在《數字識別-softmax 迴歸》和《數字識別-rnn 算法》兩篇博文中分別介紹了使用 softmax 和 rnn 算法來對數字進行識別,並且 rnn 算法相對於 softmax 迴歸的基礎得到了很大的提升,而在圖片分類中,cnn 算法一直是主導,本篇博文將介紹使用 rnn 算法來進行數字識別。
import tensorflow as tf
from tensorflow.examples.tutorials.mnist import input_data
mnist = input_data.read_data_sets('data/mnist', one_hot=True)
X = tf.placeholder(tf.float32, [None, 784])
y = tf.placeholder(tf.float32, [None, 10])
def weight_variable(shape):
return tf.Variable(tf.truncated_normal(shape, stddev=0.1))
def bias_variable(shape):
return tf.Variable(tf.constant(0.1, shape=shape))
def conv2d(X,W):
return tf.nn.conv2d(X, W, strides=[1, 1, 1, 1], padding='SAME')
def max_pool_2x2(X):
return tf.nn.max_pool(X, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME’)
X_image = tf.reshape(X, [-1, 28, 28, 1])
# 第一個卷積層
# patch_size: 5*5 input channels: 1 output channels:32
W_conv1 = weight_variable([5, 5, 1, 32])
b_conv1 = bias_variable([32])
# 第一個池化層
h_conv1 = tf.nn.relu(conv2d(X_image, W_conv1) + b_conv1)
h_pool1 = max_pool_2x2(h_conv1)
# 第二個卷積層
W_conv2 = weight_variable([5, 5, 32, 64])
b_conv2 = bias_variable([64])
# 第二個池化層
h_conv2 = tf.nn.relu(conv2d(h_pool1, W_conv2) + b_conv2)
h_pool2 = max_pool_2x2(h_conv2)
# 第一個全連接層
# image_size:7*7
W_fc1 = weight_variable([7*7*64, 1024])
b_fc1 = bias_variable([1024])
h_pool2_flat = tf.reshape(h_pool2, [-1,7*7*64])
h_fc1 = tf.nn.relu(tf.matmul(h_pool2_flat, W_fc1) + b_fc1)
keep_prob = tf.placeholder(tf.float32)
h_fc1_drop = tf.nn.dropout(h_fc1, keep_prob)
# 第二個全連接層
W_fc2 = weight_variable([1024, 10])
b_fc2 = bias_variable([10])
y_conv = tf.matmul(h_fc1_drop, W_fc2) + b_fc2
# 計算損失函數
cross_entropy = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(labels=y, logits=y_conv))
#adam 算法優化
train_step = tf.train.AdamOptimizer(1e-4).minimize(cross_entropy)
# 正確預測數目
correct_prediction = tf.equal(tf.argmax(y, 1),tf.argmax(y_conv, 1))
#計算準確率
accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))
sess = tf.InteractiveSession()
sess.run(tf.global_variables_initializer())
for i in range(1001):
xs, ys= mnist.train.next_batch(64)
if i%100 == 0:
train_accuracy = accuracy.eval(feed_dict={X: xs, y: ys, keep_prob: 1.0})
print('step %d, training accuracy %g' % (i, train_accuracy))
train_step.run(feed_dict={X: xs, y: ys, keep_prob: 0.5})
正確率爲98%左右。