There is separate detailed documentation available for this project; this readme.md file only gives a short summary.
DTW-C++ is a C++ library for dynamic time warping (DTW) and clustering of time series data. Users can input multiple time series and find clusters of similar time series. The time series can have the same or different lengths. The number of clusters to find can be fixed or specified as a range to try. DTW-C++ finds clusters in time series data using k-medoids or mixed integer programming (MIP). K-medoids is generally faster but may get stuck in local optima, while MIP can give guarantees about globally optimal clusters.
APA style:
Kumtepeli, V., Perriment, R., & Howey, D. A. (2024). DTW-C++: Fast dynamic time warping and clustering of time series data. Journal of Open Source Software, 9(101), 6881. https://doi.org/10.21105/joss.06881
BibTeX:
@article{Kumtepeli2024,
author = {Kumtepeli, Volkan and Perriment, Rebecca and Howey, David A.},
doi = {10.21105/joss.06881},
journal = {Journal of Open Source Software},
month = sep,
number = {101},
pages = {6881},
title = {{DTW-C++: Fast dynamic time warping and clustering of time series data}},
url = {https://joss.theoj.org/papers/10.21105/joss.06881},
volume = {9},
year = {2024}
}
 Becky Perriment 💡💻👀 |
 Volkan Kumtepeli 💡💻👀 |
 David Howey 💡👀 |