Start With Machine Learning On Microsoft Azure

Introduction

This article is all about starting with Machine Learning on Microsoft Azure and it’s a continuation of my previous one. We will be discussing about getting access to Azure Machine Learning, Azure Machine Learning Studio – various features on it, creating an experiment on Azure Machine Learning Studio, and mining the results from it.

Go to my previous article to know more about Machine Learning on Microsoft Azure.

Start with Machine Learning on Microsoft Azure – Part One

Access for Azure Machine Learning

Machine Learning can be accessed free of cost on a trial basis. It is designed to make the tasks easy and it's limited with their features in production level.
Azure Machine Learning can be accessed at a cost basis via your credit card, by adding it towards your Azure account. We can integrate up to four Azure accounts.

Set your work environment for Free Azure Machine Learning!

Requirement

Microsoft Account.

Move to this URL https://studio.azureml.net/home and sign in with your Microsoft account.

You will be taken to the following page of Azure Machine Learning Studio.

Note - As mentioned earlier, this is a limited account where you will not be able to use all the features.

Added, we will also get a mail on our respective Microsoft account by which we are signed in. The mail holds a few tips on how to work with Azure Machine Learning.

Adding Machine Learning to Azure Account

Requirements

Microsoft Azure Account.

This is the next option in which we can use Azure Machine Learning from our Azure account.

Note

You should already have an Azure account in order to use this option and this doesn’t have any limitation. If you don’t have an Azure account, you can create a temporary account for a one month trial valued at $200 of credits from you – click here to create an Azure account for free. You have no limitation on it.

Accessing the Machine Learning Studio from Azure Account

Go for Microsoft Azure Portal.

Here, find the Machine Learning workspace. Go for New >> Intelligence + Analytics >> Machine Learning Workspace. Here, we will be creating a workspace and a storage account, etc.

We can also access the Azure old portal using www.manage.windowsazure.com from where we can use Machine Learning too.

Click on Data Services >> Machine Learning >> Quick Create.

We will get a notification, as shown below, once the Machine Learning Workspace is created.

And here is our Machine Learning Workspace created on Microsoft Azure with a Storage account.

Clicking on it, we can find that this Machine Learning Workspace has no restrictions. Here, we can access the workspace with the help of Machine Learning Studio. We can manage user access and we can also manage the web services.

Clicking on "Sign in to Machine Learning" will help us to move to the Azure Machine Learning Studio.

Summary

To conclude this article, we can set up a Machine Learning environment in two ways. I recommend you to go with Machine Learning on Microsoft Azure by integrating your Microsoft Azure Account rather than going for a trial Machine Learning Account. Students can get it free of cost using Dreamspark, and Startups can get access to it using Bizspark.

Follow my next article to work on creating your first experiment on Azure Machine Learning Studio.