ee.Feature to pd.DataFrame

[OfirMazor]

It can be very comfortable to have a shortcut function that convert an ee.Feature to Pandas DataFrame or GeoDataFrame (if geometry is necessary). For example:

import pandas as pd
import geopandas as gpd
import geemap
import ee

my_feature = ee.FeatureCollection("FAO/GAUL/2015/level0").filterMetadata('ADM0_NAME', 'equals', 'country_name')
df = geemap.to_DataFrame(my_feature)
geo_df = geemap.to_GeoDataFrame(my_feature) 

I didn't find any straight forward function like this at the docs. I am aware I can save ee.Feature as csv/shp and than read it with pandas/geopandas. My suggestion is to build a function that skip these steps.

[giswqs]

Good suggestion! I just added the following functions. See this example. Update the package using geemap.update_package()

• ee_to_pandas()
• ee_to_geopandas()
• csv_to_pandas()
• shp_to_geopandas()

[giswqs]

Try ee_to_gdf

[Cilence-lab]

Thanks for reply but now I get the following error:
geemap.ee_to_geojson(trainingSamples,filename='points.geojson')
Unable to transform geometry into requested projection.

[Cilence-lab]

trainingSamples = clusterOutfft.stratifiedSample(**{
'numPoints': 10,

classBand: 'binary_image',

#'region': polygon,
'scale': 10,
#'projection':s2.first().select(0).projection(),
'geometries': True
})
ee_to_gdf(trainingSamples)

[Cilence-lab]

@giswqs thank you. I had to specify region parameter when I use stratifiedDample function.
All is well now. Thank you!

[gdarkwah]

Good suggestion! I just added the following functions. See this example. Update the package using geemap.update_package()

• ee_to_pandas()
• ee_to_geopandas()
• csv_to_pandas()
• shp_to_geopandas()

ee_to_pandas() is replaced with ee_to_df in the updated version.