kernel_spatial_dimensions = #vhlo.tensor_v1<dense<[0, 1]> : tensor<2xi64>>,
                                             lhs_dilation = #vhlo.tensor_v1<dense<1> : tensor<2xi64>>,
                                             output_batch_dimension = #vhlo.integer_v1<0 : i64>,

table StablehloSliceOptions{
  start_indices : [long];
  limit_indices : [long];
  strides : [long];
}

table StablehloConvolutionOptions{
  window_strides : [long];
  padding : [long];
  lhs_dilation : [long];
  rhs_dilation : [long];
  window_reversal : [bool];
  input_batch_dimension : long;
  input_feature_dimension : long;
  input_spatial_dimensions : [long];
  kernel_input_feature_dimension : long;

    // Compute ConvDimensionNumbers, dilation, and padding.
    SmallVector<int64_t, num_spatial_dims> spatial_dims;
    SmallVector<int64_t, num_spatial_dims> lhs_dilation;
    SmallVector<int64_t, num_spatial_dims> rhs_dilation;
    SmallVector<int64_t, num_spatial_dims * 2> paddings;

    for (int i : llvm::seq<int>(0, num_spatial_dims)) {
      const int64_t spatial_dim =

  { auto _e = lhs_dilation(); if (_e) { _o->lhs_dilation.resize(_e->size()); for (::flatbuffers::uoffset_t _i = 0; _i < _e->size(); _i++) { _o->lhs_dilation[_i] = _e->Get(_i); } } else { _o->lhs_dilation.resize(0); } }
  { auto _e = rhs_dilation(); if (_e) { _o->rhs_dilation.resize(_e->size()); for (::flatbuffers::uoffset_t _i = 0; _i < _e->size(); _i++) { _o->rhs_dilation[_i] = _e->Get(_i); } } else { _o->rhs_dilation.resize(0); } }

  %feature_group_count = "tf.Const"() {value = dense<1> : tensor<i32>} : () -> tensor<i32>
  %lhs_dilation = "tf.Const"() {value = dense<[4, 1, 1]> : tensor<3xi32>} : () -> tensor<3xi32>
  %rhs_dilation = "tf.Const"() {value = dense<1> : tensor<3xi32>} : () -> tensor<3xi32>
  %padding = "tf.Const"() {value = dense<0> : tensor<3x2xi32>} : () -> tensor<3x2xi32>

  if (lhs_dilation_attr.getType().getRank() != 1)
    return op.emitOpError() << "expects lhs_dilation to be a vecotr";

  if (rhs_dilation_attr.getType().getRank() != 1)
    return op.emitOpError() << "expects rhs_dilation to be a vecotr";

  if (feature_group_count_attr.getType().getRank())
    return op.emitOpError() << "expects feature_group_count to be a scalar";

  %1 = stablehlo.convolution(%arg0, %0) dim_numbers = [b, 0, 1, f]x[0, 1, i, o]->[b, 0, 1, f], window = {pad = [[1, 1], [1, 1]]} {batch_group_count = 1 : i64, feature_group_count = 1 : i64, window_strides = array<i64: 1, 1>, lhs_dilation = array<i64: 3, 3>} : (tensor<1x5x5x2x!quant.uniform<i8:f32, 2.000000e+00:0>>, tensor<2x2x2x4x!quant.uniform<i8:f32:3, {3.000000e+00, 3.000000e+00, 3.000000e+00, 3.000000e+00}>>) -> tensor<1x14x14x4x!quant.uniform<i32:f32:3, {6.000000e+00, 6.000000e+00,...

  %feature_group_count = "tf.Const"() {value = dense<1> : tensor<i32>} : () -> tensor<i32>
  %lhs_dilation = "tf.Const"() {value = dense<[4, 1, 1]> : tensor<3xi32>} : () -> tensor<3xi32>
  %rhs_dilation = "tf.Const"() {value = dense<1> : tensor<3xi32>} : () -> tensor<3xi32>
  %padding = "tf.Const"() {value = dense<0> : tensor<3x2xi32>} : () -> tensor<3x2xi32>

    StrAttr:$padding,
    DefaultValuedOptionalAttr<I64ArrayAttr, "{}">:$explicit_padding,
    DefaultValuedOptionalAttr<I64ArrayAttr, "{}">:$lhs_dilation,
    DefaultValuedOptionalAttr<I64ArrayAttr, "{}">:$rhs_dilation,
    DefaultValuedOptionalAttr<I64Attr, "1">:$batch_group_count,
    DefaultValuedOptionalAttr<I64Attr, "1">:$feature_group_count,

            mlir::cast<mlir::vhlo::TensorV1Attr>(vhlo_op.getPadding()),
            vhlo_type_converter));
        auto lhs_dialation = builder_.CreateVector(mlir::GetVector<int64_t>(
            mlir::cast<mlir::vhlo::TensorV1Attr>(vhlo_op.getLhsDilation()),
            vhlo_type_converter));
        auto rhs_dialation = builder_.CreateVector(mlir::GetVector<int64_t>(
            mlir::cast<mlir::vhlo::TensorV1Attr>(vhlo_op.getRhsDilation()),