(*node_def->mutable_attr())["_handle_dtypes"] = handle_dtypes_attr;
      (*node_def->mutable_attr())["_handle_shapes"] = handle_shapes_attr;
    }
  }

  TF_RETURN_IF_ERROR(
      SetShapeAttribute("_output_shapes", arg_type, node_def->mutable_attr()));

  DataType dtype;
  TF_RETURN_IF_ERROR(ConvertToDataType(arg_type.getElementType(), &dtype));
  AttrValue type_attr;
  type_attr.set_type(dtype);

def FuseMapAndBatch : Pat<
  (TF_BatchDatasetV2Op
     (TF_MapDatasetOp $input_dataset, $other_arguments, $f, $output_types,
        $output_shapes, $use_inter_op_parallelism, $preserve_cardinality,
        $force_synchronous, $map_dataset_metadata),
     $batch_size, $drop_remainder, $parallel_copy, $batch_output_types,

  %cst = "tf.Const"() <{value = dense<1> : tensor<i32>}> {device = ""} : () -> tensor<i32>
  %0 = "tf.MultiDeviceIteratorFromStringHandle"(%arg0) <{output_shapes = [#tf_type.shape<>], output_types = [!tf_type.string]}> {device = ""} : (tensor<!tf_type.string>) -> tensor<!tf_type.resource>

  }

  auto output_shape = xla::ShapeInference::InferGatherShape(
      input_shape, start_indices_shape, gather_dim_numbers, slice_sizes);
  if (!output_shape.ok()) {
    op->emitError() << output_shape.status().message();
    return false;
  }

  auto refined_type = xla::ConvertShapeToType<RankedTensorType>(
      *output_shape, mlir::Builder(op));
  if (!refined_type.ok()) {

    auto new_shape = rewriter.create<TF::ShapeOp>(loc, shape_type, input);
    SmallVector<int64_t, 8> output_shape(/*Size=*/1, op.getNumElements());
    for (const auto &dim : dense_elem_attr.getValues<APInt>())
      output_shape.push_back(dim.getSExtValue());
    RankedTensorType result_type = tensorflow::GetTypeFromTFTensorShape(
        output_shape, getElementTypeOrSelf(input));
    rewriter.replaceOpWithNewOp<TF::ReshapeOp>(op, result_type, input,

def LegalizeSparseToDense : Pat<
  (TF_SparseToDenseOp $sparse_indices, $output_shape, $sparse_values,
    $default_value, $validate_indices),
  (TFL_SparseToDenseOp $sparse_indices, $output_shape, $sparse_values,
    $default_value)>;

def LegalizeUnique : Pat<(TF_UniqueOp $arg0),(TFL_UniqueOp $arg0)>;

  let summary = "A container for an iterator resource.";

  let arguments = (ins
    ConfinedAttr<TypeArrayAttr, [ArrayMinCount<1>]>:$output_types,
    ConfinedAttr<TF_ShapeAttrArray, [ArrayMinCount<1>]>:$output_shapes
  );

  let results = (outs
    Res<TF_ResourceTensor, [{A handle to the iterator that can be passed to a "MakeIterator" or
"IteratorGetNext" op. In contrast to Iterator, AnonymousIterator prevents

    // Shape of the final output. (Except for dimension folding in the
    // single diagonal case.)
    Shape output_shape;
    for (int i = 0; i < num_dims - 2; i++) {
      output_shape.push_back(input_type.getDimSize(i));
    }
    output_shape.push_back(num_diags);
    output_shape.push_back(max_diag_len);

    // A slice is the shape of what GatherOp copies per lookup. So the last

      %arg1: tensor<i64>
    ) {
      %1 = "tf.ReduceDataset"(%arg0, %arg1) {
        Targuments = [],
        Tstate = [i64], device = "",
        f = @__reduce_func_1, f._tf_data_function = true,
        output_shapes = [#tf_type.shape<>],
        output_types = [i64], use_inter_op_parallelism = true, _xla_compile_device_type="TPU"} :
          (tensor<!tf_type.variant>, tensor<i64>) -> (tensor<i64>)
      func.return
    }
    ```

  let summary = "Transpose convolution operator";

  let description = [{
    Performs transpose convolution operation on input.
  }];

  let arguments = (ins
    TFL_I32Tensor:$output_shape,
    TFL_TensorOf<[F32, QI8, QUI8, QI16]>:$weights,
    TFL_TensorOf<[F32, QI8, QUI8, QI16]>:$input,
    TFL_TensorOfOrNone<[F32, QI32, I64]>:$bias,
    TFL_PaddingAttr:$padding,