Revisions to Converting Pandas DataFrame to GeoDataFrame

fix deprecated CRS definition format

Source Link

edited Jun 26, 2020 at 19:52

2.1k
1
17
19

Convert the DataFrame's content (e.g. Lat and Lon columns) into appropriate Shapely geometries first and then use them together with the original DataFrame to create a GeoDataFrame.

from geopandas import GeoDataFrame
from shapely.geometry import Point

geometry = [Point(xy) for xy in zip(df.Lon, df.Lat)]
df = df.drop(['Lon', 'Lat'], axis=1)
crs = {'init': 'epsg:4326'}
gdf = GeoDataFrame(df, crs=crscrs="EPSG:4326", geometry=geometry)

Result:

    Date/Time           ID      geometry
0   4/1/2014 0:11:00    140     POINT (-73.95489999999999 40.769)
1   4/1/2014 0:17:00    NaN     POINT (-74.03449999999999 40.7267)

Since the geometries often come in the WKT format, I thought I'd include an example for that case as well:

import geopandas as gpd
import shapely.wkt

geometry = df['wktcolumn'].map(shapely.wkt.loads)
df = df.drop('wktcolumn', axis=1)
crs = {'init': 'epsg:4326'}
gdf = gpd.GeoDataFrame(df, crs=crscrs="EPSG:4326", geometry=geometry)

Convert the DataFrame's content (e.g. Lat and Lon columns) into appropriate Shapely geometries first and then use them together with the original DataFrame to create a GeoDataFrame.

from geopandas import GeoDataFrame
from shapely.geometry import Point

geometry = [Point(xy) for xy in zip(df.Lon, df.Lat)]
df = df.drop(['Lon', 'Lat'], axis=1)
crs = {'init': 'epsg:4326'}
gdf = GeoDataFrame(df, crs=crs, geometry=geometry)

Result:

    Date/Time           ID      geometry
0   4/1/2014 0:11:00    140     POINT (-73.95489999999999 40.769)
1   4/1/2014 0:17:00    NaN     POINT (-74.03449999999999 40.7267)

Since the geometries often come in the WKT format, I thought I'd include an example for that case as well:

import geopandas as gpd
import shapely.wkt

geometry = df['wktcolumn'].map(shapely.wkt.loads)
df = df.drop('wktcolumn', axis=1)
crs = {'init': 'epsg:4326'}
gdf = gpd.GeoDataFrame(df, crs=crs, geometry=geometry)

Convert the DataFrame's content (e.g. Lat and Lon columns) into appropriate Shapely geometries first and then use them together with the original DataFrame to create a GeoDataFrame.

from geopandas import GeoDataFrame
from shapely.geometry import Point

geometry = [Point(xy) for xy in zip(df.Lon, df.Lat)]
df = df.drop(['Lon', 'Lat'], axis=1)
gdf = GeoDataFrame(df, crs="EPSG:4326", geometry=geometry)

Result:

    Date/Time           ID      geometry
0   4/1/2014 0:11:00    140     POINT (-73.95489999999999 40.769)
1   4/1/2014 0:17:00    NaN     POINT (-74.03449999999999 40.7267)

Since the geometries often come in the WKT format, I thought I'd include an example for that case as well:

import geopandas as gpd
import shapely.wkt

geometry = df['wktcolumn'].map(shapely.wkt.loads)
df = df.drop('wktcolumn', axis=1)
gdf = gpd.GeoDataFrame(df, crs="EPSG:4326", geometry=geometry)

Make the sentences more compact

Source Link

edited Oct 20, 2017 at 12:29

Martin Valgur

2.1k
1
17
19

To create a GeoDataFrame from a DataFrame you have to convertConvert the DataFrame's content (e.g. Lat and Lon columns) into appropriate Shapely geometries first and then use these geometriesthem together with the original DataFrame to initializecreate a GeoDataFrame.

from geopandas import GeoDataFrame
from shapely.geometry import Point

geometry = [Point(xy) for xy in zip(df.Lon, df.Lat)]
df = df.drop(['Lon', 'Lat'], axis=1)
crs = {'init': 'epsg:4326'}
geo_dfgdf = GeoDataFrame(df, crs=crs, geometry=geometry)

Result:

    Date/Time           ID      geometry
0   4/1/2014 0:11:00    140     POINT (-73.95489999999999 40.769)
1   4/1/2014 0:17:00    NaN     POINT (-74.03449999999999 40.7267)

Since the geometries often come in the WKT format, I thought I'd include an example for that case as well:

import geopandas as gpd
import shapely.wkt

geometry = df['wktcolumn'].map(shapely.wkt.loads)
df = df.drop('wktcolumn', axis=1)
crs = {'init': 'epsg:4326'}
gdf = gpd.GeoDataFrame(df, crs=crs, geometry=geometry)

To create a GeoDataFrame from a DataFrame you have to convert the DataFrame's content (e.g. Lat and Lon columns) into appropriate Shapely geometries first and then use these geometries together with the original DataFrame to initialize a GeoDataFrame.

from geopandas import GeoDataFrame
from shapely.geometry import Point

geometry = [Point(xy) for xy in zip(df.Lon, df.Lat)]
df = df.drop(['Lon', 'Lat'], axis=1)
crs = {'init': 'epsg:4326'}
geo_df = GeoDataFrame(df, crs=crs, geometry=geometry)

Result:

    Date/Time           ID      geometry
0   4/1/2014 0:11:00    140     POINT (-73.95489999999999 40.769)
1   4/1/2014 0:17:00    NaN     POINT (-74.03449999999999 40.7267)

Since the geometries often come in the WKT format, I thought I'd include an example for that case as well:

import geopandas as gpd
import shapely.wkt

geometry = df['wktcolumn'].map(shapely.wkt.loads)
df = df.drop('wktcolumn', axis=1)
crs = {'init': 'epsg:4326'}
gdf = gpd.GeoDataFrame(df, crs=crs, geometry=geometry)

Convert the DataFrame's content (e.g. Lat and Lon columns) into appropriate Shapely geometries first and then use them together with the original DataFrame to create a GeoDataFrame.

from geopandas import GeoDataFrame
from shapely.geometry import Point

geometry = [Point(xy) for xy in zip(df.Lon, df.Lat)]
df = df.drop(['Lon', 'Lat'], axis=1)
crs = {'init': 'epsg:4326'}
gdf = GeoDataFrame(df, crs=crs, geometry=geometry)

Result:

    Date/Time           ID      geometry
0   4/1/2014 0:11:00    140     POINT (-73.95489999999999 40.769)
1   4/1/2014 0:17:00    NaN     POINT (-74.03449999999999 40.7267)

Since the geometries often come in the WKT format, I thought I'd include an example for that case as well:

import geopandas as gpd
import shapely.wkt

geometry = df['wktcolumn'].map(shapely.wkt.loads)
df = df.drop('wktcolumn', axis=1)
crs = {'init': 'epsg:4326'}
gdf = gpd.GeoDataFrame(df, crs=crs, geometry=geometry)

More concise explanation

Source Link

edited Mar 11, 2017 at 16:48

Martin Valgur

2.1k
1
17
19

You have toTo create an array of shapely geometriesa GeoDataFrame from a DataFrame you have to convert the DataFrame's content first (e.g. the Lat and Lon columns), which you can into appropriate Shapely geometries first and then use these geometries together with the original DataFrame to constructinitialize a GeoDataFrame.

from geopandas import GeoDataFrame
from shapely.geometry import Point

geometry = [Point(xy) for xy in zip(df.Lon, df.Lat)]
df = df.drop(['Lon', 'Lat'], axis=1)
crs = {'init': 'epsg:4326'}
geo_df = GeoDataFrame(df, crs=crs, geometry=geometry)

Result:

    Date/Time           ID      geometry
0   4/1/2014 0:11:00    140     POINT (-73.95489999999999 40.769)
1   4/1/2014 0:17:00    NaN     POINT (-74.03449999999999 40.7267)

Since the geometries often come in the WKT format, I thought I'd include an example for that case as well:

import geopandas as gpd
import shapely.wkt

geometry = df['wktcolumn'].map(shapely.wkt.loads)
df = df.drop('wktcolumn', axis=1)
crs = {'init': 'epsg:4326'}
gdf = gpd.GeoDataFrame(df, crs=crs, geometry=geometry)

You have to create an array of shapely geometries from the DataFrame's content first (e.g. the Lat and Lon columns), which you can then use to construct a GeoDataFrame.

from geopandas import GeoDataFrame
from shapely.geometry import Point

geometry = [Point(xy) for xy in zip(df.Lon, df.Lat)]
df = df.drop(['Lon', 'Lat'], axis=1)
crs = {'init': 'epsg:4326'}
geo_df = GeoDataFrame(df, crs=crs, geometry=geometry)

Result:

    Date/Time           ID      geometry
0   4/1/2014 0:11:00    140     POINT (-73.95489999999999 40.769)
1   4/1/2014 0:17:00    NaN     POINT (-74.03449999999999 40.7267)

Since the geometries often come in the WKT format, I thought I'd include an example for that case as well:

import geopandas as gpd
import shapely.wkt

geometry = df['wktcolumn'].map(shapely.wkt.loads)
df = df.drop('wktcolumn', axis=1)
crs = {'init': 'epsg:4326'}
gdf = gpd.GeoDataFrame(df, crs=crs, geometry=geometry)

To create a GeoDataFrame from a DataFrame you have to convert the DataFrame's content (e.g. Lat and Lon columns) into appropriate Shapely geometries first and then use these geometries together with the original DataFrame to initialize a GeoDataFrame.

from geopandas import GeoDataFrame
from shapely.geometry import Point

geometry = [Point(xy) for xy in zip(df.Lon, df.Lat)]
df = df.drop(['Lon', 'Lat'], axis=1)
crs = {'init': 'epsg:4326'}
geo_df = GeoDataFrame(df, crs=crs, geometry=geometry)

Result:

    Date/Time           ID      geometry
0   4/1/2014 0:11:00    140     POINT (-73.95489999999999 40.769)
1   4/1/2014 0:17:00    NaN     POINT (-74.03449999999999 40.7267)

Since the geometries often come in the WKT format, I thought I'd include an example for that case as well:

import geopandas as gpd
import shapely.wkt

geometry = df['wktcolumn'].map(shapely.wkt.loads)
df = df.drop('wktcolumn', axis=1)
crs = {'init': 'epsg:4326'}
gdf = gpd.GeoDataFrame(df, crs=crs, geometry=geometry)

More concise explanation

Source Link

edited Mar 11, 2017 at 16:38

Martin Valgur

2.1k
1
17
19

Loading

Add an example for WKT parsing and drop the initial geometry columns

Source Link

edited Feb 21, 2017 at 9:12

Martin Valgur

2.1k
1
17
19

Loading

Add a CRS for a more complete example

Source Link

edited Feb 20, 2017 at 13:22

Martin Valgur

2.1k
1
17
19

Loading

Wording

Source Link

edited Dec 17, 2015 at 12:30

Martin Valgur

2.1k
1
17
19

Loading

Update the result to match the updated code.

Source Link

edited Dec 16, 2015 at 22:37

Martin Valgur

2.1k
1
17
19

Loading

DataFrame.iterrows() is slow. Either using .itertuples() or the columns directly is about 4x faster.

Source Link

edited Dec 16, 2015 at 22:20

Martin Valgur

2.1k
1
17
19

Loading

A much cleaner conversion method.

Source Link

edited Dec 16, 2015 at 22:11

Martin Valgur

2.1k
1
17
19

Loading

For clarity, the conversion did not get rid of the "Date/Time" column. I was just too lazy to include it in the test data.

Source Link

edited Dec 16, 2015 at 21:47

Martin Valgur

2.1k
1
17
19

Loading

Source Link

answered Dec 16, 2015 at 21:39

Martin Valgur

2.1k
1
17
19

Loading

Stack Exchange Network

Return to Answer