Introduction to Azure ML : ML-Series (2 of 10)

Azure ML provides a graphical tool for managing the machine learning process, a set of data preprocessing modules, a set of machine learning algorithms, and an API to expose a model to applications.

The components that Azure ML provides are the following:

  1. ML Studio, a graphical tool that can be used to control the process from beginning to end. Using this tool, people on the machine learning team can apply data pre-processing modules to raw data, run experiments on the prepared data using a machine learning algorithm, and test the resulting model. Once an effective model is found, ML Studio also helps its users deploy that model on Microsoft Azure.
  2. A set of data preprocessing modules.
  3. A set of machine learning algorithms.
  4. An Azure ML API that lets applications access the chosen model once it’s deployed on Azure.

Here is Azure ML Portal.

http://studio.azureml.net

AzureML

Once you sign in into Azure ML Portal it must look like this.

AzureMLWorkbench

Here is Azure ML Portal Workspace

AzureML Workspace

Azure ML Studio is a GUI tool that supports the complete ML process, from preprocessing data to deploying a model and even expose it as web service.

Azure ML Studio enable user drag and drop datasets on canvas and then use can processs that data again by dropping data preprocessing modules on same canvas, later soem machine learning algorithms required to be dropped on same canvas and more onto its design surface. The user can connect these together graphically using arrows and then execute the result.

For example, a data scientist might use Azure ML Studio to connect a dataset that holds prepared data with the machine learning algorithm he’s chosen for a particular experiment. Once this is done, he can use Azure ML Studio to run the experiment and evaluate the model it creates. When he has what he believes is the best possible model, he can use ML Studio to deploy this model to Microsoft Azure, where applications can use it. Rather than relying on ad hoc, largely manual mechanisms to do all of this, ML Studio provides a single tool for controlling the entire machine learning process.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.