//  noutputs - number of elements in `outputs` array
//  outputs - array of TF_Outputs that specify the outputs of the function.
//            If `noutputs` is zero (the function returns no outputs), `outputs`
//            can be null. `outputs` can contain the same tensor more than once.
//  output_names - The names of the function's outputs. `output_names` array

    TF_RETURN_IF_ERROR(
        RunAndMaybeSum(ctx, forward, theta_inputs, f_outputs, use_function));
    AbstractTensorHandlePtr fPlus(f_outputs[0]);

    // Get f(theta - eps):
    theta_inputs[input_index] = thetaMinus.get();
    TF_RETURN_IF_ERROR(
        RunAndMaybeSum(ctx, forward, theta_inputs, f_outputs, use_function));
    AbstractTensorHandlePtr fMinus(f_outputs[0]);

  output_tensors->reserve(noutputs);
  for (int i = 0; i < noutputs; ++i) {
    Node* node = outputs[i].oper ? &outputs[i].oper->node : nullptr;
    int idx = outputs[i].index;
    TF_RETURN_WITH_CONTEXT_IF_ERROR(
        fn_body->graph.IsValidOutputTensor(node, idx),
        "Encountered while processing output ", i, " from function '", fn_name,
        "'");
    TF_RETURN_WITH_CONTEXT_IF_ERROR(ValidateNonRefOutput(node, idx),

    inputs_.push_back(t->output_);
    *output = tensorflow::down_cast<TracingTensorHandle*>(outputs[0]);
    return absl::OkStatus();
  }

  absl::Status Finalize(OutputList* outputs, AbstractFunction** f) override {
    std::vector<TF_Output> graph_outputs;
    graph_outputs.reserve(outputs->outputs.size());
    for (auto* abstract_output : outputs->outputs) {
      GraphTensor* output = dyn_cast<GraphTensor>(abstract_output);

  TF_OutputListSetNumOutputs(mm_outputs, 1, status.get());
  ASSERT_EQ(TF_OK, TF_GetCode(status.get())) << TF_Message(status.get());
  TF_ExecuteOperation(fn_op, 1, &input_t, mm_outputs, status.get());
  ASSERT_EQ(TF_OK, TF_GetCode(status.get())) << TF_Message(status.get());

  ASSERT_EQ(1, TF_OutputListNumOutputs(mm_outputs));
  TF_AbstractTensor* final_result = TF_OutputListGet(mm_outputs, 0);
  TFE_TensorHandle* final =

                // input related parameters
                /*inputs*/ nullptr, /*input_values*/ nullptr, /*ninputs*/ 0,
                // output related parameters
                /*outputs*/ &output, /*output_values*/ &ret,
                /*noutputs*/ 1,
                /*targets*/ nullptr, /*ntargets*/ 0,
                /*run_metadata*/ nullptr, status);
  if (VLOG_IS_ON(1) && status->status.ok()) {
    tensorflow::Tensor tensor;

        committer: TensorFlow Release Automation <******@****.***>
        token: ${{ secrets.JENKINS_TOKEN }}
        base: ${{ github.event.inputs.release_branch }}
        branch: ${{ github.event.inputs.release_branch }}-${{ steps.cherrypick.outputs.SHORTSHA }}
        reviewers: learning-to-play
        body: |

	if [[ $? -eq 0 ]]; then
		echo "Outputs are consistent."
		break
	fi

	remaining_attempts=$((max_wait_attempts - attempt))
	if ((attempt >= max_wait_attempts)); then
		echo "Outputs remain inconsistent after $max_wait_attempts attempts. Exiting with error."
		exit 1
	else
		echo "Outputs are inconsistent. Waiting for $wait_interval seconds (attempt $attempt/$max_wait_attempts)."

  int num_retvals = 1;
  std::vector<AbstractTensorHandle*> outputs(1);
  GradientRegistry registry;
  s = RegisterGradients(&registry);
  ASSERT_EQ(errors::OK, s.code()) << s.message();
  auto tape = std::make_unique<Tape>(/*persistent=*/false);
  s = Execute(check_numerics_op.get(), ctx.get(), absl::MakeSpan(outputs),
              &num_retvals, &forward_op, tape.get(), registry);

        this.id = id;
        this.inputs = inputs;
        this.outputs = outputs;

        if (id == 0) {
            // create
            newInputs = inputs;
            newOutputs = outputs;
        }
    }

    /**
     * Gets the new input words.
     * Returns a defensive copy to prevent external modification.
     *