我正在尝试使用choropleth_geojson在choropleth映射中将geojson文件与csv数据合并,如minimal_example.py示例中所示:
import choropleth_geojson as cg
import pandas as pd
import json
import plotly.offline as offline
mapbox_access_token = 'key'
df = pd.read_csv('indicadores_ma.csv',usecols=['geocod','Impacto (2015)'],sep=';')
with open(r'county_ma.geojson') as f:
geojson = json.load(f)
northamerica = cg.choropleth(mapbox_access_token, df, geojson, 'geocod')
fig = northamerica.choroplot()
offline.plot(fig, auto_open=True)
输入数据文件为:
** indicadores_ma.csv
county_ma.geojson
{
"type": "FeatureCollection",
"features": [{
"type": "Feature",
"geometry": {
"type": "MultiPolygon",
"coordinates": [[[[-42.815, -5.3142],...]]]]
},
"properties": {
"geocod": 2112209,
"name": "Timon/MA"
}
}, {
"type": "Feature",
"geometry": {
"type": "MultiPolygon",
"coordinates": [[[[-41.8141, -2.7393],...]]]
},
"properties": {
"geocod": 2100907,
"name": "Araioses/MA"
}
}
]
}
在
fig = northamerica.choroplot()
我收到错误消息:
File "pandas/_libs/lib.pyx", line 2217, in pandas._libs.lib.map_infer
TypeError: descriptor 'lower' requires a 'str' object but received a 'int'
我做错了什么?
此消息中使用的文件可以从here下载。
答案 0 :(得分:0)
我找到了解决方案。在示例数据中,csv文件的第一列没有名称。它使大熊猫假设第一列应该用作索引列,并且它是一个字符串。
在我的数据中,我指定了所有列名,因此pandas创建了一个整数索引列,当choropleth_geojson尝试将索引值转换为小写时,该列会生成错误消息。解决方案是在dead_csv中指定索引列:
df = pd.read_csv('indicadores.csv',index_col=[0],usecols=['geocod','Impacto (2015)'],sep=';')