func.return %0 : tensor<1024xf32>

  // 1.0 * 2.0 * 512 + 4.0 = 1028.0
  // CHECK: %[[CST:.*]] = arith.constant dense<1.028000e+03> : tensor<1024xf32>
  // CHECK:  return %[[CST]]
}

// CHECK-LABEL: @ConstantFoldFullyConnectedNoBias
func.func @ConstantFoldFullyConnectedNoBias() -> tensor<1024xf32> {

// CHECK-DAG:       [[VAL_20:%.*]] = "tf.Const"() <{value = dense<-1> : tensor<1xi32>}> : () -> tensor<1xi32>
// CHECK:           [[VAL_21:%.*]] = "tf.Reshape"([[VAL_15]]#0, [[VAL_20]]) : (tensor<1x10xf32>, tensor<1xi32>) -> tensor<10xf32>

    func.return %0 : tensor<*xf32>
  }

  // CHECK-LABEL: func.func private @main_00(%arg0: tensor<?x1024xf32>) -> tensor<?x3xf32>
  func.func private @main_00(%arg0: tensor<?x1024xf32>) -> tensor<*xf32> attributes {tf._original_func_name = "main_0"} {
    %cst = "tf.Const"() <{value = dense<1.000000e+00> : tensor<1024x3xf32>}> : () -> tensor<1024x3xf32>

// CHECK:   [[ZERO:%.*]] = arith.constant dense<0> : tensor<i32>
// CHECK:   [[EXP_ITEM:%.*]] = "tf.ExpandDims"([[ITEM]], [[ZERO]]) {{.*}} -> tensor<1x10xf32>
// CHECK:   [[RESULT:%.*]] = "tf.Concat"(%cst, [[INPUT]], [[EXP_ITEM]]) : {{.*}} -> tensor<?x10xf32>
// CHECK:   return [[RESULT]] : tensor<?x10xf32>
}

// -----

  func.return %0 : tensor<2x2xf16>
}

// -----

// CHECK-LABEL: squeeze
func.func @squeeze(%arg0: tensor<1x1x10xf32>) -> tensor<1x10xf32> {
  // CHECK: mhlo.reshape
  %0 = "tf.Squeeze"(%arg0) : (tensor<1x1x10xf32>) -> tensor<1x10xf32>
  func.return %0 : tensor<1x10xf32>
}

// -----

// CHECK-LABEL: squeeze_ranked
func.func @squeeze_ranked(%arg0: tensor<?x?x?xf32>) -> tensor<?xf32> {

  func.return %0 : tensor<16x20x20x13xf32>
}

// -----

func.func @testPReluValidBroadcast(%arg0: tensor<19x7x12x14xf32>, %arg1: tensor<1x1x14xf32>) -> tensor<19x7x12x14xf32> {
  %0 = "tfl.prelu"(%arg0, %arg1) : (tensor<19x7x12x14xf32>, tensor<1x1x14xf32>) -> tensor<19x7x12x14xf32>
  func.return %0 : tensor<19x7x12x14xf32>
}

// -----