TF_EXCLUSIVE_LOCKS_REQUIRED(fn_body->mu) {
  input_tensors->reserve(ninputs);
  for (int i = 0; i < ninputs; ++i) {
    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,
        "'");

void TF_AddInput(TF_OperationDescription* desc, TF_Output input) {
  desc->node_builder.Input(&input.oper->node, input.index);
}

void TF_AddInputList(TF_OperationDescription* desc, const TF_Output* inputs,
                     int num_inputs) {
  std::vector<NodeBuilder::NodeOut> input_list;
  input_list.reserve(num_inputs);
  for (int i = 0; i < num_inputs; ++i) {
    input_list.emplace_back(&inputs[i].oper->node, inputs[i].index);
  }

  // The first one will the function's output
  std::vector<TF_Output> outputs;

  // Add while loop to func_graph_
  {
    // The inputs to the while loop
    std::vector<TF_Output> inputs = {{feed1, 0}, {feed2, 0}};
    std::unique_ptr<TF_WhileParams> params(new TF_WhileParams(
        TF_NewWhile(func_graph_, &inputs[0], inputs.size(), s_)));

  EXPECT_EQ(TF_OK, TF_GetCode(s)) << TF_Message(s);
  TF_DeleteStatus(s);
}

void CSession::SetInputs(
    std::vector<std::pair<TF_Operation*, TF_Tensor*>> inputs) {
  DeleteInputValues();
  inputs_.clear();
  for (const auto& p : inputs) {
    inputs_.emplace_back(TF_Output{p.first, 0});
    input_values_.emplace_back(p.second);
  }
}

void CSession::SetOutputs(std::initializer_list<TF_Operation*> outputs) {

  ASSERT_EQ(TF_OK, TF_GetCode(status.get())) << TF_Message(status.get());

  // Build inputs and outputs.
  TF_AbstractTensor* inputs[2] = {at, at};
  TF_OutputList* o = TF_NewOutputList();
  TF_OutputListSetNumOutputs(o, 1, status.get());
  ASSERT_EQ(TF_OK, TF_GetCode(status.get())) << TF_Message(status.get());

  // Execute.
  TF_ExecuteOperation(op, 2, inputs, o, status.get());

    for (int i = 0; i < inputs.size(); i++) {
      GraphTensor* t = dyn_cast<GraphTensor>(inputs[i]);
      if (!t) {
        return tensorflow::errors::InvalidArgument(
            "Unable to cast input to GraphTensor");
      }
      tf_outputs[i] = t->output_;
    }
    TF_AddInputList(op_.get(), tf_outputs.data(), tf_outputs.size());
    return absl::OkStatus();
  }

TF_Operation* RandomUniform(TF_Operation* shape, TF_DataType dtype,
                            TF_Graph* graph, TF_Status* s);

// Split `input` along the first dimension into 3 tensors
TF_Operation* Split3(TF_Operation* input, TF_Graph* graph, TF_Status* s,
                     const char* name = "split3");

bool IsPlaceholder(const tensorflow::NodeDef& node_def);

// checkPolicyCond returns a boolean to indicate if a condition is satisfied according
// to the passed operator
func checkPolicyCond(op string, input1, input2 string) bool {
	switch op {
	case policyCondEqual:
		return input1 == input2
	case policyCondStartsWith:
		return strings.HasPrefix(input1, input2)
	}
	return false
}

      final <T extends @Nullable Object, A extends @Nullable Object, R extends @Nullable Object>
          A result(Collector<T, A, R> collector, Iterable<T> inputs) {
        A accum = collector.supplier().get();
        for (T input : inputs) {
          collector.accumulator().accept(accum, input);
        }
        return accum;
      }
    },
    /** Get one accumulator for each element and merge the accumulators left-to-right. */

          A result(Collector<T, A, R> collector, Iterable<T> inputs) {
        List<A> stack = new ArrayList<>();
        for (T input : inputs) {
          A newAccum = collector.supplier().get();
          collector.accumulator().accept(newAccum, input);
          push(stack, newAccum);
        }
        push(stack, collector.supplier().get());