AbstractTensorHandle* loss;
  AbstractTensorHandle* backprop;
  TF_RETURN_IF_ERROR(ops::SparseSoftmaxCrossEntropyWithLogits(
      ctx, inputs[0], inputs[1], &loss, &backprop,
      "SparseSoftmaxCrossEntropyWithLogits"));
  // `gradient_checker` only works with model that returns only 1 tensor.
  // Although, `ops::SparseSoftmaxCrossEntropyWithLogits` returns 2 tensors, the
  // second tensor isn't needed for computing gradient so we could safely drop

  TF_RETURN_IF_ERROR(op_ptr->AddInputList(input));
  int num_retvals = output.size();
  return op_ptr->Execute(output, &num_retvals);
}

// Op: SparseSoftmaxCrossEntropyWithLogits()
// Summary:
//
// Description:

//
Status IdentityN(AbstractContext* ctx, absl::Span<AbstractTensorHandle* const> input, absl::Span<AbstractTensorHandle*> output, const char* name = nullptr, const char* raw_device_name = nullptr);

//
Status SparseSoftmaxCrossEntropyWithLogits(AbstractContext* ctx, AbstractTensorHandle* const features, AbstractTensorHandle* const labels, AbstractTensorHandle** loss, AbstractTensorHandle** backprop, const char* name = nullptr, const char* raw_device_name = nullptr);

 MatMul 3
 MaxPool 1
 MaxPoolGrad 1
 Mean 1
 Mul 164
 Pad 1
 ReadVariableOp 646
 Relu 49
 ReluGrad 49
 Reshape 2
 ResourceApplyKerasMomentum 161
 ShapeN 50
 Softmax 1
 SparseSoftmaxCrossEntropyWithLogits 1
 Square 55
 Squeeze 1
 Sub 106
 Sum 57
 Tile 1

 MatMul 3
 MaxPool 1
 MaxPoolGrad 1
 Mean 1
 Mul 218
 Pad 2
 ReadVariableOp 538
 Relu 49
 ReluGrad 49
 Reshape 2
 ResourceApplyKerasMomentum 161
 Slice 1
 Softmax 1
 SparseSoftmaxCrossEntropyWithLogits 1
 Squeeze 1
 Sum 1
 Tile 1
 Transpose 1
cluster 1 size 815
 AddN 1
 AssignAddVariableOp 1
 AssignSubVariableOp 106
 Const 220
 DivNoNan 1
 Identity 1
 Mul 161
 ReadVariableOp 106

  func.return %0#0, %0#1 : tensor<?xf32>, tensor<?x?xf32>
}

// CHECK-LABEL: SparseSoftmaxCrossEntropyWithLogits
// CHECK-SAME: %[[FEATURES:.*]]: tensor<2x3xf32>, %[[SPARSE_LABELS:.*]]: tensor<2xi32>
func.func @SparseSoftmaxCrossEntropyWithLogits(%features: tensor<2x3xf32>, %labels: tensor<2xi32>) -> (tensor<2xf32>, tensor<2x3xf32>) {
  // Convert SPARSE_LABELS to dense LABELS.

 ConcatV2 1
 Const 30
 ExpandDims 3
 Less 1
 Mean 2
 Minimum 1
 Mul 3
 Pack 2
 Pad 1
 Range 1
 RealDiv 1
 Reshape 8
 Shape 7
 Size 1
 Slice 2
 Snapshot 5
 SparseSoftmaxCrossEntropyWithLogits 1
 StridedSlice 2
 Sub 3
 Sum 1
 Tile 1
 Transpose 2
 Unpack 1
cluster 2 size 44
 Cast 2
 ConcatV2 2
 Const 18
 ExpandDims 1
 GatherV2 2
 Less 1
 MatMul 1

    %19 = "tf.Reshape"(%18, %7) : (tensor<2x1001xf32>, tensor<2xi32>) -> tensor<2x1001xf32>
    %loss, %backprop = "tf.SparseSoftmaxCrossEntropyWithLogits"(%19, %13) : (tensor<2x1001xf32>, tensor<2xi64>) -> (tensor<2xf32>, tensor<2x1001xf32>)
    %20 = "tf.Sum"(%loss, %6) {keep_dims = false} : (tensor<2xf32>, tensor<1xi32>) -> tensor<f32>

}

// -----

//===--------------------------------------------------------------------===//
//  tf.SparseSoftmaxCrossEntropyWithLogits
//===--------------------------------------------------------------------===//

// Test valid tf.SparseSoftmaxCrossEntropyWithLogits
// CHECK-LABEL: func @testSparseSoftmaxCrossEntropyWithLogits

          // clang-format on
          {"PWRED",
           {"ArgMax", "ArgMin", "DiagPart", "Softmax",
            "SparseSoftmaxCrossEntropyWithLogits", "LogSoftmax"}},
          {"REDUCEWINDOW",
           {"ArgMax", "ArgMin", "DiagPart", "Softmax",
            "SparseSoftmaxCrossEntropyWithLogits", "LogSoftmax"}},
          {"REDUCEWINDOWPW", {"BiasAddGrad", "LRN", "LRNGrad"}},
          {"BN",