AI-900 Question 152
Select 4A company wants to build an image classification solution to identify different types of fruits. Which features of an Azure-based image classification solution should they consider to ensure accurate and efficient results?
- A
The ability to train a custom model using labeled images specific to the fruits being classified.
- B
The availability of pre-trained models for generic image classification tasks.
- C
The requirement to manually write algorithms for feature extraction.
- D
The ability to evaluate the model's accuracy using a validation dataset.
- E
The option to deploy the model to edge devices for real-time classification.
Show answer and explanation
Correct answers: A, B, D, E
Explanation
Azure image classification solutions provide a variety of features, including the ability to train custom models, use pre-trained models, validate model accuracy, and deploy models to edge devices. These features help build robust and scalable solutions without requiring users to manually handle complex processes like feature extraction.
- A. Correct.
Correct: Custom models allow the company to train the solution on a dataset of labeled images specific to their use case, improving accuracy for their specific classification task.
- B. Correct.
Correct: Pre-trained models can be used as a starting point or for generic classifications, reducing the time and effort required to build a solution.
- C. Incorrect.
Incorrect: Azure image classification solutions, such as those using Azure Custom Vision, do not require manual algorithm writing for feature extraction as this process is automated.
- D. Correct.
Correct: Evaluating the model's accuracy using a validation dataset is a key step in ensuring the model performs well before deployment.
- E. Correct.
Correct: Azure image classification solutions, like those using Azure IoT Edge, support deploying models to edge devices for scenarios requiring real-time processing.