          基于Azure的AI应用程序开发培训

Introduction to Artificial
IntelligenceThis module introduces
Artificial Intelligence and Machine learning.
Next, we talk about machine learning types and tasks.
This leads into a discussion of machine learning algorithms.
Finally we explore python as a popular language for machine learning solutions
and share some scientific ecosystem packages which
will help you implement machine learning. By the end of this unit
you will be able to implement machine learning models
in at least one of the available python machine learning libraries.
Standardized AI Processes and Azure Resources
This module introduces machine learning tools available
in Microsoft Azure.
It then looks at standardized approaches developed to help data analytics projects to be successful.
Finally, it gives you specific guidance on
Microsoft's Team Data Science Approach to include roles and tasks involved with the process.
The exercise at the end of this unit points you to Microsoft's documentation to implement this process
in their DevOps solution if you don't have your own.Azure Cognitive APIs
This module introduces you to Microsoft's pretrained and managed machine learning offered as
REST API's in their suite of cognitive services.
We specifically implement solutions using the computer vision api,
the facial recognition api, and do sentiment analysis by calling the natural language service.
Azure Machine Learning Service:
Model Training
This module introduces you to the capabilities
of the Azure Machine Learning Service. We explore how to create and then reference
an ML workspace. We then talk about how to train a machine learning model using the Azure
ML service. We talk about the purpose and role of experiments, runs, and models.
Finally, we talk about
Azure resources available to train your machine learning models with.
Exercises in this unit include creating a workspace,
building a compute target, and executing a training run using the Azure
ML service.Azure Machine Learning Service: Model Management and Deployment
This module covers how to connect to your workspace.
Next, we discuss how the model registry works and how to register
a trained model locally and from a workspace training run.
In addition, we show you the steps to prepare a model for deployment including identifying dependencies,
configuring a deployment target, building a container image.
Finally, we deploy a trained model as a webservice and test it by sending JSON objects to the API.