A notebook fits a scaler on the entire hop-harvest file, then splits. Validation rows have already leaked into the transform. What did the team do wrong?
Select an answer to reveal the explanation.
Short Explanation
Think of a scaler fitted on the entire hop-harvest file, then a split. Validation already leaked into the transform. Fit encoders and scalers on the training slice only.
Full Explanation
Fit feature transforms on training data only so validation and test rows cannot leak into scaling or encoding. Fitting on the whole file is that leak. It is a Domain 1 transform rule, not a Domain 2 metric item.