%43 = "quantfork.stats"(%42) {layerStats = dense<[-1.0, 1.0]> : tensor<2xf32>} : (tensor<1x4x3xf32>) -> tensor<1x4x3xf32>
  %44 = "tfl.strided_slice"(%43, %1, %2, %3) {begin_mask = 7 : i32, ellipsis_mask = 0 : i32, end_mask = 5 : i32, new_axis_mask = 0 : i32, shrink_axis_mask = 0 : i32, offset = false} : (tensor<1x4x3xf32>, tensor<3xi32>, tensor<3xi32>, tensor<3xi32>) -> tensor<1x2x3xf32>

  %6 = mhlo.divide %3, %5 : tensor<1x3x4x4xf32>
  return %6 : tensor<1x3x4x4xf32>
}

// CHECK-LABEL:   func.func @avg_pool2d_1(
// CHECK-SAME:                            %[[VAL_0:.*]]: tensor<1x3x6x6xf32>) -> tensor<*xf32> {
// CHECK:           %[[VAL_1:.*]] = arith.constant dense<[0, 2, 3, 1]> : tensor<4xi32>

        version = 5 : i64
      } : (tensor<?x3x4x3xf32>, tensor<2x3x3x2xf32>, tensor<1x1x1x2xf32>) -> tensor<?x3x4x2xf32>
    %2 = "quantfork.stats"(%1) {layerStats = dense<[5.00000000e-6, 7.00000000e-1]> : tensor<2xf32>} : (tensor<?x3x4x2xf32>) -> tensor<?x3x4x2xf32>
    return %2 : tensor<?x3x4x2xf32>
  }

  func.return %0, %1, %2 : tensor<1x4x4xf32>, tensor<1x4x4xf32>, tensor<14x4x4xf32>
}

// -----

func.func @testSplitVOpWithBadSizeSplitsConstantSum(%arg0: tensor<16x4x4xf32>) -> (tensor<0x4x4xf32>, tensor<16x4x4xf32>) {

func.func @no_qdq_except_conv_and_dot_general(%arg0: tensor<2x3x2xi64>) -> tensor<2x3x2x2xf32> attributes {tf._original_func_name = "main_0"} {
  %cst = "tf.Const"() {value = dense<3.000000e-01> : tensor<3x4x2xf32>} : () -> tensor<3x4x2xf32>
  %0 = "tf.XlaCallModule"(%cst, %arg0) {
    Sout = [#tf_type.shape<1x3>], _entry_function = @composite_gather_fn,
    _original_entry_function = "composite_gather_fn", _quantization_method = "weight_only_ptq { }",

  %gather = "tf.TensorListGather"(%tl, %arg1, %elem_shape) : (tensor<!tf_type.variant<tensor<8x9xf32>>>, tensor<3xi32>, tensor<2xi32>) -> tensor<3x8x9xf32>
  // CHECK-NEXT: return %[[GATHER]] : tensor<3x8x9xf32>
  func.return %gather: tensor<3x8x9xf32>
}

// -----

// Test scatter into existing tensor list.

  // expected-error @+1 {{op requires indices operand to have at least 1 dimension}}
  %0 = "tf.TensorScatterUpdate"(%tensor, %indices, %updates) : (tensor<4x4x4xf32>, tensor<i32>, tensor<4x4xf32>) -> tensor<4x4x4xf32>
  func.return %0 : tensor<4x4x4xf32>
}

// -----

func.func @tensor_scatter_update(%tensor: tensor<4x4x4xf32>, %indices: tensor<4x2xi32>, %updates: tensor<f32>) -> tensor<4x4x4xf32> {

}

// -----

// CHECK-LABEL: tensorlistConcat
func.func @tensorlistConcat(%arg0: tensor<3x2x2xf32>, %lead: tensor<i64>) -> (tensor<?x2xf32>, tensor<0xi64>) {
  %cst = arith.constant dense<[2, 2]> : tensor<2xi32>
  %list = "tf.TensorListFromTensor"(%arg0, %cst) : (tensor<3x2x2xf32>, tensor<2xi32>) -> tensor<!tf_type.variant<tensor<f32>>>

func.func @leaky_relu_grad(%arg0: tensor<1x4x4xf32>, %arg1: tensor<1x4x4xf32>) -> tensor<1x4x4xf32> attributes {tf.entry_function = {}} {
    // CHECK-NEXT: %[[ALPHA:.*]] = "chlo.constant_like"(%arg1) <{value = 2.000000e-01 : f32}> : (tensor<1x4x4xf32>) -> tensor<1x4x4xf32>
    // CHECK-NEXT: %[[ZERO:.*]] = "chlo.constant_like"(%arg1) <{value = 0.000000e+00 : f32}> : (tensor<1x4x4xf32>) -> tensor<1x4x4xf32>

  %cst = "tf.Const"() {value = dense<1.000000e+00> : tensor<3x4xf32>} : () -> tensor<3x4xf32>
  %cst_0 = "tf.Const"() {value = dense<0> : tensor<i32>} : () -> tensor<i32>
  %0 = "tf.Abs"(%cst) {device = ""} : (tensor<3x4xf32>) -> tensor<3x4xf32>
  %1 = "tf.Max"(%0, %cst_0) {device = "", keep_dims = false} : (tensor<3x4xf32>, tensor<i32>) -> tensor<4xf32>
  %2 = "tf.Neg"(%1) {device = ""} : (tensor<4xf32>) -> tensor<4xf32>