tensorflow（三）

最新推荐文章于 2025-05-31 14:15:49 发布

原创最新推荐文章于 2025-05-31 14:15:49 发布 · 150 阅读

0 ·

CC 4.0 BY-SA版权

文章标签：

#MNIST

计算机视觉专栏收录该内容

17 篇文章

订阅专栏

本文介绍使用TensorFlow构建卷积神经网络进行Mnist手写数字识别的全过程，包括数据集加载、网络结构定义、训练过程及准确率评估。通过调整超参数和网络结构，实现对Mnist数据集的高效识别。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

tensorflow（三）——Mnist

代码

import tensorflow as tf
from tensorflow.examples.tutorials.mnist import input_data
import numpy as np
import matplotlib.pyplot as plt


mnist = input_data.read_data_sets("MNIST_data/", one_hot=True)




n_input = 784
n_output = 10

stddev = 0.1

weigths = {
	'wc1':tf.Variable(tf.random_normal([3,3,1,64],stddev=stddev)),
	'wc2':tf.Variable(tf.random_normal([3,3,64,128],stddev=stddev)),
	'wf1':tf.Variable(tf.random_normal([7*7*128,1024])),
	'out':tf.Variable(tf.random_normal([1024,n_output],stddev=stddev))
}

biases = {
	'bc1':tf.Variable(tf.random_normal([64],stddev=stddev)),
	'bc2':tf.Variable(tf.random_normal([128],stddev=stddev)),
	'bf1':tf.Variable(tf.random_normal([1024],stddev=stddev)),
	'out':tf.Variable(tf.random_normal([n_output],stddev=stddev))
}

def conv_basic(_input,_w,_b,_keepratio):
	#把数据转成tensorflow的四维输入
	_input = tf.reshape(_input,shape=[-1,28,28,1])
	#第一卷积层
	_conv1 = tf.nn.conv2d(_input,_w['wc1'],strides=[1,1,1,1],padding='SAME')
	#relu激活层
	_conv1 = tf.nn.relu(tf.nn.bias_add(_conv1,_b['bc1']))
	_pool1 = tf.nn.max_pool(_conv1,ksize=[1,2,2,1],strides=[1,2,2,1],padding='SAME')
	_pool_dr1 = tf.nn.dropout(_pool1,_keepratio)
	#第二层
	_conv2 = tf.nn.conv2d(_pool_dr1,_w['wc2'],strides=[1,1,1,1],padding='SAME')
	#relu激活层
	_conv2 = tf.nn.relu(tf.nn.bias_add(_conv2,_b['bc2']))
	_pool2 = tf.nn.max_pool(_conv2,ksize=[1,2,2,1],strides=[1,2,2,1],padding='SAME')
	_pool_dr2 = tf.nn.dropout(_pool2,_keepratio)
	#转成全连接层的输入数据
	_densel = tf.reshape(_pool_dr2,[-1,_w['wf1'].get_shape().as_list()[0]])
	_fc1 = tf.nn.relu(tf.add(tf.matmul(_densel,_w['wf1']),_b['bf1']))
	_fc1_dr1 = tf.nn.dropout(_fc1,_keepratio)
	_out = tf.add(tf.matmul(_fc1_dr1,_w['out']),_b['out'])
	out = {'input':_input,'conv1':_conv1,'pool1':_pool1,'pool_dr1':_pool_dr1,'conv2':_conv2,'pool2':_pool2,
	'pool_dr2':_pool_dr2,'densel':_densel,'fc1':_fc1,'fc1_dr1':_fc1_dr1,'out':_out}
	return out


x = tf.placeholder(tf.float32,[None,n_input])
y = tf.placeholder(tf.float32,[None,n_output])
keepratio = tf.placeholder(tf.float32)

_pred = conv_basic(x,weigths,biases,keepratio)['out']
cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits_v2(labels=y,logits=_pred))
opmt = tf.train.GradientDescentOptimizer(0.001).minimize(cost)
corr = tf.equal(tf.argmax(_pred,1),tf.argmax(y,1))
accr = tf.reduce_mean(tf.cast(corr,"float"))

#初始化参数
init = tf.global_variables_initializer()
sess = tf.Session()
sess.run(init)


batch_size = 10
trianing_epochs = 50
display_step = 5

for epoch in range(trianing_epochs):
	avg_cost = 0.
	total_batch = int(mnist.train.num_examples/batch_size)

	for i in range(total_batch):
		batch_xs,batch_ys = mnist.train.next_batch(batch_size)

		sess.run(opmt,feed_dict={x:batch_xs,y:batch_ys,keepratio:0.7})

		feeds = {x:batch_xs,y:batch_ys,keepratio:1.0}
		avg_cost += sess.run(cost,feed_dict=feeds)
	avg_cost = avg_cost/total_batch

	if epoch%display_step ==0:
		print("epoch:%03d/%03d cost : %.9f "%(epoch,trianing_epochs,avg_cost))
		feeds_train = {x:batch_xs,y:batch_ys,keepratio:1.0}
		feeds_test = {x:mnist.test.images,y:mnist.test.labels,keepratio:1.0}
		train_acc = sess.run(accr,feed_dict=feeds_train)
		print("train_acc:%.3f  "%train_acc)
		test_acc = sess.run(accr,feed_dict=feeds_test)
		print("test_acc:%.3f"%test_acc)

print("dome")