%1 = "tf.Mul"(%arg1, %arg1) : (tensor<2xf32>, tensor<2xf32>) -> tensor<2xf32>
  %2 = "tf.Div"(%arg1, %arg1) : (tensor<2xf32>, tensor<2xf32>) -> tensor<2xf32>

  // CHECK: "tf.If"{{.+}}else_branch = @tf.IfRegion_else{{.+}}then_branch = @tf.IfRegion_then
  %3 = "tf.IfRegion"(%arg0) ({
     %4 = "tf.Select"(%arg0, %0, %1) : (tensor<i1>, tensor<2xf32>, tensor<2xf32>) -> tensor<2xf32>
     "tf.Yield"(%4) : (tensor<2xf32>) -> ()

    func.return %0#0 : tensor<8x8x8x8xf32>
  }

  // CHECK-LABEL: fusedBatchNormV3_training

  %10 = "tfl.pseudo_const"() {value = dense<0.000000e+00> : tensor<3xf32>} : () -> tensor<3xf32>
  %11 = "tfl.pseudo_const"() {value = dense<1.000000e+00> : tensor<3xf32>} : () -> tensor<3xf32>
  %recurrent_input = "tfl.pseudo_const"() {value = dense<0.000000e+00> : tensor<1x3xf32>} : () -> tensor<1x3xf32>
  %recurrent_stats = "quantfork.stats"(%recurrent_input) {layerStats = dense<[0.0, 1.0]> : tensor<2xf32>} : (tensor<1x3xf32>) -> tensor<1x3xf32>

  func.return %0 : tensor<*xf32>
}

// CHECK-NOT: "tfl.reshape"
// CHECK: "tfl.pack"

// -----

func.func @squaredDifference(%arg0: tensor<4xf32>, %arg1: tensor<4xf32>) -> tensor<4xf32> {
  %0 = "tfl.squared_difference"(%arg0, %arg1) : (tensor<4xf32>, tensor<4xf32>) -> tensor<4xf32>
  func.return %0 : tensor<4xf32>
}

  %7 = "tfl.add"(%2, %1) {fused_activation_function = "NONE"} : (tensor<4xf32>, tensor<  f32>) -> tensor<4xf32>
  %8 = "tfl.add"(%2, %3) {fused_activation_function = "NONE"} : (tensor<4xf32>, tensor<4xf32>) -> tensor<4xf32>
  %9 = "tfl.add"(%2, %3) {fused_activation_function = "SIGN_BIT"} : (tensor<4xf32>, tensor<4xf32>) -> tensor<4xf32>

  %1:2 = "tfl.split"(%cst, %0) {num_splits = 2 : i32} : (tensor<i32>, tensor<4xf32>) -> (tensor<2xf32>, tensor<2xf32>)
  %2 = "tfl.quantize"(%1#0) {qtype = tensor<2x!quant.uniform<u8:f32, 1.0>>} : (tensor<2xf32>) -> tensor<2x!quant.uniform<u8:f32, 1.0>>
  %3 = "tfl.quantize"(%1#1) {qtype = tensor<2x!quant.uniform<u8:f32, 1.0>>} : (tensor<2xf32>) -> tensor<2x!quant.uniform<u8:f32, 1.0>>

// CHECK-LABEL: floor_mod
func.func @floor_mod(%arg0: tensor<5xf32>, %arg1: tensor<5xf32>) -> tensor<5xf32> {
  %0 = "tf.FloorMod"(%arg0, %arg1) : (tensor<5xf32>, tensor<5xf32>) -> tensor<5xf32>
  func.return %0 : tensor<5xf32>
// CHECK: %[[CUSTOM_0:.*]] = "tfl.custom"(%arg0, %arg1) <{custom_code = "FlexFloorMod", custom_option = #tfl<const_bytes : "{{.*}}">}> : (tensor<5xf32>, tensor<5xf32>) -> tensor<5xf32>
// CHECK: return %[[CUSTOM_0]] : tensor<5xf32>
}

  %1:4 = "tfl.split"(%cst, %0) {num_splits = 4 : i32} : (tensor<i32>, tensor<4xf32>) -> (tensor<2xf32>, tensor<2xf32>,tensor<2xf32>, tensor<2xf32>)
  %2 = "tfl.quantize"(%1#0) {qtype = tensor<2x!quant.uniform<u8:f32, 1.0>>} : (tensor<2xf32>) -> tensor<2x!quant.uniform<u8:f32, 1.0>>
  %3 = "tfl.quantize"(%1#1) {qtype = tensor<2x!quant.uniform<u8:f32, 1.0>>} : (tensor<2xf32>) -> tensor<2x!quant.uniform<u8:f32, 1.0>>

  %0 = "quantfork.stats"(%arg0) {
    layerStats = dense<[-1.28e-5, 1.27e-5]> : tensor<2xf32>
  } : (tensor<1x5x5x2xf32>) -> tensor<1x5x5x2xf32>
  %w = arith.constant dense<[[[[-1.0, 1.0]]], [[[1.0, 2.0]]], [[[-2.0, 1.0]]]]> : tensor<3x1x1x2xf32>
  %b = arith.constant dense<0.0> : tensor<3xf32>
  %b2 = arith.constant dense<[1.0e-2, 2.1473647e1, -2.1473647e2]> : tensor<3xf32>
  %conv = "tfl.conv_2d"(%0, %w, %b) {

  %cst_1 = "tf.Const"() {value = dense<[0.3, 0.4, 0.3, 0.4, 0.3, 0.4, 0.3, 0.4]> : tensor<8xf32>} : () -> tensor<8xf32>