blr_active/active_learning/mod.rs
1//! Active learning module for precision-driven sensor calibration.
2//!
3//! Implements Algorithms 1–5 from Plan 03: online active learning that guides
4//! users to measure where model uncertainty is highest, detecting noise floors,
5//! and iterating until a user-specified precision goal is reached.
6//!
7//! # Module Structure
8//!
9//! - [`variance`] — Algorithm 1: posterior variance computation
10//! - [`acquisition`] — Algorithm 2: variance-maximizing acquisition function
11//! - [`precision`] — Algorithm 3: precision assessment and goal checking
12//! - [`noise_floor`] — Algorithm 4: noise floor (plateau) detection
13//! - [`orchestration`] — Algorithm 5: synchronous calibration state machine
14//! - [`multi_sensor`] — multi-sensor session management
15pub mod acquisition;
16pub mod multi_sensor;
17pub mod noise_floor;
18pub mod orchestration;
19pub mod precision;
20pub mod variance;
21
22// Convenience re-exports for the most commonly used items
23pub use acquisition::RecommendedSample;
24pub use noise_floor::{
25 detect_noise_floor, detect_noise_floor_default, NoiseFloorConfig, NoiseFloorDiagnostic,
26};
27pub use orchestration::{
28 CalibrationSession, IterationOutcome, PrecisionRecord, SampleRecord, SessionConfig,
29};
30pub use precision::{assess_precision, percentile, PrecisionAssessment, PrecisionStatus};
31pub use variance::{posterior_std, posterior_std_grid, posterior_variance};