%paddings = "tf.Const"() {value = dense<[[1, 2], [1, 2]]> : tensor<2x2xi64>} : () -> tensor<2x2xi64>
  // expected-error @+1 {{requires block_shape[i] divides input_shape[i + 1] + paddings[i, 0] + paddings[i, 1]; failed for i=1}}
  %1 = "tf.SpaceToBatchND"(%input, %block_shape, %paddings) : (tensor<3x5x7x10xf32>, tensor<2xi64>, tensor<2x2xi64>) -> tensor<36x2x3x10xf32>

    for padding `input`.

    The padded size of each dimension D of the output is:

      `paddings(D, 0) + input.dim_size(D) + paddings(D, 1)`

    For example:

    ```
    # 't' is [[1, 1], [2, 2]]
    # 'paddings' is [[1, 1], [2, 2]]
    # rank of 't' is 2
    pad(t, paddings) ==> [[0, 0, 0, 0, 0, 0]
                          [0, 0, 1, 1, 0, 0]

    return failure();

  ElementsAttr padding;
  if (matchPattern(op.getPadding(), m_Constant(&padding))) {
    const ShapedType &padding_ty = cast<ShapedType>(padding.getType());
    if (padding_ty.getRank() != 2 || padding_ty.getDimSize(1) != 2) {
      return op.emitOpError()
             << "expects padding to be a matrix with minor dimension 2, got "
             << padding.getShapedType().getShape();
    }
  }

-2,.px-2{padding-right:.5rem!important}.pb-2,.py-2{padding-bottom:.5rem!important}.pl-2,.px-2{padding-left:.5rem!important}.p-3{padding:1rem!important}.pt-3,.py-3{padding-top:1rem!important}.pr-3,.px-3{padding-right:1rem!important}.pb-3,.py-3{padding-bottom:1rem!important}.pl-3,.px-3{padding-left:1rem!important}.p-4{padding:1.5rem!important}.pt-4,.py-4{padding-top:1.5rem!important}.pr-4,.px-4{padding-right:1.5rem!important}.pb-4,.py-4{padding-bottom:1.5rem!important}.pl-4,.px-4{padding-left:1.5r...

-2,.px-2{padding-right:.5rem!important}.pb-2,.py-2{padding-bottom:.5rem!important}.pl-2,.px-2{padding-left:.5rem!important}.p-3{padding:1rem!important}.pt-3,.py-3{padding-top:1rem!important}.pr-3,.px-3{padding-right:1rem!important}.pb-3,.py-3{padding-bottom:1rem!important}.pl-3,.px-3{padding-left:1rem!important}.p-4{padding:1.5rem!important}.pt-4,.py-4{padding-top:1.5rem!important}.pr-4,.px-4{padding-right:1.5rem!important}.pb-4,.py-4{padding-bottom:1.5rem!important}.pl-4,.px-4{padding-left:1.5r...

-2,.px-2{padding-right:.5rem!important}.pb-2,.py-2{padding-bottom:.5rem!important}.pl-2,.px-2{padding-left:.5rem!important}.p-3{padding:1rem!important}.pt-3,.py-3{padding-top:1rem!important}.pr-3,.px-3{padding-right:1rem!important}.pb-3,.py-3{padding-bottom:1rem!important}.pl-3,.px-3{padding-left:1rem!important}.p-4{padding:1.5rem!important}.pt-4,.py-4{padding-top:1.5rem!important}.pr-4,.px-4{padding-right:1.5rem!important}.pb-4,.py-4{padding-bottom:1.5rem!important}.pl-4,.px-4{padding-left:1.5r...

      // stablehlo.convolution op needs explicit padding to be set to model any
      // Transposed-Convolution in JAX/PT. Checking to see if-
      // 1. Pre set padding matches to the desired padding
      // 2. Output size respects the `VALID` padding scenario
      if ((padding[2 * i] == padding[2 * i + 1]) &&
          (((kernel_size - 1) != padding[2 * i]) ||

    SmallVector<int64_t> padded_shape(tensor_shape.begin(), tensor_shape.end());
    for (int i = 0; i < padded_shape.size(); ++i) {
      // Left padding + right padding.
      int32_t padding = tfl_pad_values[i * 2] + tfl_pad_values[i * 2 + 1];
      padded_shape[i] += padding;
    }

    return padded_shape;
  }

  std::pair<int64_t, int64_t> GetDimSize(

    return emitOptionalError(op.getLoc(), "Invalid data format provided");
  }

  const StringRef paddings = op.getPadding();
  tensorflow::Padding padding;
  auto padding_is_valid = GetPaddingFromString(paddings.str(), &padding);
  if (!padding_is_valid.ok()) {
    return emitOptionalError(op.getLoc(), "Invalid padding format provided");
  }

  // Verifies that,
  // * Ranks of operands and result are valid

  // CHECK: "VALID"