API reference

The pyFIES public API is intentionally small. Most users only need:

• RaschModel — the sklearn-style class wrapping the full pipeline.
• FAO_2014_2016 — the global reference standard.
• DEFAULT_FIES_ITEMS — canonical item names.

For users who want direct access to the numerical primitives (e.g., to build custom workflows or to validate intermediate quantities), the numerical core is also documented.

Module layout

pyfies
├── RaschModel         — sklearn-style fit/equate/prevalence pipeline
├── FAO_2014_2016      — reference standard constant
├── DEFAULT_FIES_ITEMS — the 8 standard FIES item names
└── core
    ├── cml            — weighted Conditional Maximum Likelihood Rasch fit
    ├── gamma          — log-stable elementary symmetric functions
    ├── person         — post-hoc person parameter MLE
    ├── equating       — iterative scale-and-shift equating
    └── prevalence     — Gaussian-mixture prevalence assignment