小编No *_*Lie的帖子

我shapefile从data.gov下载了县。我可以在传单中使用它们，如下所示来绘制 WA、ID 和 OR 中的县。我想让州边界更厚。我怎样才能做到这一点？

counties <- readOGR(paste0(dir, "/tl_2017_us_county.shp"),
                    layer = "tl_2017_us_county", GDAL1_integer64_policy = TRUE)

counties <- counties[counties@data$STATEFP %in% c("16", "41", "53"), ]
counties <- subset(counties, STATEFP %in% c("16", "41", "53") )
counties <- rmapshaper::ms_simplify(counties)

counties %>%
leaflet() %>%
setView(lng = -118.4942, lat = 47.2149, zoom = 5) %>%
addPolygons( fillColor = "green", fillOpacity = 0.5,
             color = "black", opacity = 1.0, weight = .6, smoothFactor = 0.5,
             highlightOptions = highlightOptions(color="white", weight=2, bringToFront = TRUE),
             label= …

我有一个数据表，dt看起来像

   location year value
       NYC 2026     1
       NYC 2026     2
       NYC 2026     3
       NYC 2026     4
       NYC 2026     5
        LA 2026     6
        LA 2026     7
        LA 2026     8
        LA 2026     9
        LA 2026    10

我想对它们进行分组，city并year在value每组中的列中找到第二小的元素，所需的结果如下所示：

   location year value
        NYC  2026     2
         LA  2026     7

dt %>% grou_by(location, year) %>% nth(value, 2)

将无法正常工作。任何帮助表示赞赏。

上面的数据表可以通过以下方式创建：

dt <- structure(list(location = c("NYC", "NYC", "NYC","NYC", "NYC", 
                                   "LA", "LA", "LA", "LA", "LA"), 
                 year = c(2026, 2026, 2026, 2026, …

我想从 dataframe 中找到kdataframe 中所有点的最近邻居。这怎么可行？它似乎只需要一组数据，并且只需要一个查询点。ABsklearn.neighbors.NearestNeighbors

喜欢：

samples = [[0., 0., 0.], [0., .5, 0.], [1., 1., .5]]
from sklearn.neighbors import NearestNeighbors
neigh = NearestNeighbors(n_neighbors=1)
neigh.fit(samples) 

print(neigh.kneighbors([[1., 1., 1.]])) 

来自Python 文档

我想要一个包含多个查询点的数据框而不是 [[1., 1., 1.]]

PS我想要的指标是Mahalanobis其sklearn.neighbors.KDTree不接受，而且scipy.spatial.KDTree甚至没有任何指标的选择。