Question

我想将数据表中的列从因子转换为数值，但是我正在使用“字典”（列表），所以我希望能够通过“字典”引用列变量“：

#my "dictionary"
outcomes <- list("heart attack"="Hospital 30-Day Death (Mortality) Rates from Heart Attack",
               "heart failure"="Number of Patients - Hospital 30-Day Death (Mortality) Rates from Heart Failure",
               "pneumonia"="Number of Patients - Hospital 30-Day Death (Mortality) Rates from Pneumonia")
#named variable
outcome <- outcomes[["heart attack"]]

#conversion function
as.numeric.factor <- function(x) {as.numeric(levels(x))[x]}

#convert my data table column from factor to numeric
DT[, outcome := as.numeric.factor(outcome)]

理想情况下，

DT[, outcome := as.numeric.factor(outcome)]是我想要做的，但我知道我不能只在那里粘贴一个命名变量outcome，希望数据表知道我指的是具有值outcome。对不起，我对数据表很新，我对引用列时如何处理命名变量感到有些困惑。

Answer 1

使用tidyverse方法：

library(tidyverse)
Your_df_mod <- Your_df %>% 
    mutate_at(.vars=outcome,.funs=funs(as.numeric(.)))

Answer 2

你可以试试这个：

library(data.table)

DT <- data.table(mtcars)[, mpg := as.factor(mpg)]
class(DT$mpg)
# [1] "factor"

as.numeric.factor <- function(x) {as.numeric(levels(x))[x]}
outcome <- "mpg"
# Reference column by assigned variable
DT[, (outcome) := as.numeric.factor(get(outcome))]
class(DT$mpg)
# [1] "numeric"

在data.table中具有指定变量的引用列

2 个答案: