- Clone this repo:
git clone https://github.com/kimwin2/RRN_AnomalyDetection.git
- Download datase:
python3 0_download_dataset.py
- For training:
python3 1_train.py
- For testing:
python3 2_anomaly.py
- Display result:
python3 3_display_result.py
- preprocess_data.py : This code is for preprocessing dataset.
- model/model.py: This code is for the lstm based encoder-decoder architecture.
Y.-H. Yoo, U.-H. Kim and J.-H. Kim, "Recurrent Reconstructive Network for Sequential Anomaly Detection," IEEE Transactions on Cybernetics, Accepted, DOI: 10.1109/TCYB.2019.2933548, Jul. 2019.