"'' if a single data type is skipped. The data type from "
                   "the import graph is used if it is skipped."),
    llvm::cl::init(""));

// NOLINTNEXTLINE
opt<std::string> input_shapes(
    "tf-input-shapes",
    llvm::cl::desc(
        "Input tensor shapes. Shapes for different tensors are separated by "
        "':', and dimension sizes for the same tensor are separated by ','"),

      Op("Cast"), Inputs(Out(NodeWith(Op("Placeholder"), Name("begin"))))));
  auto m_input_shape = Out(NodeWith(Op("Shape"), Inputs(m_input)));
  auto m_slice_size_0 = Out(NodeWith(
      Op("Sub"), AssignedDevice(kHostName),
      Inputs(
          Out(NodeWith(Op("Slice"), AssignedDevice(kHostName),
                       Inputs(m_input_shape, Const(zero_64), Const(one_64)))),
          Out(NodeWith(Op("Slice"), AssignedDevice(kHostName),

                                           Builder* builder) {
  const auto& input_dtype = input_tensor.dtype();
  const auto& input_shape = input_tensor.shape();
  Type elt_type;
  TF_RETURN_IF_ERROR(ConvertDataType(input_dtype, *builder, &elt_type));
  SmallVector<int64_t, 4> shape;
  ConvertToMlirShape(input_shape, &shape);
  auto type = RankedTensorType::get(shape, elt_type);

#define CONVERT_FLAT(DTYPE, CTYPE) \

    ArrayRef<int64_t> input_shape = input_tensor.getShapedType().getShape();
    auto output_type = mlir::cast<ShapedType>(op.getOutput().getType());

    SmallVector<int32_t, 4> perm;
    SmallVector<int64_t, 4> output_shape;
    for (int i = 0; i < num_dimensions; ++i) {
      perm.push_back(perm_tensor.getValues<IntegerAttr>()[i].getInt());
      output_shape.push_back(input_shape[perm[i]]);

    return errors::InvalidArgument(absl::StrCat(
        "Length of input node array and data shape doesn't match (#arrays ",
        node_names.size(), ", #input_shapes ", node_shapes.size(), ")"));
  }

  // StringMap doesn't support reserve else reserve input map size here.
  for (int i = 0, end = node_names.size(); i < end; i++) {
    auto& name = node_names[i];

      auto dim_attr = DenseIntElementsAttr::get(dim_type, dims);
      auto dim = rewriter.create<arith::ConstantOp>(op.getLoc(), dim_attr);

      input_shape.insert(input_shape.begin() + dim_to_expand, 1);
      TensorType expanded_type = input_type.clone(input_shape);
      input = rewriter.create<TFL::ExpandDimsOp>(op.getLoc(), expanded_type,
                                                 input, dim);

  // the input tensor's dimensions, return 0-valued tensor of the requested
  // shape.
  ArrayRef<int64_t> input_shape = GetRankedTensorShape(input);
  for (int i = 0, end = input_shape.size(); i < end; i++) {
    if (begin_values[i] < 0 ||
        (begin_values[i] + size_values[i] > input_shape[i])) {
      return CreateF32SplatConst(builder, size_shape, 0, location);
    }
  }

    int num_input_dims = ranked_input_type.getRank();
    SmallVector<int32_t, 4> padding_begin(num_input_dims, 0);
    auto input_shape = ranked_input_type.getShape();
    SmallVector<int32_t, 4> padding_end(input_shape.begin(), input_shape.end());
    SmallVector<int32_t, 4> padding_strides(num_input_dims, 1);

    int begin_mask = strided_slice_op.getBeginMask();

        size_diff, scalar_zero);

    // Build the argument/result types for if branch function.
    auto input_shape = rewriter.create<TF::ShapeOp>(
        loc, tensorflow::GetTypeFromTFTensorShape({-1}, shape_dtype),
        input_handle);

    Type branch_args_type[] = {input_handle.getType(), input_shape.getType(),
                               size_diff.getType(), size.getType()};
    Type branch_result_type[] = {result_type};

        op->getLoc(), permuation_tensor_type,
        DenseElementsAttr::get(permuation_tensor_type, permute));

    auto input_shape = input_type.getShape();
    llvm::SmallVector<int64_t, 4> permuted_shape(input_shape.begin(),
                                                 input_shape.end());
    // Swaps z dimension and x dimension to get permuted shape.
    std::iter_swap(permuted_shape.begin() + input_rank - 1,