pyTigerGraph is a Python package for connecting to TigerGraph databases. We offer two versions of the package: pyTigerGraph and pyTigerGraph[gds].
pyTigerGraph is the default version and contains the core functionality of pyTigerGraph, including the following:
Data manipulation functions:inserting, updating, upserting, deleting, and retrieving vertices and edges.
Query functions: running and managing queries inside the TigerGraph database
Metadata functions: fetching details of graphs/schemas, vertex and edge types, and other schema objects and object types
Authentication and authorization functions
Miscellaneous utility functions
The pyTigerGraph[gds] version of pyTigerGraph is a drop-in replacement for pyTigerGraph, but adds support for Graph Data Science and Graph machine learning capabilities. This includes:
Graph feature engineering using algorithms from the GSQL Graph Data Science Library.
Data loaders for training and inference of Graph Neural Network (GNN) models using PyTorch Geometric and DGL.
Machine learning utilities for splitting vertices into training, validation, and testing sets.
Checkout the Getting Started section for a quick introduction to pyTigerGraph. It walks you through how to perform the following:
In addition, we also provide a video tutorial and a Google Colab notebook that walks you through all core functions of pyTigergraph:
There are many community resources that you can use for help and support using pyTigerGraph:
Checkout the Contributing section for instructions on how to contribute. pyTigerGraph was started as an open-source community project, and we welcome contributions from the community.