Skip to main content

FEERCI: A python package for EER confidence intervals

Project description

FEERCI: A Package for Fast non-parametric confidence intervals for Equal Error Rates
******************************************


**FEERCI** is an opinionated, easy-to-use package for calculating EERs and non-parametric confidence intervals efficiently. It offers a single method, ``feerci.feerci``, that returns both an EER and CI for provided impostor and genuine scores. The only dependency is numpy.

Installation
=======
``pip install feerci``

What's New
=======
0.2.0
--------
- Switched output arguments around, to make more intuitive sense
0.1.0
--------
- Initial release of package


License
=====
**FEERCI** is distributed under the MIT license

Version
=====
0.2.0

Examples
======
Calculating just an EER::

import feerci
import numpy as np
impostors = np.random.rand(100)
genuines = np.random.rand(100)
eer,_,_,_ = feerci.feerci(impostors,genuines,is_sorted=False,m=-1)

Calculating an EER and 95% confidence interval::

eer,ci_lower,ci_upper,bootstrapped_eers = feerci.feerci(impostors,genuines,is_sorted=False)

Calculating an EER and 99% confidence interval::

eer,ci_lower,ci_upper,bootstrapped_eers = feerci.feerci(impostors,genuines,is_sorted=False,ci=.99)

Calculating an EER and 99% confidence interval on 1000 bootstrap iterations::

eer,ci_lower,ci_upper,bootstrapped_eers = feerci.feerci(impostors,genuines,is_sorted=False,m=1000,ci=.99)



Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Filename, size & hash SHA256 hash help File type Python version Upload date
feerci-0.2.0-cp34-cp34m-manylinux1_x86_64.whl (388.9 kB) Copy SHA256 hash SHA256 Wheel cp34
feerci-0.2.0-cp35-cp35m-manylinux1_x86_64.whl (386.2 kB) Copy SHA256 hash SHA256 Wheel cp35
feerci-0.2.0-cp36-cp36m-manylinux1_x86_64.whl (395.2 kB) Copy SHA256 hash SHA256 Wheel cp36

Supported by

Elastic Elastic Search Pingdom Pingdom Monitoring Google Google BigQuery Sentry Sentry Error logging AWS AWS Cloud computing DataDog DataDog Monitoring Fastly Fastly CDN SignalFx SignalFx Supporter DigiCert DigiCert EV certificate StatusPage StatusPage Status page