1 # 感知机 2 3 #导入相关库并载入数据 4 from tensorflow.examples.tutorials.mnist import input_data 5 import tensorflow as tf 6 mnist = input_data.read_data_sets("MNIST_data/",one_hot = True) 7 sess = tf.InteractiveSession() 8 9 #对隐含层参数进行初始化 10 in_units = 784 11 h1_units = 300 12 w1 = tf.Variable(tf.truncated_normal([in_units,h1_units],stddev=0.1)) 13 b1 = tf.Variable(tf.zeros([h1_units])) 14 w2 = tf.Variable(tf.zeros([h1_units,10])) 15 b2 = tf.Variable(tf.zeros([10])) 16 17 # 定义输入x及dropout比率 18 x = tf.placeholder(tf.float32,[None,in_units]) 19 keep_prob = tf.placeholder(tf.float32) 20 21 22 #定义模型结构 23 hidden1 = tf.nn.relu(tf.matmul(x,w1) + b1) 24 hidden1_drop = tf.nn.dropout(hidden1, keep_prob) 25 y = tf.nn.softmax(tf.matmul(hidden1_drop,w2) + b2) 26 27 28 #定义损失函数和优化器 29 y_ = tf.placeholder(tf.float32, [None,10]) 30 cross_entropy = tf.reduce_mean(-tf.reduce_sum(y_ * tf.log(y),reduction_indices=[1])) 31 train_step = tf.train.AdagradOptimizer(0.3).minimize(cross_entropy) 32 33 34 #训练 35 tf.global_variables_initializer().run() 36 for i in range(3000): 37 batch_xs, batch_ys = mnist.train.next_batch(100) 38 train_step.run({x:batch_xs,y_:batch_ys,keep_prob:0.75}) 39 40 41 #对模型进行准确率评测 42 correct_prediction = tf.equal(tf.argmax(y,1), tf.argmax(y_,1)) 43 accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32)) 44 print(accuracy.eval({x:mnist.test.images,y_:mnist.test.labels,keep_prob:1.0}))
参考书籍：
1.《Tensorflow实战》黄文坚 唐源 著
作者：舟华520
出处：https://www.cnblogs.com/xfzh193/
本文以学习，分享，研究交流为主，欢迎转载，请标明作者出处！