implementation.cc

/*
 * SPDX-License-Identifier: Apache-2.0
 */

#include "onnx/shape_inference/implementation.h"
#include <fstream>
#include <list>
#include "onnx/checker.h"
#include "onnx/defs/data_type_utils.h"
#include "onnx/string_utils.h"

namespace ONNX_NAMESPACE {
namespace shape_inference {
namespace {

std::string GetValueCaseString(const TypeProto& type) {
  switch (type.value_case()) {
    case TypeProto::ValueCase::kTensorType:
      return "tensor_type";
    case TypeProto::ValueCase::kSequenceType:
      return "sequence_type";
    case TypeProto::ValueCase::kMapType:
      return "map_type";
    case TypeProto::ValueCase::kOptionalType:
      return "optional_type";
#ifdef ONNX_ML
    case TypeProto::ValueCase::kOpaqueType:
      return "opaque_type";
#endif
    case TypeProto::ValueCase::kSparseTensorType:
      return "sparse_tensor_type";
    case TypeProto::ValueCase::VALUE_NOT_SET:
      return "NOT_SET";
    default:
      return ONNX_NAMESPACE::to_string(type.value_case());
  }
}

std::string GetElemTypeString(const TypeProto_Tensor& type) {
#ifndef ONNX_USE_LITE_PROTO
  const std::string type_str = TensorProto::DataType_Name(static_cast<TensorProto_DataType>(type.elem_type()));
  if (!type_str.empty()) {
    return type_str;
  }
#endif
  return ONNX_NAMESPACE::to_string(type.elem_type());
}

std::string GetElemTypeString(const TypeProto_SparseTensor& type) {
#ifndef ONNX_USE_LITE_PROTO
  const std::string type_str = TensorProto::DataType_Name(static_cast<TensorProto_DataType>(type.elem_type()));
  if (!type_str.empty()) {
    return type_str;
  }
#endif
  return ONNX_NAMESPACE::to_string(type.elem_type());
}

} // namespace

template<class T>
void CheckTensorShapesAndTypes(const T& inferred_type, const T& existing_type) {
  if (inferred_type.elem_type() != TensorProto::UNDEFINED && existing_type.elem_type() != TensorProto::UNDEFINED &&
      existing_type.elem_type() != inferred_type.elem_type()) {
    std::stringstream ss;
    ss << "Inferred elem type differs from existing elem type: (" << GetElemTypeString(inferred_type) << ") vs ("
       << GetElemTypeString(existing_type) << ")";
    fail_type_inference(ss.str());
  }

  if (!inferred_type.has_shape() || !existing_type.has_shape()) {
    return;
  }

  if (inferred_type.shape().dim_size() != existing_type.shape().dim_size()) {
    std::stringstream ss;
    ss << "Inferred shape and existing shape differ in rank: (" << inferred_type.shape().dim_size() << ") vs ("
       << existing_type.shape().dim_size() << ")";
    fail_shape_inference(ss.str());
  }

  for (int i = 0; i < inferred_type.shape().dim_size(); ++i) {
    const auto& inferred_dim = inferred_type.shape().dim(i);
    const auto& existing_dim = existing_type.shape().dim(i);
    if (inferred_dim.has_dim_value() && existing_dim.has_dim_value() &&
        inferred_dim.dim_value() != existing_dim.dim_value()) {
      std::stringstream ss;
      ss << "Inferred shape and existing shape differ in dimension " << i << ": (" << inferred_dim.dim_value()
         << ") vs (" << existing_dim.dim_value() << ")";
      fail_shape_inference(ss.str());
    }
  }
}

void checkShapesAndTypes(const TypeProto& inferred_type, const TypeProto& existing_type) {
  const auto inferred_value_case = inferred_type.value_case();
  const auto existing_value_case = existing_type.value_case();
  if (inferred_value_case == TypeProto::ValueCase::VALUE_NOT_SET ||
      existing_value_case == TypeProto::ValueCase::VALUE_NOT_SET) {
    // nothing to check; will assign inferredType to undefined existingType
    return;
  }
  if (inferred_value_case != existing_value_case) {
    fail_type_inference(
        "type case mismatch. existing=",
        GetValueCaseString(existing_type),
        " inferred=",
        GetValueCaseString(inferred_type));
  }

  if (inferred_value_case == TypeProto::kTensorType && existing_value_case == TypeProto::kTensorType) {
    CheckTensorShapesAndTypes(inferred_type.tensor_type(), existing_type.tensor_type());
  } else if (inferred_value_case == TypeProto::kSparseTensorType && existing_value_case == TypeProto::kSparseTensorType) {
    CheckTensorShapesAndTypes(inferred_type.sparse_tensor_type(), existing_type.sparse_tensor_type());
  } else if (inferred_value_case == TypeProto::kSequenceType && existing_value_case == TypeProto::kSequenceType) {
    checkShapesAndTypes(inferred_type.sequence_type().elem_type(), existing_type.sequence_type().elem_type());
  } else if (inferred_value_case == TypeProto::kOptionalType && existing_value_case == TypeProto::kOptionalType) {
    checkShapesAndTypes(inferred_type.optional_type().elem_type(), existing_type.optional_type().elem_type());
  } else if (inferred_value_case == TypeProto::TypeProto::kMapType && existing_value_case == TypeProto::TypeProto::kMapType) {
    if (inferred_type.map_type().key_type() != existing_type.map_type().key_type()) {
      fail_type_inference(
          "key type mismatch from MapProto. existing=",
          Utils::DataTypeUtils::ToDataTypeString(existing_type.map_type().key_type()),
          " inferred=",
          Utils::DataTypeUtils::ToDataTypeString(inferred_type.map_type().key_type()));
    }
    checkShapesAndTypes(inferred_type.map_type().value_type(), existing_type.map_type().value_type());
  } else {
    fail_type_inference("type case unsupported. existing=", existing_value_case, " inferred=", inferred_value_case);
  }
}

void mergeShapesAndTypes(const TypeProto_Tensor& inferred_type, TypeProto_Tensor* existing_type) {
  if (existing_type->elem_type() == TensorProto::UNDEFINED) {
    existing_type->set_elem_type(inferred_type.elem_type());
  }

  if (!inferred_type.has_shape()) {
    return;
  }

  if (!existing_type->has_shape()) {
    *existing_type->mutable_shape() = inferred_type.shape();
    return;
  }

  for (int i = 0; i < inferred_type.shape().dim_size(); ++i) {
    const auto& inferred_dim = inferred_type.shape().dim(i);
    auto* existing_dim = existing_type->mutable_shape()->mutable_dim(i);
    if ((!existing_dim->has_dim_value() && !existing_dim->has_dim_param()) ||
        inferred_dim.has_dim_value()) {
      *existing_dim = inferred_dim;
    }
  }
}

void mergeShapesAndTypes(const TypeProto_SparseTensor& inferred_type, TypeProto_SparseTensor* existing_type) {
  if (existing_type->elem_type() == TensorProto::UNDEFINED) {
    existing_type->set_elem_type(inferred_type.elem_type());
  }

  if (!inferred_type.has_shape()) {
    return;
  }

  if (!existing_type->has_shape()) {
    *existing_type->mutable_shape() = inferred_type.shape();
    return;
  }

  for (int i = 0; i < inferred_type.shape().dim_size(); ++i) {
    const auto& inferred_dim = inferred_type.shape().dim(i);
    auto* existing_dim = existing_type->mutable_shape()->mutable_dim(i);
    if ((!existing_dim->has_dim_value() && !existing_dim->has_dim_param()) ||
        inferred_dim.has_dim_value()) {
      *existing_dim = inferred_dim;
    }
  }
}

void mergeShapesAndTypes(const TypeProto& inferred_type, TypeProto* existing_type) {
  // Check before merge
  checkShapesAndTypes(inferred_type, *existing_type);
  const auto inferred_val_case = inferred_type.value_case();
  if (inferred_val_case == TypeProto::kTensorType) {
    mergeShapesAndTypes(inferred_type.tensor_type(), existing_type->mutable_tensor_type());
  } else if (inferred_val_case == TypeProto::kSparseTensorType) {
    mergeShapesAndTypes(inferred_type.sparse_tensor_type(), existing_type->mutable_sparse_tensor_type());
  } else if (inferred_val_case == TypeProto::kSequenceType) {
    mergeShapesAndTypes(
        inferred_type.sequence_type().elem_type(), existing_type->mutable_sequence_type()->mutable_elem_type());
  } else if (inferred_val_case == TypeProto::kOptionalType) {
    mergeShapesAndTypes(
        inferred_type.optional_type().elem_type(), existing_type->mutable_optional_type()->mutable_elem_type());
  } else if (inferred_val_case == TypeProto::kMapType) {
    mergeShapesAndTypes(
      inferred_type.map_type().value_type(), existing_type->mutable_map_type()->mutable_value_type());
  }
}

// TypeProto_Tensor or TypeProto_SparseTensor
template <typename TensorTypeProto>
void GenerateSymbolicShape(TensorTypeProto* inferred_type, SymbolTable& symbol_table) {
  if (!inferred_type->has_shape()) {
    return;
  }
  for (int i = 0; i < inferred_type->shape().dim_size(); ++i) {
    // set a symbol if it doesn't have dim_value and dim_param
    auto* dim = inferred_type->mutable_shape()->mutable_dim(i);
    if (!dim->has_dim_value() && !dim->has_dim_param()) {
      dim->set_dim_param(symbol_table.createNew("unk__"));
    }
  }
}

void MaterializeSymbolicShape(TypeProto* inferred_type, SymbolTable& symbol_table) {
  const auto inferred_val_case = inferred_type->value_case();
  if (inferred_val_case == TypeProto::ValueCase::VALUE_NOT_SET) {
    return;
  }

  if (inferred_val_case == TypeProto::kTensorType) {
    GenerateSymbolicShape(inferred_type->mutable_tensor_type(), symbol_table);
  } else if (inferred_val_case == TypeProto::kSparseTensorType) {
    GenerateSymbolicShape(inferred_type->mutable_sparse_tensor_type(), symbol_table);
  } else if (inferred_val_case == TypeProto::kSequenceType) {
    MaterializeSymbolicShape(inferred_type->mutable_sequence_type()->mutable_elem_type(), symbol_table);
  } else if (inferred_val_case == TypeProto::kOptionalType) {
    MaterializeSymbolicShape(inferred_type->mutable_optional_type()->mutable_elem_type(), symbol_table);
  } else {
    fail_shape_inference("type case unsupported for symbolic shape inference. inferred=", inferred_val_case);
  }
}

std::string GetModelLocalFunctionsMapIdentifier(const std::string& domain, const std::string& func_name) {
  return domain + ":" + func_name;
}

static void InferShapesImpl(
    GraphProto* g,
    const std::unordered_map<std::string, TypeProto*>& outer_scope_value_types_by_name,
    const std::unordered_map<std::string, int>& opset_imports,
    const ShapeInferenceOptions& options,
    SymbolTable* symbol_table,
    const ModelLocalFunctionsMap& model_local_functions_map,
    const ISchemaRegistry* schema_registry = OpSchemaRegistry::Instance(),
    const int ir_version = IR_VERSION // default the latest one
) {
  std::unordered_map<std::string, TypeProto*> value_types_by_name{outer_scope_value_types_by_name};
  std::unordered_map<std::string, TypeProto*> undefined_value_types_by_name{outer_scope_value_types_by_name};
  std::unordered_map<std::string, TensorShapeProto> generated_shape_data_by_name;

  GraphInferenceContext graph_inference_context{
      value_types_by_name, opset_imports, symbol_table, schema_registry, ir_version, model_local_functions_map};
  for (auto& vi : *g->mutable_value_info()) {
    if (vi.has_type()) {
      value_types_by_name[vi.name()] = vi.mutable_type();
    }
  }
  for (auto& vi : *g->mutable_input()) {
    if (vi.has_type()) {
      value_types_by_name[vi.name()] = vi.mutable_type();
    }
  }
  for (auto& vi : *g->mutable_output()) {
    if (vi.has_type()) {
      value_types_by_name[vi.name()] = vi.mutable_type();
    } else {
      // Some output type might be undefined in subgraph. e.g., Loop Op
      // Saving names of outputs with undefined types to allow assigning inferred types to them
      undefined_value_types_by_name[vi.name()] = vi.mutable_type();
    }
  }

  // Holds the contructed type protos for graph initializers
  std::list<TypeProto> initializer_type_list;
  // Create TypeProtos for all graph initializers including sparse initializers
  std::unordered_map<std::string, const TensorProto*> input_data_by_name;
  for (const auto& tp : g->initializer()) {
    input_data_by_name[tp.name()] = &tp;
    TypeProto initializer_type;
    TypeProto_Tensor* initializer_tensor_type = initializer_type.mutable_tensor_type();
    initializer_tensor_type->set_elem_type(tp.data_type());
    // set the shape according to the initializer shape info
    auto* shape = initializer_tensor_type->mutable_shape();
    for (int i = 0; i < tp.dims_size(); ++i) {
      shape->add_dim()->set_dim_value(tp.dims(i));
    }

    auto iter = value_types_by_name.find(tp.name());
    // If it already exists in input, check input and initializer is sync
    // use shape info from input (input has priority over initializer)
    if (iter != value_types_by_name.end()) {
      CheckTensorShapesAndTypes(*initializer_tensor_type, *iter->second->mutable_tensor_type());
    }
    // Support IR>=4: some tensors can only exist in initializer and not in input
    // So shape_inference should make use of initializer shapes
    // Store initializer shape info in value_info as well
    else if (ir_version >= 4) {
      initializer_type_list.push_back(std::move(initializer_type));
      value_types_by_name[tp.name()] = &initializer_type_list.back();
    }
  }

  std::unordered_map<std::string, const SparseTensorProto*> input_sparse_data_by_name;
  for (const auto& tp : g->sparse_initializer()) {
    const auto& name = tp.values().name();
    input_sparse_data_by_name[name] = &tp;
    // Create TypeProto for sparse initializer
    TypeProto initializer_type;
    auto* initializer_sparse_tensor_type = initializer_type.mutable_sparse_tensor_type();
    initializer_sparse_tensor_type->set_elem_type(tp.values().data_type());
    // set the shape according to the initializer shape info
    auto* shape = initializer_sparse_tensor_type->mutable_shape();
    for (int i = 0; i < tp.dims_size(); ++i) {
      shape->add_dim()->set_dim_value(tp.dims(i));
    }

    auto iter = value_types_by_name.find(name);
    // If it already exists in input, check input and initializer is sync
    // use shape info from input (input has priority over initializer)
    if (iter != value_types_by_name.end()) {
      CheckTensorShapesAndTypes(*initializer_sparse_tensor_type, *iter->second->mutable_sparse_tensor_type());
    }
    // Support IR>=4: some tensors can only exist in initializer and not in input
    // So shape_inference should make use of initializer shapes
    // Store initializer shape info in value_info as well
    else if (ir_version >= 4) {
      initializer_type_list.push_back(std::move(initializer_type));
      value_types_by_name[name] = &initializer_type_list.back();
    }
  }

  bool has_experimental_op = false;
  // Collect data from constant nodes and check if any experimental ops exist
  for (const auto& n : g->node()) {
    if (checker::check_is_experimental_op(n.op_type())) {
      has_experimental_op = true;
    } else if (n.op_type() == "Constant" &&  n.output().size() == 1) {
      for (const auto& attr : n.attribute()) {
        if (attr.name() == "value") {
          if (attr.type() == AttributeProto::TENSOR && attr.has_t()) {
            input_data_by_name[n.output(0)] = &attr.t();
          } else if (attr.type() == AttributeProto::SPARSE_TENSOR && attr.has_sparse_tensor()) {
            input_sparse_data_by_name[n.output(0)] = &attr.sparse_tensor();
          }
        }
      }
    }
  }

  std::vector<std::string> inference_errors;
  bool has_unsupported_op = false; // check whether exist unsupported ops
  for (auto& n : *g->mutable_node()) {
    // Resolve domain for node
    auto dit = opset_imports.find(n.domain());
    if (dit == opset_imports.end()) {
      fail_type_inference("Cannot infer type and shape for node name ", n.name(), ". No opset import for domain",
          n.domain(), " optype ", n.op_type());
    }
    auto domain_version = dit->second;
    const auto schema = schema_registry->GetSchema(n.op_type(), domain_version, n.domain());
    InferenceContextImpl ctx(
        n,
        value_types_by_name,
        input_data_by_name,
        input_sparse_data_by_name,
        &generated_shape_data_by_name,
        &graph_inference_context);

    ONNX_TRY {
      if (schema) {
        if (schema->has_type_and_shape_inference_function()) {
          schema->GetTypeAndShapeInferenceFunction()(ctx);
        } else if (schema->HasFunction()) {
          InferShapeForFunctionNode(
              *(schema->GetFunction()),
              schema_registry,
              ctx,
              options,
              model_local_functions_map,
              symbol_table,
              &generated_shape_data_by_name);
        } else {
          // Continue with inference for remaining nodes
          continue;
        }
      } else if (model_local_functions_map.size() > 0) {
        auto iter = model_local_functions_map.find(GetModelLocalFunctionsMapIdentifier(n.domain(), n.op_type()));
        if (iter != model_local_functions_map.end()) {
          InferShapeForFunctionNode(
              *(iter->second),
              schema_registry,
              ctx,
              options,
              model_local_functions_map,
              symbol_table,
              &generated_shape_data_by_name);
        } else {
          has_unsupported_op = true;
          continue;
        }
      } else {
        has_unsupported_op = true;
        continue;
      }
    }
    ONNX_CATCH(const ONNX_NAMESPACE::InferenceError& ex) {
      ONNX_HANDLE_EXCEPTION([&]() {
        // onnx does not support unsupported/experimental operators
        // so it won't consider it as an error
        if (!has_unsupported_op && !has_experimental_op) {
          inference_errors.push_back(GetErrorWithNodeInfo(n, ex));
        }
      });
      // Continue with inference for remaining nodes
      continue;
    }

    ONNX_TRY {
      // check the type-equality for input and output
      if (options.check_type && schema) {
        schema->CheckInputOutputType(ctx);
      }

      for (int i = 0; i < n.output_size(); ++i) {
        // skip type and shape propagation for missing optional outputs.
        if (n.output(i).empty()) {
          continue;
        }
        auto* inferred_type = ctx.getOutputType(i);
        if (inferred_type->value_case() == TypeProto::ValueCase::VALUE_NOT_SET) {
          continue;
        }

        if (symbol_table) {
          MaterializeSymbolicShape(inferred_type, *symbol_table);
        }

        // Find any pre-existing type and shape info. If there is such,
        // then check for compatibility with the inferred
        // information. Otherwise, initialize it in an empty state.
        auto iter = value_types_by_name.find(n.output(i));
        TypeProto* existing_type = nullptr;
        if (iter != value_types_by_name.end()) {
          existing_type = iter->second;
        } else {
          // Create a new value_info if defined type does not exist
          auto vi = g->add_value_info();
          vi->set_name(n.output(i));
          existing_type = vi->mutable_type();
          // For undefined output type, update both value_info and output for now
          // Update existing output with undefined type: assign inferred type to it
          iter = undefined_value_types_by_name.find(n.output(i));
          if (iter != undefined_value_types_by_name.end()) {
            *iter->second = *inferred_type;
          }
        }

        // Now we can merge pre-existing and inferred info
        mergeShapesAndTypes(*inferred_type, existing_type);

        // If data propagation is enabled, propagate shape data if it exists.
        if (options.enable_data_propagation && schema && schema->has_data_propagation_function()) {
          DataPropagationContextImpl data_propagation_ctx(
              n, value_types_by_name, input_data_by_name, generated_shape_data_by_name);
          schema->GetDataPropagationFunction()(data_propagation_ctx);
        }

        // Make merged info available to further inference.
        value_types_by_name[n.output(i)] = existing_type;
      }
    }
    ONNX_CATCH(const std::runtime_error& err) {
      ONNX_HANDLE_EXCEPTION([&]() { fail_shape_inference(GetErrorWithNodeInfo(n, err)); });
    }
  }
  // Throw shape inference error if any. Error mode right now only supports 0 and 1. 
  // When set to 0, any node level shape inference errors are not thrown. This is to support backward compatiblity
  // with 1.7 and earlier releases. When set to 1 it will throw all exceptions.
  // TODO: Add a more granular way for exception handling.
  if (options.error_mode > 0 && !inference_errors.empty()) {
    std::string full_errors = "Shape inference error(s): ";
    for (const std::string& error : inference_errors) {
      full_errors += error + "\n";
    }
    fail_shape_inference(full_errors);
  }
}

void InferShapes(
    GraphProto* g,
    const std::unordered_map<std::string, int>& opset_imports,
    const ISchemaRegistry* schema_registry,
    const ShapeInferenceOptions& options,
    const std::unordered_map<std::string, const FunctionProto*>& model_local_functions) {
  SymbolTableImpl symbol_table;
  TraverseGraphsToAddExistingSymbols(*g, symbol_table);
  InferShapesImpl(
      g,
      std::unordered_map<std::string, TypeProto*>(0),
      opset_imports,
      options,
      &symbol_table,
      model_local_functions,
      schema_registry);
}

void InferShapes(
    ModelProto& m,
    const ISchemaRegistry* schema_registry,
    const ShapeInferenceOptions& options) {
  std::unordered_map<std::string, int> opset_imports;
  for (const auto& opset_import : m.opset_import()) {
    opset_imports[opset_import.domain()] = static_cast<int>(opset_import.version());
  }

  ModelLocalFunctionsMap model_local_functions_by_id;
  for (const auto& function_proto : m.functions()) {
    model_local_functions_by_id.insert(
        {GetModelLocalFunctionsMapIdentifier(function_proto.domain(), function_proto.name()), &function_proto});
  }

  auto* g = m.mutable_graph();
  SymbolTableImpl symbol_table;
  TraverseGraphsToAddExistingSymbols(*g, symbol_table);
  InferShapesImpl(
      g,
      std::unordered_map<std::string, TypeProto*>(0),
      opset_imports,
      options,
      &symbol_table,
      model_local_functions_by_id,
      schema_registry,
      m.ir_version());
}

void InferShapes(
    const std::string& model_path,
    const std::string& save_path,
    const ISchemaRegistry* schema_registry,
    const ShapeInferenceOptions& options) {
  ModelProto model;
  std::fstream model_stream(model_path, std::ios::in | std::ios::binary);
  if (!model_stream.good()) {
    fail_check("Unable to open model file:", model_path, ". Please check if it is a valid file.");
  }
  std::string data{std::istreambuf_iterator<char>{model_stream}, std::istreambuf_iterator<char>{}};
  if (!ParseProtoFromBytes(&model, data.c_str(), data.size())) {
    fail_check(
        "Unable to parse model from file:", model_path, ". Please check if it is a valid protobuf file of model.");
  }
  InferShapes(model, schema_registry, options);
  // Save the inferred model to the original model path
  // Use SerializeToString instead of SerializeToOstream due to LITE_PROTO
  std::fstream output(save_path, std::ios::out | std::ios::trunc | std::ios::binary);
  std::string model_string;
  ONNX_TRY {
    model.SerializeToString(&model_string);
    output << model_string;
  }
  ONNX_CATCH(...) {
    fail_check("Unable to save inferred model to the target path:", save_path);
  }
}

// Infer shape for functions.
void InferShapeForFunctionNode(
    const FunctionProto& func_proto,
    const std::unordered_map<std::string, int>& func_opset_imports,
    const ISchemaRegistry* schema_registry,
    InferenceContext& ctx,
    const ShapeInferenceOptions& options,
    const std::unordered_map<std::string, const FunctionProto*>& model_local_functions_map,
    SymbolTable* symbol_table,
    std::unordered_map<std::string, TensorShapeProto>* generated_shape_data_by_name) {

  if (options.enable_data_propagation && generated_shape_data_by_name == nullptr) {
    fail_shape_inference(
        "Container for generated shape data cannot be nullptr when enable_data_propagation option is set.");
  }

  GraphProto g;
  // Get a temporary tensor-shape map
  const auto num_func_inputs = func_proto.input_size();
  std::unordered_map<std::string, TypeProto*> value_types_by_name;
  std::vector<TypeProto> types_cache(func_proto.input_size());
  for (int i = 0; i < num_func_inputs; ++i) {
    types_cache[i] = *ctx.getInputType(i);
    value_types_by_name[func_proto.input().Get(i)] = &types_cache[i];
  }

  // Create a temporary initializer value map
  std::unordered_map<std::string, const TensorProto*> initializers_by_name;
  std::unordered_map<std::string, const SparseTensorProto*> sparse_initializers_by_name;
  for (int i = 0; i < static_cast<int>(ctx.getNumInputs()) && i < num_func_inputs; ++i) {
    const TypeProto* type = ctx.getInputType(i);
    if (type->value_case() == TypeProto::kTensorType && ctx.getInputData(i) != nullptr) {
      initializers_by_name[func_proto.input().Get(i)] = ctx.getInputData(i);
    } else if (type->value_case() == TypeProto::kSparseTensorType && ctx.getInputSparseData(i) != nullptr) {
      sparse_initializers_by_name[func_proto.input().Get(i)] = ctx.getInputSparseData(i);
    }
  }

  std::unordered_map<std::string, const AttributeProto*> attr_map;
  for (auto& attr : func_proto.attribute()) {
    if (ctx.getAttribute(attr) != nullptr) {
      attr_map[attr] = ctx.getAttribute(attr);
    }
  }

  for (auto& n : func_proto.node()) {
    // Resolve domain for node
    auto it = func_opset_imports.find(n.domain());
    if (it == func_opset_imports.end()) {
      fail_type_inference(
          "Cannot infer type and shape for function",
          func_proto.name(),
          ". No opset import for domain",
          n.domain(),
          " referenced by function body node ",
          n.name(),
          " optype ",
          n.op_type());
    }
    auto domain_version = it->second;
    const auto schema = schema_registry->GetSchema(n.op_type(), domain_version, n.domain());

    NodeProto copy_n(n);
    // Add attribute information into the temporary node
    copy_n.clear_attribute();
    for (const auto& attr : n.attribute()) {
      if (attr.has_ref_attr_name()) {
        if (attr_map.count(attr.ref_attr_name())) {
          auto copy_attr = *attr_map[attr.ref_attr_name()];
          copy_attr.set_name(attr.name());
          copy_n.add_attribute()->CopyFrom(copy_attr);
        }
      } else {
        copy_n.add_attribute()->CopyFrom(attr);
      }
    }
    ONNX_NAMESPACE::shape_inference::InferenceContextImpl func_node_ctx(
        copy_n, value_types_by_name, initializers_by_name, sparse_initializers_by_name, {});

    if (schema && schema->has_type_and_shape_inference_function()) {
      schema->GetTypeAndShapeInferenceFunction()(func_node_ctx);
    } else if (schema && schema->HasFunction()) {
      InferShapeForFunctionNode(
          *(schema->GetFunction()),
          schema_registry,
          func_node_ctx,
          options,
          model_local_functions_map,
          symbol_table,
          generated_shape_data_by_name);
    } else if (model_local_functions_map.size() > 0) {
      // check model local functions for FunctionProto
      auto iter = model_local_functions_map.find(GetModelLocalFunctionsMapIdentifier(n.domain(), n.op_type()));
      if (iter == model_local_functions_map.end()) {
        return;
      }

      InferShapeForFunctionNode(
          *iter->second,
          schema_registry,
          func_node_ctx,
          options,
          model_local_functions_map,
          symbol_table,
          generated_shape_data_by_name);
    } else {
      // Cannot find the function definition in onnx defined schemas and model local functions map, so return.
      return;
    }

    for (int i = 0; i < copy_n.output_size(); ++i) {
      TypeProto* inferred_output_type = func_node_ctx.getOutputType(i);
      // validate and merge the inferred type
      TypeProto* existing_type = nullptr;
      auto iter = value_types_by_name.find(n.output(i));
      if (iter != value_types_by_name.end()) {
        existing_type = iter->second;
        checkShapesAndTypes(*inferred_output_type, *existing_type);
      } else {
        // Store the inferred type info in the temporary subgraph
        auto vi = g.add_value_info();
        vi->set_name(copy_n.output(i));
        existing_type = vi->mutable_type();
      }

      if (symbol_table) {
        MaterializeSymbolicShape(inferred_output_type, *symbol_table);
      }
      mergeShapesAndTypes(*inferred_output_type, existing_type);
      if (options.enable_data_propagation && schema && schema->has_data_propagation_function()) {
        DataPropagationContextImpl data_propagation_ctx(
            copy_n, value_types_by_name, initializers_by_name, *generated_shape_data_by_name);
        schema->GetDataPropagationFunction()(data_propagation_ctx);
      }

      // Make merged info available to downstream inference.
      value_types_by_name[copy_n.output(i)] = existing_type;
    }
  }

  for (int i = 0; i < func_proto.output_size(); ++i) {
    const std::string& output_name = func_proto.output().Get(i);
    // Skip if no type inferred for the tensor
    auto iter = value_types_by_name.find(output_name);
    if (iter != value_types_by_name.cend()) {
      // Copy the type info to ctx
      // to pass back to main graph
      auto type_proto = ctx.getOutputType(i);
      type_proto->CopyFrom(*(iter->second));
    }
  }
}

void InferShapeForFunctionNode(
    const FunctionProto& function_proto,
    const ISchemaRegistry* schema_registry,
    InferenceContext& ctx,
    const ShapeInferenceOptions& options,
    const std::unordered_map<std::string, const FunctionProto*>& model_local_functions_map,
    SymbolTable* symbol_table,
    std::unordered_map<std::string, TensorShapeProto>* generated_shape_data_by_name) {

  std::unordered_map<std::string, int> opset_imports;
  for (const auto& opset_import : function_proto.opset_import()) {
    opset_imports[opset_import.domain()] = static_cast<int>(opset_import.version());
  }

  InferShapeForFunctionNode(
      function_proto,
      opset_imports,
      schema_registry,
      ctx,
      options,
      model_local_functions_map,
      symbol_table,
      generated_shape_data_by_name);
}

std::vector<const TypeProto*> GraphInferencerImpl::doInferencing(
    const std::vector<const TypeProto*>& input_types,
    const std::vector<const TensorProto*>& input_data) {
  SymbolTable* symbol_table = getSymbolTable();
  int num_inputs = int(input_types.size());
  std::unordered_set<std::string> initializer_name_set;
  for (const auto& tp : g_->initializer()) {
    initializer_name_set.insert(tp.name());
  }

  if (context_->ir_version >= 4) {
    if (g_->input_size() != num_inputs) {
      fail_shape_inference("Graph has ", g_->input_size(), " inputs but ", num_inputs, " were provided");
    }
    for (int i = 0; i < g_->input_size(); ++i) {
      if (initializer_name_set.count(g_->input(i).name()) > 0) {
        fail_shape_inference("Cannot use the same name as both a subgraph initializer and subgraph input: ",
          g_->input(i).name());
      }
    }
  } else {
    // IR < 4 requires all initializers to be optional inputs
    // So the number of graph input can be larger than the number of node input 
    if (num_inputs > g_->input_size()) {
      fail_shape_inference(
          "Graph has ",
          g_->input_size(),
          " inputs but ",
          num_inputs,
          " were provided.",
        "The number of graph input cannot be smaller than the number of node input" );
    } else if (num_inputs < g_->input_size()) {
      for (int i = 0; i < g_->input_size(); ++i) {
        if (i < num_inputs && initializer_name_set.count(g_->input(i).name()) > 0) {
          fail_shape_inference("Graph initializer names must appear after the actual inputs: ",
            g_->input(i).name());
        } else if (i >= num_inputs && initializer_name_set.count(g_->input(i).name()) == 0) {
          // Further check whether the additional input is in initializers
          fail_shape_inference("Cannot find missing input: ", g_->input(i).name(), "in initializers. ");
        }
      }
    }
  }

  for (int i = 0, end = num_inputs; i < end; ++i) {
    const TypeProto* inferred_input = input_types[i];

    if (!inferred_input)
      continue;

    TypeProto* graph_input = g_->mutable_input(i)->mutable_type();
    // Even if graphInput doesn't have defined type, it will assign inferredType to it
    mergeShapesAndTypes(*inferred_input, graph_input);

    if (symbol_table) {
      MaterializeSymbolicShape(graph_input, *symbol_table);
    }
  }

  // future: pass inputData into InferShapes either directly, or indirectly by
  // updating initializers that match subgraph inputs.
  (void)input_data;
  ShapeInferenceOptions options {};
  InferShapesImpl(
      g_,
      *context_->outer_scope_value_types_by_name, // never null
      context_->opset_imports,
      options,
      symbol_table,
      context_->model_local_functions,
      context_->schema_registry);

  std::vector<const TypeProto*> graph_output_types;
  graph_output_types.reserve(g_->output().size());
  for (const ValueInfoProto& output : g_->output()) {
    graph_output_types.push_back(&output.type());
  }

  return graph_output_types;
}

std::string GetErrorWithNodeInfo(NodeProto n, std::runtime_error err) {
  std::string op_name = n.has_name() ? (", node name: " + n.name()) : "";
  return "(op_type:" + n.op_type() + op_name + "): " + err.what();
}

void TraverseGraphsToAddExistingSymbols(const GraphProto& g, SymbolTable& symbol_table) {
  symbol_table.addFromGraph(g);
  for (const auto& n : g.node()) {
    for (auto& attr : n.attribute()) {
      if (attr.has_g()) {
        TraverseGraphsToAddExistingSymbols(attr.g(), symbol_table);
      }
    }
  }
}

} // namespace shape_inference
} // namespace ONNX_NAMESPACE