MI-Graph

MI-graph (Machine learning graph) is a pip module supporting various configurations for knowledge extraction.

Usage

To download the module run the following command:

pip install --extra-index-url https://repos.knox.cs.aau.dk knox-mi-graph

List of commandline arguments

MI-Graph supports the following arguments.

usage: mi-graph [-h] [-v] [--file] [--visualisation] {file,flask}

positional arguments:
  {file,flask}     Choose if you want to process a file, or run the program as a rest api

optional arguments:
  -h, --help       show this help message and exit
  -v, --version    show program's version number and exit
  --file , -f      Please indicate the json file you want to process.
  --visualisation  This option visualizes the graph with plotly, after the script has run

Run under development

There is two alternatives to run the program when developing:

Option 1

python -c "from mi_graph import cli; cli()"

Option 2

Wrap option 1 into a Python file which then can be executed:

from mi_graph import cli

cli()

Now run the file:

python <file path>

Setup: virtualenv environment

Follow this guide to setup a virtual environment for development

Setup: conda environment

Install conda from their website. (We recommend the mini version / miniconda)

Initialize the environment with:

• conda create --name knox-env python=3.8
• activate knox-env
• pip install -r requirements.txt

Test dependencies

For developers, you also need to install the test requirements:

• pip install -r tests/requirements.txt

Pylint

Before you make a pull request to master, you should run branch though pylint.

you can use pylint_runner to run all folders in the solution, or with pylint like so Run:

pip install pylint

And then

pylint folder/

Build module

You cannot be in a virtual environment, when building

• python3 setup.py sdist bdist_wheel

Output artifacts:

• knox-mi-graph-1.0.2.tar.gz
• knox_mi_graph-1.0.2-py3-none-any.whl

When the command is finished, it outputs a set of artifacts in a dist folder. Copy the artifacts to the knox server knox-master02.srv.aau.dk in the following path: /srv/web/repos.knox.cs.aau.dk/https/

The final result after completing the steps above:

• knox-master02.srv.aau.dk/srv/web/repos.knox.cs.aau.dk/https/knox-mi-graph-1.0.2.tar.gz
• knox-master02.srv.aau.dk/srv/web/repos.knox.cs.aau.dk/https/knox_mi_graph-1.0.2-py3-none-any.whl