)## Col_1 Col_2 Col_3 Col_4 Col_5 Col_6
## 0.07389782 -0.28224671 -0.22575555 -0.35505915 0.10132837 0.50260884
## Col_7 Col_8 Col_9 Col_10
## 0.09821195 0.05646223 -0.19066573 0.34130073# Calculating standard deviation
apply(mt,2, sd)## Col_1 Col_2 Col_3 Col_4 Col_5 Col_6 Col_7
## 1.2488349 0.8685868 0.7180313 1.4720801 1.3703386 0.8003884 0.9481447
## Col_8 Col_9 Col_10
## 0.8103711 0.9292266 1.5184065# Calculating median
apply(mt,2, median)## Col_1 Col_2 Col_3 Col_4 Col_5 Col_6
## -0.36679628 -0.44211458 -0.14176502 -0.06962635 0.45663236 0.20510032
## Col_7 Col_8 Col_9 Col_10
## 0.37193682 0.12832925 -0.19719723 0.57398151# Find maximum or minimum values
apply(mt,2,min)## Col_1 Col_2 Col_3 Col_4 Col_5 Col_6
## -1.1521364 -1.6085024 -1.3440913 -2.5177170 -2.1533960 -0.8038404
## Col_7 Col_8 Col_9 Col_10
## -1.8190443 -1.5953229 -1.7766103 -2.0183743apply(mt,2,max)## Col_1 Col_2 Col_3 Col_4 Col_5 Col_6 Col_7
## 2.5567545 1.4023026 0.6753913 1.7159783 1.8635011 1.6018389 1.3827442
## Col_8 Col_9 Col_10
## 1.2739718 0.8765213 2.3340020When using apply family functions, we don’t need to convert it to a matrix
–lapply function: This function returns its result in list
library(ggplot2)
lapply(trees,mean)## $Girth
## [1] 13.24839
##
## $Height
## [1] 76
##
## $Volume
## [1] 30.17097Define a function
# Calculate mean
my_func<- function(x){
mean(x)/min(x)
}
lapply(trees,my_func)## $Girth
## [1] 1.596191
##
## $Height
## [1] 1.206349
##
## $Volume
## [1] 2.957938sapply function returns its output in vector# Calculating mean or sd using sapply
sapply(trees,mean)## Girth Height Volume
## 13.24839 76.00000 30.17097sapply(df,mean)## Warning in mean.default(X[[i]], ...): argument is not numeric or logical:
## returning NA
## Warning in mean.default(X[[i]], ...): argument is not numeric or logical:
## returning NA
## Warning in mean.default(X[[i]], ...): argument is not numeric or logical:
## returning NA
## Warning in mean.default(X[[i]], ...): argument is not numeric or logical:
## returning NA
## Warning in mean.default(X[[i]], ...): argument is not numeric or logical:
## returning NA## x df1 df2 ncp log
## NA NA NA NA 0 NAsapply(trees, sd)## Girth Height Volume
## 3.138139 6.371813 16.437846Define a function to calculate mean of all tree columns
sapply(trees, function(x){sum(x)/length(x)})## Girth Height Volume
## 13.24839 76.00000 30.17097If dataset is a list, using lapply to get summary
mylist<-list(trees)
lapply(mylist,summary)## [[1]]
## Girth Height Volume
## Min. : 8.30 Min. :63 Min. :10.20
## 1st Qu.:11.05 1st Qu.:72 1st Qu.:19.40
## Median :12.90 Median :76 Median :24.20
## Mean :13.25 Mean :76 Mean :30.17
## 3rd Qu.:15.25 3rd Qu.:80 3rd Qu.:37.30
## Max. :20.60 Max. :87 Max. :77.00If you want to multiply 3 to all your dataset, using lapply to do that
mytrees<-trees
my_tree<-sapply(mytrees, function(x){x*3})
head(my_tree)## Girth Height Volume
## [1,] 24.9 210 30.9
## [2,] 25.8 195 30.9
## [3,] 26.4 189 30.6
## [4,] 31.5 216 49.2
## [5,] 32.1 243 56.4
## [6,] 32.4 249 59.1Some other interesting features of sapply and lapply
# Displaying first six rows
sapply(trees,head)## Girth Height Volume
## [1,] 8.3 70 10.3
## [2,] 8.6 65 10.3
## [3,] 8.8 63 10.2
## [4,] 10.5 72 16.4
## [5,] 10.7 81 18.8
## [6,] 10.8 83 19.7sapply(trees,colnames)## $Girth
## NULL
##
## $Height
## NULL
##
## $Volume
## NULLtapply used for calculating mean, sd, median ... for each group
library(caret)## Loading required package: lattice# Calculating the mean of Sepal length for each species
tapply(iris$Sepal.Length,iris$Species, mean)## setosa versicolor virginica
## 5.006 5.936 6.588