gnn
gnn copied to clipboard
tensorflow
TensorFlow GNN is a library to build Graph Neural Networks on the TensorFlow platform.
![copybara-service[bot] avatar](https://avatars.githubusercontent.com/in/44061?v=4 "copybara-service[bot] avatar"){kind=avatar}
Implement full GATv2 model for in-memory examples using GATv2 Layer.
Add multitask support to orchestration. Runner `task` may now be a nested structure of Tasks. Note: the introduction of a `name` argument to Tasks is by virtue of Keras layer...
Adds support for converting of GraphTensorSpec to GraphSchema.
API for constructing a homogeneous GraphTensor
Initial support for graph pieces shapes relaxation (nodes and edges). Helper function to easy dataset creation from scalar graph pieces generator.
Fix the padding api to handle dataset creation from the distribute_datasets_from_function, and use the same SizeConstraints for padding and filtering.
keras_fit.py: s/tf.keras.callbacks.experimental.BackupAndRestore/tf.keras.callbacks.BackupAndRestore/
Metadata
Owner
Metadata
TensorFlow GNN is a library to build Graph Neural Networks on the TensorFlow platform.