#   are all settings for the scripts. TF's CI jobs are all made of a combination
#   of these env files.
#
#   If you've clicked on a test result from our CI (via a dashboard or GitHub link),
#   click to "Invocation Details" and find BUILD_CONFIG, which will contain a TFCI
#   value in the "env_vars" list that you can choose to copy that environment.
#      Ex. 1: TFCI=py311,linux_x86_cuda,nightly_upload  (nightly job)

system-installed Python packages. \
Instead, an independent Python toolchain is registered, ensuring the right
dependencies are always used. \
See https://github.com/bazelbuild/rules_python/ for more details.

### Specifying the Python version

Note: Only a number of minor Python versions are supported at any given time.

By default, the lowest supported version is used.

# Do an eager load for Keras' code so that any function/method that needs to
# happen at load time will trigger, eg registration of optimizers in the
# SavedModel registry.
# See b/196254385 for more details.
try:
  if _tf_uses_legacy_keras:
    importlib.import_module("tf_keras.src.optimizers")
  else:
    importlib.import_module("keras.src.optimizers")
except (ImportError, AttributeError):
  pass

    tensorflow/tools/ci_build/ci_build.sh CPU bazel test //tensorflow/...
    ```

    See
    [TensorFlow Builds](https://github.com/tensorflow/tensorflow/tree/master/tensorflow/tools/ci_build)
    for details.

#### Running doctest for testable docstring

There are two ways to test the code in the docstring locally:

1.  If you are only changing the docstring of a class/function/method, then you

    which is a NumPy-compatible API for writing TF programs. See the
    [detailed guide](https://www.tensorflow.org/guide/tf_numpy) to learn more.
    Additional details below.

*   Adds Support for
    [TensorFloat-32](https://blogs.nvidia.com/blog/2020/05/14/tensorfloat-32-precision-format/)
    on Ampere based GPUs. TensorFloat-32, or TF32 for short, is a math mode for

    echo 'behavior, create a new shell.'
    cat "$FROM_ENV"
    source "$FROM_ENV"
    rm "$FROM_ENV"
  fi
fi

# Mac builds have some specific setup needs. See setup_macos.sh for details
if [[ "${OSTYPE}" =~ darwin* ]]; then
  source ./ci/official/utilities/setup_macos.sh
fi

# Create and expand to the full path of TFCI_OUTPUT_DIR
export TFCI_OUTPUT_DIR=$(realpath "$TFCI_OUTPUT_DIR")

# MACOS X86 PYCPP
# These are defined as build configs so that we can run a build only job. See
# the note under "ARM64 PYCPP" for more details.
build:macos_x86_pycpp_test_filters --test_tag_filters=-no_oss,-oss_excluded,-oss_serial,-no_oss_py38,-no_oss_py39,-no_oss_py310,-nomac,-no_mac,-mac_excluded,-v1only,-gpu,-tpu,-benchmark-test

  configure_ios(environ_cp)

  print('Preconfigured Bazel build configs. You can use any of the below by '
        'adding "--config=<>" to your build command. See .bazelrc for more '
        'details.')
  config_info_line('mkl', 'Build with MKL support.')
  config_info_line(
      'mkl_aarch64',
      'Build with oneDNN and Compute Library for the Arm Architecture (ACL).')

      gtl::ArraySlice<Gradient*> output_gradients, absl::Span<Gradient*> result,
      bool build_default_zeros_grads = true);

  // Whether the tape is persistent. See ctor for detailed description.
  bool IsPersistent() const { return persistent_; }

 private:
  TensorTape tensor_tape_;
  OpTape<BackwardFunction, TapeTensor> op_tape_;
  int64_t next_op_id_{0};