A091005 周秉宇
- 改以人口點圖(散布圖或隨機點子圖)或密度面量圖顯示以避免誤判
- 以其他顏色繪製周圍的直轄市/縣界輪廓
- 玩了一下 Windows 的字體問題,發現一定要透過
windowsFonts
來解決 - 一開始誤會
fill
跟color
的差異,另外對於aes
、+、%+% 等 ggplot2 的使用也不是太理解,有遇到一些小問題 - 好奇
scale_fill_distiller
跟scale_fill_brewer
的使用方式詳情
- 調整 label 位置
- 調整配色、車站大小
ggsfrepel
實在是非常難纏- 原先以為
dplyr
的可以吃,所以雖然測試時用了 vectorized 的 filter 函數,但後來去掉之後就噴錯了
警告訊息:
1: 於 st_point_on_surface.sfc(data$geometry):
st_point_on_surface may not give correct results for longitude/latitude data
2: 於 st_point_on_surface.sfc(data$geometry):
st_point_on_surface may not give correct results for longitude/latitude data
3: 於 st_point_on_surface.sfc(data$geometry):
st_point_on_surface may not give correct results for longitude/latitude data
4: 於 st_point_on_surface.sfc(data$geometry):
st_point_on_surface may not give correct results for longitude/latitude data
5: ggrepel: 5 unlabeled data points (too many overlaps). Consider increasing max.overlaps
6: ggrepel: 8 unlabeled data points (too many overlaps). Consider increasing max.overlaps
7: ggrepel: 5 unlabeled data points (too many overlaps). Consider increasing max.overlaps
8: ggrepel: 5 unlabeled data points (too many overlaps). Consider increasing max.overlaps