First TensorFlow Graphs

In this notebook, we execute elementary TensorFlow computational graphs.

Load dependencies



In [1]:

    
import numpy as np
import tensorflow as tf

Simple arithmetic



In [2]:

    
x1 = tf.placeholder(tf.float32)
x2 = tf.placeholder(tf.float32)



In [3]:

    
sum_op = tf.add(x1, x2)
product_op = tf.multiply(x1, x2)



In [4]:

    
with tf.Session() as session:
    sum_result = session.run(sum_op, feed_dict={x1: 2.0, x2: 0.5}) # run again with {x1: [2.0, 2.0, 2.0], x2: [0.5, 1.0, 2.0]}
    product_result = session.run(product_op, feed_dict={x1: 2.0, x2: 0.5}) # ...and with {x1: [2.0, 4.0], x2: 0.5}



In [5]:

    
sum_result









    Out[5]:





2.5



In [6]:

    
product_result









    Out[6]:





1.0

Simple array arithmetic



In [7]:

    
with tf.Session() as session:
    sum_result = session.run(sum_op, feed_dict={x1: [2.0, 2.0, 2.0], x2: [0.5, 1.0, 2.0]})
    product_result = session.run(product_op, feed_dict={x1: [2.0, 4.0], x2: 0.5})



In [8]:

    
sum_result









    Out[8]:





array([ 2.5,  3. ,  4. ], dtype=float32)



In [9]:

    
product_result









    Out[9]:





array([ 1.,  2.], dtype=float32)



In [ ]: