Azure Machine Learning을 사용하여 AI 솔루션 빌드

데이터 과학자
Machine Learning
Azure Portal

Azure Machine Learning은 기계 학습 모델을 학습, 배포, 관리 및 모니터링하기 위한 클라우드 플랫폼입니다. Azure Machine Learning Python SDK를 사용하여 엔터프라이즈급 AI 솔루션을 만드는 방법에 대해 알아봅니다.

필수 조건

이 학습 경로는 Python과 Scikit-Learn, PyTorch, Tensorflow와 같은 오픈 소스 프레임워크로 기계 학습 모델을 교육한 경험이 있다고 가정합니다. 그렇지 않은 경우, 이 학습 경로를 시작하기 전에 기계 학습 모델 만들기 학습 경로를 먼저 완료하는 것이 좋습니다.

이 학습 경로의 모듈

Azure Machine Learning 소개

Azure Machine Learning을 사용하여 모델을 학습시키고 작업 영역에 등록하는 방법을 알아봅니다.

Data is the foundation of machine learning. In this module, you will learn how to work with datastores and datasets in Azure Machine Learning, enabling you to build scalable, cloud-based model training solutions.

One of the key benefits of the cloud is the ability to use scalable, on-demand compute resources for cost-effective processing of large data. In this module, you'll learn how to use cloud compute in Azure Machine Learning to run training experiments at scale.

Orchestrating machine learning training with pipelines is a key element of DevOps for machine learning. In this module, you'll learn how to create, publish, and run pipelines to train models in Azure Machine Learning.

Azure Machine Learning Service를 사용한 ML 모델 등록 및 배포 방법을 알아봅니다.

Machine learning models are often used to generate predictions from large numbers of observations in a batch process. To accomplish this, you can use Azure Machine Learning to publish a batch inference pipeline.

Choosing optimal hyperparameter values for model training can be difficult, and usually involved a great deal of trial and error. With Azure Machine Learning, you can leverage cloud-scale experiments to tune hyperparameters.

Azure MAchine Learning에서 자동화된 Machine Learning을 사용하여 데이터에 가장 적합한 모델을 찾는 방법을 알아봅니다.

Data scientists have an ethical (and often legal) responsibility to protect sensitive data. Differential privacy is a leading edge approach that enables useful analysis while protecting individually identifiable data values.

Many decisions made by organizations and automated systems today are based on predictions made by machine learning models. It's increasingly important to be able to understand the factors that influence the predictions models make.

Machine learning models can often encapsulate unintentional bias that results in unfairness. With Fairlearn and Azure Machine Learning, you can detect and mitigate unfairness in your models.

After a machine learning model has been deployed into production, it's important to understand how it is being used by capturing and viewing telemetry.

Changing trends in data over time can reduce the accuracy of the predictions made by a model. Monitoring for this data drift is an important way to ensure your model continues to predict accurately.