The indices of the two GeoSeries are different - Understanding Indices

user_3935035 0
Viewed 3135

I am working with GeoPandas and I have two GeoDataframes with the same CRS. One of them contains a geometry column with a polygon geometry, the other one a column with point geometry. I want to check which points are inside the polygon.

Naively I tried

shape.contains(points)

This gave me

>  The indices of the two GeoSeries are different

I do not understand this message. When I check the documentation, it says

We can also check two GeoSeries against each other, row by row. The GeoSeries above have different indices. We can either align both GeoSeries based on index values and compare elements with the same index using align=True or ignore index and compare elements based on their matching order using align=False:

What are these Indices? Why are they checked against each other and not the geometry columns? Online I read, I have to convert my geometries into shapely geometries. But isn't the whole point of using GeoPandas that I can perfrom geographical operations on the data?

I am confused about this. How to check if the geometries in shape contain any of the geometries in points?

geopandas geo geospatial python
3 Answers

What you are describing is effectively a spatial join. Below example constructs points from lon/lat of cities in UK and then finds which administrational area polygon the city is in. This is an NxM comparison

import pandas as pd
import numpy as np
import geopandas as gpd
import shapely.geometry
import requests

# source some points and polygons
# fmt: off
dfp = pd.read_html("https://www.latlong.net/category/cities-235-15.html")[0]
dfp = gpd.GeoDataFrame(dfp, geometry=dfp.loc[:,["Longitude", "Latitude",]].apply(shapely.geometry.Point, axis=1))
res = requests.get("https://opendata.arcgis.com/datasets/69dc11c7386943b4ad8893c45648b1e1_0.geojson")
df_poly = gpd.GeoDataFrame.from_features(res.json())
# fmt: on

gpd.sjoin(dfp, df_poly)
user_94414040

What are these Indices?

Simply speaking index is name of row of pandas.DataFrames or entry of pandas.Series. Aligning using indices is useful if you have data which overlap only partially, consider following example: let say you have daily data from two sensors but 2nd was turned on later then you might prepare pandas.DataFrame as follows.

import pandas as pd
s1 = pd.Series({'day1':100,'day2':110,'day3':105,'day4':120,'day5':110})
s2 = pd.Series({'day3':100,'day4':105,'day5':100})
df = pd.DataFrame({'sensor1':s1,'sensor2':s2})
print(df)

output

      sensor1  sensor2
day1      100      NaN
day2      110      NaN
day3      105    100.0
day4      120    105.0
day5      110    100.0

Observe that NaN (denoting unknown value) were inserted for day1 and day2 for sensor2.

user_107859750

The contains operation doesn't do the full NxM comparison. It can check which of the polygons contain a single point. Or if you provide it with two series it will chack whether the first polygon contains the first point, the second polygon contains the second point, etc..

The error you're getting is geopandas trying the second option, but the points and polygons don't align so it can't do the element-wise comparison.

It sounds like you want the full all polygons times all points comparison. You can do that by iterating over all of the points:

point_series = gpd.GeoSeries([Point(1, 2), Point(2, 1)])

polygon_series = gpd.GeoSeries(
    [
        Polygon([(0, 0), (0, 3), (2, 3), (2, 0), (0, 0)]),
        Polygon([(0, 0), (0, 2), (3, 2), (3, 0), (0, 0)]),
    ]
)

pd.concat([polygon_series.contains(point) for point in point_series], axis=1)

Output:

    0       1
0   True    False
1   False   True
user_48000860
Related