---
title: "pyTigerGraph GDS Module"
component: "ml-workbench"
version: "1.4C"
module: "faq"
html_url: "/ml-workbench/1.4C/faq/reference"
canonical_url: "https://www.tigergraph.com/docs/ml-workbench/1.4C/faq/reference"
markdown_url: "https://www.tigergraph.com/docs/ml-workbench/1.4C/faq/reference.md"
---

[View as HTML](https://www.tigergraph.com/docs/ml-workbench/1.4C/faq/reference)

# pyTigerGraph GDS Module

The centerpiece of the ML Workbench is its Python client:[pyTigerGraph](https://www.tigergraph.com/docs/pytigergraph/1.8/intro/index.md).

This Python package contains all the essential utilities you need to bootstrap your Graph Machine Learning journey with these key functions:

* Data loading/export
* Graph partitioning for preparing your training, validation, and test data set for your supervised graph machine learning model
* Featurizer for generating unique graph features
* Subgraph sampling for your stochastic training process.

Please refer to the [GDS Functions](https://www.tigergraph.com/docs/pytigergraph/1.8/gds/index.md) section of the pyTigerGraph documentation to learn more about the submodules, classes, and methods, and to see working examples and code snippets.
