r - 如何仅在数字大于阈值时绘制 ggplots 六边形

Question

我想用 ggplot 的漂亮框架创建一个绘图。这是一个六边形的密度图。我使用了 https://www.r-graph-gallery.com/329-hexbin-map-for-distribution/ 中的示例代码

图形很好，但如果达到阈值，我想拥有这些六边形。例如:如果数字大于 4，则绘制所有值。

是否有机会保存底层聚合数据？我想用它们来进一步测试模式相似性。因此我想删除四次或更少观察的点。

通常可以通过以下方式提取数据

 object <- Function_that_produces_object
 object$Data_I_Want_have

我查看了文档，但是写了如何增加 Letters 的大小，但没有写明级别的数量和范围。

包

library(tidyverse)
library(viridis)
library(ggplot2)
# Get the GPS coordinates of a set of 200k tweets:
data=read.table("https://www.r-graph-gallery.com/wp-content/uploads/2017/12/Coordinate_Surf_Tweets.csv", sep=",", header=T)

# Get the world polygon
library(mapdata)
world <- map_data("world")



data %>%
  filter(homecontinent=='Europe') %>%
  ggplot( aes(x=homelon, y=homelat)) + 
  geom_hex(bins=65) +
  theme_void() +
  xlim(-30, 70) +
  ylim(24, 72) +
  scale_fill_viridis(option="B",
                     trans = "log", 
                     name="Number of Tweet recorded in 8 months", 
                     guide = guide_legend( keyheight = unit(3, units = "mm"), keywidth=unit(12, units = "mm"), label.position = "bottom", title.position = 'top', nrow=1) 
  )  +
  ggtitle( "Where people tweet about #Surf" ) +
  theme(
    legend.position = c(0.5, 0.09),
    text = element_text(color = "#22211d"),
    plot.background = element_rect(fill = "#f5f5f2", color = NA), 
    panel.background = element_rect(fill = "#f5f5f2", color = NA), 
    legend.background = element_rect(fill = "#f5f5f2", color = NA),
    plot.title = element_text(size= 22, hjust=0.1, color = "#4e4d47", margin = margin(b = -0.1, t = 0.4, l = 2, unit = "cm")),
  )

Answer 1

如评论中所述，您可以使用 ggplot_build 提取绘制的数据。

获得所需情节的一种方法是使用 cut，就像这里提到的:https://unconj.ca/blog/not-all-population-maps-are-boring.html对数据进行分箱。

如果您从 4 而不是 0 开始，则 5 以下的所有内容都将映射到 NA，不会绘制这些点，然后您可以在 中使用 breaks code>scale_fill_viridis 从图例中删除 NA 因素，然后您再次从 ggplot_build 中获取绘图数据。

我的意思是:

df <- read.table("https://www.r-graph-gallery.com/wp-content/uploads/2017/12/Coordinate_Surf_Tweets.csv", sep=",", header=T)
df %>%
  filter(homecontinent=='Europe') %>% 
  ggplot( ) + 
  geom_hex(aes(x=homelon, y=homelat, 
               fill = cut(..count.., c(4, 10, 50, 100, 500, 1000, 2000, Inf))), 
           bins=65) +
  theme_void() +
  xlim(-30, 70) +
  ylim(24, 72) + 
  scale_fill_viridis(option="B",
                     breaks = cut(c(5, 10, 50, 100, 500, 1000, 2000), 
                                  c(4, 10, 50, 100, 500, 1000, 2000, Inf)),
                     labels = c("5-9 ", "10-49 ", "50-99 ", "100-499 ", "500-999 ", "1000-1999", '2000+'), 
                     name="Number of Tweet recorded in 8 months",
                     discrete = TRUE,
                     guide = guide_legend( keyheight = unit(3, units = "mm"), 
                                           keywidth=unit(12, units = "mm"), 
                                           label.position = "bottom", 
                                           title.position = 'top', 
                                           nrow=1) ) +
  ggtitle( "Where people tweet about #Surf" ) +
  theme(
    legend.position = c(0.5, 0.09),
    text = element_text(color = "#22211d"),
    plot.background = element_rect(fill = "#f5f5f2", color = NA), 
    panel.background = element_rect(fill = "#f5f5f2", color = NA), 
    legend.background = element_rect(fill = "#f5f5f2", color = NA),
    plot.title = element_text(size= 22, hjust=0.1, color = "#4e4d47", margin = margin(b = -0.1, t = 0.4, l = 2, unit = "cm")),
  )

最终我得到了这个: