package onnxruntime.server

Get desktop application:
View/edit binary Protocol Buffers messages

PredictRequest specifies how inputs are mapped to tensors and how outputs are filtered before returning to user.

• map<string, onnx.TensorProto> inputs = 2

  Input Tensors. This is a mapping between output name and tensor.

• repeated string output_filter = 3

  Output Filters. This field is to specify which output fields need to be returned. If the list is empty, all outputs will be included.

Response for PredictRequest on successful run.

• map<string, onnx.TensorProto> outputs = 1

  Output Tensors. This is a mapping between output name and tensor.