TF_Variable Sharing

Reference: http://jermmy.xyz/2017/08/25/2017-8-25-learn-tensorflow-shared-variables/

Tensorflow doesn't know what nodes should be grouped together, unless you tell it to.

name scope VS variable scope   tf.name_scope() VS tf.variable_scope()    variable scope facilitates variable sharing

tf.variable_scope() implicitly creates a name scope.

 """ Examples to demonstrate variable sharing
 CS 20: 'TensorFlow for Deep Learning Research'
 cs20.stanford.edu
 Chip Huyen (chiphuyen@cs.stanford.edu)
 Lecture 05
 """
 import os
 os.environ['TF_CPP_MIN_LOG_LEVEL']=''
 
 import tensorflow as tf
 
 x1 = tf.truncated_normal([200, 100], name='x1')
 x2 = tf.truncated_normal([200, 100], name='x2')
 
 def two_hidden_layers(x):
     assert x.shape.as_list() == [200, 100]
     w1 = tf.Variable(tf.random_normal([100, 50]), name='h1_weights')
     b1 = tf.Variable(tf.zeros([50]), name='h1_biases')
     h1 = tf.matmul(x, w1) + b1
     assert h1.shape.as_list() == [200, 50]
     w2 = tf.Variable(tf.random_normal([50, 10]), name='h2_weights')
     b2 = tf.Variable(tf.zeros([10]), name='2_biases')
     logits = tf.matmul(h1, w2) + b2
     return logits
 
 def two_hidden_layers_2(x):
     assert x.shape.as_list() == [200, 100]
     w1 = tf.get_variable('h1_weights', [100, 50], initializer=tf.random_normal_initializer())
     b1 = tf.get_variable('h1_biases', [50], initializer=tf.constant_initializer(0.0))
     h1 = tf.matmul(x, w1) + b1
     assert h1.shape.as_list() == [200, 50]
     w2 = tf.get_variable('h2_weights', [50, 10], initializer=tf.random_normal_initializer())
     b2 = tf.get_variable('h2_biases', [10], initializer=tf.constant_initializer(0.0))
     logits = tf.matmul(h1, w2) + b2
     return logits
 
 # logits1 = two_hidden_layers(x1)
 # logits2 = two_hidden_layers(x2)

#### Two sets of variables are created. You want all your inputs to use the same weights and biases! 
 # logits1 = two_hidden_layers_2(x1)
 # logits2 = two_hidden_layers_2(x2)
 
####ValueError: Variable h1_weights already exists,disallowed.Did you mean to set reuse=True in VarScope?
 
 # with tf.variable_scope('two_layers') as scope:
 #     logits1 = two_hidden_layers_2(x1)
 #     scope.reuse_variables()
 #     logits2 = two_hidden_layers_2(x2)
 
 # with tf.variable_scope('two_layers') as scope:
 #     logits1 = two_hidden_layers_2(x1)
 #     scope.reuse_variables()
 #     logits2 = two_hidden_layers_2(x2)
 
 def fully_connected(x, output_dim, scope):
     with tf.variable_scope(scope, reuse=tf.AUTO_REUSE) as scope:
         w = tf.get_variable('weights', [x.shape[1], output_dim], initializer=tf.random_normal_initializer())
         b = tf.get_variable('biases', [output_dim], initializer=tf.constant_initializer(0.0))
         return tf.matmul(x, w) + b
 
 def two_hidden_layers(x):
     h1 = fully_connected(x, 50, 'h1')
     h2 = fully_connected(h1, 10, 'h2')
 
 with tf.variable_scope('two_layers') as scope:
     logits1 = two_hidden_layers(x1)
     # scope.reuse_variables()
     logits2 = two_hidden_layers(x2)
 
 writer = tf.summary.FileWriter('./graphs/cool_variables', tf.get_default_graph())
 writer.close()