神经网络的结构：
输入层--> 隐藏层 ---> 输出层
输入层 ： 28*28
隐藏层 ： 10 个
输出层 ：10
LSTM 的原理在这里不做赘述，仅仅是为了了解代码的实现过程
代码如下：
# -*- coding: utf-8 -*-
"""
Created on Sat May 25 20:19:28 2019
@author: 666
"""
import tensorflow as tf
from tensorflow.examples.tutorials.mnist import input_data
from tensorflow.contrib import rnn
#载入数据集
mnist = input_data.read_data_sets('F:/AI/AItest/MNIST_data/',one_hot=True)
#定义参数
#输入一行，一行有28个数据
n_input = 28
#一共有28行
max_time = 28
#100个隐藏单元
lstm_size = 10
#10分类
n_class = 10
#每批次分50个样本
batch_size = 50
#一共有n_batch个批次
n_batch = mnist.train.num_examples
#定义输入
with tf.name_scope('input'):
x = tf.placeholder(tf.float32,[None,784],name='x-input')
y = tf.placeholder(tf.float32,[None,10],name='y-input')
#初始化权重值
weights = tf.Variable(tf.truncated_normal([lstm_size,n_class],stddev=0.1))
biases = tf.Variable(tf.constant(0.1,shape=[n_class]))
#定义RNN网络
def RNN(X,weights,biases):
with tf.name_scope('RNN'):
#input = [batch_size,max_time,n_input]
inputs = tf.reshape(X,[-1,max_time,n_input])
#定义LSTM基本的cell
lstm_cell = rnn.BasicLSTMCell(lstm_size)
#final_state[0] 是 cell state
#final_state[1] 是 hidden_state
outputs,final_state = tf.nn.dynamic_rnn(lstm_cell, inputs, dtype=tf.float32)
results = tf.nn.softmax(tf.matmul(final_state[1],weights) + biases)
return results
prediction = RNN(x,weights,biases)
#损失函数
with tf.name_scope('loss'):
loss = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(logits= prediction,labels=y))
#优化器
with tf.name_scope('optimizer'):
train_step = tf.train.AdamOptimizer(1e-4).minimize(loss)
#结果存在一个bool类型的值中
correct_prediction = tf.equal(tf.argmax(prediction,1),tf.argmax(y,1))
#求准确率
with tf.name_scope('accuracy'):
accuracy = tf.reduce_mean(tf.cast(correct_prediction,tf.float32))
init = tf.global_variables_initializer()
with tf.Session() as sess:
sess.run(init)
for epoch in range(21):
for batch in range(n_batch):
batch_xs,batch_ys = mnist.train.next_batch(batch_size)
sess.run(train_step,feed_dict={x:batch_xs,y:batch_ys})
acc = sess.run(accuracy,feed_dict={x:mnist.test.images,y:mnist.train.labels})
print('Iter '+ str(epoch) + " , Test accuracy = " + str(acc) )
