# `Monty.Frameworks.Models.EvidenceMatching.FeatureEvidence.Calculator`
[🔗](https://github.com/d-led/ex.tbp.monty/blob/v0.1.0/packages/ex_tbp_monty_core/lib/monty/frameworks/models/evidence_matching/feature_evidence/calculator.ex#L1)

How much of a match a node's stored features are — the calculator family.

Mirrors `tbp/monty/frameworks/models/evidence_matching/feature_evidence/calculator.py`,
a module holding a protocol and its default implementation. The Elixir split
has the same shape:

  * `Monty.Frameworks.Models.EvidenceMatching.FeatureEvidence.Calculator.FeatureEvidenceCalculator`
    is the protocol, as a behaviour;
  * `...Calculator.DefaultFeatureEvidenceCalculator` is the implementation an
    experiment gets unless it names another.

A calculator is handed the stored features, the observed features and the
tolerances, and answers with a score per node. It never touches graph memory:
Monty injects the calculator into the hypotheses updater precisely so that
scoring can be replaced without touching the search.

---

*Consult [api-reference.md](api-reference.md) for complete listing*