import tensorflow as tf
from numpy.random import RandomState
batch_size = 8
#随机生成两个变量,服从正态分布
w1 = tf.Variable(tf.random_normal([2,3],stddev=1,seed = 1))
w2 = tf.Variable(tf.random_normal([3,1],stddev=1,seed = 1))
#定义两个占位符,用来训练时输入
x= tf.placeholder(tf.float32, shape=(None , 2) ,name= 'x-input')
y_ = tf.placeholder(tf.float32 , shape=(None ,1),name='y-input')
#定义神经网络向前传播的过程
a= tf.matmul(x,w1)
y= tf.matmul(a,w2)
#定义损失函数与反向传播的算法
cross_entropy = -tf.reduce_mean(y_*tf.log(tf.clip_by_value(y,1e-10,1.0)))
train_step = tf.train.AdadeltaOptimizer(0.001).minimize(cross_entropy)
#通过随机数生成一个模拟的数据集
rdm = RandomState(1)
dataset_size = 64
X = rdm.rand(dataset_size,2)
Y = [[int(x1+x2<1)] for (x1,x2) in X]
#创建初始化函数,对所有变量进行初始化
init_op = tf.initialize_all_variables()
#创建会话
with tf.Session() as sess:
sess.run(init_op)
print(sess.run(w1))
print(sess.run(w2))
step = 5000
for i in range(step):
start = (i* batch_size) %dataset_size
end = min(start+batch_size,dataset_size)
total_cross_entropy = sess.run(cross_entropy,feed_dict={x:X,y_:Y})
print("After %d trainint step(s),cross entropy on all data is %g" % (i, total_cross_entropy))
cross_entropy = -tf.reduce_mean(y_*tf.log(tf.clip_by_value(y,1e-10,1.0)))
定义交叉熵,随着训练的不断进行,交叉熵变小,其中随着训练,神经网络参数的值趋向确定,也就是
W1 和 W2 ,在这里w1 w2 所谓的向前传播的过程其实可以通过矩阵运算来表示,其中w1,w2就是权值