python画好看的图python如何画出漂亮的地图？

推荐一个超好用的python包folium, 专门用于地理数据可视化&＃xff0c;官方英文教程教程点击这里&＃xff0c;查看本文源代码请点击这里。 />

使用方法很简单&＃xff0c;操作如下&＃xff1a;导入包&＃xff0c;创建一副世界地图

import folium

import pandas as pd

# define the world map

world_map &＃61; folium.Map()

# display world map

world_map />

2. 输入经纬度&＃xff0c;尺度&＃xff0c;在这里我们以旧金山&＃xff08;37.7749° N, 122.4194° W&＃xff09;为例。

# San Francisco latitude and longitude values

latitude &＃61; 37.77

longitude &＃61; -122.42

# Create map and display it

san_map &＃61; folium.Map(location&＃61;[latitude, longitude], zoom_start&＃61;12)

# Display the map of San Francisco

san_map />

更改地图显示&＃xff0c;默认为&＃39;OpenStreetMap&＃39;风格&＃xff0c;我们还可以选择&＃39;Stamen Terrain&＃39;, &＃39;Stamen Toner&＃39;等。

# Create map and display it

san_map &＃61; folium.Map(location&＃61;[latitude, longitude], zoom_start&＃61;12,tiles&＃61;&＃39;Stamen Toner&＃39;) />

3. 读取数据集&＃xff08;旧金山犯罪数据集&＃xff09;

# Read Dataset

cdata &＃61; pd.read_csv(&＃39;https://cocl.us/sanfran_crime_dataset&＃39;)

cdata.head() />

4. 在地图上显示前200条犯罪数据

# get the first 200 crimes in the cdata

limit &＃61; 200

data &＃61; cdata.iloc[0:limit, :]

# Instantiate a feature group for the incidents in the dataframe

incidents &＃61; folium.map.FeatureGroup()

# Loop through the 200 crimes and add each to the incidents feature group

for lat, lng, in zip(cdata.Y, data.X):

incidents.add_child(

folium.CircleMarker(

[lat, lng],

radius&＃61;7, # define how big you want the circle markers to be

color&＃61;&＃39;yellow&＃39;,

fill&＃61;True,

fill_color&＃61;&＃39;red&＃39;,

fill_opacity&＃61;0.4

)

)

# Add incidents to map

san_map &＃61; folium.Map(location&＃61;[latitude, longitude], zoom_start&＃61;12)

san_map.add_child(incidents) />

5. 添加地理标签

# add pop-up text to each marker on the map

latitudes &＃61; list(data.Y)

longitudes &＃61; list(data.X)

labels &＃61; list(data.Category)

for lat, lng, label in zip(latitudes, longitudes, labels):

folium.Marker([lat, lng], popup&＃61;label).add_to(san_map)

# add incidents to map

san_map.add_child(incidents) />

6. 统计区域犯罪总数

from folium import plugins

# let&＃39;s start again with a clean copy of the map of San Francisco

san_map &＃61; folium.Map(location &＃61; [latitude, longitude], zoom_start &＃61; 12)

# instantiate a mark cluster object for the incidents in the dataframe

incidents &＃61; plugins.MarkerCluster().add_to(san_map)

# loop through the dataframe and add each data point to the mark cluster

for lat, lng, label, in zip(data.Y, data.X, cdata.Category):

folium.Marker(

location&＃61;[lat, lng],

icon&＃61;None,

popup&＃61;label,

).add_to(incidents)

# add incidents to map

san_map.add_child(incidents) />

7. 读取geojson文件&＃xff0c;可视化旧金山市10个不同Neighborhood的边界

import json

import requests

url &＃61; &＃39;https://cocl.us/sanfran_geojson&＃39;

san_geo &＃61; f&＃39;{url}&＃39;

san_map &＃61; folium.Map(location&＃61;[37.77, -122.4], zoom_start&＃61;12)

folium.GeoJson(

san_geo,

style_function&＃61;lambda feature: {

&＃39;fillColor&＃39;: &＃39;#ffff00&＃39;,

&＃39;color&＃39;: &＃39;black&＃39;,

&＃39;weight&＃39;: 2,

&＃39;dashArray&＃39;: &＃39;5, 5&＃39;

}

).add_to(san_map)

#display map

san_map />

8. 统计每个区域的犯罪事件数目

# Count crime numbers in each neighborhood

disdata &＃61; pd.DataFrame(cdata[&＃39;PdDistrict&＃39;].value_counts())

disdata.reset_index(inplace&＃61;True)

disdata.rename(columns&＃61;{&＃39;index&＃39;:&＃39;Neighborhood&＃39;,&＃39;PdDistrict&＃39;:&＃39;Count&＃39;},inplace&＃61;True)

disdata />

9. 创建Choropleth Map &＃xff08;颜色深浅代表各区犯罪事件数目&＃xff09;

m &＃61; folium.Map(location&＃61;[37.77, -122.4], zoom_start&＃61;12)

folium.Choropleth(

geo_data&＃61;san_geo,

data&＃61;disdata,

columns&＃61;[&＃39;Neighborhood&＃39;,&＃39;Count&＃39;],

key_on&＃61;&＃39;feature.properties.DISTRICT&＃39;,

#fill_color&＃61;&＃39;red&＃39;,

fill_color&＃61;&＃39;YlOrRd&＃39;,

fill_opacity&＃61;0.7,

line_opacity&＃61;0.2,

highlight&＃61;True,

legend_name&＃61;&＃39;Crime Counts in San Francisco&＃39;

).add_to(m)

m />

10. 创建热力图

from folium.plugins import HeatMap

# let&＃39;s start again with a clean copy of the map of San Francisco

san_map &＃61; folium.Map(location &＃61; [latitude, longitude], zoom_start &＃61; 12)

# Convert data format

heatdata &＃61; data[[&＃39;Y&＃39;,&＃39;X&＃39;]].values.tolist()

# add incidents to map

HeatMap(heatdata).add_to(san_map)

san_map />

本文源代码Jupyter notebook地址&＃xff1a;Jupyter Notebook Viewer​nbviewer.jupyter.org

GitHub下载地址&＃xff1a;gaonanlee/Visualization-Practice​github.com

最后&＃xff0c;folium还可以用来创建动态热力图&＃xff0c;动态路径图等&＃xff0c;具体可参考Medium上的这篇文章。

实现效果如下图所示 (直接从Medium上抱过来的图&＃xff0c;详细代码请点击上述链接&＃xff09;。 />

我的其他回答

欢迎大家关注我的机器学习笔记专栏&＃xff0c;我将用小白也能听懂的语言&＃xff0c;为大家讲述机器学习中那些有趣好玩的知识 (●&＃39;◡&＃39;●)