}
	globalCtx, globalSpan := Start(ctx, rootSpan)
	return globalCtx, func() {
		globalSpan.End()
		shutdown()
	}, nil
}

func Start(ctx context.Context, span string) (context.Context, traceapi.Span) {
	return tracer().Start(ctx, span)

  virtual Status SetDeviceName(const char* name) = 0;

  virtual Status AddInput(AbstractTensorHandle* input) = 0;
  virtual Status AddInputList(
      absl::Span<AbstractTensorHandle* const> inputs) = 0;
  virtual Status Execute(absl::Span<AbstractTensorHandle*> retvals,
                         int* num_retvals) = 0;

  virtual Status SetAttrString(const char* attr_name, const char* data,

<div class="badge-wrapper">
    <a class="badge" href="https://dpeuniversity.gradle.com/app/courses/012de84f-fcd3-45d4-9c4c-284382eb3f3f/" target="_blank">
        <span class="badge-type button--blue">LEARN</span>
        <span class="badge-text">Introduction to Gradle for Developers&nbsp;&nbsp;&nbsp;&gt;</span>
    </a>
</div>
++++

Everyone has to start somewhere, and if you're new to Gradle, this is where to begin.

To get started using Gradle:

apiVersion: release-notes/v2
kind: feature
area: telemetry
issue:
  - 48336
releaseNotes:
  - |

==============================================================================*/

#include "tensorflow/c/experimental/saved_model/core/revived_types/flat_tensor_function.h"

#include <memory>
#include <string>

#include "absl/types/span.h"
#include "tensorflow/c/eager/abstract_tensor_handle.h"
#include "tensorflow/c/eager/immediate_execution_operation.h"
#include "tensorflow/c/eager/immediate_execution_tensor_handle.h"

#include "tensorflow/c/experimental/gradients/not_differentiable.h"

namespace tensorflow {
namespace gradients {
Status NotDifferentiableGradientFunction::Compute(
    AbstractContext* ctx, absl::Span<AbstractTensorHandle* const> grad_outputs,
    absl::Span<AbstractTensorHandle*> grad_inputs) {
  for (int i = 0; i < grad_inputs.size(); i++) {
    grad_inputs[i] = nullptr;
  }
  return absl::OkStatus();
}

class NotDifferentiableGradientFunction : public GradientFunction {
  Status Compute(AbstractContext* ctx,
                 absl::Span<AbstractTensorHandle* const> grad_outputs,
                 absl::Span<AbstractTensorHandle*> grad_inputs) override;
};
// Shorthand for registry->Register(op, new NotDifferentiableGradientFunction)

																	<span class="badge bg-primary"><la:message
                                                                            key="labels.enabled"/></span>
                                                    </c:if> <c:if test="${data.available=='false'}">
																	<span class="badge bg-gray"><la:message

  if (attr_.full_type() == "list(string)") {
    return "absl::Span<string const>";
  }

  // Normal path: translate base type to C++ ...
  string c_base_type = attr_.base_type();
  if (attr_.base_type() == "type") {
    c_base_type = "DataType";
  } else if (attr_.base_type() == "shape") {
    c_base_type = "const PartialTensorShape";
  }

  // ... and wrap in a Span<> if it's a list.
  if (attr_.is_list()) {

  }
};

class PassThroughGradientFunction : public GradientFunction {
 public:
  Status Compute(AbstractContext* ctx,
                 absl::Span<AbstractTensorHandle* const> grad_outputs,
                 absl::Span<AbstractTensorHandle*> grad_inputs) override {
    CHECK_EQ(grad_outputs.size(), 1);
    CHECK_EQ(grad_inputs.size(), 1);
    grad_inputs[0] = grad_outputs[0];
    if (grad_inputs[0]) {