This course explores the Azure Custom Vision service and how you can use it to create and customize vision recognition solutions. You'll get some background info on what the service is before looking at the various steps for creating image classification and object detection models, uploading and tagging images, and then training and deploying your models.
Learning Objectives
- Use the Custom Vision portal to create new Vision models
- Create both Classification and Object Detection models
- Upload and tag images according to your requirements
- Train and deploy the configured models
Intended Audience
- Developers or architects who want to learn how to use Azure Custom Vision to tailor a vision recognition solution to their needs
Prerequisites
To get the most out of this course, you should have:
- Basic Azure experience, at least with topics such as subscriptions and resource groups
- Basic knowledge of Azure Cognitive Services, especially vision-related services
- Some developer experience, including familiarity with terms such as REST API and SDK
So we're in the Azure portal, and I want to start by creating a cognitive services resource. We're not going to the end of this resource creation, but I just would like to illustrate a couple of things that we have mentioned on the slides. To do that, let's click here at the top, and select Create a resource.
On the new page that opens, I'll type Cognitive on the search bar, and Cognitive Services is the first option available. So let's click on it, and then click on the Create button. On the next page that opens, notice the message here at the top saying that this resource will consolidate vision, language, search, and speech within a single API key. That's a compromise between simplicity and granularity on your resource management, right? Although it will be very simple to manage a single Azure resource, that also means that you'll have to distribute the same API key to potentially multiple people in your company.
If I scroll down to pricing tier, notice that I only have Standard S0 as an option. Although cognitive services is probably one of the cheapest Azure products, there will be some cost associated with it. So if you want to play with a free tier, you need to create a Custom Vision resource instead. Let's do that. I'll click here at the top and select again Create a resource. But this time, I'll type Custom Vision, click on it, and then click on the Create button.
On the next screen, notice that I can create a resource for Training, Prediction, or Both. I'll leave Both selected. This will create both endpoints with the same name, with a -prediction suffix for the prediction endpoint. So if this is not what you want, you might prefer to create each endpoint separately. I'll leave my subscription as is and select my Cloud Academy resource group. Under name, I'll type CACustomVision. Then I have the option to select both the location and pricing tier for each resource. I'll select Southeast Asia for the location, as this is my closest Azure region, and if I click here on Pricing Tier, notice that I do have a free tier, so let me click here on F0 to select it. I'll also repeat the same operation for the prediction resource. Then I'll click here on Tags, and there's not much to configure here, so I'll click on Next: Review and Create, and then on the Create button.
This should take a few seconds, but when it's done, let's click here on the Go to resource button. But apart from the Keys and Endpoints button here, where you can see your access information, there's barely anything else to do on this page. All the other options are pretty standard and common to any other Azure resource.
Where are we going to spend most of the time after the resource is created is on the Custom Vision portal. So let's go there. I'll open a new tab and type www.customvision.ai. Then here on the center of the page, I'll click on Sign in, and now I'm on the home page of the portal, where I can see my projects, currently none, and have the option to create a new one. If I click here on the gear at the top, I can see all my endpoints. Let's expand both training and prediction endpoints, and as you can see, they're using independent API addresses and keys.
If I want to change the pricing tier for any of them, I can use this button over here to do it. Let's click to come back to the home page, and then click on the New Project button. Here's where we're going to create our Custom Vision models, but that's a subject for the next videos.
Emilio Melo has been involved in IT projects in over 15 countries, with roles ranging across support, consultancy, teaching, project and department management, and sales—mostly focused on Microsoft software. After 15 years of on-premises experience in infrastructure, data, and collaboration, he became fascinated by Cloud technologies and the incredible transformation potential it brings. His passion outside work is to travel and discover the wonderful things this world has to offer.