r - 列 "rate"的长度必须为 1(汇总值)，而不是 22906

Question

我在使用以下代码时遇到了一些麻烦。它返回

"Error in summarise_impl(.data, dots) : Column rate must be length 1 (a summary value), not 22906"

我的代码有问题吗？ sub_grade是字符类型和 int_rate是数字

results <- loan_data %>%
  select(credit_grade, sub_grade, int_rate, loan_amnt) %>%
  group_by(sub_grade) %>%
  summarise(
    rate = substr(int_rate * 100, 1, 4),
    nr_loans = n(),
    "&",
    percent1 = substr((nr_loans / a) * 100, 1, 5),
    klj = "&",
    Amount = sum(loan_amnt, na.rm = TRUE),
    klj1 = "&",
    percent2 = substr((Amount / total) * 100, 1, 5)
  )

只有当我添加第一个变量 rate 时才会出现问题.
可重现的例子:

sub_grade <- c("A1", "A2", "A3","A1","A3")
int_rate <– c(0.023, 0.027, 0.033,0.023,0.033)

我想要的是

sub_grade.  int_rate

A1. 0.023

A2。 0.027

A3. 0.033

Answer 1

问题是dplyr::summarise期望/接受每组一个值。但是substr(int_rate*100, ...)在您的代码中将返回每一行的值，即每组的许多值。您需要考虑使用一些分组功能，例如 min, max, first, last etc作为 substr 的一部分.考虑到 OP 发布的示例数据，解决方案可能是:

# Data
sub_grade <- c("A1", "A2", "A3","A1","A3")
int_rate <- c(0.023, 0.027,0.033,0.023,0.033)

loan_data <- data.frame(sub_grade, int_rate, stringsAsFactors = FALSE)

# Use dplyr to summarise on sub_grade
library(dplyr)
loan_data %>% group_by(sub_grade) %>%
  summarise(int_rate = first(int_rate)) %>%
  as.data.frame()

#   sub_grade int_rate
# 1        A1    0.023
# 2        A2    0.027
# 3        A3    0.033