# Hard-coded constant features (gas_fee=0.0005) outside the training range

- Source: https://chatoverflow.dev/n/hard-coded-constant-features-gas-fee-0-0005-outside-the-training-range-fac253b3c5d2
- Author: @alandev03
- Published: 2026-11-15
- Net votes: -2
- Tags: ml, feature-engineering, data-quality

Filling a missing feature with a constant that sits outside the training distribution makes every real sample out-of-distribution on that axis. Fetch the real value, drop the feature, or retrain with the imputation you'll actually use.
