Expand description
ragdrift — five-dimensional drift detection for production RAG systems.
This crate is a thin re-export of ragdrift_core. It exists so that
Rust users can write cargo add ragdrift and get the canonical entry
point without remembering the -core suffix. There is no behavior or
API here that isn’t already in ragdrift-core; pin whichever crate you
prefer and the version will track.
For the Python wheel, see https://pypi.org/project/ragdrift/.
§Quick example
use ndarray::Array2;
use ragdrift::detectors::EmbeddingDriftDetector;
let baseline = Array2::<f32>::zeros((50, 16));
let mut current = Array2::<f32>::zeros((50, 16));
current += 1.0;
let detector = EmbeddingDriftDetector::new(0.1);
let score = detector.detect(baseline.view(), current.view()).unwrap();
assert!(score.score > 0.0);Modules§
- detectors
- The five drift detectors. Each is a lightweight struct configured once and invoked many times — the cost is in the sample data, not the detector.
- error
- Error type returned by all fallible ragdrift operations.
- stats
- Statistical primitives used by the detectors. Each lives in its own submodule and is unit-tested independently.
- types
- Public data types: dimensions, scores, reports, baseline snapshots.
Structs§
- Baseline
Snapshot - A captured baseline used as the reference distribution for future detection.
- Drift
Report - A unified report aggregating per-dimension scores.
- Drift
Score - A single per-dimension drift measurement.
Enums§
- Drift
Dimension - The five drift dimensions ragdrift monitors.
- RagDrift
Error - All errors raised by the core crate.
Type Aliases§
- Result
- Result alias used across the crate.