基礎知識
- 使用圖 (graph) 來表示計算任務.
- 在被稱之爲 會話 (Session) 的上下文 (context) 中執行圖.
- 使用 tensor 表示數據.
- 通過 變量 (Variable) 維護狀態.
- 使用 feed 和 fetch 可以爲任意的操作(arbitrary operation) 賦值或者從其中獲取數據.
圖
A Graph
contains a set of Operation
objects, which represent units of computation; and Tensor
objects, which represent the units of data that flow between operations.
- 創建圖:
構建圖的第一步, 是創建源 op (source op). 源 op 不需要任何輸入, 例如 常量 (Constant). 源 op 的輸出被傳遞給其它 op 做運算.
Python 庫中, op 構造器的返回值代表被構造出的 op 的輸出, 這些返回值可以傳遞給其它 op 構造器作爲輸入. - 在一個會話中啓動圖
會話
- session 類 (http://wiki.jikexueyuan.com/project/tensorflow-zh/api_docs/python/client.html)
- InteractiveSession 類
獲取中間變量
sess.run(variable)
構建多層卷積網絡模型
- 數據集
這裏使用mnist數據集,使用寫好的代碼 input_data 導入數據集
數據集分爲三個部分:- 訓練數據集55000:mnist.train;訓練圖片:mnist.train.images;訓練標籤:mnist.train.labels.
- 測試數據集10000:mnist.test。例子導入測試集圖像和標籤:
feed_dict={x:mnist.test.images,y_:mnist.test.labels
- 驗證數據集5000: mnist.validation
from tensorflow.examples.tutorials.mnist import input_data
import tensorflow as tf
mnist=input_data.read_data_sets("/home/yangshuhui/code/dataset/MNIST_data",one_hot=True)
# initial function
def weight_variable(shape):
initial=tf.truncated_normal(shape,stddev=0.1)
return tf.Variable(initial)
def bias_variable(shape):
initial=tf.constant(0.1,shape=shape)
return tf.Variable(initial)
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')
#1
x=tf.placeholder("float",[None,784])
x_image=tf.reshape(x,[-1,28,28,1])
w_conv1=weight_variable([5,5,1,32])
b_conv1=bias_variable([32])
h_conv1=tf.nn.relu(conv2d(x_image,w_conv1)+b_conv1)
h_pool1=max_pool_2x2(h_conv1)
#2
W_conv2=weight_variable([5,5,32,64])
b_conv2=bias_variable([64])
h_conv2=tf.nn.relu(conv2d(h_pool1,W_conv2)+b_conv2)
h_pool2=max_pool_2x2(h_conv2)
#3
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("float")
h_fc1_drop=tf.nn.dropout(h_fc1,keep_prob)
#output
W_fc2=weight_variable([1024,10])
b_fc2=bias_variable([10])
y_conv=tf.nn.softmax(tf.matmul(h_fc1_drop,W_fc2)+b_fc2)
# train and evaluate
y_=tf.placeholder("float",[None,10])
cross_entropy=-tf.reduce_sum(y_*tf.log(y_conv))
train_step=tf.train.AdamOptimizer(1e-4).minimize(cross_entropy)
correct_prediction=tf.equal(tf.argmax(y_conv,1),tf.argmax(y_,1))
accuracy=tf.reduce_mean(tf.cast(correct_prediction,"float"))
sess=tf.Session()
sess.run(tf.initialize_all_variables())
for i in range(20000):
batch=mnist.train.next_batch(50)
if i%100==0:
train_accuracy=accuracy.eval(session=sess,feed_dict={x:batch[0],y_:batch[1],keep_prob:1.0})
print "step %d, training accuracy %g"%(i,train_accuracy)
train_step.run(session=sess,feed_dict={x:batch[0],y_:batch[1],keep_prob:0.5})
print "test accuracy %g"%accuracy.eval(session=sess,feed_dict={x:mnist.test.images,y_:mnist.test.labels,keep_prob:1.0})