%2 = "quantfork.stats"(%1) {layerStats = dense<[5.00000000e-6, 7.00000000e-1]> : tensor<2xf32>} : (tensor<2x2x3xf32>) -> tensor<2x2x3xf32>
    return %2 : tensor<2x2x3xf32>
  }
// CHECK: func.func private @quantize_dot_general_batch_per_tensor_quantized_fn(%[[ARG_0:.+]]: tensor<2x2x2xf32>) -> tensor<2x2x3xf32> attributes {tf._original_func_name = "main_0"}

func.func @testBatchMatMulToV2(%arg0: tensor<2x3x5xf32>, %arg1: tensor<2x5x7xf32>) -> tensor<2x3x7xf32> {
  // CHECK: "tf.BatchMatMulV2"(%arg0, %arg1) <{adj_x = false, adj_y = false, grad_x = false, grad_y = false}> {device = "/job:localhost/replica:0/task:0/device:GPU:0"}
  %0 = "tf.BatchMatMul"(%arg0, %arg1) <{adj_x = false, adj_y = false}> {device = "/job:localhost/replica:0/task:0/device:GPU:0"} : (tensor<2x3x5xf32>, tensor<2x5x7xf32>) -> tensor<2x3x7xf32>

}

// -----

func.func @trivial_dynamic_update_slice_wrong_update_shape(%arg0: tensor<2x7x14xf32>, %arg1: tensor<2x7x7xf32>) -> tensor<2x7x14xf32> {
  %0 = arith.constant dense<0> : tensor<3xi32>
  %1 = "tfl.dynamic_update_slice"(%arg0, %arg1, %0) : (tensor<2x7x14xf32>, tensor<2x7x7xf32>, tensor<3xi32>) -> tensor<2x7x14xf32>
  // CHECK: "tfl.dynamic_update_slice"
  func.return %1 : tensor<2x7x14xf32>
}

  func.func @tf_mod(%arg1: tensor<2x2xf32>) -> tensor<2x2xf32> {
    %cst = "tf.Const"() {value = dense<7.000000e+00> : tensor<f32>} : () -> tensor<f32>
    // CHECK: "mhlo.dynamic_broadcast_in_dim"
    // CHECK: mhlo.remainder
    %6 = "tf.Mod"(%arg1, %cst) {_global_shape = [#tf_type.shape<4x8>], device = ""} : (tensor<2x2xf32>, tensor<f32>) -> tensor<2x2xf32>
    return %6 : tensor<2x2xf32>
  }

  // CHECK-LABEL: func @concat_v2

      %2 = "tf.Cast"(%arg0) {Truncate = false} : (tensor<1x8x2xi32>) -> tensor<1x8x2xf32>
      tf_device.return %2 : tensor<1x8x2xf32>
    // CHECK: () -> tensor<1x8x2xf32>
    }) {device = "/device:CPU:0"} : () -> tensor<*xf32>
    // CHECK: "tf.Cast"(%{{.*}}) <{Truncate = false}> : (tensor<1x8x2xf32>) -> tensor<*xf32>
    // CHECK: (tensor<i32>, tensor<1x8x2xf32>) -> (tensor<1x8x1xf32>, tensor<1x8x1xf32>)

  // CHECK: [[RESHAPE:%.*]] = mhlo.reshape [[SLICE]] : (tensor<2x16x2xf32>) -> tensor<2x16x2xf32>
  %0 = "tf.StridedSlice"(%input, %begin, %end, %strides) {Index = i32, T = f32} : (tensor<10x16x2xf32>, tensor<1xi32>, tensor<1xi32>, tensor<1xi32>) -> tensor<2x16x2xf32>
  // CHECK: return [[RESHAPE]] : tensor<2x16x2xf32>
  func.return %0 : tensor<2x16x2xf32>
}

// -----

  %44 = "tf.AddV2"(%43, %37) {device = ""} : (tensor<2x16x1xf32>, tensor<2x16x1xf32>) -> tensor<2x16x1xf32>
  %45 = "tf.Mul"(%42, %35) {device = ""} : (tensor<2x16x1xf32>, tensor<2x16x1xf32>) -> tensor<2x16x1xf32>
  %46 = "tf.AddV2"(%45, %34) {device = ""} : (tensor<2x16x1xf32>, tensor<2x16x1xf32>) -> tensor<2x16x1xf32>

//   %arg1)
//     : (tensor<i32>, tensor<2x2xi32>, tensor<2xf32>, tensor<2x2x2xf32>)
//     -> tensor<5x2xf32>
//
// is lowered to
//
//   %shape = "tf.Const"() {value = dense<[-1, 2]> : tensor<2xi64>}
//   %inp0 = "tf.Reshape"(%arg0, %shape)
//     : (tensor<2xf32>, tensor<2xi64>) -> tensor<1x2xf32>
//   %inp1 = "tf.Reshape"(%arg1, %shape)
//     : (tensor<2x2x2xf32>, tensor<2xi64>) -> tensor<4x2xf32>

// -----

func.func @testDynamicStitch(%arg0: tensor<2x2xf32>) -> tensor<2x2xf32> {
  %indices = "tf.Const"() {value = dense<[1, 0]> : tensor<2xi32>} : () -> tensor<2xi32>
  %0 = "tf.DynamicStitch"(%indices, %arg0) : (tensor<2xi32>, tensor<2x2xf32>) -> tensor<2x2xf32>
  func.return %0 : tensor<2x2xf32>
}

// -----

func.func @testDynamicStitch() -> tensor<2x2xf32> {

func.func @torch_index_select(%arg0: tensor<2x1xf32>, %arg1: tensor<2xi32>) -> tensor<2x1xf32> {
  %0 = "mhlo.torch_index_select"(%arg0, %arg1) {
    batch_dims = 0 : i64, dim = 0 : i64
  } : (tensor<2x1xf32>, tensor<2xi32>) -> tensor<2x1xf32>
  func.return %0 : tensor<2x1xf32>
}

// CHECK-LABEL:   func @lowered_cumsum(
// CHECK-SAME:      %[[VAL_0:.*]]: tensor<4x12xf32>) -> tensor<4x12xf32> {