The original paper of this repository is Ruff, Lukas, et al. "Deep one-class classification." International conference on machine learning. PMLR, 2018.

https://proceedings.mlr.press/v80/ruff18a.html https://github.com/lukasruff/Deep-SVDD-PyTorch

Environment

torch : 2.2.2+cu118 torchaudio : 2.2.2+cu118 torchvision : 0.17.2+cu118 numpy : 1.24.1 opencv-python : 4.9.0.80 opencv-contrib-python : 4.9.0.80 matplotlib : 3.8.4 torchsummary : 1.4

Train and test

To train SVDD, type next prompt. python train_SVDD.py --training_mode 1 To test SVDD, type next prompt. python train_SVDD.py --training_mode 0

ordering

This file make paths clear.

dataloader

This file plays the role of loading data.

optimization

This folder contains two files, each tasked with training an autoencoder and SVDD, respectively.

model

This folder comprises two files, each designated for modeling an autoencoder and SVDD, respectively.