# AI-102 — Question 3

**Type:** drag_and_drop
**Topics:** custom_vision_api, drag_and_drop, sequence_workflow

## Question

You are developing an application that will recognize faults in components produced on a factory production line. The components are specific to your business. You need to use the Custom Vision API to help detect common faults. Which three actions should you perform in sequence? To answer, move the appropriate actions from the list of actions to the answer area and arrange them in the correct order.

## Correct Answer

_See scenario._

## Explanation

The correct sequence starts with initializing the training dataset because you need to define a structure for the data that will be used to train the Custom Vision model. The dataset acts as the container for your images, labels, and related metadata, which are essential to training the classifier. Without this initial step, the subsequent actions would lack the foundational data organization required for model training.

Next, uploading and tagging images is crucial because the model needs labeled examples (tags) to learn how to classify faults. This step supplies the algorithm with sufficient data to identify patterns related to faults. Finally, training the classifier model uses this tagged dataset to build a model capable of recognizing common component defects. Skipping any of these actions would make the training process incomplete, preventing the algorithm from functioning effectively.

**Reference:** Azure Custom Vision — Training Basics

---
Source: https://hiexam.net/q/microsoft/AI-102/3  
Practice (tracked): https://hiexam.net/study/AI-102/practice