如何处理 Folium 地图上的极值答案

【问题标题】：How to deal with extreme values on a Folium map如何处理 Folium 地图上的极值
【发布时间】：2022-01-16 10:20:26
【问题描述】：

我对folium 地图的经验很少。我需要用每个部门的机构数量制作一张地图，问题是首都的机构数量远远多于内部，所以当我创建颜色层时，我将首都设为深蓝色，其余的都一样较浅的颜色。这样地图就没那么有用了…… 我怎么能解决这个问题？我想也许将值除以人口，但最好使用原始值。在文档中，我没有找到参数化颜色的方法。

    df1 = pd.DataFrame({'code':['75','77','78','91','92','93','94','95'],'value':['13000','2000','2500','2300','2150','2600','1630','1300']})
    dep_geo = geopandas.read_file('./dep.json', driver="JSON") #geodata taken from https://github.com/gregoiredavid/france-geojson/blob/master/departements.geojson
departments = {'75', '77', '78', '91', '92', '93', '94', '95'}
dep_geo = dep_geo[dep_geo['code'].isin(departments)]

df_map = dep_geo.merge(df1, how="left", left_on=['code'], right_on=['code'])

my_map = folium.Map(location=[48.856614, 2.3522219],
                    zoom_start = 9, tiles='cartodbpositron')

folium.Choropleth(
geo_data=df_map,
data=df_map,
columns=['code',"value"],
key_on="feature.properties.code",
fill_color='YlGnBu',
fill_opacity=0.5,
line_opacity=0.2,
legend_name="value ",
smooth_factor=0,
Highlight= True,
line_color = "black",
name = "value",
show=False,
overlay=True,
nan_fill_color = "White"
).add_to(my_map)

结果：

感谢您的帮助！

【问题讨论】：

标签： python maps geopandas folium

【解决方案1】：

就像使用 vmax 参数一样简单。我已设置为第 85 个百分位
还使用 geopandas explore() 生成 folium 地图

import geopandas as gpd
import pandas as pd
import folium

df1 = pd.DataFrame(
    {
        "code": ["75", "77", "78", "91", "92", "93", "94", "95"],
        "value": ["13000", "2000", "2500", "2300", "2150", "2600", "1630", "1300"],
    }
)
# dep_geo = geopandas.read_file('./dep.json', driver="JSON") #geodata taken from https://github.com/gregoiredavid/france-geojson/blob/master/departements.geojson
dep_geo = gpd.read_file(
    "https://github.com/gregoiredavid/france-geojson/raw/master/departements.geojson"
)  # geodata taken from https://github.com/gregoiredavid/france-geojson/blob/master/departements.geojson


departments = {"75", "77", "78", "91", "92", "93", "94", "95"}
dep_geo = dep_geo[dep_geo["code"].isin(departments)]
df_map = dep_geo.merge(df1, how="left", left_on=["code"], right_on=["code"])
df_map["value"] = pd.to_numeric(df_map["value"])

df_map.explore(
    column="value",
    cmap="YlGnBu",
    vmax=df_map["value"].quantile(0.85),
    style_kwds=dict(
        color="rgba(0,0,0,.2)",
    ),
    location=[48.856614, 2.3522219],
    zoom_start=9,
    tiles="cartodbpositron",
)

【讨论】：