probcal

Universal post-hoc probability calibration for binary classifiers: methods, metrics, diagnostics, and auditable offsetting — numpy-only.

probcal unifies the binary calibration literature (Platt, temperature, beta, isotonic, centered isotonic, histogram binning, scaling-binning, BBQ, ENIR, Venn–Abers, spline calibration), an extensive catalog of calibration evaluation metrics and statistical tests, calibration visualization on both probability and logit scales, an auditable logit-offset (central tendency) adjustment, automatic method selection under nested validation, and two data flows (prefit and cross-validation). Primary application domain: credit-risk PD models; the package is fully general.

Status: released on PyPI, beta. The API is stable enough to build on; breaking changes bump the minor version until 1.0.

Installation

pip install probcal            # runtime: numpy only
pip install "probcal[viz]"     # + matplotlib for probcal.plots

Development setup (tests, lint, type-check):

git clone https://github.com/wlazlod/probcal && cd probcal
uv sync --extra dev

Quickstart

from probcal import BetaCalibrator, make_pd_portfolio
from probcal.metrics import calibration_guardrails

port = make_pd_portfolio(n=8000, random_state=42)   # synthetic 3% PD portfolio

g_before = calibration_guardrails(port.y, port.scores)
print(f"before: slope={g_before.slope:.3f}  intercept={g_before.intercept:+.3f}  ok={g_before.all_ok}")

cal = BetaCalibrator().fit(port.scores, port.y)
p = cal.predict_proba(port.scores)

g_after = calibration_guardrails(port.y, p)
print(f"after:  slope={g_after.slope:.3f}  intercept={g_after.intercept:+.3f}  ok={g_after.all_ok}")
print()
print(cal.interpret())

Output:

before: slope=0.968  intercept=-0.765  ok=False
after:  slope=1.000  intercept=+0.000  ok=True

Interpretation[BetaCalibrator]
parameter  value
---------  --------
a          0.875054
b          1.58922
c          -1.15227
- a = 0.875: sensitivity near s -> 0; a < 1 raises the smallest probabilities (model was overconfident in the low tail), a > 1 deepens them
- b = 1.589: sensitivity near s -> 1; the mirrored reading for the high tail
- c = -1.152: base-rate shift of -1.152 log-odds, odds factor 0.316
- identity map corresponds to (a, b, c) = (1, 1, 0)
- a != b (gap -0.714): asymmetric tail distortion that no symmetric (Platt/temperature) map could express

Automatic selection, model wrapping, offsetting, and threshold translation:

from probcal import CalibratedModel, CalibratorSelector, PlattCalibrator

sel = CalibratorSelector().fit(s_cal, y_cal)             # nested CV, log-loss criterion
wrapped = CalibratedModel(model, PlattCalibrator(), flow="prefit").fit(X_cal, y_cal)
wrapped.offset_to(target_mean=0.031)                     # auditable central-tendency stage
lo_z, hi_z = wrapped.interval_inverse(0.0, 0.02, space="logit")   # "PD <= 2%" in raw margins

Why probcal

¹ betacal, venn-abers, ml-insights. ² prdm0/probcal (P. R. Diniz Marinho), unaffiliated — see the FAQ. Verified against v0.2.0, 2026-08-08. ³ Platt, temperature, beta, isotonic, histogram binning; its multiclass methods (Dirichlet, vector scaling, one-vs-rest) are out of probcal's binary scope.

Performance note: the PAVA family and special functions are hand-rolled numpy/stdlib; profiling on real workloads has not shown them to be a bottleneck. Rust acceleration is deliberately out of scope unless benchmarks on production-sized portfolios say otherwise.

Documentation

Built with mkdocs-material; run locally with uv run mkdocs serve. Start with Getting started, then the Concepts chapters — the package's theoretical foundation — and the executed PD calibration walkthrough notebook. The Visualization chapter is a gallery of every plot, regenerated deterministically by docs/scripts/generate_figures.py.

License

MIT. See LICENSE and docs/LICENSING.md for the conceptual-reference policy on GPL-licensed R packages.