Skip to content

DeepGNN is a framework for training machine learning models on large scale graph data.

License

Notifications You must be signed in to change notification settings

microsoft/DeepGNN

Folders and files

NameName
Last commit message
Last commit date
Jan 28, 2025
May 25, 2024
Jan 28, 2025
Jan 28, 2025
Jan 28, 2025
May 12, 2022
Jan 16, 2024
Aug 21, 2023
Jun 30, 2022
Sep 11, 2023
Jan 20, 2023
Aug 16, 2022
Apr 22, 2022
Jan 28, 2025
Apr 21, 2022
Aug 15, 2024
Apr 21, 2022
Oct 10, 2023
Apr 22, 2022
Apr 22, 2022
May 25, 2024
Feb 11, 2025

Repository files navigation

DeepGNN Overview

DeepGNN is a framework for training machine learning models on large scale graph data. DeepGNN contains all the necessary features including:

  • Distributed GNN training and inferencing on both CPU and GPU.
  • Custom graph neural network design.
  • Online Sampling: Graph Engine (GE) will load all graph data, each training worker will call GE to get node/edge/neighbor features and labels.
  • Automatic graph partitioning.
  • Highly performant and scalable.

Project is in alpha version, there might be breaking changes in the future and they will be documented in the changelog.

Usage

Install pip package:

python -m pip install deepgnn

If you want to build package from source, see instructions in CONTRIBUTING.md.

Train and evaluate a graphsage model with pytorch on cora dataset:

cd examples/pytorch
python sage.py

Migrating Scripts

We provide a python module to help you upgrade your scripts to new deepgnn versions.

pip install google-pasta
python -m deepgnn.migrate.0_1_56 --script_dir directory_to_migrate

See CHANGELOG.md for full change details.

About

DeepGNN is a framework for training machine learning models on large scale graph data.

Resources

License

Code of conduct

Security policy

Stars

Watchers

Forks

Packages

No packages published