vocabulary file.
*   Added `index_to_string_table` which returns a lookup table that maps indices
    to strings.
*   Add `string_to_index_table`, which returns a lookup table that matches
    strings to indices.
*   Add a `ParallelForWithWorkerId` function.
*   Add `string_to_index_table`, which returns a lookup table that matches
    strings to indices.

        "//tensorflow/c/eager:dlpack",
        "//tensorflow/c/eager:tape",
        "//tensorflow/c/eager:tfe_context_internal",
        "//tensorflow/c/eager:tfe_op_internal",
        "//tensorflow/c/eager:tfe_tensorhandle_internal",
        "//tensorflow/c/experimental/gradients",
        "//tensorflow/c/experimental/gradients/tape",
        "//tensorflow/c/experimental/next_pluggable_device:c_api",

             absl::Span<AbstractTensorHandle*> outputs) -> Status {
    Tape tape(/*persistent=*/false);
    for (size_t i{}; i < inputs.size(); ++i) {
      tape.Watch(inputs[i]);
    }
    std::vector<AbstractTensorHandle*> temp_outputs(1);
    AbstractContextPtr tape_ctx(new TapeContext(ctx, &tape, grad_registry));
    TF_RETURN_IF_ERROR(
        forward_model(tape_ctx.get(), inputs, absl::MakeSpan(temp_outputs)));

  Tape tape(/*persistent=*/false);
  tape.Watch(inputs[0]);  // Watch x.
  AbstractTensorHandle* exp_output;
  TF_RETURN_IF_ERROR(ops::Exp(ctx, inputs[0], &exp_output, "Exp"));
  std::unique_ptr<GradientFunction> gradient_function(
      new PassThroughGradientFunction);
  tape.RecordOperation(inputs, {exp_output}, gradient_function.release());
  TF_RETURN_IF_ERROR(tape.ComputeGradient(ctx,

    deps = [
        "//tensorflow/c/eager:gradient_checker",
        "//tensorflow/c/eager:gradients_internal",
        "//tensorflow/c/eager:unified_api_testutil",
        "//tensorflow/c/experimental/gradients/tape:tape_context",
        "//tensorflow/core:test",
        "//tensorflow/core:test_main",
    ],
)

tf_cuda_cc_test(
    name = "nn_grad_test",
    size = "small",
    srcs = [

#include "tensorflow/c/eager/c_api_unified_experimental_internal.h"
#include "tensorflow/c/eager/unified_api_testutil.h"
#include "tensorflow/c/experimental/gradients/grad_test_helper.h"
#include "tensorflow/c/experimental/gradients/tape/tape_context.h"
#include "tensorflow/c/experimental/ops/math_ops.h"
#include "tensorflow/c/tf_status_helper.h"
#include "tensorflow/core/platform/tensor_float_32_utils.h"
#include "tensorflow/core/platform/test.h"

      ParallelTensor::FromTensorHandles(parallel_device,
                                        std::move(mixed_handles), status.get());
  ASSERT_TRUE(TF_GetCode(status.get()) == TF_OK);
  // Can't take the shape of a parallel tensor with varying numbers of axes, but
  // running operations on them is OK.
  TF_ASSERT_OK(unknown_length_vector->Shape(&shape));
  EXPECT_THAT(*shape, ElementsAre(-1));

# only executabe on the RBE (x86) machine, so the strategy_regexp options are
# added to override and run the actions using remote strategy.
build:rbe_cross_compile_base_xla --strategy_regexp='Generating code from table.*=remote'
build:rbe_cross_compile_base_xla --strategy_regexp='Generating flatbuffer files.*=remote'
build:rbe_cross_compile_base_xla --strategy_regexp='Executing genrule @llvm-project.*=remote'

  std::vector<Tensor> all_input_tensors;
  // Update the vector with information from `input_tensors` if provided.
  if (input_tensors != nullptr) {
    // Note that we take the address of the elements in `all_input_tensors`
    // below. Allocate enough space so that no reallocation happens, which will
    // make the pointers invalid.
    all_input_tensors.reserve(num_inputs);

             'compilation to reduce the compilation time?'),
      'Eigen strong inline overridden.', 'Not overriding eigen strong inline, '
      'some compilations could take more than 20 mins.'):
    # Due to a known MSVC compiler issue
    # https://github.com/tensorflow/tensorflow/issues/10521
    # Overriding eigen strong inline speeds up the compiling of