CHECK_EQ(TF_OK, TF_GetCode(status)) << TF_Message(status);
  TFE_OpAddInput(op, axis, status);
  CHECK_EQ(TF_OK, TF_GetCode(status)) << TF_Message(status);
  TFE_OpSetAttrBool(op, "keep_dims", 1);
  TFE_OpSetAttrType(op, "Tidx", TF_INT32);
  TF_DeleteStatus(status);
  TFE_OpSetAttrType(op, "T", TFE_TensorHandleDataType(input));

  return op;
}

TFE_Op* AllReduceOp(TFE_Context* ctx, TFE_TensorHandle* in, int group_size) {

    input_tensors->emplace_back(node, idx);

    const auto& iter = input_nodes->find(node);
    if (iter == input_nodes->end()) {
      input_nodes->insert({node, {idx}});
    } else {
      auto& indices = iter->second;
      if (std::find(indices.begin(), indices.end(), idx) != indices.end()) {
        return InvalidArgument("TF_Output ", node->name(), ":", idx,

  tensorflow::AttrValueMap::const_iterator attr_found = attr_values.find("T");
  EXPECT_NE(attr_found, attr_values.cend());
  EXPECT_EQ(attr_found->second.type(), tensorflow::DataType::DT_FLOAT);
  attr_found = attr_values.find("Tidx");
  EXPECT_NE(attr_found, attr_values.cend());
  EXPECT_EQ(attr_found->second.type(), tensorflow::DataType::DT_INT32);

  TFE_TensorHandle* retvals[1] = {nullptr};
  int num_retvals = 1;

  float results[2];
  for (int idx = 0; idx < 2; ++idx) {
    TF_AbstractTensor* result = TF_OutputListGet(func_outputs, idx);
    TFE_TensorHandle* handle = TF_AbstractTensorGetEagerTensor(result, s);
    ASSERT_EQ(TF_OK, TF_GetCode(s)) << TF_Message(s);
    TF_Tensor* f_t = TFE_TensorHandleResolve(handle, s);
    ASSERT_EQ(TF_OK, TF_GetCode(s)) << TF_Message(s);
    results[idx] = *static_cast<float*>(TF_TensorData(f_t));