v3.14.0
Quick links
New feature highlights
Pipeline editor: Configure pipeline parameters for Kubeflow Pipelines
Pipeline parameters allow for customization of pipeline runs and pipeline exports, without to need to modify the pipeline. A pipeline parameter is a typed variable that can be applied to generic or custom nodes. The Kubeflow Pipelines pipeline editor includes a new PIPELINE PARAMETERS tab, where pipeline parameters can be defined.
To make a parameter value available to a generic node, select it from the pipeline parameters list:
Jupyter notebooks and scripts can access the selected parameters and their associated values using environment variables.
To pass a parameter value to a custom node, choose parameter as input and select the desired parameter:
Note that for custom nodes the parameter list only includes parameters that are type-compatible.
You can customize parameter values in the pipeline editor during pipeline submission and pipeline export:
You can also customize parameter values in the pipelines UI in the Kubeflow Central Dashboard.
Pipeline editor: Customize GPU vendor (Kubeflow Pipelines only)
For generic components components you can now specify a custom GPU vendor. Note that the appropriate device plugin must be installed in Kubernetes or node execution will fail. The default vendor remains nvidia.com/gpu.
Pipeline editor: Specify custom pipeline export filename
You can now customize the filename when exporting pipelines. Custom filenames make it easier to retain multiple versions of the same pipeline.
Pipeline editor: Use markdown in comments
You can now use markdown in comments.
Support for Python 3.11
Python 3.11 is now officially supported:
- Elyra can be deployed in Python 3.11 environments
- Generic components can be configured to use runtime images that have Python 3.11 installed
What's Changed
New Features
- Allow for custom export file name by @salonee13 in #2999
- Support config GPU vendor by @typhoonzero in #3029
- Add support for pipeline parameters by @kiersten-stokes and @marthacryan in #3001
Bug Fixes
- Fix code of conduct by @ptitzler in #3027
- Validation: reject binary property input files for custom components by @ptitzler in #3026
- Properly escape string property inputs by @ptitzler in #3048
Other
- Remove broken doc badge and add openssf best practices badge by @ptitzler in #3013
- Update GH pull request template by @akchinSTC in #3015
- Removed unused KeyValueList class by @salonee13 in #3003
- Add reference to latest Elyra blog post to the documentation by @ptitzler in #3011
- Update repository with LFAI onboarding requirements by @akchinSTC in #2980
- Address some code scan issues by @kevin-bates in #3021
- Add and update KFP processor tests by @ptitzler in #3022
- lint Jupyter notebooks by @ptitzler in #3031
- Address warnings produced by tests by @kevin-bates in #3032
- Add Python 3.11 support by @ptitzler in #3033
- Fix generic component schema to unblock frontend changes to file browse feature by @kiersten-stokes in #3040
- Update test dependencies to min of project, address dependabot warnings by @kevin-bates in #3042
- Remove deprecated legacy python package checks by @akchinSTC in #3050
- Add version cap to Jupyterlab by @akchinSTC in #3051
- Export documentation by @salonee13 in #3039
- Add new integration test for testing custom export filename by @kiersten-stokes in #3052
- Cap jupyter-events to <0.5.0 due to jsonschema version conflict by @akchinSTC in #3055
- Add documentation for pipeline parameters by @kiersten-stokes in #3053
- Move build system to use pyproject.toml by @akchinSTC in #3038
New Contributors
- @typhoonzero made their first contribution in #3029
Full Changelog: v3.13.0...v3.14.0
