rocci 0.3.1

Distribution-free simultaneous confidence bands for ROC curves

rocci

Distribution-free simultaneous confidence bands for ROC curves.

rocci is a simple interface to easily add uncertainty estimates to your ROC curve. It draws a simultaneous confidence band, which maintains the specified confidence of capturing the entire true (population) ROC. rocci uses a new nonparametric method by default that is likely to work with nearly all common data distributions.

from rocci import roc_band

band = roc_band(y_true, y_score)
band.plot()
print(band.summary())

Why rocci

rocci is designed to:

• just work. It does the right thing off the shelf for nearly any data set.
• drop in to your workflow. It natively integrates with sklearn, torch, statsmodels, PyMC/arviz, and pandas/polars data.
• be fast. A rust backend blazes through the algorithm.
• make a minimal footprint. The only hard dependency is numpy.
• clear an unreasonably high bar of rigor. Method is validated with millions of simulations across diverse data sets, implementation is verified with exacting tests.
• support an open ecosystem. Permissive MIT license, easy extensibility.

If you are comfortable adding a normality assumption to get tighter bands, rocci yields the tighter "Working-Hotelling" band, but also carefully checks the normality assumption and warns you when it looks dicey.

Installation

pip install rocci            # prebuilt wheels (no need for rust toolchain)

# optional plotting support
pip install 'rocci[plot]'

Details in the installation guide.

Docs: https://ndelaneybusch.github.io/rocci · Changelog: CHANGELOG.md · Contributing (including the release process): CONTRIBUTING.md