TF_CAPI_EXPORT extern void TFE_MonitoringStringGaugeCellSet(
    TFE_MonitoringStringGaugeCell* cell, const char* value);
// Retrieves the string value and saves it in the buffer.
TF_CAPI_EXPORT extern const void TFE_MonitoringStringGaugeCellValue(
    TFE_MonitoringStringGaugeCell* cell, TF_Buffer* buf);

// APIs for String Gauge without label.
typedef struct TFE_MonitoringStringGauge0 TFE_MonitoringStringGauge0;

  *result = outputs[0];
  return absl::OkStatus();
}
}  // namespace

Status GradientRegistry::Register(
    const string& op_name, GradientFunctionFactory gradient_function_factory) {
  auto iter = registry_.find(op_name);
  if (iter != registry_.end()) {
    const string error_msg = "Gradient already exists for op: " + op_name + ".";
    return errors::AlreadyExists(error_msg);
  }

limitations under the License.
==============================================================================*/

#include "tensorflow/c/eager/dlpack.h"

#include <string>

#include "include/dlpack/dlpack.h"  // from @dlpack
#include "tensorflow/c/eager/c_api.h"
#include "tensorflow/c/eager/c_api_experimental.h"
#include "tensorflow/c/eager/tfe_tensorhandle_internal.h"

                       subdiv_offsets.size());

  return op;
}

TFE_Op* SendOp(TFE_Context* ctx, TFE_TensorHandle* in,
               const std::string& op_name, const std::string& send_device,
               const std::string& recv_device,
               tensorflow::uint64 send_device_incarnation) {
  TF_Status* status = TF_NewStatus();
  TFE_Op* op = TFE_NewOp(ctx, op_name.c_str(), status);

#include "tensorflow/c/experimental/ops/math_ops.h"
#include "tensorflow/c/experimental/ops/nn_ops.h"

using std::vector;
using tensorflow::ops::AddV2;
using tensorflow::ops::Div;
using tensorflow::ops::DivNoNan;
using tensorflow::ops::MatMul;
using tensorflow::ops::Mul;
using tensorflow::ops::Neg;
using tensorflow::ops::OnesLike;
using tensorflow::ops::SqrtGrad;

namespace tensorflow {
namespace gradients {
namespace {

#include "tensorflow/core/lib/llvm_rtti/llvm_rtti.h"
#include "tensorflow/core/platform/errors.h"
#include "tensorflow/core/platform/strcat.h"
#include "tensorflow/core/platform/types.h"

using tensorflow::dyn_cast;
using tensorflow::string;
using tensorflow::gtl::ArraySlice;

namespace tensorflow {
namespace tracing {
namespace graph {

class GraphContext;
class GraphOperation;
class GraphTensor;

  status->status = status_or_value.status();
  if (!status_or_value.ok()) return;

  const std::string& value_string = status_or_value.value();
  void* data = tensorflow::port::Malloc(value_string.length());
  value_string.copy(static_cast<char*>(data), value_string.length(), 0);
  value_buf->data = data;
  value_buf->length = value_string.length();
  value_buf->data_deallocator = [](void* data, size_t length) {

#include "tensorflow/core/public/version.h"

#if !defined(IS_MOBILE_PLATFORM)
#include "tensorflow/core/common_runtime/eager/context_distributed_manager.h"
#endif  // !IS_MOBILE_PLATFORM

using tensorflow::string;

namespace {

string DeviceName(const tensorflow::Device* d) {
  return (d == nullptr) ? "cpu:0" : d->name();
}

// Annotate eager runtime construction context to the given `function_def` as
// an attribute.

// ignored (and hence can be None). The backprop algorithm can then leverage
// this by not constructing zeros to pass for those indices.
std::unordered_map<string, std::unordered_set<int>>*
FunctionsAcceptingNoneForIndicesMap() {
  static auto* const m =
      new std::unordered_map<string, std::unordered_set<int>>({
          {"SoftmaxCrossEntropyWithLogits", {1}},
          {"SparseSoftmaxCrossEntropyWithLogits", {1}},

                     TF_Status* status) {
  using tensorflow::NodeDef;
  using tensorflow::OpRegistrationData;
  using tensorflow::Tensor;
  using tensorflow::shape_inference::DimensionHandle;
  using tensorflow::shape_inference::InferenceContext;
  using tensorflow::shape_inference::ShapeAndType;
  using tensorflow::shape_inference::ShapeHandle;

  const int num_inputs = input_shapes->num_items;