// CHECK: %[[INPUT00:.*]] = "tf.IteratorGetNext"
    // CHECK-DAG: %[[SPACETODEPTH00:.*]] = "tf.SpaceToDepth"([[INPUT00:.*]]#0) <{block_size = 2 : i64, data_format = "NHWC"}> : (tensor<2x224x224x3xf32>) -> tensor<2x112x112x12xf32>

  EXPECT_THAT(input2->type, Eq(TensorType_INT8));

  // Check if the quantization params of the minimum/maximum inputs match
  // after requantization
  EXPECT_THAT(input1->quantization->scale, Eq(input2->quantization->scale));
  EXPECT_THAT(input1->quantization->zero_point,
              Eq(input2->quantization->zero_point));

  // Check the input quantization params match the output ones.

  // TFL lstm only supports time-majored inputs, so if it's not time-majored,
  // we will transpose the inputs and outputs.
  auto time_major_attr = func_op->getAttrOfType<BoolAttr>("tf.time_major");
  if (time_major_attr == nullptr) return failure();

  bool time_majored = time_major_attr.getValue();
  auto input_type = mlir::dyn_cast_or_null<RankedTensorType>(input.getType());
  if (!input_type) {

    // from inferring output shape for Merge node (we need shapes for all its
    // inputs).
    // For loop invariant resource input's Merge node, we set output resource
    // shape as Enter node's resource shape.
    // TODO(b/129367850): clean this up.
    if (n->IsMerge() && n->output_type(0) == DT_RESOURCE) {
      // Check if this is a loop invariant input's Merge node. We do it by

    private final String[] inputs;

    private final String output;

    private String[] newInputs;

    private String newOutput;

    public CharMappingItem(final long id, final String[] inputs, final String output) {
        this.id = id;
        this.inputs = inputs;
        this.output = output == null ? null : output.replace("\n", " ");
        Arrays.sort(inputs);

        if (id == 0) {

    protected AbstractModelAction(ModelReference<T> subject, ModelRuleDescriptor descriptor, ModelReference<?>... inputs) {
        this(subject, descriptor, inputs == null ? EMPTY_MODEL_REF_LIST : Arrays.asList(inputs));
    }

    protected AbstractModelAction(ModelReference<T> subject, ModelRuleDescriptor descriptor, List<? extends ModelReference<?>> inputs) {

        return new AbstractModelAction<T>(reference, descriptor, input1, input2) {
            @Override
            public void execute(MutableModelNode modelNode, List<ModelView<?>> inputs) {
                action.execute(modelNode,
                    Cast.<I>uncheckedCast(inputs.get(0).getInstance()),
                    Cast.<J>uncheckedCast(inputs.get(1).getInstance())
                );
            }
        };
    }

        private final ModelType<T> type;
        private final List<ModelReference<?>> inputs;

        private UnmanagedElementCreationAction(ModelRuleDescriptor descriptor, ModelReference<?> subject, List<ModelReference<?>> inputs, ModelType<T> type) {
            this.subject = subject;
            this.inputs = inputs;
            this.descriptor = descriptor;
            this.type = type;
        }

        @Override

                && Objects.equals(inputs, that.inputs);
    }

    @Override
    public int hashCode() {
        return Objects.hash(modelId, location, inputs);
    }

    Stream<InputSource> sources() {
        return inputs != null ? inputs.stream() : Stream.of(this);
    }

    @Override
    public String toString() {
        if (inputs != null) {

  std::vector<NodeDefBuilder::NodeOut> inputs;
  inputs.emplace_back("min", 0, DT_FLOAT);
  inputs.emplace_back("max", 0, DT_FLOAT);
  inputs.emplace_back("histogram", 0, DT_INT64);
  inputs.emplace_back("min", 0, DT_FLOAT);
  inputs.emplace_back("max", 0, DT_FLOAT);
  inputs.emplace_back("histogram", 0, DT_INT64);

  const std::string dir = testing::TmpDir();