- Services: Data Science
- Release Date: March 2, 2023
The following changes were made in ADS 2.8.2:
Removed support for Python 3.7.
DataScienceMode.create()to support the timeout argument and automatically extract the region from the signer and signer configuration.
Support Jupyter Notebook as
entrypointwhen defining Data Science jobs with
Support environment variable substitution in Data Science job names and output URI.
Support JSON serialization of list and dictionary when assigning them as Data Science jobs environment variables.
Support saving the notebook to output URI even if the job run failed when running a Data Science job using
job.build()method to Data Science job to load default values from the environment.
DataScienceJob.fast_launch_shapes()method to list fast launch shapes available for Data Science job.
HuggingFacePipelineModelclass to support prepare, save, deploy, and predict for Hugging Face pipelines.
Updated Data Science job run YAML representation to include configurations inherited from the job.
Fixed custom conda environment not showing in Data Science Job YAML specification.
Fixed an issue where model saving was failing in notebook sessions when
Fixed “Unknown archive format” error in
ads.jobs.PythonRuntime, when the source code folder name ends with
zip. The supported archive files are
For more information, see Data Science and take a look at our Data Science blog.