Calibration of Neural Networks using Splines

This repository is the official implementation of ICLR 2021 paper: Calibration of Neural Networks using Splines.

This code is for research purposes only.

Any questions or discussions are welcomed!

Installation

Setup python virtual environment.

virtualenv -p python3 venv
source venv/bin/activate                                 
pip3 install -r requirements.txt
mkdir saved_logits

Setup

Download the logits for different data and network combinations from here and put them under saved_logits folder.

Recalibration

To find a recalibration function and evaluate the calibration:

python recalibrate.py

The results for pre-calibration and post-calibration with various metrics will be saved in csv format under out/{dataset}/{network}/beforeCALIB_results.csv and out/{dataset}/{network}/afterCALIBsplinenatual6_results.csv. Calibration graphs such as Figure 1 in the main paper will be generated under out/{dataset}/{network} folder.

Cite

If you make use of this code in your own work, please cite our paper:

@inproceedings{
gupta2021calibration,
title={Calibration of Neural Networks using Splines},
author={Kartik Gupta and Amir Rahimi and Thalaiyasingam Ajanthan and Thomas Mensink and Cristian Sminchisescu and Richard Hartley},
booktitle={International Conference on Learning Representations},
year={2021},
url={https://openreview.net/forum?id=eQe8DEWNN2W}
}

Contact