namespace tensorflow {
namespace gradients {
// Ignores `grad_outputs` and sets all entries in grad_inputs to nullptr.
class NotDifferentiableGradientFunction : public GradientFunction {
  Status Compute(AbstractContext* ctx,
                 absl::Span<AbstractTensorHandle* const> grad_outputs,
                 absl::Span<AbstractTensorHandle*> grad_inputs) override;
};

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 OkStatus();
}

Status RegisterNotDifferentiable(GradientRegistry* registry, const string& op) {

    float* expected_grad, int num_grad, bool use_function,
    double abs_error = 1e-2) {
  Status s;
  AbstractTensorHandlePtr numerical_grad;
  {
    AbstractTensorHandle* numerical_grad_raw;
    s = CalcNumericalGrad(ctx, model, inputs, input_index, use_function,
                          &numerical_grad_raw);
    ASSERT_EQ(errors::OK, s.code()) << s.message();

namespace tensorflow {
namespace gradients {

/* Returns numerical grad inside `dtheta_approx` given `forward` model and
 * parameter specified by `input_index`.
 *
 * I.e. if y = <output of the forward model> and w = inputs[input_index],
 * this will calculate dy/dw numerically.
 *
 * `use_function` indicates whether to use graph mode(true) or eager(false).
 *
 * `numerical_grad` is the pointer to the AbstractTensorHandle* which will

  if (TF_GetCode(status) != TF_OK) return;
  if (!grad) return;

  status->status = g->graph.AddFunctionDef(grad->record->fdef(),
                                           grad->record->stack_traces());
  if (TF_GetCode(status) != TF_OK) return;

  tensorflow::GradientDef gdef;
  gdef.set_function_name(func->record->fdef().signature().name());
  gdef.set_gradient_func(grad->record->fdef().signature().name());

#include "tensorflow/c/eager/gradients_internal.h"
#include "tensorflow/c/eager/unified_api_testutil.h"
#include "tensorflow/c/experimental/gradients/array_grad.h"
#include "tensorflow/c/experimental/gradients/math_grad.h"
#include "tensorflow/c/experimental/gradients/not_differentiable.h"
#include "tensorflow/c/experimental/gradients/tape/tape_context.h"
#include "tensorflow/c/experimental/ops/array_ops.h"

// Adds a copy of function `func` and optionally its gradient function `grad`
// to `g`. Once `func`/`grad` is added to `g`, it can be called by creating
// an operation using the function's name.
// Any changes to `func`/`grad` (including deleting it) done after this method
// returns, won't affect the copy of `func`/`grad` in `g`.
// If `func` or `grad` are already in `g`, TF_GraphCopyFunction has no

// graph_def.library().gradient()
std::vector<std::pair<string, string>> GetGradDefs(
    const tensorflow::GraphDef& graph_def);

// Returns a sorted vector of names contained in `grad_def`
std::vector<string> GetFuncNames(const tensorflow::GraphDef& graph_def);

class CSession {
 public:
  CSession(TF_Graph* graph, TF_Status* s, bool use_XLA = false);
  explicit CSession(TF_Session* session);

[**sandbox**](https://developers.google.com/code-sandboxing). Memory corruptions
in TensorFlow ops can be recognized as security issues only if they are
reachable and exploitable through production-grade, benign models.

### Compilation

Compiling models via the recommended entry points described in
[XLA](https://www.tensorflow.org/xla) and
[JAX](https://jax.readthedocs.io/en/latest/jax-101/02-jitting.html)

the project's [log viewer](https://cloud.google.com/logging/docs/view/overview) and/or the `gcloud logging read`
capability.

The following example grabs the last 10 `kubelet` logs containing the string "cmdAdd" in the log message.

```console