A python package to extract gee data for machine learning.
Explore the documentation »
View Demo
·
Report Bug
·
Request Feature
Table of Contents
This python package makes it easier to extract satellite data from Google Earth Engine using parallel processing and the Google Earth Engine high volume end point.
In its current state it supports the extraction of data for traditional machine learning (tabular data) in the form of csv's and the extraction of GeoTiff image patches for Deep Neural Networks.
The Machine learning capabilities in the GEE JS code editor remain limited. For example, there is no support for XGBoost, LightGBM, NGBoost, etc. Moreover, the python ecosystem has much more support for training, valdation and hyperparameter tuning. However, for this functionality to be leveraged, data needs to be downloaded locally or stored in Google Drive or Google Cloud Storage to benefit from the Machine learning python ecosystem. Therfore, this package aims to make it easier and faster to download GEE-processed data in a machine learning-ready format.
Features
- Parallel export of images or sparse images (for example, GEDI).
- Export raster values at points or polygons (ee.FeatureCollection).
- Summarise raster data within polygons (ee.FeatureCollections).
- Extract both tabular and Deep Neural Network (DNN) type datasets.
To install this package:
- pip
pip install geeml
- Build from source (latest version)
pip install git+https://github.com/Geethen/geeml.git
Download the NASADEM elevation data for Kenya.
#import packages
import ee
from geeml.utils import getCountry
from geeml.extract import extractor
# Authenticate GEE
ee.Authenticate()
# Initialize GEE with high-volume end-point
ee.Initialize(opt_url='https://earthengine-highvolume.googleapis.com')
# Import datasets from GEE
nasadem = ee.Image("NASA/NASADEM_HGT/001")
# A point in Kenya
poi = ee.Geometry.Point([37.857884,-0.002197])
kenya = getCountry(poi)
# Download directory
dd = '/content/drive/MyDrive/geeml_example'
# Prepare for data extraction
trialExtractor = extractor(covariates=nasadem, aoi = kenya, scale= 5000, dd= dd)
# Extract data
trialExtractor.extractAoi()For more examples, please refer to the Documentation
- Support the export of additional formats (TFrecords)
- Download data from GEE based on local shapefiles
- Add more examples for using the package
See the open issues for a full list of proposed features (and known issues).
Contributions are what make the open source community such an amazing place to learn, inspire, and create. Any contributions you make are greatly appreciated.
If you have a suggestion that would make this better, please fork the repo and create a pull request. You can also simply open an issue with the tag "enhancement". Don't forget to give the project a star! Thanks again!
- Fork the Project
- Create your Feature Branch (
git checkout -b feature/AmazingFeature) - Commit your Changes (
git commit -m 'Add some AmazingFeature') - Push to the Branch (
git push origin feature/AmazingFeature) - Open a Pull Request
Distributed under the MIT License. See LICENSE.txt for more information.
Geethen Singh - @Geethen - [email protected]
Project Link: https://github.com/Geethen/geeml
This package was created with Cookiecutter and the giswqs/pypackage project template.
This package uses the geedim package for extracting image data at an AOI