Skip to content
Free public beta until December 4, 2026: every course is open, no card needed. After that, the first chapter of each course stays free and full access is a one-time $39 ($19 for up to 100 founding members chosen for helpful feedback). Your progress is kept. How the beta works

feature vector

A short, fixed-length list of numbers that sums up one window of sensor readings.

A feature vector is the ordered list of features computed from one window of samples (for example mean, RMS, peak-to-peak and zero crossings per axis), used as the input to a detector or a classifier in place of the raw window.

Examples

  • The fan's feature vector is the mean, RMS, peak-to-peak and zero crossings of each of three axes: 12 numbers, 24 bytes, every second.

  • Batching 60 feature vectors into one message a minute sends the same information in about 2.2 MB a day instead of 10.3 MB.

See also

data reduction edge computing batching
Read the full section →