AI-102 Question 140
Select 3You are tasked with building a custom image classification model for identifying different types of fruits (e.g., apples, oranges, bananas). You decide to use Azure Custom Vision. Which of the following steps are necessary to successfully train and deploy this model?
- A
Create a Custom Vision project and upload labeled images of each fruit category.
- B
Manually write the algorithm to classify images based on the uploaded dataset.
- C
Train the model within the Custom Vision portal or via the Custom Vision API.
- D
Export the trained model for offline use or integrate it with an Azure service.
- E
Assign a unique key to each image in the dataset to ensure correct training.
Show answer and explanation
Correct answers: A, C, D
Explanation
To train a custom image classification model in Azure Custom Vision, you need to create a project, upload labeled images, train the model, and then deploy or export it for use. Custom Vision simplifies the process by providing an intuitive interface and tools, removing the need for manual algorithm development or unique key assignments for individual images.
- A. Correct.
Correct. Creating a Custom Vision project and uploading labeled images is the first step in training a custom image classification model.
- B. Incorrect.
Incorrect. Azure Custom Vision automatically handles the algorithm creation; you don't need to manually write the classification algorithm.
- C. Correct.
Correct. Training the model is a necessary step and can be done either through the Custom Vision portal or via its API.
- D. Correct.
Correct. Exporting the trained model or integrating it with an Azure service is a valid step after training to deploy the model for use.
- E. Incorrect.
Incorrect. Assigning unique keys to each image is not a requirement for training a Custom Vision model. The system uses labels and metadata for training.