R Data Manipulation
数据集(Datasets)
Iris
Iris数据集是著名的分类实验数据集，由Fisher或Anderson于1936收集整理. 该数据集由3种不同类型的鸢尾花的构成，各50个样本。
其中的一个种类与另外两个种类是线性可分离的，后两个种类是非线性可分离的。
该数据集包含 了5个属性：
- Sepal.Length（花萼长度）
- Sepal.Width（花萼宽度）
- Petal.Length（花瓣长度）
- Petal.Width（花瓣宽度）
- Species(种类)：Setosa（山鸢尾）、Versicolour（杂色鸢尾），Virginica（维吉尼亚鸢尾）
rbind(head(iris,5) , tail(iris,5)) |
## Sepal.Length Sepal.Width Petal.Length Petal.Width Species ## 1 5.1 3.5 1.4 0.2 setosa ## 2 4.9 3.0 1.4 0.2 setosa ## 3 4.7 3.2 1.3 0.2 setosa ## 4 4.6 3.1 1.5 0.2 setosa ## 5 5.0 3.6 1.4 0.2 setosa ## 146 6.7 3.0 5.2 2.3 virginica ## 147 6.3 2.5 5.0 1.9 virginica ## 148 6.5 3.0 5.2 2.0 virginica ## 149 6.2 3.4 5.4 2.3 virginica ## 150 5.9 3.0 5.1 1.8 virginica |
Converting data between wide and long (长宽数据转换)
长数据(Long Format)的变量不是在各个列上, 而是拍成一列, 每一个变量都分别占其中的几行, 这样便于对每个变量分组, R偏向于使用长数据。
宽数据(Wide Format)的变量则由每一列表示，比如常见的excel，spss等。
tidyr中gather()和spread()能够聚集(gather)和伸展(spread)数据框中的数据。
reshape2中melt()和cast()能够灵活的融合(melt)和重铸(cast，dcast)数据框中的数据。
R中自带的reshape(),stack()和unstack()也可以实现, 不过微繁, 这里不加赘述
library(reshape2) library(tidyr) library(dplyr) library(ggplot2) |
Wide to Long fromat(宽转长)
Tidyr
gather(data, key = “key”, value = “value”, …, na.rm = FALSE, convert = FALSE, factor_key = FALSE)
## factor_key : Treat the new key column as a factor (instead of character vector) long_iris <- gather(iris, Attributes, Measurement, Sepal.Length:Petal.Width, factor_key = T) ## These would have the same result as above # long_iris <- iris %>% gather(Attributes,Measurement,Sepal.Length:Petal.Width,factor_key = T) # long_iris <- gather(iris,Attributes,Measurement,-Species,factor_key = T) # long_iris <- gather(iris,"Attributes","Measurement",colnames(iris)[1:4],factor_key = T) # long_iris <- gather(iris,Attributes,Measurement,Sepal.Length,Sepal.Width,Petal.Length,Petal.Width,factor_key = T) |
## Species Attributes Measurement ## 1 setosa Sepal.Length 5.1 ## 2 setosa Sepal.Length 4.9 ## 3 setosa Sepal.Length 4.7 ## 4 setosa Sepal.Length 4.6 ## 5 setosa Sepal.Length 5.0 ## 596 virginica Petal.Width 2.3 ## 597 virginica Petal.Width 1.9 ## 598 virginica Petal.Width 2.0 ## 599 virginica Petal.Width 2.3 ## 600 virginica Petal.Width 1.8 |
Reshape2
melt(data, …, na.rm = FALSE, value.name = “value”)
long_iris <- melt(iris, # ID variables - all the variables to keep id.vars = colnames(iris)[5], # The split columns measure.vars = colnames(iris)[1:4], # Name of the destination column variable.name="Attributes", value.name="Measurement" ) ## This would have the same result as above # long_iris <- melt(iris, "Species", variable.name = "Attributes",vaule.name = "Measurement") |
## Species Attributes Measurement ## 1 setosa Sepal.Length 5.1 ## 2 setosa Sepal.Length 4.9 ## 3 setosa Sepal.Length 4.7 ## 4 setosa Sepal.Length 4.6 ## 5 setosa Sepal.Length 5.0 ## 596 virginica Petal.Width 2.3 ## 597 virginica Petal.Width 1.9 ## 598 virginica Petal.Width 2.0 ## 599 virginica Petal.Width 2.3 ## 600 virginica Petal.Width 1.8 |
Long to Wide fromat(长转宽)
宽数据的列顺序由变量的factor levels决定
如果你的数据有duplicated row, 需建立索引
显然long_iris含有duplicated rows
Find duplicated rows
long_iris1 <- long_iris[!duplicated(long_iris),] long_iris2 <- unique(long_iris) identical(long_iris1, long_iris2) # TRUE dim(long_iris1) # 175 3 |
## # A tibble: 10 x 4 ## rowname Species Attributes Measurement ## * <fct> <fct> <fct> <dbl> ## 1 1 setosa Sepal.Length 5.10 ## 2 2 setosa Sepal.Length 4.90 ## 3 3 setosa Sepal.Length 4.70 ## 4 4 setosa Sepal.Length 4.60 ## 5 5 setosa Sepal.Length 5.00 ## 6 565 virginica Petal.Width 2.40 ## 7 566 virginica Petal.Width 2.30 ## 8 570 virginica Petal.Width 1.50 ## 9 580 virginica Petal.Width 1.60 ## 10 585 virginica Petal.Width 1.40 |
## OR using dplyr library(dplyr) long_iris3 <- long_iris %>% distinct() ##== # long_iris %>% # distinct(Species, Attributes, Measurement, .keep_all = TRUE) |
## # A tibble: 10 x 4 ## rowname Species Attributes Measurement ## * <fct> <fct> <fct> <dbl> ## 1 1 setosa Sepal.Length 5.10 ## 2 2 setosa Sepal.Length 4.90 ## 3 3 setosa Sepal.Length 4.70 ## 4 4 setosa Sepal.Length 4.60 ## 5 5 setosa Sepal.Length 5.00 ## 6 171 virginica Petal.Width 2.40 ## 7 172 virginica Petal.Width 2.30 ## 8 173 virginica Petal.Width 1.50 ## 9 174 virginica Petal.Width 1.60 ## 10 175 virginica Petal.Width 1.40 |
Note: all dplyr methods ignore rownames
all.equal(long_iris1, long_iris3) # "Attributes: < Component “row.names”: Mean relative difference: 0.6602903 " |
Tidyr
spread(data, key, value, fill = NA, convert = FALSE, drop = TRUE, sep = NULL)
# wide_iris <- spread(long_iris,Attributes,Measurement) ## ERROR! duplicate identifiers for rows ## To build indices long_iris$index <- rep(1:150,4) wide_iris <- spread(long_iris, Attributes, Measurement) wide_iris <- wide_iris[,c(3:6,1)] identical(wide_iris,iris) # TRUE |
Reshape2
# wide_iris <- dcast(long_iris, Species ~ Attributes, value.var="Measurement") ## ERROR！Aggregation function missing: defaulting to length ## Try fun.aggregate wide_iris <- dcast(long_iris, Species ~ Attributes, value.var="Measurement",fun.aggregate=function(x) toString(unique(x))) |
wide_iris <- dcast(long_iris, Species + index ~ Attributes, value.var="Measurement") wide_iris <- wide_iris[,c(3:6,1)] identical(wide_iris,iris) |
## [1] TRUE |
# TRUE |
Merging data
Merge two data frames by common columns or row names
merge(x, y, by = intersect(names(x), names(y)), by.x = by, by.y = by, all = FALSE, all.x = all, all.y = all, sort = TRUE, suffixes = c(“.x”,“.y”), incomparables = NULL, …)
# Make up some data animals <- read.table(header=T, text=' size type name small cat Dragen-Li small cat lynx big cat tiger small dog chihuahua big dog "great dane" middle dog "siberian husky" small dog dachshund big dog "chinese kunmin" small dig "shetland sheepdog" big dog "German Shepherd" ') observations <- read.table(header=T, text=' number size type name 1 big cat tiger 2 small dog dachshund 3 small dog chihuahua 4 big dog "chinese kunmin" 5 middle dog poodle 6 big dog samoyed 7 middle dog "chow chow" 8 small dog papillon ') merge(observations, animals) ## These results are equal # merge(animals, observations) # merge(observations, animals,by=c("size","type","name")) # size type name number # 1 big cat tiger 1 # 2 big dog chinese kunmin 4 # 3 small dog chihuahua 3 # 4 small dog dachshund 2 |
merge(observations, animals,by.x="name",by.y="name") |
## name number size.x type.x size.y type.y ## 1 chihuahua 3 small dog small dog ## 2 chinese kunmin 4 big dog big dog ## 3 dachshund 2 small dog small dog ## 4 tiger 1 big cat big cat |
merge(observations, animals,by.x="name",by.y="name",all.x=T) |
## name number size.x type.x size.y type.y ## 1 chihuahua 3 small dog small dog ## 2 chinese kunmin 4 big dog big dog ## 3 chow chow 7 middle dog <NA> <NA> ## 4 dachshund 2 small dog small dog ## 5 papillon 8 small dog <NA> <NA> ## 6 poodle 5 middle dog <NA> <NA> ## 7 samoyed 6 big dog <NA> <NA> ## 8 tiger 1 big cat big cat |
merge(observations, animals,by.x="name",by.y="name",all.x=T,sort=FALSE) |
## name number size.x type.x size.y type.y ## 1 tiger 1 big cat big cat ## 2 dachshund 2 small dog small dog ## 3 chihuahua 3 small dog small dog ## 4 chinese kunmin 4 big dog big dog ## 5 poodle 5 middle dog <NA> <NA> ## 6 samoyed 6 big dog <NA> <NA> ## 7 chow chow 7 middle dog <NA> <NA> ## 8 papillon 8 small dog <NA> <NA> |
merge(observations, animals, all=T) |
## size type name number ## 1 big cat tiger 1 ## 2 big dog chinese kunmin 4 ## 3 big dog samoyed 6 ## 4 big dog German Shepherd NA ## 5 big dog great dane NA ## 6 middle dog chow chow 7 ## 7 middle dog poodle 5 ## 8 middle dog siberian husky NA ## 9 small cat Dragen-Li NA ## 10 small cat lynx NA ## 11 small dog chihuahua 3 ## 12 small dog dachshund 2 ## 13 small dog papillon 8 ## 14 small dig shetland sheepdog NA |
Dplyr
dplyr provides a flexible grammar of data manipulation. It’s the next iteration of plyr, focused on tools for working with data frames (hence the d in the name).
df <- iris ## dplyr approach: better formatting, and adapts to your screen width glimpse(df) |
## Observations: 150 ## Variables: 5 ## $ Sepal.Length <dbl> 5.1, 4.9, 4.7, 4.6, 5.0, 5.4, 4.6, 5.0, 4.4, 4.9,... ## $ Sepal.Width <dbl> 3.5, 3.0, 3.2, 3.1, 3.6, 3.9, 3.4, 3.4, 2.9, 3.1,... ## $ Petal.Length <dbl> 1.4, 1.4, 1.3, 1.5, 1.4, 1.7, 1.4, 1.5, 1.4, 1.5,... ## $ Petal.Width <dbl> 0.2, 0.2, 0.2, 0.2, 0.2, 0.4, 0.3, 0.2, 0.2, 0.1,... ## $ Species <fct> setosa, setosa, setosa, setosa, setosa, setosa, s... |
Ordering data
To reverse the direction of a particular column, the method depends on the data type:
Numbers: put a
-in front of the variable name, e.g.df[ order(-df$weight), ].Factors: convert to integer and put a
-in front of the variable name, e.g.df[ order(-xtfrm(df$size)), ].Characters: there isn’t a simple way to do this. One method is to convert to a factor first and then sort as above.
Note: all dplyr methods ignore rownames
## Sort by length, then by width arrange(df, Sepal.Length,Sepal.Width) # Use arrange from dplyr package df[order(df$Sepal.Length,df$Sepal.Width), ] # Use built-in R functions ## Sort by all columns in the data frame, from left to right df[ do.call(order, as.list(df)), ] ## Reverse sort by weight column. These all have the same effect: arrange(df, -Sepal.Length) df[order(df$Sepal.Length, decreasing=TRUE), ] df[order(-df$Sepal.Length), ] ## Sort by length, then by width (decreasing) arrange(df, Sepal.Length,-Sepal.Width) # Use arrange from plyr package df[ order(df$Sepal.Length,-df$Sepal.Width), ] ## Sort by Species (decreasing), then by Sepal.Length (increasing) ## The call to xtfrm() which is needed for factors arrange(df, -xtfrm(Species), Sepal.Length) # Use arrange from plyr package df[ order(-xtfrm(df$Species), df$Sepal.Length), ] # Use built-in R functions |
Recoding data
The easiest way is to use revalue() or mapvalues() in plyr
Alternative, recode in dplyr
cut in R will split the data to specified groups
Note: For numeric vectors, revalue() won’t work, since it uses a named vector, and the names are always strings, not numbers. mapvalues() will work, though.
library(plyr) library(dplyr) |
# The following two groups are equivalent: df$flora <- recode(df$Species, "setosa"=1, "versicolor"=2, "virginica"=3) df$flora[df$Species=="setosa"] <- "1" df$flora[df$Species=="versicolor"] <- "2" df$flora[df$Species=="virginica"] <- "3" # flora the column to a factor df$flora <- factor(df$flora) df$flora <- revalue(df$Species, c("setosa"="1", "versicolor"="2", "virginica"="3")) df$flora <- mapvalues(df$Species, from = c("setosa", "versicolor", "virginica"), to = c("1", "2", "3")) oldvalues <- c("setosa", "versicolor", "virginica") newvalues <- factor(c("1", "2", "3")) # Make this a factor df$flora <- newvalues[ match(df$Species, oldvalues) ] |
## # A tibble: 10 x 7 ## rowname Sepal.Length Sepal.Width Petal.Length Petal.Width Species flora ## * <fct> <dbl> <dbl> <dbl> <dbl> <fct> <fct> ## 1 1 5.10 3.50 1.40 0.200 setosa 1 ## 2 2 4.90 3.00 1.40 0.200 setosa 1 ## 3 3 4.70 3.20 1.30 0.200 setosa 1 ## 4 4 4.60 3.10 1.50 0.200 setosa 1 ## 5 5 5.00 3.60 1.40 0.200 setosa 1 ## 6 146 6.70 3.00 5.20 2.30 virgin~ 3 ## 7 147 6.30 2.50 5.00 1.90 virgin~ 3 ## 8 148 6.50 3.00 5.20 2.00 virgin~ 3 ## 9 149 6.20 3.40 5.40 2.30 virgin~ 3 ## 10 150 5.90 3.00 5.10 1.80 virgin~ 3 |
When using cut, by default, the ranges are open on the left, and closed on the right, as in (7,9].
To set it so that ranges are closed on the left and open on the right, like [7,9), use right=FALSE.
df$category <- cut(df$Sepal.Length, breaks=c(-Inf, 5, 7, Inf), labels=c("low","medium","high")) |
rename in dplyr and plyr
# rename the columns in a data frame. # names(df)[names(df)=="."] <- ".." df <- rename(df, c("Sepal.Length"="S.Length", "Sepal.Width"="S.Width","Petal.Length"="P.Length", "Petal.Width"="P.Width")) |
## # A tibble: 10 x 8 ## rowname S.Length S.Width P.Length P.Width Species flora category ## * <fct> <dbl> <dbl> <dbl> <dbl> <fct> <fct> <fct> ## 1 1 5.10 3.50 1.40 0.200 setosa 1 medium ## 2 2 4.90 3.00 1.40 0.200 setosa 1 low ## 3 3 4.70 3.20 1.30 0.200 setosa 1 low ## 4 4 4.60 3.10 1.50 0.200 setosa 1 low ## 5 5 5.00 3.60 1.40 0.200 setosa 1 low ## 6 146 6.70 3.00 5.20 2.30 virginica 3 medium ## 7 147 6.30 2.50 5.00 1.90 virginica 3 medium ## 8 148 6.50 3.00 5.20 2.00 virginica 3 medium ## 9 149 6.20 3.40 5.40 2.30 virginica 3 medium ## 10 150 5.90 3.00 5.10 1.80 virginica 3 medium |
Select & Filter & mutate
names(df) # [1] "S.Length" "S.Width" "P.Length" "P.Width" "Species" # [6] "flora" "category" ## using select method df %>% select(S.Length, P.Length:P.Width, contains("flora"), -starts_with("P.Length"), -S.Length, ends_with("ory"), matches("S\\.W.*")) %>% head # P.Width flora category S.Width # 1 0.2 1 medium 3.5 # 2 0.2 1 low 3.0 # 3 0.2 1 low 3.2 # 4 0.2 1 low 3.1 # 5 0.2 1 low 3.6 # 6 0.4 1 medium 3.9 ## using filter method df %>% filter(category=="medium" & as.numeric(flora) >1) %>% head(.,5) # S.Length S.Width P.Length P.Width Species flora category # 1 7.0 3.2 4.7 1.4 versicolor 2 medium # 2 6.4 3.2 4.5 1.5 versicolor 2 medium # 3 6.9 3.1 4.9 1.5 versicolor 2 medium # 4 5.5 2.3 4.0 1.3 versicolor 2 medium # 5 6.5 2.8 4.6 1.5 versicolor 2 medium |
Using mutate
df %>% select(S.Width, flora) %>% arrange(desc(flora)) %>% mutate(mesc = paste0(S.Width,"-",flora)) %>% separate(mesc, into = c("Sepal.Width", "Flora"), sep = "-") %>% # separate in tidyr unite("mesc", sep = "-", Sepal.Width, Flora) %>% head(., 5) |
## S.Width flora mesc ## 1 3.3 3 3.3-3 ## 2 2.7 3 2.7-3 ## 3 3.0 3 3-3 ## 4 2.9 3 2.9-3 ## 5 3.0 3 3-3 |
Summarise
Basic
df %>% group_by(Species) %>% select(matches("S.Length")) %>% summarise_all(funs(min(., na.rm=TRUE), Q1 = quantile(., probs = 0.25), med = median, Q3 = quantile(., probs = 0.75), max(., na.rm=TRUE), mean(., na.rm=TRUE), counts = n())) |
## # A tibble: 3 x 8 ## Species min Q1 med Q3 max mean counts ## <fct> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <int> ## 1 setosa 4.30 4.80 5.00 5.20 5.80 5.01 50 ## 2 versicolor 4.90 5.60 5.90 6.30 7.00 5.94 50 ## 3 virginica 4.90 6.22 6.50 6.90 7.90 6.59 50 |
Note:
Lead and lag : Find the “next” or “previous” values in a vector. Useful for comparing values ahead of or behind the current values.
df %>% group_by(category) %>% tally %>% mutate(change = n- lag(n)) |
## # A tibble: 3 x 3 ## category n change ## <fct> <int> <int> ## 1 low 32 NA ## 2 medium 106 74 ## 3 high 12 -94 |
# df %>% # group_by(category) %>% # summarise(n=n()) %>% # mutate(change=n - lag(n)) |
df %>% group_by(category) %>% select(P.Length, P.Width, Species) %>% filter(min_rank(desc(P.Length)) <= 3) %>% ##== top_n(3) arrange(category, desc(P.Length)) |
## # A tibble: 11 x 4 ## # Groups: category [3] ## category P.Length P.Width Species ## <fct> <dbl> <dbl> <fct> ## 1 low 4.50 1.70 virginica ## 2 low 3.50 1.00 versicolor ## 3 low 3.30 1.00 versicolor ## 4 low 3.30 1.00 versicolor ## 5 medium 6.00 2.50 virginica ## 6 medium 5.90 2.30 virginica ## 7 medium 5.80 2.20 virginica ## 8 medium 5.80 1.80 virginica ## 9 high 6.90 2.30 virginica ## 10 high 6.70 2.20 virginica ## 11 high 6.70 2.00 virginica |
## randomly sample a fixed number of rows, without replacement set.seed(0) df %>% sample_n(5) |
## S.Length S.Width P.Length P.Width Species flora category ## 135 6.1 2.6 5.6 1.4 virginica 3 medium ## 40 5.1 3.4 1.5 0.2 setosa 1 medium ## 56 5.7 2.8 4.5 1.3 versicolor 2 medium ## 85 5.4 3.0 4.5 1.5 versicolor 2 medium ## 133 6.4 2.8 5.6 2.2 virginica 3 medium |
## randomly sample a fraction of rows, with replacement df %>% sample_frac(0.05, replace=TRUE) |
## S.Length S.Width P.Length P.Width Species flora category ## 31 4.8 3.1 1.6 0.2 setosa 1 low ## 135 6.1 2.6 5.6 1.4 virginica 3 medium ## 142 6.9 3.1 5.1 2.3 virginica 3 medium ## 100 5.7 2.8 4.1 1.3 versicolor 2 medium ## 95 5.6 2.7 4.2 1.3 versicolor 2 medium ## 10 4.9 3.1 1.5 0.1 setosa 1 low ## 31.1 4.8 3.1 1.6 0.2 setosa 1 low ## 27 5.0 3.4 1.6 0.4 setosa 1 low |
Putting some NAs
## Using ddply in plyr ddply(df, c("Species", "category"), summarise, N = length(S.Length), mean = mean(S.Length), sd = sd(S.Length), se = sd / sqrt(N) ) |
## Species category N mean sd se ## 1 setosa low 28 4.764286 0.22146697 0.04185332 ## 2 setosa medium 22 5.313636 0.22317077 0.04758017 ## 3 versicolor low 3 4.966667 0.05773503 0.03333333 ## 4 versicolor medium 47 5.997872 0.46741163 0.06817899 ## 5 virginica low 1 4.900000 NA NA ## 6 virginica medium 37 6.345946 0.35558605 0.05845799 ## 7 virginica high 12 7.475000 0.27010099 0.07797144 |
## Using dlpyr df %>% group_by(Species, category) %>% select(S.Length) %>% summarise_all(funs(N = n(), mean = mean(.), sd = sd(.), se = sd / sqrt(N))) |
## # A tibble: 7 x 6 ## # Groups: Species [?] ## Species category N mean sd se ## <fct> <fct> <int> <dbl> <dbl> <dbl> ## 1 setosa low 28 4.76 0.221 0.0419 ## 2 setosa medium 22 5.31 0.223 0.0476 ## 3 versicolor low 3 4.97 0.0577 0.0333 ## 4 versicolor medium 47 6.00 0.467 0.0682 ## 5 virginica low 1 4.90 NaN NaN ## 6 virginica medium 37 6.35 0.356 0.0585 ## 7 virginica high 12 7.48 0.270 0.0780 |
dfNA <- df dfNA$S.Length[11:20] <- NA ## Using ddply in plyr ddply(dfNA, c("Species", "category"), summarise, N = sum(!is.na(S.Length)), mean = mean(S.Length, na.rm=TRUE), sd = sd(S.Length, na.rm=TRUE), se = sd / sqrt(N) ) |
## Species category N mean sd se ## 1 setosa low 25 4.780000 0.21408721 0.04281744 ## 2 setosa medium 15 5.246667 0.15522641 0.04007929 ## 3 versicolor low 3 4.966667 0.05773503 0.03333333 ## 4 versicolor medium 47 5.997872 0.46741163 0.06817899 ## 5 virginica low 1 4.900000 NA NA ## 6 virginica medium 37 6.345946 0.35558605 0.05845799 ## 7 virginica high 12 7.475000 0.27010099 0.07797144 |
ddply(dfNA, c("Species", "category"), .fun = function(., col) { c(N = sum(!is.na(.[[col]])), mean = mean (.[[col]], na.rm=T), sd = sd (.[[col]], na.rm=T) ) }, "S.Length" ) |
## Species category N mean sd ## 1 setosa low 25 4.780000 0.21408721 ## 2 setosa medium 15 5.246667 0.15522641 ## 3 versicolor low 3 4.966667 0.05773503 ## 4 versicolor medium 47 5.997872 0.46741163 ## 5 virginica low 1 4.900000 NA ## 6 virginica medium 37 6.345946 0.35558605 ## 7 virginica high 12 7.475000 0.27010099 |
## Using dlpyr dfNA %>% filter(!is.na(S.Length)) %>% group_by(Species, category) %>% select(S.Length) %>% summarise_all(funs(N = n(), mean = mean(.), sd = sd(.), se = sd / sqrt(N))) |
## # A tibble: 7 x 6 ## # Groups: Species [?] ## Species category N mean sd se ## <fct> <fct> <int> <dbl> <dbl> <dbl> ## 1 setosa low 25 4.78 0.214 0.0428 ## 2 setosa medium 15 5.25 0.155 0.0401 ## 3 versicolor low 3 4.97 0.0577 0.0333 ## 4 versicolor medium 47 6.00 0.467 0.0682 ## 5 virginica low 1 4.90 NaN NaN ## 6 virginica medium 37 6.35 0.356 0.0585 ## 7 virginica high 12 7.48 0.270 0.0780 |
Keep rownames in dplyr
name_rows : provides a way to preserve them by converting them to an explicit column in the data fram
df %>% name_rows %>% filter(category=="medium", S.Width>=4) |
## S.Length S.Width P.Length P.Width Species flora category .rownames ## 1 5.8 4.0 1.2 0.2 setosa 1 medium 15 ## 2 5.7 4.4 1.5 0.4 setosa 1 medium 16 ## 3 5.2 4.1 1.5 0.1 setosa 1 medium 33 ## 4 5.5 4.2 1.4 0.2 setosa 1 medium 34 |
Pipelining(chaining)
The Treachery of Images - Ceci n’est pas une pipe
library(magrittr) library(babynames) |
iris %>% { n <- 5 rbind(head(., n), tail(., n)) } %T>% plot %>% summary |
## Sepal.Length Sepal.Width Petal.Length Petal.Width ## Min. :4.600 Min. :2.50 Min. :1.300 Min. :0.200 ## 1st Qu.:4.925 1st Qu.:3.00 1st Qu.:1.400 1st Qu.:0.200 ## Median :5.500 Median :3.05 Median :3.250 Median :1.000 ## Mean :5.590 Mean :3.13 Mean :3.290 Mean :1.130 ## 3rd Qu.:6.275 3rd Qu.:3.35 3rd Qu.:5.175 3rd Qu.:1.975 ## Max. :6.700 Max. :3.60 Max. :5.400 Max. :2.300 ## Species ## setosa :5 ## versicolor:0 ## virginica :5 ## ## ## |
iris %>% subset(Sepal.Length > mean(Sepal.Length)) %$% cor(Sepal.Length, Sepal.Width) |
## [1] 0.3361992 |
# iris %>% # subset(Sepal.Length > mean(Sepal.Length)) %>% # select(Sepal.Length, Sepal.Width) %>% # cor iris %>% subset(Sepal.Length > mean(Sepal.Length)) %$% add(Sepal.Length, Sepal.Width) %>% head |
## [1] 10.2 9.6 10.0 9.3 9.6 9.5 |
iris$Sepal.Length %<>% sqrt # iris$Sepal.Length <- # iris$Sepal.Length %>% # sqrt() |
dat <- filter(babynames, substr(babynames$name, 1, 2)=="Sa") # # A tibble: 6 x 5 # year sex name n prop # <dbl> <chr> <chr> <int> <dbl> # 1 1880 F Sarah 1288 0.0132 # 2 1880 F Sallie 404 0.00414 # 3 1880 F Sadie 317 0.00325 # 4 1880 F Sara 165 0.00169 # 5 1880 F Sally 80 0.000820 # 6 1880 F Samantha 21 0.000215 dim(dat) # 27093 5 |
| extract | [ |
|---|---|
| extract2 | [[ |
| inset | [<- |
| inset2 | [[<- |
| use_series | $ |
| add | + |
| subtract | - |
| multiply_by | * |
| raise_to_power | ^ |
| multiply_by_matrix | %*% |
| divide_by | / |
| divide_by_int | %/% |
| mod | %% |
| is_in | %in% |
| and | & |
| or | | |
| equals | == |
| is_greater_than | > |
| is_weakly_greater_than | >= |
| is_less_than | < |
| is_weakly_less_than | <= |
not (n'est pas) |
! |
| set_colnames | colnames<- |
| set_rownames | rownames<- |
| set_names | names<- |
## qplot # babynames %>% # filter(name %>% substr(1, 2) %>% equals("Sa")) %>% # group_by(year, sex) %>% # select(n) %>% # summarise_all(funs(total = sum(n)))%>% # qplot(year, total, color = sex, data=., geom="line") %>% # add(ggtitle("Names starting with Sa by pipe")) %>% # add(theme_minimal()) %>% # print babynames %>% filter(name %>% substr(1, 2) %>% equals("Sa")) %>% group_by(year,sex) %>% select(n) %>% summarise_all(funs(total = sum(n)))%>% {ggplot(.) + geom_line(aes(year, total, color = sex))} %>% add(ggtitle("Names starting with Sa by pipe")) %>% add(theme_minimal()) %>% print |
All articles in this blog are licensed under CC BY-NC-SA 4.0 unless stating additionally.
Related Articles
Comment
