From 32434854439203ba78a7b7ae7d30daa9fa7d491e Mon Sep 17 00:00:00 2001 From: beckyperriment <93582518+beckyperriment@users.noreply.github.com> Date: Thu, 7 Dec 2023 11:16:18 +0000 Subject: [PATCH] Update paper.md --- joss/paper.md | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/joss/paper.md b/joss/paper.md index 75145e1..efe04fa 100644 --- a/joss/paper.md +++ b/joss/paper.md @@ -127,6 +127,8 @@ Finding global optimality can increase the computation time, depending on the nu We compared our approach with two other DTW clustering packages, \texttt{DTAIDistance} [@Meert2020Dtaidistance] and \texttt{TSlearn} [@Tavenard2020TslearnData]. The datasets used for the comparison are from the UCR Time Series Classification Archive [@Dau2018TheArchive], and consist of 128 time series datasets with up to 16,800 data series of lengths up to 2,844. The full results can be found in the Appendix. Benchmarking against \texttt{TSlearn} was stopped after the first 22 datasets because the results were consistently over 20 times slower than \texttt{DTW-C++}. Table \ref{tab:small_table} shows the results for datasets downselected to have a number of time series ($N$) greater than 100 and a length of each time series greater than 500 points. This is because \texttt{DTW-C++} is aimed at larger datasets where the speed improvements are more relevant. + +Table: Computational time comparison of \texttt{DTW-C++} using MIP and k-medoids, vs.\ \texttt{DTAIDistance}, and \texttt{TSlearn}, on datasets in the UCR Time Series Classification Archive where $N>100$ and $L>500$. \label{tab} | | Number of time series | Length of time series | DTW-C++ MIP (s) | DTW-C++ k-Medoids (s) | DTAI Distance (s) | Time decrease (%) | |----------------------------|-----------------------|-----------------------|-----------------|-----------------------|-------------------|--------------------| @@ -161,7 +163,8 @@ We compared our approach with two other DTW clustering packages, \texttt{DTAIDis | SmallKitchenAppliances | 375 | 720 | 41.7 | **23.8** | 30.1 | 21 | | StarLightCurves | 8236 | 1024 | N/A | **18551.7** | 27558.1 | 33 | | UWaveGestureLibraryAll | 3582 | 945 | N/A | **1194.6** | 4436.9 | 73 | -[Computational time comparison of \texttt{DTW-C++} using MIP and k-medoids, vs.\ \texttt{DTAIDistance}, and \texttt{TSlearn}, on datasets in the UCR Time Series Classification Archive where $N>100$ and $L>500$. \label{tab}] + + # Acknowledgements