const char* name) {
  const auto& m = reader->GetVariableToDataTypeMap();
  return static_cast<TF_DataType>(m.at(name));
}

TF_Tensor* TF_CheckpointReaderGetTensor(TF_CheckpointReader* reader,
                                        const char* name, TF_Status* status) {
  std::unique_ptr<tensorflow::Tensor> tensor;
  reader->GetTensor(name, &tensor, status);

  if (fflush(posix_file->handle) != 0)
    TF_SetStatusFromIOError(status, errno, posix_file->filename);
}

static void Sync(const TF_WritableFile* file, TF_Status* status) {
  // For historical reasons, this does the same as `Flush` at the moment.
  // TODO(b/144055243): This should use `fsync`/`sync`.
  Flush(file, status);
}

static void Close(const TF_WritableFile* file, TF_Status* status) {

  return op_->SetAttrInt(attr_name, value);
}
Status SetAttrFloat(AbstractOperation* op_, const char* attr_name, float value,
                    ForwardOperation* forward_op_) {
  forward_op_->attrs.Set(attr_name, value);
  return op_->SetAttrFloat(attr_name, value);
}
Status SetAttrBool(AbstractOperation* op_, const char* attr_name, bool value,

  tensorflow::DeviceNameUtils::ParseFullName(device_name, &parsed_name);
  std::string device_type = parsed_name.type;
  int device_id = 0;
  if (parsed_name.has_id) {
    device_id = parsed_name.id;
  }

  ctx.device_id = device_id;
  if (device_type == "CPU") {
    ctx.device_type = DLDeviceType::kDLCPU;
  } else if (device_type == "GPU") {
#if TENSORFLOW_USE_ROCM

    control_output_nodes.push_back(&control_outputs[i]->node);
  }

  // Do the actual function creation.
  DCHECK(append_hash_to_fn_name <= 1);
  tensorflow::FunctionDef fdef;
  status->status = tensorflow::GraphToFunctionDef(
      fn_body->graph, fn_name, append_hash_to_fn_name != 0,
      /*set_stateful_from_nodes=*/true,
      /*copy_placeholder_attrs_from_nodes=*/true, body_nodes, input_tensors,

  const char task0_name[] = "/job:localhost/replica:0/task:0/device:CPU:0";
  const char task1_name[] = "/job:localhost/replica:0/task:1/device:CPU:0";
  const char task2_name[] = "/job:localhost/replica:0/task:2/device:CPU:0";

  // Create one variable per task.
  TFE_TensorHandle* h0 = TestVariable(ctx, 1.0, task1_name);
  TFE_TensorHandle* h1 = TestVariable(ctx, 2.0, task2_name);

        Execute(context_, operation_name_, std::move(op_inputs_), attributes_,
                expected_max_outputs_, &op_outputs_, status_.get());
        execution_state_ = ExecutionState::kHasResult;
      }
    }
    finished_execute_.notify_one();
  }
}

void DeviceThread::StartExecute(TFE_Context* context,
                                const char* operation_name,

  }
}

void TFE_OpSetAttrInt(TFE_Op* op, const char* attr_name, int64_t value) {
  auto s = tensorflow::unwrap(op)->SetAttrInt(attr_name, value);
  if (!s.ok()) {
    LOG(WARNING) << "Unable to set attribute: " << attr_name;
  }
}

void TFE_OpSetAttrFloat(TFE_Op* op, const char* attr_name, float value) {
  auto s = tensorflow::unwrap(op)->SetAttrFloat(attr_name, value);
  if (!s.ok()) {

    tensorflow::NameAttrList func_name;
    func_name.set_name(string(value, value + length));
    op_->node_builder.Attr(attr_name, func_name);
    return absl::OkStatus();
  }
  Status SetAttrTensor(const char* attr_name,
                       AbstractTensorInterface* tensor) override {
    return tensorflow::errors::Unimplemented(
        "SetAttrTensor has not been implemented yet.");
  }

    std::string name = "Conj_Log1p_Grad_X";
    TF_RETURN_IF_ERROR(SafeConj(ctx, X, &temp_output, name.c_str()));

    AbstractTensorHandlePtr Conj_X(temp_output);

    // Creates Ones
    name = "OnesLike_Log1p_Grad_X";
    TF_RETURN_IF_ERROR(OnesLike(ctx, Conj_X.get(), &temp_output, name.c_str()));

    AbstractTensorHandlePtr Ones_X(temp_output);

    name = "Add_Log1p_Grad_X";