>- **🍨 本文为[🔗365天深度学习训练营]中的学习记录博客**
>- **🍖 原作者：[K同学啊]**
📌本周任务：📌
– 1.请根据本文TensorFlow代码，编写出相应的Pytorch代码（建议使用上周的数据测试一下模型是否构建正确）
– 2.了解ResNetV2与ResNetV的区别
– 3.改进思路是否可以迁移到其他地方呢（自由探索）
🚀我的环境：
- 语言环境：Python3.11.7
- 编译器：jupyter notebook
- 深度学习框架：TensorFlow2.13.0
一、论文解读
论文原文，何凯明大神在这篇论文中提出了一种新的残差单元。我们将这篇论文中的ResNet结构称为ResNetV2：
Identity Mappings in Deep Residual Networks.pdf
论文针对ResNet的中残差和恒等映射进行了进一步的分析，提出了一个改进版本ResNetV2，不过本人认为大多数情况下用原来的ResNet50或者ResNet101就已经够用，ResNetV2主要是针对CNN特别特别深时的改进，如大于100层，到1000层时，这时候再换ResNetV2即可。
1、ResNetV2结构与ResNet结构对比
为了构建 f(yl)=yl 的恒等映射，本文将激活函数（ReLU 和 BN）移到权值层（Conv）之前，形成一种“预激活（pre-activation）”的方式，而不是常规的“后激活（post-activation）”方式，这样就设计出了一种新的残差单元（见图 1(b)）。基于这种新的单元我们在 CIFAR-10/100 数据集上使用1001 层残差网络进行训练，发现新的残差网络比之前（ResNet）的更容易训练并且泛化性能更好。同时还考察了 200 层新残差网络在 ImageNet 上的表现，原先的残差网络在这个层数之后开始出现过拟合的现象。这些结果表明网络深度这个维度还有很大探索空间，毕竟深度是现代神经网络成功的关键。
**🧲 改进点： **(a)original表示原始的ResNet的残差结构，(b)proposed表示新的ResNet的残差结构。主要差别就是(a)结构先卷积后进行BN和激活函数计算，最后执行addition后再进行ReLU计算；(b)结构先进性BN和激活函数计算后卷积，把addition后的ReLU计算放到了残差结构内部。
📌 改进结果：作者使用这两种不同的结构在CIFAR-10数据集上做测试，模型用的是1001层的ResNet模型。从图中的结果我们可以看出，(b)proposed的测试集错误率明显更低一些，达到了4.92%的错误率。(a)original的测试集错误率是7.61%。
2、关于残差结构的不同尝试
(b-f)中的快捷连接被不同的组件阻碍。为了简化插图，我们不显示BN层，这里所有的单位均采用权值层之后的BN层。图中(a-f)都是作者对残差结构的shortcut部分进行的不同尝试，作者对不同shortcut结构的尝试结果如下表所示。
作者用不同的shortcut结构的ResNet-110在CIFAR-10数据集上做测试，发现最原始的(a)original结构是最好的，也就是identity mapping恒等映射是最好的。
3、关于激活的尝试
其中weight指conv层，BN指Batch Normalization层，ReLU指激活层，addition指相加；
对于每个图右侧部分我们称作“residual”分支，左侧部分我们称作“identity”分支。
对于图（3），如果ReLU作为“residual”分支的结尾，我们不难发现“residual”分支的结果永远非负，这样前向的时候输入会单调递增，从而会影响特征的表达能力，所以我们希望“residual”分支的结果应该在（-， +）；这点也是我们以后设计网络时所要注意的。
对于图（3）不OK的情况，那如果把BN也挪到addition后面呢？如图（2），同时也保证了“residual”分支的取值范围。但这里BN改变了“identity”分支的分布，影响了信息的传递，在训练的时候会阻碍loss的下降。
ResNet要尽量保证两点：1）不轻易改变”identity“分支的值，也就是输入与输出一致；2）addition之后不再接改变信息分布的层；所以图（2）的结构会阻碍反向传播时的信息。
在分析图（4）和图（5）之前，我们引出一个概念：”Post-activation”和”Pre-activation”，其中Post和Pre的概念是相对于weight(conv)层来说的，那么我们不难发现，图(1), (2), (3)都是”Post-activation”，图(4), (5)都是”Pre-activation”，那么两种模式哪一个更好呢？这里我们就用实验结果说话。
上图是5种结构在Cifar10上的实验结果，一共实验了两种网络ResNet110和ResNet164
从实验结果上，我们可以发现图(4)的结构与ResNet原结构伯仲之间，稍稍逊色，然而图(5)的结构却好于ResNet原结构。图5的结构好的原因在于两点：1）反向传播基本符合假设，信息传递无阻碍；2）BN层作为pre-activation，起到了正则化的作用；
最后我们通常把图5的结构称作ResNetV2。这里我们把ResNetV1和ResNetV2结构再次show：
更多详尽内容可参考：
Backbone 网络-ResNetv2 论文解读 - 知乎 (zhihu.com)
[DL-架构-ResNet系] 002 ResNet-v2 - 知乎 (zhihu.com)
卷积神经网络的网络结构——ResNet V2_resnetv2结构-CSDN博客
Identity Mappings in Deep Residual Networks（译）-CSDN博客
二、前期工作
1、设置CPU（也可以是GPU）
import tensorflow as tf
gpus=tf.config.list_physical_devices("GPU")
if gpus:
tf.config.experimental.set_memory_growth(gpus[0],True)
tf.config.set_visible_devices([gpus[0]],"GPU")2、导入数据
再次使用J1中的数据
import pathlib
data_dir=r"D:\THE MNIST DATABASE\J-series\J1\bird_photos"
data_dir=pathlib.Path(data_dir)3、查看数据
image_count=len(list(data_dir.glob('*/*')))
print("图片总数为：",image_count)运行结果：
图片总数为： 565查看数据集中的图片
import PIL
roses=list(data_dir.glob('Black Skimmer\*.jpg'))
PIL.Image.open(str(roses[0]))运行结果：
三、数据预处理
| 文件夹 | 数量 |
| Bananaquit | 166张 |
| Black Skimmer | 111张 |
| Black Throated Bushtiti | 122张 |
| Cockatoo | 166张 |
1、加载数据
加载训练集：
train_ds=tf.keras.preprocessing.image_dataset_from_directory(
data_dir,
validation_split=0.2,
subset="training",
seed=123,
image_size=(224,224),
batch_size=8
)运行结果：
Found 565 files belonging to 4 classes.
Using 452 files for training.加载验证集：
val_ds=tf.keras.preprocessing.image_dataset_from_directory(
data_dir,
validation_split=0.2,
subset="validation",
seed=123,
image_size=(224,224),
batch_size=8
)运行结果：
Found 565 files belonging to 4 classes.
Using 113 files for validation.查看分类名称
class_names=train_ds.class_names
class_names运行结果：
['Bananaquit', 'Black Skimmer', 'Black Throated Bushtiti', 'Cockatoo']2、可视化数据
import matplotlib.pyplot as plt
plt.rcParams['font.sans-serif']=['SimHei'] #正常显示中文标签
plt.rcParams['axes.unicode_minus']=False #正常显示负号
plt.figure(figsize=(10,5))
plt.suptitle("OreoCC的案例")
for images,labels in train_ds.take(1):
for i in range(8):
ax=plt.subplot(2,4,i+1)
plt.imshow(images[i].numpy().astype("uint8"))
plt.title(class_names[labels[i]])
plt.axis("off")运行结果：
单独查看其中一张图片：
plt.imshow(images[6].numpy().astype("uint8"))运行结果：
3、再次检查数据
for image_batch,labels_batch in train_ds:
print(image_batch.shape)
print(labels_batch.shape)
break运行结果：
(8, 224, 224, 3)
(8,)image_batch是形状的张量(8,224,224,3)。这是一批形状224*224*4的8张图片（最后一维指的是彩色通道RGB）
labels_batch是形状(8,)的张量，这些标签对应8张图片。
4、配置数据集
shuffle() ： 打乱数据，关于此函数的详细介绍可以参考：https://zhuanlan.zhihu.com/p/42417456
prefetch() ：预取数据，加速运行，其详细介绍可以参考我前两篇文章，里面都有讲解。
cache() ：将数据集缓存到内存当中，加速运行
AUTOTUNE=tf.data.AUTOTUNE
train_ds=train_ds.cache().shuffle(1000).prefetch(buffer_size=AUTOTUNE)
val_ds=val_ds.cache().prefetch(buffer_size=AUTOTUNE)四、构建ResNet-50v2网络模型
官方调用：
tf.keras.applications.resnet_v2.ResNet50V2(
include_top=True,
weights='imagenet',
input_tensor=None,
input_shape=None,
pooling=None,
classes=1000,
classifier_activation='softmax'
)因本人单位网络因素，代码运行后自动下载官方模型由于超时而失败，后联系其他同学通过地址帮忙下载成功。已将ResNet50v2的官方模型放于置顶处，如有需要可随时下载使用。也可自行通过官方地址下载：
https://storage.googleapis.com/tensorflow/keras-applications/resnet/resnet50v2_weights_tf_dim_ordering_tf_kernels.h5
注意：ResNet50V2、ResNet101V2、ResNet152V2的搭建方式完全一样，区别在于Residual Block的数量不同。
1、Residual Block
import tensorflow as tf
import tensorflow.keras.layers as layers
from tensorflow.keras.models import Model
"""
残差块
Arguments:
x:输入张量
filters:integer,filters of the bottleneck layer.
kernel_size:默认是3，kernel size of the bottleneck layer.
stride:default 1,stride of the first layer.
conv_shortcut:default False,use convolution shortcut if True,
otherwise identity shortcut.
name:string,block label.
Returns:
Output tensor for the residual block.
"""
def block2(x,filters,kernel_size=3,stride=1,conv_shortcut=False,name=None):
preact=layers.BatchNormalization(name=name+'_preact_bn')(x)
preact=layers.Activation('relu',name=name+'_preact_relu')(preact)
if conv_shortcut:
shortcut=layers.Conv2D(4*filters,1,strides=stride,name=name+'_0_conv')(preact)
else:
shortcut=layers.MaxPooling2D(1,strides=stride)(x) if stride>1 else x
x=layers.Conv2D(filters,1,strides=1,use_bias=False,name=name+'_1_conv')(preact)
x=layers.BatchNormalization(name=name+'_1_bn')(x)
x=layers.Activation('relu',name=name+'_1_relu')(x)
x=layers.ZeroPadding2D(padding=((1,1),(1,1)),name=name+'_2_pad')(x)
x=layers.Conv2D(filters,
kernel_size,
strides=stride,
use_bias=False,
name=name+'_2_conv')(x)
x=layers.BatchNormalization(name=name+'_2_bn')(x)
x=layers.Activation('relu',name=name+'_2_relu')(x)
x=layers.Conv2D(4*filters,1,name=name+'_3_conv')(x)
x=layers.Add(name=name+'_out')([shortcut,x])
return x我大概对比了下ResNet50的源代码，发现关于identity shortcut这里稍微有点区别。ResNet50中是这样的：
if conv_shortcut:
shortcut=layers.Conv2D(4*filters,1,strides=stride,name=name+'_0_conv')(x)
shortcut=layers.Conv2D.BatchNormalization(
axis=bn_axis,epsilon=1.001e-5,name=name+'_0_bn')(shortcut)
else:
shortcut=x而ResNet50V2中是这样的：
if conv_shortcut:
shortcut=layers.Conv2D(4*filters,1,strides=stride,name=name+'_0_conv')(preact)
else:
#注意后面还多了if语句
shortcut=layers.MaxPooling2D(1,strides=stride)(x) if stride>1 else x除此之外，在ResNet50源代码中是没有预激活的（即上面代码中的preact）。区别很容易发现，但我还是感觉原版的ResNet50比ResNet50V2的代码看起来更优雅。
2、堆叠Residual Block
def stack2(x,filters,blocks,stride1=2,name=None):
x=block2(x,filters,conv_shortcut=True,name=name+'_block1')
for i in range(2,blocks):
x=block2(x,filters,name=name+'_block'+str(i))
x=block2(x,filters,stride=stride1,name=name+'_block'+str(blocks))
return x3、ResNet50V2架构复现
更详细的网络结构图可以直接拉到文章底部查看
def ResNet50V2(
include_top=True, #是否包含位于网络顶部的全连接层
preact=True, #是否使用预激活
use_bias=True, #是否对卷积层使用偏置
weights='imagenet',
input_tensor=None, #可选的keras张量，用作模型的图像输入
input_shape=None,
pooling=None,
classes=4, #用于分类图像的可选类数
classifier_activation='softmax' #分类层激活函数
):
img_input=layers.Input(shape=input_shape)
x=layers.ZeroPadding2D(padding=((3,3),(3,3)),name='conv1_pad')(img_input)
x=layers.Conv2D(64,7,strides=2,use_bias=use_bias,name='conv1_conv')(x)
if not preact:
x=layers.BatchNormalization(name='conv1_bn')(x)
x=layers.Activation('relu',name='conv1_relu')(x)
x=layers.ZeroPadding2D(padding=((1,1),(1,1)),name='pool1_pad')(x)
x=layers.MaxPooling2D(3,strides=2,name='pool1_pool')(x)
x=stack2(x,64,3,name='conv2')
x=stack2(x,128,4,name='conv3')
x=stack2(x,256,6,name='conv4')
x=stack2(x,512,3,stride1=1,name='conv5')
if preact:
x=layers.BatchNormalization(name='post_bn')(x)
x=layers.Activation('relu',name='post_relu')(x)
if include_top:
x=layers.GlobalAveragePooling2D(name='avg_pool')(x)
#x=layers.Flatten()(x)
x=layers.Dense(classes,activation=classifier_activation,name='predictions')(x)
else:
if pooling=='avg':
#GlobalAveragePooling2D就是将每张图片的每个通道值各自加起来再求平均
#最后结果是没有了宽高维度，只剩下个数与平均值两个维度
#可以理解为变成了多张单像素图片
x=layers.GlobalAveragePooling2D(name='avg_pool')(x)
elif pooling=='max':
x=layers.GlobalAveragePooling2D(name='max_pool')(x)
model=Model(img_input,x)
return model查看网络结构：
if __name__=='__main__':
model=ResNet50V2(input_shape=(224,224,3))
model.summary()运行结果：
Model: "model"
__________________________________________________________________________________________________
Layer (type) Output Shape Param # Connected to
==================================================================================================
input_7 (InputLayer) [(None, 224, 224, 3)] 0 []
conv1_pad (ZeroPadding2D) (None, 230, 230, 3) 0 ['input_7[0][0]']
conv1_conv (Conv2D) (None, 112, 112, 64) 9472 ['conv1_pad[0][0]']
pool1_pad (ZeroPadding2D) (None, 114, 114, 64) 0 ['conv1_conv[0][0]']
pool1_pool (MaxPooling2D) (None, 56, 56, 64) 0 ['pool1_pad[0][0]']
conv2_block1_preact_bn (Ba (None, 56, 56, 64) 256 ['pool1_pool[0][0]']
tchNormalization)
conv2_block1_preact_relu ( (None, 56, 56, 64) 0 ['conv2_block1_preact_bn[0][0]
Activation) ']
conv2_block1_1_conv (Conv2 (None, 56, 56, 64) 4096 ['conv2_block1_preact_relu[0][
D) 0]']
conv2_block1_1_bn (BatchNo (None, 56, 56, 64) 256 ['conv2_block1_1_conv[0][0]']
rmalization)
conv2_block1_1_relu (Activ (None, 56, 56, 64) 0 ['conv2_block1_1_bn[0][0]']
ation)
conv2_block1_2_pad (ZeroPa (None, 58, 58, 64) 0 ['conv2_block1_1_relu[0][0]']
dding2D)
conv2_block1_2_conv (Conv2 (None, 56, 56, 64) 36864 ['conv2_block1_2_pad[0][0]']
D)
conv2_block1_2_bn (BatchNo (None, 56, 56, 64) 256 ['conv2_block1_2_conv[0][0]']
rmalization)
conv2_block1_2_relu (Activ (None, 56, 56, 64) 0 ['conv2_block1_2_bn[0][0]']
ation)
conv2_block1_0_conv (Conv2 (None, 56, 56, 256) 16640 ['conv2_block1_preact_relu[0][
D) 0]']
conv2_block1_3_conv (Conv2 (None, 56, 56, 256) 16640 ['conv2_block1_2_relu[0][0]']
D)
conv2_block1_out (Add) (None, 56, 56, 256) 0 ['conv2_block1_0_conv[0][0]',
'conv2_block1_3_conv[0][0]']
conv2_block2_preact_bn (Ba (None, 56, 56, 256) 1024 ['conv2_block1_out[0][0]']
tchNormalization)
conv2_block2_preact_relu ( (None, 56, 56, 256) 0 ['conv2_block2_preact_bn[0][0]
Activation) ']
conv2_block2_1_conv (Conv2 (None, 56, 56, 64) 16384 ['conv2_block2_preact_relu[0][
D) 0]']
conv2_block2_1_bn (BatchNo (None, 56, 56, 64) 256 ['conv2_block2_1_conv[0][0]']
rmalization)
conv2_block2_1_relu (Activ (None, 56, 56, 64) 0 ['conv2_block2_1_bn[0][0]']
ation)
conv2_block2_2_pad (ZeroPa (None, 58, 58, 64) 0 ['conv2_block2_1_relu[0][0]']
dding2D)
conv2_block2_2_conv (Conv2 (None, 56, 56, 64) 36864 ['conv2_block2_2_pad[0][0]']
D)
conv2_block2_2_bn (BatchNo (None, 56, 56, 64) 256 ['conv2_block2_2_conv[0][0]']
rmalization)
conv2_block2_2_relu (Activ (None, 56, 56, 64) 0 ['conv2_block2_2_bn[0][0]']
ation)
conv2_block2_3_conv (Conv2 (None, 56, 56, 256) 16640 ['conv2_block2_2_relu[0][0]']
D)
conv2_block2_out (Add) (None, 56, 56, 256) 0 ['conv2_block1_out[0][0]',
'conv2_block2_3_conv[0][0]']
conv2_block3_preact_bn (Ba (None, 56, 56, 256) 1024 ['conv2_block2_out[0][0]']
tchNormalization)
conv2_block3_preact_relu ( (None, 56, 56, 256) 0 ['conv2_block3_preact_bn[0][0]
Activation) ']
conv2_block3_1_conv (Conv2 (None, 56, 56, 64) 16384 ['conv2_block3_preact_relu[0][
D) 0]']
conv2_block3_1_bn (BatchNo (None, 56, 56, 64) 256 ['conv2_block3_1_conv[0][0]']
rmalization)
conv2_block3_1_relu (Activ (None, 56, 56, 64) 0 ['conv2_block3_1_bn[0][0]']
ation)
conv2_block3_2_pad (ZeroPa (None, 58, 58, 64) 0 ['conv2_block3_1_relu[0][0]']
dding2D)
conv2_block3_2_conv (Conv2 (None, 28, 28, 64) 36864 ['conv2_block3_2_pad[0][0]']
D)
conv2_block3_2_bn (BatchNo (None, 28, 28, 64) 256 ['conv2_block3_2_conv[0][0]']
rmalization)
conv2_block3_2_relu (Activ (None, 28, 28, 64) 0 ['conv2_block3_2_bn[0][0]']
ation)
max_pooling2d_18 (MaxPooli (None, 28, 28, 256) 0 ['conv2_block2_out[0][0]']
ng2D)
conv2_block3_3_conv (Conv2 (None, 28, 28, 256) 16640 ['conv2_block3_2_relu[0][0]']
D)
conv2_block3_out (Add) (None, 28, 28, 256) 0 ['max_pooling2d_18[0][0]',
'conv2_block3_3_conv[0][0]']
conv3_block1_preact_bn (Ba (None, 28, 28, 256) 1024 ['conv2_block3_out[0][0]']
tchNormalization)
conv3_block1_preact_relu ( (None, 28, 28, 256) 0 ['conv3_block1_preact_bn[0][0]
Activation) ']
conv3_block1_1_conv (Conv2 (None, 28, 28, 128) 32768 ['conv3_block1_preact_relu[0][
D) 0]']
conv3_block1_1_bn (BatchNo (None, 28, 28, 128) 512 ['conv3_block1_1_conv[0][0]']
rmalization)
conv3_block1_1_relu (Activ (None, 28, 28, 128) 0 ['conv3_block1_1_bn[0][0]']
ation)
conv3_block1_2_pad (ZeroPa (None, 30, 30, 128) 0 ['conv3_block1_1_relu[0][0]']
dding2D)
conv3_block1_2_conv (Conv2 (None, 28, 28, 128) 147456 ['conv3_block1_2_pad[0][0]']
D)
conv3_block1_2_bn (BatchNo (None, 28, 28, 128) 512 ['conv3_block1_2_conv[0][0]']
rmalization)
conv3_block1_2_relu (Activ (None, 28, 28, 128) 0 ['conv3_block1_2_bn[0][0]']
ation)
conv3_block1_0_conv (Conv2 (None, 28, 28, 512) 131584 ['conv3_block1_preact_relu[0][
D) 0]']
conv3_block1_3_conv (Conv2 (None, 28, 28, 512) 66048 ['conv3_block1_2_relu[0][0]']
D)
conv3_block1_out (Add) (None, 28, 28, 512) 0 ['conv3_block1_0_conv[0][0]',
'conv3_block1_3_conv[0][0]']
conv3_block2_preact_bn (Ba (None, 28, 28, 512) 2048 ['conv3_block1_out[0][0]']
tchNormalization)
conv3_block2_preact_relu ( (None, 28, 28, 512) 0 ['conv3_block2_preact_bn[0][0]
Activation) ']
conv3_block2_1_conv (Conv2 (None, 28, 28, 128) 65536 ['conv3_block2_preact_relu[0][
D) 0]']
conv3_block2_1_bn (BatchNo (None, 28, 28, 128) 512 ['conv3_block2_1_conv[0][0]']
rmalization)
conv3_block2_1_relu (Activ (None, 28, 28, 128) 0 ['conv3_block2_1_bn[0][0]']
ation)
conv3_block2_2_pad (ZeroPa (None, 30, 30, 128) 0 ['conv3_block2_1_relu[0][0]']
dding2D)
conv3_block2_2_conv (Conv2 (None, 28, 28, 128) 147456 ['conv3_block2_2_pad[0][0]']
D)
conv3_block2_2_bn (BatchNo (None, 28, 28, 128) 512 ['conv3_block2_2_conv[0][0]']
rmalization)
conv3_block2_2_relu (Activ (None, 28, 28, 128) 0 ['conv3_block2_2_bn[0][0]']
ation)
conv3_block2_3_conv (Conv2 (None, 28, 28, 512) 66048 ['conv3_block2_2_relu[0][0]']
D)
conv3_block2_out (Add) (None, 28, 28, 512) 0 ['conv3_block1_out[0][0]',
'conv3_block2_3_conv[0][0]']
conv3_block3_preact_bn (Ba (None, 28, 28, 512) 2048 ['conv3_block2_out[0][0]']
tchNormalization)
conv3_block3_preact_relu ( (None, 28, 28, 512) 0 ['conv3_block3_preact_bn[0][0]
Activation) ']
conv3_block3_1_conv (Conv2 (None, 28, 28, 128) 65536 ['conv3_block3_preact_relu[0][
D) 0]']
conv3_block3_1_bn (BatchNo (None, 28, 28, 128) 512 ['conv3_block3_1_conv[0][0]']
rmalization)
conv3_block3_1_relu (Activ (None, 28, 28, 128) 0 ['conv3_block3_1_bn[0][0]']
ation)
conv3_block3_2_pad (ZeroPa (None, 30, 30, 128) 0 ['conv3_block3_1_relu[0][0]']
dding2D)
conv3_block3_2_conv (Conv2 (None, 28, 28, 128) 147456 ['conv3_block3_2_pad[0][0]']
D)
conv3_block3_2_bn (BatchNo (None, 28, 28, 128) 512 ['conv3_block3_2_conv[0][0]']
rmalization)
conv3_block3_2_relu (Activ (None, 28, 28, 128) 0 ['conv3_block3_2_bn[0][0]']
ation)
conv3_block3_3_conv (Conv2 (None, 28, 28, 512) 66048 ['conv3_block3_2_relu[0][0]']
D)
conv3_block3_out (Add) (None, 28, 28, 512) 0 ['conv3_block2_out[0][0]',
'conv3_block3_3_conv[0][0]']
conv3_block4_preact_bn (Ba (None, 28, 28, 512) 2048 ['conv3_block3_out[0][0]']
tchNormalization)
conv3_block4_preact_relu ( (None, 28, 28, 512) 0 ['conv3_block4_preact_bn[0][0]
Activation) ']
conv3_block4_1_conv (Conv2 (None, 28, 28, 128) 65536 ['conv3_block4_preact_relu[0][
D) 0]']
conv3_block4_1_bn (BatchNo (None, 28, 28, 128) 512 ['conv3_block4_1_conv[0][0]']
rmalization)
conv3_block4_1_relu (Activ (None, 28, 28, 128) 0 ['conv3_block4_1_bn[0][0]']
ation)
conv3_block4_2_pad (ZeroPa (None, 30, 30, 128) 0 ['conv3_block4_1_relu[0][0]']
dding2D)
conv3_block4_2_conv (Conv2 (None, 14, 14, 128) 147456 ['conv3_block4_2_pad[0][0]']
D)
conv3_block4_2_bn (BatchNo (None, 14, 14, 128) 512 ['conv3_block4_2_conv[0][0]']
rmalization)
conv3_block4_2_relu (Activ (None, 14, 14, 128) 0 ['conv3_block4_2_bn[0][0]']
ation)
max_pooling2d_19 (MaxPooli (None, 14, 14, 512) 0 ['conv3_block3_out[0][0]']
ng2D)
conv3_block4_3_conv (Conv2 (None, 14, 14, 512) 66048 ['conv3_block4_2_relu[0][0]']
D)
conv3_block4_out (Add) (None, 14, 14, 512) 0 ['max_pooling2d_19[0][0]',
'conv3_block4_3_conv[0][0]']
conv4_block1_preact_bn (Ba (None, 14, 14, 512) 2048 ['conv3_block4_out[0][0]']
tchNormalization)
conv4_block1_preact_relu ( (None, 14, 14, 512) 0 ['conv4_block1_preact_bn[0][0]
Activation) ']
conv4_block1_1_conv (Conv2 (None, 14, 14, 256) 131072 ['conv4_block1_preact_relu[0][
D) 0]']
conv4_block1_1_bn (BatchNo (None, 14, 14, 256) 1024 ['conv4_block1_1_conv[0][0]']
rmalization)
conv4_block1_1_relu (Activ (None, 14, 14, 256) 0 ['conv4_block1_1_bn[0][0]']
ation)
conv4_block1_2_pad (ZeroPa (None, 16, 16, 256) 0 ['conv4_block1_1_relu[0][0]']
dding2D)
conv4_block1_2_conv (Conv2 (None, 14, 14, 256) 589824 ['conv4_block1_2_pad[0][0]']
D)
conv4_block1_2_bn (BatchNo (None, 14, 14, 256) 1024 ['conv4_block1_2_conv[0][0]']
rmalization)
conv4_block1_2_relu (Activ (None, 14, 14, 256) 0 ['conv4_block1_2_bn[0][0]']
ation)
conv4_block1_0_conv (Conv2 (None, 14, 14, 1024) 525312 ['conv4_block1_preact_relu[0][
D) 0]']
conv4_block1_3_conv (Conv2 (None, 14, 14, 1024) 263168 ['conv4_block1_2_relu[0][0]']
D)
conv4_block1_out (Add) (None, 14, 14, 1024) 0 ['conv4_block1_0_conv[0][0]',
'conv4_block1_3_conv[0][0]']
conv4_block2_preact_bn (Ba (None, 14, 14, 1024) 4096 ['conv4_block1_out[0][0]']
tchNormalization)
conv4_block2_preact_relu ( (None, 14, 14, 1024) 0 ['conv4_block2_preact_bn[0][0]
Activation) ']
conv4_block2_1_conv (Conv2 (None, 14, 14, 256) 262144 ['conv4_block2_preact_relu[0][
D) 0]']
conv4_block2_1_bn (BatchNo (None, 14, 14, 256) 1024 ['conv4_block2_1_conv[0][0]']
rmalization)
conv4_block2_1_relu (Activ (None, 14, 14, 256) 0 ['conv4_block2_1_bn[0][0]']
ation)
conv4_block2_2_pad (ZeroPa (None, 16, 16, 256) 0 ['conv4_block2_1_relu[0][0]']
dding2D)
conv4_block2_2_conv (Conv2 (None, 14, 14, 256) 589824 ['conv4_block2_2_pad[0][0]']
D)
conv4_block2_2_bn (BatchNo (None, 14, 14, 256) 1024 ['conv4_block2_2_conv[0][0]']
rmalization)
conv4_block2_2_relu (Activ (None, 14, 14, 256) 0 ['conv4_block2_2_bn[0][0]']
ation)
conv4_block2_3_conv (Conv2 (None, 14, 14, 1024) 263168 ['conv4_block2_2_relu[0][0]']
D)
conv4_block2_out (Add) (None, 14, 14, 1024) 0 ['conv4_block1_out[0][0]',
'conv4_block2_3_conv[0][0]']
conv4_block3_preact_bn (Ba (None, 14, 14, 1024) 4096 ['conv4_block2_out[0][0]']
tchNormalization)
conv4_block3_preact_relu ( (None, 14, 14, 1024) 0 ['conv4_block3_preact_bn[0][0]
Activation) ']
conv4_block3_1_conv (Conv2 (None, 14, 14, 256) 262144 ['conv4_block3_preact_relu[0][
D) 0]']
conv4_block3_1_bn (BatchNo (None, 14, 14, 256) 1024 ['conv4_block3_1_conv[0][0]']
rmalization)
conv4_block3_1_relu (Activ (None, 14, 14, 256) 0 ['conv4_block3_1_bn[0][0]']
ation)
conv4_block3_2_pad (ZeroPa (None, 16, 16, 256) 0 ['conv4_block3_1_relu[0][0]']
dding2D)
conv4_block3_2_conv (Conv2 (None, 14, 14, 256) 589824 ['conv4_block3_2_pad[0][0]']
D)
conv4_block3_2_bn (BatchNo (None, 14, 14, 256) 1024 ['conv4_block3_2_conv[0][0]']
rmalization)
conv4_block3_2_relu (Activ (None, 14, 14, 256) 0 ['conv4_block3_2_bn[0][0]']
ation)
conv4_block3_3_conv (Conv2 (None, 14, 14, 1024) 263168 ['conv4_block3_2_relu[0][0]']
D)
conv4_block3_out (Add) (None, 14, 14, 1024) 0 ['conv4_block2_out[0][0]',
'conv4_block3_3_conv[0][0]']
conv4_block4_preact_bn (Ba (None, 14, 14, 1024) 4096 ['conv4_block3_out[0][0]']
tchNormalization)
conv4_block4_preact_relu ( (None, 14, 14, 1024) 0 ['conv4_block4_preact_bn[0][0]
Activation) ']
conv4_block4_1_conv (Conv2 (None, 14, 14, 256) 262144 ['conv4_block4_preact_relu[0][
D) 0]']
conv4_block4_1_bn (BatchNo (None, 14, 14, 256) 1024 ['conv4_block4_1_conv[0][0]']
rmalization)
conv4_block4_1_relu (Activ (None, 14, 14, 256) 0 ['conv4_block4_1_bn[0][0]']
ation)
conv4_block4_2_pad (ZeroPa (None, 16, 16, 256) 0 ['conv4_block4_1_relu[0][0]']
dding2D)
conv4_block4_2_conv (Conv2 (None, 14, 14, 256) 589824 ['conv4_block4_2_pad[0][0]']
D)
conv4_block4_2_bn (BatchNo (None, 14, 14, 256) 1024 ['conv4_block4_2_conv[0][0]']
rmalization)
conv4_block4_2_relu (Activ (None, 14, 14, 256) 0 ['conv4_block4_2_bn[0][0]']
ation)
conv4_block4_3_conv (Conv2 (None, 14, 14, 1024) 263168 ['conv4_block4_2_relu[0][0]']
D)
conv4_block4_out (Add) (None, 14, 14, 1024) 0 ['conv4_block3_out[0][0]',
'conv4_block4_3_conv[0][0]']
conv4_block5_preact_bn (Ba (None, 14, 14, 1024) 4096 ['conv4_block4_out[0][0]']
tchNormalization)
conv4_block5_preact_relu ( (None, 14, 14, 1024) 0 ['conv4_block5_preact_bn[0][0]
Activation) ']
conv4_block5_1_conv (Conv2 (None, 14, 14, 256) 262144 ['conv4_block5_preact_relu[0][
D) 0]']
conv4_block5_1_bn (BatchNo (None, 14, 14, 256) 1024 ['conv4_block5_1_conv[0][0]']
rmalization)
conv4_block5_1_relu (Activ (None, 14, 14, 256) 0 ['conv4_block5_1_bn[0][0]']
ation)
conv4_block5_2_pad (ZeroPa (None, 16, 16, 256) 0 ['conv4_block5_1_relu[0][0]']
dding2D)
conv4_block5_2_conv (Conv2 (None, 14, 14, 256) 589824 ['conv4_block5_2_pad[0][0]']
D)
conv4_block5_2_bn (BatchNo (None, 14, 14, 256) 1024 ['conv4_block5_2_conv[0][0]']
rmalization)
conv4_block5_2_relu (Activ (None, 14, 14, 256) 0 ['conv4_block5_2_bn[0][0]']
ation)
conv4_block5_3_conv (Conv2 (None, 14, 14, 1024) 263168 ['conv4_block5_2_relu[0][0]']
D)
conv4_block5_out (Add) (None, 14, 14, 1024) 0 ['conv4_block4_out[0][0]',
'conv4_block5_3_conv[0][0]']
conv4_block6_preact_bn (Ba (None, 14, 14, 1024) 4096 ['conv4_block5_out[0][0]']
tchNormalization)
conv4_block6_preact_relu ( (None, 14, 14, 1024) 0 ['conv4_block6_preact_bn[0][0]
Activation) ']
conv4_block6_1_conv (Conv2 (None, 14, 14, 256) 262144 ['conv4_block6_preact_relu[0][
D) 0]']
conv4_block6_1_bn (BatchNo (None, 14, 14, 256) 1024 ['conv4_block6_1_conv[0][0]']
rmalization)
conv4_block6_1_relu (Activ (None, 14, 14, 256) 0 ['conv4_block6_1_bn[0][0]']
ation)
conv4_block6_2_pad (ZeroPa (None, 16, 16, 256) 0 ['conv4_block6_1_relu[0][0]']
dding2D)
conv4_block6_2_conv (Conv2 (None, 7, 7, 256) 589824 ['conv4_block6_2_pad[0][0]']
D)
conv4_block6_2_bn (BatchNo (None, 7, 7, 256) 1024 ['conv4_block6_2_conv[0][0]']
rmalization)
conv4_block6_2_relu (Activ (None, 7, 7, 256) 0 ['conv4_block6_2_bn[0][0]']
ation)
max_pooling2d_20 (MaxPooli (None, 7, 7, 1024) 0 ['conv4_block5_out[0][0]']
ng2D)
conv4_block6_3_conv (Conv2 (None, 7, 7, 1024) 263168 ['conv4_block6_2_relu[0][0]']
D)
conv4_block6_out (Add) (None, 7, 7, 1024) 0 ['max_pooling2d_20[0][0]',
'conv4_block6_3_conv[0][0]']
conv5_block1_preact_bn (Ba (None, 7, 7, 1024) 4096 ['conv4_block6_out[0][0]']
tchNormalization)
conv5_block1_preact_relu ( (None, 7, 7, 1024) 0 ['conv5_block1_preact_bn[0][0]
Activation) ']
conv5_block1_1_conv (Conv2 (None, 7, 7, 512) 524288 ['conv5_block1_preact_relu[0][
D) 0]']
conv5_block1_1_bn (BatchNo (None, 7, 7, 512) 2048 ['conv5_block1_1_conv[0][0]']
rmalization)
conv5_block1_1_relu (Activ (None, 7, 7, 512) 0 ['conv5_block1_1_bn[0][0]']
ation)
conv5_block1_2_pad (ZeroPa (None, 9, 9, 512) 0 ['conv5_block1_1_relu[0][0]']
dding2D)
conv5_block1_2_conv (Conv2 (None, 7, 7, 512) 2359296 ['conv5_block1_2_pad[0][0]']
D)
conv5_block1_2_bn (BatchNo (None, 7, 7, 512) 2048 ['conv5_block1_2_conv[0][0]']
rmalization)
conv5_block1_2_relu (Activ (None, 7, 7, 512) 0 ['conv5_block1_2_bn[0][0]']
ation)
conv5_block1_0_conv (Conv2 (None, 7, 7, 2048) 2099200 ['conv5_block1_preact_relu[0][
D) 0]']
conv5_block1_3_conv (Conv2 (None, 7, 7, 2048) 1050624 ['conv5_block1_2_relu[0][0]']
D)
conv5_block1_out (Add) (None, 7, 7, 2048) 0 ['conv5_block1_0_conv[0][0]',
'conv5_block1_3_conv[0][0]']
conv5_block2_preact_bn (Ba (None, 7, 7, 2048) 8192 ['conv5_block1_out[0][0]']
tchNormalization)
conv5_block2_preact_relu ( (None, 7, 7, 2048) 0 ['conv5_block2_preact_bn[0][0]
Activation) ']
conv5_block2_1_conv (Conv2 (None, 7, 7, 512) 1048576 ['conv5_block2_preact_relu[0][
D) 0]']
conv5_block2_1_bn (BatchNo (None, 7, 7, 512) 2048 ['conv5_block2_1_conv[0][0]']
rmalization)
conv5_block2_1_relu (Activ (None, 7, 7, 512) 0 ['conv5_block2_1_bn[0][0]']
ation)
conv5_block2_2_pad (ZeroPa (None, 9, 9, 512) 0 ['conv5_block2_1_relu[0][0]']
dding2D)
conv5_block2_2_conv (Conv2 (None, 7, 7, 512) 2359296 ['conv5_block2_2_pad[0][0]']
D)
conv5_block2_2_bn (BatchNo (None, 7, 7, 512) 2048 ['conv5_block2_2_conv[0][0]']
rmalization)
conv5_block2_2_relu (Activ (None, 7, 7, 512) 0 ['conv5_block2_2_bn[0][0]']
ation)
conv5_block2_3_conv (Conv2 (None, 7, 7, 2048) 1050624 ['conv5_block2_2_relu[0][0]']
D)
conv5_block2_out (Add) (None, 7, 7, 2048) 0 ['conv5_block1_out[0][0]',
'conv5_block2_3_conv[0][0]']
conv5_block3_preact_bn (Ba (None, 7, 7, 2048) 8192 ['conv5_block2_out[0][0]']
tchNormalization)
conv5_block3_preact_relu ( (None, 7, 7, 2048) 0 ['conv5_block3_preact_bn[0][0]
Activation) ']
conv5_block3_1_conv (Conv2 (None, 7, 7, 512) 1048576 ['conv5_block3_preact_relu[0][
D) 0]']
conv5_block3_1_bn (BatchNo (None, 7, 7, 512) 2048 ['conv5_block3_1_conv[0][0]']
rmalization)
conv5_block3_1_relu (Activ (None, 7, 7, 512) 0 ['conv5_block3_1_bn[0][0]']
ation)
conv5_block3_2_pad (ZeroPa (None, 9, 9, 512) 0 ['conv5_block3_1_relu[0][0]']
dding2D)
conv5_block3_2_conv (Conv2 (None, 7, 7, 512) 2359296 ['conv5_block3_2_pad[0][0]']
D)
conv5_block3_2_bn (BatchNo (None, 7, 7, 512) 2048 ['conv5_block3_2_conv[0][0]']
rmalization)
conv5_block3_2_relu (Activ (None, 7, 7, 512) 0 ['conv5_block3_2_bn[0][0]']
ation)
conv5_block3_3_conv (Conv2 (None, 7, 7, 2048) 1050624 ['conv5_block3_2_relu[0][0]']
D)
conv5_block3_out (Add) (None, 7, 7, 2048) 0 ['conv5_block2_out[0][0]',
'conv5_block3_3_conv[0][0]']
post_bn (BatchNormalizatio (None, 7, 7, 2048) 8192 ['conv5_block3_out[0][0]']
n)
post_relu (Activation) (None, 7, 7, 2048) 0 ['post_bn[0][0]']
avg_pool (GlobalAveragePoo (None, 2048) 0 ['post_relu[0][0]']
ling2D)
predictions (Dense) (None, 4) 8196 ['avg_pool[0][0]']
==================================================================================================
Total params: 23572996 (89.92 MB)
Trainable params: 23527556 (89.75 MB)
Non-trainable params: 45440 (177.50 KB)
__________________________________________________________________________________________________对应各层结构图如下：
其中，最后的Dense(softmax)输出结果应调整为：4
五、编译
在准备对模型进行训练之前，还需要再对其进行一些设置。以下内容是在模型的编译步骤中添加的：
损失函数（loss）：用于衡量模型在训练期间的准确率。
优化器（optimizer）：决定模型如何根据其看到的数据和自身的损失函数进行更新。
指标（metrics）：用于监控训练和测试步骤。以下示例使用了准确率，即被正确分类的图像的比率。
#设置优化器
opt=tf.keras.optimizers.Adam(learning_rate=1e-3)
model.compile(optimizer=opt,
loss=tf.keras.losses.SparseCategoricalCrossentropy(),
metrics=['accuracy'])六、训练模型
epochs=10
history=model.fit(
train_ds,
validation_data=val_ds,
epochs=epochs
)运行结果：
Epoch 1/10
57/57 [==============================] - 142s 2s/step - loss: 1.2355 - accuracy: 0.5774 - val_loss: 984.8382 - val_accuracy: 0.2301
Epoch 2/10
57/57 [==============================] - 127s 2s/step - loss: 0.8714 - accuracy: 0.6792 - val_loss: 21.0315 - val_accuracy: 0.3451
Epoch 3/10
57/57 [==============================] - 122s 2s/step - loss: 0.7507 - accuracy: 0.7257 - val_loss: 4.6712 - val_accuracy: 0.3982
Epoch 4/10
57/57 [==============================] - 112s 2s/step - loss: 0.6349 - accuracy: 0.7611 - val_loss: 5.8396 - val_accuracy: 0.4513
Epoch 5/10
57/57 [==============================] - 109s 2s/step - loss: 0.6026 - accuracy: 0.7611 - val_loss: 16.5968 - val_accuracy: 0.2743
Epoch 6/10
57/57 [==============================] - 108s 2s/step - loss: 0.5709 - accuracy: 0.8230 - val_loss: 17.4454 - val_accuracy: 0.4336
Epoch 7/10
57/57 [==============================] - 108s 2s/step - loss: 0.5856 - accuracy: 0.7942 - val_loss: 24.0665 - val_accuracy: 0.2301
Epoch 8/10
57/57 [==============================] - 111s 2s/step - loss: 0.3669 - accuracy: 0.8739 - val_loss: 1.6654 - val_accuracy: 0.7080
Epoch 9/10
57/57 [==============================] - 113s 2s/step - loss: 0.3580 - accuracy: 0.8850 - val_loss: 1.1793 - val_accuracy: 0.7257
Epoch 10/10
57/57 [==============================] - 109s 2s/step - loss: 0.2346 - accuracy: 0.9159 - val_loss: 4.7485 - val_accuracy: 0.4248八、模型评估
acc=history.history['accuracy']
val_acc=history.history['val_accuracy']
loss=history.history['loss']
val_loss=history.history['val_loss']
epochs_range=range(epochs)
plt.figure(figsize=(12,4))
plt.suptitle("OreoCC")
plt.subplot(1,2,1)
plt.plot(epochs_range,acc,label='Training Accuracy')
plt.plot(epochs_range,val_acc,label='Validation Accuracy')
plt.legend(loc='lower right')
plt.title('Training and Validation Accuracy')
plt.subplot(1,2,2)
plt.plot(epochs_range,loss,label='Training Loss')
plt.plot(epochs_range,val_loss,label='Validation Loss')
plt.legend(loc='upper right')
plt.title('Training and Validation Loss')
plt.show()运行结果：
九、心得体会
通过搭建ResNet50V2模型，深入了解了其与ResNet50的区别仅在与BN和ReLU的位置不同，我们也通过论文学习到激活前置后，模型的准确率出现了一定程度的提升，依据此原理可以得出，其他模型我们也可以此中前置激活模式修改模型，提升模型的准确率。
十、ResNet50V2模型结构放大图
820
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
