Create a gist now

Instantly share code, notes, and snippets.

What would you like to do?
import tensorflow as tf
import numpy as np
from tensorflow.examples.tutorials.mnist import input_data
mnist = input_data.read_data_sets('MNIST_data', one_hot=True)
train_images = np.reshape(mnist.train.images, (-1, 28, 28, 1))
train_labels = mnist.train.labels
test_images = np.reshape(mnist.test.images, (-1, 28, 28, 1))
test_labels = mnist.test.labels
graph = tf.Graph()
with graph.as_default():
input = tf.placeholder(tf.float32, shape=(None, 28, 28, 1))
labels = tf.placeholder(tf.float32, shape=(None, 10))
padded_input = tf.image.resize_image_with_crop_or_pad(input, target_height=32, target_width=32)
layer1_weights = tf.get_variable("layer1_weights", [3, 3, 1, 64], initializer=tf.contrib.layers.variance_scaling_initializer())
layer1_bias = tf.Variable(tf.zeros([64]))
layer1_conv = tf.nn.conv2d(padded_input, filter=layer1_weights, strides=[1,1,1,1], padding='SAME')
layer1_out = tf.nn.relu(layer1_conv + layer1_bias)
layer2_weights = tf.get_variable("layer2_weights", [3, 3, 64, 64], initializer=tf.contrib.layers.variance_scaling_initializer())
layer2_bias = tf.Variable(tf.zeros([64]))
layer2_conv = tf.nn.conv2d(layer1_out, filter=layer2_weights, strides=[1,1,1,1], padding='SAME')
layer2_out = tf.nn.relu(layer2_conv + layer2_bias)
pool1 = tf.nn.max_pool(layer2_out, ksize=[1,2,2,1], strides=[1,2,2,1], padding='VALID')
layer3_weights = tf.get_variable("layer3_weights", [3, 3, 64, 128], initializer=tf.contrib.layers.variance_scaling_initializer())
layer3_bias = tf.Variable(tf.zeros([128]))
layer3_conv = tf.nn.conv2d(pool1, filter=layer3_weights, strides=[1,1,1,1], padding='SAME')
layer3_out = tf.nn.relu(layer3_conv + layer3_bias)
layer4_weights = tf.get_variable("layer4_weights", [3, 3, 128, 128], initializer=tf.contrib.layers.variance_scaling_initializer())
layer4_bias = tf.Variable(tf.zeros([128]))
layer4_conv = tf.nn.conv2d(layer3_out, filter=layer4_weights, strides=[1,1,1,1], padding='SAME')
layer4_out = tf.nn.relu(layer4_conv + layer4_bias)
pool2 = tf.nn.max_pool(layer4_out, ksize=[1,2,2,1], strides=[1,2,2,1], padding='VALID')
layer5_weights = tf.get_variable("layer5_weights", [3, 3, 128, 256], initializer=tf.contrib.layers.variance_scaling_initializer())
layer5_bias = tf.Variable(tf.zeros([256]))
layer5_conv = tf.nn.conv2d(pool2, filter=layer5_weights, strides=[1,1,1,1], padding='SAME')
layer5_out = tf.nn.relu(layer5_conv + layer5_bias)
layer6_weights = tf.get_variable("layer6_weights", [3, 3, 256, 256], initializer=tf.contrib.layers.variance_scaling_initializer())
layer6_bias = tf.Variable(tf.zeros([256]))
layer6_conv = tf.nn.conv2d(layer5_out, filter=layer6_weights, strides=[1,1,1,1], padding='SAME')
layer6_out = tf.nn.relu(layer6_conv + layer6_bias)
layer7_weights = tf.get_variable("layer7_weights", [3, 3, 256, 256], initializer=tf.contrib.layers.variance_scaling_initializer())
layer7_bias = tf.Variable(tf.zeros([256]))
layer7_conv = tf.nn.conv2d(layer6_out, filter=layer7_weights, strides=[1,1,1,1], padding='SAME')
layer7_out = tf.nn.relu(layer7_conv + layer7_bias)
pool3 = tf.nn.max_pool(layer7_out, ksize=[1,2,2,1], strides=[1,2,2,1], padding='VALID')
layer8_weights = tf.get_variable("layer8_weights", [3, 3, 256, 512], initializer=tf.contrib.layers.variance_scaling_initializer())
layer8_bias = tf.Variable(tf.zeros([512]))
layer8_conv = tf.nn.conv2d(pool3, filter=layer8_weights, strides=[1,1,1,1], padding='SAME')
layer8_out = tf.nn.relu(layer8_conv + layer8_bias)
layer9_weights = tf.get_variable("layer9_weights", [3, 3, 512, 512], initializer=tf.contrib.layers.variance_scaling_initializer())
layer9_bias = tf.Variable(tf.zeros([512]))
layer9_conv = tf.nn.conv2d(layer8_out, filter=layer9_weights, strides=[1,1,1,1], padding='SAME')
layer9_out = tf.nn.relu(layer9_conv + layer9_bias)
layer10_weights = tf.get_variable("layer10_weights", [3, 3, 512, 512], initializer=tf.contrib.layers.variance_scaling_initializer())
layer10_bias = tf.Variable(tf.zeros([512]))
layer10_conv = tf.nn.conv2d(layer9_out, filter=layer10_weights, strides=[1,1,1,1], padding='SAME')
layer10_out = tf.nn.relu(layer10_conv + layer10_bias)
pool4 = tf.nn.max_pool(layer10_out, ksize=[1,2,2,1], strides=[1,2,2,1], padding='VALID')
shape = pool4.shape.as_list()
newShape = shape[1] * shape[2] * shape[3]
reshaped_pool4 = tf.reshape(pool4, [-1, newShape])
fc1_weights = tf.get_variable("layer11_weights", [newShape, 4096], initializer=tf.contrib.layers.variance_scaling_initializer())
fc1_bias = tf.Variable(tf.zeros([4096]))
fc1_out = tf.nn.relu(tf.matmul(reshaped_pool4, fc1_weights) + fc1_bias)
fc2_weights = tf.get_variable("layer12_weights", [4096, 10], initializer=tf.contrib.layers.xavier_initializer())
fc2_bias = tf.Variable(tf.zeros([10]))
logits = tf.matmul(fc1_out, fc2_weights) + fc2_bias
cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(logits=logits, labels=labels))
learning_rate = 0.001
optimizer = tf.train.AdamOptimizer(learning_rate).minimize(cost)
#Add a few nodes to calculate accuracy and optionally retrieve predictions
predictions = tf.nn.softmax(logits)
correct_prediction = tf.equal(tf.argmax(labels, 1), tf.argmax(predictions, 1))
accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))
with tf.Session(graph=graph) as session:
tf.global_variables_initializer().run()
num_steps = 1000
batch_size = 100
for step in range(num_steps):
offset = (step * batch_size) % (train_labels.shape[0] - batch_size)
batch_images = train_images[offset:(offset + batch_size), :]
batch_labels = train_labels[offset:(offset + batch_size), :]
feed_dict = {input: batch_images, labels: batch_labels}
_, c, acc = session.run([optimizer, cost, accuracy], feed_dict=feed_dict)
if step % 100 == 0:
print("Cost: ", c)
print("Accuracy: ", acc * 100.0, "%")
#Test
num_test_batches = int(len(test_images) / 100)
total_accuracy = 0
total_cost = 0
for step in range(num_test_batches):
offset = (step * batch_size) % (train_labels.shape[0] - batch_size)
batch_images = test_images[offset:(offset + batch_size)]
batch_labels = test_labels[offset:(offset + batch_size)]
feed_dict = {input: batch_images, labels: batch_labels}
_, c, acc = session.run([optimizer, cost, accuracy], feed_dict=feed_dict)
total_cost = total_cost + c
total_accuracy = total_accuracy + acc
print("Test Cost: ", total_cost / num_test_batches)
print("Test accuracy: ", total_accuracy * 100.0 / num_test_batches, "%")
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment