Pytorch implementation of RED-SDS (NeurIPS 2021).
This repository contains a reference implementation of RED-SDS, a non-linear state space model proposed in the NeurIPS 2021 paper Deep Explicit Duration Switching Models for Time Series.
pip install -r requirements.txt
.python reevaluate.py --ckpt <model-path>.pt
../data/
. Alternatively, you can download the datasets from this link../data/bee.sh
.python run_segmentation.py --config configs/<config>.yaml --device cuda:0
to train the RED-SDS model.tensorboard --logdir /path/to/results/dir
to visualize results.python run_gts_univariate.py --config configs/<config>.yaml --device cuda:0
to train the RED-SDS model. The dataset will be downloaded automatically.tensorboard --logdir /path/to/results/dir
to visualize results.For any questions regarding the code or the paper, please email Fatir, Konstantinos, or Richard.
If you find this repository or the ideas presented in our paper useful for your research, please consider citing our paper.
@inproceedings{ansari2021deep,
author = {Abdul Fatir Ansari and Konstantinos Benidis and Richard Kurle and Ali Caner Turkmen and Harold Soh and Alex Smola and Bernie Wang and Tim Januschowski},
title = {Deep Explicit Duration Switching Models for Time Series},
year = {2021},
booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
}
This repo contains parts of code based on the following repos:
Repo | Copyright (c) | License |
---|---|---|
google-research/google-research/snlds | The Google Research Authors | Apache 2.0 |
mattjj/pyslds | Matthew James Johnson | MIT |