onnxruntime/server/protobuf/predict.proto

syntax = "proto3";

import "onnx-ml.proto";

package onnxruntime.server;

// PredictRequest specifies how inputs are mapped to tensors
// and how outputs are filtered before returning to user.
message PredictRequest {
  reserved 1;

  // Input Tensors.
  // This is a mapping between output name and tensor.
  map<string, onnx.TensorProto> inputs = 2;

  // Output Filters.
  // This field is to specify which output fields need to be returned.
  // If the list is empty, all outputs will be included.
  repeated string output_filter = 3;
}

// Response for PredictRequest on successful run.
message PredictResponse {
  // Output Tensors.
  // This is a mapping between output name and tensor.
  map<string, onnx.TensorProto> outputs = 1;
}