Tfx-bsl

Major Features and Improvements

* `InferTensorRepresentationsFromSchema`, `TensorAdapter` and
`TensorsToRecordBatchConverter` now support `SparseTensor`s with unknown
`dense_shape`.

Bug Fixes and Other Changes

* Depends on `tensorflow>=2.11,<3`

Breaking Changes

* N/A

Deprecations

* N/A

Major Features and Improvements

* `TensorAdapter` now processes `tf.RaggedTensor`s in TF 2 ~10x faster.
* `InferTensorRepresentationsFromSchema` now infers `RaggedTensor`s for
`STRUCT` features.
* `TFSequenceExampleRecord` now supports schemas with features not covered or
partially covered by `TensorRepresentation`s.

* This is the last version that supports TensorFlow 1.15.x. TF 1.15.x support
will be removed in the next version. Please check the
[TF2 migration guide](https://www.tensorflow.org/guide/migrate) to migrate
to TF2.

Bug Fixes and Other Changes

* Depends on `tensorflow>=1.15.5,<2` or `tensorflow>=2.10,<3`
* Depends on `protobuf>=3.13,<4`
* Various `TFXIO` implementations now infer `TensorRepresentations` for
provided schema `Features` even if some `TensorRepresentations` are provided
as well.

Breaking Changes

* N/A

Deprecations

* N/A

Major Features and Improvements

* `ExamplesToRecordBatchDecoder` is now picklable.
* `ParquetTFXIO` can now be used as `RecordBasedTFXIO`.
* Introduces `CreateTfSequenceExampleParserConfig` that takes TFMD schema as
input and produces configs for `tf.SequenceExample` parsing.
* `TFSequenceExampleRecord` can now produce an equivalent tf.data.Dataset.
* Introduces an api: `CreateModelHandler` that produces a model handler
suitable for apache_beam.ml.inference.
* Quantiles sketch supports GetQuantilesAndCumulativeWeights, which returns
the sum of weights in each quantiles bin along with boundaries.

Bug Fixes and Other Changes

* Depends on `apache-beam[gcp]>=2.40,<3`.
* Depends on `pyarrow>=6,<7`.
* Depends on `tensorflow-metadata>=1.10,<1.11`.
* Depends on `tensorflow>=1.15.5,<2` or `tensorflow>=2.9,<3`.

Breaking Changes

* GenerateQuantiles removed from weighted_quantiles_summary.h and replaced
with GenerateQuantilesAndCumulativeWeights.

Deprecations

* N/A

Major Features and Improvements

* N/A

Bug Fixes and Other Changes

* Depends on `tensorflow-metadata>=1.9,<1.10`.
* Depends on `tensorflow>=1.15.5,<2` or `tensorflow>=2.9,<3`.
* Depends on `protobuf>=3.13,<3.21`.

Breaking Changes

* N/A

Deprecations

* N/A

Major Features and Improvements

* Introduced `RunInferencePerModel` PTransform, which is a vectorized variant
of `RunInference` (useful for ensembles).
* Introduced `ParquetTFXIO` that allows reading data from Parquet files in
`pyarrow.RecordBatch` format.
* From this version we will be releasing python 3.9 wheels.
* Depends on `apache-beam[gcp]>=2.38,<3`.

Bug Fixes and Other Changes

* Depends on `tensorflow-metadata>=1.8,<1.9`.

Breaking Changes

* N/A

Deprecations

* N/A

Major Features and Improvements

* N/A

Bug Fixes and Other Changes

* Depends on `apache-beam[gcp]>=2.36,<3`.
* Depends on `tensorflow-metadata>=1.7,<1.8`.
* Depends on
`tensorflow>=1.15.5,!=2.0.*,!=2.1.*,!=2.2.*,!=2.3.*,!=2.4.*,!=2.5.*,!=2.6.*,!=2.7.*,<3`.
* Depends on
`tensorflow-serving-api>=1.15,!=2.0.*,!=2.1.*,!=2.2.*,!=2.3.*,!=2.4.*,!=2.5.*,!=2.6.*,!=2.7.*,<3`.
* Added a TFXIO where the user defines the beam source.

Breaking Changes

* N/A

Deprecations

* N/A