Clever Geek Handbook

R - get the highest value for each identifier

I have the following df:

>> animals_df:

animal_name    age
cat             1
cat             1
cat             2
cat             3
cat             3
dog             1
dog             1
dog             3
dog             4
dog             4
dog             4
horse           1
horse           3
horse           5
horse           5
horse           5

I only want to hunt animals with the highest ages of each species. So I want to get the following output:

animal_name    age
    cat         3
    cat         3
    dog         4
    dog         4
    dog         4
    horse       5
    horse       5
    horse       5

I've tried using:

animals_df = do.call(rbind,lapply(split(animals_df, animals_df$animal_name), function(x) tail(x, 1) ) )

But this will only give one instance of each animal, which will be the following:

animals_name    age
    cat          3
    dog          4
    horse        5
+3


source to share


2 answers


It's easy with dplyr

/ tidyverse

:



library(tidyverse)

# How I read your data in, ignore since you already have your data available
df = read.table(file="clipboard", header=TRUE)
df %>%
    group_by(animal_name) %>%
    filter(age == max(age))

# Output:
Source: local data frame [8 x 2]
Groups: animal_name [3]

  animal_name   age
       <fctr> <int>
1         cat     3
2         cat     3
3         dog     4
4         dog     4
5         dog     4
6       horse     5
7       horse     5
8       horse     5
+4


source


Another parameter data.table

:



library(data.table)
setDT(df)
df[, .SD[which(age == max(age))], by = animal_name]

#       animal_name age
#1:         cat   3
#2:         cat   3
#3:         dog   4
#4:         dog   4
#5:         dog   4
#6:       horse   5
#7:       horse   5
#8:       horse   5
+1


source






More articles:


All Articles