Failed potash-kiln batches are 2 percent of rows. Accuracy looks fine if the model always predicts pass. Which official pre-training metric should the team compute first?
Select an answer to reveal the explanation.
Short Explanation
Think of failed kiln batches at 2 percent, and accuracy looking great if the model always says pass. Compute class imbalance first. A constant-pass score is a hide, not a balance check.
Full Explanation
Class imbalance (CI) is the official pre-training metric for a rare numeric or tabular label. A constant-pass model hides that 2 percent. Rekognition and a Domain 2 ROC curve are not that prep check.