```bash
    docker run -it --rm -v $PWD:/tmp -w /tmp tensorflow/build:2.15-python3.10
    ```

    Once you have the packages installed, you can run a specific unit test in
    bazel by doing as follows:

    ```bash
    export flags="--config=opt -k"
    ```

    If the tests are to be run on the GPU:

                                            TF_Status* status) const;

  // The number of devices operations run on.
  size_t num_underlying_devices() const { return underlying_devices_.size(); }

  // The devices operations run on.
  const std::vector<std::string>& underlying_devices() const {
    return underlying_devices_;
  }

  TFE_DeleteContextOptions(opts);

  TF_ExecutionContextRegisterFunction(eager_execution_ctx, func, status.get());
  ASSERT_EQ(TF_OK, TF_GetCode(status.get())) << TF_Message(status.get());

  // Build the abstract op to run the function.
  TF_AbstractOp* fn_op = TF_NewAbstractOp(eager_execution_ctx);
  TF_AbstractOpSetOpType(fn_op, fn_name.c_str(), status.get());
  ASSERT_EQ(TF_OK, TF_GetCode(status.get())) << TF_Message(status.get());

      case_name = f'{command_type} invocation {invocation_id}'
    else:
      case_name = f' Invocation {invocation_id}'
    case_attrib.update({'name': case_name,
                        'status': 'run', 'result': 'completed'})

    testcase = ElemTree.SubElement(testsuite, 'testcase', attrib=case_attrib)
    if status in (InvokeStatus.tests_failed, InvokeStatus.build_failed):
      if status == InvokeStatus.tests_failed:

#
# This file is autogenerated by pip-compile with Python 3.10
# by the following command:
#
#    bazel run //ci/official/requirements_updater:requirements.update
#
absl-py==2.1.0 \
    --hash=sha256:526a04eadab8b4ee719ce68f204172ead1027549089702d99b9059f129ff1308 \
    --hash=sha256:7820790efbb316739cde8b4e19357243fc3608a152024288513dd968d7d959ff
    # via
    #   keras-nightly
    #   tb-nightly
astor==0.7.1 \

#
# This file is autogenerated by pip-compile with Python 3.9
# by the following command:
#
#    bazel run //ci/official/requirements_updater:requirements.update
#
absl-py==2.1.0 \
    --hash=sha256:526a04eadab8b4ee719ce68f204172ead1027549089702d99b9059f129ff1308 \
    --hash=sha256:7820790efbb316739cde8b4e19357243fc3608a152024288513dd968d7d959ff
    # via
    #   keras-nightly
    #   tb-nightly
astor==0.7.1 \

#
# This file is autogenerated by pip-compile with Python 3.11
# by the following command:
#
#    bazel run //ci/official/requirements_updater:requirements.update
#
absl-py==2.1.0 \
    --hash=sha256:526a04eadab8b4ee719ce68f204172ead1027549089702d99b9059f129ff1308 \
    --hash=sha256:7820790efbb316739cde8b4e19357243fc3608a152024288513dd968d7d959ff
    # via
    #   keras-nightly
    #   tb-nightly
astor==0.7.1 \

  TensorHandlePtr parallel_value = CreatePerDeviceValues(
      context.get(), components, device_name, status.get());
  ASSERT_EQ(TF_GetCode(status.get()), TF_OK) << TF_Message(status.get());

  // Run a collective sum, so each component should now be the same.
  TensorHandlePtr reduced(
      CollectiveSum(context.get(), parallel_value.get(), 2, status.get()));

        op_(nullptr),
        thread_(tensorflow::Env::Default()->StartThread(
            tensorflow::ThreadOptions(), "parallel_device_execute",
            std::bind(&DeviceThread::Run, this))) {}
  ~DeviceThread();

  // Requests that the worker thread execute the specified operation. Blocks
  // until the previously pending operation (a StartExecute without a Join) has
  // finished, if any.
  //

// TF_NewAttrBuilder() returns an object that you can set attributes on as
// though it were an op. This allows querying properties of that op for
// type-checking purposes like if the op will run on a particular device type.
typedef struct TF_AttrBuilder TF_AttrBuilder;
TF_CAPI_EXPORT extern TF_AttrBuilder* TF_NewAttrBuilder(const char* op_name);
TF_CAPI_EXPORT extern void TF_DeleteAttrBuilder(TF_AttrBuilder* builder);