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

  // Initialize a input_tensor vector with `nullptr` values.
  std::vector<const Tensor*> input_tensors_vector(num_inputs, nullptr);
  // A vector to keep track of newly created `tf::Tensor` objects.
  std::vector<Tensor> all_input_tensors;
  // Update the vector with information from `input_tensors` if provided.
  if (input_tensors != nullptr) {

  device->delete_device = &DeleteParallelDevice;
  device->execute = &ParallelDeviceExecute;
  std::vector<std::string> underlying_devices_vector;
  underlying_devices_vector.reserve(num_underlying_devices);
  for (int device_index = 0; device_index < num_underlying_devices;
       ++device_index) {
    underlying_devices_vector.push_back(underlying_devices[device_index]);
  }
  std::unique_ptr<ParallelDevice> parallel_device(

    // Input tensors
    const std::vector<std::pair<string, Tensor>>& input_pairs,
    // Output tensors
    const std::vector<string>& output_tensor_names, TF_Tensor** c_outputs,
    // Target nodes
    const std::vector<string>& target_oper_names, TF_Buffer* run_metadata,
    TF_Status* status) {
  const int noutputs = output_tensor_names.size();
  std::vector<Tensor> outputs(noutputs);
  Status result;

  // This cache has space for 4 blocks; we'll read from two files.
  const size_t n = 3;
  tf_gcs_filesystem::RamFileBlockCache cache(8, 32, 0, fetcher);
  std::vector<char> out;
  std::vector<char> a(n, 'a');
  std::vector<char> b(n, 'b');
  std::vector<char> A(n, 'A');
  std::vector<char> B(n, 'B');
  // Fill the cache.
  TF_EXPECT_OK(ReadCache(&cache, "a", 0, n, &out));
  EXPECT_EQ(out, a);
  EXPECT_EQ(calls, 1);

                           const string& op_name) {
  std::vector<int64_t> input_ids(inputs.size());
  std::vector<tensorflow::DataType> input_dtypes(inputs.size());
  for (int i = 0; i < inputs.size(); i++) {
    input_ids[i] = ToId(inputs[i]);
    input_dtypes[i] = inputs[i]->DataType();
  }
  std::vector<TapeTensor> tape_tensors;
  tape_tensors.reserve(outputs.size());
  for (auto t : outputs) {

  int num_elems = TF_TensorElementCount(theta_tensor);
  vector<float> theta_data(num_elems);
  memcpy(theta_data.data(), TF_TensorData(theta_tensor),
         TF_TensorByteSize(theta_tensor));

  // Initialize space for the numerical gradient.
  vector<float> dtheta_approx(num_elems);

  // Get theta shape and store in theta_dims.
  int num_dims = TF_NumDims(theta_tensor);
  vector<int64_t> theta_dims(num_dims);

  // Process inputs.
  std::vector<tensorflow::OutputTensor> input_tensors;
  std::unordered_map<const Node*, std::vector<int>> input_nodes;
  status->status = tensorflow::ProcessInputs(fn_body, fn_name, ninputs, inputs,
                                             &input_tensors, &input_nodes);
  if (TF_GetCode(status) != TF_OK) return nullptr;

  // Process outputs.
  std::vector<tensorflow::OutputTensor> output_tensors;

            if ( mechs != null ) {
                ASN1EncodableVector vector = new ASN1EncodableVector();
                for ( int i = 0; i < mechs.length; i++ ) {
                    vector.add(mechs[ i ]);
                }
                fields.add(new DERTaggedObject(true, 0, new DERSequence(vector)));
            }
            int ctxFlags = getContextFlags();
            if ( ctxFlags != 0 ) {

  std::vector<tensorflow::string> tensor_id_data;
};

struct TF_ImportGraphDefResults {
  std::vector<TF_Output> return_tensors;
  std::vector<TF_Operation*> return_nodes;
  std::vector<const char*> missing_unused_key_names;
  std::vector<int> missing_unused_key_indexes;

  // Backing memory for missing_unused_key_names values.
  std::vector<tensorflow::string> missing_unused_key_names_data;

#include "tensorflow/core/platform/status.h"

using tensorflow::string;

namespace {

std::vector<int64_t> TensorShapeAsVector(const tensorflow::TensorHandle& handle,
                                         tensorflow::Status* status) {
  std::vector<int64_t> shape;
  int rank = -1;
  *status = handle.NumDims(&rank);
  if (!status->ok()) {
    return shape;
  }
  shape.reserve(rank);