Using Azure Machine Learning Studio for data analysis

Using Microsoft Azure Machine Learning Studio for data analysis

In previous  posts, we discussed the advantages of cloud-based analytics platforms like Google Cloud Platform. In this post, we look at using Azure Machine Learning Studio for data analysis. Azure ML offers access to data science development tools like R and Python Jupyter Notebooks, where you can run your R and Python scripts. There is a free tier plan and we encourage you to sign up. With the plan, you can run your own analytics experiments and try out the example code on

Benefits of Using Microsoft Azure Machine Learning Studio For Data Analysis

1. Easy and Intuitive workspace for analytics development

Azure ML offers a simple and easy to understand interface. Access to the tools are point and click and there is no need to run any server commands. When you enter the platform, you are presented with easy to understand options:

Using Azure Machine Learning Studio for data analysis

2.  Develop your own machine learning process and workflow

By selecting the Experiments option, you can define the workflow of your analytics project. For example,  you will define:

  • how datasets are accessed or imported,
  • Python or R scripts to run against your data,
  • machine learning or statistical modules to apply against your data,

See the sample below:

Design your data analytics workflow with Azure ML

3. Store your datasets
The free tier allows you to import and store 10GB of data.

4. Notebooks allow you to run python scripts and libraries via your web browser
A feature that I particularly like is the access to cloud hosted Jupyter Notebooks. This feature provides data scientists with the flexibility to run python scripts and explore and investigate their datasets via the web browser. You can also install and access additional Python Libraries like Pandas, Numpy.  Installing Jupyter on your local computer can be a pain and this cloud hosted option is a good alternative.

Python data analysis with Azure Notebooks5. Store your trained machine learning models
Once you used your data and trained your models, you can now store and save them easily.

6. Web services
Once you have built your predictive model, you can publish it via Azure ML studio’s web services option. Data can be posted to these web services and users will receive the predictions.


About The Author