Choosing the Right Machine Learning Service in Azure (Part 3 of 4)

 





To read part 1, please click here
To read part 2, please click here
To read part 4, please click here






Custom ML Services

Since platform services are built on the top of IaaS services containing useful abstractions and functionalities for the relevant domain, Azure makes sure to provide as many PaaS services for different specialized domains as possible. A domain called ML is also laced with various services in order to build custom ML models. Now we will discuss some of the most popular Custom ML PaaS services.

Azure Machine Learning Studio (Classic)

  • It's Azure's most widely used tool to build, train, optimize, and deploy ML models via a GUI as well as drag and drop, block-based programming model. 

  • It offers a robust and large number of features, algorithms, and extensions through R and Python support; and comes under one of the oldest managed cloud services for ML in Azure.

  • It also provide built-in building blocks for clustering, regression, classification, anomaly detection, and recommendation as well as data and statistical and text analysis.

  • Its functionality can also be extended with the help of custom code blocks for Python or R.  

Azure Machine Learning Designer

  • As it is fully integrated with Azure Machine Learning, it can easily access as well as share all the resources and assets within the workspace.

  • You can also use GUI-based creation of ML pipelines while collaborating with the other data engineers and data scientists in the same workspace. The same compute resources can be shared among them that can adjust according to the need of the developers.

  • Data ingestion, preprocessing, cleaning, and feature extraction stages can also be shared with the other users in the workspace while focusing solely on ML tasks in the designer.

  • Although GUIs to create block-based ML training pipelines are not for everyone, but, if you still go for it, then, Azure Machine Learning Designer is the best choice for you. 

Azure Automated Machine Learning

  • It's a no-code tool, that allows you to define a dataset, a target column, and ML tasks in order to train an ML model from a spreadsheet.

  • This service can also be created by the Azure ML workspace, and thus have access to all the resources and assets present in the workspace.

  • It can also support regression, classification, and time-series forecasting tasks along with the explanations for each one of them, which in turn helps Excel users as well as ML engineers to quickly build and deploy a great baseline model.

  • It also allows access to all the training runs, trained models, training scores, useful built-in metrics, visualization, and insights.








To read part 1, please click here
To read part 2, please click here
To read part 4, please click here






































Comments

Popular posts from this blog

Query, Visualize, & Monitor Data in Azure Sentinel

Planning for Implementing SAP Solutions on Azure (Part 2 of 5)

Work with String Data Using KQL Statements