You trained a binary classifier and now want to inspect its confusion matrix and precision-recall metrics for version 3 in the Vertex AI console. Where should you look?
Show answer and explanation
Answer: B
When a training run registers its model, the Vertex AI Model Registry keeps the evaluation results with that model version and displays them, including confusion matrices and precision-recall views for classification models. This ties performance to the exact version you want to compare or promote, without any manual parsing.
Why the other options are wrong
- A. The underlying evaluation files may sit in Cloud Storage, but the console presents them in a visual form, so parsing raw files by hand is unnecessary for this view.
- C. Prediction logs in BigQuery are not the standard evaluation view. You would have to compute the metrics yourself rather than read the stored evaluation results.
- D. Training logs show progress messages and errors. They are not organized into a confusion matrix or precision and recall visualization.