/klcpd

PyTorch package for KL-CPD algorithm for change point and anomaly detection in time series

Primary LanguagePython

Unofficial PyTorch implementation of KL-CPD

License

KL-CPD is an algorithm for change point and anomaly detection in time series.

More information can be found in the 2019 paper Kernel Change-point Detection with Auxiliary Deep Generative Models.

Usage

from klcpd import KL_CPD

dim, seq_length = 1, 100
ts = np.random.randn(seq_length,dim)
device = torch.device('cuda')
model = KL_CPD(dim).to(device)
model.fit(ts)
preds = model.predict(ts)
print(preds)

Installation

pip install git+https://github.com/HolyBayes/klcpd

Authors

@article{chang2019kernel,
  title={Kernel change-point detection with auxiliary deep generative models},
  author={Chang, Wei-Cheng and Li, Chun-Liang and Yang, Yiming and P{\'o}czos, Barnab{\'a}s},
  journal={arXiv preprint arXiv:1901.06077},
  year={2019}
}

Contacts

Artem Ryzhikov, LAMBDA laboratory, Higher School of Economics, Yandex School of Data Analysis

E-mail: artemryzhikoff@yandex.ru

Linkedin: https://www.linkedin.com/in/artem-ryzhikov-2b6308103/

Link: https://www.hse.ru/org/persons/190912317