TF之LoR：基于tensorflow利用逻辑回归算LoR法实现手写数字图片识别提高准确率
目录
输出结果
设计代码
输出结果
设计代码
#TF之LoR：基于tensorflow实现手写数字图片识别准确率import tensorflow as tffrom tensorflow.examples.tutorials.mnist import input_dataimport numpy as np import matplotlib.pyplot as plt mnist = input_data.read_data_sets('MNIST_data', one_hot=True)print(mnist)#设置超参数lr=0.001 #学习率training_iters=100 #训练次数batch_size=100#每轮训练数据的大小，如果一次训练5000张图片，电脑会卡死，分批次训练会更好display_step=1#tf Graph的输入x=tf.placeholder(tf.float32, [None,784])y=tf.placeholder(tf.float32, [None, 10])#设置权重和偏置w =tf.Variable(tf.zeros([784,10]))b =tf.Variable(tf.zeros([10]))#设定运行模式pred =tf.nn.softmax(tf.matmul(x,w)+b) ##设置cost function为cross entropycost =tf.reduce_mean(-tf.reduce_sum(y*tf.log(pred),reduction_indices=1))#GD算法optimizer=tf.train.GradientDescentOptimizer(lr).minimize(cost) #初始化权重init=tf.global_variables_initializer() #开始训练with tf.Session() as sess: sess.run(init)avg_cost_list=[]for epoch in range(training_iters): #输入所有训练数据avg_cost=0.total_batch=int(mnist.train.num_examples/batch_size)for i in range(total_batch): #遍历每个batch……if (epoch+1) % display_step ==0: #显示每次迭代日志print("迭代次数Epoch:","%04d" % (epoch+1),"下降值cost=","{:.9f}".format(avg_cost))avg_cost_list.append(avg_cost)print("Optimizer Finished!")print(avg_cost_list)#测试模型correct_prediction=tf.equal(tf.argmax(pred,1),tf.argmax(y,1))accuracy=tf.reduce_mean(tf.cast(correct_prediction,tf.float32))print("Accuracy:",accuracy.eval({x:mnist.test.images[:3000],y:mnist.test.labels[:3000]}))xdata=np.linspace(0,training_iters,num=len(avg_cost_list)) plt.figure() plt.plot(xdata,avg_cost_list,'r')plt.xlabel('训练轮数')plt.ylabel('损失函数')plt.title('TF之LiR：基于tensorflow实现手写数字图片识别准确率——Jason Niu')plt.show()
