https://github.com/yeonghyeon/skip-ganomaly

Implementation of Skip-GANomaly with MNIST dataset
https://github.com/yeonghyeon/skip-ganomaly

anomaly-detection generative-adversarial-network generative-neural-network mnist-dataset

Last synced: 27 days ago
JSON representation

Implementation of Skip-GANomaly with MNIST dataset

• Host: GitHub
• URL: https://github.com/yeonghyeon/skip-ganomaly
• Owner: YeongHyeon
• License: mit
• Created: 2019-11-28T04:47:00.000Z (over 5 years ago)
• Default Branch: master
• Last Pushed: 2019-11-28T04:47:59.000Z (over 5 years ago)
• Last Synced: 2025-04-04T16:42:36.312Z (about 2 months ago)
• Topics: anomaly-detection, generative-adversarial-network, generative-neural-network, mnist-dataset
• Language: Python
• Homepage:
• Size: 421 KB
• Stars: 10
• Watchers: 2
• Forks: 2
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
Skip-GANomaly

=====

Implementation of Skip-GANomaly with MNIST dataset [Related repository].

## Architecture



    

  
Simplified Skip-GANomaly architecture.



## Graph in TensorBoard



    

  
Graph of Skip-GANomaly.



## Problem Definition



    

  
'Class-1' is defined as normal and the others are defined as abnormal.



## Results



    

  
Restoration result by Skip-GANomaly.





  

  
Box plot with encoding loss of test procedure.



## Environment

* Python 3.7.4  

* Tensorflow 1.14.0  

* Numpy 1.17.1  

* Matplotlib 3.1.1  

* Scikit Learn (sklearn) 0.21.3  

## Reference

[1] S Akcay, et al. (2018). Skip-ganomaly: Skip connected and adversarially trained encoder-decoder anomaly detection.. arXiv preprint arXiv:1901.08954.