Visualizing CNN + Summary
import numpy as np
import tensorflow as tf
import matplotlib.pyplot as plt
from tensorflow.examples.tutorials.mnist import input_data
%matplotlib inline
mnist = input_data.read_data_sets('data/', one_hot=True)
trainimg = mnist.train.images
trainlabel = mnist.train.labels
testimg = mnist.test.images
testlabel = mnist.test.labels
print ("Packages loaded.")
Extracting data/train-images-idx3-ubyte.gz
Extracting data/train-labels-idx1-ubyte.gz
Extracting data/t10k-images-idx3-ubyte.gz
Extracting data/t10k-labels-idx1-ubyte.gz
Packages loaded.
Define Networks
tf.variable_scope("NAME"):
learning_rate = 0.001
training_epochs = 5
batch_size = 100
display_step = 1
n_input = 784
n_output = 10
with tf.variable_scope("CNN_WEIGHTS"):
weights = {
'wc1': tf.Variable(tf.random_normal([3, 3, 1, 64], stddev=0.1)),
'wc2': tf.Variable(tf.random_normal([3, 3, 64, 128], stddev=0.1)),
'wd1': tf.Variable(tf.random_normal([7*7*128, 1024], stddev=0.1)),
'wd2': tf.Variable(tf.random_normal([1024, n_output], stddev=0.1))
}
with tf.variable_scope("CNN_BIASES"):
biases = {
'bc1': tf.Variable(tf.random_normal([64], stddev=0.1)),
'bc2': tf.Variable(tf.random_normal([128], stddev=0.1)),
'bd1': tf.Variable(tf.random_normal([1024], stddev=0.1)),
'bd2': tf.Variable(tf.random_normal([n_output], stddev=0.1))
}
def conv_basic(_input, _w, _b, _keepratio):
with tf.variable_scope("INPUT_LAYER"):
_input_r = tf.reshape(_input, shape=[-1, 28, 28, 1])
with tf.variable_scope("CNN_CONV_1"):
_conv1 = tf.nn.relu(tf.nn.bias_add(tf.nn.conv2d(_input_r, _w['wc1']
, strides=[1, 1, 1, 1], padding='SAME'), _b['bc1']))
with tf.variable_scope("CNN_POOL_1"):
_pool1 = tf.nn.max_pool(_conv1, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1]
, padding='SAME')
_pool_dr1 = tf.nn.dropout(_pool1, _keepratio)
with tf.variable_scope("CNN_CONV_2"):
_conv2 = tf.nn.relu(tf.nn.bias_add(tf.nn.conv2d(_pool_dr1, _w['wc2']
, strides=[1, 1, 1, 1], padding='SAME'), _b['bc2']))
with tf.variable_scope("CNN_POOL_2"):
_pool2 = tf.nn.max_pool(_conv2, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1]
, padding='SAME')
_pool_dr2 = tf.nn.dropout(_pool2, _keepratio)
with tf.variable_scope("FC_1"):
_dense1 = tf.reshape(_pool_dr2, [-1, _w['wd1'].get_shape().as_list()[0]])
_fc1 = tf.nn.relu(tf.nn.bias_add(tf.matmul(_dense1, _w['wd1']), _b['bd1']))
_fc_dr1 = tf.nn.dropout(_fc1, _keepratio)
with tf.variable_scope("FC_2"):
_out = tf.add(tf.matmul(_fc_dr1, _w['wd2']), _b['bd2'])
out = {
'input_r': _input_r, 'conv1': _conv1, 'pool1': _pool1, 'pool1_dr1': _pool_dr1,
'conv2': _conv2, 'pool2': _pool2, 'pool_dr2': _pool_dr2, 'dense1': _dense1,
'fc1': _fc1, 'fc_dr1': _fc_dr1, 'out': _out }
return out
print ("Network ready")
Network ready
Define functions
x = tf.placeholder(tf.float32, [None, n_input], name="CNN_INPUT_x")
y = tf.placeholder(tf.float32, [None, n_output], name="CNN_TARGET_y")
keepratio = tf.placeholder(tf.float32, name="CNN_DROPOUT_keepratio")
pred = conv_basic(x, weights, biases, keepratio)['out']
cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(pred, y))
optm = tf.train.AdamOptimizer(learning_rate=learning_rate).minimize(cost)
corr = tf.equal(tf.argmax(pred,1), tf.argmax(y,1))
accr = tf.reduce_mean(tf.cast(corr, tf.float32))
init = tf.initialize_all_variables()
print ("Functions ready")
Functions ready
Summary
sess = tf.Session()
sess.run(init)
tf.scalar_summary('cross entropy', cost)
tf.scalar_summary('accuracy', accr)
merged = tf.merge_all_summaries()
summary_writer = tf.train.SummaryWriter('/tmp/tf_logs/cnn_mnist'
, graph=sess.graph)
print ("Summary ready")
Summary ready
Run
print ("Start!")
for epoch in range(training_epochs):
avg_cost = 0.
total_batch = int(mnist.train.num_examples/batch_size)
for i in range(total_batch):
batch_xs, batch_ys = mnist.train.next_batch(batch_size)
summary, _ = sess.run([merged, optm]
, feed_dict={x: batch_xs, y: batch_ys, keepratio:0.7})
avg_cost += sess.run(cost
, feed_dict={x: batch_xs, y: batch_ys, keepratio:1.})/total_batch
summary_writer.add_summary(summary, epoch*total_batch+i)
if epoch % display_step == 0:
print ("Epoch: %03d/%03d cost: %.9f" % (epoch, training_epochs, avg_cost))
train_acc = sess.run(accr, feed_dict={x: batch_xs, y: batch_ys, keepratio:1.})
print (" Training accuracy: %.3f" % (train_acc))
test_acc = sess.run(accr, feed_dict={x: testimg, y: testlabel, keepratio:1.})
print (" Test accuracy: %.3f" % (test_acc))
print ("Optimization Finished.")
Start!
Epoch: 000/005 cost: 0.352900247
Training accuracy: 0.990
Test accuracy: 0.981
Epoch: 001/005 cost: 0.047190106
Training accuracy: 0.980
Test accuracy: 0.988
Epoch: 002/005 cost: 0.031958400
Training accuracy: 1.000
Test accuracy: 0.990
Epoch: 003/005 cost: 0.022927662
Training accuracy: 1.000
Test accuracy: 0.990
Epoch: 004/005 cost: 0.018374201
Training accuracy: 0.990
Test accuracy: 0.991
Optimization Finished.
Run the command line
tensorboard --logdir=/tmp/tf_logs/cnn_mnist