Data for the preprint of "Layer-by-layer unsupervised clustering of statistically relevant fluctuations in noisy time-series data of complex dynamical systems"

README: description of the files. 

This Zenodo repository contains all the data and original code necessary to reproduce the results of the paper https://doi.org/10.48550/arXiv.2402.07786. The code (continuously mantained and updated) is available open-source as a Python package at https://pypi.org/project/onion-clustering/ and on GitHub (https://github.com/matteobecchi/timeseries_analysis). 

The repository contains the folders "Fig1", "Fig2" etc, which contain the corresponding Datasets, together with the code to reproduce the figures. Additionally, the folder "FigS1 contains code and data for FigS1. 

The repository also contains the Supplementary Movies S1 to S4, in .mp4 format.