Google InceptionNet介绍

可以看到每个分支都采用了1*1的卷积网络，因为这是一个优秀的网络，可以跨通道组织信息，提高网络的表达能力，提供更多的非线性变换，性价比很高。同时网络中的卷积和大小也不一样，可以增加网络对不同尺度的适应性。所以，InceptionNet通过分支的方式增大网络的宽度和深度能够很好的提高网络的性能，避免过拟合。

    InceptionNet有22层，除了最后一层，其中间节点的分类效果也非常好，因此在这个网络中，还使用到了辅助分类节点，即将中间的某一层的输出用作分类，并按一个较小的权重（0.3）加到最终的分类结果中。

    提出并采用了著名的BatchNormalization(BN)。BN是一种非常有用的正则化方法，可以让卷积网络的训练速度加快很多，同时收敛后的分类准确率也可以提高很多。BN就是神经网络在训练的时候对每个Batch数据的内部进行标准化，使输出规范化到N(0,1)的正态分布。BN在某种程度上还起到正则的作用，所以可以减小或者取消Dropout，优化网络结构。同时添加BN后，还需要对超参数做一些调整，增大学速率，加快学习速率的衰减，去除Drought，减轻正则，去除LRN，对样本进行洗牌等等。

4 代码实现

#%%
# Copyright 2015 The TensorFlow Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ==============================================================================
import tensorflow as tfslim = tf.contrib.slim
#产生截断的正态分布
trunc_normal = lambda stddev: tf.truncated_normal_initializer(0.0, stddev)#生成Inception V3的卷积部分
def inception_v3_base(inputs, scope=None):
'''
inputs: 图片的tensor
scope: 函数默认的参数环境
'''#保存某些关键节点的信息end_points = {}#定义一个变量的命名空间，可以使变量有相同的名称with tf.variable_scope(scope, 'InceptionV3', [inputs]):#为参数设置默认值with slim.arg_scope([slim.conv2d, slim.max_pool2d, slim.avg_pool2d],stride=1, padding='VALID'):#用小尺寸卷积和1*1的卷积，减小参数，增加非线性# 299 x 299 x 3net = slim.conv2d(inputs, 32, [3, 3], stride=2, scope='Conv2d_1a_3x3')# 149 x 149 x 32net = slim.conv2d(net, 32, [3, 3], scope='Conv2d_2a_3x3')# 147 x 147 x 32net = slim.conv2d(net, 64, [3, 3], padding='SAME', scope='Conv2d_2b_3x3')# 147 x 147 x 64net = slim.max_pool2d(net, [3, 3], stride=2, scope='MaxPool_3a_3x3')# 73 x 73 x 64net = slim.conv2d(net, 80, [1, 1], scope='Conv2d_3b_1x1')# 73 x 73 x 80.net = slim.conv2d(net, 192, [3, 3], scope='Conv2d_4a_3x3')# 71 x 71 x 192.net = slim.max_pool2d(net, [3, 3], stride=2, scope='MaxPool_5a_3x3')# 35 x 35 x 192.#大小大幅压缩，通道大幅增加，增加抽象性# Inception blockswith slim.arg_scope([slim.conv2d, slim.max_pool2d, slim.avg_pool2d],stride=1, padding='SAME'):# mixed: 35 x 35 x 256.with tf.variable_scope('Mixed_5b'):with tf.variable_scope('Branch_0'):branch_0 = slim.conv2d(net, 64, [1, 1], scope='Conv2d_0a_1x1')with tf.variable_scope('Branch_1'):branch_1 = slim.conv2d(net, 48, [1, 1], scope='Conv2d_0a_1x1')branch_1 = slim.conv2d(branch_1, 64, [5, 5], scope='Conv2d_0b_5x5')with tf.variable_scope('Branch_2'):branch_2 = slim.conv2d(net, 64, [1, 1], scope='Conv2d_0a_1x1')branch_2 = slim.conv2d(branch_2, 96, [3, 3], scope='Conv2d_0b_3x3')branch_2 = slim.conv2d(branch_2, 96, [3, 3], scope='Conv2d_0c_3x3')with tf.variable_scope('Branch_3'):branch_3 = slim.avg_pool2d(net, [3, 3], scope='AvgPool_0a_3x3')branch_3 = slim.conv2d(branch_3, 32, [1, 1], scope='Conv2d_0b_1x1')#将4个通道合并，合并维度为3，即在输出通道上合并net = tf.concat([branch_0, branch_1, branch_2, branch_3], 3)# mixed_1: 35 x 35 x 288.with tf.variable_scope('Mixed_5c'):with tf.variable_scope('Branch_0'):branch_0 = slim.conv2d(net, 64, [1, 1], scope='Conv2d_0a_1x1')with tf.variable_scope('Branch_1'):branch_1 = slim.conv2d(net, 48, [1, 1], scope='Conv2d_0b_1x1')branch_1 = slim.conv2d(branch_1, 64, [5, 5], scope='Conv_1_0c_5x5')with tf.variable_scope('Branch_2'):branch_2 = slim.conv2d(net, 64, [1, 1], scope='Conv2d_0a_1x1')branch_2 = slim.conv2d(branch_2, 96, [3, 3], scope='Conv2d_0b_3x3')branch_2 = slim.conv2d(branch_2, 96, [3, 3], scope='Conv2d_0c_3x3')with tf.variable_scope('Branch_3'):branch_3 = slim.avg_pool2d(net, [3, 3], scope='AvgPool_0a_3x3')branch_3 = slim.conv2d(branch_3, 64, [1, 1], scope='Conv2d_0b_1x1')net = tf.concat([branch_0, branch_1, branch_2, branch_3], 3)# mixed_2: 35 x 35 x 288.with tf.variable_scope('Mixed_5d'):with tf.variable_scope('Branch_0'):branch_0 = slim.conv2d(net, 64, [1, 1], scope='Conv2d_0a_1x1')with tf.variable_scope('Branch_1'):branch_1 = slim.conv2d(net, 48, [1, 1], scope='Conv2d_0a_1x1')branch_1 = slim.conv2d(branch_1, 64, [5, 5], scope='Conv2d_0b_5x5')with tf.variable_scope('Branch_2'):branch_2 = slim.conv2d(net, 64, [1, 1], scope='Conv2d_0a_1x1')branch_2 = slim.conv2d(branch_2, 96, [3, 3], scope='Conv2d_0b_3x3')branch_2 = slim.conv2d(branch_2, 96, [3, 3], scope='Conv2d_0c_3x3')with tf.variable_scope('Branch_3'):branch_3 = slim.avg_pool2d(net, [3, 3], scope='AvgPool_0a_3x3')branch_3 = slim.conv2d(branch_3, 64, [1, 1], scope='Conv2d_0b_1x1')net = tf.concat([branch_0, branch_1, branch_2, branch_3], 3)#网络每经过一个Inception Mode输出尺寸变化不大，但是特征都被重新精炼了一遍#丰富的卷积和非线性化对网络的提升很大# mixed_3: 17 x 17 x 768.with tf.variable_scope('Mixed_6a'):with tf.variable_scope('Branch_0'):branch_0 = slim.conv2d(net, 384, [3, 3], stride=2,padding='VALID', scope='Conv2d_1a_1x1')with tf.variable_scope('Branch_1'):branch_1 = slim.conv2d(net, 64, [1, 1], scope='Conv2d_0a_1x1')branch_1 = slim.conv2d(branch_1, 96, [3, 3], scope='Conv2d_0b_3x3')branch_1 = slim.conv2d(branch_1, 96, [3, 3], stride=2,padding='VALID', scope='Conv2d_1a_1x1')with tf.variable_scope('Branch_2'):branch_2 = slim.max_pool2d(net, [3, 3], stride=2, padding='VALID',scope='MaxPool_1a_3x3')net = tf.concat([branch_0, branch_1, branch_2], 3)# mixed4: 17 x 17 x 768.with tf.variable_scope('Mixed_6b'):with tf.variable_scope('Branch_0'):branch_0 = slim.conv2d(net, 192, [1, 1], scope='Conv2d_0a_1x1')with tf.variable_scope('Branch_1'):branch_1 = slim.conv2d(net, 128, [1, 1], scope='Conv2d_0a_1x1')#串联1*7和7*1卷积网络，相当于7*7的卷积网络，参数减小，减小过拟合branch_1 = slim.conv2d(branch_1, 128, [1, 7], scope='Conv2d_0b_1x7')branch_1 = slim.conv2d(branch_1, 192, [7, 1], scope='Conv2d_0c_7x1')with tf.variable_scope('Branch_2'):branch_2 = slim.conv2d(net, 128, [1, 1], scope='Conv2d_0a_1x1')branch_2 = slim.conv2d(branch_2, 128, [7, 1], scope='Conv2d_0b_7x1')branch_2 = slim.conv2d(branch_2, 128, [1, 7], scope='Conv2d_0c_1x7')branch_2 = slim.conv2d(branch_2, 128, [7, 1], scope='Conv2d_0d_7x1')branch_2 = slim.conv2d(branch_2, 192, [1, 7], scope='Conv2d_0e_1x7')with tf.variable_scope('Branch_3'):branch_3 = slim.avg_pool2d(net, [3, 3], scope='AvgPool_0a_3x3')branch_3 = slim.conv2d(branch_3, 192, [1, 1], scope='Conv2d_0b_1x1')net = tf.concat([branch_0, branch_1, branch_2, branch_3], 3)# mixed_5: 17 x 17 x 768.with tf.variable_scope('Mixed_6c'):with tf.variable_scope('Branch_0'):branch_0 = slim.conv2d(net, 192, [1, 1], scope='Conv2d_0a_1x1')with tf.variable_scope('Branch_1'):branch_1 = slim.conv2d(net, 160, [1, 1], scope='Conv2d_0a_1x1')branch_1 = slim.conv2d(branch_1, 160, [1, 7], scope='Conv2d_0b_1x7')branch_1 = slim.conv2d(branch_1, 192, [7, 1], scope='Conv2d_0c_7x1')with tf.variable_scope('Branch_2'):branch_2 = slim.conv2d(net, 160, [1, 1], scope='Conv2d_0a_1x1')branch_2 = slim.conv2d(branch_2, 160, [7, 1], scope='Conv2d_0b_7x1')branch_2 = slim.conv2d(branch_2, 160, [1, 7], scope='Conv2d_0c_1x7')branch_2 = slim.conv2d(branch_2, 160, [7, 1], scope='Conv2d_0d_7x1')branch_2 = slim.conv2d(branch_2, 192, [1, 7], scope='Conv2d_0e_1x7')with tf.variable_scope('Branch_3'):branch_3 = slim.avg_pool2d(net, [3, 3], scope='AvgPool_0a_3x3')branch_3 = slim.conv2d(branch_3, 192, [1, 1], scope='Conv2d_0b_1x1')net = tf.concat([branch_0, branch_1, branch_2, branch_3], 3)# mixed_6: 17 x 17 x 768.with tf.variable_scope('Mixed_6d'):with tf.variable_scope('Branch_0'):branch_0 = slim.conv2d(net, 192, [1, 1], scope='Conv2d_0a_1x1')with tf.variable_scope('Branch_1'):branch_1 = slim.conv2d(net, 160, [1, 1], scope='Conv2d_0a_1x1')branch_1 = slim.conv2d(branch_1, 160, [1, 7], scope='Conv2d_0b_1x7')branch_1 = slim.conv2d(branch_1, 192, [7, 1], scope='Conv2d_0c_7x1')with tf.variable_scope('Branch_2'):branch_2 = slim.conv2d(net, 160, [1, 1], scope='Conv2d_0a_1x1')branch_2 = slim.conv2d(branch_2, 160, [7, 1], scope='Conv2d_0b_7x1')branch_2 = slim.conv2d(branch_2, 160, [1, 7], scope='Conv2d_0c_1x7')branch_2 = slim.conv2d(branch_2, 160, [7, 1], scope='Conv2d_0d_7x1')branch_2 = slim.conv2d(branch_2, 192, [1, 7], scope='Conv2d_0e_1x7')with tf.variable_scope('Branch_3'):branch_3 = slim.avg_pool2d(net, [3, 3], scope='AvgPool_0a_3x3')branch_3 = slim.conv2d(branch_3, 192, [1, 1], scope='Conv2d_0b_1x1')net = tf.concat([branch_0, branch_1, branch_2, branch_3], 3)# mixed_7: 17 x 17 x 768.with tf.variable_scope('Mixed_6e'):with tf.variable_scope('Branch_0'):branch_0 = slim.conv2d(net, 192, [1, 1], scope='Conv2d_0a_1x1')with tf.variable_scope('Branch_1'):branch_1 = slim.conv2d(net, 192, [1, 1], scope='Conv2d_0a_1x1')branch_1 = slim.conv2d(branch_1, 192, [1, 7], scope='Conv2d_0b_1x7')branch_1 = slim.conv2d(branch_1, 192, [7, 1], scope='Conv2d_0c_7x1')with tf.variable_scope('Branch_2'):branch_2 = slim.conv2d(net, 192, [1, 1], scope='Conv2d_0a_1x1')branch_2 = slim.conv2d(branch_2, 192, [7, 1], scope='Conv2d_0b_7x1')branch_2 = slim.conv2d(branch_2, 192, [1, 7], scope='Conv2d_0c_1x7')branch_2 = slim.conv2d(branch_2, 192, [7, 1], scope='Conv2d_0d_7x1')branch_2 = slim.conv2d(branch_2, 192, [1, 7], scope='Conv2d_0e_1x7')with tf.variable_scope('Branch_3'):branch_3 = slim.avg_pool2d(net, [3, 3], scope='AvgPool_0a_3x3')branch_3 = slim.conv2d(branch_3, 192, [1, 1], scope='Conv2d_0b_1x1')net = tf.concat([branch_0, branch_1, branch_2, branch_3], 3)#存储Mixed_6e辅助模型的分类end_points['Mixed_6e'] = net# mixed_8: 8 x 8 x 1280.with tf.variable_scope('Mixed_7a'):with tf.variable_scope('Branch_0'):branch_0 = slim.conv2d(net, 192, [1, 1], scope='Conv2d_0a_1x1')branch_0 = slim.conv2d(branch_0, 320, [3, 3], stride=2,padding='VALID', scope='Conv2d_1a_3x3')with tf.variable_scope('Branch_1'):branch_1 = slim.conv2d(net, 192, [1, 1], scope='Conv2d_0a_1x1')branch_1 = slim.conv2d(branch_1, 192, [1, 7], scope='Conv2d_0b_1x7')branch_1 = slim.conv2d(branch_1, 192, [7, 1], scope='Conv2d_0c_7x1')branch_1 = slim.conv2d(branch_1, 192, [3, 3], stride=2,padding='VALID', scope='Conv2d_1a_3x3')with tf.variable_scope('Branch_2'):branch_2 = slim.max_pool2d(net, [3, 3], stride=2, padding='VALID',scope='MaxPool_1a_3x3')net = tf.concat([branch_0, branch_1, branch_2], 3)# mixed_9: 8 x 8 x 2048.with tf.variable_scope('Mixed_7b'):with tf.variable_scope('Branch_0'):branch_0 = slim.conv2d(net, 320, [1, 1], scope='Conv2d_0a_1x1')with tf.variable_scope('Branch_1'):branch_1 = slim.conv2d(net, 384, [1, 1], scope='Conv2d_0a_1x1')#合并1*3和3*1的2个分支branch_1 = tf.concat([slim.conv2d(branch_1, 384, [1, 3], scope='Conv2d_0b_1x3'),slim.conv2d(branch_1, 384, [3, 1], scope='Conv2d_0b_3x1')], 3)with tf.variable_scope('Branch_2'):branch_2 = slim.conv2d(net, 448, [1, 1], scope='Conv2d_0a_1x1')branch_2 = slim.conv2d(branch_2, 384, [3, 3], scope='Conv2d_0b_3x3')branch_2 = tf.concat([slim.conv2d(branch_2, 384, [1, 3], scope='Conv2d_0c_1x3'),slim.conv2d(branch_2, 384, [3, 1], scope='Conv2d_0d_3x1')], 3)with tf.variable_scope('Branch_3'):branch_3 = slim.avg_pool2d(net, [3, 3], scope='AvgPool_0a_3x3')branch_3 = slim.conv2d(branch_3, 192, [1, 1], scope='Conv2d_0b_1x1')net = tf.concat([branch_0, branch_1, branch_2, branch_3], 3)# mixed_10: 8 x 8 x 2048.with tf.variable_scope('Mixed_7c'):with tf.variable_scope('Branch_0'):branch_0 = slim.conv2d(net, 320, [1, 1], scope='Conv2d_0a_1x1')with tf.variable_scope('Branch_1'):branch_1 = slim.conv2d(net, 384, [1, 1], scope='Conv2d_0a_1x1')branch_1 = tf.concat([slim.conv2d(branch_1, 384, [1, 3], scope='Conv2d_0b_1x3'),slim.conv2d(branch_1, 384, [3, 1], scope='Conv2d_0c_3x1')], 3)with tf.variable_scope('Branch_2'):branch_2 = slim.conv2d(net, 448, [1, 1], scope='Conv2d_0a_1x1')branch_2 = slim.conv2d(branch_2, 384, [3, 3], scope='Conv2d_0b_3x3')branch_2 = tf.concat([slim.conv2d(branch_2, 384, [1, 3], scope='Conv2d_0c_1x3'),slim.conv2d(branch_2, 384, [3, 1], scope='Conv2d_0d_3x1')], 3)with tf.variable_scope('Branch_3'):branch_3 = slim.avg_pool2d(net, [3, 3], scope='AvgPool_0a_3x3')branch_3 = slim.conv2d(branch_3, 192, [1, 1], scope='Conv2d_0b_1x1')net = tf.concat([branch_0, branch_1, branch_2, branch_3], 3)#返回卷积模块的最终输出，和end_pointreturn net, end_pointsdef inception_v3(inputs,num_classes=1000,is_training=True,dropout_keep_prob=0.8,prediction_fn=slim.softmax,spatial_squeeze=True,reuse=None,scope='InceptionV3'):
'''
构建 Inception V3网络
inputs 输入图片的tensor
num_classes 预测类别数
is_traininf 是否训练，对drop_out和BN有影响
dropout_keep_prob drop比例
prediction_fn 分类函数
spatial_squeeze 是否对输出去除1的维度，5*3*1变成 5*3
reuse 网络参数复用信息
scope 变量空间
'''with tf.variable_scope(scope, 'InceptionV3', [inputs, num_classes],reuse=reuse) as scope:with slim.arg_scope([slim.batch_norm, slim.dropout],is_training=is_training):#拿到卷积网络的输出net, end_points = inception_v3_base(inputs, scope=scope)# Auxiliary Head logits 设置默认参数with slim.arg_scope([slim.conv2d, slim.max_pool2d, slim.avg_pool2d],stride=1, padding='SAME'):aux_logits = end_points['Mixed_6e']with tf.variable_scope('AuxLogits'):#17*17*768aux_logits = slim.avg_pool2d(aux_logits, [5, 5], stride=3, padding='VALID',scope='AvgPool_1a_5x5')#5*5*768aux_logits = slim.conv2d(aux_logits, 128, [1, 1],scope='Conv2d_1b_1x1')# Shape of feature map before the final layer.#5*5*768aux_logits = slim.conv2d(aux_logits, 768, [5,5],weights_initializer=trunc_normal(0.01),padding='VALID', scope='Conv2d_2a_5x5')#1*1*768aux_logits = slim.conv2d(aux_logits, num_classes, [1, 1], activation_fn=None,normalizer_fn=None, weights_initializer=trunc_normal(0.001),scope='Conv2d_2b_1x1')#1*1*1000 去除前两个为1的维度if spatial_squeeze:aux_logits = tf.squeeze(aux_logits, [1, 2], name='SpatialSqueeze')#存储辅助分类节点end_points['AuxLogits'] = aux_logits# Final pooling and predictionwith tf.variable_scope('Logits'):net = slim.avg_pool2d(net, [8, 8], padding='VALID',scope='AvgPool_1a_8x8')# 1 x 1 x 2048net = slim.dropout(net, keep_prob=dropout_keep_prob, scope='Dropout_1b')#存储辅助分类节点end_points['PreLogits'] = net# 2048logits = slim.conv2d(net, num_classes, [1, 1], activation_fn=None,normalizer_fn=None, scope='Conv2d_1c_1x1')if spatial_squeeze:logits = tf.squeeze(logits, [1, 2], name='SpatialSqueeze')# 1000end_points['Logits'] = logitsend_points['Predictions'] = prediction_fn(logits, scope='Predictions')return logits, end_pointsdef inception_v3_arg_scope(weight_decay=0.00004,stddev=0.1,batch_norm_var_collection='moving_vars'):
'''
L2 的权重衰减为0.00004
权重初始化的方差为0.1'''batch_norm_params = {# 定义batch normalization（标准化）的参数字典'decay': 0.9997, # 定义参数衰减系数'epsilon': 0.001,'updates_collections': tf.GraphKeys.UPDATE_OPS,'variables_collections': {'beta': None,'gamma': None,'moving_mean': [batch_norm_var_collection],'moving_variance': [batch_norm_var_collection],}}# 使用slim.arg_scope后就不需要每次都重复设置参数了，只需要在有修改时设置with slim.arg_scope([slim.conv2d, slim.fully_connected],weights_regularizer=slim.l2_regularizer(weight_decay)):# 嵌套一个slim.arg_scope对卷积层生成函数slim.conv2d的几个参数赋予默认值with slim.arg_scope([slim.conv2d],weights_initializer=trunc_normal(stddev),activation_fn=tf.nn.relu,normalizer_fn=slim.batch_norm,# 标准化器的参数设置为前面定义的batch_norm_paramsnormalizer_params=batch_norm_params) as sc:return scfrom datetime import datetime
import math
import time
def time_tensorflow_run(session, target, info_string):num_steps_burn_in = 10total_duration = 0.0total_duration_squared = 0.0for i in range(num_batches + num_steps_burn_in):start_time = time.time()_ = session.run(target)duration = time.time() - start_timeif i >= num_steps_burn_in:if not i % 10:print ('%s: step %d, duration = %.3f' %(datetime.now(), i - num_steps_burn_in, duration))total_duration += durationtotal_duration_squared += duration * durationmn = total_duration / num_batchesvr = total_duration_squared / num_batches - mn * mnsd = math.sqrt(vr)print ('%s: %s across %d steps, %.3f +/- %.3f sec / batch' %(datetime.now(), info_string, num_batches, mn, sd))batch_size = 32
height, width = 299, 299
inputs = tf.random_uniform((batch_size, height, width, 3))
with slim.arg_scope(inception_v3_arg_scope()):logits, end_points = inception_v3(inputs, is_training=False)init = tf.global_variables_initializer()
sess = tf.Session()
sess.run(init)  
num_batches=100
time_tensorflow_run(sess, logits, "Forward")

5 参考文献

[1]黄文坚,唐源.TensorFlow实战[M].北京:电子工业出版社,2017.

[2] https://arxiv.org/abs/1512.00567

[3]https://github.com/tensorflow/models/blob/master/research/slim/nets/inception_v3_test.py