Getting Started: Descartes Labs Platform

Let's make sure you have everything you need to get up and running with the DL Platform

By now you should have been able to create an account, verify your email, and log in to your account.

Login to your portal

Ensure that you are logged in by going to and signing in using the credentials you created during the setup process.

Install the Python Client

NOTE: If you are using the Workbench service, you can skip to Authentication. 

To start using the DL Platform, you need to install the Python client and perform a quick test to ensure everything is functioning correctly. Whether you are using a Virtual Machine (VM) or your local maching, you can run the DL Python client by following these steps: 

Example Notebooks to get your started

The notebooks are located in our Jupyter Lab environment Workbench. Alternatively, you can visit our GitHub repository to learn how to use the DL Platform. The repository covers everything from basic concepts and API usage to creating a simple web application that takes advantage of DL Platform services. To get started, simply clone the repository to your local machine and explore the Jupyter notebooks in the "notebooks" folder.

Sample Notebooks include examples of working with the API, models on demand, models at scale. See the Github repo for more information.

Key Features

Create, manage, search, and visualize geospatial data with the Catalog.

The DL Catalog serves as a comprehensive repository for geo-referenced data. It offers access to a vast collection of DL-provided geospatial data, amounting to approximately 20 Pedabytes. Additionally, users can augment the Catalog with their own data products and any derivative data generated on the Platform.

One of the Catalog's key advantages is its ability to simplify access to earth observation imagery. This enables customers to swiftly search for and extract the most pertinent data for their specific needs. Moreover, the Catalog provides a highly scalable data feed and storage mechanism, which, when paired with DL-provided compute capabilities, empowers global-scale modeling for the world's most intricate problems.

Furthermore, the Catalog offers specialized storage and access to geo-referenced vector data. It also provides flexible file storage for model weights, results, and other types of data. Each dataset within the Catalog can be independently managed, searched, and shared, making it an advanced collaboration platform for geospatial projects.

Utilize Dynamic Compute, our powerful on-demand geospatial analysis engine, to quickly prototype and develop your projects.

Dynamic Compute allows users to focus on their specific problem rather than getting caught up in the details of the data. It provides users with a live-updating interactive map that displays their analysis. Users can combine different operations on Catalog data and add them to the map. These operations are computed as the user explores their area of interest on the map. This interactive approach allows users to develop analyses without needing to know specific geographic coordinates or deal with the complexities of the dataset.

Leverage the power of scalable compute in Descartes Lab's cloud infrastructure to parallelize your computations.

When you need to deploy a model across a wide area, the Batch Compute service offers users the ability to utilize cloud computing infrastructure to parallelize and run their code on a large scale. Users' Python code is packaged and executed on nodes that are hosted within DL's cloud infrastructure. This provides a flexible foundation for running complex machine learning and artificial intelligence algorithms on Catalog data.

Explorer Interface

The Descartes Labs Explorer interface provides users with the ability to search and visualize data products that are available on the Platform. By utilizing Explorer, users can easily become familiar with different data products and their associated metadata. Additionally, Explorer leverages the Catalog to enable users to manage their own datasets, offering a comprehensive solution for data exploration and management.

High resolution Airbus OneAtlas Spot 6/7 imagery in Explorer.

Workbench Interface

Workbench is a hosted development environment specially tuned to work with the DL Platform. In addition to being a one-stop-shop for interacting with DL’s APIs, Workbench provides a suite of pre-loaded example Jupyter notebooks and tutorials that can be run and modified to jump-start development.

API Documentation and Guides

To learn more about the Python API's, you can refer to the comprehensive API documentation available at Additionally, Descartes Labs provides detailed Guides that offer additional information and functionality. You can access the Guides at

Next: Understanding Your Usage