A Graph Learning library for Humans

Overview

A Graph Learning library for Humans

These novel algorithms include but are not limited to:

  • A graph construction and graph searching class can be found here (NodeGraph). It was developed and invented as a faster alternative for hierarchical DAG construction and searching.
  • A fast DBSCAN method utilizing my connectivity code as invented during my PhD.
  • A NLP pattern matching algorithm useful for sequence alignment clustering.
  • High dimensional alignment code for aligning models to data.
  • An SVD based variant of the Distance Geometry algorithm. For going from relative to absolute coordinates.

License DOI Downloads

Visit the active code via : https://github.com/richardtjornhammar/graphtastic

Pip installation with :

pip install graphtastic

Version controlled installation of the Graphtastic library

The Graphtastic library

In order to run these code snippets we recommend that you download the nix package manager. Nix package manager links from Februari 2022:

https://nixos.org/download.html

$ curl -L https://nixos.org/nix/install | sh

If you cannot install it using your Wintendo then please consider installing Windows Subsystem for Linux first:

https://docs.microsoft.com/en-us/windows/wsl/install-win10

In order to run the code in this notebook you must enter a sensible working environment. Don't worry! We have created one for you. It's version controlled against python3.9 (and experimental python3.10 support) and you can get the file here:

https://github.com/richardtjornhammar/graphtastic/blob/master/env/env39.nix

Since you have installed Nix as well as WSL, or use a Linux (NixOS) or bsd like system, you should be able to execute the following command in a termnial:

$ nix-shell env39.nix

Now you should be able to start your jupyter notebook locally:

$ jupyter-notebook graphhaxxor.ipynb

and that's it.

EXAMPLE 0

Running

import graphtastic.graphs as gg
import graphtastic.clustering as gl
import graphtastic.fit as gf
import graphtastic.convert as gc

Should work if the install was succesful

Example 1 : Absolute and relative coordinates

In this example, we will use the SVD based distance geometry method to go between absolute coordinates, relative coordinate distances and back to ordered absolute coordinates. Absolute coordinates are float values describing the position of something in space. If you have several of these then the same information can be conveyed via the pairwise distance graph. Going from absolute coordinates to pairwise distances is simple and only requires you to calculate all the pairwise distances between your absolute coordinates. Going back to mutually orthogonal ordered coordinates from the pariwise distances is trickier, but a solved problem. The distance geometry can be obtained with SVD and it is implemented in the graphtastic.fit module under the name distance_matrix_to_absolute_coordinates. We start by defining coordinates afterwhich we can calculate the pair distance matrix and transforming it back by using the code below

import numpy as np

coordinates = np.array([[-23.7100 ,  24.1000 ,  85.4400],
  [-22.5600 ,  23.7600 ,  85.6500],
  [-21.5500 ,  24.6200 ,  85.3800],
  [-22.2600 ,  22.4200 ,  86.1900],
  [-23.2900 ,  21.5300 ,  86.4800],
  [-20.9300 ,  22.0300 ,  86.4300],
  [-20.7100 ,  20.7600 ,  86.9400],
  [-21.7900 ,  19.9300 ,  87.1900],
  [-23.0300 ,  20.3300 ,  86.9600],
  [-24.1300 ,  19.4200 ,  87.2500],
  [-23.7400 ,  18.0500 ,  87.0000],
  [-24.4900 ,  19.4600 ,  88.7500],
  [-23.3700 ,  19.8900 ,  89.5200],
  [-24.8500 ,  18.0000 ,  89.0900],
  [-23.9600 ,  17.4800 ,  90.0800],
  [-24.6600 ,  17.2400 ,  87.7500],
  [-24.0800 ,  15.8500 ,  88.0100],
  [-23.9600 ,  15.1600 ,  86.7600],
  [-23.3400 ,  13.7100 ,  87.1000],
  [-21.9600 ,  13.8700 ,  87.6300],
  [-24.1800 ,  13.0300 ,  88.1100],
  [-23.2900 ,  12.8200 ,  85.7600],
  [-23.1900 ,  11.2800 ,  86.2200],
  [-21.8100 ,  11.0000 ,  86.7000],
  [-24.1500 ,  11.0300 ,  87.3200],
  [-23.5300 ,  10.3200 ,  84.9800],
  [-23.5400 ,   8.9800 ,  85.4800],
  [-23.8600 ,   8.0100 ,  84.3400],
  [-23.9800 ,   6.5760 ,  84.8900],
  [-23.2800 ,   6.4460 ,  86.1300],
  [-23.3000 ,   5.7330 ,  83.7800],
  [-22.7300 ,   4.5360 ,  84.3100],
  [-22.2000 ,   6.7130 ,  83.3000],
  [-22.7900 ,   8.0170 ,  83.3800],
  [-21.8100 ,   6.4120 ,  81.9200],
  [-20.8500 ,   5.5220 ,  81.5200],
  [-20.8300 ,   5.5670 ,  80.1200],
  [-21.7700 ,   6.4720 ,  79.7400],
  [-22.3400 ,   6.9680 ,  80.8000],
  [-20.0100 ,   4.6970 ,  82.1500],
  [-19.1800 ,   3.9390 ,  81.4700] ]);

if __name__=='__main__':

    import graphtastic.fit as gf

    distance_matrix = gf.absolute_coordinates_to_distance_matrix( coordinates )
    ordered_coordinates = gf.distance_matrix_to_absolute_coordinates( distance_matrix , n_dimensions=3 )

    print ( ordered_coordinates )

You will notice that the largest variation is now aligned with the X axis, the second most variation aligned with the Y axis and the third most, aligned with the Z axis while the graph topology remained unchanged.

Example 2 : Deterministic DBSCAN

DBSCAN is a clustering algorithm that can be seen as a way of rejecting points, from any cluster, that are positioned in low dense regions of a point cloud. This introduces holes and may result in a larger segment, that would otherwise be connected via a non dense link to become disconnected and form two segments, or clusters. The rejection criterion is simple. The central concern is to evaluate a distance matrix with an applied cutoff this turns the distances into true or false values depending on if a pair distance between point i and j is within the distance cutoff. This new binary Neighbour matrix tells you wether or not two points are neighbours (including itself). The DBSCAN criterion states that a point is not part of any cluster if it has fewer than minPts neighbors. Once you've calculated the distance matrix you can immediately evaluate the number of neighbors each point has and the rejection criterion, via . If the rejection vector R value of a point is True then all the pairwise distances in the distance matrix of that point is set to a value larger than epsilon. This ensures that a distance matrix search will reject those points as neighbours of any other for the choosen epsilon. By tracing out all points that are neighbors and assessing the connectivity (search for connectivity) you can find all the clusters.

import numpy as np
from graphtastic.clustering import dbscan, reformat_dbscan_results
from graphtastic.fit import absolute_coordinates_to_distance_matrix

N   = 100
N05 = int ( np.floor(0.5*N) )
R   = 0.25*np.random.randn(N).reshape(N05,2) + 1.5
P   = 0.50*np.random.randn(N).reshape(N05,2)

coordinates = np.array([*P,*R])

results = dbscan ( distance_matrix = absolute_coordinates_to_distance_matrix(coordinates,bInvPow=True) , eps=0.45 , minPts=4 )
clusters = reformat_dbscan_results(results)
print ( clusters )

Example 3 : NodeGraph, distance matrix to DAG

Here we demonstrate how to convert the graph coordinates into a hierarchy. The leaf nodes will correspond to the coordinate positions.

import numpy as np

coordinates = np.array([[-23.7100 ,  24.1000 ,  85.4400],
  [-22.5600 ,  23.7600 ,  85.6500],
  [-21.5500 ,  24.6200 ,  85.3800],
  [-22.2600 ,  22.4200 ,  86.1900],
  [-23.2900 ,  21.5300 ,  86.4800],
  [-20.9300 ,  22.0300 ,  86.4300],
  [-20.7100 ,  20.7600 ,  86.9400],
  [-21.7900 ,  19.9300 ,  87.1900],
  [-23.0300 ,  20.3300 ,  86.9600],
  [-24.1300 ,  19.4200 ,  87.2500],
  [-23.7400 ,  18.0500 ,  87.0000],
  [-24.4900 ,  19.4600 ,  88.7500],
  [-23.3700 ,  19.8900 ,  89.5200],
  [-24.8500 ,  18.0000 ,  89.0900],
  [-23.9600 ,  17.4800 ,  90.0800],
  [-24.6600 ,  17.2400 ,  87.7500],
  [-24.0800 ,  15.8500 ,  88.0100],
  [-23.9600 ,  15.1600 ,  86.7600],
  [-23.3400 ,  13.7100 ,  87.1000],
  [-21.9600 ,  13.8700 ,  87.6300],
  [-24.1800 ,  13.0300 ,  88.1100],
  [-23.2900 ,  12.8200 ,  85.7600],
  [-23.1900 ,  11.2800 ,  86.2200],
  [-21.8100 ,  11.0000 ,  86.7000],
  [-24.1500 ,  11.0300 ,  87.3200],
  [-23.5300 ,  10.3200 ,  84.9800],
  [-23.5400 ,   8.9800 ,  85.4800],
  [-23.8600 ,   8.0100 ,  84.3400],
  [-23.9800 ,   6.5760 ,  84.8900],
  [-23.2800 ,   6.4460 ,  86.1300],
  [-23.3000 ,   5.7330 ,  83.7800],
  [-22.7300 ,   4.5360 ,  84.3100],
  [-22.2000 ,   6.7130 ,  83.3000],
  [-22.7900 ,   8.0170 ,  83.3800],
  [-21.8100 ,   6.4120 ,  81.9200],
  [-20.8500 ,   5.5220 ,  81.5200],
  [-20.8300 ,   5.5670 ,  80.1200],
  [-21.7700 ,   6.4720 ,  79.7400],
  [-22.3400 ,   6.9680 ,  80.8000],
  [-20.0100 ,   4.6970 ,  82.1500],
  [-19.1800 ,   3.9390 ,  81.4700] ]);


if __name__=='__main__':

    import graphtastic.graphs as gg
    import graphtastic.fit as gf
    GN = gg.NodeGraph()
    #
    # bInvPow refers to the distance type. If True then R distances are returned
    # instead of R2 (R**2) distances. That is also computing the square root if True
    #
    distm = gf.absolute_coordinates_to_distance_matrix( coordinates , bInvPow=True )
    #
    # Now a Graph DAG is constructed from the pairwise distances
    GN.distance_matrix_to_graph_dag( distm )
    #
    # And write it to a json file so that we may employ JS visualisations
    # such as D3 or other nice packages to view our hierarchy
    GN.write_json( jsonfile='./graph_hierarchy.json' )

Manually updated code backups for this library :

GitLab | https://gitlab.com/richardtjornhammar/graphtastic

CSDN | https://codechina.csdn.net/m0_52121311/graphtastic

You might also like...
Fastest Gephi's ForceAtlas2 graph layout algorithm implemented for Python and NetworkX
Fastest Gephi's ForceAtlas2 graph layout algorithm implemented for Python and NetworkX

ForceAtlas2 for Python A port of Gephi's Force Atlas 2 layout algorithm to Python 2 and Python 3 (with a wrapper for NetworkX and igraph). This is the

🐍PyNode Next allows you to easily create beautiful graph visualisations and animations
🐍PyNode Next allows you to easily create beautiful graph visualisations and animations

PyNode Next A complete rewrite of PyNode for the modern era. Up to five times faster than the original PyNode. PyNode Next allows you to easily create

LabGraph is a a Python-first framework used to build sophisticated research systems with real-time streaming, graph API, and parallelism.
LabGraph is a a Python-first framework used to build sophisticated research systems with real-time streaming, graph API, and parallelism.

LabGraph is a a Python-first framework used to build sophisticated research systems with real-time streaming, graph API, and parallelism.

Automatization of BoxPlot graph usin Python MatPlotLib and Excel

BoxPlotGraphAutomation Automatization of BoxPlot graph usin Python / Excel. This file is an automation of BoxPlot-Graph using python graph library mat

Library for exploring and validating machine learning data

TensorFlow Data Validation TensorFlow Data Validation (TFDV) is a library for exploring and validating machine learning data. It is designed to be hig

Library for exploring and validating machine learning data

TensorFlow Data Validation TensorFlow Data Validation (TFDV) is a library for exploring and validating machine learning data. It is designed to be hig

Declarative statistical visualization library for Python
Declarative statistical visualization library for Python

Altair http://altair-viz.github.io Altair is a declarative statistical visualization library for Python. With Altair, you can spend more time understa

Plotting library for IPython/Jupyter notebooks
Plotting library for IPython/Jupyter notebooks

bqplot 2-D plotting library for Project Jupyter Introduction bqplot is a 2-D visualization system for Jupyter, based on the constructs of the Grammar

Cartopy - a cartographic python library with matplotlib support
Cartopy - a cartographic python library with matplotlib support

Cartopy is a Python package designed to make drawing maps for data analysis and visualisation easy. Table of contents Overview Get in touch License an

Releases(v0.12.0)
Owner
Richard Tjörnhammar
PhD in Biological physics https://richardtjornhammar.github.io
Richard Tjörnhammar
HW 2: Visualizing interesting datasets

HW 2: Visualizing interesting datasets Check out the project instructions here! Mean Earnings per Hour for Males and Females My first graph uses data

7 Oct 27, 2021
Set of matplotlib operations that are not trivial

Matplotlib Snippets This repository contains a set of matplotlib operations that are not trivial. Histograms Histogram with bins adapted to log scale

Raphael Meudec 1 Nov 15, 2021
Learn Data Science with focus on adding value with the most efficient tech stack.

DataScienceWithPython Get started with Data Science with Python An engaging journey to become a Data Scientist with Python TL;DR Download all Jupyter

Learn Python with Rune 110 Dec 22, 2022
Rockstar - Makes you a Rockstar C++ Programmer in 2 minutes

Rockstar Rockstar is one amazing library, which will make you a Rockstar Programmer in just 2 minutes. In last decade, people learned C++ in 21 days.

4k Jan 05, 2023
Type-safe YAML parser and validator.

StrictYAML StrictYAML is a type-safe YAML parser that parses and validates a restricted subset of the YAML specification. Priorities: Beautiful API Re

Colm O'Connor 1.2k Jan 04, 2023
Smoking Simulation is an app to simulate the spreading of smokers and non-smokers, their interactions and population during certain amount of time.

Smoking Simulation is an app to simulate the spreading of smokers and non-smokers, their interactions and population during certain

Bohdan Ruban 5 Nov 08, 2022
A Python toolbox for gaining geometric insights into high-dimensional data

"To deal with hyper-planes in a 14 dimensional space, visualize a 3D space and say 'fourteen' very loudly. Everyone does it." - Geoff Hinton Overview

Contextual Dynamics Laboratory 1.8k Dec 29, 2022
NW 2022 Hackathon Project by Angelique Clara Hanzel, Aryan Sonik, Damien Fung, Ramit Brata Biswas

Spiral-Data-Visualizer NW 2022 Hackathon Project by Angelique Clara Hanzell, Aryan Sonik, Damien Fung, Ramit Brata Biswas Description This project vis

Damien Fung 2 Jan 16, 2022
This is my favourite function - the Rastrigin function.

This is my favourite function - the Rastrigin function. What sparked my curiosity and interest in the function was its complexity in terms of many local optimum points, which makes it particularly in

1 Dec 27, 2021
High performance, editable, stylable datagrids in jupyter and jupyterlab

An ipywidgets wrapper of regular-table for Jupyter. Examples Two Billion Rows Notebook Click Events Notebook Edit Events Notebook Styling Notebook Pan

J.P. Morgan Chase 75 Dec 15, 2022
Visualization of the World Religion Data dataset by Correlates of War Project.

World Religion Data Visualization Visualization of the World Religion Data dataset by Correlates of War Project. Mostly personal project to famirializ

Emile Bangma 1 Oct 15, 2022
Active Transport Analytics Model (ATAM) is a new strategic transport modelling and data visualization framework for Active Transport as well as emerging micro-mobility modes

{ATAM} Active Transport Analytics Model Active Transport Analytics Model (“ATAM”) is a new strategic transport modelling and data visualization framew

Peter Stephan 0 Jan 12, 2022
Generate visualizations of GitHub user and repository statistics using GitHub Actions.

GitHub Stats Visualization Generate visualizations of GitHub user and repository statistics using GitHub Actions. This project is currently a work-in-

JoelImgu 3 Dec 14, 2022
ScisorWiz: Differential Isoform Visualizer for Long-Read RNA Sequencing Data

ScisorWiz: Vizualizer for Differential Isoform Expression README ScisorWiz is a linux-based R-package for visualizing differential isoform expression

Alexander Stein 6 Oct 04, 2022
OpenStats is a library built on top of streamlit that extracts data from the Github API and shows the main KPIs

Open Stats Discover and share the KPIs of your OpenSource project. OpenStats is a library built on top of streamlit that extracts data from the Github

Pere Miquel Brull 4 Apr 03, 2022
A guide for using Bootstrap 5 classes in Dash Bootstrap Components V1

dash-bootstrap-cheatsheet This handy interactive cheatsheet makes it easy to use the Bootstrap 5 classes with your Dash app made with the latest versi

10 Dec 22, 2022
The official colors of the FAU as matplotlib/seaborn colormaps

FAU - Colors The official colors of Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU) as matplotlib / seaborn colormaps. We support the old colo

Machine Learning and Data Analytics Lab FAU 9 Sep 05, 2022
A python visualization of the A* path finding algorithm

A python visualization of the A* path finding algorithm. It allows you to pick your start, end location and make obstacles and then view the process of finding the shortest path. You can also choose

Kimeon 4 Aug 02, 2022
A deceptively simple plotting library for Streamlit

🍅 Plost A deceptively simple plotting library for Streamlit. Because you've been writing plots wrong all this time. Getting started pip install plost

Thiago Teixeira 192 Dec 29, 2022
Draw datasets from within Jupyter.

drawdata This small python app allows you to draw a dataset in a jupyter notebook. This should be very useful when teaching machine learning algorithm

vincent d warmerdam 505 Nov 27, 2022