I do this a lot:
library(tidyverse)
iris %>%
group_by(Species) %>%
summarise(num_Species = n_distinct(Species)) %>%
mutate(perc_Species = 100 * num_Species / sum(num_Species))
So I would like to create a function that outputs the same thing but with dynamically named num_ and perc_ columns:
num_perc <- function(df, group_var, summary_var) {
}
I found this resource useful but it did not directly address how to reuse newly created column names in the way I want.
What you can do is use as_label(enquo()) on your group_var to extract variable passed as a character vector to generate your new columns. You can see a clear example of this is 6.1.3 in the linked document you sent. In this way, we can dynamically prepend num_ and perc_ to your summary variable, and just have to pass in df and group_var.
library(dplyr)
num_perc <- function(df, group_var) {
summary_lbl <- as_label(enquo(group_var))
num_lbl <- paste0("num_", summary_lbl)
perc_lbl <- paste0("perc_", summary_lbl)
df %>%
group_by({{ group_var }}) %>%
summarize(!!num_lbl := n_distinct({{ group_var }})) %>%
mutate(!!perc_lbl := 100 * .data[[num_lbl]] / sum(.data[[num_lbl]]))
}
num_perc(iris, Species)
#> # A tibble: 3 × 3
#> Species num_Species perc_Species
#> <fct> <int> <dbl>
#> 1 setosa 1 33.3
#> 2 versicolor 1 33.3
#> 3 virginica 1 33.3
In this case where group_var and summary_var actually differ, it's the same solution essentially.
num_perc <- function(df, group_var, summary_var) {
summary_lbl <- as_label(enquo(summary_var))
num_lbl <- paste0("num_", summary_lbl)
perc_lbl <- paste0("perc_", summary_lbl)
df %>%
group_by({{ group_var }}) %>%
summarize(!!num_lbl := n_distinct({{ summary_var }})) %>%
mutate(!!perc_lbl := 100 * .data[[num_lbl]] / sum(.data[[num_lbl]]))
}
num_perc(iris, Species, Species)
Another possible solution, which uses deparse(substitute(...)) to get the name of the function parameters as strings:
library(tidyverse)
f <- function(df, group_var, summary_var)
{
group_var <- deparse(substitute(group_var))
summary_var <- deparse(substitute(summary_var))
df %>%
group_by(!!sym(group_var)) %>%
summarise(!!str_c("num_", summary_var) := n_distinct(summary_var)) %>%
mutate(!!str_c("per_", summary_var) := 100 * !!sym(str_c("num_", summary_var)) / sum(!!sym(str_c("num_", summary_var))))
}
f(iris, Species, Species)
#> # A tibble: 3 × 3
#> Species num_Species per_Species
#> <fct> <int> <dbl>
#> 1 setosa 1 33.3
#> 2 versicolor 1 33.3
#> 3 virginica 1 33.3
Are you sure n_distinct is what you want to do? In the case of the iris dataset, there are three Species - setosa, versicolor, virginica. Therefore, each species is 1/3 unique species. The Iris dataset is balanced in the sense that there are 50 of each species, so each species represents 1/3 of the data set but more generally this will not be the case.
A function with data masking will cover imbalanced datasets for you:
library(dplyr)
my_func <- function(df, var, percent){
df %>%
count({{var}}) %>%
mutate(percent = 100 * n/sum(n))
}
my_func(iris, Species, percent)
iris %>%
my_func(Species, percent) #or with pipe
Related
Consider this code:
iris %>% count(Species) %>% group_by(Species)
# A tibble: 3 x 2
# Groups: Species [3]
Species n
<fct> <int>
1 setosa 50
2 versicolor 50
3 virginica 50
I want to define a function which does the same task, something like this :
table_freq <- function(Table, Var) {
freq <- NA
freq <- Table %>%
dplyr::count(Var) %>%
group_by(Var)
return(freq)
}
table_freq(iris, "Species")
But it does not work :
> table_freq(iris, "Species")
Error in `group_by_prepare()`:
! Must group by variables found in `.data`.
* Column `Var` is not found.
Any ideas?
Please do not write alternate solutions, I need to define a function that takes the table and the column name for which we need the freq. table.
You can use table to create a function that will take the table and the column name.
table_freq <- function(Table, Var) {
setNames(data.frame(table(Table[,Var])), c(Var, "n"))
}
table_freq(iris, "Species")
Output
Species n
1 setosa 50
2 versicolor 50
3 virginica 50
The secret sauce is {{}}, we simply write :
table_freq <- function(Table, Var) {
freq <- NA
freq <- Table %>%
dplyr::count({{Var}}) %>%
group_by({{Var}})
return(freq)
}
table_freq(iris, Species)
I have a parent dataset nesting multiple datasets (i.e. a tibble where each cell is a tibble) , where I want for each dataset, to find the number of rows of each group. Standard way, using a single dataset, would simply be to do group_by(var) %>% mutate(nrow=n()).
But now that I do this for multiple datasets with a map() call, it looks like the n() call refers to the (implicit) grouping made by map(), not the explicit grouping within my local dataset made by group_by?
Standard way for one single dataset, n() returns 50:
iris %>%
group_by(., Species) %>%
mutate(nrow=n())
Dataset of datasets:
df <- data_frame(name=c("a", "b"), Data=list(iris, iris))
df2 <- df %>%
mutate(Data2=map(Data, ~group_by(., Species) %>%
mutate(nrow=n()) %>%
ungroup()))
but now n() returned 2?
df2[1,]$Data2[[1]]
If you define the function outside of mutate it works fine (I assume this output is what you have in mind...)
fun <- function(x) {
df <- group_by(x, Species) %>%
summarise(nrow = n())
}
df2 <- df %>%
mutate(Data2=map(Data, fun))
df2$Data2
# [[1]]
# # A tibble: 3 x 2
# Species nrow
# <fctr> <int>
# 1 setosa 50
# 2 versicolor 50
# 3 virginica 50
#
# [[2]]
# # A tibble: 3 x 2
# Species nrow
# <fctr> <int>
# 1 setosa 50
# 2 versicolor 50
# 3 virginica 50
Another option, available since version 0.7.0 is to use add_count(), which will not conflict with the map(), and anyway simplifies the code:
# standard case:
iris %>%
add_count(Species)
## df of df:
df2 <- df %>%
mutate(Data2=map(Data, ~add_count(., Species)))
I have a data frame called "Something". I am doing an aggregation on one of the numeric columns using summarise, and I want the name of that column to contain "Something" - data frame title in the column name.
Example:
temp <- Something %>%
group_by(Month) %>%
summarise(avg_score=mean(score))
But i would like to name the aggregate column as "avg_Something_score". Did that make sense?
We can use the devel version of dplyr (soon to be released 0.6.0) that does this with quosures
library(dplyr)
myFun <- function(data, group, value){
dataN <- quo_name(enquo(data))
group <- enquo(group)
value <- enquo(value)
newName <- paste0("avg_", dataN, "_", quo_name(value))
data %>%
group_by(!!group) %>%
summarise(!!newName := mean(!!value))
}
myFun(mtcars, cyl, mpg)
# A tibble: 3 × 2
# cyl avg_mtcars_mpg
# <dbl> <dbl>
#1 4 26.66364
#2 6 19.74286
#3 8 15.10000
myFun(iris, Species, Petal.Width)
# A tibble: 3 × 2
# Species avg_iris_Petal.Width
# <fctr> <dbl>
#1 setosa 0.246
#2 versicolor 1.326
#3 virginica 2.026
Here, the enquo takes the input arguments like substitute from base R and converts to quosure, with quo_name, we can convert it to string, evaluate the quosure by unquoting (!! or UQ) inside group_by/summarise/mutate etc. The column names on the lhs of assignment (:=) can also evaluated by unquoting to get the columns of interest
You can use rename_ from dplyr with deparse(substitute(Something)) like this:
Something %>%
group_by(Month) %>%
summarise(avg_score=mean(score))%>%
rename_(.dots = setNames("avg_score",
paste0("avg_",deparse(substitute(Something)),"_score") ))
It seems like it makes more sense to generate the new column name dynamically so that you don't have to hard-code the name of the data frame inside setNames. Maybe something like the function below, which takes a data frame, a grouping variable, and a numeric variable:
library(dplyr)
library(lazyeval)
my_fnc = function(data, group, value) {
df.name = deparse(substitute(data))
data %>%
group_by_(group) %>%
summarise_(avg = interp(~mean(v), v=as.name(value))) %>%
rename_(.dots = setNames("avg", paste0("avg_", df.name, "_", value)))
}
Now let's run the function on two different data frames:
my_fnc(mtcars, "cyl", "mpg")
cyl avg_mtcars_mpg
<dbl> <dbl>
1 4 26.66364
2 6 19.74286
3 8 15.10000
my_fnc(iris, "Species", "Petal.Width")
Species avg_iris_Petal.Width
1 setosa 0.246
2 versicolor 1.326
3 virginica 2.026
library(dplyr)
# Take mtcars as an example
# Calculate the mean of mpg using cyl as group
data(mtcars)
Something <- mtcars
# Create a list of expression
dots <- list(~mean(mpg))
# Apply the function, Use setNames to name the column
temp <- Something %>%
group_by(cyl) %>%
summarise_(.dots = setNames(dots,
paste0("avg_", as.character(quote(Something)), "_score")))
You could use colnames(Something)<-c("score","something_avg_score")
I want to use use the dplyr::group_by function inside another function, but I do not know how to pass the arguments to this function.
Can someone provide a working example?
library(dplyr)
data(iris)
iris %.% group_by(Species) %.% summarise(n = n()) #
## Source: local data frame [3 x 2]
## Species n
## 1 virginica 50
## 2 versicolor 50
## 3 setosa 50
mytable0 <- function(x, ...) x %.% group_by(...) %.% summarise(n = n())
mytable0(iris, "Species") # OK
## Source: local data frame [3 x 2]
## Species n
## 1 virginica 50
## 2 versicolor 50
## 3 setosa 50
mytable1 <- function(x, key) x %.% group_by(as.name(key)) %.% summarise(n = n())
mytable1(iris, "Species") # Wrong!
# Error: unsupported type for column 'as.name(key)' (SYMSXP)
mytable2 <- function(x, key) x %.% group_by(key) %.% summarise(n = n())
mytable2(iris, "Species") # Wrong!
# Error: index out of bounds
For programming, group_by_ is the counterpart to group_by:
library(dplyr)
mytable <- function(x, ...) x %>% group_by_(...) %>% summarise(n = n())
mytable(iris, "Species")
# or iris %>% mytable("Species")
which gives:
Species n
1 setosa 50
2 versicolor 50
3 virginica 50
Update At the time this was written dplyr used %.% which is what was originally used above but now %>% is favored so have changed above to that to keep this relevant.
Update 2 regroup is now deprecated, use group_by_ instead.
Update 3 group_by_(list(...)) now becomes group_by_(...) in new version of dplyr as per Roberto's comment.
Update 4 Added minor variation suggested in comments.
Update 5: With rlang/tidyeval it is now possible to do this:
library(rlang)
mytable <- function(x, ...) {
group_ <- syms(...)
x %>%
group_by(!!!group_) %>%
summarise(n = n())
}
mytable(iris, "Species")
or passing Species unevaluated, i.e. no quotes around it:
library(rlang)
mytable <- function(x, ...) {
group_ <- enquos(...)
x %>%
group_by(!!!group_) %>%
summarise(n = n())
}
mytable(iris, Species)
Update 6: There is now a {{...}} notation that works if there is just one grouping variable:
mytable <- function(x, group) {
x %>%
group_by({{group}}) %>%
summarise(n = n())
}
mytable(iris, Species)
UPDATE: As of dplyr 0.7.0 you can use tidy eval to accomplish this.
See http://dplyr.tidyverse.org/articles/programming.html for more details.
library(tidyverse)
data("iris")
my_table <- function(df, group_var) {
group_var <- enquo(group_var) # Create quosure
df %>%
group_by(!!group_var) %>% # Use !! to unquote the quosure
summarise(n = n())
}
my_table(iris, Species)
> my_table(iris, Species)
# A tibble: 3 x 2
Species n
<fctr> <int>
1 setosa 50
2 versicolor 50
3 virginica 50
As a complement to the Update 6 in the answer by #G. Grothendieck, if you want to use a string as an argument in your summary function, instead of embracing the argument with doubled braces ({{), you should use the .data pronoun as described in the Programming vignette: Loop over multiple variables:
mytable <- function( x, group ) {
x %>%
group_by( .data[[group]] ) %>%
summarise( n = n() )
}
group_string <- 'Species'
mytable( iris, group_string )
`summarise()` ungrouping output (override with `.groups` argument)
# A tibble: 3 x 2
Species n
<fct> <int>
1 setosa 50
2 versicolor 50
3 virginica 50
Ugly as they come, but she works:
mytable3 <- function(x, key) {
my.call <- bquote(summarise(group_by(.(substitute(x)), NULL), n = n()))
my.call[[2]][[3]] <- as.name(key)
eval(my.call, parent.frame())
}
mytable3(iris, "Species")
# Source: local data frame [3 x 2]
#
# Species n
# 1 virginica 50
# 2 versicolor 50
# 3 setosa 50
There are almost certainly cases that will cause this to break, but you get the idea. I don't think you can get around messing with the call. One other thing that did work but was even uglier is:
mytable4 <- function(x, key) summarise(group_by(x, x[[key]]), n = n())
I want to use use the dplyr::group_by function inside another function, but I do not know how to pass the arguments to this function.
Can someone provide a working example?
library(dplyr)
data(iris)
iris %.% group_by(Species) %.% summarise(n = n()) #
## Source: local data frame [3 x 2]
## Species n
## 1 virginica 50
## 2 versicolor 50
## 3 setosa 50
mytable0 <- function(x, ...) x %.% group_by(...) %.% summarise(n = n())
mytable0(iris, "Species") # OK
## Source: local data frame [3 x 2]
## Species n
## 1 virginica 50
## 2 versicolor 50
## 3 setosa 50
mytable1 <- function(x, key) x %.% group_by(as.name(key)) %.% summarise(n = n())
mytable1(iris, "Species") # Wrong!
# Error: unsupported type for column 'as.name(key)' (SYMSXP)
mytable2 <- function(x, key) x %.% group_by(key) %.% summarise(n = n())
mytable2(iris, "Species") # Wrong!
# Error: index out of bounds
For programming, group_by_ is the counterpart to group_by:
library(dplyr)
mytable <- function(x, ...) x %>% group_by_(...) %>% summarise(n = n())
mytable(iris, "Species")
# or iris %>% mytable("Species")
which gives:
Species n
1 setosa 50
2 versicolor 50
3 virginica 50
Update At the time this was written dplyr used %.% which is what was originally used above but now %>% is favored so have changed above to that to keep this relevant.
Update 2 regroup is now deprecated, use group_by_ instead.
Update 3 group_by_(list(...)) now becomes group_by_(...) in new version of dplyr as per Roberto's comment.
Update 4 Added minor variation suggested in comments.
Update 5: With rlang/tidyeval it is now possible to do this:
library(rlang)
mytable <- function(x, ...) {
group_ <- syms(...)
x %>%
group_by(!!!group_) %>%
summarise(n = n())
}
mytable(iris, "Species")
or passing Species unevaluated, i.e. no quotes around it:
library(rlang)
mytable <- function(x, ...) {
group_ <- enquos(...)
x %>%
group_by(!!!group_) %>%
summarise(n = n())
}
mytable(iris, Species)
Update 6: There is now a {{...}} notation that works if there is just one grouping variable:
mytable <- function(x, group) {
x %>%
group_by({{group}}) %>%
summarise(n = n())
}
mytable(iris, Species)
UPDATE: As of dplyr 0.7.0 you can use tidy eval to accomplish this.
See http://dplyr.tidyverse.org/articles/programming.html for more details.
library(tidyverse)
data("iris")
my_table <- function(df, group_var) {
group_var <- enquo(group_var) # Create quosure
df %>%
group_by(!!group_var) %>% # Use !! to unquote the quosure
summarise(n = n())
}
my_table(iris, Species)
> my_table(iris, Species)
# A tibble: 3 x 2
Species n
<fctr> <int>
1 setosa 50
2 versicolor 50
3 virginica 50
As a complement to the Update 6 in the answer by #G. Grothendieck, if you want to use a string as an argument in your summary function, instead of embracing the argument with doubled braces ({{), you should use the .data pronoun as described in the Programming vignette: Loop over multiple variables:
mytable <- function( x, group ) {
x %>%
group_by( .data[[group]] ) %>%
summarise( n = n() )
}
group_string <- 'Species'
mytable( iris, group_string )
`summarise()` ungrouping output (override with `.groups` argument)
# A tibble: 3 x 2
Species n
<fct> <int>
1 setosa 50
2 versicolor 50
3 virginica 50
Ugly as they come, but she works:
mytable3 <- function(x, key) {
my.call <- bquote(summarise(group_by(.(substitute(x)), NULL), n = n()))
my.call[[2]][[3]] <- as.name(key)
eval(my.call, parent.frame())
}
mytable3(iris, "Species")
# Source: local data frame [3 x 2]
#
# Species n
# 1 virginica 50
# 2 versicolor 50
# 3 setosa 50
There are almost certainly cases that will cause this to break, but you get the idea. I don't think you can get around messing with the call. One other thing that did work but was even uglier is:
mytable4 <- function(x, key) summarise(group_by(x, x[[key]]), n = n())