ParseReduce.cxx

Go to the documentation of this file.

#include "TMVA/RModelParser_ONNX.hxx"
#include "TMVA/ROperator_Reduce.hxx"
#include "onnx.hxx"
#include <stdexcept>
 
namespace TMVA {
namespace Experimental {
namespace SOFIE {
 
template <EReduceOpMode Op>
std::unique_ptr<ROperator> ParseReduce(RModelParser_ONNX &parser, const onnx::NodeProto &nodeproto)
{
   ETensorType input_type;
 
   EReduceOpMode op_mode = InvalidReduceOp;
 
   if (nodeproto.op_type() == "ReduceMean")
      op_mode = ReduceMean;
   else if (nodeproto.op_type() == "ReduceSumSquare")
      op_mode = ReduceSumSquare;
   else if (nodeproto.op_type() == "ReduceProd")
      op_mode = ReduceProd;
   else if (nodeproto.op_type() == "ReduceSum")
      op_mode = ReduceSum;
   else if (nodeproto.op_type() == "ReduceMax")
      op_mode = ReduceMax;
   else if (nodeproto.op_type() == "ReduceMin")
      op_mode = ReduceMin;
 
   if (op_mode == InvalidReduceOp) {
      throw std::runtime_error("TMVA::SOFIE - Reduce op mode not supported.");
   }
 
   auto input_name = nodeproto.input(0);
   if (parser.IsRegisteredTensorType(input_name)) {
      input_type = parser.GetTensorType(input_name);
   } else {
      throw std::runtime_error("TMVA::SOFIE ONNX Parser Reduce  op has input tensor" + input_name +
                               " but its type is not yet registered");
   }
   //in the latest version of ONNX axis is not an attribute but an input
   std::string axes_name;
   if (nodeproto.input_size() > 1) {
      axes_name = nodeproto.input(1);
      if (!parser.IsRegisteredTensorType(axes_name)) {
         throw std::runtime_error("TMVA::SOFIE ONNX Parser Reduce  op has input tensor" + axes_name +
                               " but its type is not yet registered");
      }
   }
 
   std::unique_ptr<ROperator> op;
   std::string output_name = nodeproto.output(0);
   int attr_keepdims = 1;
   std::vector<int64_t> attr_axes;
   for (int_t i = 0; i < nodeproto.attribute_size(); i++) {
      std::string attribute_name = nodeproto.attribute(i).name();
      if (attribute_name == "keepdims")
         attr_keepdims = nodeproto.attribute(i).i();
      if (attribute_name == "axes") {
         attr_axes =
            std::vector<int64_t>({nodeproto.attribute(i).ints().begin(), nodeproto.attribute(i).ints().end()});
      }
   }
   op.reset(new ROperator_Reduce<Op>(attr_keepdims, attr_axes, input_name, axes_name, output_name));
   // switch (input_type) {
   // case ETensorType::FLOAT:
   //    op.reset(new ROperator_Reduce<float, Op>(attr_keepdims, attr_axes, input_name, axes_name, output_name));
   //    break;
   // default:
   //    throw std::runtime_error("TMVA::SOFIE - Unsupported - Reduce Operator does not yet support input type " +
   //                             std::to_string(static_cast<int>(input_type)));
   // }
 
   if (!parser.IsRegisteredTensorType(output_name)) {
      parser.RegisterTensorType(output_name, input_type);
   }
   return op;
}
 
void RegisterReduceParsers(RModelParser_ONNX &parser)
{
   parser.RegisterOperator("ReduceMean", ParseReduce<EReduceOpMode::ReduceMean>);
   parser.RegisterOperator("ReduceSumSquare", ParseReduce<EReduceOpMode::ReduceSumSquare>);
   parser.RegisterOperator("ReduceProd", ParseReduce<EReduceOpMode::ReduceProd>);
   parser.RegisterOperator("ReduceSum", ParseReduce<EReduceOpMode::ReduceSum>);
   parser.RegisterOperator("ReduceMax", ParseReduce<EReduceOpMode::ReduceMax>);
   parser.RegisterOperator("ReduceMin", ParseReduce<EReduceOpMode::ReduceMin>);
}
 
} // namespace SOFIE
} // namespace Experimental
} // namespace TMVA