https://github.com/kamilest/conformal-rnn

Implementation for Stankevičiūtė et al. "Conformal time-series forecasting", NeurIPS 2021.
https://github.com/kamilest/conformal-rnn

Last synced: about 1 month ago
JSON representation

Implementation for Stankevičiūtė et al. "Conformal time-series forecasting", NeurIPS 2021.

• Host: GitHub
• URL: https://github.com/kamilest/conformal-rnn
• Owner: kamilest
• Created: 2021-05-06T15:53:56.000Z (over 3 years ago)
• Default Branch: master
• Last Pushed: 2022-08-30T10:20:39.000Z (over 2 years ago)
• Last Synced: 2024-04-09T21:32:25.061Z (9 months ago)
• Language: Jupyter Notebook
• Homepage:
• Size: 14.8 MB
• Stars: 60
• Watchers: 3
• Forks: 13
• Open Issues: 1
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

• awesome-conformal-prediction - Conformal time-series forecasting

README

        
# Conformal time-series forecasting

Implementation for [Stankevičiūtė et al. 

"Conformal time-series forecasting", NeurIPS 2021](https://proceedings.neurips.cc/paper/2021/hash/312f1ba2a72318edaaa995a67835fad5-Abstract.html).

This codebase builds on the implementation for

"Frequentist Uncertainty in Recurrent Neural Networks

via Blockwise Influence Functions" (ICML 2020), available at

https://github.com/ahmedmalaa/rnn-blockwise-jackknife

under the BSD 3-clause license. 

## Installation

Python 3.6+ is recommended. Install the dependencies from [`requirements.txt`](./requirements.txt).

## Replicating Results

To replicate experiment results, run the notebooks:

* [`synthetic.ipynb`](./synthetic.ipynb)

* [`synthetic_bjrnn.ipynb`](./synthetic.ipynb) (**Note:** this notebook should be executed with requirements as per [`requirements_bjrnn.txt`](./requirements_bjrnn.txt).)

* [`medical.ipynb`](./medical.ipynb)

You can download the publicly available data for this work at the [UKHSA data dashboard](https://ukhsa-dashboard.data.gov.uk/respiratory-viruses/covid-19) and [UCI Machine Learning Repository](https://archive.ics.uci.edu/dataset/121/eeg+database) (note the data format may have changed by maintainers of the datasets). As the MIMIC-III dataset [requires PhysioNet credentialing](https://mimic.mit.edu/docs/gettingstarted/) to access, you must become a credentialed user on PhysioNet before accessing the data. To get access to the dataset as used in this work, please contact @DrShushen or @ahmedmalaa and provide proof of your PhysioNet credentialing.

## Citing

If you use our code in your research, please cite:

```

@inproceedings{stankeviciute2021conformal,

  author = {Stankevičiūtė, Kamilė and Alaa, Ahmed M. and {van der Schaar}, Mihaela},

  title = {Conformal time-series forecasting},

  booktitle = {Advances in Neural Information Processing Systems},

  year = {2021}

}

```