deepgnn-ge 0.1.65

Graph engine - distributed graph engine to host graphs.

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.