If you have used Azure Machine Learning before, you probably have tried up to making the prediction models.

However, if you want to start consuming Web APIs you will need to start programming to develop apps and websites – and that increases the difficulty for non-coders.

So today, I will be showing you how to develop a website that uses Azure Machine Learning without any coding at all – in matter of just 5 minutes.

 

First, you need to prepare a web service in Azure Machine Learning.

Go to Microsoft Azure Machine Learning Studio https://studio.azureml.net/ and create one by clicking on “Deploy Web Service” at the bottom of the screen.


 

The screen probably switched to something like below.

Make note of the jumbled letters below the “API key” and keep it in a notepad to use it for later. Once done, click on the “REQUEST/RESPONSE” link.


 

Now copy the web address below “Request URI” as you will need this later as well.

 

Next we will deploy a new website to consume the web services from Azure Machine Learning.

From the Azure Portal, click on “New” and “Web + Mobile”.


Search for “azure ml” and you will find “Azure ML Request-Response Service Web App”. Click on it.


 

Click on “Create”


 

Enter the address for the website, subscription, resource groups. If you are using this by yourself for testing, you can choose the “Free” plan and it will work fine.


 

Now wait for the deploy to finish…


 

Once done, now we will access the newly created website.


 

You can directly enter the URL if you remember, or click on the link from Azure Portal.


 

If you access the website for the first time, initial setup is launched.

Here, you need to enter the URL and the API Key you had taken note of from Azure Machine Learning Studio.


 

Setup the title and description for your site.

You can setup “Alias” to show different field names instead of what you had setup in Azure ML.

Max and Min values can also be setup for fields that are numeric.


 

Once setup is finished, click on “Save Changes”.


 

If saving goes well, it will show “Success!” like below.

To try out the website, click on “Go to Home Page”

If you want to access the settings again, you can always access by going to URL + settings.aspx


 

Now lets try the website out!

Fill in the details, and click on “Submit”.


And there you go, the results from Azure ML is shown as “Scored Label”.

If you want to adjust the details to be shown for results, you need to change the settings in Azure Machine Learning Studio and not the website. You can change the results by using the “Select Columns” module within Azure ML studio to restrict the columns to extract.