// Recover data format from forward pass for gradient.
    std::string data_format;
    TF_RETURN_IF_ERROR(forward_attrs_.Get("data_format", &data_format));

    // Grad for A
    grad_inputs[0] = upstream_grad;
    grad_inputs[0]->Ref();

    // Grad for bias
    std::string name = "bias_add_grad";
    TF_RETURN_IF_ERROR(BiasAddGrad(ctx, upstream_grad, &grad_inputs[1],

using tensorflow::string;
using tensorflow::Tensor;
using tensorflow::TensorId;
using tensorflow::TensorShapeProto;
using tensorflow::VersionDef;
using tensorflow::errors::FailedPrecondition;
using tensorflow::errors::InvalidArgument;
using tensorflow::errors::OutOfRange;
using tensorflow::gtl::ArraySlice;
using tensorflow::strings::StrCat;

extern "C" {

      ASSERT_NEAR(dnumerical[j], danalytical[j], abs_error);
    }
    TF_DeleteTensor(analytical_tensor);
    TF_DeleteTensor(numerical_tensor);
    delete[] danalytical;
    delete[] dnumerical;
    outputs[i]->Unref();
  }
}

void CheckTensorValue(AbstractTensorHandle* t, absl::Span<const float> manuals,
                      absl::Span<const int64_t> dims, double abs_error) {
  TF_Tensor* analytical_tensor;

  tape.Watch(inputs[0]);  // Watch x.
  AbstractTensorHandle* exp_output;
  TF_RETURN_IF_ERROR(ops::Exp(ctx, inputs[0], &exp_output, "Exp"));
  std::unique_ptr<GradientFunction> gradient_function(
      new PassThroughGradientFunction);
  tape.RecordOperation(inputs, {exp_output}, gradient_function.release());
  TF_RETURN_IF_ERROR(tape.ComputeGradient(ctx,
                                          /*targets*/ {exp_output},

static void RenameFile(const TF_Filesystem* filesystem, const char* src,
                       const char* dst, TF_Status* status) {
  // If target is a directory return TF_FAILED_PRECONDITION.
  // Target might be missing, so don't error in that case.
  struct stat st;
  if (stat(dst, &st) != 0) {
    if (errno != ENOENT) {
      TF_SetStatusFromIOError(status, errno, dst);
      return;
    }
  } else if (S_ISDIR(st.st_mode)) {

  if (!tensorflow::protobuf::TextFormat::ParseFromString(text_proto, &gdef)) {
    status->status = tensorflow::errors::Internal(
        "Invalid text proto for GraphDef: ", text_proto);
    return {};
  }
  const auto& fdef_lib = gdef.library();
  if (fdef_lib.gradient_size() > 0) {
    status->status = tensorflow::errors::Internal(
        "GradientDef is not supported in reading Dataset related functions: ",
        text_proto);

    status_ = StatusFromTF_Status(status.get());
    ASSERT_EQ(errors::OK, status_.code()) << status_.message();

    {
      AbstractContext* ctx_raw = nullptr;
      status_ =
          BuildImmediateExecutionContext(std::get<1>(GetParam()), &ctx_raw);
      ASSERT_EQ(errors::OK, status_.code()) << status_.message();
      immediate_execution_ctx_.reset(ctx_raw);
    }

    status_ = StatusFromTF_Status(status.get());
    ASSERT_EQ(errors::OK, status_.code()) << status_.message();

    {
      AbstractContext* ctx_raw = nullptr;
      status_ =
          BuildImmediateExecutionContext(std::get<1>(GetParam()), &ctx_raw);
      ASSERT_EQ(errors::OK, status_.code()) << status_.message();
      immediate_execution_ctx_.reset(ctx_raw);
    }

    Node* node = inputs[i].oper ? &inputs[i].oper->node : nullptr;
    int idx = inputs[i].index;

    TF_RETURN_WITH_CONTEXT_IF_ERROR(
        fn_body->graph.IsValidOutputTensor(node, idx),
        "Encountered while processing input ", i, " into function '", fn_name,
        "'");
    TF_RETURN_WITH_CONTEXT_IF_ERROR(ValidateNonRefOutput(node, idx),
                                    "Encountered while processing input ", i,

  EXPECT_EQ(TF_ALREADY_EXISTS,
            static_cast<TF_Code>(tensorflow::error::ALREADY_EXISTS));
  EXPECT_EQ(TF_PERMISSION_DENIED,
            static_cast<TF_Code>(tensorflow::error::PERMISSION_DENIED));
  EXPECT_EQ(TF_UNAUTHENTICATED,
            static_cast<TF_Code>(tensorflow::error::UNAUTHENTICATED));
  EXPECT_EQ(TF_RESOURCE_EXHAUSTED,