原理就是在第一次学习的过程中,随即忽略一些神经元和神经的链接。使得神经网络变得不完整。一次一次。。。。。每一次得出的结果不依赖某一个参数。这样就解决了过拟合问题。
import tensorflow as tf from sklearn.datasets import load_digits from sklearn.model_selection import train_test_split from sklearn.preprocessing import LabelBinarizer #load data digits = load_digits() X = digits.data y = digits.target y = LabelBinarizer().fit_transform(y) X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=.3) # 传入的参数有输入层,输入大小,输出大小,还有一个激励函数,默认是NONE(线性函数) def add_layer(inputs, in_size, out_size, layer_name, activation_function=None, ): Weights = tf.Variable(tf.random_normal([in_size, out_size])) # 定义权重为随机变量,因为随机变量生成初始变量要比0好很多。形状是【2】【3】:2行3列 # 机器学习推荐变量不为0.他的size是:1行our_size列 biases = tf.Variable(tf.zeros([1, out_size]) + 0.1) Wx_plus_b = tf.matmul(inputs, Weights) + biases # matmul是矩阵的乘法。还没被激活的值存在这个变量中 #将这个结果50%不考虑 其实就是dropout Wx_plus_b = tf.nn.dropout(Wx_plus_b, keep_prob) if activation_function is None: outputs = Wx_plus_b else: outputs = activation_function(Wx_plus_b) tf.summary.histogram(layer_name + '/outputs', outputs) return outputs #define placeholder for input.784个像素点 keep_prob = tf.placeholder(tf.float32)#需要定义一个参数,保持多少的如果不被drop掉 xs = tf.placeholder(tf.float32, [None, 64]) #X的是8X8的64个单位 ys = tf.placeholder(tf.float32, [None, 10]) #输出是十个单位,分别描述0123456789 #add output layer. softmax一般是用来做分类的函数 l1 = add_layer(xs, 64, 50, 'l1',activation_function=tf.nn.tanh) prediction = add_layer(l1, 50, 10, 'l2', activation_function=tf.nn.softmax) #the error between prediction and real data.在softmax来说,这个cross_entropy算法做分类,生成分类算法 cross_entropy = tf.reduce_mean(-tf.reduce_sum(ys * tf.log(prediction), reduction_indices=[1]))#loss tf.summary.scalar('loss', cross_entropy) train_step = tf.train.GradientDescentOptimizer(0.5).minimize(cross_entropy) sess = tf.Session() merged = tf.summary.merge_all() #summary writer goes in here train_writer = tf.summary.FileWriter("A://logs/train", sess.graph) test_writer = tf.summary.FileWriter("A://logs/test", sess.graph) #important stetp if int((tf.__version__).split('.')[1]) < 12 and int((tf.__version__).split('.')[0]) < 1: init = tf.initialize_all_variables() else: init = tf.global_variables_initializer() sess.run(init) for i in range(500): sess.run(train_step, feed_dict={xs: X_train, ys: y_train, keep_prob: 0.5})#一般会有50%的结果(被)drop if i % 50 == 0: #记录loss train_result = sess.run(merged, feed_dict={xs: X_train, ys: y_train, keep_prob: 1}) #记录result的时候不要drop任何东西 test_result = sess.run(merged, feed_dict={xs: X_test, ys: y_test, keep_prob: 1}) #加载到writer,第i次学习 train_writer.add_summary(train_result, i) test_writer.add_summary(test_result, i)