单层bp神经网络mnist手写体识别

最新推荐文章于 2023-04-12 11:24:00 发布

没有那把剑，

最新推荐文章于 2023-04-12 11:24:00 发布

阅读量252

点赞数

分类专栏： Tensorflow

本文链接：https://blog.csdn.net/qq_43065776/article/details/89597146

版权

Tensorflow 专栏收录该内容

3 篇文章 0 订阅

订阅专栏

# coding=utf8
import tensorflow as tf
import numpy as np
import matplotlib.pyplot as plt
from tensorflow.examples.tutorials.mnist import input_data
import  numpy as np
#Activation function : don't Know  没卷积没池化，我也不知道为啥，也不敢问
#lost Function : - log(p)  对数损失函数softmax分类
def sigmoid(x,deriv = False):
    if(deriv==True):
        return x*(1-x)
    return 1/(1+np.exp(-x))
mnist=input_data.read_data_sets('data/',one_hot=True)
#print
#mnist  type是dataSet ，分train，test，validation （训练，测试，验证）
trainimg=mnist.train.images
#print(np.array(trainimg).shape)   #np.ndarray ,可以转化为np.array 读
trainlab=mnist.train.labels
testimg = mnist.test.images
testlab = mnist.test.labels
# print("Train image : %s"%(trainimg.shape,))
# print("Train labels: %s"%(trainlab.shape,))
# print("Test  image : %s"%(testimg.shape,))
# print("Test  labels: %s"%(testlab.shape,))
#plt module
# nsample = 5
# randidx = np.random.randint(trainimg.shape[0],size = nsample)#取五个数，[0,55000)返回一个列表
#print("randint is ")

# for i in randidx:
#     curr_img = np.reshape(trainimg[i,:],[28,28])   #
#     #print("curr_img is ")
#     #print(curr_img)
#     curr_lable = np.argmax(trainlab[i,:])
#     #print("curr_lable is :")
#     #print(curr_lable)
#     plt.matshow(curr_img, cmap = plt.get_cmap('gray'))
#     plt.title(str(i)+"th Train Data."+ " Label id "+str(curr_lable))
#     #print (str(i)+"th Train Data."+ " Label id "+str(curr_lable))
#     plt.show()
#  plt在上面
x = tf.placeholder('float',[None,784])
y = tf.placeholder('float',[None,10])
W = tf.Variable(tf.zeros([784,10]))
b = tf.zeros([10])
yo = tf.nn.softmax(tf.matmul(x,W)+b)  # wx+b  = yo
#print("actv is :",yo)
learing_reat = 0.01
cost = tf.reduce_mean(tf.reduce_sum(-y*tf.log(yo),reduction_indices=1))  # - log(p)

optm = tf.train.GradientDescentOptimizer(learing_reat).minimize(cost)

#prediction is correct ?
pred = tf.equal(tf.argmax(yo,1),tf.argmax(y,1))
# Accuracy
accr = tf.reduce_mean(tf.cast(pred,'float'))
# print(accr)
# accr = tf.cast(pred,'float')  错了？
# print(accr)

train_number = 51   #迭代次数
batch_size = 100    #分批次处理
display_step = 5    #显示频率
init = tf.global_variables_initializer()
sess = tf.Session()
sess.run(init)
for step in range(train_number):
    avg_cost=0
    numbatch = int(mnist.train.num_examples/batch_size)  # 批数
    for i in range(numbatch):
        batch_xs ,batch_ys = mnist.train.next_batch(batch_size)
        feeds = {x: batch_xs, y: batch_ys}
        sess.run(optm,feed_dict=feeds)
        avg_cost += sess.run(cost,feed_dict=feeds)/numbatch
    if step % display_step == 0:
        feeds_train = {x:batch_xs,y:batch_ys}
        feed_test = {x:mnist.test.images,y:mnist.test.labels}
        train_acc = sess.run(accr, feed_dict=feeds_train)
        test_acc = sess.run(accr, feed_dict = feed_test)
        print("Epoch: %03d/%03d cost: %9f train_acc: %3f  test_acc: %3f"%(step,train_number,avg_cost,train_acc,test_acc))
sess.close()

萌新微菜，欢迎讨论