Product Updates
      Back to home
      1. Knowledge Base
      2. Descartes Labs Platform
      3. Product Updates

      May 2024 Release

      3.1.0 Python Client Release Notes

      Python client updates

      We are excited to announce the v3.1.0 release of the descarteslabs Python client.

      General

      • Due to declining support for Python 3.8 across the ecosystem, we have discontinued our support for Python 3.8. It is expected that the client will continue to function until Python 3.8 is End of Life (October 2024), but we can no longer test new releases against this version.

      Catalog

      • The Catalog Storage Blob deletion methods have been enhanced to support waiting for completion of the operation. When a blob is deleted, it is removed immediately from the catalog and a background asynchronous task is launched to clean up the contents of the blob from the backing storage. If a blob is deleted and then a new blob with the identical id is immediately created and uploaded before this background task completes, it is possible for the background task to end up deleting the new blob contents. As of this release the Blob instance and class delete methods return a BlobDeletionTaskStatus object which provides a wait_for_completion method which can be used to wait until the background task completes and it is safe to create a new blob with the same id. For the Blob.delete_many method, the wait_for_completion=True parameter can be used to wait for all the supplied blobs to be completely deleted. Note that in the case of the Blob.delete class method, this is a very slight breaking change, as it used to return True or False, and now instead returns a BlobDeletionTaskStatus or None, which have the same truthiness and hence are very likely to behave identically in practical use.

      Interface updates

      Workbench

      New server options with descarteslabs v3.1.0 and dynamic compute v1.1.4 are now available. The GPU server for this new release now utilizes CUDA 12. Most all Python packages have been updated to their most recent compatible releases. We've also added in support for Torch, Torchvision, and ONNX (including the runtime and various utilities for converting other model formats to ONNX). The examples notebooks contain some minor updates and enhancements.

      Data Updates

      The following datasets have been added to the Catalog and are available to all users:

      • nasa-opera:sentinel-1:rtc:v1: NASA-OPERA generated Radiometric Terrain Correction (RTC) 30m data based on Sentinel-1 burst SLCS, hosted at Alaska Satellite Facility (ASF)

      • nasa-opera:sentinel-1:cslc:v1: NASA-OPERA generated geocoded Sentinel-1 burst SLCS, hosted at Alaska Satellite Facility (ASF)

      • sar:umbra:GEC-latlon:v0: UMBRA's open data bucket. Data is geocoded assuming a constant ellipsoid height surface and has been scaled to Byte for best visualization

      • kgclim:historical:v1: KGClim is a 1-km-resolution global dataset of Köppen-Geiger climate classification maps and bioclimatic variables for ten historical periods from 1979 to 2017

      • kgclim:future:rcp2.6:v1: KGClim is a 1-km-resolution global dataset of Köppen-Geiger climate classification maps and bioclimatic variables predicted for four future periods from 2020 to 2100

      • kgclim:future:rcp4.5:v1: KGClim is a 1-km-resolution global dataset of Köppen-Geiger climate classification maps and bioclimatic variables predicted for four future periods from 2020 to 2100

      • kgclim:future:rcp6.0:v1: KGClim is a 1-km-resolution global dataset of Köppen-Geiger climate classification maps and bioclimatic variables predicted for four future periods from 2020 to 2100

      • kgclim:future:rcp8.5:v1: KGClim is a 1-km-resolution global dataset of Köppen-Geiger climate classification maps and bioclimatic variables predicted for four future periods from 2020 to 2100

      • meta:wri:chm:v1: Global high-resolution Canopy Height Maps. Canopy Height Maps were developed using AI models on high-resolution worldwide Maxar satellite imagery for over half of the globe

      • meta:ciesin:hrsl:v1: Estimates the number of people living within 30-meter grid tiles for more than 160 countries

      • gfw:agbd:2000:v1: Global, wall-to-wall map of aboveground biomass (AGB) at approximately 30-meter resolution


      Note: For more details please refer to the Github release notes