dcasting a dataframe using reshape2::dcast-CodePudding

I am trying to reshape a dataframe using dcast function of reshape2 package

library(reshape2)
dat = data.frame(aa = c('A', 'B', 'C', 'C', 'B'))
dcast(dat, aa~aa)

This generates below output

However I wanted to keep the number of rows same as my original dataframe as below

Is there any direct function to get my desired shape?

Thanks for your time.

CodePudding user response：

With tidyverse or base R you can do:

dat = data.frame(aa = c('A', 'B', 'C', 'C', 'B'))
                  
library(tidyverse)
dat %>% 
  mutate(id = seq_along(aa), val = 1) %>% 
  tidyr::pivot_wider(names_from = aa, values_from = val, values_fill = 0) %>% 
  select(-id)
#> # A tibble: 5 × 3
#>       A     B     C
#>   <dbl> <dbl> <dbl>
#> 1     1     0     0
#> 2     0     1     0
#> 3     0     0     1
#> 4     0     0     1
#> 5     0     1     0

as.data.frame(sapply(unique(dat$aa), \(x) as.numeric(x == dat$aa)))
#>   A B C
#> 1 1 0 0
#> 2 0 1 0
#> 3 0 0 1
#> 4 0 0 1
#> 5 0 1 0

CodePudding user response：

You could add first an id column with row_number and spread the aa column to your desired output but with the columnames as values. To replace these you can replace everything across the columns if it is not 0 with 1 like this:

dat = data.frame(aa = c('A', 'B', 'C', 'C', 'B'))

library(dplyr)
library(tidyr)
dat %>% 
  mutate(id = row_number()) %>% 
  spread(aa, aa, fill = 0) %>% 
  select(-id) %>%
  mutate(across(everything(), ~ replace(., . != 0, 1)))
#>   A B C
#> 1 1 0 0
#> 2 0 1 0
#> 3 0 0 1
#> 4 0 0 1
#> 5 0 1 0

^{Created on 2022-12-26 with reprex v2.0.2}