How can you identify Input and Output name in tensorboard graph like this one in the pictures attached to this post?

Question

I used the MobileNet_v1_1.0_224 tensorflow model for object detection. Now, I have my custom frozen graph (.pb file) that I need to convert to tflite extension so that I can use my model for mobile devices.

Can someone help me to identify the input and the output names in this tensorboard graph? I need them to use as input and output parameters to convert my frozen graph (.pb file) to tensorflow lite (.tflite) file

graph from tensorboard

same graph

Answer 1

You can use this code:

import tensorflow as tf
gf = tf.GraphDef()   
m_file = open('frozen_inference_graph.pb','rb')
gf.ParseFromString(m_file.read())

with open('somefile.txt', 'a') as the_file:
    for n in gf.node:
        the_file.write(n.name+'\n')

file = open('somefile.txt','r')
data = file.readlines()
print ("\noutput name = ")
print (data[len(data)-1])

print ("Input name = ")
file.seek ( 0 )
print (file.readline())

In my case I had

output name: SemanticPredictions
input name: ImageTensor