tensorflow手写数字体识别（Tensorflow深度学习笔记手写数字识别-MNIST数据测试）
tensorflow手写数字体识别（Tensorflow深度学习笔记手写数字识别-MNIST数据测试）
MNIST的结果是0-9，常用softmax函数进行分类，输出结果。
softmax函数常用于分类，定义如下：
# coding: utf-8 import tensorflow as tf from tensorflow.examples.tutorials.mnist import input_data #载入数据集 mnist = input_data.read_data_sets("MNIST_data" one_hot=True) #每个批次的大小 batch_size = 100 #计算一共有多少个批次 n_batch = mnist.train.num_examples // batch_size print( mnist.train.num_examples) #定义两个placeholder x = tf.placeholder(tf.float32 [None 784])#输入 y = tf.placeholder(tf.float32 [None 10])#输出 #创建一个简单的神经网络 W = tf.Variable(tf.zeros([784 10])) b = tf.Variable(tf.zeros([10])) prediction = tf.nn.softmax(tf.matmul(x W) b) #二次代价函数 loss = tf.reduce_mean(tf.square(y-prediction)) #使用梯度下降法 train_step = tf.train.GradientDescentOptimizer(3).minimize(loss) #train_step = tf.train.GradientDescentOptimizer(0.2).minimize(loss) #初始化变量 init = tf.global_variables_initializer() #结果存放在一个布尔型列表中 correct_prediction = tf.equal(tf.argmax(y 1) tf.argmax(prediction 1))#argmax返回一维张量中最大的值所在的位置 #求准确率 accuracy = tf.reduce_mean(tf.cast(correct_prediction tf.float32))#布尔转成float32，然后求平均值 with tf.Session() as sess: sess.run(init) for epoch in range(100): 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.test.labels}) print("Iter " str(epoch) " Testing Accuracy " str(acc)) if acc > 0.99: break #输出 # Iter 0 Testing Accuracy 0.9074 # Iter 1 Testing Accuracy 0.9168 # Iter 2 Testing Accuracy 0.9221 # Iter 3 Testing Accuracy 0.9223 # Iter 4 Testing Accuracy 0.9237 # Iter 5 Testing Accuracy 0.9248 # Iter 6 Testing Accuracy 0.9246 # Iter 7 Testing Accuracy 0.925 # Iter 8 Testing Accuracy 0.9257 # Iter 9 Testing Accuracy 0.9278 # Iter 10 Testing Accuracy 0.9268 # Iter 11 Testing Accuracy 0.928 # Iter 12 Testing Accuracy 0.9269 # Iter 13 Testing Accuracy 0.9288 # Iter 14 Testing Accuracy 0.9273 # Iter 15 Testing Accuracy 0.9282 # Iter 16 Testing Accuracy 0.9301 # Iter 17 Testing Accuracy 0.9287 # Iter 18 Testing Accuracy 0.9297 # Iter 19 Testing Accuracy 0.9296 # Iter 20 Testing Accuracy 0.9285 # Iter 21 Testing Accuracy 0.9286 # Iter 22 Testing Accuracy 0.9288 # Iter 23 Testing Accuracy 0.9285 # Iter 24 Testing Accuracy 0.9311 # Iter 25 Testing Accuracy 0.9298 # Iter 26 Testing Accuracy 0.9294 # Iter 27 Testing Accuracy 0.9299 # Iter 28 Testing Accuracy 0.9298 # Iter 29 Testing Accuracy 0.9298 # Iter 30 Testing Accuracy 0.9307 # Iter 31 Testing Accuracy 0.9305 # Iter 32 Testing Accuracy 0.9291 # Iter 33 Testing Accuracy 0.9295 # Iter 34 Testing Accuracy 0.9289 # Iter 35 Testing Accuracy 0.9301 # Iter 36 Testing Accuracy 0.93 # Iter 37 Testing Accuracy 0.9293 # Iter 38 Testing Accuracy 0.93 # Iter 39 Testing Accuracy 0.9298 # Iter 40 Testing Accuracy 0.9299 # Iter 41 Testing Accuracy 0.9304 # Iter 42 Testing Accuracy 0.9303 # Iter 43 Testing Accuracy 0.93 # Iter 44 Testing Accuracy 0.9308 # Iter 45 Testing Accuracy 0.9296 # Iter 46 Testing Accuracy 0.9291 # Iter 47 Testing Accuracy 0.9306 # Iter 48 Testing Accuracy 0.9311 # Iter 49 Testing Accuracy 0.9301 # Iter 50 Testing Accuracy 0.93 # Iter 51 Testing Accuracy 0.9301 # Iter 52 Testing Accuracy 0.9306 # Iter 53 Testing Accuracy 0.9303 # Iter 54 Testing Accuracy 0.9307 # Iter 55 Testing Accuracy 0.9295 # Iter 56 Testing Accuracy 0.9313 # Iter 57 Testing Accuracy 0.9295 # Iter 58 Testing Accuracy 0.9303 # Iter 59 Testing Accuracy 0.9299 # Iter 60 Testing Accuracy 0.9286 # Iter 61 Testing Accuracy 0.9301 # Iter 62 Testing Accuracy 0.9303 # Iter 63 Testing Accuracy 0.9289 # Iter 64 Testing Accuracy 0.9301 # Iter 65 Testing Accuracy 0.9296 # Iter 66 Testing Accuracy 0.9303 # Iter 67 Testing Accuracy 0.9313 # Iter 68 Testing Accuracy 0.9301 # Iter 69 Testing Accuracy 0.9312 # Iter 70 Testing Accuracy 0.9294 # Iter 71 Testing Accuracy 0.9283 # Iter 72 Testing Accuracy 0.9295 # Iter 73 Testing Accuracy 0.9305 # Iter 74 Testing Accuracy 0.929 # Iter 75 Testing Accuracy 0.9315 # Iter 76 Testing Accuracy 0.9306 # Iter 77 Testing Accuracy 0.9288 # Iter 78 Testing Accuracy 0.9312 # Iter 79 Testing Accuracy 0.9309 # Iter 80 Testing Accuracy 0.9298 # Iter 81 Testing Accuracy 0.9293 # Iter 82 Testing Accuracy 0.9295 # Iter 83 Testing Accuracy 0.9292 # Iter 84 Testing Accuracy 0.9291 # Iter 85 Testing Accuracy 0.9294 # Iter 86 Testing Accuracy 0.9298 # Iter 87 Testing Accuracy 0.9296 # Iter 88 Testing Accuracy 0.9301 # Iter 89 Testing Accuracy 0.9306 # Iter 90 Testing Accuracy 0.9297 # Iter 91 Testing Accuracy 0.9307 # Iter 92 Testing Accuracy 0.9289 # Iter 93 Testing Accuracy 0.931 # Iter 94 Testing Accuracy 0.9301 # Iter 95 Testing Accuracy 0.9302 # Iter 96 Testing Accuracy 0.9297 # Iter 97 Testing Accuracy 0.9299 # Iter 98 Testing Accuracy 0.9317 # Iter 99 Testing Accuracy 0.9297
转载请注明出处，Juyin@2017/11/26
