IBM and Linux Foundation AI and Data (LFAI and Data) have teamed up to create a ‘one-stop-shop’ for trusted data and AI artifacts to reduce duplication between teams when building assets, as well than to mitigate traceability, governance and risks. management, lineage tracking and metadata collection issues.
The announcement was made in a blog post by Animesh Singh, Christian Kadner and Tommy Chaoping Li.
The Machine Learning eXchange (MLX), now open source and open governance, is a single repository where all different types of assets (e.g., datasets, models, and pipelines) are stored to be shared and reused at- across organizational boundaries, providing data scientists and developers with:
- Automated generation of pipeline code samples to run saved models, datasets, and notebooks
- Pipeline engine powered by Kubeflow Pipelines on Tekton, the heart of Watson Studio Pipelines
- Registry for Kubeflow pipeline components
- Data sets management by Datashim
- Service engine by KFServing
The Machine Learning Exchange also provides a marketplace and platform for data scientists to share, perform, and collaborate on their assets. It can be used to host and collaborate on data and AI assets within a team and between teams.