Azure Cloud AI Services

Like Amazon and Google, Microsoft provides many AI tools and services for data scientists and engineers to develop ML models in the Azure cloud. Microsoft cloud AI services include Azure Machine Learning workspaces, which is a fully managed platform for data scientists to build, train, and deploy machine learning models, and Azure Cognitive Services, which helps developers build cognitive intelligence into applications, based on pre-trained AI models and APIs for common ML tasks. Azure cloud AI services integrate seamlessly with other Microsoft cloud services and tools. In this chapter, we will cover the following topics:

• Azure Machine Learning workspace, which is an end-to-end platform for data scientists to develop ML models, including data collection, model training and deploying, and other AI capabilities.
• Azure Cognitive Services, which enables developers to easily add cognitive features into their applications. The Azure Cognitive Services...

Azure ML workspaces

An Azure ML workspace allows you to build, deploy, and manage ML models at scale. It provides a centralized workspace for data scientists, machine learning engineers, and developers to collaborate on machine learning projects, with the following features:

• An Azure ML workspace is an end-to-end suite for organizing and managing ML assets such as datasets, models, notebooks, experiments, and pipelines/resources. It provides a centralized location for team collaboration, version control, and resource management.
• It integrates with Jupyter notebooks and provides an interactive environment for developing and running code, visualizing data, and documenting the ML process.
• It supports dataset versioning and management so you can register and track different versions of datasets for ML model training and evaluation. Datasets can be stored within the workspace or referenced from external data sources.
• It allows you to organize and track different iterations...

Azure Cognitive Services

Azure Cognitive Services is a collection of cloud-based AI services provided by Microsoft Azure. Like AWS AI services and Google AI APIs, Azure Cognitive Services offers pre-built AI capabilities that can be easily integrated into applications, thus allowing developers to enhance their applications with intelligent features using pre-trained AI algorithms. Some key features of Azure Cognitive Services are the following:

• Vision: Azure Cognitive Services includes Computer Vision, which can analyze images and extract information such as objects, texts, and facial expressions. The Face service enables facial detection, recognition, and emotion analysis, and the custom vision option allows you to train custom image classification models.
• Speech: The Speech service provides speech recognition, text-to-speech, and speech translation capabilities.
• Language: Azure offers several language-related services. Text Analytics performs sentiment analysis, key...

Azure OpenAI Service

Azure OpenAI Service is newly launched at the time of writing this book. Azure OpenAI Service supports many common AI workloads including ML, computer vision, NLP, conversational AI, anomaly detection, and knowledge mining. It also supports generative AI tasks including the following:

• Generating natural language:
  • Text completion: generate and edit text
  • Embeddings: search, classify, and compare text
• Generating code: generate, edit, and explain code
• Generating images: generate and edit images

We will explore some of these tasks now:

1. Create an Azure OpenAI resource:

Log in to the Azure portal and choose Azure OpenAI. Click Create Azure OpenAI. Fill in the basics as shown in Figure 14.18:

Figure 14.18 – Create Azure OpenAI

2. Get into Azure OpenAI:

Click the ayeopenai resource we have created to enter the Azure OpenAI landing page, where you can explore, develop, and deploy the OpenAI...

Summary

In this chapter, we focused on three main topics of machine learning in Azure. We started with the Azure ML workspace, which is an end-to-end ML development suite. Next, we looked at Azure Cognitive Services, which allows us to leverage prebuilt ML models for application development. Lastly, we checked out Azure OpenAI Service, which provides access to GPT-3, Codex, DALL·E, and other pretrained Large Language Models (LLMs). By the end of this chapter, you have learned the basic concepts of Azure ML and acquired the necessary skills for ML model training and application development using the Azure AI services. In the next chapter, we will discuss Azure cloud security.

Practice questions

1. A cloud engineer team need access to the Azure ML workspace to run a script as a job. What role should they be granted?

A. Azure ML Data Scientist

B. Azure ML Compute Operator

C. Reader

D. Contributor

2. A cloud engineer team need to run a single script to train a model. What job best fits their requirements?

A. command

B. pipeline

C. sweep

D. archive

3. And what role should they be assigned?

A. Azure ML Data Scientist

B. Azure ML Compute Operator

C. Reader

D. Contributor

4. A cloud engineer team need to create/delete Azure ML registries. What role should they be assigned?

A. Azure ML Data Scientist

B. Azure ML Compute Operator

C. Reader

D. Contributor

Questions 5-8 are based on the following use case:

A cloud engineer is developing applications using Azure Cognitive Services to extract insights from unstructured text. The cloud engineer has created some Azure Cognitive Services resources.

5. The...

Answers to the practice questions

1. A
2. A
3. B
4. D
5. B
6. A
7. A
8. A
9. A
10. C

Further reading

For further insights into what you’ve learned in this chapter, refer to the following links:

• https://learn.microsoft.com/en-us/azure/machine-learning/
• https://azure.microsoft.com/en-in/products/machine-learning
• https://azure.microsoft.com/en-us/products/cognitive-services
• https://learn.microsoft.com/en-us/azure/cognitive-services/
• https://azure.microsoft.com/en-us/products/cognitive-services/openai-service
• https://learn.microsoft.com/en-us/azure/cognitive-services/openai/
• https://learn.microsoft.com/en-us/azure/cognitive-services/openai/concepts/models
• https://azure.microsoft.com/en-us/blog/product/azure-openai-service/