https://github.com/jmrozanec/ts2g2

Generate graphs from time series and time series from graphs.
https://github.com/jmrozanec/ts2g2

graph-algorithms graphs machine-learning time-series

Last synced: 4 months ago
JSON representation

Generate graphs from time series and time series from graphs.

• Host: GitHub
• URL: https://github.com/jmrozanec/ts2g2
• Owner: jmrozanec
• License: apache-2.0
• Created: 2023-05-19T10:53:22.000Z (almost 3 years ago)
• Default Branch: main
• Last Pushed: 2024-07-30T07:02:43.000Z (over 1 year ago)
• Last Synced: 2025-07-27T22:07:07.244Z (8 months ago)
• Topics: graph-algorithms, graphs, machine-learning, time-series
• Language: Jupyter Notebook
• Homepage: https://timeseriestographs.com
• Size: 11.5 MB
• Stars: 1
• Watchers: 3
• Forks: 7
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# ts2g2

TS2G2 stands for "timeseries to graphs and back". The library implements a variety of strategies to convert timeseries into graphs, and convert graphs into sequences.

    stream = TimeseriesArrayStream([2, 1, 3, 2, 1, 3, 2, 1, 3, 2, 1, 3])

    timeseries = Timeseries(stream)

    g = timeseries.to_graph(NaturalVisibilityGraphStrategy())

    sequence = g.to_sequence(RandomWalkSequenceGenerationStrategy(), sequence_length=500)

For a more detailed example, look at the [Amazon stocks demo](https://github.com/graph-massivizer/ts2g2/blob/main/tutorials/demo-amazon-stocks.ipynb).

Many of the methods implemented in this library are described in _Silva, Vanessa Freitas, et al. "Time series analysis via network science: Concepts and algorithms." Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery 11.3 (2021): e1404._ Nevertheless, the library also includes additional techniques found in other works from the scientific literature.

This package is being developed as part of the [Graph-Massivizer](https://graph-massivizer.eu/) project. 

The package is a joint effort between the [Jožef Stefan Institute](https://www.ijs.si/), the [University of Twente](https://www.utwente.nl/en/), the [Vrije Universiteit Amsterdam](https://vu.nl/en), the [University of Klagenfurt](https://www.aau.at/en/), the [University of Bologna](https://www.unibo.it/en), and [Peracton](https://peracton.com/).

### Timeseries to graph conversion

#### Implemented features

  

    #

    Visibility Graph

    Graph type

    Constraints

  

  

    Undirected

    Directed

    Weighted

  

  

    Penetration

    Angle

  

  

    1

    Natural Visibility Graph

    

        X

    

    

      

      X

    

    

      

      X

    

    

      

      X

    

    

      

      X

    

  

  

    2

    Horizontal Visibility Graph

    

      X

    

    

        

        X

    

    

      

      X 

    

    

      

      X

    

    

      

      X

    

  

  

    3

    Difference Visibility Graph

    

        

    

    

      

    

    

      

    

    

      

    

    

      

    

  

#### References table

  

    #

    Visibility Graph

    Graph type

    Constraints

  

  

    Undirected

    Directed

    Weighted

  

  

    Penetration

    Angle

  

  

    1

    Natural Visibility Graph

    

        ref

    

    

      

      ref

    

    

      

    

    

      

      ref

    

    

      

      ref, 

      ref

    

  

  

    2

    Horizontal Visibility Graph

    

      ref

    

    

      ref

      

    

    

      

      ref 

    

    

      

      ref, 

      ref

    

    

      

    

  

  

    3

    Difference Visibility Graph

    

        

    

    

      

    

    

      

    

    

      

    

    

      

    

  

### Graphs to timeseries conversion

Graphs are converted back to timeseries by sampling node values from the graph following different strategies. The following strategies have been implemented so far:

 - random node

 - random node neighbour

 - random node degree 

 - random walk

 - random walk with restart

 - random walk with jump

## Publications

When using this work for research purposes, we would appreciate it if the following references could be included:

Below we provide a curated list of papers related to our research in this area: