https://github.com/pris-cv/advanceddropout
Advanced Dropout: A Model-free Methodology for Bayesian Dropout Optimization (IEEE TPAMI 2021)
https://github.com/pris-cv/advanceddropout
dropout
Last synced: about 1 year ago
JSON representation
Advanced Dropout: A Model-free Methodology for Bayesian Dropout Optimization (IEEE TPAMI 2021)
- Host: GitHub
- URL: https://github.com/pris-cv/advanceddropout
- Owner: PRIS-CV
- License: mit
- Created: 2021-03-29T02:35:16.000Z (about 5 years ago)
- Default Branch: main
- Last Pushed: 2021-06-04T02:22:00.000Z (about 5 years ago)
- Last Synced: 2025-04-24T05:46:32.395Z (about 1 year ago)
- Topics: dropout
- Language: Python
- Homepage:
- Size: 17.6 KB
- Stars: 17
- Watchers: 2
- Forks: 1
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# AdvancedDropout
Advanced Dropout: A Model-free Methodology for Bayesian Dropout Optimization (IEEE TPAMI 2021) [IEEE Xplore](https://ieeexplore.ieee.org/document/9439951 "IEEE Xplore") or [ArXiv](https://arxiv.org/abs/2010.05244 "ArXiv")
## Code List
+ main.py
+ Main file for running
+ mlp.py
+ Fully connected (FC) layers with advanced dropout
+ variationalBayesDropout.py
+ Advanced dropout
## Dataset
### CIFAR-10 (and others)
## Requirements
- python >= 3.6
- PyTorch >= 1.1.0
- torchvision >= 0.3.0
- GPU memory >= 3500MiB (GTX 1080Ti)
## Training
- Download datasets
- Train and evaluate: `python main.py` or use nohup `nohup python main.py >1.out 2>&1 &`
## Citation
If you find this paper useful in your research, please consider citing:
```
@ARTICLE{9439951,
author={Xie, Jiyang and Ma, Zhanyu and Lei, Jianjun and Zhang, Guoqiang and Xue, Jing-Hao and Tan, Zheng-Hua and Guo, Jun},
journal={IEEE Transactions on Pattern Analysis and Machine Intelligence},
title={Advanced Dropout: A Model-free Methodology for Bayesian Dropout Optimization},
year={2021},
volume={},
number={},
pages={1-1},
doi={10.1109/TPAMI.2021.3083089}}
```