$quant_input,
    $bias, $padding, $stride_h, $stride_w, $faf),
  (TFL_TransposeConvOp $output_shape, $quant_weights,
    $quant_input, $bias, $padding, $stride_h, $stride_w, $faf), 

include "tensorflow/compiler/mlir/tensorflow/ir/tf_ops.td"
include "mhlo/IR/hlo_ops.td"

def SignedIntTensor : TensorOf<[I1, I8, I16, I32, I64]>;
def UnsignedIntTensor : TensorOf<[UI8, UI16, UI32, UI64]>;

// IEEE compliant floating point tensors.
def IEEEFloatTensor : TensorOf<[F16, F32, F64]>;

//===----------------------------------------------------------------------===//
// BatchNorm op patterns.

// TFL Relu doesn't support I32/I64 type, so legalizes TF Relu to TFL Maximum.
def LegalizeReluI32 :
  Pat<(TF_ReluOp TensorOf<[I32]>:$arg),
      (TFL_MaximumOp $arg,
        (Arith_ConstantOp ConstantAttr<RankedI32ElementsAttr<[]>,"0">))>;
def LegalizeReluI64 :
  Pat<(TF_ReluOp TensorOf<[I64]>:$arg),
      (TFL_MaximumOp $arg,
        (Arith_ConstantOp

// Square op patterns.
//===----------------------------------------------------------------------===//

// Restrict the pattern to the types supported by SquaredDifferenceOp.
def TF_NumberTensorNoSmallInt: TensorOf<[TF_Bfloat16, TF_Complex128, TF_Complex64,
    TF_Float16, TF_Float32, TF_Int32, TF_Int64]>;

def SquareOfSub : Pat<
  (TF_SquareOp:$src (TF_SubOp TF_NumberTensorNoSmallInt:$arg0,
    TF_NumberTensorNoSmallInt:$arg1)),

  "allowed tfr.call result types">;

// standard tensor type and tfr.tensor types can be casted to each other.
def TFR_singleTensorType : Type<Or<[
    TFR_TensorType.predicate,
    TF_Tensor.predicate,
    TensorOf<[quant_QuantizedType]>.predicate]>, "single tensor or tfr.tensor type">;

// all allowed build list input types
def TFR_allowedBuiltListType : Type<Or<[
    TFR_TensorType.predicate,
    TF_ElementType.predicate,

class TFL_TensorOf<list<Type> allowedRuntimeTypes,
                   list<Type> allowedOpTypes = [AnyType]> :
  TensorOf<allowedOpTypes>, TFL_RuntimeType<TensorOf<allowedRuntimeTypes>> {
  // Set the summary equal to that representing the runtime types.
  let summary = TensorOf<allowedRuntimeTypes>.summary;
}

class TFL_TensorOfOrNone<list<Type> allowedRuntimeTypes, string description = "",