%3 = "tf.Identity"(%2) {device = ""} : (tensor<1x3x1x1xf32>) -> tensor<1x3x1x1xf32>
    return %3 : tensor<1x3x1x1xf32>
  }

// CHECK-LABEL: func @multiple_quantizable_ops_in_graph

  %8 = "tf.FloorMod"(%7, %4) {device = ""} : (tensor<1x1x2x1xi32>, tensor<i32>) -> tensor<1x1x2x1xi32>
  %9 = "tf.FloorDiv"(%arg1, %4) {device = ""} : (tensor<1x1x2x1xi32>, tensor<i32>) -> tensor<1x1x2x1xi32>
  %10 = "tf.Pack"(%2, %9, %8, %2) {axis = 0 : i64, device = ""} : (tensor<1x1x2x1xi32>, tensor<1x1x2x1xi32>, tensor<1x1x2x1xi32>, tensor<1x1x2x1xi32>) -> tensor<4x1x1x2x1xi32>

  // Unsupported strides
  %2 = "tf.MaxPool"(%arg0) {T = "tfdtype$DT_FLOAT", data_format = "NHWC", ksize = [1, 3, 6, 1], padding = "VALID", strides = [1, 3, 1, 3]} : (tensor<1x1x1x16xf32>) -> tensor<1x1x1x16xf32>

  %5 = arith.addf %0, %1 : tensor<1x1x1x16xf32>
  %6 = arith.addf %2, %5 : tensor<1x1x1x16xf32>
  func.return %6 : tensor<1x1x1x16xf32>

}

// CHECK-LABEL: prepareConv2D
// PerTensor-LABEL: prepareConv2D
func.func @prepareConv2D(%arg0: tensor<1x5x5x1xf32>) -> tensor<1x5x5x3xf32> {
  %w = arith.constant dense<[[[[0.0]]], [[[127.0]]], [[[-127.0]]]]> : tensor<3x1x1x1xf32>
  %b = arith.constant dense<0.0> : tensor<3xf32>

func.func @InvalidFuseTileAlreadyBroadcastAlongTileDim(%arg0: tensor<1x1x1x1xf32>) -> tensor<1x6x8x1xf32> {
  %cst_1 = arith.constant dense<[1, 6, 8, 1]> : tensor<4xi32>
  %cst_2 = arith.constant dense<[1, 1, 1, 46]> : tensor<4xi32>
  %cst_20 = arith.constant dense<4.600000e+01> : tensor<f32>
  %0 = "tfl.tile"(%arg0, %cst_1) : (tensor<1x1x1x1xf32>, tensor<4xi32>) -> tensor<1x6x8x1xf32>

  func.func @simple_folding(%arg0: tensor<1x1x1x1xi32>, %arg1: tensor<1x1x1x1xf32>) -> tensor<?x?x?x?xf32> {
    // CHECK: %[[SHAPE:.*]] = "tf.Shape"
    // CHECK: %[[CONV:.*]] = "tf.Conv2DBackpropInput"(%[[SHAPE]]
    // CHECK-SAME: (tensor<4xi32>, tensor<1x1x1x1xf32>, tensor<1x1x1x1xf32>) -> tensor<1x1x1x1xf32>
    // CHECK: return %[[CONV]] : tensor<1x1x1x1xf32>
    %0 = "tf.Shape"(%arg0) : (tensor<1x1x1x1xi32>) -> tensor<4xi32>

    } : (tensor<1x6x6x16xf32>) -> tensor<1x1x1x16xf32>
  func.return %1 : tensor<1x1x1x16xf32>

// CHECK: %0 = "tfl.dequantize"(%arg0)
// CHECK: %1 = "tfl.average_pool_2d"(%0)
// CHECK: %2 = "tfl.quantize"(%1)
// CHECK: %3 = "tfl.dequantize"(%2)
// CHECK: return %3 : tensor<1x1x1x16xf32>
}

// CHECK-LABEL: QuantizeMaximum

    quantization_axis = -1 : i64, quantization_min_val = -128 : i64, quantization_max_val = 127 : i64
  } : (tensor<1x2x2x1x!tf_type.qint32>, tensor<f32>, tensor<i32>) -> tensor<1x2x2x1xf32>

  return %output : tensor<1x2x2x1xi32>

// CHECK-LABEL: func @tfAssertTrue
func.func @tfAssertTrue(%arg0: tensor<1x1x6x2xf32>) {
  %t = arith.constant dense<true> : tensor<i1>
  // CHECK-NOT: tf.Assert
  "tf.Assert"(%t, %arg0) {summarize = 3} : (tensor<i1>, tensor<1x1x6x2xf32>) -> ()
  func.return
}

// CHECK-LABEL: func @tfAssertFalse
func.func @tfAssertFalse(%arg0: tensor<1x1x6x2xf32>) {
  %f = arith.constant dense<false> : tensor<i1>
  // CHECK: tf.Assert

  // CHECK-NOT: tf.Lgamma
  %0 = "tf.Lgamma"(%arg0) : (tensor<4xf32>) -> tensor<4xf32>
  func.return %0 : tensor<4xf32>
}

// CHECK-LABEL: func @imag_resize_nearest
func.func @imag_resize_nearest(%arg0: tensor<1x7x7x1xi32>) -> tensor<1x3x3x1xi32> {
  %shape = "tf.Const"() {device = "", value = dense<3> : tensor<2xi32>} : () -> tensor<2xi32>

  // CHECK-DAG: [[VAL0:%.+]] = "tf.Const"() <{value = dense<1> : tensor<i32>}>