func.func @broadcast_mul_implicit_no_fold(%arg0: tensor<5x7xf32>, %arg1: tensor<5xf32>) -> tensor<3x5x7xf32> {
  %cst = arith.constant dense<[3, 5, 7]> : tensor<3xi32>
  %0 = "tf.BroadcastTo"(%arg1, %cst) : (tensor<5xf32>, tensor<3xi32>) -> tensor<3x5x7xf32>
  %1 = "tf.Mul"(%arg0, %0) : (tensor<5x7xf32>, tensor<3x5x7xf32>) -> tensor<3x5x7xf32>
  func.return %1 : tensor<3x5x7xf32>
  // CHECK: %[[C0:.*]] = arith.constant dense<[3, 5, 7]> : tensor<3xi32>

func.func @mean_4d_no_keepdim(%arg0: tensor<1x48x48x512xf32>) -> tensor<1x512xf32> {
  %cst = arith.constant dense<[1, 2]> : tensor<2xi32>
  %0 = "tfl.mean"(%arg0, %cst) {keep_dims = false} : (tensor<1x48x48x512xf32>, tensor<2xi32>) -> tensor<1x512xf32>
  func.return %0 : tensor<1x512xf32>
}

// CHECK:       func @mean_4d_no_keepdim([[VAL_0:%.*]]: tensor<1x48x48x512xf32>) -> tensor<1x512xf32> {

  %flow_1 = "tf.TensorArrayWriteV3"(%handle, %index, %value, %flow) {device = "/job:localhost/replica:0/task:0/device:CPU:0"} : (tensor<2x!tf_type.resource<tensor<1x512xf32>>>, tensor<i32>, tensor<1x512xf32>, tensor<f32>) -> tensor<f32>

// CHECK:           [[LHSBCAST:%.*]] = "mhlo.dynamic_broadcast_in_dim"([[LHS]], [[LHSSHAPEEXTENTS]]) <{broadcast_dimensions = dense<[0, 1, 2]> : tensor<3xi64>}> : (tensor<1x4x2xf32>, tensor<3xindex>) -> tensor<3x4x2xf32>
// CHECK:           [[RHSBCASTSHAPE:%.*]] = shape.concat [[BCASTHEAD]], [[RHSTAIL]]
// CHECK:           [[RESULT:%.*]] = "mhlo.dot_general"([[LHSBCAST]], [[RHS]])
// CHECK:           return [[RESULT]] : tensor<3x4x4xf32>