keras 实现轻量级网络ShuffleNet教程


Posted in Python onJune 19, 2020

ShuffleNet是由旷世发表的一个计算效率极高的CNN架构,它是专门为计算能力非常有限的移动设备(例如,10-150 MFLOPs)而设计的。该结构利用组卷积和信道混洗两种新的运算方法,在保证计算精度的同时,大大降低了计算成本。ImageNet分类和MS COCO对象检测实验表明,在40 MFLOPs的计算预算下,ShuffleNet的性能优于其他结构,例如,在ImageNet分类任务上,ShuffleNet的top-1 error 7.8%比最近的MobileNet低。在基于arm的移动设备上,ShuffleNet比AlexNet实际加速了13倍,同时保持了相当的准确性。

Paper:ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile

Github:https://github.com/zjn-ai/ShuffleNet-keras

网络架构

组卷积

组卷积其实早在AlexNet中就用过了,当时因为GPU的显存不足因而利用组卷积分配到两个GPU上训练。简单来讲,组卷积就是将输入特征图按照通道方向均分成多个大小一致的特征图,如下图所示左面是输入特征图右面是均分后的特征图,然后对得到的每一个特征图进行正常的卷积操作,最后将输出特征图按照通道方向拼接起来就可以了。

keras 实现轻量级网络ShuffleNet教程

目前很多框架都支持组卷积,但是tensorflow真的不知道在想什么,到现在还是不支持组卷积,只能自己写,因此效率肯定不及其他框架原生支持的方法。组卷积层的代码编写思路就与上面所说的原理完全一致,代码如下。

def _group_conv(x, filters, kernel, stride, groups):
 """
 Group convolution
 # Arguments
  x: Tensor, input tensor of with `channels_last` or 'channels_first' data format
  filters: Integer, number of output channels
  kernel: An integer or tuple/list of 2 integers, specifying the
   width and height of the 2D convolution window.
  strides: An integer or tuple/list of 2 integers,
   specifying the strides of the convolution along the width and height.
   Can be a single integer to specify the same value for
   all spatial dimensions.
  groups: Integer, number of groups per channel
  
 # Returns
  Output tensor
 """
 channel_axis = 1 if K.image_data_format() == 'channels_first' else -1
 in_channels = K.int_shape(x)[channel_axis]
 
 # number of input channels per group
 nb_ig = in_channels // groups
 # number of output channels per group
 nb_og = filters // groups
 
 gc_list = []
 # Determine whether the number of filters is divisible by the number of groups
 assert filters % groups == 0
 
 for i in range(groups):
  if channel_axis == -1:
   x_group = Lambda(lambda z: z[:, :, :, i * nb_ig: (i + 1) * nb_ig])(x)
  else:
   x_group = Lambda(lambda z: z[:, i * nb_ig: (i + 1) * nb_ig, :, :])(x)
  gc_list.append(Conv2D(filters=nb_og, kernel_size=kernel, strides=stride, 
        padding='same', use_bias=False)(x_group))
  
 return Concatenate(axis=channel_axis)(gc_list)

通道混洗

通道混洗是这篇paper的重点,尽管组卷积大量减少了计算量和参数,但是通道之间的信息交流也受到了限制因而模型精度肯定会受到影响,因此作者提出通道混洗,在不增加参数量和计算量的基础上加强通道之间的信息交流,如下图所示。

keras 实现轻量级网络ShuffleNet教程

通道混洗层的代码实现很巧妙参考了别人的实现方法。通过下面的代码说明,d代表特征图的通道序号,x是经过通道混洗后的通道顺序。

>>> d = np.array([0,1,2,3,4,5,6,7,8]) 
>>> x = np.reshape(d, (3,3)) 
>>> x = np.transpose(x, [1,0]) # 转置
>>> x = np.reshape(x, (9,)) # 平铺
'[0 1 2 3 4 5 6 7 8] --> [0 3 6 1 4 7 2 5 8]'

利用keras后端实现代码:

def _channel_shuffle(x, groups):
 """
 Channel shuffle layer
 
 # Arguments
  x: Tensor, input tensor of with `channels_last` or 'channels_first' data format
  groups: Integer, number of groups per channel
  
 # Returns
  Shuffled tensor
 """
 
 if K.image_data_format() == 'channels_last':
  height, width, in_channels = K.int_shape(x)[1:]
  channels_per_group = in_channels // groups
  pre_shape = [-1, height, width, groups, channels_per_group]
  dim = (0, 1, 2, 4, 3)
  later_shape = [-1, height, width, in_channels]
 else:
  in_channels, height, width = K.int_shape(x)[1:]
  channels_per_group = in_channels // groups
  pre_shape = [-1, groups, channels_per_group, height, width]
  dim = (0, 2, 1, 3, 4)
  later_shape = [-1, in_channels, height, width]
 
 x = Lambda(lambda z: K.reshape(z, pre_shape))(x)
 x = Lambda(lambda z: K.permute_dimensions(z, dim))(x) 
 x = Lambda(lambda z: K.reshape(z, later_shape))(x)
 
 return x

ShuffleNet Unit

ShuffleNet的主要构成单元。下图中,a图为深度可分离卷积的基本架构,b图为1步长时用的单元,c图为2步长时用的单元。

keras 实现轻量级网络ShuffleNet教程

ShuffleNet架构

注意,对于第二阶段(Stage2),作者没有在第一个1×1卷积上应用组卷积,因为输入通道的数量相对较少。

keras 实现轻量级网络ShuffleNet教程

环境

Python 3.6

Tensorlow 1.13.1

Keras 2.2.4

实现

支持channel first或channel last

# -*- coding: utf-8 -*-
"""
Created on Thu Apr 25 18:26:41 2019
@author: zjn
"""
import numpy as np
from keras.callbacks import LearningRateScheduler
from keras.models import Model
from keras.layers import Input, Conv2D, Dropout, Dense, GlobalAveragePooling2D, Concatenate, AveragePooling2D
from keras.layers import Activation, BatchNormalization, add, Reshape, ReLU, DepthwiseConv2D, MaxPooling2D, Lambda
from keras.utils.vis_utils import plot_model
from keras import backend as K
from keras.optimizers import SGD
 
def _group_conv(x, filters, kernel, stride, groups):
 """
 Group convolution
 
 # Arguments
  x: Tensor, input tensor of with `channels_last` or 'channels_first' data format
  filters: Integer, number of output channels
  kernel: An integer or tuple/list of 2 integers, specifying the
   width and height of the 2D convolution window.
  strides: An integer or tuple/list of 2 integers,
   specifying the strides of the convolution along the width and height.
   Can be a single integer to specify the same value for
   all spatial dimensions.
  groups: Integer, number of groups per channel
  
 # Returns
  Output tensor
 """
 
 channel_axis = 1 if K.image_data_format() == 'channels_first' else -1
 in_channels = K.int_shape(x)[channel_axis]
 
 # number of input channels per group
 nb_ig = in_channels // groups
 # number of output channels per group
 nb_og = filters // groups
 
 gc_list = []
 # Determine whether the number of filters is divisible by the number of groups
 assert filters % groups == 0
 
 for i in range(groups):
  if channel_axis == -1:
   x_group = Lambda(lambda z: z[:, :, :, i * nb_ig: (i + 1) * nb_ig])(x)
  else:
   x_group = Lambda(lambda z: z[:, i * nb_ig: (i + 1) * nb_ig, :, :])(x)
  gc_list.append(Conv2D(filters=nb_og, kernel_size=kernel, strides=stride, 
        padding='same', use_bias=False)(x_group))
  
 return Concatenate(axis=channel_axis)(gc_list)
def _channel_shuffle(x, groups):
 """
 Channel shuffle layer
 
 # Arguments
  x: Tensor, input tensor of with `channels_last` or 'channels_first' data format
  groups: Integer, number of groups per channel
  
 # Returns
  Shuffled tensor
 """
 if K.image_data_format() == 'channels_last':
  height, width, in_channels = K.int_shape(x)[1:]
  channels_per_group = in_channels // groups
  pre_shape = [-1, height, width, groups, channels_per_group]
  dim = (0, 1, 2, 4, 3)
  later_shape = [-1, height, width, in_channels]
 else:
  in_channels, height, width = K.int_shape(x)[1:]
  channels_per_group = in_channels // groups
  pre_shape = [-1, groups, channels_per_group, height, width]
  dim = (0, 2, 1, 3, 4)
  later_shape = [-1, in_channels, height, width]
 
 x = Lambda(lambda z: K.reshape(z, pre_shape))(x)
 x = Lambda(lambda z: K.permute_dimensions(z, dim))(x) 
 x = Lambda(lambda z: K.reshape(z, later_shape))(x)
 
 return x
 
def _shufflenet_unit(inputs, filters, kernel, stride, groups, stage, bottleneck_ratio=0.25):
 """
 ShuffleNet unit
 
 # Arguments
  inputs: Tensor, input tensor of with `channels_last` or 'channels_first' data format
  filters: Integer, number of output channels
  kernel: An integer or tuple/list of 2 integers, specifying the
   width and height of the 2D convolution window.
  strides: An integer or tuple/list of 2 integers,
   specifying the strides of the convolution along the width and height.
   Can be a single integer to specify the same value for
   all spatial dimensions.
  groups: Integer, number of groups per channel
  stage: Integer, stage number of ShuffleNet
  bottleneck_channels: Float, bottleneck ratio implies the ratio of bottleneck channels to output channels
   
 # Returns
  Output tensor
  
 # Note
  For Stage 2, we(authors of shufflenet) do not apply group convolution on the first pointwise layer 
  because the number of input channels is relatively small.
 """
 channel_axis = 1 if K.image_data_format() == 'channels_first' else -1
 in_channels = K.int_shape(inputs)[channel_axis]
 bottleneck_channels = int(filters * bottleneck_ratio)
 
 if stage == 2:
  x = Conv2D(filters=bottleneck_channels, kernel_size=kernel, strides=1,
     padding='same', use_bias=False)(inputs)
 else:
  x = _group_conv(inputs, bottleneck_channels, (1, 1), 1, groups)
 x = BatchNormalization(axis=channel_axis)(x)
 x = ReLU()(x)
 
 x = _channel_shuffle(x, groups)
 x = DepthwiseConv2D(kernel_size=kernel, strides=stride, depth_multiplier=1, 
      padding='same', use_bias=False)(x)
 x = BatchNormalization(axis=channel_axis)(x)
  
 if stride == 2:
  x = _group_conv(x, filters - in_channels, (1, 1), 1, groups)
  x = BatchNormalization(axis=channel_axis)(x)
  avg = AveragePooling2D(pool_size=(3, 3), strides=2, padding='same')(inputs)
  x = Concatenate(axis=channel_axis)([x, avg])
 else:
  x = _group_conv(x, filters, (1, 1), 1, groups)
  x = BatchNormalization(axis=channel_axis)(x)
  x = add([x, inputs])
 return x
 
def _stage(x, filters, kernel, groups, repeat, stage):
 """
 Stage of ShuffleNet
 
 # Arguments
  x: Tensor, input tensor of with `channels_last` or 'channels_first' data format
  filters: Integer, number of output channels
  kernel: An integer or tuple/list of 2 integers, specifying the
   width and height of the 2D convolution window.
  strides: An integer or tuple/list of 2 integers,
   specifying the strides of the convolution along the width and height.
   Can be a single integer to specify the same value for
   all spatial dimensions.
  groups: Integer, number of groups per channel
  repeat: Integer, total number of repetitions for a shuffle unit in every stage
  stage: Integer, stage number of ShuffleNet
  
 # Returns
  Output tensor
 """
 x = _shufflenet_unit(x, filters, kernel, 2, groups, stage)
 
 for i in range(1, repeat):
  x = _shufflenet_unit(x, filters, kernel, 1, groups, stage)
 return x
 
def ShuffleNet(input_shape, classes):
 """
 ShuffleNet architectures
 
 # Arguments
  input_shape: An integer or tuple/list of 3 integers, shape
   of input tensor
  k: Integer, number of classes to predict
  
 # Returns
  A keras model
 """
 inputs = Input(shape=input_shape)
 
 x = Conv2D(24, (3, 3), strides=2, padding='same', use_bias=True, activation='relu')(inputs)
 x = MaxPooling2D(pool_size=(3, 3), strides=2, padding='same')(x)
 
 x = _stage(x, filters=384, kernel=(3, 3), groups=8, repeat=4, stage=2)
 x = _stage(x, filters=768, kernel=(3, 3), groups=8, repeat=8, stage=3)
 x = _stage(x, filters=1536, kernel=(3, 3), groups=8, repeat=4, stage=4)
 
 x = GlobalAveragePooling2D()(x)
 
 x = Dense(classes)(x)
 predicts = Activation('softmax')(x)
 model = Model(inputs, predicts)
 return model
 
if __name__ == '__main__':
 model = ShuffleNet((224, 224, 3), 1000)
 #plot_model(model, to_file='ShuffleNet.png', show_shapes=True)

以上这篇keras 实现轻量级网络ShuffleNet教程就是小编分享给大家的全部内容了,希望能给大家一个参考,也希望大家多多支持三水点靠木。

Python 相关文章推荐
一篇文章入门Python生态系统(Python新手入门指导)
Dec 11 Python
Ruby元编程基础学习笔记整理
Jul 02 Python
Python利用IPython提高开发效率
Aug 10 Python
python中实现迭代器(iterator)的方法示例
Jan 19 Python
python通过opencv实现批量剪切图片
Nov 13 Python
python正则表达式之对号入座篇
Jul 24 Python
Python实现查找字符串数组最长公共前缀示例
Mar 27 Python
python语言基本语句用法总结
Jun 11 Python
python基础 range的用法解析
Aug 23 Python
服务器端jupyter notebook映射到本地浏览器的操作
Apr 14 Python
OpenCV+python实现实时目标检测功能
Jun 24 Python
零基础学python应该从哪里入手
Aug 11 Python
Python爬虫实现HTTP网络请求多种实现方式
Jun 19 #Python
Keras设置以及获取权重的实现
Jun 19 #Python
Python包和模块的分发详细介绍
Jun 19 #Python
浅谈Keras中shuffle和validation_split的顺序
Jun 19 #Python
Python爬虫headers处理及网络超时问题解决方案
Jun 19 #Python
sklearn和keras的数据切分与交叉验证的实例详解
Jun 19 #Python
Python虚拟环境的创建和包下载过程分析
Jun 19 #Python
You might like
深入PHP与浏览器缓存的分析
2013/06/03 PHP
php htmlspecialchars()与shtmlspecialchars()函数的深入分析
2013/06/05 PHP
php中通过数组进行高效随机抽取指定条记录的算法
2013/09/09 PHP
Zend Framework入门教程之Zend_Db数据库操作详解
2016/12/08 PHP
php array_reverse 以相反的顺序返回数组实例代码
2017/04/11 PHP
JavaScript入门教程(9) Document文档对象
2009/01/31 Javascript
利用javascript实现全部删或清空所选的操作
2014/05/27 Javascript
jquery网页回到顶部效果(图标渐隐,自写)
2014/06/16 Javascript
原生js实现百叶窗效果及原理介绍
2016/04/12 Javascript
JS常见疑难点分析之match,charAt,charCodeAt,map,search用法分析
2016/12/25 Javascript
jQuery仿写百度百科的目录树
2017/01/03 Javascript
js实现产品缩略图效果
2017/03/10 Javascript
BootStrap Select清除选中的状态恢复默认状态
2017/06/20 Javascript
基于LayUI实现前端分页功能的方法
2017/07/22 Javascript
nodejs 图片预览和上传的示例代码
2017/09/30 NodeJs
vue自定义移动端touch事件之点击、滑动、长按事件
2018/07/10 Javascript
angularjs实现table表格td单元格单击变输入框/可编辑状态示例
2019/02/21 Javascript
微信小程序关键字变色实现代码实例
2019/12/13 Javascript
PHP 502bad gateway原因及解决方案
2020/11/13 Javascript
Python实现多线程HTTP下载器示例
2017/02/11 Python
Django REST framework视图的用法
2019/01/16 Python
python实现从本地摄像头和网络摄像头截取图片功能
2019/07/11 Python
Python+OpenCv制作证件图片生成器的操作方法
2019/08/21 Python
Python+appium框架原生代码实现App自动化测试详解
2020/03/06 Python
Python分析微信好友性别比例和省份城市分布比例的方法示例【基于itchat模块】
2020/05/29 Python
HTML5中通过li-canvas轻松实现单图、多图、圆角图绘制,单行文字、多行文字等
2018/11/30 HTML / CSS
俄罗斯美容和健康网上商店:Созвездие Красоты
2019/07/23 全球购物
Etam德国:内衣精品店
2019/08/25 全球购物
Java面试题:为什么要用Java
2012/05/11 面试题
剪彩仪式主持词
2014/03/19 职场文书
《大自然的语言》教学反思
2014/04/08 职场文书
2015元旦主持词开场白和结束语
2014/12/14 职场文书
小王子读书笔记
2015/06/29 职场文书
mysql 联合索引生效的条件及索引失效的条件
2021/11/20 MySQL
Redis+Lua脚本实现计数器接口防刷功能(升级版)
2022/02/12 Redis
Tomcat配置访问日志和线程数
2022/05/06 Servers