Merged commit includes the following changes:
433332091  by A. Unique TensorFlower<gardener@tensorflow.org>:

    [tf.data] Ramp up `initial_parallelism_value` experiment to 100%

--
433332037  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Return conversion error if `StatefulPartitionedCall` or `ParitionedCall` is in the result model.

    Neither `StatefulPartitionedCall` nor `ParitionedCall` can be executed by TF Lite, so we should raise a conversion error if such an op appears in the result model.

--
433331676  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Add ContainerMap forward type inference function

    Helper for ops with semantics of applying some transformation to the elements of a container, e.g. changing the batch size of a dataset.

    `<t>[PRODUCT[T1, ..., Tn]] -> <t>[PRODUCT[U1, ..., Un]]`

    where Ui is obtained by applying a map T -> U. Both <t> and the "map"
    function are parameterized by this factory.

    Note that when the "map" function is the identity function, this becomes equivalent to ReplicateInput.

--
433331338  by A. Unique TensorFlower<gardener@tensorflow.org>:

    [NNAPI] Allow dequantizing fp16 scalar input

--
433327321  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Go: Update generated wrapper functions for TensorFlow ops.

--
433324664  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Update tf.py_function and tf.numpy_function docstring.

--
433324643  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Implement try_generalizing_trace_type for optimize_tracing_count

--
433323662  by A. Unique TensorFlower<gardener@tensorflow.org>:

    [XNNPACK] Rename concatenation tests to include number of ouputs in name

    Also introduce a helper to create shapes that match in all dimensions except the axis.

--
433316824  by A. Unique TensorFlower<gardener@tensorflow.org>:

    [XNNPACK] Change ConcatenationTester to be able to take multiple input shapes

    This will allow us to test concatenation with more than 2 inputs.

--
433316507  by A. Unique TensorFlower<gardener@tensorflow.org>:

    PR #53796: Add linux_aarch64 target to AOT.

    Imported from GitHub PR https://github.com/tensorflow/tensorflow/pull/53796

    The lack of a target for `linux_aarch64` on AOT breaks tests under `//tensorflow/python/tools`.
    Copybara import of the project:

    --
    e50a9a79b73b1a5b5ef258fc28bde3d6545043e1 by Everton Constantino <everton.constantino@linaro.org>:

    Add linux_aarch64 target to AOT.

--
433314278  by A. Unique TensorFlower<gardener@tensorflow.org>:

    mutate xplane for kernel launch flows, not enabled in OSS, because missing event and json support.

--
433313910  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Export Tracing Protocol

--
433305742  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Add XlaOptimizationBarrier tensorflow op.

    See https://www.tensorflow.org/xla/operation_semantics#optimizationbarrier.

    The use case for this is to support the new XLA remat op in jax2tf.

--
433303419  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Add missing Copyright notice in the new files

--
433290108  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Go: Update generated wrapper functions for TensorFlow ops.

--
433289730  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Change call sites from deprecated `mlir::parseSourceFile()` to `mlir::parseSourceFile<::mlir::ModuleOp>()`.

--
433285889  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Implement parser/printer as member methods.

    See https://reviews.llvm.org/D121090.

--
433284791  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Update TFRT dependency to use revision
    http://github.com/tensorflow/runtime/commit/d7dd5e431e9319a690d898f81193d09e87bc8643.

--
433283527  by A. Unique TensorFlower<gardener@tensorflow.org>:

    [XLIR] Support bf16 type for gemm.

--
433280976  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Set print_operand_shape=false for HloPrintOptions::Fingerprint(). This reduces fingerprint computation time by 10-100x.

--
433280889  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Add Remapper pass for TFG.

    Implements ReplaceSigmoidMulWithSwish as a beginning pattern, which is migrated from grappler

--

PiperOrigin-RevId: 433332091
70 files changed
tree: c1f3f41ee6ca3c32612ffc3a9a2cd85872c83b69
  1. .github/
  2. tensorflow/
  3. third_party/
  4. tools/
  5. .bazelrc
  6. .bazelversion
  7. .clang-format
  8. .gitignore
  9. .zenodo.json
  10. arm_compiler.BUILD
  11. AUTHORS
  12. BUILD
  13. CITATION.cff
  14. CODE_OF_CONDUCT.md
  15. CODEOWNERS
  16. configure
  17. configure.cmd
  18. configure.py
  19. CONTRIBUTING.md
  20. ISSUE_TEMPLATE.md
  21. ISSUES.md
  22. LICENSE
  23. models.BUILD
  24. README.md
  25. RELEASE.md
  26. SECURITY.md
  27. WORKSPACE
README.md

Python PyPI DOI

Documentation
Documentation

TensorFlow is an end-to-end open source platform for machine learning. It has a comprehensive, flexible ecosystem of tools, libraries, and community resources that lets researchers push the state-of-the-art in ML and developers easily build and deploy ML-powered applications.

TensorFlow was originally developed by researchers and engineers working on the Google Brain team within Google's Machine Intelligence Research organization to conduct machine learning and deep neural networks research. The system is general enough to be applicable in a wide variety of other domains, as well.

TensorFlow provides stable Python and C++ APIs, as well as non-guaranteed backward compatible API for other languages.

Keep up-to-date with release announcements and security updates by subscribing to announce@tensorflow.org. See all the mailing lists.

Install

See the TensorFlow install guide for the pip package, to enable GPU support, use a Docker container, and build from source.

To install the current release, which includes support for CUDA-enabled GPU cards (Ubuntu and Windows):

$ pip install tensorflow

A smaller CPU-only package is also available:

$ pip install tensorflow-cpu

To update TensorFlow to the latest version, add --upgrade flag to the above commands.

Nightly binaries are available for testing using the tf-nightly and tf-nightly-cpu packages on PyPi.

Try your first TensorFlow program

$ python
>>> import tensorflow as tf
>>> tf.add(1, 2).numpy()
3
>>> hello = tf.constant('Hello, TensorFlow!')
>>> hello.numpy()
b'Hello, TensorFlow!'

For more examples, see the TensorFlow tutorials.

Contribution guidelines

If you want to contribute to TensorFlow, be sure to review the contribution guidelines. This project adheres to TensorFlow's code of conduct. By participating, you are expected to uphold this code.

We use GitHub issues for tracking requests and bugs, please see TensorFlow Discuss for general questions and discussion, and please direct specific questions to Stack Overflow.

The TensorFlow project strives to abide by generally accepted best practices in open-source software development:

Fuzzing Status CII Best Practices Contributor Covenant

Continuous build status

You can find more community-supported platforms and configurations in the TensorFlow SIG Build community builds table.

Official Builds

Build TypeStatusArtifacts
Linux CPUStatusPyPI
Linux GPUStatusPyPI
Linux XLAStatusTBA
macOSStatusPyPI
Windows CPUStatusPyPI
Windows GPUStatusPyPI
AndroidStatusDownload
Raspberry Pi 0 and 1StatusPy3
Raspberry Pi 2 and 3StatusPy3
Libtensorflow MacOS CPUStatus Temporarily UnavailableNightly Binary Official GCS
Libtensorflow Linux CPUStatus Temporarily UnavailableNightly Binary Official GCS
Libtensorflow Linux GPUStatus Temporarily UnavailableNightly Binary Official GCS
Libtensorflow Windows CPUStatus Temporarily UnavailableNightly Binary Official GCS
Libtensorflow Windows GPUStatus Temporarily UnavailableNightly Binary Official GCS

Resources

Learn more about the TensorFlow community and how to contribute.

License

Apache License 2.0