Question 31
Domain 4You are employed as an ML engineer at a social media company, and your current project involves creating a visual filter for users' profile photos. This entails training an ML model to identify bounding boxes around human faces. Your goal is to integrate this filter into your company's iOS-based mobile application with minimal code development while ensuring that the model is optimized for efficient inference on mobile devices. What steps should you take?
Correct answer: A
Explanation
Vertex AI AutoML Vision can train an object detection model to find "bounding boxes around human faces," which matches the task. Using the "export for Core ML" option produces a model optimized for iOS inference, enabling "minimal code development" because Core ML is the native Apple framework for on-device ML.
Why each option is right or wrong
A. Train a model using Vertex AI AutoML Vision and use the "export for Core ML" option.
Vertex AI AutoML Vision is the managed object-detection service for training models that return bounding boxes, which is the required output for locating faces in profile photos. The iOS deployment requirement is met by exporting the trained model with the Core ML option, since Core ML is Apple’s native on-device ML format and is designed for efficient inference on iPhone/iPad with minimal app-side integration work.
B. Train a model using Vertex AI AutoML Vision and use the "export for Coral" option.
C. Train a model using Vertex AI AutoML Vision and use the “export for TensorFlow.js” option.
D. Train a custom TensorFlow model and convert it to TensorFlow Lite (TFLite).