NodeWith(Op("Add"), Name("add"),
                       Inputs(Out(NodeWith(Name("placeholder_a"))),
                              Out(NodeWith(Name("placeholder_b"))))));

  EXPECT_EQ(Explain(add.node(), NodeWith(Inputs())),
            "\nexpected 0 inputs but node has 2");
  EXPECT_EQ(
      Explain(add.node(), NodeWith(Inputs(Out(NodeWith(Name("blah"))), _))),

// Creates parameters (placeholders) in the tracing `ctx` using the shape and
// dtype of `inputs`.
Status CreateParamsForInputs(AbstractContext* ctx,
                             absl::Span<AbstractTensorHandle* const> inputs,
                             std::vector<AbstractTensorHandle*>* params);

// A callable that takes tensor inputs and returns zero or more tensor outputs.
using Model = std::function<Status(AbstractContext*,

                  const std::vector<Output>& inputs,
                  std::vector<Output>* outputs) {
  DCHECK(outputs != nullptr);
  DCHECK(outputs->empty());

  // The control dependency is analogous to that in CreateCond().
  Scope body_scope =
      scope.NewSubScope("body").WithControlDependencies(inputs[0]);
  TF_RETURN_IF_ERROR(body(body_scope, inputs, outputs));

  const size_t num_loop_vars = inputs.size();

  bool UseFunction() const { return std::get<2>(GetParam()); }
};

// Checks that inputs[0] is a scalar.
Status TestScalarShape(AbstractContext* ctx,
                       absl::Span<AbstractTensorHandle* const> inputs,
                       absl::Span<AbstractTensorHandle*> outputs) {
  PartialTensorShape shape;
  TF_RETURN_IF_ERROR(inputs[0]->Shape(&shape));
  if (shape.dims() != 0) {
    return errors::InvalidArgument(

  if (ops.empty()) {
    return;
  }

  std::vector<Value> inputs;
  std::vector<Value> outputs;
  computeInputsOutputs(ops, &inputs, &outputs);

  std::vector<Type> input_types;
  std::vector<Type> output_types;

  input_types.reserve(inputs.size());
  for (Value v : inputs) {
    input_types.push_back(v.getType());
  }
  output_types.reserve(outputs.size());

//   return tf.exp(input), grad
// outputs = [f(inputs[0])]
Status ExpWithPassThroughGrad(AbstractContext* ctx,
                              absl::Span<AbstractTensorHandle* const> inputs,
                              absl::Span<AbstractTensorHandle*> outputs) {
  Tape tape(/*persistent=*/false);
  tape.Watch(inputs[0]);  // Watch x.
  AbstractTensorHandle* exp_output;
  TF_RETURN_IF_ERROR(ops::Exp(ctx, inputs[0], &exp_output, "Exp"));

// CHECK-NEXT:       has_rank: true
// CHECK-NEXT:     } ],
// CHECK-NEXT:     inputs: [ 0, 1 ],
// CHECK-NEXT:     outputs: [ 3 ],
// CHECK-NEXT:     operators: [ {
// CHECK-NEXT:       inputs: [ 0, 1 ],
// CHECK-NEXT:       outputs: [ 2 ]
// CHECK-NEXT:     }, {
// CHECK-NEXT:       opcode_index: 1,
// CHECK-NEXT:       inputs: [ 2, 0, 1 ],
// CHECK-NEXT:       outputs: [ 3 ],

  Example:
    @composite.Composite('AddN')
    def _compose_add_n(inputs, N):
      if N == 1:
        ....
  """

  # TODO(fengliuai): support input_binding and output_binding so the arguments
  # are not positional.
  def __init__(self,
               op_name,
               inputs=None,
               attrs=None,
               derived_attrs=None,
               outputs=None):

        "inputs.file('a')"                                          | "TaskInputs.file(Object)"
        "inputs.files('a')"                                         | "TaskInputs.files(Object...)"
        "inputs.dir('a')"                                           | "TaskInputs.dir(Object)"
        "inputs.property('key', 'value')"                           | "TaskInputs.property(String, Object)"

func (f *Selector) Selects(inputs Set) bool {
	return !inputs.containsAny(f.absent) && inputs.containsAll(f.present)
}

// Excludes returns false, if the given set of labels, even combined with new ones, could end up satisfying the Selector.
// It returns false, if Matches would never return true, even if new labels are added to the input set.
func (f *Selector) Excludes(inputs Set) bool {
	return inputs.containsAny(f.absent)
}