Tensorflow: Basics -I

Let us start with some basics of addition and multiplication program and represent them in a tensorboard.

import tensorflow as tf

x=tf.constant(1,name='x')
m=tf.constant(2,name='y')
z=tf.constant(3,name='z')
prod=tf.multiply(x,m,name='multiply')
add=tf.add(prod,z,name='add')

with tf.Session() as sess:
writer=tf.summary.FileWriter(logdir='./graph',graph=sess.graph)
print(sess.run(add))

Output is :
5

Now, if we want to check the graph, we have to run following command in Linux(Note: I have used Kali linux), read different types of tensor in

Next Step is to create a scope in tensorflow, this helps in maintaining the code complexity.

#Add Variables
a=tf.constant(2,name='a')
b=tf.constant(2,name='b')
c=tf.constant(3,name='c')
d=tf.constant(4,name='d')

with tf.name_scope("Addition"):
with tf.name_scope("Addition1"):
add1=tf.add(a,b,name='Addition')
with tf.name_scope("Addition2"):
add2=tf.add(c,d,name='Addition')
with tf.name_scope("Multiplication"):
mul=tf.multiply(add1,add2,name="multiply")
#Run the graph
with tf.Session() as sess1:
writer= tf.summary.FileWriter(logdir='./graph1',graph=sess1.graph)
print(sess1.run(mul))
tf.reset_default_graph()

If we want to check in detail:

Leave a Reply

Your email address will not be published. Required fields are marked *