The Tables API allows users to store, query, and access tabular data. This includes but is not limited to vector or spatially referenced tabular data. Tables is replacing the older Vectors service which will be deprecated.
As part of the 1.10.0 release, Descartes Labs (DL) is excited to announce the new Tables API! The Tables API allows users to store, query, and access tabular data. This includes but is not limited to vector or spatially referenced tabular data. Tables is replacing the older Vectors service which has been deprecated. This overview will walk you through an introduction to Tables and provide answers to frequently asked questions. For more detailed information on Tables you can refer to the API documentation or the Tables Guide. For any additional help with Tables please reach out to support@descarteslabs.com.
A GeoPandas dataframe of MODIS Active Fire Data uploaded to and pulled from the DL Platform using the Tables API and plotted interactively using Folium. Getting this data onto the Platform only took a few cells of code in a Jupyter notebook on the DL Workbench!
What is a table (in DL Tables)?
A Table is one or more columns of data where each column has a specified data type. A variety of familiar python data types can be assigned to your Tables’ columns. Additionally geometry columns mirror GeoJSON specification.
What can I do with Tables?
Tables can be used to store any tabular data relevant to your data science work on the DL Platform. Data in a table can be queried both spatially (when relevant) and by the various columns. All queries are executed server-side. This allows you to use “table expressions” similar to those used in Pandas and Geopandas without needing to have all features/rows loaded into memory!
Can I share Tables with my colleagues?
Tables is fully integrated with Discover. This means that any tables you have stored can be shared with other users at your discretion. You can grant two roles to other users when sharing: “editor” and “viewer”. Editors can upload and insert data into the table while viewers only have permission to query the table.
What limitations should I expect when using Tables?
Tables are intended for data up to the scale of millions of features (rows). The number of columns you can create in a table is limited to 1600. We also recommend not creating tables larger than 100GB. If you have tabular data with hundreds of millions or even billions of rows or expect you may run into some of the limits above then please reach out to and let us know about your specific use case and needs. You can always get in touch at support@descarteslabs.com!
What vector file formats are supported with Tables?
ESRI shapefile in a .zip file archive, GeoJSON, line-delimited GeoJSON (also known as GeoJSON Text Sequences), and CSV are all supported file types you can upload to one of your Tables. Tables also supports insertions/uploads from Pandas and GeoPandas dataframes.
How do I get access to Tables?
Contact customer support (support@descarteslabs.com) to have Tables enabled for your account or organization.