if (model_outputs[i]) {
          output_list.outputs.emplace_back(model_outputs[i]);
          output_list.expected_num_outputs += 1;
        } else {
          null_indices.insert(i);
        }
      }
      TF_RETURN_IF_ERROR(dyn_cast<tracing::TracingContext>(func_ctx.get())
                             ->Finalize(&output_list, &func));
      scoped_func.reset(func);

                           TF_DataType value) {
  auto iter = builder->attr_names.insert(attr_name).first;
  builder->Set(*iter, static_cast<tensorflow::DataType>(value));
}

void TF_AttrBuilderSetTypeList(TF_AttrBuilder* builder, const char* attr_name,
                               const TF_DataType* values, int num_values) {
  auto iter = builder->attr_names.insert(attr_name).first;

        missing_it->second--;
        VLOG(1) << "Op " << op_id << " missing " << missing_it->second
                << " output gradients";
        if (missing_it->second == 0) {
          op_stack.insert(op_stack.begin(), op_id);
        }
      }
    }
  }
  if (!state.op_tape.empty()) {
    return tensorflow::errors::Internal("Invalid tape state.");
  }
  if (result.size() != source_tensor_ids.size()) {

  // Create a simple graph.
  Placeholder(graph, s);
  ASSERT_EQ(TF_OK, TF_GetCode(s)) << TF_Message(s);
  ASSERT_TRUE(TF_GraphOperationByName(graph, "feed") != nullptr);
  TF_Operation* oper = ScalarConst(3, graph, s);
  ASSERT_EQ(TF_OK, TF_GetCode(s)) << TF_Message(s);
  ASSERT_TRUE(TF_GraphOperationByName(graph, "scalar") != nullptr);
  Neg(oper, graph, s);
  ASSERT_EQ(TF_OK, TF_GetCode(s)) << TF_Message(s);

    ASSERT_EQ(errors::OK, status_.code()) << status_.message();
    Y.reset(Y_raw);
  }

  std::vector<AbstractTensorHandle*> outputs(1);
  status_ = RunModel(ReluGradModel, immediate_execution_ctx_.get(), {Y.get()},
                     absl::MakeSpan(outputs), UseFunction());
  ASSERT_EQ(errors::OK, status_.code()) << status_.message();

      return;
    }
    desc->colocation_constraints.clear();
    for (const string& location : attr_value.list().s()) {
      desc->colocation_constraints.insert(location);
    }
  } else {
    desc->node_builder.Attr(attr_name, std::move(attr_value));
  }

  status->status = absl::OkStatus();
}

                                    " into function '", fn_name, "'");

    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,

## Thanks to our Contributors

This release contains contributions from many people at Google, as well as:

<INSERT>, <NAME>, <HERE>, <USING>, <GITHUB>, <HANDLE>

# Release 2.16.0

## TensorFlow

<INSERT SMALL BLURB ABOUT RELEASE FOCUS AREA AND POTENTIAL TOOLCHAIN CHANGES>

*   TensorFlow Windows Build:

    # Check whether all capabilities from the input is valid
    all_valid = True
    # Remove all whitespace characters before splitting the string
    # that users may insert by accident, as this will result in error
    tf_cuda_compute_capabilities = ''.join(tf_cuda_compute_capabilities.split())
    for compute_capability in tf_cuda_compute_capabilities.split(','):

    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);
    }