Azure AI & ML: Optimize Language Models for AI Applications

本课程是 Exam Prep DP-100: Microsoft Azure Data Scientist Associate 专项课程的一部分

位教师：Whizlabs Instructor

访问权限由 Coursera Learning Team 提供

2个模块

深入了解一个主题并学习基础知识。

中级等级

推荐体验

8 小时完成

灵活的计划

自行安排学习进度

2个模块

深入了解一个主题并学习基础知识。

中级等级

推荐体验

8 小时完成

灵活的计划

自行安排学习进度

您将获得的技能

MLOps (Machine Learning Operations)
Model Evaluation
Model Deployment
Artificial Intelligence and Machine Learning (AI/ML)
Retrieval-Augmented Generation
Performance Tuning
Data Pipelines
Microsoft Azure
AI Workflows
Prompt Engineering
Scalability
Cloud Deployment
Continuous Monitoring
技能部分已折叠。显示 6 项技能，共 13 项。

要了解的详细信息

可分享的证书

添加到您的领英档案

作业

5 项作业

授课语言：英语（English）

了解顶级公司的员工如何掌握热门技能

了解关于 Coursera for Business 的更多信息

Petrobras, TATA, Danone, Capgemini, P&G 和 L'Oreal 的徽标

积累特定领域的专业知识

本课程是 Exam Prep DP-100: Microsoft Azure Data Scientist Associate 专项课程专项课程的一部分

在注册此课程时，您还会同时注册此专项课程。

向行业专家学习新概念
获得对主题或工具的基础理解
通过实践项目培养工作相关技能
获得可共享的职业证书

该课程共有2个模块

This course is designed to provide a comprehensive foundation in Azure Machine Learning, equipping learners with the skills to deploy, manage, and optimize ML models efficiently. Participants will begin by exploring model deployment and consumption in Azure ML, understanding how to operationalize machine learning solutions in production environments.

The course progresses to managing and evaluating models, covering key concepts such as performance monitoring, retraining strategies, and best practices for ensuring model accuracy. Learners will gain expertise in Azure AutoML workflows, from data preparation to model selection and evaluation, ensuring automated yet effective ML development. Additionally, the course covers key aspects of MLOps, enabling seamless integration with Azure services for scalable and secure machine learning operations. This course is structured into multiple modules, each featuring lessons and video lectures that provide theoretical insights and hands-on practice. Participants will engage with approximately 3:00–4:00 hours of instructional content, ensuring both conceptual understanding and practical application. To reinforce learning, graded and ungraded assignments are included within each module to test the ability of learners in real-world scenarios. Module 1: Azure AI Foundry: End-to-End Model Development & Optimization Module 2: Optimize model training with Azure Machine Learning By end of this course, you will be able to learn Understand the concepts of Azure AI Foundry, including its role in model optimization, fine-tuning, and retrieval-augmented generation (RAG) strategies. Learn how to explore and manage the Model Catalog and Collections within Azure AI Foundry and ML, and use compute resources effectively. Gain practical experience testing and manually evaluating prompts in the Azure AI Foundry portal playground, including tracking prompt variants. Discover how to create and configure search indexes in the Azure portal, using Azure AI Search for enhanced data retrieval and model deployment.

This module provides a comprehensive understanding of Azure AI Foundry and its capabilities, equipping learners with the skills to leverage AI models for advanced applications. Participants will explore key concepts such as Retrieval Augmented Generation (RAG) for enhancing AI-driven responses, fine-tuning strategies for optimizing model performance, and best practices for deploying AI models in production environments. The module covers the Azure AI Foundry model catalog, compute considerations, and how to test and refine language models using the interactive playground. Learners will gain expertise in manually evaluating prompts, defining and tracking prompt variants, and utilizing Azure AI Search to create efficient search indexes. By the end of this module, participants will be prepared to work with Azure AI Foundry and ML tools, ensuring scalable and high-performing AI solutions for various enterprise applications.

涵盖的内容

9个视频2篇阅读材料2个作业1个讨论话题

9个视频总计51分钟

Azure AI Foundry: Overview and Demo 5分钟
Retrieval Augmented Generation (RAG) in Azure AI and ML: Overview 5分钟
Optimizing Models: Fine-Tuning, RAG and Application Strategies 6分钟
Model Catalog and Collections [Azure AI Foundry and ML]-Overview 5分钟
Model Catalog and Collections [Azure AI Foundry and ML]-Compute 4分钟
Test a deployed language model in the playground 7分钟
How to manually evaluate prompts in Azure AI Foundry portal playground 6分钟
Define and track prompt variants 5分钟
Quickstart: Create a search index in the Azure portal - Azure AI Search 8分钟

2篇阅读材料总计60分钟

Welcome to the Course 30分钟
Azure AI Foundry: End-to-End Model Development & Optimization - Overview 30分钟

2个作业总计70分钟

Azure AI Foundry: End-to-End Model Development & Optimization - Graded Assignment 40分钟
Optimize language models for AI applications - Practice Assignment 30分钟

1个讨论话题总计20分钟

Meet & Greet 20分钟

This module provides a comprehensive understanding of preparing machine learning workflows for production using Azure Machine Learning, equipping learners with the skills needed for scalable and efficient deployment. Participants will explore best practices for transitioning from notebooks to scripts, executing command jobs with parameters, and integrating MLflow for model tracking and evaluation. The module covers pipeline creation, custom components, and prebuilt workflows—including an Automobile Price Prediction pipeline—to automate and optimize ML processes. Learners will gain expertise in working with metrics, hyperparameters, and data transformation techniques, ensuring model performance and reliability. Additionally, the module emphasizes key aspects of production readiness, such as managing resources, tracking ML models, and refining training workflows for real-world applications. By the end of this module, participants will be equipped with practical knowledge to implement and manage robust ML pipelines within Azure Machine Learning effectively

涵盖的内容

19个视频2篇阅读材料3个作业

19个视频总计119分钟

Preparing code for production scenarios 8分钟
Convert a notebook to a script 7分钟
Run a script as a command job 8分钟
Use parameters in a command job 6分钟
Exploring The use of MLflow For Tracking Models 7分钟
Track Metrics with Machine Learning Flow 7分钟
Integrating ML Flow in Model Training Flow 7分钟
Viewing Metrics and Evaluating Models 8分钟
Creating and Using components in Azure Machine Learning 6分钟
Creating Pipelines in Azure Machine Learning 5分钟
Creating a Custom Pipeline 7分钟
Components in Azure Machine Learning 7分钟
Prebuilt Pipelines (Automobile Price Prediction) 10分钟
Understanding Metrics 4分钟
Load and Transform Data in Azure Machine Learning using notebooks 4分钟
Parameters and Hyperparameters 4分钟
Key Hyperparameters in Machine Learning 6分钟
Exam Tips 5分钟
Conclusion, What's Next, Job Roles, and Best Practices 3分钟

2篇阅读材料总计60分钟

Preparing Code, Integrating ML Flow, and Tracking Models - Overview 30分钟
Key Takeaways 30分钟

3个作业总计100分钟

Optimize model training with Azure Machine Learning - Graded Assignment 40分钟
Preparing Code, Integrating MLflow, and Tracking Models - Practice Assignment 30分钟
Building Pipelines, Components, and Optimizing Models - Practice Assignment 30分钟