In [17]:

    

import numpy as np 
import tensorflow as tf 
import matplotlib.pyplot as plt


% matplotlib inline
plt.style.use('ggplot')


    

In [18]:

    

from tensorflow.examples.tutorials.mnist import input_data
mnist = input_data.read_data_sets("MNIST_data", one_hot = True)


    

    




Extracting MNIST_data/train-images-idx3-ubyte.gz
Extracting MNIST_data/train-labels-idx1-ubyte.gz
Extracting MNIST_data/t10k-images-idx3-ubyte.gz
Extracting MNIST_data/t10k-labels-idx1-ubyte.gz

In [19]:

    

print(mnist.train.images.shape)


    

    




(55000, 784)

In [20]:

    

plt.imshow(mnist.train.images[0].reshape(28,28))
for i in range(0,9):
    if mnist.train.labels[0][i] == 1:
        print('LABEL:{}' .format(i))


    

    




LABEL:7






    

In [23]:

    

input_nodes = 28*28
output_nodes = 10
learning_rate = 0.001

# Creating the graph
tf.reset_default_graph()

with tf.name_scope('Placeholders'):
    x = tf.placeholder(dtype=tf.float32, shape=[None, input_nodes], name ='Input')
    y = tf.placeholder(dtype=tf.float32, shape=[None, output_nodes], name = 'Labels')
    x_image = tf.reshape(x, [-1, 28, 28, 1])
tf.summary.image('Input_images', x_image, 3)

with tf.name_scope('Operations'):
    W = tf.get_variable(dtype=tf.float32,
                initializer=tf.random_normal_initializer(mean = 0, stddev=0.1),
                shape=[input_nodes, output_nodes],
                   name = 'W')
    b = tf.get_variable(dtype=tf.float32,
                   initializer=tf.constant_initializer(0.0),
                    shape =[output_nodes],
                    name = 'b'
                   )
    h = tf.nn.softmax(tf.matmul(x,W) + b)

# Cross Entropy 
with tf.name_scope('xEnt'):
    cross_entropy = tf.reduce_mean(-tf.reduce_sum(y * tf.log(h), reduction_indices=[1]))
tf.summary.scalar('Cross_Entropy', cross_entropy) # TB

# Optimiser 
with tf.name_scope('Train'):
    optimiser = tf.train.GradientDescentOptimizer(learning_rate).minimize(cross_entropy)
    
# Accuracy
with tf.name_scope('Accuracy'):
    correct_pediction = tf.equal(tf.arg_max(y,1), tf.arg_max(h,1))
    accuracy = tf.reduce_mean(tf.cast(correct_pediction, tf.float32))
tf.summary.scalar('Accuracy', accuracy) # TB


    

    
Out[23]:





<tf.Tensor 'Accuracy_1:0' shape=() dtype=string>

In [24]:

    

%%time 

sess = tf.Session()
sess.run(tf.global_variables_initializer())

# TensorBoard Filewriter 
merged_summary = tf.summary.merge_all()
writer = tf.summary.FileWriter("./tmp/mnist/1")
writer.add_graph(sess.graph)

# Train
batch_size = 100
epoch = 10
epoch_size = 500
n_iter = epoch * epoch_size

for iter in range(n_iter+1):
    xs, ys = mnist.train.next_batch(batch_size)
    sess.run(optimiser, feed_dict={x: xs, y: ys})
    
    if iter % 5 ==0:
        s = sess.run(merged_summary, feed_dict={x:xs, y:ys})
        writer.add_summary(s,iter)
    
    if iter%epoch_size ==0:
        print("Epoch: {} Cross Entropy: {:.2f} " .format(int(iter/epoch_size),
              cross_entropy.eval(feed_dict={x:xs, y:ys}, session=sess)))


    

    




Epoch: 0 Cross Entropy: 2.96 
Epoch: 1 Cross Entropy: 2.11 
Epoch: 2 Cross Entropy: 1.81 
Epoch: 3 Cross Entropy: 1.57 
Epoch: 4 Cross Entropy: 1.26 
Epoch: 5 Cross Entropy: 1.16 
Epoch: 6 Cross Entropy: 1.13 
Epoch: 7 Cross Entropy: 1.01 
Epoch: 8 Cross Entropy: 1.07 
Epoch: 9 Cross Entropy: 0.86 
Epoch: 10 Cross Entropy: 0.78 
CPU times: user 19.1 s, sys: 2.66 s, total: 21.8 s
Wall time: 14.8 s

In [ ]:

    

t_accur = sess.run(accuracy, feed_dict={x:mnist.train.images, y:mnist.train.labels})
print('Test Accuracy: {:4.2f}%' .format(100*t_accur))


    

In [ ]:

    

v_accur = sess.run(accuracy, feed_dict={x:mnist.validation.images, y: mnist.validation.labels})
print('Valid Accuracy: {:4.2f}%' .format(100*v_accur))


    

In [ ]:

    

t_accur = sess.run(accuracy, feed_dict={x:mnist.test.images, y: mnist.test.labels})
print('Test Accuracy: {:4.2f}%' .format(100*t_accur))