Getting Started with EarthDaily EarthOne Python API

This repository is designed to serve as an introduction to the EarthOne Python API and its modules.

Installation

Please follow the steps located in our Documentation page to install the client or simply run:

pip install earthdaily-earthone

Note for working with the Dynamic Compute API outside of a Workbench environment

For the time being this package must be installed separate from the core EarthOne client:

!pip install earthdaily-earthone-dynamic-compute

Overview

A general outline of the tutorial notebooks located in this repo is as follows:

Guides

Quickstart examples outlined as a general overview for each of the core services within the EarthOne Platform.

1. Logging in to your local client installation for the first time
2. Catalog - Create, manage, search, share, and visualize data:
   • 01 Searching and Rastering Catalog Products.ipynb
   • 02 Creating and Managing Products.ipynb
   • 03 Advanced Catalog Product Operations.ipynb
   • 04 Introduction to Blob Storage.ipynb
   • 05 Searching and Retrieving Vector Tables.ipynb
   • 06 Creating and Managing Vector Tables.ipynb
   • 07 Introduction to Events.ipynb
3. Dynamic Compute - An interactive geospatial data processing engine:
   • 01 Interactive Computing with Mosaics.ipynb
   • 02 Interactive Computing with ImageStacks.ipynb
   • 03 GeoContexts and Array Computation.ipynb
   • 04 Advanced Computing Concepts.ipynb
   • 05 Managing and Sharing in Dynamic Compute.ipynb
4. Batch Compute - A highly scalable asynchronous compute service:
   • 01 Hello World.ipynb
   • 02 Create Imagery.ipynb
   • 03 Extracting Timeseries Data.ipynb

Demos

End-to-end example analytic pipelines oriented towards specific applications in remote sensing.

1. Unsupervised Machine Learning - Scale a kmeans clustering algorithm:

   • 01a Training an Unsupervised Classifier.ipynb
   • 01b Deploying an Unsupervised Classifier.ipynb
   • 01c Interactive Deployment with Dynamic Compute.ipynb

2. Supervised Machine Learning - Predict land cover by training and deploying a random forest classifier:

   • 02a Generate Training Data.ipynb
   • 02b Training a Supervised Classifier.ipynb
   • 02c Deploying a Supervised Classifier.ipynb
   • 02d Interactive Deployment with Dynamic Compute.ipynb

3. Image Segmentation - Train and deploy a simple U-Net model to detect well pads in West Texas:

   • 03a Generate Training Data.ipynb
   • 03b Training a Segmentation Model.ipynb
   • 03c Deploying a Segmentation Model.ipynb
   • 03d Interactive Deployment with Dynamic Compute.ipynb

4. Hurricane Case Study - Analyze the impacts of Hurricane Ida on roughly 7500 offshore oil rigs in the Gulf of Mexico:

   • 04 Hurricane Ida Offshore Rigs.ipynb

5. Vessel Detection Pipeline - Deploy a sample vessel detection pipeline which responds to new image upload events:

   • 01 Kaohsiung Port New Image Events.ipynb

6. Dynamic Compute Demos - Various case studies implementing scientific methodologies on open data through Dynamic Compute:
   • 01 Se2WaQ Neuchatel Lake
   • 02 Sentinel-1 Vessel Detection Shanghai

Support

If you have any questions please reach out to [email protected].