The Atlas Trends API is an implementation of a novel method to cluster RTT time series using nonparametric Bayesian models. The API allows producing humanlike segmentation of RIPE Atlas RTT time series.
This repository contains the following Python notebooks demonstrating the API usage:
| Name | Description | Online Notebook |
|---|---|---|
| Atlas Trends API | Overview of the API |  |
M. Mouchet, S. Vaton, T. Chonavel, E. Aben and J. D. Hertog, "Large-Scale Characterization and Segmentation of Internet Path Delays With Infinite HMMs," in IEEE Access, vol. 8, pp. 16771-16784, 2020.
@article{mouchet2019large,
author={M. {Mouchet} and S. {Vaton} and T. {Chonavel} and E. {Aben} and J. {Den Hertog}},
journal={IEEE Access},
title={Large-Scale Characterization and Segmentation of Internet Path Delays With Infinite HMMs},
year={2020},
volume={8},
pages={16771-16784},
doi={10.1109/ACCESS.2020.2968380},
ISSN={2169-3536}
}You can run the notebooks on Google Colab by following the links at the top, or locally by running the following in a terminal:
git clone https://github.com/maxmouchet/atlas-trends-demo.git
cd atlas-trends-demo
python3 -m venv trends-env; source trends-env/bin/activate
pip install -r requirements.txt
jupyter lab