Variables and Placeholders

     Graphs are sets of connected nodes - vertices

The connections are referred to as edges

each node is an operation with possible inputs that can supply some output

In [1]:

import tensorflow as tf

In [2]:

n1 = tf.constant(1)

In [3]:

n2 = tf.constant(2)

In [4]:

n3 = n1 + n2

In [5]:

with tf.Session() as sess:
    result = sess.run(n3)

In [6]:

print (result)

3

n3 은 여기서 뭘까 -> Tensor 이다

In [7]:

print (tf.get_default_graph())

<tensorflow.python.framework.ops.Graph object at 0xb30d53b00>

In [8]:

g = tf.Graph()

In [9]:

print(g)

<tensorflow.python.framework.ops.Graph object at 0xb30d89080>

In [10]:

graph_one = tf.get_default_graph()

In [11]:

print(graph_one)

<tensorflow.python.framework.ops.Graph object at 0xb30d53b00>

In [12]:

graph_two = tf.Graph()

In [13]:

print(graph_two)

<tensorflow.python.framework.ops.Graph object at 0xb3109bd30>

In [15]:

with graph_two.as_default():
    print(graph_two is tf.get_default_graph())

True

In [16]:

print (graph_two is tf.get_default_graph())

False

In [ ]: