/graphium

Graphium: Scaling molecular GNNs to infinity.

Primary LanguagePythonApache License 2.0Apache-2.0

Scaling molecular GNNs to infinity


Run on Gradient PyPI Conda PyPI - Downloads Conda license GitHub Repo stars GitHub Repo stars test test-ipu release code-check doc codecov hydra

A deep learning library focused on graph representation learning for real-world chemical tasks.

  • ✅ State-of-the-art GNN architectures.
  • 🐍 Extensible API: build your own GNN model and train it with ease.
  • ⚗️ Rich featurization: powerful and flexible built-in molecular featurization.
  • 🧠 Pretrained models: for fast and easy inference or transfer learning.
  • ⮔ Read-to-use training loop based on Pytorch Lightning.
  • 🔌 Have a new dataset? Graphium provides a simple plug-and-play interface. Change the path, the name of the columns to predict, the atomic featurization, and you’re ready to play!

Documentation

Visit https://graphium-docs.datamol.io/.

Run on Gradient

You can try running Graphium on Graphcore IPUs for free on Gradient by clicking on the button above.

Installation for developers

For CPU and GPU developers

Use mamba:

# Install Graphium's dependencies in a new environment named `graphium`
mamba env create -f env.yml -n graphium

# Install Graphium in dev mode
mamba activate graphium
pip install --no-deps -e .

For IPU developers

# Install Graphcore's SDK and Graphium dependencies in a new environment called `.graphium_ipu`
./install_ipu.sh .graphium_ipu

The above step needs to be done once. After that, enable the SDK and the environment as follows:

source enable_ipu.sh .graphium_ipu

The Graphium CLI

Installing graphium makes two CLI tools available: graphium and graphium-train. These CLI tools make it easy to access advanced functionality, such as training a model, extracting fingerprints from a pre-trained model or precomputing the dataset. For more information, visit the documentation.

License

Under the Apache-2.0 license. See LICENSE.

Documentation

  • Diagram for data processing in Graphium.

Data Processing Chart

  • Diagram for Muti-task network in Graphium

Full Graph Multi-task Network