// CHECK:           })
// CHECK-SAME:        -> tensor<2x4x6x7xf32>
// CHECK:           %[[RESULT:.*]] = mhlo.divide %[[DIVIDEND]], %[[DIVISOR]] : tensor<2x4x6x7xf32>
// CHECK:           return %[[RESULT]] : tensor<2x4x6x7xf32>
// CHECK:         }
func.func @avgpool_same_padding(%arg0: tensor<2x12x21x7xf32>) -> tensor<2x4x6x7xf32> {

}

func.func @einsum_reshapetail(%arg0: tensor<3x4x5xf32>, %arg1: tensor<5x6x2xf32>) -> tensor<3x4x6x2xf32> {
  %0 = "tf.Einsum"(%arg0, %arg1) {T = "tfdtype$DT_FLOAT", equation = "bfd,dnh->bfnh"}: (tensor<3x4x5xf32>, tensor<5x6x2xf32>) -> tensor<3x4x6x2xf32>
  func.return %0 : tensor<3x4x6x2xf32>
  // CHECK-LABEL: einsum_reshapetail
  // CHECK-DAG: %[[cst:.*]] = arith.constant dense<[5, 12]> : tensor<2xi64>

// CHECK-SAME:      %arg0: tensor<2x4x4x1xf32>,
// CHECK-SAME:      %arg1: tensor<2x4x4x2xf32>) -> tensor<2x4x4x1xf32> attributes {tf._implements = "DenseImageWarp"} {
// CHECK-NEXT:    %0 = "tfl.custom"(%arg0, %arg1) <{custom_code = "DenseImageWarp", custom_option = #tfl<const_bytes : "0x">}> : (tensor<2x4x4x1xf32>, tensor<2x4x4x2xf32>) -> tensor<2x4x4x1xf32>
// CHECK-NEXT:    return %0 : tensor<2x4x4x1xf32>
// CHECK-NEXT:  }
}

// -----

  attributes  {tf.entry_function = {inputs = "args_0,args_1", outputs = "rets_0,rets_1"}} {
    %0:2 = tf_executor.graph {
      %1:3 = tf_executor.island wraps "tf.IdentityN"(%arg0, %arg1) {T = ["tfdtype$DT_FLOAT", "tfdtype$DT_INT32"], device = "", name = "identity"} : (tensor<*xf32>, tensor<2x4x6x8xi32>) -> (tensor<*xf32>, tensor<2x4x6x8xi32>)

  attributes  {tf.entry_function = {inputs = "args_0,args_1", outputs = "rets_0,rets_1"}} {
    %0:2 = tf_executor.graph {
      %1:3 = tf_executor.island wraps "tf.IdentityN"(%arg0, %arg1) {T = ["tfdtype$DT_FLOAT", "tfdtype$DT_INT32"], device = "", name = "identity"} : (tensor<*xf32>, tensor<2x4x6x8xi32>) -> (tensor<*xf32>, tensor<2x4x6x8xi32>)

module attributes {tf.versions = {producer = 946 : i32}, tf_saved_model.semantics, tfl.description = "MLIR Converted.", tfl.schema_version = 3 : i32}  {

// -----

// CHECK-LABEL: doubleTposeInputPermNotEqualNoChange
func.func @doubleTposeInputPermNotEqualNoChange(%arg0: tensor<2x4x3x5xf32>, %arg1: tensor<2x3x4x5xf32>) -> tensor<5x2x3x4xf32> {
  %perm = arith.constant dense<[3, 0, 2, 1]> : tensor<4xi32>
  %0 = "tfl.transpose"(%arg0, %perm) : (tensor<2x4x3x5xf32>, tensor<4xi32>) -> tensor<5x2x3x4xf32>
  %perm1 = arith.constant dense<[3, 0, 1, 2]> : tensor<4xi32>

# Verify arg and ret devices are added as arg and ret attributes.

# CHECK-LABEL: func @main
# CHECK-SAME:  (%[[ARG_0:[a-z0-9]+]]: tensor<*xf32> {tf.device = "/CPU:0"}, %[[ARG_1:[a-z0-9]+]]: tensor<2x4x6x8xi32>) -> (tensor<*xf32>, tensor<*xi32> {tf.device = "/CPU:1"})

node {
  name: "args_0"
  op: "_Arg"
  device: "/CPU:0"
  attr {
    key: "T"
    value {
      type: DT_FLOAT
    }
  }
  attr {

// RUN: tf-mlir-translate -mlir-to-graphdef %s -tf-graph-as-function -o - | FileCheck %s

func.func @main(%arg0: tensor<*x!tf_type.resource>, %arg1: tensor<*x!tf_type.resource<tensor<3x3x1x32xf32>>>, %arg2: tensor<*xf32>, %arg3: tensor<2x4x6x8xi32>) -> (tensor<f32>, tensor<f32>)
attributes {tf.entry_function = {inputs = "args_0,args_1,args_2,args_3", outputs = "rets_0_RetVal,rets_1_RetVal"}} {
  %graph:2 = tf_executor.graph {

    %20 = "tf.Mul"(%19, %cst_0) {device = ""} : (tensor<2x4x3x6xf32>, tensor<f32>) -> tensor<2x4x3x6xf32>
    %21 = "tf.Relu"(%20) {device = ""} : (tensor<2x4x3x6xf32>) -> tensor<2x4x3x6xf32>
    %22 = "tf.Minimum"(%21, %cst) {device = ""} : (tensor<2x4x3x6xf32>, tensor<f32>) -> tensor<2x4x3x6xf32>
    %23 = "tf.Identity"(%22) {device = ""} : (tensor<2x4x3x6xf32>) -> tensor<2x4x3x6xf32>