Free Webinar: Using Jupyter/IPython Notebooks in Azure ML
Jupyter notebooks (formerly IPython) provide a multi-lingual REPL (Read Eval Print Loop) canvas for data scientists and developers to explore ideas. You can enter some code and get a response. The response can be program output, graph, etc. The notebook can mix code and text for documentation purposes.
Jupyter notebooks are now available as a service on Azure ML, fully integrated with the Azure ML Studio, and it works from any modern browser on any OS.
In this webinar we'll cover what Jupyter notebooks are, their integration with Azure ML Studio, and the operationalization of code to run on the Azure ML backend.
To register, click here or on the image below:
ML Blog Team