Skip to content

Latest commit

 

History

History
26 lines (19 loc) · 1.04 KB

README.md

File metadata and controls

26 lines (19 loc) · 1.04 KB

TensorFlow GNN: API docs

TensorFlow GNN is a Python package that consists of multiple sub-libraries.

  • Core library [API docs]:
    import tensorflow_gnn as tfgnn

  • Training orchestration [API docs]:
    from tensorflow_gnn import runner

  • Graph sampler:
    from tensorflow_gnn.experimental import sampler

  • Models from the models collection:
    from tensorflow_gnn.models import <model name>

The TensorFlow GNN package uses semantic versioning for its numbered releases. Its stable Python API consists of the Python identifiers exposed in these imported modules and their sub-modules, with the following exceptions:

  • identifiers or modules that contain experimental in their name, that are expressly documented to be unstable, or not documented at all;
  • private identifiers beginning with a single underscore (_);
  • models from the models collection whose documentation opts out of semantic versioning.