if (value_shaped_type != nullptr) {
        auto operand_or_result = value_shaped_type.getElementType();
        std::string dtype;

        TypeSwitch<Type>(operand_or_result)
            .Case<IntegerType>([&](Type) {
              dtype =
                  absl::StrCat("i", operand_or_result.getIntOrFloatBitWidth());
            })
            .Case<FloatType>([&](Type) {
              dtype =

    *dtype = DT_FLOAT8_E4M3FN;
    return absl::OkStatus();
  } else if (type.isFloat8E5M2()) {
    *dtype = DT_FLOAT8_E5M2;
    return absl::OkStatus();
  } else if (auto itype = mlir::dyn_cast<mlir::IntegerType>(type)) {
    switch (itype.getWidth()) {
      case 1:
        *dtype = DT_BOOL;
        return absl::OkStatus();
      case 4:
        *dtype = itype.isUnsigned() ? DT_UINT4 : DT_INT4;

    } else if (complex_type.getElementType().isF64()) {
      return tflite::TensorType_COMPLEX128;
    }
    llvm_unreachable("invalid complex Type in conversion");
  } else if (auto itype = mlir::dyn_cast<mlir::IntegerType>(type)) {
    switch (itype.getWidth()) {
      case 1:
        return tflite::TensorType_BOOL;
      case 4:
        if (itype.isUnsigned())

  return rewriter.create<mhlo::TransposeOp>(value.getLoc(), type, value,
                                            permutation);
}

Value CreateCastToInt32(Value val, Location loc, PatternRewriter& rewriter) {
  IntegerType new_ele_type = rewriter.getIntegerType(32);
  if (auto shaped_type = mlir::dyn_cast<RankedTensorType>(val.getType())) {
    ShapedType new_type =
        RankedTensorType::get(shaped_type.getShape(), new_ele_type);

        if (auto attr = inst.getAttrOfType<mlir::ElementsAttr>(attr_id)) {
          mlir::Attribute rand_val;
          mlir::Type element_type = attr.getShapedType().getElementType();
          if (mlir::isa<mlir::IntegerType>(element_type)) {
            rand_val = mlir::IntegerAttr::get(element_type, std::rand());
          } else if (element_type.isF16() || element_type.isF32() ||
                     element_type.isF64()) {

		x.mode = value
		x.typ = &emptyInterface
		if check.recordTypes() {
			check.recordBuiltinType(call.Fun, makeSig(x.typ))
		}

	case _Add:
		// unsafe.Add(ptr unsafe.Pointer, len IntegerType) unsafe.Pointer
		check.verifyVersionf(call.Fun, go1_17, "unsafe.Add")

		check.assignment(x, Typ[UnsafePointer], "argument to unsafe.Add")
		if x.mode == invalid {
			return
		}

		y := args[1]

		x.mode = value
		x.typ = &emptyInterface
		if check.recordTypes() {
			check.recordBuiltinType(call.Fun, makeSig(x.typ))
		}

	case _Add:
		// unsafe.Add(ptr unsafe.Pointer, len IntegerType) unsafe.Pointer
		check.verifyVersionf(call.Fun, go1_17, "unsafe.Add")

		check.assignment(x, Typ[UnsafePointer], "argument to unsafe.Add")
		if x.mode == invalid {
			return
		}

		y := args[1]

  /// Performs the lowering to TFLite dialect.
  void runOnOperation() override;
};

// Util that casts 'val' to Int32 by adding a cast Op.
Value CreateCastToInt32(Value val, Location loc, PatternRewriter& rewriter) {
  IntegerType new_ele_type = rewriter.getIntegerType(32);
  if (auto shaped_type = mlir::dyn_cast<RankedTensorType>(val.getType())) {
    ShapedType new_type =
        RankedTensorType::get(shaped_type.getShape(), new_ele_type);

      mlir::cast<ComplexType>(elemType).getElementType().isF32())
    return true;

  // QUINT8 and UI8 are allowed.
  if (mlir::isa<TF::Quint8Type>(elemType) ||
      (elemType.isInteger(8) && mlir::cast<IntegerType>(elemType).isUnsigned()))
    return true;

  return false;
}

func::FuncOp CreateOutlineFunc(StringRef name, Region& region,
                               bool passthru_extra_args, int num_loop_carried,

    return func.emitError() << "First input should be a string tensor";
  }
  auto row_splits_type = GetInputType(func, 1);
  if (!row_splits_type ||
      !mlir::isa<IntegerType>(row_splits_type.getElementType())) {
    return func.emitError() << "Second input should be an integer tensor";
  }

  auto hash_seed =
      mlir::dyn_cast_or_null<ArrayAttr>(attr.getAttrs().get("hash_seed"));