TIRE is an autoencoder-based change point detection algorithm for time series data that uses a TIme-Invariant Representation (TIRE). More information can be found in the 2020 preprint Change Point Detection in Time Series Data using Autoencoders with a Time-Invariant Representation.
from TIRE import DenseTIRE as TIRE
import torch
import numpy as np
seq_length = 4500
dim = 1
ts = np.random.randn(seq_length, dim)
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
model = TIRE(dim).to(device)
model.fit(ts, epoches=1)
dissimilarities, change_point_scores = model.predict(ts)
# plt.plot(dissimilarities)
pip install git+https://github.com/HolyBayes/TIRE_pytorch
@misc{ryck2020change,
title={Change Point Detection in Time Series Data using Autoencoders with a Time-Invariant Representation},
author={Tim De Ryck and Maarten De Vos and Alexander Bertrand},
year={2020},
eprint={2008.09524},
archivePrefix={arXiv},
primaryClass={cs.LG}}
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/