protocol buffers - connect input and output tensors of two different graphs tensorflow

Question

I have 2 ProtoBuf Files, I currently load and forward pass each of them separately, by calling-

out1=session.run(graph1out, feed_dict={graph1inp:inp1})

followed by

final=session.run(graph2out, feed_dict={graph2inp:out1})

where graph1inp and graph1out are input node and output node of graph 1 and similar terminology for graph 2

Now, I want to connect graph1out with graph2inp such that I only have to run graph2out while feeding graph1inp with inp1. In other words connecting the input and output tensors of the 2 involved graphs in such a way that one run is sufficient to run inference on both trained ProtoBuf files.

Answer 1

Assuming that your Protobuf files contain serialized tf.GraphDef protos, you can use the input_map argument of tf.import_graph_def() to connect the two graphs:

# Import graph1.
graph1_def = ...  # tf.GraphDef object
out1_name = "..."  # name of the graph1out tensor in graph1_def.
graph1out, = tf.import_graph_def(graph1_def, return_elements=[out_name])

# Import graph2 and connect it to graph1.
graph2_def = ...  # tf.GraphDef object
inp2_name = "..."  # name of the graph2inp tensor in graph2_def.
out2_name = "..."  # name of the graph2out tensor in graph2_def.
graph2out, = tf.import_graph_def(graph2_def, input_map={inp2_name: graph1out},
                                 return_elements=[out2_name])