In [1]:
import tensorflow as tf
In [2]:
def weight_variable(shape):
initial = tf.truncated_normal(shape, stddev=0.1)
return tf.Variable(initial)
In [3]:
def bias_variable(shape):
initial = tf.constant(0.1, shape=shape)
return tf.Variable(initial)
In [4]:
def conv2d(x, W):
return tf.nn.conv2d(x, W, strides=[1, 1, 1, 1], padding='SAME')
In [5]:
def max_pool_2x2(x):
return tf.nn.max_pool(x, ksize=[1, 2, 2, 1],
strides=[1, 2, 2, 1], padding='SAME')
In [6]:
# parametric rectified linear unit
def prelu(x, alphas):
return tf.nn.relu(x) + tf.mul(alphas, tf.sub(x, tf.abs(x)))
In [7]:
x = tf.placeholder(tf.float32, shape=[None, 784])
y_ = tf.placeholder(tf.float32, shape=[None, 10])
In [8]:
x_image = tf.reshape(x, [-1, 28, 28, 1])
In [9]:
W_conv1 = weight_variable([5, 5, 1, 32])
b_conv1 = bias_variable([32])
alphas_conv1 = bias_variable([32])
layer_conv_1 = prelu(conv2d(x_image, W_conv1) + b_conv1, alphas_conv1)
In [10]:
W_conv2 = weight_variable([5, 5, 32, 1])
b_conv2 = bias_variable([32])
alphas_conv2 = bias_variable([32])
layer_conv_2 = prelu(conv2d(layer_conv_1, W_conv2) + b_conv2, alphas_conv2)
stage_1_pool = max_pool_2x2(layer_conv_2)
In [11]:
layer_conv_2.get_shape()
Out[11]:
In [12]:
W_conv3 = weight_variable([5, 5, 32, 64])
b_conv3 = bias_variable([64])
alphas_conv3 = bias_variable([64])
layer_conv_3 = prelu(conv2d(layer_conv_2, W_conv3) + b_conv3, alphas_conv3)
W_conv4 = weight_variable([5, 5, 64, 1])
b_conv4 = bias_variable([64])
alphas_conv4 = bias_variable([64])
layer_conv_4 = prelu(conv2d(layer_conv_3, W_conv4) + b_conv4, alphas_conv4)
stage_2_pool = max_pool_2x2(layer_conv_4)
In [13]:
W_conv5 = weight_variable([5, 5, 64, 128])
b_conv5 = bias_variable([128])
alphas_conv5 = bias_variable([128])
layer_conv_5 = prelu(conv2d(layer_conv_4, W_conv5) + b_conv5, alphas_conv5)
W_conv6 = weight_variable([5, 5, 128, 1])
b_conv6 = bias_variable([128])
alphas_conv6 = bias_variable([128])
layer_conv_6 = prelu(conv2d(layer_conv_5, W_conv6) + b_conv6, alphas_conv6)
stage_3_pool = max_pool_2x2(layer_conv_6)
In [14]:
stage_3_pool.get_shape()
Out[14]:
In [15]:
W_fc1 = weight_variable([14 * 14 * 128, 2])
b_fc1 = bias_variable([2])
alphas_fc1 = bias_variable([2])
stage_3_pool_flat = tf.reshape(stage_3_pool, [-1, 14 * 14 * 128])
h_fc1 = prelu(tf.matmul(stage_3_pool_flat, W_fc1) + b_fc1, alphas_fc1)
In [16]:
# keep_prob = tf.placeholder(tf.float32)
# h_fc1_drop = tf.nn.dropout(h_fc1, keep_prob)
# W_fc2 = weight_variable([2, 10])
# b_fc2 = bias_variable([10])
# y_conv = tf.nn.softmax(tf.matmul(h_fc1_drop, W_fc2) + b_fc2)
In [17]:
W_fc2 = weight_variable([2, 10])
b_fc2 = bias_variable([10])
y_conv = tf.nn.softmax(tf.matmul(h_fc1, W_fc2) + b_fc2)
In [18]:
from tensorflow.examples.tutorials.mnist import input_data
mnist = input_data.read_data_sets('MNIST_data', one_hot=True)
In [19]:
sess = tf.InteractiveSession()
In [25]:
cross_entropy = tf.reduce_mean(-tf.reduce_sum(y_ * tf.log(y_conv), reduction_indices=[1]))
correct_prediction = tf.equal(tf.argmax(y_conv,1), tf.argmax(y_,1))
train_step = tf.train.AdamOptimizer(1e-1).minimize(cross_entropy)
accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))
sess.run(tf.initialize_all_variables())
In [26]:
%time
for i in range(2000):
batch = mnist.train.next_batch(250)
if i%100 == 0:
train_accuracy = accuracy.eval(feed_dict={
x:batch[0], y_: batch[1]}) # , keep_prob: 1.0
print("step %d, training accuracy %g"%(i, train_accuracy))
train_step.run(feed_dict={x: batch[0], y_: batch[1]})#, keep_prob: 0.5
In [30]:
%%time
print("test accuracy %g"%accuracy.eval(feed_dict={
x: mnist.test.images[:500], y_: mnist.test.labels[:500]}))#, keep_prob: 1.0
In [ ]:
In [ ]: