tensorflow/gnn
TensorFlow library for building Graph Neural Networks with heterogeneous graph support, Keras layers, and training orchestration.
Not currently ranked — collecting fresh signals.
star history
TensorFlow GNN provides a GraphTensor type for representing graphs with multiple node and edge types, tools for data preparation including graph samplers to handle large-scale subgraph sampling, and a collection of ready-to-use GNN models and Keras layers. It also offers a high-level API for training orchestration and is based on a Google-internal library used for scalable graph mining tasks.
Frequently asked
- What is tensorflow/gnn?
- TensorFlow library for building Graph Neural Networks with heterogeneous graph support, Keras layers, and training orchestration.
- Is gnn open source?
- Yes — tensorflow/gnn is open source, released under the Apache-2.0 license.
- What language is gnn written in?
- tensorflow/gnn is primarily written in Python.
- How popular is gnn?
- tensorflow/gnn has 1.5k stars on GitHub.
- Where can I find gnn?
- tensorflow/gnn is on GitHub at https://github.com/tensorflow/gnn.