func.return %2, %3 : tensor<128x16xf32>, tensor<128xi32>
}

// CHECK-LABEL: PruneUnusedLstm
func.func @PruneUnusedLstm(%arg0: tensor<1x28x28xf32>) -> (tensor<1x28x28xf32>) {
    %input = "tfl.quantize"(%arg0) {qtype = tensor<1x28x28x!quant.uniform<i8:f32, 0.003:-128>>} : (tensor<1x28x28xf32>) -> tensor<1x28x28x!quant.uniform<i8:f32, 0.003:-128>>

        tensor<20x20xf32>, none,
        tensor<1x20xf32>, tensor<1x20xf32>,
        none, none, none, none) -> tensor<1x28x20xf32>
    %1 = "quantfork.stats"(%0) {layerStats = dense<[-1.0, 2.0]> : tensor<2xf32>} : (tensor<1x28x20xf32>) -> tensor<1x28x20xf32>
    func.return %1 : tensor<1x28x20xf32>

  // CHECK-NEXT: [[RESULT:%.*]] = "tf.BiasAdd"([[SQUEEZE]], [[BIAS]]) : (tensor<1x128x128xf32>, tensor<128xf32>) -> tensor<1x128x128xf32>
  // CHECK-NEXT: return [[RESULT]] : tensor<1x128x128xf32>
}

func.func @testDilatedDepthWiseConvWithExpandSqueeze1(%arg0: tensor<1x128x128xf32>, %arg1: tensor<5x5x1x1xf32>, %arg2: tensor<128xf32>) -> tensor<1x128x128xf32> {

module {
func.func private @func_20_GPU_FLOAT(%arg0: tensor<128x128xf32>, %arg1: tensor<3xi32>) -> tensor<1x128x128xf32> attributes {tac.device = "GPU", tac.inference_type = "FLOAT", tac.interface_name = "func_20"} {
  %0 = "tfl.reshape"(%arg0, %arg1) {tac.device = "GPU", tac.inference_type = "FLOAT"} : (tensor<128x128xf32>, tensor<3xi32>) -> tensor<1x128x128xf32>
  func.return %0 : tensor<1x128x128xf32>
}

   "tfl.yield"(%4) : (tensor<128x128xf32>) -> ()
  }) {_tfl_quant_trait = "fully_quantizable", device = ""} : (tensor<128x128xf32>, tensor<1xf32>, tensor<1xf32>, tensor<1xi32>) -> tensor<128x128xf32>

   "tfl.yield"(%4) : (tensor<128x128xf32>) -> ()
  }) {_tfl_quant_trait = "fully_quantizable", device = ""} : (tensor<128x128xf32>, tensor<1xf32>, tensor<1xf32>, tensor<1xi32>) -> tensor<128x128xf32>

}

// CHECK-LABEL: test_gather
func.func @test_gather(%arg0 : tensor<1x2x20xf32>, %arg1 : tensor<3x5xi32>) -> (tensor<1x3x5x20xf32>){
    // expected-remark@+1 {{Cost: 1}}
    %0 = "tf.Const"() { value = dense<[1]> : tensor<1xi32> } : () -> tensor<1xi32>
    // expected-remark@+1 {{Cost: 300}}
    %1 = "tf.GatherV2"(%arg0, %arg1, %0) : (tensor<1x2x20xf32>, tensor<3x5xi32>, tensor<1xi32>) -> tensor<1x3x5x20xf32>

func.func @div_dense_different_rank() -> tensor<1x2x2xf32> {
  %cst_0 = arith.constant dense<[[[1.0],[2.0]]]> : tensor<1x2x1xf32>
  %cst_1 = arith.constant dense<[[2.0, 3.0]]> : tensor<1x2xf32>

  %0 = "tfl.div"(%cst_0, %cst_1) {fused_activation_function = "NONE"} : (tensor<1x2x1xf32>, tensor<1x2xf32>) -> tensor<1x2x2xf32>

  func.return %0 : tensor<1x2x2xf32>

  // tf.XlaReduceScatter legalization
  //===--------------------------------------------------------------------===//
  // CHECK-LABEL: func @xla_reduce_scatter
  func.func @xla_reduce_scatter(%arg0: tensor<128x128xf32>) -> tensor<64x128xf32> {
      %cst = "tf.Const"() {value = dense<0> : tensor<i32>} : () -> tensor<i32>
      %cst_0 = "tf.Const"() {value = dense<[[0, 4], [1, 5], [2, 6], [3, 7]]> : tensor<4x2xi32>} : () -> tensor<4x2xi32>