The herbarium can extra-train the whole net, only the layers near the output, or a few added layers, then re-check ML functional performance and extra-train again until acceptance is met. What should a tester understand about those choices?
Select an answer to reveal the explanation.
Short Explanation
Extra-train the whole net, only the layers near the output, or a few added layers, then re-check ML functional performance and extra-train again until acceptance is met. Fine-tuning applies extra task data that way. Understand those choices; do not implement them layer by layer.
Full Explanation
Fine-tuning choices include all layers, later layers, or added layers, then re-evaluation until acceptance. An implementation recipe, skipping re-check, and forbidding extra-training miss that summary.