Exercise 1

Learn how to use tensorflow basic concepts and variables

First start to learn about the graph structure, tensorflow is built upon the nodes



In [9]:

    
# basic imported headers
import tensorflow as tf



In [ ]:

    
Second, learn how to fetch the data from the result



In [10]:

    
input1 = tf.constant(3.0)
input2 = tf.constant(2.0)
input3 = tf.constant(5.0)

intermd = tf.add(input1, input2)
mult = tf.multiply(input3, intermd)

with tf.Session() as sess:
        result = sess.run([mult, intermd])
        print result









    



[25.0, 5.0]



In [3]:

    
# Create a constant op and adds as a node into the default graph
matrix1 = tf.constant([[3., 3.]])

## Pay attention to this wrong one, DIMENSION
# matrix1 = tf.constant([3., 3.])

matrix2 = tf.constant([[2.], [2.]])

product = tf.matmul(matrix1, matrix2)

with tf.Session() as sess:
    print sess.run(product)









    



[[ 12.]]

Next, we are going to show how to feed data as the parameters



In [6]:

    
input1 = tf.placeholder(tf.float32)
input2 = tf.placeholder(tf.float32)
output = tf.multiply(input1, input2)

with tf.Session() as sess:
    print (sess.run([output], feed_dict={input1:[7.], input2:[2.]}))









    



[array([ 14.], dtype=float32)]

At last, we are going to learn how to use variable, unlike the placeholder



In [12]:

    
state = tf.Variable(0, name="counter")
one = tf.constant(1)
new_value = tf.add(state, one)

#define the op, or rule to update/assign value
update = tf.assign(state, new_value)

init_op = tf.global_variables_initializer()

with tf.Session() as sess:
    sess.run(init_op)
    
    # print the initial state of state
    print sess.run(state)
    
    # use loop to output interatively
    for _ in range(3):
        sess.run(update)
        print sess.run(state)



In [ ]: