func.return %1 : tensor<5x2x3x4xf32>
}

// CHECK: %cst = arith.constant dense<1.000000e+00> : tensor<5x2x3x4xf32>
// CHECK: %0 = "tfl.transpose"(%arg0, %arg1) : (tensor<2x3x4x5xf32>, tensor<4xi32>) -> tensor<5x2x3x4xf32>
// CHECK: %1 = tfl.add %0, %cst {fused_activation_function = "NONE"} : tensor<5x2x3x4xf32>
// CHECK: return %1 : tensor<5x2x3x4xf32>

// -----

// CHECK-LABEL: doubleTposeInput

// CHECK-LABEL: transpose_simple_4d
func.func @transpose_simple_4d() -> tensor<5x2x3x4xf32> {
  %0 = stablehlo.constant dense<1.000000e+0> : tensor<2x3x4x5xf32>
  %1 = stablehlo.transpose %0, dims = [3, 0, 1, 2] : (tensor<2x3x4x5xf32>) -> tensor<5x2x3x4xf32>
  return %1 : tensor<5x2x3x4xf32>
}
// CHECK-DAG: %[[CONST_0:.+]] = stablehlo.constant dense<1.000000e+00> : tensor<5x2x3x4xf32>
// CHECK-NOT: transpose
// CHECK: return %[[CONST_0]] : tensor<5x2x3x4xf32>

  %0 = stablehlo.constant() {value = dense<7.000000e+00> : tensor<8x3x3x8xf32>} : () -> tensor<8x3x3x8xf32>
  %2 = stablehlo.convolution(%arg0, %0) dim_numbers = [b, f, 0, 1]x[i, 0, 1, o]->[b, f, 0, 1], window = {pad = [[1, 1], [1, 1]]} {batch_group_count = 1 : i64, feature_group_count = 1 : i64} : (tensor<1x8x4x4xf32>, tensor<8x3x3x8xf32>) -> tensor<1x8x4x4xf32>
  return %2 : tensor<1x8x4x4xf32>
}

  // CHECK: return %[[v2]] : tensor<2x5x3xf32>
}

func.func @einsum_transposereduceddim(%arg0: tensor<2x5x7xf32>, %arg1: tensor<2x5x3x7xf32>) -> tensor<2x5x3xf32> {
  %0 = "tf.Einsum"(%arg0, %arg1) {T = "tfdtype$DT_FLOAT", equation = "bij,binj->bin"}: (tensor<2x5x7xf32>, tensor<2x5x3x7xf32>) -> tensor<2x5x3xf32>
  func.return %0 : tensor<2x5x3xf32>
  // CHECK-LABEL: einsum_transposereduceddim

  %0 = "tf.Gather"(%arg0, %arg1) : (tensor<2x3x6xf32>, tensor<4x5xi32>) -> tensor<4x5x3x6xf32>
  func.return %0 : tensor<4x5x3x6xf32>

// CHECK-LABEL:gatherHigherRankIndices
// CHECK:  "tfl.gather"(%arg0, %arg1) <{axis = 0 : i32, batch_dims = 0 : i32}> : (tensor<2x3x6xf32>, tensor<4x5xi32>) -> tensor<4x5x3x6xf32>
}

// CHECK:           }) : (tensor<5x4x3x7xf32>, tensor<2x2xi32>, tensor<2x5x3xf32>) -> tensor<5x4x3x7xf32>
// CHECK:           return %[[VAL_3]] : tensor<5x4x3x7xf32>
// CHECK:         }
func.func @convert_scatter_update_to_non_trailing_operand_dimensions(
  %arg0: tensor<5x4x3x7xf32>,
  %arg1: tensor<2x2xi32>,
  %arg2: tensor<2x5x3xf32>) -> tensor<5x4x3x7xf32>
{
  %0 = "mhlo.scatter"(%arg0, %arg1, %arg2) ({

  func.return %47 : tensor<*xf32>

  // CHECK-NOT: %[[slice:.*]] = "tfl.slice"
}

// CHECK-LABEL: @StridedSliceToSliceBeginNeg
func.func @StridedSliceToSliceBeginNeg(%arg0: tensor<5x5x5x5xf32>) -> tensor<*xf32> {
  %44 = arith.constant dense<[-5, 0, 0, 0]> : tensor<4xi32>

}

// -----

func.func @unpack(%arg0: tensor<?x2x5x3xf32>) -> () {
  %0:2 = "tfl.unpack"(%arg0) {axis = 1 : i32, num = 2 : i32} : (tensor<?x2x5x3xf32>) -> (tensor<5x5x3xf32>, tensor<5x5x3xf32>)
  func.return
}

// -----

// CHECK-LABEL: testMean
func.func @testMean(%arg0: tensor<2x2xf32>, %arg1 : tensor<1xi32>) -> tensor<1x2xf32> {

  // CHECK: %[[BROADCASTED:.*]] = "mhlo.broadcast_in_dim"(%arg0) <{broadcast_dimensions = dense<[1, 3]> : tensor<2xi64>}> : (tensor<4x8xf32>) -> tensor<7x4x3x8xf32>
  // CHECK: %[[RESULT:.*]] = mhlo.reshape %[[BROADCASTED]] : (tensor<7x4x3x8xf32>) -> tensor<28x24xf32>
  // CHECK: return %[[RESULT]] : tensor<28x24xf32>
  %multiples = "tf.Const"() { value = dense<[7,3]> : tensor<2xi64> } : () -> tensor<2xi64>