This post is co-authored by Sharon Xu Program Manager, Azure Notebooks.
Today we are very proud to announce the next set of productivity features and improvements for the notebook experience. Since we announced the GA release of Notebooks in Azure Machine Learning (Azure ML), we have learned a lot from our customers. Over the past few months, we have incrementally improved the notebook experience while simultaneously contributing back to the open source nteract project. The Azure ML team recently released a robust set of new functionalities designed to improve data scientist productivity and collaboration in Azure ML Notebooks.
We have spoken to several data scientists and developers to fully understand the additional features needed to improve productivity while developing machine learning projects. From feedback, we have found that users constantly needed the following enhancements to speed up their workflow: a clear indication that a cell has finished running, a way to templatize common code excerpts, a way to check variable contents, and more. The following list is a culmination of the most highly requested productivity features:
Figure 1: (1) Cell status bar (2) Variable explorer
Figure 2 (1) Notebook snippets panel, showing all useful snippets
Figure 3: IntelliCode in Azure ML Notebooks
Figure 4: (1) Table of content pane (2) Markdown side by side
An increasing number of data scientists and developers are creating notebooks collaboratively and sharing these notebooks across their team We heard feedback that most users feel like they are missing adequate tools to edit notebooks simultaneously or share their notebooks with a broader audience. Users often resort to screen shares and calls to complete or present work within a notebook. We recently just release a few new features to help combat some of these issues:
Figure 5: Live Co-editing in Azure ML
Figure 6: Export Notebooks as Python and more in Azure ML
To begin using these features in Azure ML Notebooks, you will first need to create an Azure Machine Learning. Your Azure ML workspace serves as your one-stop-shop for all your machine learning needs, where you can create and share all your machine learning assets.
Once you have your workspace set up, you can get started using all the features in the Azure ML Notebooks experience. The notebooks experience aims to provide you with an integrated suite of data science tools. Users can start working with a highly productive and collaborative Jupyter notebook editor directly in their workspace as well as quickly access other ML assets such as experiment details, datasets, models, and more.
With the addition of this host of features, notebooks in Azure ML aims to improve every aspect of your development needs – collaboration, code editing, debugging. Give these features a try and leave your feedback. The feedback provided by our community is what drives us to improve and build new features. As we continue to push out new releases, keep an eye out, because the team has a few more exciting features coming out soon.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.