Mutate column as input to sample

问题

I want create a distribution using the sample() function with the probability of each value defined by a data.frame() column. When I try the code below however it produces the error:

Error in sample.int(length(x), size, replace, prob) : 
  incorrect number of probabilities

I'm guessing this is because mutate is passing the entire column and not just one integer. Can anyone help me with this?

library(dplyr)

input = data.frame(input = 1:100)
output = input %>% mutate(output = 
                            sum(
                              sample(0:1, 
                                     10, 
                                     replace = T,
                                     prob = c(input, 100-input)
                                     )))

回答1:

I have determined that the solution was to use the rowwise() function in the pipe expression:

library(dplyr)

input = data.frame(input = 1:100)
output = input %>% rowwise() %>% mutate(output = 
                            sum(
                              sample(0:1, 
                                     100, 
                                     replace = T,
                                     prob = c(input, 1)
                                     )))

来源：https://stackoverflow.com/questions/57617305/mutate-column-as-input-to-sample