|Elyra JupyterLab AI Extension Improves Usability|
|Written by Kay Ewbank|
|Thursday, 13 August 2020|
Elyra AI Toolkit is now available in a 1.0 release. The open source project extends the JupyterLab user interface to simplify the development of data science and AI models.
Elyra is an open-source project that provides a set of AI-centric extensions to JupyterLab. Launched iat the end of April, 2020 as we reported at the time, Elyra was developed by IBM, and has features including a Notebook Pipelines visual editor; the ability to run notebooks as batch jobs; and hybrid runtime support (based on Jupyter Enterprise Gateway). It also lets you execute Python scripts within the editor, and has notebook versioning based on Git integration.
Elyra’s Notebook Pipeline Editor is designed to make it easier to convert multiple notebooks into batch jobs or workflows. This means Jupyter users can move their experiments to run on cloud-based resources so they run faster. The Notebook Pipeline Editor now has enhanced inline user documentation and validation capabilities. It also has improved performance for pipeline submission to Kubeflow Pipelines runtime. New optimized dependency handling providing a much faster submission of pipelines, and there's support for “bring your own image” to be used as the environment to execute Notebooks on the external runtime.
Reusable code snippet support has now been added so researchers can reuse task-oriented blocks of code. The new code snippets extension lets you discover, create, and insert reusable snippets of code into your Notebooks, Python Scripts, or even Markdown files used for documentation directly from the JupyterLab workspace. The available code snippets as shown as a list in the left side pane, including a preview of each snippet, and an option to either copy a snippet or insert it directly inline.
Elyra also has a Table of Contents extension for Notebooks and Python Scripts. This makes it easier to find specific sections in Notebooks or function definitions in Python Scripts. The extension has been enhanced in this release to support navigating Python Scripts.
or email your comment to: firstname.lastname@example.org