6.1.17 Index

TensorFlow's public C++ API includes only the API for executing graphs, as of version 0.5. To control the execution of a graph from C++:

1. Build the computation graph using the Python API.
2. Use tf.train.write_graph() to write the graph to a file.

3. Load the graph using the C++ Session API. For example:

   // Reads a model graph definition from disk, and creates a session object you
   // can use to run it.
   Status LoadGraph(string graph_file_name, Session** session) {
    GraphDef graph_def;
    TF_RETURN_IF_ERROR(
        ReadBinaryProto(Env::Default(), graph_file_name, &graph_def));
    TF_RETURN_IF_ERROR(NewSession(SessionOptions(), session));
    TF_RETURN_IF_ERROR((*session)->Create(graph_def));
    return Status::OK();
   }
   

4. Run the graph with a call to session->Run()

Classes

• tensorflow::Env
• tensorflow::EnvWrapper
• tensorflow::RandomAccessFile
• tensorflow::Session
• tensorflow::Status
• tensorflow::Tensor
• tensorflow::TensorBuffer
• tensorflow::TensorShape
• tensorflow::TensorShapeIter
• tensorflow::TensorShapeUtils
• tensorflow::Thread
• tensorflow::WritableFile

Structs

• tensorflow::SessionOptions
• tensorflow::Status::State
• tensorflow::TensorShapeDim
• tensorflow::ThreadOptions