Higra tutorial

Higra tutorial 


Higra is a hierarchical graph analysis library.

Preliminaries


The tutorial can be done online with Google Colab (you just need a Google account, no paid plan needed) on your machine with a local installation of Python 3.8, 3.9, 3.10, 3.11, 3.12 (amd64).
In this second case, you can make sure you have a working environment by installing the packages specified in the attached requirements.txt file. This can be done with the following command (in a shell with the python environment variables properly initialized):

$> pip install -r requirements.txt

If you are using an Anaconda distribution for Python, you can create a properly configured conda environment with the attached environment.yml file:

$> conda env create -f environment.yml

Then you activate the environment

$> conda activate tutohigra

News: higra can now be installed with conda. Notably, macOS ARM64 is currently only supported through conda conda install higra -c conda-forge. If your are on ARM64, you should install higra using conda, and not with pip as currently done in the environment.yml or in the requirements.txt.
 
Whether you are using pip or conda, you can check that the installation was successful by running Jupyter notebook:

$> jupyter notebook

Tutorial

 

The notebooks used in the tutorial are available by clicking on the name (Intro, Connected filters or Hierarchical segmentation), while the data is available in thisattached file (but the data file should not be needed, it is contained in the notebooks)

Colab users can use the following links:


Solutions

 
The solutions of the last 2 notebooks are available in the attached file (Hierarchical segmentation and Connected filters) or in the following notebooks on Colab: