tensorflow的基本用法(八)——dropout的作用

文章作者：Tyan
博客：noahsnail.com  |  CSDN  |  简书

本文主要是介绍tensorflow中dropout的作用，dropout主要是用来防止过拟合，即提供模型的泛化能力。

#!/usr/bin/env python# _*_ coding: utf-8 _*_import tensorflow as tffrom sklearn.datasets import load_digitsfrom sklearn.cross_validation import train_test_splitfrom sklearn.preprocessing import LabelBinarizer# 加载数据 digits = load_digits()# 输入数据X = digits.data# 输出数据y = digits.target# 标签变换y = LabelBinarizer().fit_transform(y)X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=.3)# 创建一个神经网络层def add_layer(input, in_size, out_size, layer_name, activation_function = None):    """    :param input:        神经网络层的输入    :param in_zize:        输入数据的大小    :param out_size:        输出数据的大小    :param layer_name        神经网络层的名字    :param activation_function:        神经网络激活函数，默认没有    """    # 定义神经网络的初始化权重    Weights = tf.Variable(tf.random_normal([in_size, out_size]))    # 定义神经网络的偏置    biases = tf.Variable(tf.zeros([1, out_size]) + 0.1)    # 计算w*x+b    W_mul_x_plus_b = tf.matmul(input, Weights) + biases    # 进行dropout，可以注释和不注释来对比dropout的效果#   W_mul_x_plus_b = tf.nn.dropout(W_mul_x_plus_b, keep_prob)    # 根据是否有激活函数进行处理    if activation_function is None:        output = W_mul_x_plus_b    else:        output = activation_function(W_mul_x_plus_b)    # 查看权重变化    tf.summary.histogram(layer_name + '/output', output)    return output# 定义dropout的placeholderkeep_prob = tf.placeholder(tf.float32)# 输入数据64个特征xs = tf.placeholder(tf.float32, [None, 64])  # 8x8ys = tf.placeholder(tf.float32, [None, 10])# 添加隐藏层和输出层l1 = add_layer(xs, 64, 50, 'l1', activation_function=tf.nn.tanh)prediction = add_layer(l1, 50, 10, 'l2', activation_function=tf.nn.softmax)# 计算losscross_entropy = tf.reduce_mean(-tf.reduce_sum(ys * tf.log(prediction), reduction_indices=[1]))# 存储losstf.summary.scalar('loss', cross_entropy)# 神经网络训练train_step = tf.train.GradientDescentOptimizer(0.5).minimize(cross_entropy)# 定义Sessionsess = tf.Session()# 收集所有的数据merged = tf.summary.merge_all()# 将数据写入到tensorboard中train_writer = tf.summary.FileWriter("logs/train", sess.graph)test_writer = tf.summary.FileWriter("logs/test", sess.graph)# 根据tensorflow版本选择初始化函数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()# 执行初始化sess.run(init)# 进行训练迭代for i in range(500):    # 执行训练，dropout为1-0.5=0.5    sess.run(train_step, feed_dict={xs: X_train, ys: y_train, keep_prob: 0.5})    if i % 50 == 0:        # 记录损失        train_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) 

执行结果如下：

• 没有dropout

测试误差与训练误差的损失差的较大，说明模型更拟合训练数据。

• 有dropout

测试误差与训练误差相差不大，说明模型泛化能力较好。