What is target leakage and how do you prevent it?
mediumAnswer
- Target leakage is when a feature contains information about the label that would not be available at prediction time (e.g., using post-outcome features, or fitting a scaler/encoder on the full dataset before splitting).
- Prevent it by fitting all transformations only on the training fold (use pipelines), splitting temporally when time matters, and reviewing features for future information.
Check yourself — multiple choice
- It's when the model is too small
- It's when features leak information from the label or the future
- It's the same as underfitting
- It only happens in deep learning
Leakage = using information at train time that won't be available at inference time.
#data-quality#features
Practise Supervised Learning
215 interview questions in this topic.