TensorFlow GNN is a library to build Graph Neural Networks on the TensorFlow platform.
APACHE-2.0 License
Bot releases are visible (Hide)
Release 1.0 is the first with a stable public API.
>=2.12,<2.16
, incompatible Keras v3 raises a clear error.GraphSchema
protobuf are now exposed undertfgnn.proto.*
.tfgnn.experimental.in_memory
is removed in favor of the new sampler.tfgnn_sampler
replaces the old tfgnn_graph_sampler
.tfgnn.*
functions on GraphTensor now work in Keras' Functional API, including the factory methods GraphTensor.from_pieces(...)
etc.tfgnn.disable_graph_tensor_validation()
.tfgnn.enable_graph_tensor_validation_at_runtime()
.GraphTensor
maintains .row_splits_dtype
separately from .indices_dtype
.GraphSchema
and the I/O functions for tf.Example
now support all non-quantized, non-complex floating-point and integer types as well as bool
and string
.tfgnn.pool_neighbors_to_node()
.tfgnn.random_graph_tensor()
, now respects component boundaries.models/contrastive_losses
gets multiple extensions, including a triplet loss and API docs.models/multi_head_attention
replaces sigmoid with elu+1 in trained scaling.Full Changelog: https://github.com/tensorflow/gnn/compare/v0.6.1...v1.0.0
tf_keras
is installed but not used as tf.keras
(https://github.com/tensorflow/gnn/commit/ffa453f712fcf07a2a918726fd3cf622e39f9a50) (https://github.com/tensorflow/gnn/commit/e1d921051d22c10d6f322017f496408b78a98452).Full Changelog: https://github.com/tensorflow/gnn/compare/v1.0.1...v1.0.2
runner.GraphMean*Error
: the reduce_type
is again passed through correctly (https://github.com/tensorflow/gnn/commit/19c10f2489e3e62efad0fefe27aa9f5002bd2ce5).Full Changelog: https://github.com/tensorflow/gnn/compare/v1.0.0...v1.0.1
Published by arnoegw 9 months ago
Release 1.0 is the first with a stable public API.
>=2.12,<2.16
, incompatible Keras v3 raises a clear error.GraphSchema
protobuf are now exposed undertfgnn.proto.*
.tfgnn.experimental.in_memory
is removed in favor of the new sampler.tfgnn_sampler
replaces the old tfgnn_graph_sampler
.tfgnn.*
functions on GraphTensor now work in Keras' Functional API, including the factory methods GraphTensor.from_pieces(...)
etc.tfgnn.disable_graph_tensor_validation()
.tfgnn.enable_graph_tensor_validation_at_runtime()
.GraphTensor
maintains .row_splits_dtype
separately from .indices_dtype
.GraphSchema
and the I/O functions for tf.Example
now support all non-quantized, non-complex floating-point and integer types as well as bool
and string
.tfgnn.pool_neighbors_to_node()
.tfgnn.random_graph_tensor()
, now respects component boundaries.models/contrastive_losses
gets multiple extensions, including a triplet loss and API docs.models/multi_head_attention
replaces sigmoid with elu+1 in trained scaling.Full Changelog: https://github.com/tensorflow/gnn/compare/v0.6.1...v1.0.0
tf_keras
is installed but not used as tf.keras
(https://github.com/tensorflow/gnn/commit/ffa453f712fcf07a2a918726fd3cf622e39f9a50) (https://github.com/tensorflow/gnn/commit/e1d921051d22c10d6f322017f496408b78a98452).Full Changelog: https://github.com/tensorflow/gnn/compare/v1.0.1...v1.0.2rc1
runner.GraphMean*Error
: the reduce_type
is again passed through correctly (https://github.com/tensorflow/gnn/commit/19c10f2489e3e62efad0fefe27aa9f5002bd2ce5).Full Changelog: https://github.com/tensorflow/gnn/compare/v1.0.0...v1.0.1
Published by arnoegw 9 months ago
Release 1.0 is the first with a stable public API.
>=2.12,<2.16
, incompatible Keras v3 raises a clear error.GraphSchema
protobuf are now exposed undertfgnn.proto.*
.tfgnn.experimental.in_memory
is removed in favor of the new sampler.tfgnn_sampler
replaces the old tfgnn_graph_sampler
.tfgnn.*
functions on GraphTensor now work in Keras' Functional API, including the factory methods GraphTensor.from_pieces(...)
etc.tfgnn.disable_graph_tensor_validation()
.tfgnn.enable_graph_tensor_validation_at_runtime()
.GraphTensor
maintains .row_splits_dtype
separately from .indices_dtype
.GraphSchema
and the I/O functions for tf.Example
now support all non-quantized, non-complex floating-point and integer types as well as bool
and string
.tfgnn.pool_neighbors_to_node()
.tfgnn.random_graph_tensor()
, now respects component boundaries.models/contrastive_losses
gets multiple extensions, including a triplet loss and API docs.models/multi_head_attention
replaces sigmoid with elu+1 in trained scaling.Full Changelog: https://github.com/tensorflow/gnn/compare/v0.6.1...v1.0.0
tf_keras
is installed but not used as tf.keras
(https://github.com/tensorflow/gnn/commit/ffa453f712fcf07a2a918726fd3cf622e39f9a50).Full Changelog: https://github.com/tensorflow/gnn/compare/v1.0.1...v1.0.2rc0
runner.GraphMean*Error
: the reduce_type
is again passed through correctly (https://github.com/tensorflow/gnn/commit/19c10f2489e3e62efad0fefe27aa9f5002bd2ce5).Full Changelog: https://github.com/tensorflow/gnn/compare/v1.0.0...v1.0.1
Published by arnoegw 9 months ago
Release 1.0 is the first with a stable public API.
>=2.12,<2.16
, incompatible Keras v3 raises a clear error.GraphSchema
protobuf are now exposed undertfgnn.proto.*
.tfgnn.experimental.in_memory
is removed in favor of the new sampler.tfgnn_sampler
replaces the old tfgnn_graph_sampler
.tfgnn.*
functions on GraphTensor now work in Keras' Functional API, including the factory methods GraphTensor.from_pieces(...)
etc.tfgnn.disable_graph_tensor_validation()
.tfgnn.enable_graph_tensor_validation_at_runtime()
.GraphTensor
maintains .row_splits_dtype
separately from .indices_dtype
.GraphSchema
and the I/O functions for tf.Example
now support all non-quantized, non-complex floating-point and integer types as well as bool
and string
.tfgnn.pool_neighbors_to_node()
.tfgnn.random_graph_tensor()
, now respects component boundaries.models/contrastive_losses
gets multiple extensions, including a triplet loss and API docs.models/multi_head_attention
replaces sigmoid with elu+1 in trained scaling.Full Changelog: https://github.com/tensorflow/gnn/compare/v0.6.1...v1.0.0
runner.GraphMean*Error
: the reduce_type
is again passed through correctlyFull Changelog: https://github.com/tensorflow/gnn/compare/v1.0.0...v1.0.1
Published by arnoegw 10 months ago
First release with a stable public API.
>=2.12,<2.16
, incompatible Keras v3 raises a clear error.GraphSchema
protobuf are now exposed undertfgnn.proto.*
.tfgnn.experimental.in_memory
is removed in favor of the new sampler.tfgnn_sampler
replaces the old tfgnn_graph_sampler
.tfgnn.*
functions on GraphTensor now work in Keras' Functional API, including the factory methods GraphTensor.from_pieces(...)
etc.tfgnn.disable_graph_tensor_validation()
.tfgnn.enable_graph_tensor_validation_at_runtime()
.GraphTensor
maintains .row_splits_dtype
separately from .indices_dtype
.GraphSchema
and the I/O functions for tf.Example
now support all non-quantized, non-complex floating-point and integer types as well as bool
and string
.tfgnn.pool_neighbors_to_node()
.tfgnn.random_graph_tensor()
, now respects component boundaries.models/contrastive_losses
gets multiple extensions, including a triplet loss and API docs.models/multi_head_attention
replaces sigmoid with elu+1 in trained scaling.Full Changelog: https://github.com/tensorflow/gnn/compare/v0.6.1...v1.0.0
Published by arnoegw 10 months ago
First release with a stable public API.
>=2.12,<2.16
, incompatible Keras v3 raises a clear error.GraphSchema
protobuf are now exposed undertfgnn.proto.*
.tfgnn.experimental.in_memory
is removed in favor of the new sampler.tfgnn_sampler
replaces the old tfgnn_graph_sampler
.tfgnn.*
functions on GraphTensor now work in Keras' Functional API, including the factory methods GraphTensor.from_pieces(...)
etc.tfgnn.disable_graph_tensor_validation()
.tfgnn.enable_graph_tensor_validation_at_runtime()
.GraphTensor
maintains .row_splits_dtype
separately from .indices_dtype
.GraphSchema
and the I/O functions for tf.Example
now support all non-quantized, non-complex floating-point and integer types as well as bool
and string
.tfgnn.pool_neighbors_to_node()
.tfgnn.random_graph_tensor()
, now respects component boundaries.models/contrastive_losses
gets multiple extensions, including a triplet loss and API docs.models/multi_head_attention
replaces sigmoid with elu+1 in trained scaling.Full Changelog: https://github.com/tensorflow/gnn/compare/v0.6.1...v1.0.0
Published by arnoegw 10 months ago
Initial release candidate for v1.0.0.
Published by arnoegw 10 months ago
Early developmental release of tensorflow-gnn 1.0.0 code; docs still unfinished.
Published by arnoegw 11 months ago
import tensorflow_gnn
now checks if the version of tf.keras
is compatible.Published by arnoegw 11 months ago
Early developmental release of tensorflow-gnn 1.0.0 code; docs still outdated.
Published by mihirparadkar about 1 year ago
ResidualNextState
. by @copybara-service in https://github.com/tensorflow/gnn/pull/353
RunResult
. by @copybara-service in https://github.com/tensorflow/gnn/pull/395
attribution.py
testing: the preprocessing model created by orchestration returns a Sequence[GraphTensor]
. by @copybara-service in https://github.com/tensorflow/gnn/pull/442
ReadoutNamedIntoFeature
and AddReadoutFromFirstNode
by @copybara-service in https://github.com/tensorflow/gnn/pull/452
maxval += 1
in the case of an integer random counterfactual. by @copybara-service in https://github.com/tensorflow/gnn/pull/457
graph_sampler.py
by @copybara-service in https://github.com/tensorflow/gnn/pull/469
indices_dtype
as two concepts: by @copybara-service in https://github.com/tensorflow/gnn/pull/486
SimpleConv(..., receiver_feature=None)
is supported by @copybara-service in https://github.com/tensorflow/gnn/pull/506
tfgnn
core package. by @copybara-service in https://github.com/tensorflow/gnn/pull/519
Full Changelog: https://github.com/tensorflow/gnn/compare/v0.5.1...v0.6.0
Published by mihirparadkar about 1 year ago
Full Changelog: https://github.com/tensorflow/gnn/compare/v0.6.0rc0...v0.6.0rc1
Published by mihirparadkar about 1 year ago
ResidualNextState
. by @copybara-service in https://github.com/tensorflow/gnn/pull/353
RunResult
. by @copybara-service in https://github.com/tensorflow/gnn/pull/395
attribution.py
testing: the preprocessing model created by orchestration returns a Sequence[GraphTensor]
. by @copybara-service in https://github.com/tensorflow/gnn/pull/442
ReadoutNamedIntoFeature
and AddReadoutFromFirstNode
by @copybara-service in https://github.com/tensorflow/gnn/pull/452
maxval += 1
in the case of an integer random counterfactual. by @copybara-service in https://github.com/tensorflow/gnn/pull/457
graph_sampler.py
by @copybara-service in https://github.com/tensorflow/gnn/pull/469
indices_dtype
as two concepts: by @copybara-service in https://github.com/tensorflow/gnn/pull/486
SimpleConv(..., receiver_feature=None)
is supported by @copybara-service in https://github.com/tensorflow/gnn/pull/506
tfgnn
core package. by @copybara-service in https://github.com/tensorflow/gnn/pull/519
Full Changelog: https://github.com/tensorflow/gnn/compare/v0.5.1...v0.6.0rc0
Published by mihirparadkar over 1 year ago
Full Changelog: https://github.com/tensorflow/gnn/compare/v0.5.0...v0.5.1
Published by mihirparadkar over 1 year ago
Full Changelog: https://github.com/tensorflow/gnn/compare/v0.5.0rc0...v0.5.0
Published by mihirparadkar over 1 year ago
Full Changelog: https://github.com/tensorflow/gnn/compare/v0.4.0...v0.5.0rc0
Published by mihirparadkar almost 2 years ago
Full Changelog: https://github.com/tensorflow/gnn/compare/v0.4.0...v0.4.1
Published by mihirparadkar almost 2 years ago
Full Changelog: https://github.com/tensorflow/gnn/compare/v0.4.0rc2...v0.4.0
Published by mihirparadkar almost 2 years ago
Bugfix for python 3.7 compatibility
Full Changelog: https://github.com/tensorflow/gnn/compare/v0.4.0rc1...v0.4.0rc2
Published by mihirparadkar almost 2 years ago
Removes explicit requirement for google-vizier to stay compatible with google Colab.
Full Changelog: https://github.com/tensorflow/gnn/compare/v0.4.0rc0...v0.4.0rc1