在普通神经网络的基础上,加入隐藏层,减轻过拟合的Dropout,自适应学习速率的Adagrad,以及可以解决梯度你三的激活函数Relu.
首先是载入Tensorflow并加载MNIST数据集,创建一个Tensorflow默认的InteractiveSession,这样后面执行各项操作就无需指定Session。
from tensorflow.examples.tutorials.mnist import input_data
import tensorflow as tf
mnist=input_data.read_data_sets('/tmp/data',one_hot=True)
sess=tf.InteractiveSession()
in_units=784
hl_units=300
w1=tf.Variable(tf.truncated_normal([in_units,hl_units],stddev=0.1))
b1=tf.Variable(tf.zeros([hl_units]))
w2=tf.Variable(tf.zeros([hl_units,10]))
b2=tf.Variable(tf.zeros([10]))
接下来定义输入x的placeholder,另外因为在训练和预测时,Dropout的比率keep_prob(即保留节点的概率)是不一样的,通常训练是小于1,而预测时是等于1,所以也把Dropout的比率作为计算图的输入,并定义一个placeholder。
x=tf.placeholder(tf.float32,[None,in_units])
keep_prob=tf.placeholder(tf.float32)
hiddenl=tf.nn.relu(tf.matmul(x,w1)+b1)
hiddenl_drop=tf.nn.dropout(hiddenl,keep_prob)
y=tf.nn.softmax(tf.matmul(hiddenl_drop,w2)+b2)
y_=tf.placeholder(tf.float32,[None,10])
cross_entropy=tf.reduce_mean(-tf.reduce_sum(y_*tf.log(y),reduction_indices=[1]))
train_step=tf.train.AdagradOptimizer(0.3).minimize(cross_entropy)
tf.global_variables_initializer().run()
for i in range(3000):
batch_xs,batch_ys=mnist.train.next_batch(100)
train_step.run({x:batch_xs,y_:batch_ys,keep_prob:0.75})
correct_prediction=tf.equal(tf.argmax(y,1),tf.argmax(y_,1))
accuracy=tf.reduce_mean(tf.cast(correct_prediction,tf.float32))
print accuracy.eval({x:mnist.test.images,y_:mnist.test.labels,keep_prob:1.0})
z