ASSERT_EQ(num_elem_numerical, num_grad);

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

  for (int j = 0; j < num_grad; j++) {
    ASSERT_NEAR(dnumerical[j], expected_grad[j], abs_error);
  }
  delete[] dnumerical;
  TF_DeleteTensor(numerical_tensor);
}

  std::unique_ptr<float[]> actual_values(new float[expected_values.size()]);
  EXPECT_EQ(sizeof(float) * expected_values.size(), TF_TensorByteSize(t));
  memcpy(actual_values.get(), TF_TensorData(t), TF_TensorByteSize(t));
  TF_DeleteTensor(t);

  for (int i = 0; i < expected_values.size(); i++) {
    EXPECT_EQ(expected_values[i], actual_values[i])

      ASSERT_EQ(sizeof(int32_t), TF_TensorByteSize(out));
      int32_t* output_contents = static_cast<int32_t*>(TF_TensorData(out));
      EXPECT_EQ(expected_results[i], *output_contents);
    }
  }

  void CompareInt32Tensor(const std::vector<int32_t>& expected, TF_Tensor* t) {
    int32_t* data = static_cast<int32_t*>(TF_TensorData(t));
    size_t size = TF_TensorByteSize(t);

// Allocate and return a new Tensor on the host.
//
// The caller must set the Tensor values by writing them to the pointer returned
// by TF_TensorData with length TF_TensorByteSize.
TF_CAPI_EXPORT extern TF_Tensor* TFE_AllocateHostTensor(TFE_Context* ctx,
                                                        TF_DataType dtype,