IBM and Linux Foundation AI launched Machine Learning eXchange (MLX) as a one-stop shop for trusted data and AI artefacts in open source and open governance.
MLX provides a collection of free, open-source, state-of-the-art deep learning models for common application domains. The curated list includes deployable models that can be run as a microservice on Kubernetes or OpenShift and trainable models where users can provide their own data to train the models.
It provides developers and data scientists with automated sample pipeline code generation to execute registered models, datasets, and notebooks, and a pipelines engine powered by Kubeflow Pipelines on Tekton, the core of Watson Studio Pipelines.
It also provides a registry for Kubeflow Pipeline Components, dataset management by Datashim, and a serving engine by KFServing. The contributors to the project said:
“Due to the large number of steps that need to be worked on in the Data and AI lifecycle, the process of building a model can be bifurcated amongst various teams and large amounts of duplication can arise when creating similar Datasets, Features, Models, Pipelines, Pipeline tasks, etc. This also poses a strong challenge for traceability, governance, risk management, lineage tracking, and metadata collection.
To solve the problems mentioned above, contributors need a central repository where all the different asset types like Datasets, Models, and Pipelines are stored to be shared and reused across organizational boundaries.