Configuring Azure Custom Vision for Image Classification
Short answer
Project Type: Classification; Classification Types: Multiclass (Single tag per image); Domains: General (compact).
Question
You are building a custom vision model that will be deployed as part of an iOS app.
You have images of cats and dogs. Each image contains either a cat or a dog.
You need to use the Azure Custom Vision service to detect whether the image is of a cat or a dog.
How should you configure the project in the Azure Custom Vision portal? To answer, select the appropriate options in the answer area.
NOTE: Each correct selection is worth one point.
Key Takeaways
The key is recognizing that classifying entire images as cat or dog requires Classification > Multiclass, not Object Detection.
Learn how to configure Azure Custom Vision projects for image classification tasks, specifically distinguishing between multiclass and object detection settings.
Common mistake: Learners often confuse Classification with Object Detection, selecting the latter when they only need to identify what is in the image rather than locate it.
Expert Analysis
Why the Answer Is Correct
The scenario involves categorizing images as either cats or dogs, which is a classic classification task. Since each image contains only one animal type, 'Multiclass (Single tag per image)' is the correct configuration. This setting ensures the model learns to assign a single label to the entire image.
Why Other Candidate Answers Are Wrong
Object Detection would be appropriate if the task required drawing bounding boxes around multiple animals within a single image. Binary classification is used for two outcomes like 'spam/not spam', but here we have distinct classes (cat/dog) potentially expanding later, making Multiclass the standard choice for this specific exam context. General Domain is suitable for general photos, whereas specialized domains are for specific industries.
Community Comment Notes
As Amelia233 noted, the correct selections are Project Type: Classification, Classification Types: Multiclass, and Domains: General. The explanation confirms that since the goal is to categorize the overall image rather than locate bounding boxes, Classification is the correct path.
Exam Strategy
Always read the requirement carefully: if you need to find WHERE an object is, choose Object Detection. If you need to know WHAT the image is, choose Classification.
Frequently Asked Questions
When should I use Object Detection vs Classification in Custom Vision?
What is the difference between Binary and Multiclass classification?
Discussion 1 contributions
Compiled from public community threads and reader submissions, lightly edited for clarity. Handles and vote tallies are shown as recorded, and comments reflect the views of individual contributors.
From public community threads 1
-
Amelia233 👍 1Correct Answers Project Types: Classification Classification Types: Multiclass (Single tag per image) Domains: General (compact) Explanations Project Types: Classification is selected because the goal is to categorize the overall image rather than locate bounding boxes around objects. Classification Types: Multiclass (Single tag per image) is selected because each image contains either a cat or a dog, so only one label applies per image. Domains: General (compact) is required because the model will be deployed as part of an iOS app (on-device edge export like CoreML requires a compact domain).
Reader comments & corrections
No reader corrections yet — spotted an error or have a note? Share it below.
Related Analysis
Practice All AI-103 Questions
Access 88 questions with complete answers and detailed explanations.
View Full AI-103 Practice Test →