mixle.inference.production.monitor module

Production model monitoring: watch for drift, retrain, and swap – with provenance and a DOE hook.

A Monitor wraps a fitted model + its estimator + a reference (training) sample. Feed it each production batch via check() (drift report only) or update() (check, and on drift retrain a fresh model – recording a new provenance header – and swap it in). Every action is appended to a history for audit. suggest_samples() ties in mixle.doe so a drift signal (or any model objective) can drive where to collect new data – space-filling by default, or active-learning against a model objective.

class Monitor(model, estimator, reference, *, psi_threshold=0.25, ks_threshold=0.2, loglik_shift_threshold=-0.5)[source]

Bases: object

Stateful monitor for one deployed model: drift detection + retrain-and-swap + DOE-driven sampling.

Parameters:

  • model (Any)

  • estimator (Any)

  • reference (Any)

  • psi_threshold (float)

  • ks_threshold (float)

  • loglik_shift_threshold (float)

check(current)[source]

Drift report of current (production) data against the reference under the current model.

Parameters:

current (Any)

Return type:

DriftReport

update(current, *, retrain=True, combine_reference=True, **fit_kw)[source]

Check drift on current and, if drift is flagged and retrain, fit a fresh model (with a new provenance header) and swap it in. Returns {report, action, model, header} and appends to history. combine_reference retrains on reference + current (else current only).

Parameters:
Return type:

dict

suggest_samples(bounds, n=10, *, method='lhs', objective=None, seed=None)[source]

Use mixle.doe to propose where to collect new data (e.g. after drift, or to meet an objective). method='lhs'/'sobol' gives a space-filling batch over bounds (list of (lo, hi)); pass an objective(x)->float to switch to active learning (ALC/ALM) that targets where the model is most informative.

Parameters:
Return type:

Any