graphomaly 0.3

Anomaly detection in graphs modeling financial transactions and computer networks.

Graphomaly

Automatic tool for Anti-Money Laundering (AML) and detecting abnormal behavior in computer networks. Find abnormal data in graph and network structures.

Official package documentation here.

This work was initially supported by the Graphomaly Research Grant and later partially supported by the Netalert Research Grant.

Installation and setup

Install via pip from the PyPi repository:

pip install graphomaly

or for the latest changes not yet in the official release:

pip install git+https://gitlab.com/unibuc/graphomaly/graphomaly

Install via docker from the DockerHub repository

docker pull pirofti/graphomaly

For using the GPU pull the dedicated image:

docker pull pirofti/graphomaly:latest_gpu

Usage

The package follows the sklearn API and can be included in your projects via

from graphomaly.estimator import GraphomalyEstimator

which will provide you with a standard scikit-learn estimator that you can use in your pipeline.

For configuration and tweaks please consult the YAML file for now until documentation matures.

Development and testing

First clone the repository and change directory to the root of your fresh checkout.

0. Install Prerequisites

Install PyPA’s build:

python3 -m pip install --upgrade build

1. Build

Inside the Graphomaly directory

python -m build

2. Virtual Environment

Create a virtual environment with Python:

python -m venv venv

Activate the environment:

source venv/bin/activate

For Windows execute instead:

venv\Scripts\activate

3. Install

Inside the virtual environment execute:

pip install dist/graphomaly-*.whl

Running unit tests

First create the results directory:

mkdir -p tests/results/synthetic

Run the initial test on synthetic data to make sure things installed ok:

cd tests && python test_synthetic

Then run the other unit tests by hand as above or via pytest:

pytest  # add -v for verbose, add -s to print(...) to console from tests