1000 Empirical Time series

A diverse selection of 1000 empirical time series, along with results of an <i>hctsa</i> feature extraction, using v1.03 of <i>hctsa</i> and Matlab 2019b, computed on a linux server at Sydney University.<br>The results of the computation are in the <i>hctsa</i> file, <b>HCTSA_Empirical1000.mat</b> for use in Matlab using v1.03 of <i>hctsa</i>.<br><b><br></b>The same data is available in <b>.csv</b> format (e.g., for use with non-Matlab computing environments) for the <b>hctsa_datamatrix.csv</b> (results of feature computation), with information about rows (time series) in <b>hctsa_timeseries-info.csv</b>, information about columns (features) in <b>hctsa_features.csv</b> and the data of individual time series (each line a time series, for time series described in <b>hctsa_timeseries-info.csv</b>) is in <b>hctsa_timeseries-data.csv</b>. Note that these files were produced by running <b>&gt;&gt;OutputToCSV(HCTSA_Empirical1000.mat,true);</b> in <i>hctsa</i>.<b><br></b>The input file, <b>INP_Empirical1000.mat</b>,<b> </b>is for use with <i>hctsa</i>, and contains the time-series data and metadata for the 1000 time series. For example, massive feature extraction from these data on the user's machine, using <i>hctsa</i>, can proceed as<b>&gt;&gt; TS_Init('INP_Empirical1000.mat');</b><br>Some visualizations of the dataset are in <b>CarpetPlot.png </b>(first 1000 samples of all time series as a carpet (color) plot) and <b>150TS-250samples.png</b> (conventional time-series plots of the first 250 samples of a sample of 150 time series from the dataset). More visualizations can be performed by the user using <b>TS_PlotTimeseries</b> from the <i>hctsa</i> package.<br><br>See links in references for more comprehensive documentation for performing methodological comparison using this dataset, and on how to download and use v1.03 of <i>hctsa</i>.