Python ts2vg package provides high-performance algorithm implementations to build visibility graphs from time series data.

Last update: Dec 17, 2022

Overview

ts2vg: Time series to visibility graphs

The Python ts2vg package provides high-performance algorithm implementations to build visibility graphs from time series data.

The visibility graphs and some of their properties (e.g. degree distributions) are computed quickly and efficiently, even for time series with millions of observations thanks to the use of NumPy and a custom C backend (via Cython) developed for the visibility algorithms.

The visibility graphs are provided according to the mathematical definitions described in:

Lucas Lacasa et al., "From time series to complex networks: The visibility graph", 2008.
Lucas Lacasa et al., "Horizontal visibility graphs: exact results for random time series", 2009.

An efficient divide-and-conquer algorithm is used to compute the graphs, as described in:

Xin Lan et al., "Fast transformation from time series to visibility graphs", 2015.

Installation

The latest released ts2vg version is available at the Python Package Index (PyPI) and can be easily installed by running:

pip install ts2vg

For other advanced uses, to build ts2vg from source Cython is required.

Basic usage

Visibility graph

Building visibility graphs from time series is very simple:

from ts2vg import NaturalVG

ts = [1.0, 0.5, 0.3, 0.7, 1.0, 0.5, 0.3, 0.8]

g = NaturalVG()
g.build(ts)

edges = g.edges

The time series passed can be a list, a tuple, or a numpy 1D array.

Horizontal visibility graph

We can also obtain horizontal visibility graphs in a very similar way:

from ts2vg import HorizontalVG

ts = [1.0, 0.5, 0.3, 0.7, 1.0, 0.5, 0.3, 0.8]

g = HorizontalVG()
g.build(ts)

edges = g.edges

Degree distribution

If we are only interested in the degree distribution of the visibility graph we can pass only_degrees=True to the build method. This will be more efficient in time and memory than computing the whole graph.

g = NaturalVG()
g.build(ts, only_degrees=True)

ks, ps = g.degree_distribution

Directed visibility graph

g = NaturalVG(directed='left_to_right')
g.build(ts)

Weighted visibility graph

g = NaturalVG(weighted='distance')
g.build(ts)

For more information and options see: Examples and API Reference.

Interoperability with other libraries

The graphs obtained can be easily converted to graph objects from other common Python graph libraries such as igraph, NetworkX and SNAP for further analysis.

The following methods are provided:

as_igraph()
as_networkx()
as_snap()

For example:

g = NaturalVG()
g.build(ts)

nx_g = g.as_networkx()

Command line interface

ts2vg can also be used as a command line program directly from the console:

ts2vg ./timeseries.txt -o out.edg

For more help and a list of options run:

ts2vg --help

Contributing

ts2vg can be found on GitHub. Pull requests and issue reports are welcome.

License

ts2vg is licensed under the terms of the MIT License.

You might also like...

The Spectral Diagram (SD) is a new tool for the comparison of time series in the frequency domain

The Spectral Diagram (SD) is a new tool for the comparison of time series in the frequency domain. The SD provides a novel way to display the coherence function, power, amplitude, phase, and skill score of discrete frequencies of two time series. Each SD summarises these quantities in a single plot for multiple targeted frequencies.

3 Oct 10, 2022

The windML framework provides an easy-to-use access to wind data sources within the Python world, building upon numpy, scipy, sklearn, and matplotlib. Renewable Wind Energy, Forecasting, Prediction

windml Build status : The importance of wind in smart grids with a large number of renewable energy resources is increasing. With the growing infrastr

125 Dec 24, 2022

Comments

help getting started

I am playing around with ts2vg and I am having a hard time with the plotting using igraph. I try to compute the natural vg for a short time series, but when trying to plot it I get this error:

Traceback (most recent call last):
  File "\anaconda3\envs\DK_01\lib\site-packages\IPython\core\interactiveshell.py", line 3398, in run_code
    exec(code_obj, self.user_global_ns, self.user_ns)
  File "<ipython-input-1-9a1fdcf342e8>", line 1, in <cell line: 1>
    ig.plot(nx_g, target='graph.pdf')
  File "\anaconda3\envs\DK_01\lib\site-packages\igraph\drawing\__init__.py", line 512, in plot
    result.save()
  File "\anaconda3\envs\DK_01\lib\site-packages\igraph\drawing\__init__.py", line 309, in save
    self._ctx.show_page()
igraph.drawing.cairo.MemoryError: out of memory

The file created is corrupted.

Here is my code:

import numpy as np
from ts2vg import NaturalVG
import igraph as ig

import matplotlib.pyplot as plt

# time domain
t = np.linspace(1, 40)
dt = np.diff(t)

# build series
x1 = np.sin(2*np.pi/10*t)
x2 = np.sin(2*np.pi/15*t)

y = x1 + x2

plt.plot(t, y, '.-')
plt.show()

# build HVG
g = NaturalVG()
g.build(y)

nx_g = g.as_igraph()

# plotting
ig.plot(nx_g, target='graph.pdf')

I am using ts2vg 1.0.0, igraph 0.9.11, and pycairo 1.21.0

opened by ACatAC 1

Python ts2vg package provides high-performance algorithm implementations to build visibility graphs from time series data.

Related tags

Overview

ts2vg: Time series to visibility graphs

Installation

Basic usage

Visibility graph

Horizontal visibility graph

Degree distribution

Directed visibility graph

Weighted visibility graph

Interoperability with other libraries

Command line interface

Contributing

License

You might also like...

The Spectral Diagram (SD) is a new tool for the comparison of time series in the frequency domain

The windML framework provides an easy-to-use access to wind data sources within the Python world, building upon numpy, scipy, sklearn, and matplotlib. Renewable Wind Energy, Forecasting, Prediction

Kglab - an abstraction layer in Python for building knowledge graphs

Extensible, parallel implementations of t-SNE

Extensible, parallel implementations of t-SNE

Graphical display tools, to help students debug their class implementations in the Carcassonne family of projects

Draw interactive NetworkX graphs with Altair

Draw interactive NetworkX graphs with Altair

Generate graphs with NetworkX, natively visualize with D3.js and pywebview

Comments

help getting started

Releases(v1.0.0)

v1.0.0(Aug 19, 2021)

Owner

Carlos Bergillos

Jupyter Notebook extension leveraging pandas DataFrames by integrating DataTables and ChartJS.

Realtime Viewer Mandelbrot set with Python and Taichi (cpu, opengl, cuda, vulkan, metal)

Small U-Net for vehicle detection

PyFlow is a general purpose visual scripting framework for python

Pebble is a stat's visualization tool, this will provide a skeleton to develop a monitoring tool.

Massively parallel self-organizing maps: accelerate training on multicore CPUs, GPUs, and clusters

erdantic is a simple tool for drawing entity relationship diagrams (ERDs) for Python data model classes

Python library that makes it easy for data scientists to create charts.

GitHub Stats Visualizations : Transparent

Graphing communities on Twitch.tv in a visually intuitive way

Type-safe YAML parser and validator.

Runtime analysis of code with plotting

Machine learning beginner to Kaggle competitor in 30 days. Non-coders welcome. The program starts Monday, August 2, and lasts four weeks. It's designed for people who want to learn machine learning.

Apache Superset is a Data Visualization and Data Exploration Platform

High performance, editable, stylable datagrids in jupyter and jupyterlab

A high performance implementation of HDBSCAN clustering. http://hdbscan.readthedocs.io/en/latest/

An automatic prover for tautologies in Metamath

Simple spectra visualization tool for astronomers

An application that allows you to design and test your own stock trading algorithms in an attempt to beat the market.

Advanced_Data_Visualization_Tools - The present hands-on lab mainly uses Immigration to Canada dataset and employs advanced visualization tools such as word cloud, and waffle plot to display relations between features within the dataset.