AI-102 Question 146
Select 3You have trained a custom vision classification model in Azure AI to identify different types of fruits. After training, you need to evaluate the model's performance to ensure it meets the required accuracy and reliability for deployment. Which metrics should you analyze to effectively evaluate the model's performance?
- A
Precision
- B
Recall
- C
F1 Score
- D
Training Iteration Count
- E
Prediction Time
Show answer and explanation
Correct answers: A, B, C
Explanation
To evaluate a custom vision model's performance, it is crucial to analyze metrics that measure its accuracy and capability to correctly identify the target classes. Precision, Recall, and F1 Score are key metrics that provide insights into the model's effectiveness and reliability. While Training Iteration Count and Prediction Time may be relevant for other aspects of model development and deployment, they do not directly measure the model's classification accuracy or overall performance in identifying fruits.
- A. Correct.
Precision measures the percentage of correct positive predictions out of all positive predictions made by the model, making it a critical metric for evaluating model accuracy.
- B. Correct.
Recall measures the percentage of actual positive cases correctly identified by the model, making it essential for evaluating how well the model captures all relevant cases.
- C. Correct.
F1 Score is the harmonic mean of Precision and Recall, providing a balanced measure of the model's accuracy and reliability, particularly when dealing with imbalanced datasets.
- D. Incorrect.
Training Iteration Count refers to the number of times the model has been trained on the dataset, which is not a direct metric for evaluating model performance.
- E. Incorrect.
Prediction Time measures how long the model takes to make a prediction, which is a performance-related metric but not directly relevant to evaluating model accuracy.