The following changes were made in ADS 2.8.4:

• Added support for creating ADSDataset from pandas dataframe.
• Added support for multi-model deployment using Triton.
• Added support for model deployment local testing in ads opctl CLI.
• Added support in ads opctl CLI to generate starter YAML specification for the Data Science job, Data Flow application, Data Science model deployment, and pipeline resources.
• Added support for invoking model prediction locally with predict(local=True).
• Added support for attaching customized score.py when preparing model.
• Added status check for model deployment delete, activate, or deactivate APIs.
• Added support for training and verifying SparkPipelineModel in Data Flow.
• Added support for generating score.py for GPU model deployment.
• Added support for setting defined tags in Data Science jobs.
• Improved model deployment progress bar.
• Fixed bug when using ads opctl CLI to run jobs locally.
• Fixed bug in Dataflow magic when using archive_uri in Data Flow configuration.

For more information, see Data Science and take a look at our Data Science blog.