%b = arith.constant dense<0.0> : tensor<16xf32>
  %conv_3d = "tfl.conv_3d"(%arg0, %w, %b) {dilation_d_factor = 1 : i32, dilation_h_factor = 1 : i32, dilation_w_factor = 1 : i32, fused_activation_function = "NONE", padding = "SAME", stride_d = 1 : i32, stride_h = 1 : i32, stride_w = 1 : i32} : (tensor<1x32x32x32x8xf32>, tensor<1x1x1x8x16xf32>, tensor<16xf32>) -> tensor<1x32x32x32x16xf32>
  func.return %conv_3d : tensor<1x32x32x32x16xf32>

// CHECK:           %[[VAL_4:.*]]:2 = call @func_0_GPU_FLOAT(%[[VAL_0]], %[[VAL_1]], %[[VAL_2]], %[[VAL_3]]) {tac.device = "GPU", tac.inference_type = "FLOAT", tac.interface_name = "func_0"} : (tensor<1xf32>, tensor<1xf32>, tensor<1xf32>, tensor<1xf32>) -> (tensor<1xf32>, tensor<1xf32>)

// CHECK-LABEL: div
// CHECK:  tfl.div %arg0, %arg1 {fused_activation_function = "NONE"} : tensor<1xf32>
// CHECK:  return
}

func.func @squaredDifferenceRelu(tensor<1xf32>, tensor<1xf32>) -> tensor<1xf32> {
^bb0(%arg0: tensor<1xf32>, %arg1: tensor<1xf32>):
  %0 = "tf.SquaredDifference"(%arg0, %arg1) : (tensor<1xf32>, tensor<1xf32>) -> tensor<1xf32>
  %1 = "tf.Relu6"(%0) : (tensor<1xf32>) -> tensor<1xf32>

        tensor<20xf32>, tensor<20xf32>, tensor<20xf32>,
        tensor<20xf32>, tensor<20xf32>, tensor<20xf32>, tensor<20xf32>,
        tensor<20x20xf32>, none,
        tensor<1x20xf32>, tensor<1x20xf32>,
        none, none, none, none) -> tensor<1x28x20xf32>
    %1 = "quantfork.stats"(%0) {layerStats = dense<[-1.0, 2.0]> : tensor<2xf32>} : (tensor<1x28x20xf32>) -> tensor<1x28x20xf32>

  func.return %0#0 : tensor<?x?x?x?xf32>
}

// -----

// CHECK-LABEL: fusedBatchNormV3_training_dynamic_unsupported1

func.func @tf2xla_fallback_op_approx_top_k(%arg0: tensor<16xf32>) -> (tensor<?xf32>, tensor<?xi32>) {
  %0:2 = "tf_device.cluster"() ({
    // CHECK: tf.ApproxTopK
    // CHECK-NOT: _xla_outside_compilation
    %1:2 = "tf.ApproxTopK"(%arg0) {k = 2} : (tensor<16xf32>) -> (tensor<?xf32>, tensor<?xi32>)
    tf_device.return %1#0, %1#1 : tensor<?xf32>, tensor<?xi32>

func.func @testCompatibleCastType(%arg0: tensor<?xf32>) -> (tensor<10xf32>, tensor<10xf32>) {
  %0 = "tf.Cast"(%arg0) {Truncate = false} : (tensor<?xf32>) -> tensor<10xf32>
  %1 = "tf.Cast"(%arg0) {Truncate = true} : (tensor<?xf32>) -> tensor<10xf32>
  func.return %0, %1: tensor<10xf32>, tensor<10xf32>

  // CHECK: %0 = "tf.Cast"(%arg0) <{Truncate = false}> : (tensor<?xf32>) -> tensor<10xf32>

// CHECK:           %[[VAL_3:.*]] = "tf.Atan2"(%[[VAL_1]], %[[VAL_0]]) : (tensor<4xf32>, tensor<1xf32>) -> tensor<4xf32>
// CHECK:           return %[[VAL_2]], %[[VAL_3]] : tensor<4xf32>, tensor<4xf32>
// CHECK:         }
func.func @broadcast_atan2(%arg0: tensor<1xf32>, %arg1: tensor<4xf32>) -> (tensor<4xf32>, tensor<4xf32>) {

// CHECK:           }) : (tensor<1x?xf32>, tensor<1x0xf32>, tensor<1x0xf32>, tensor<1x0xf32>, tensor<1x0xf32>, tensor<1x3xf32>, tensor<1x3xf32>, tensor<1x3xf32>, tensor<1x3xf32>, none, none, none, tensor<1xf32>, tensor<1xf32>, tensor<1xf32>, tensor<1xf32>, tensor<3x1xf32>, tensor<3xf32>, tensor<1x3xf32>, tensor<1x1xf32>, tensor<1xf32>, tensor<1xf32>, tensor<1xf32>, tensor<1xf32>) -> tensor<1x3xf32>

    // CHECK: %[[MUL:.*]] = "tf.Mul"{{.*}} (tensor<1xf32>, tensor<1xf32>) -> tensor<1xf32>
    // CHECK: %[[UNKNOWN:.*]] = "tf.Unknown"(%[[MUL]], %[[MUL]]) : (tensor<1xf32>, tensor<1xf32>) -> tensor<*xf32>
    // CHECK: return %[[UNKNOWN]] : tensor<*xf32>
    %0 = "tf.Mul"(%arg0, %arg0) : (tensor<1xf32>, tensor<1xf32>) -> tensor<*xf32>
    %1 = "tf.Unknown"(%0, %0) : (tensor<*xf32>, tensor<*xf32>) -> tensor<*xf32>