Part I: Foundations of Sensory AI
Chapter 5  [F][C]

Sensor Data Engineering and Leakage-Safe Datasets

Placeholder chapter page. Sections are produced by the book-skills pipeline.

Sections

  1. 5.1 Formats for telemetry, waveforms, event streams, and point clouds
  2. 5.2 Window construction and labeling
  3. 5.3 Leakage in sensor datasets (the field's most common error)
  4. 5.4 Device, user, site, session, and machine splits
  5. 5.5 Normalization and per-device calibration
  6. 5.6 Annotation quality, weak labels, and label delay
  7. 5.7 Reproducible, versioned sensor pipelines and data contracts

Lab 5

build leakage-safe subject- and device-disjoint splits for human activity recognition; measure the accuracy gap vs a naive random split.