pkg image, method (*Alpha16) SetRGBA64(int, int, color.RGBA64)
pkg image, method (*CMYK) RGBA64At(int, int) color.RGBA64
pkg image, method (*CMYK) SetRGBA64(int, int, color.RGBA64)
pkg image, method (*Gray) RGBA64At(int, int) color.RGBA64
pkg image, method (*Gray) SetRGBA64(int, int, color.RGBA64)
pkg image, method (*Gray16) RGBA64At(int, int) color.RGBA64
pkg image, method (*Gray16) SetRGBA64(int, int, color.RGBA64)

                                                        </c:when>
                                                        <c:otherwise>
                                                            <span class="badge bg-gray">${f:h(data.jobStatus)}</span>
                                                        </c:otherwise>
                                                    </c:choose></td>

pkg image, method (*Alpha) AlphaAt(int, int) color.Alpha
pkg image, method (*Alpha16) Alpha16At(int, int) color.Alpha16
pkg image, method (*Gray) GrayAt(int, int) color.Gray
pkg image, method (*Gray16) Gray16At(int, int) color.Gray16
pkg image, method (*NRGBA) NRGBAAt(int, int) color.NRGBA
pkg image, method (*NRGBA64) NRGBA64At(int, int) color.NRGBA64

                 absl::Span<AbstractTensorHandle* const> grad_outputs,
                 absl::Span<AbstractTensorHandle*> grad_inputs) override {
    // TODO(b/161805092): Support broadcasting.

    DCHECK(grad_outputs[0]);
    grad_inputs[0] = grad_outputs[0];
    grad_inputs[1] = grad_outputs[0];

    grad_inputs[0]->Ref();
    grad_inputs[1]->Ref();
    return absl::OkStatus();
  }

==============================================================================*/
#include "tensorflow/c/experimental/gradients/math_grad.h"

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

    int64_t tensor_id = source_tensor_ids[i];
    auto grad_it = gradients.find(tensor_id);
    if (grad_it == gradients.end()) {
      result[i] = nullptr;
    } else {
      if (grad_it->second.size() > 1) {
        Gradient* grad = vspace.AggregateGradients(grad_it->second);
        grad_it->second.clear();
        grad_it->second.push_back(grad);
      }
      result[i] = grad_it->second[0];
      used_gradient_ids.insert(tensor_id);

  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());

// TODO: Not implemented
func (m *muxClient) RequestStateless(h HandlerID, req []byte, out chan<- Response) {
	if m.init {
		out <- Response{Err: errors.New("mux client already used")}
	}
	m.init = true

	// Try to grab an initial block.
	m.singleResp = false
	msg := message{
		Op:         OpConnectMux,
		Handler:    h,
		Flags:      FlagEOF,
		Payload:    req,
		DeadlineMS: uint32(m.deadline.Milliseconds()),
	}

      } finally {
        monitor.leave();
      }
    } else {
      // It is possible due to races that we are currently in the expected state even though we
      // timed out. e.g. if we weren't event able to grab the lock within the timeout we would never
      // even check the guard. I don't think we care too much about this use case but it could lead
      // to a confusing error message.

	dashboardCmd.AddCommand(prom)

	graf := grafanaDashCmd(cliContext)
	graf.PersistentFlags().IntVar(&grafanaPort, "ui-port", defaultGrafanaPort, "The component dashboard UI port.")
	dashboardCmd.AddCommand(graf)

	jaeger := jaegerDashCmd(cliContext)