1.前言

Overfitting 也被称为过度学习，过度拟合。我们总是希望在机器学习训练时，机器学习模型能在新样本上很好的表现。过拟合时，通常是因为模型过于复杂，学习器把训练样本学得“太好了”，很可能把一些训练样本自身的特性当成了所有潜在样本的共性了，这样一来模型的泛化性能就下降了。我们形象的打个比方吧，你考试复习，复习题都搞懂了，但是一到考试就不会了，那是过拟合。

2.对比drop前后的loss

2.1.导入必要模块

import tensorflow as tffrom sklearn.datasets import load_digitsfrom sklearn.model_selection import train_test_splitfrom sklearn.preprocessing import LabelBinarizer #处理标签为二进制

2.2.加载数据

digits = load_digits()X = digits.datay = digits.targety = LabelBinarizer().fit_transform(y) #转化标签为二进制形式X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=.3)

2.3.定义添加层函数

def add_layer(inputs, in_size, out_size, layer_name, activation_function=None, ):Weights = tf.Variable(tf.random_normal([in_size, out_size])) #系数biases = tf.Variable(tf.zeros([1, out_size]) + 0.1, ) #偏置Wx_plus_b = tf.matmul(inputs, Weights) + biases# here to dropoutWx_plus_b = tf.nn.dropout(Wx_plus_b, keep_prob)if activation_function is None:outputs = Wx_plus_belse:outputs = activation_function(Wx_plus_b, )tf.summary.histogram(layer_name + '/outputs', outputs)return outputs

2.4.损失函数与优化器

keep_prob = tf.placeholder(tf.float32)xs = tf.placeholder(tf.float32, [None, 64]) # 8x8ys = tf.placeholder(tf.float32, [None, 10])

这里的keep_prob是保留概率，即我们要保留的结果所占比例，它作为一个placeholder，在run时传入， 当keep_prob=1的时候，相当于100%保留，也就是dropout没有起作用。

添加隐含层和输出层：

l1 = add_layer(xs, 64, 50, 'l1', activation_function=tf.nn.tanh)prediction = add_layer(l1, 50, 10, 'l2', activation_function=tf.nn.softmax)

cross_entropy = tf.reduce_mean(-tf.reduce_sum(ys * tf.log(prediction),reduction_indices=[1])) # 交叉熵函数损失函数tf.summary.scalar('loss', cross_entropy)train_step = tf.train.GradientDescentOptimizer(0.5).minimize(cross_entropy) #优化函数sess = tf.Session()merged = tf.summary.merge_all()# summary writer goes in heretrain_writer = tf.summary.FileWriter("logs/train", sess.graph)test_writer = tf.summary.FileWriter("logs/test", sess.graph)

2.5.训练

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):# here to determine the keeping probabilitysess.run(train_step, feed_dict={xs: X_train, ys: y_train, keep_prob: 0.5})if i % 50 == 0:# record losstrain_result = sess.run(merged, feed_dict={xs: X_train, ys: y_train, keep_prob: 1})test_result = sess.run(merged, feed_dict={xs: X_test, ys: y_test, keep_prob: 1})train_writer.add_summary(train_result, i)test_writer.add_summary(test_result, i)