Skip to content

This is a tutorial for anomaly detection with Autoencoder

Notifications You must be signed in to change notification settings

Lekhika97/AnomalydetectionAutoencoder

 
 

Repository files navigation

AnomalydetectionAutoencoder

"This is a tutorial for anomaly detection with Autoencoder. The tutorial is based on 'Anomalies, Representations, and Self-Supervision' (arXiv:2301.04660). The aim here is to reproduce the upper left panel of Figure 1 from arXiv:2301.04660. Please first ensure that the necessary modules are installed. The notebook is created for a dataset provided as based on JHEP 05 (2019) 036, arXiv:1811.10276 (see also arXiv:2107.02157). The original training data can be found here: https://zenodo.org/record/5046428. The A > 4l new physics samples are available at https://zenodo.org/record/5046428. Please utilize the 20k training samples for a less computationally intensive environment."

For creating the representation utlize AnoCLR.ipynb. The augmentation and loss funtions can be found from the modules shared in [https://github.com/bmdillon/AnomalyCLR] .

About

This is a tutorial for anomaly detection with Autoencoder

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Jupyter Notebook 100.0%