TF_RETURN_IF_ERROR(GetValue(theta, &theta_tensor));

  // Get number of elements and fill data.
  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);

  EXPECT_EQ(TF_FLOAT, TFE_TensorHandleDataType(h));

  TF_Tensor* t = TFE_TensorHandleResolve(h, status.get());
  ASSERT_EQ(16, TF_TensorByteSize(t));
  float data[4] = {0};
  memcpy(&data[0], TF_TensorData(t), TF_TensorByteSize(t));
  EXPECT_EQ(1.0, data[0]);
  EXPECT_EQ(2.0, data[1]);
  EXPECT_EQ(3.0, data[2]);
  EXPECT_EQ(4.0, data[3]);
  TF_DeleteTensor(t);
  TFE_DeleteTensorHandle(h);

    ASSERT_EQ(TF_OK, TF_GetCode(s)) << TF_Message(s);

    memcpy(&result_data[0], TF_TensorData(f_t), TF_TensorByteSize(f_t));

    // Verify results for each output
    for (int j = 0; j < 4; j++) {
      ASSERT_EQ(result_data[j], expected_outputs[idx][j]);
    }

    TF_DeleteTensor(f_t);
  }

  // Free memory associated with add and MatMul outputs
  for (int idx = 0; idx < 3; ++idx) {

    ASSERT_EQ(num_elem_numerical, num_elem_analytical);

    float* dnumerical = new float[num_elem_numerical]{0};
    memcpy(&dnumerical[0], TF_TensorData(numerical_tensor),
           TF_TensorByteSize(numerical_tensor));
    float* danalytical = new float[num_elem_analytical]{0};
    memcpy(&danalytical[0], TF_TensorData(analytical_tensor),
           TF_TensorByteSize(analytical_tensor));

      // The block extends past the end of the slice we're reading.
      end -= (pos + data.size()) - (offset + n);
    }
    if (begin < end) {
      size_t bytes_to_copy = end - begin;
      memcpy(&buffer[total_bytes_transferred], &*begin, bytes_to_copy);
      total_bytes_transferred += bytes_to_copy;
    }
    if (data.size() < block_size_) {

  ASSERT_EQ(TF_GetCode(status), TF_OK) << TF_Message(status);
  TFE_DeleteTensorHandle(retvals[0]);
  float sum = 0;
  EXPECT_EQ(sizeof(sum), TF_TensorByteSize(t));
  memcpy(&sum, TF_TensorData(t), TF_TensorByteSize(t));
  TF_DeleteTensor(t);
  EXPECT_EQ(sum, 6.0);

  TFE_DeleteTensorHandle(h0);
  TFE_DeleteTensorHandle(h1);
  TFE_DeleteTensorHandle(h2);

  TFE_DeleteExecutor(old_executor);
  TFE_DeleteContext(ctx);
  ASSERT_EQ(TF_OK, TF_GetCode(status)) << TF_Message(status);
  float product[4] = {0};
  EXPECT_EQ(sizeof(product), TF_TensorByteSize(t));
  memcpy(&product[0], TF_TensorData(t), TF_TensorByteSize(t));
  TF_DeleteTensor(t);
  EXPECT_EQ(7, product[0]);
  EXPECT_EQ(10, product[1]);
  EXPECT_EQ(15, product[2]);
  EXPECT_EQ(22, product[3]);

  status->status = s->session->PRunSetup(input_names, output_names,
                                         target_oper_names, &new_handle);
  if (status->status.ok()) {
    char* buf = new char[new_handle.size() + 1];
    memcpy(buf, new_handle.c_str(), new_handle.size() + 1);
    *handle = buf;
  }
}

void TF_PRun(TF_DeprecatedSession* s, const char* handle,
             // Input tensors

TFE_TensorHandle* TestScalarTensorHandle(TFE_Context* ctx, float value) {
  float data[] = {value};
  TF_Status* status = TF_NewStatus();
  TF_Tensor* t = TFE_AllocateHostTensor(ctx, TF_FLOAT, nullptr, 0, status);
  memcpy(TF_TensorData(t), &data[0], TF_TensorByteSize(t));
  TFE_TensorHandle* th = TFE_NewTensorHandleFromTensor(ctx, t, status);
  CHECK_EQ(TF_OK, TF_GetCode(status)) << TF_Message(status);
  TF_DeleteTensor(t);

                                           const int64_t* dims, int num_dims) {
  TF_Status* status = TF_NewStatus();
  TF_Tensor* t = TFE_AllocateHostTensor(ctx, datatype, dims, num_dims, status);
  memcpy(TF_TensorData(t), data, TF_TensorByteSize(t));
  TFE_TensorHandle* th = TFE_NewTensorHandleFromTensor(ctx, t, status);
  CHECK_EQ(TF_OK, TF_GetCode(status)) << TF_Message(status);
  TF_DeleteTensor(t);
  TF_DeleteStatus(status);