https://github.com/fxia22/neuralfdr

Software accompanying "NeuralFDR: Learning Discovery Thresholds from Hypothesis Features"
https://github.com/fxia22/neuralfdr

Last synced: 1 day ago
JSON representation

Software accompanying "NeuralFDR: Learning Discovery Thresholds from Hypothesis Features"

• Host: GitHub
• URL: https://github.com/fxia22/neuralfdr
• Owner: fxia22
• Created: 2017-04-14T23:19:48.000Z (over 7 years ago)
• Default Branch: master
• Last Pushed: 2018-06-14T20:12:05.000Z (over 6 years ago)
• Last Synced: 2024-04-16T01:53:49.921Z (7 months ago)
• Language: Python
• Homepage: https://arxiv.org/pdf/1711.01312.pdf
• Size: 17.6 MB
• Stars: 5
• Watchers: 4
• Forks: 5
• Open Issues: 1
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# NeuralFDR

Software accompanying "NeuralFDR: Learning Discovery Thresholds from Hypothesis Features", NIPS 2017

## Dependencies 

You will have to install PyTorch to run the code, follow the instructions from http://pytorch.org

## Download the data

Download the data used in the paper from this [dropbox folder](https://www.dropbox.com/sh/wtp58wd60980d6b/AAA4wA60ykP-fDfS5BNsNkiGa?dl=0).

## Train a NeuralFDR model

```

python train.py --data data/data_airway.csv --dim 1 --out airway

```

The report will be available in airway folder

## Citation

If you use this code, please cite

```

@inproceedings{xia2017neuralfdr,

  title={NeuralFDR: Learning Discovery Thresholds from Hypothesis Features},

  author={Xia, Fei and Zhang, Martin J and Zou, James Y and Tse, David},

  booktitle={Advances in Neural Information Processing Systems},

  pages={1540--1549},

  year={2017}

}

```