by Andrie de Vries
Earlier today Microsoft announced that Jupyter Notebooks are now available with the R Kernel as a service in Azure Machine Learning (ML) Studio.
- Jupyter is an easy to use and convenient way of mixing code and text in the same document.
- Unlike other reporting systems like RMarkdown and LaTex, Jupyter notebooks are interactive – you can run the code snippets directly in the document
- This makes it easy to share and publish code samples as well as reports.
- You can create and edit R notebooks directly in the Azure ML cloud, without having to create virtual machines.
- Using the AzureML R package, you can interact between R notebooks and your Azure ML environment. Specifically, you can:
- Download and upload datasets
- Download intermediate datasets from Azure ML experiments
- Publish and consume web services
- You can easily publish your notebook in the Cortana Intelligence Gallery, either privately or publicly.
See notebooks in action
This video has a gentle introduction on how to use Jupyter Notebooks:
See some sample notebooks
We have created some examples that demonstrates the possibilities:
- Introduction to Azure ML R notebooks
- Predicting breast cancer using data from AzureML
- Changing R plot options in Jupyter
- Accessing datasets with R using the AzureML R package
- Connect Azure ML Studio with R using the AzureML package
- Fitting a Gradient Boosting Machine (GBM) and publishing to AzureML using R
- Fitting a LASSO regression model and publishing to Azure ML using R
You can find these notebooks, and others, in the Cortana Intelligence Gallery:
How to try the notebooks in Azure ML
You can try the notebook service, as well as all of the Azure ML Studio, using a free guest account.
- Go to http://studio.azureml.net/
- Click “Sign up”.
- For completely anonymous access, choose “Guest Workspace”
- However, note that in a guest workspace you won't be able to save your work.
- In a “Free Workspace” you can save your work and publish notebooks.
- Take the option to do the quick tour, if you wish, otherwise skip this step
- Click the Notebook tab on the left
- You now have a range of sample notebooks to choose from. You can select either a new R notebook, or one of the sample notebooks.
- Notice that some of the notebooks are Python and some are R.
- You can tell the difference by looking for the small “R” icon in the top right hand corner of each tile
- Click on “Blank R Notebook”, or open one of the other R sample notebooks
The Jupyter project is an open source project with many contributors. Thank you to the Jupyter team, as well as the IRKernel team who did the integration work between Jupyter and R.
Tell us what you think
You can help us prioritize features and otherwise tell us what you think by completing this one-minute survey