- Linux wheels now specifically conform to manylinux2014, an upgrade from manylinux2010. This is aligned with TensorFlow 2.9 requirement.
- Depends on
tensorflow>=2.9.0,<2.10
. - Depends on
tensorflow-metadata>=1.9.0,<1.10.0
.
- N/A
- N/A
- From this version we will be releasing python 3.9 wheels.
- Depends on
tensorflow-metadata>=1.8.0,<1.9.0
.
- N/A
- N/A
- Added equi_join_any_indices_op.
- Added broadcast for subtrees.
- Depends on
tensorflow>=2.8.0,<2.9
. - Depends on
tensorflow-metadata>=1.7.0,<1.8.0
.
- N/A
- N/A
- N/A
- Depends on
tensorflow-metadata>=1.6.0,<1.7.0
.
- N/A
- N/A
- N/A
- Depends on
tensorflow>=2.7.0,<2.8
. - Depends on
tensorflow-metadata>=1.5.0,<1.6.0
.
- N/A
- N/A
- N/A
- Fix bug in which expression.apply_schema mutated its input schema
- N/A
- Deprecated python 3.6 support.
- N/A
- Depends on
tensorflow>=2.6.0,<2.7
. - Depends on
pyarrow>=1,<6
.
- N/A
- N/A
- N/A
- Added doc with benchmark numbers. Also added the benchmark code and test data.
- Depends on
tensorflow-metadata>=1.2.0,<1.3.0
.
- N/A
- N/A
- N/A
- Depends on
protobuf>=3.13,<4
. - Depends on
tensorflow-metadata>=1.1.0,<1.2.0
.
- N/A
- N/A
- N/A
- Introduced DecodeProtoSparseV4. It is same as V3 and will replace V3 soon.
- DecodeProtoSparseV3 is now the default (instead of V2).
- Bumped tf version for statically linked libraries to TF 2.5.0.
- Depends on
tensorflow>=2.5.0,<2.6
. - Depends on
tensorflow-metadata>=1.0.0,<1.1.0
.
- N/A
- N/A
- N/A
- Deprecate
get_ragged_tensors()
andget_sparse_tensors()
in prensor_util. - Expose
get_ragged_tensors()
andget_sparse_tensors()
asPrensor
methods. - Expose
get_positional_index
as a method ofNodeTensor
. - Depends on
tensorflow-metadata>=0.30,<0.31
- N/A
- N/A
- N/A
- Allow path to concat with string.
- Bumped the mininum bazel version required to build
struct2tensor
to 3.7.2. - Depends on
tensorflow-metadata>=0.29,<0.30
- N/A
- N/A
- N/A
- Depends on
tensorflow-metadata>=0.28,<0.29
- N/A
- N/A
- N/A
- Depends on
pyarrow>=1,<3
- Depends on
tensorflow>=2.4.0,<2.5
- Depends on
tensorflow-metadata>=0.27,<0.28
- N/A
- N/A
- Created a docker image that contains a TF model server with struct2tensor
ops linked. This docker image is available at
gcr.io/tfx-oss-public/s2t_tf_serving
. - Add support for string_views for intermediate serialized protos. To use, set the option "use_string_view" in CalculateOptions to true. string_views are potentially more memory bandwidth efficient depending on the depth and complexity of the input proto.
- Depends on
tensorflow-metadata>=0.26,<0.27
.
- N/A
- N/A
-
From this release Struct2Tensor will also be hosting nightly packages on https://pypi-nightly.tensorflow.org. To install the nightly package use the following command:
pip install --extra-index-url https://pypi-nightly.tensorflow.org/simple struct2tensor
Note: These nightly packages are unstable and breakages are likely to happen. The fix could often take a week or more depending on the complexity involved for the wheels to be available on the PyPI cloud service. You can always use the stable version of struct2tensor available on PyPI by running the command
pip install struct2tensor
.
- Update init.py to import the API, instead of just the modules.
- Provide an init.py for struct2tensor.expression_impl directory. This is meant for power users.
- Update python notebook to use import style.
- Fix bug in prensor_to_structured_tensor.
- Depends on
tensorflow-metadata>=0.25,<0.26
. - Depends on
pyarrow>=0.17,<1
.
- N/A
- N/A
- Add support for converting prensor to
StructuredTensor
.
- Depends on
tensorflow>=2.3.0,<2.4
- Depends on
tensorflow-metadata>=0.24,<0.25
- N/A
- Deprecated py3.5 support
- Add promote for substructures.
- Add support for converting
StructuredTensor
to prensor.
- Depends on
tensorflow>=2.3.0,<2.4
- Depends on
tensorflow-metadata>=0.23,<0.24
- Drop snappy support for parquet dataset.
- Deprecating Py2 support.
- Depends on `tensorflow>=2.2.0,<2.3
- Bumped Tensorflow version for statically linked libraries from 1.5 to 2.1.
- Added tests for statically linked libraries.
- Statically linked libraries build now.
- Parquet dataset that can apply expressions to a parquet schema, allowing for reading data from IO to tensors directly.
- Now requires tensorflow>=2.1.0,<2.2.
- Initial release of struct2tensor.