One of the biggest strengths of Azure ML Studio is the ease with which you can deploy models to the cloud, to be consumed by other applications. Once an ML model is trained, as demonstrated in the previous section, it can be exported to ML Studio Web Services with just a few clicks. Deployment creates a web API for the model, which can be called from any internet-connected application. The model takes the features as input data and produces a predicted value as output. By deploying models to the ML Studio Web Service, there is no need to worry about the underlying server infrastructure. The computing resources and maintenance are handled entirely by Azure.
The following subsections show how to deploy an already trained model to the web service and how to test a model with user input. In the final subsection, we'll show how to import and...
