|IBM Contributes Two AI Libraries To Linux Foundation|
|Written by Kay Ewbank|
|Thursday, 02 March 2023|
IBM has contributed two open source projects to the Linux Foundation. The Elyra and Claimed libraries are used in IBM’s AI products, including IBM Watson Studio and IBM Cloud Pak for Data.
IBM Watson is a platform that uses natural language processing and machine learning to discover insights in large sets of unstructured data. It is a question-answering system that can answer questions phrased using natural language.
IBM Cloud Pak for Data is a data and AI platform with a data fabric that makes all data available for AI and analytics, on any cloud. Elyra is a collection of AI-centric JupyterLab extensions that can be used to create reproducible, scalable, component-based, low-/no-code data science pipelines.
Elyra’s tools can be used to create, modify, and execute script, notebook, and component-based pipelines via a visual pipeline editor using drag and drop. The editor comes with script editor extensions and debuggers. Elyra was added as a top-level image/project in the Open Data Hub with Red Hat. It is being integrated with the installation of Kubeflow pipelines (data science pipelines) as part of the stack.
Claimed stands for Component Library for AI, ML, ETL and Data Science. It provides coarse-grained, opinionated, tested, and verified processing components. with the aim of enablinf low-code/no-code rapid prototyping-style programming.
The library provides ready-made components for various business domains, supports multiple computer languages, works on different data flow editors and command-line tools, and runs on various execution engines, including Kubernetes, Knative, Kubeflow, Apache Airflow, or plain Docker.
IBM wants developers and data scientists to use the libraries for the daily work, including production-ready software, and to contribute new components to Elyra and Claimed. They say that because Elyra is modular, users can pick and choose which extensions they want to use. IBM's main request is that users try Elyra's visual pipeline editor with Kubeflow pipelines, and to a lesser extent Apache Airflow.
or email your comment to: firstname.lastname@example.org
|Last Updated ( Wednesday, 01 March 2023 )|