// Only 2d pooling is supported in TFLite.
    if (op.getWindowDimensions().size() != 4) {
      return failure();
    }

    // reduce_window op with dilations or padding will supported later.
    // TODO: b/321099943 - Support reduce_window op with dilations and padding.
    if (op.getBaseDilations().has_value() ||
        op.getWindowDilations().has_value() || op.getPadding().has_value()) {
      return failure();
    }

    %0 = "tf.Conv2D"(%arg0, %arg1) {attr_map = "0:strides,1:use_cudnn_on_gpu,2:padding,3:explicit_paddings,4:dilations", data_format = "NHWC", device = "", dilations = [1, 1, 1, 1], explicit_paddings = [], padding = "VALID", strides = [1, 1, 2, 1], use_cudnn_on_gpu = true} : (tensor<*xf32>, tensor<*xf32>) -> tensor<*xf32>
    %1 = "tf.Relu6"(%0) : (tensor<*xf32>) -> tensor<*xf32>

  attr {
    key: "T"
    value {
      type: DT_FLOAT
    }
  }
  attr {
    key: "data_format"
    value {
      s: "NHWC"
    }
  }
  attr {
    key: "dilations"
    value {
      list {
        i: 1
        i: 1
        i: 1
        i: 1
      }
    }
  }
  attr {
    key: "explicit_paddings"
    value {
      list {
      }

  attr {
    key: "T"
    value {
      type: DT_FLOAT
    }
  }
  attr {
    key: "data_format"
    value {
      s: "NHWC"
    }
  }
  attr {
    key: "dilations"
    value {
      list {
        i: 1
        i: 1
        i: 1
        i: 1
      }
    }
  }
  attr {
    key: "explicit_paddings"
    value {
      list {
      }

  %side_input_scale = "tf.Const"() {value = dense<2.0> : tensor<f32>} : () -> tensor<f32>
  %conv2d = "tf._FusedConv2D"(%input, %filter, %bias, %act, %input_scale, %side_input_scale) {

  %0 = "tf.Conv2DBackpropInput"(%arg0, %arg1, %arg2) {strides = [1, 2, 2, 1], padding="SAME", dilations=[1, 1, 1, 1]}: (tensor<4xi32>, tensor<3x3x1x32xf32>, tensor<15x14x14x32xf32>) -> tensor<15x28x28x1xf32>
  %1 = "tf.Conv2DBackpropInput"(%arg0, %arg1, %arg2) {strides = [1, 2, 2, 1], padding="VALID", dilations=[1, 1, 1, 1]}: (tensor<4xi32>, tensor<3x3x1x32xf32>, tensor<15x14x14x32xf32>) -> tensor<15x28x28x1xf32>

  // CHECK-SAME: batch_group_count = 1
  // CHECK-SAME: feature_group_count = 2

  %0 = "tf.Conv2D"(%arg0, %arg1) {data_format = "NHWC", dilations = [1, 2, 3, 1], padding = "SAME", strides = [1, 4, 5, 1]} : (tensor<256x32x32x6xf32>, tensor<3x3x3x16xf32>) -> tensor<256x8x7x16xf32>
  func.return %0 : tensor<256x8x7x16xf32>
}

// -----

     IsIntList1XY1:$strides,
     BoolAttr:$use_cudnn_on_gpu,
     IsSameOrValid:$padding,
     I64ArrayAttr:$explicit_paddings,
     IsDataFormatNHWC:$data_format,
     IsAllOnes:$dilations),
  (TFL_TransposeConvOp $input_sizes,
     (TFL_TransposeOp $filter,
       (Arith_ConstantOp ConstantAttr<I32VectorElementsAttr<4>, "{2, 0, 1, 3}">)),
     $out_backprop,
     /*bias=*/ (CreateNoneValue $input_sizes),

        verify_size(window_strides.size(), "window strides") &&
        verify_size(padding.size(), "padding entries") &&
        verify_size(lhs_dilation.size(), "lhs dilation factors") &&
        verify_size(rhs_dilation.size(), "rhs dilation factors")))
    return std::nullopt;

  xla::Window window;
  for (size_t i = 0; i < window_dimensions.size(); i++) {
    auto dim = window.add_dimensions();

}

// -----

func.func @testConv2D(%arg0: tensor<256x32x32x3xf32>, %arg1: tensor<3x3x3x16xf32>) -> tensor<256x30x30x16xf32> {
  // expected-error @+1 {{requires dilations attribute length to be 4}}
  %0 = "tf.Conv2D"(%arg0, %arg1) {padding = "SAME", strides = [1, 1, 1, 1], dilations = [1, 1]} : (tensor<256x32x32x3xf32>, tensor<3x3x3x16xf32>) -> tensor<256x30x30x16xf32>
  func.return %0 : tensor<256x30x30x16xf32>
}

// -----