Is there any difference between the two codes? - machine-learning

Question

I am currently still relatively new to Tensorflow. I am having some trouble with these two pieces of code.
Code A:
self.h1_layer = tf.layers.dense(self.x, self.n_nodes_hl1, activation=tf.nn.relu, name="h1")
self.h2_layer = tf.layers.dense(self.h1_layer, self.n_nodes_hl2, activation=tf.nn.relu, name="h2")
self.h3_layer = tf.layers.dense(self.h2_layer, self.n_nodes_hl3, activation=tf.nn.relu, name="h3")
self.logits = tf.layers.dense(self.h3_layer, self.num_of_classes, name="output")
Code B:
self.hidden_1_layer = {
'weights': tf.Variable(tf.random_normal([self.num_of_words, self.h1])),
'biases' : tf.Variable(tf.random_normal([self.h1]))
}
self.hidden_2_layer = {
'weights': tf.Variable(tf.random_normal([self.h1, self.h2])),
'biases' : tf.Variable(tf.random_normal([self.h2]))
}
self.hidden_3_layer = {
'weights': tf.Variable(tf.random_normal([self.h2, self.h3])),
'biases' : tf.Variable(tf.random_normal([self.h3]))
}
self.final_output_layer = {
'weights': tf.Variable(tf.random_normal([self.h3, self.num_of_classes])),
'biases' : tf.Variable(tf.random_normal([self.num_of_classes]))
}
layer1 = tf.add(tf.matmul(data, self.hidden_1_layer['weights']), self.hidden_1_layer['biases'])
layer1 = tf.nn.relu(layer1)
layer2 = tf.add(tf.matmul(layer1, self.hidden_2_layer['weights']), self.hidden_2_layer['biases'])
layer2 = tf.nn.relu(layer2)
layer3 = tf.add(tf.matmul(layer2, self.hidden_3_layer['weights']), self.hidden_3_layer['biases'])
layer3 = tf.nn.relu(layer3)
output = tf.matmul(layer3, self.final_output_layer['weights']) + self.final_output_layer['biases']
Are they the same thing? Can both Codes A & B weights and biases be saved with tf.train.Saver() ?
Thanks
Edit:
I am facing issues using Code A to generate prediction. It seems that logits of Code A is always changing.
The full code:
import tensorflow as tf
import os
from utils import Utils as utils
os.environ['TF_CPP_MIN_LOG_LEVEL'] = '3'
class Neural_Network:
# Neural Network Setup
num_of_epoch = 50
n_nodes_hl1 = 500
n_nodes_hl2 = 500
n_nodes_hl3 = 500
def __init__(self):
self.num_of_classes = utils.get_num_of_classes()
self.num_of_words = utils.get_num_of_words()
# placeholders
self.x = tf.placeholder(tf.float32, [None, self.num_of_words])
self.y = tf.placeholder(tf.int32, [None, self.num_of_classes])
with tf.name_scope("model"):
self.h1_layer = tf.layers.dense(self.x, self.n_nodes_hl1, activation=tf.nn.relu, name="h1")
self.h2_layer = tf.layers.dense(self.h1_layer, self.n_nodes_hl2, activation=tf.nn.relu, name="h2")
self.h3_layer = tf.layers.dense(self.h2_layer, self.n_nodes_hl3, activation=tf.nn.relu, name="h3")
self.logits = tf.layers.dense(self.h3_layer, self.num_of_classes, name="output")
def predict(self):
return self.logits
def make_prediction(self, query):
result = None
with tf.Session() as sess:
sess.run(tf.global_variables_initializer())
saver = tf.train.import_meta_graph('saved_models/testing.meta')
saver.restore(sess, 'saved_models/testing')
# for variable in tf.trainable_variables():
# print sess.run(variable)
prediction = self.predict()
pre, prediction = sess.run([self.logits, prediction], feed_dict={self.x : query})
print pre
prediction = prediction.tolist()
prediction = tf.nn.softmax(prediction)
prediction = sess.run(prediction)
print prediction
return utils.get_label_from_encoding(prediction[0])
def train(self, data):
prediction = self.predict()
cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(logits=prediction, labels=self.y))
optimizer = tf.train.AdamOptimizer().minimize(cost)
with tf.Session() as sess:
sess.run(tf.global_variables_initializer())
writer = tf.summary.FileWriter("mygraph/logs", tf.get_default_graph())
for epoch in range(self.num_of_epoch):
optimised, loss = sess.run([optimizer, cost],
feed_dict={self.x: data['values'], self.y: data['labels']})
if epoch % 1 == 0:
print("Completed Training Cycle: " + str(epoch) + " out of " + str(self.num_of_epoch))
print("Current Loss: " + str(loss))
saver = tf.train.Saver()
saver.save(sess, 'saved_models/testing')
print("Model saved")

Answer 1

TLDR: The operations are essentially the same but the variables creation and initialization methods are different.
If you trace the code from here, you will eventually get to a stage where the code is calling tf.get_variable to initialize variables. In your example above, since kernel_initializer and bias_initializer is not set, they will default to None and tf.zeros_initializer() respectively (see Dense API). When None is passed to tf.get_variable as an initializer, a glorot_uniform_initializer will be used:
If initializer is None (the default), the default initializer passed
in the variable scope will be used. If that one is None too, a
glorot_uniform_initializer will be used. The initializer can also be a
Tensor, in which case the variable is initialized to this value and
shape.
More on tf.get_variable can be found here.
For one case, you used a tf.random_normal initializer for both kernel weights and bias weights, but for the other, you used tf.layers.dense and will result in a glorot_uniform_initializer for kernel weights and zeros_initializer for bias weights as no parameters were passed to tf.layers.dense.
To your second question on whether they can be saved, yes they can.
As a last note, you have to be careful when using tf.Variable as it might complicate things when the scopes are not properly set.