如何将 geom_smooth() 与 gganimate 一起使用

Question

我很难理解如何将geom_smooth() 与gganimate 一起使用。

reprex 在下面可用。运行不包括transition_time 函数的绘图语句时，我按预期可视化静态绘图。

当我尝试使用transition_time() 和shadow_mark() 使该图动态化以保留点时，就会出现问题。出现以下错误：

$<-.data.frame(*tmp*, "group", value = "") 中的错误：替换 有 1 行，数据有 0

library(readr)
library(tidyr)
library(ggplot2)
library(gifski)
library(gganimate)
library(dplyr)

tst <- readr::read_csv("https://elpais.com/especiales/2019/elecciones-generales/encuestas-electorales/ficheros/tabla.csv?1554807308", 
                       col_types = cols(
                         date = col_date(format = "%d/%m/%y"),
                         house = col_character(),
                         sample = col_double(),
                         turnout = col_double(),
                         PP = col_double(),
                         PSOE = col_double(),
                         UP = col_double(),
                         Cs = col_double(),
                         ERC = col_double(),
                         PDC = col_double(),
                         PNV = col_double(),
                         PAC = col_double(),
                         BIL = col_double(),
                         CC = col_double(),
                         VOX = col_double(),
                         COM = col_double()
                       ))

tst %>% 
  select(date, house, PP, PSOE, UP, Cs, VOX) %>% 
  gather(key = partido, voto, PP, PSOE, UP, Cs, VOX) %>% 
  ggplot(aes(x = date, y = voto, color = partido)) +
  geom_point() + 
  geom_smooth(method = 'loess', formula = 'y ~ x', se = FALSE) +
  transition_time(time = date) + shadow_mark()

Error in `$<-.data.frame`(`*tmp*`, "group", value = ""): replacement
has 1 row, data has 0
In addition: There were 16 warnings (use warnings() to see them)

^{由reprex package (v0.2.1) 于 2019-04-10 创建}

我期待的是与此类似的内容 (source)：

Answer 1

试试这个：

tmp <- tst %>% 
  select(date, house, PP, PSOE, UP, Cs, VOX) %>% 
  gather(key = partido, voto, PP, PSOE, UP, Cs, VOX) %>%
  filter(!is.na(voto))

ggplot(tmp, aes(x = date, y = voto, color = partido)) + 
  geom_smooth(method = "loess", formula = 'y ~ x', se = FALSE) + 
  geom_point(aes(group = rev(seq_along(date)))) +
  transition_reveal(rev(seq_along(date))) + shadow_mark()