A retail company uses Azure Computer Vision to analyze customer traffic in stores. They deploy a custom object detection model to count customers and detect occupancy. After deployment, the model consistently underestimates the number of customers during peak hours. The company has retrained the model with more data but the issue persists. What is the most likely cause?
Trap 1: The model is not being batch-processed for inference.
Batch processing affects throughput, not accuracy.
Trap 2: The model is overfitting to the training data.
Overfitting would cause poor generalization, but the issue is specific to peak hours, not overall.
Trap 3: The Computer Vision API version is outdated.
API version does not systematically affect model accuracy for custom models.
- A
The model is not being batch-processed for inference.
Why wrong: Batch processing affects throughput, not accuracy.
- B
The training data does not adequately represent peak-hour scenarios.
Data drift or lack of representative samples for peak hours leads to underestimation during those times.
- C
The model is overfitting to the training data.
Why wrong: Overfitting would cause poor generalization, but the issue is specific to peak hours, not overall.
- D
The Computer Vision API version is outdated.
Why wrong: API version does not systematically affect model accuracy for custom models.