func.func @batchmatmulv2_basic(%arg0: tensor<1x4x2xf32>, %arg1: tensor<3x2x4xf32>) -> tensor<3x4x4xf32> {
// CHECK-LABEL:   func @batchmatmulv2_basic
// CHECK-SAME:        ([[LHS:%.*]]: tensor<1x4x2xf32>, [[RHS:%.*]]: tensor<3x2x4xf32>) -> tensor<3x4x4xf32>
// CHECK:           [[LHSSHAPE:%.*]] = shape.shape_of [[LHS]] : tensor<1x4x2xf32>
// CHECK:           [[RHSSHAPE:%.*]] = shape.shape_of [[RHS]] : tensor<3x2x4xf32>
// CHECK:           [[CM2:%.*]] = arith.constant -2 : index

  // CHECK-LABEL: test_batch_matmul_to_einsum
  // CHECK: "tf.Einsum"(%arg0, %arg1) <{equation = "...mk,...kn->...mn"}> : (tensor<1x2x3xf32>, tensor<3x4xf32>) -> tensor<1x2x4xf32>
  %0 = "tf.BatchMatMul"(%arg0, %arg1) {adj_x = false, adj_y = false} : (tensor<1x2x3xf32>, tensor<3x4xf32>) -> tensor<1x2x4xf32>
  func.return %0: tensor<1x2x4xf32>
}

  %0 = "tf.BatchMatMulV2"(%arg0, %arg1) {T = f32, adj_x = false, adj_y = false, grad_x = false, grad_y = false, device = ""} : (tensor<1x4x2xf32>, tensor<3x2x4xf32>) -> tensor<3x4x4xf32>
  func.return %0 : tensor<3x4x4xf32>
}

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>

# CHECK:           %[[VAL_9:.*]] = "tfl.transpose"(%[[VAL_1]], %[[VAL_2]]) : (tensor<3x7xf32>, tensor<2xi32>) -> tensor<7x3xf32>
# CHECK:           %[[VAL_10:.*]] = "tfl.fully_connected"(%[[VAL_7]]#0, %[[VAL_9]], %[[VAL_3]]) <{fused_activation_function = "NONE", keep_num_dims = false, weights_format = "DEFAULT"}> : (tensor<1x5x3xf32>, tensor<7x3xf32>, none) -> tensor<5x7xf32>