func.func @serving_default(%arg0: tensor<1x4xf32> {tf_saved_model.index_path = ["x"]}) -> (tensor<1x3xf32> {tf_saved_model.index_path = ["output"]}) attributes {tf.entry_function = {control_outputs = "", inputs = "serving_default_x:0", outputs = "PartitionedCall:0"}, tf_saved_model.exported_names = ["serving_default"]} {

# ==============================================================================

# LINT.IfChange(savedmodel_to_stablehlo)
def savedmodel_to_stablehlo(
    input_path: str,
    exported_model_signatures: list[str] = ["serving_default"],
    tag_names: list[str] = ["serve"],
    input_arg_shapes_str: str = "",
) -> bytes: ...
# LINT.ThenChange()

# LINT.IfChange(tensorflow_module_to_stablehlo)
def tensorflow_module_to_stablehlo(

        exported_model_signatures=["serving_default"],
        tag_names=["serve"],
        input_arg_shapes_str="1,28,28,3::32"
)

```

#### Arguments:

* `input_path` (required): Path to your SavedModel directory.
* `exported_model_signatures` (optional): List of signature names to convert.
                                          Defaults to ["serving_default"].

#include "mlir/IR/BuiltinOps.h"  // from @llvm-project

namespace mlir::quant {

// Returns a public `func::FuncOp` in `module_op` whose name matches either
// `main` or `serving_default`. If `func::FuncOps` with both names exist, the
// function with name "main" takes precedence. Returns null if no such a
// function exists.
func::FuncOp FindMainFuncOp(ModuleOp module_op);

}  // namespace mlir::quant

    tf.quantization.experimental.TfRecordRepresentativeDatasetSaver(
          path_map={'serving_default': '/tmp/representative_dataset_path'}
      ).save({'serving_default': representative_dataset})
  )

  # Using in QuantizationOptions.
  quantization_options = tf.quantization.experimental.QuantizationOptions(
      signature_keys=['serving_default'],
      representative_datasets=dataset_file_map,
  )

      self, output_saved_model_path: str
  ) -> str:
    """Extracts the first XlaCallModule op from output saved model to string."""
    root = load.load(output_saved_model_path)
    tf_graph_def = root.signatures['serving_default'].graph.as_graph_def()
    for function in tf_graph_def.library.function:
      for node_def in function.node_def:
        if node_def.op == 'XlaCallModule':
          with ir.Context() as context:

opt<std::string> exported_model_signatures(
    "exported-model-signatures",
    llvm::cl::desc(
        "Comma-separated list of exported model signatures to convert"),
    llvm::cl::Optional, llvm::cl::init("serving_default"));

// NOLINTNEXTLINE
opt<std::string> tag_names(
    "tags",
    llvm::cl::desc("Comma-separated list of tags for loading SavedModel. "
                   "Ignored for MLIR input"),

    %cst = "tf.Const"() {device = "", value = dense<[[[[-0.315365672, 0.27481091], [0.0901821703, -0.382271349], [-0.105572946, -0.354302853]], [[-0.47703138, -0.307006568], [0.306320101, -0.209111646], [0.252869487, 0.449634969]], [[0.167675957,...

  // within the SavedModel to analyze. If not specified, ["serve"] is used.
  repeated string tags = 5;

  // Sequence of keys identifying SignatureDef containing inputs and outputs.
  // If not specified, ["serving_default"] is used.
  repeated string signature_keys = 6;

  // A map from signature keys to the corresponding representative dataset.
  map<string, RepresentativeDatasetFile> representative_datasets = 7;

// NOLINTNEXTLINE
opt<std::string> exported_model_signatures(
    "exported_model_signatures", llvm::cl::desc("model signature names"),
    llvm::cl::Optional, llvm::cl::init("serving_default"));

namespace mlir {
namespace odml {

absl::StatusOr<OwningOpRef<mlir::ModuleOp>> ImportSavedModelOrMLIR(
    const std::string& input_path, MLIRContext* context,
    llvm::SourceMgr* source_mgr,