Question

我有一个如下所示的数据框，我想根据周来对此进行热议，

这是我理想的输出：

id    week1   week2    week3    week4    week5
345    1        2       0         0        1
351    0        1       0         0        0
222    0        0       0         1        0
264    0        0       1         0        1
277    0        1       0         0        0 
...

我对这个问题的想法是基于组合此数据帧的一种热编码的，但是它非常复杂，

任何人都知道我可以在R中获得此输出吗？

Answer 1

我相信这可以做得更优雅，但这可以完成工作。

# Libraries
library(dplyr)
library(tidyr)

# Dataframe
data <- "id    week
345    1
351    2
222    4
264    3
345    5
277    2
345    2
345    2
264    5"
df <- read.table(text = data, header = TRUE)

# All at once
df <- df %>% 
  group_by(id, week) %>% 
  summarise(count = n()) %>%
  mutate(week = paste0("week", week)) %>%
  spread(week, count)

# Setting NA to zero
df[is.na(df)] <- 0

Answer 2

使用tidyverse：

df %>%
  mutate(week = paste("week", week, sep = "")) %>%
  group_by(id, week) %>% 
  summarise(n = n()) %>%
  ungroup() %>%
  spread(key = week, value = n) %>% 
  mutate_all(funs(replace(., is.na(.), 0)))

# A tibble: 5 x 6
     id week1 week2 week3 week4 week5
  <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
1  222.    0.    0.    0.    1.    0.
2  264.    0.    0.    1.    0.    1.
3  277.    0.    1.    0.    0.    0.
4  345.    1.    2.    0.    0.    1.
5  351.    0.    1.    0.    0.    0.

R中的许多热门编码器

2 个答案: