检查R中数据帧中不同列的数量

Question

I am using R and have a Dataset where each column is a production unit and each row is a time unit. Each variable is a crop rotation sequence applied to the production unit.

The dataset looks like this:

land_use_1 land_use_2 land_use_3 land_use_4 land_use_5 land_use_6
 <chr>      <chr>      <chr>      <chr>      <chr>      <chr>     
1 PAST       PAST       PAST       PAST       SOY       PAST      
2 PAST       PAST       PAST       PAST       SOY       PAST      
3 PAST       PAST       PAST       PAST       PAST      PAST      
4 PAST       PAST       PAST       PAST       PAST      SOY       
5 PAST       PAST       PAST       PAST       CORN      SOY      
6 PAST       PAST       PAST       PAST       CORN      PAST      

I need to check how many of these columns (crop sequences) are unique, by i cannot do it one by one (doing something like land_use_1!=land_use_2, doing something like land_use_1!=land_use_3, etc) because there are hundreds of columns in the dataset.

I tried to use this command

dataset %>% unique(, MARGIN=2) %>% dim()

but it returns the same number of columns of the dataset and therefore doesnt detect which columns are identical (i know that some are identical, because i checked using some of them).

How can i do that in an efficient way?

Thanks

Answer 1

you can use data.table fuction duplicated :

library(data.table)
DT<-data.table(yourdataframe)

DT$duplicated_rows<- duplicate(DT)

here a reporoductible example:

DT <- data.table(A = rep(1:3, each=4), B = rep(1:4, each=3),
                  C = rep(1:2, 6), key = "A,B")
DT
    A B C
 1: 1 1 1
 2: 1 1 2
 3: 1 1 1
 4: 1 2 2
 5: 2 2 1
 6: 2 2 2
 7: 2 3 1
 8: 2 3 2
 9: 3 3 1
10: 3 4 2
11: 3 4 1
12: 3 4 2
duplicated(DT)
 [1] FALSE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE  TRUE
> 

Answer 2

Here's a solution to generate a list of names of the unique columns and the total number of unique columns:

library(tidyverse)

df <- data.frame(land_use_1 = rep("PAST", 6),
                 land_use_2 = rep("PAST", 6),
                 land_use_3 = rep("PAST", 6),
                 land_use_4 = rep("PAST", 6),
                 land_use_5 = c("SOY", "SOY", "PAST", "PAST", "CORN", "CORN"),
                 land_use_6 = c("PAST", "PAST", "PAST", "SOY", "SOY", "PAST"))

unique_vars <- data.frame(t(df)) %>%
  rownames_to_column() %>%
  distinct_at(vars(-rowname), .keep_all = T)

unique_vars$rowname
# [1] [1] "land_use_1" "land_use_5" "land_use_6"

length(unique_vars$rowname)
# [1] 3

Answer 3

Use:

 unique(as.list(dataset))

This coerces the dataframe into a list of columns, then counts the number of unique elements of the list.

eg:

> d <- data.frame(a=c(1,1,0) , b=c(1,1,0), c=c(1,0,1))
> unique(as.list(d))
[[1]]
[1] 1 1 0

[[2]]
[1] 1 0 1

> length(unique(as.list(d)))
[1] 2