keras 截取 resnet50 现有网络层结构（含权重）-截取部分【原创】

最新推荐文章于 2022-11-04 19:50:01 发布
最新推荐文章于 2022-11-04 19:50:01 发布
阅读量2.3k 收藏 4
点赞数 1
分类专栏： python 深度学习 文章标签： Keras 网络拼接 深度学习
版权声明：本文为博主原创文章，遵循 CC 4.0 BY-SA 版权协议，转载请附上原文出处链接和本声明。
本文链接：https://blog.csdn.net/Monster_H7/article/details/103241977
版权

其中截取主要是使用get_layer()方法来设置截取网络的起点与终点！

1.get_layer()用法

例如：*(from:网上一个例子）https://blog.csdn.net/zhzhx1204/article/details/77393245

from keras.models import Model

model = ...  # create the original model

layer_name = 'my_layer'
intermediate_layer_model = Model(input=model.input,
                                 output=model.get_layer(layer_name).output)
intermediate_output = intermediate_layer_model.predict(data..

后来又在keras 官方中文文档中找到该函数的用法，写的很详细：

https://keras.io/zh/models/model/#get_layer

get_layer

get_layer(self, name=None, index=None)

根据名称（唯一）或索引值查找网络层。

如果同时提供了 name 和 index，则 index 将优先。

索引值来自于水平图遍历的顺序（自下而上）。

参数

  • name: 字符串，层的名字。
  • index: 整数，层的索引。

返回

一个层实例。

异常

  • ValueError: 如果层的名称或索引不正确。

2.截取代码如下

参考：https://blog.csdn.net/u013249853/article/details/89211832

from __future__ import print_function
import numpy as np
import matplotlib.pyplot as plt
import os
import keras
from keras import Model
from keras.applications import vgg16, inception_v3, resnet50, mobilenet
from keras.preprocessing.image import ImageDataGenerator, load_img
os.environ['CUDA_VISIBLE_DEVICES']='0'
resnet_model = resnet50.ResNet50(weights = 'imagenet', include_top=False,input_shape = (480,640,3))
resnet_model.summary()
layer_name = 'activation_10'
res10 = Model(inputs=resnet_model.input, outputs=resnet_model.get_layer(layer_name).output)
res10.summary()

从开始输入层一直截取到layer_name = 'activation_10'层 

 

 打印出来的输出是这样的：

E:\Anaconda3\envs\CNN_New\lib\site-packages\keras_applications\resnet50.py:265: UserWarning: The output shape of `ResNet50(include_top=False)` has been changed since Keras 2.2.0.
  warnings.warn('The output shape of `ResNet50(include_top=False)` '
__________________________________________________________________________________________________
Layer (type)                    Output Shape         Param #     Connected to                     
==================================================================================================
input_1 (InputLayer)            (None, 480, 640, 3)  0                                            
__________________________________________________________________________________________________
conv1_pad (ZeroPadding2D)       (None, 486, 646, 3)  0           input_1[0][0]                    
__________________________________________________________________________________________________
conv1 (Conv2D)                  (None, 240, 320, 64) 9472        conv1_pad[0][0]                  
__________________________________________________________________________________________________
bn_conv1 (BatchNormalization)   (None, 240, 320, 64) 256         conv1[0][0]                      
__________________________________________________________________________________________________
activation_1 (Activation)       (None, 240, 320, 64) 0           bn_conv1[0][0]                   
__________________________________________________________________________________________________
pool1_pad (ZeroPadding2D)       (None, 242, 322, 64) 0           activation_1[0][0]               
__________________________________________________________________________________________________
max_pooling2d_1 (MaxPooling2D)  (None, 120, 160, 64) 0           pool1_pad[0][0]                  
__________________________________________________________________________________________________
res2a_branch2a (Conv2D)         (None, 120, 160, 64) 4160        max_pooling2d_1[0][0]            
__________________________________________________________________________________________________
bn2a_branch2a (BatchNormalizati (None, 120, 160, 64) 256         res2a_branch2a[0][0]             
__________________________________________________________________________________________________
activation_2 (Activation)       (None, 120, 160, 64) 0           bn2a_branch2a[0][0]              
__________________________________________________________________________________________________
res2a_branch2b (Conv2D)         (None, 120, 160, 64) 36928       activation_2[0][0]               
__________________________________________________________________________________________________
bn2a_branch2b (BatchNormalizati (None, 120, 160, 64) 256         res2a_branch2b[0][0]             
__________________________________________________________________________________________________
activation_3 (Activation)       (None, 120, 160, 64) 0           bn2a_branch2b[0][0]              
__________________________________________________________________________________________________
res2a_branch2c (Conv2D)         (None, 120, 160, 256 16640       activation_3[0][0]               
__________________________________________________________________________________________________
res2a_branch1 (Conv2D)          (None, 120, 160, 256 16640       max_pooling2d_1[0][0]            
__________________________________________________________________________________________________
bn2a_branch2c (BatchNormalizati (None, 120, 160, 256 1024        res2a_branch2c[0][0]             
__________________________________________________________________________________________________
bn2a_branch1 (BatchNormalizatio (None, 120, 160, 256 1024        res2a_branch1[0][0]              
__________________________________________________________________________________________________
add_1 (Add)                     (None, 120, 160, 256 0           bn2a_branch2c[0][0]              
                                                                 bn2a_branch1[0][0]               
__________________________________________________________________________________________________
activation_4 (Activation)       (None, 120, 160, 256 0           add_1[0][0]                      
__________________________________________________________________________________________________
res2b_branch2a (Conv2D)         (None, 120, 160, 64) 16448       activation_4[0][0]               
__________________________________________________________________________________________________
bn2b_branch2a (BatchNormalizati (None, 120, 160, 64) 256         res2b_branch2a[0][0]             
__________________________________________________________________________________________________
activation_5 (Activation)       (None, 120, 160, 64) 0           bn2b_branch2a[0][0]              
__________________________________________________________________________________________________
res2b_branch2b (Conv2D)         (None, 120, 160, 64) 36928       activation_5[0][0]               
__________________________________________________________________________________________________
bn2b_branch2b (BatchNormalizati (None, 120, 160, 64) 256         res2b_branch2b[0][0]             
__________________________________________________________________________________________________
activation_6 (Activation)       (None, 120, 160, 64) 0           bn2b_branch2b[0][0]              
__________________________________________________________________________________________________
res2b_branch2c (Conv2D)         (None, 120, 160, 256 16640       activation_6[0][0]               
__________________________________________________________________________________________________
bn2b_branch2c (BatchNormalizati (None, 120, 160, 256 1024        res2b_branch2c[0][0]             
__________________________________________________________________________________________________
add_2 (Add)                     (None, 120, 160, 256 0           bn2b_branch2c[0][0]              
                                                                 activation_4[0][0]               
__________________________________________________________________________________________________
activation_7 (Activation)       (None, 120, 160, 256 0           add_2[0][0]                      
__________________________________________________________________________________________________
res2c_branch2a (Conv2D)         (None, 120, 160, 64) 16448       activation_7[0][0]               
__________________________________________________________________________________________________
bn2c_branch2a (BatchNormalizati (None, 120, 160, 64) 256         res2c_branch2a[0][0]             
__________________________________________________________________________________________________
activation_8 (Activation)       (None, 120, 160, 64) 0           bn2c_branch2a[0][0]              
__________________________________________________________________________________________________
res2c_branch2b (Conv2D)         (None, 120, 160, 64) 36928       activation_8[0][0]               
__________________________________________________________________________________________________
bn2c_branch2b (BatchNormalizati (None, 120, 160, 64) 256         res2c_branch2b[0][0]             
__________________________________________________________________________________________________
activation_9 (Activation)       (None, 120, 160, 64) 0           bn2c_branch2b[0][0]              
__________________________________________________________________________________________________
res2c_branch2c (Conv2D)         (None, 120, 160, 256 16640       activation_9[0][0]               
__________________________________________________________________________________________________
bn2c_branch2c (BatchNormalizati (None, 120, 160, 256 1024        res2c_branch2c[0][0]             
__________________________________________________________________________________________________
add_3 (Add)                     (None, 120, 160, 256 0           bn2c_branch2c[0][0]              
                                                                 activation_7[0][0]               
__________________________________________________________________________________________________
activation_10 (Activation)      (None, 120, 160, 256 0           add_3[0][0]                      
__________________________________________________________________________________________________
res3a_branch2a (Conv2D)         (None, 60, 80, 128)  32896       activation_10[0][0]              
__________________________________________________________________________________________________
bn3a_branch2a (BatchNormalizati (None, 60, 80, 128)  512         res3a_branch2a[0][0]             
__________________________________________________________________________________________________
activation_11 (Activation)      (None, 60, 80, 128)  0           bn3a_branch2a[0][0]              
__________________________________________________________________________________________________
res3a_branch2b (Conv2D)         (None, 60, 80, 128)  147584      activation_11[0][0]              
__________________________________________________________________________________________________
bn3a_branch2b (BatchNormalizati (None, 60, 80, 128)  512         res3a_branch2b[0][0]             
__________________________________________________________________________________________________
activation_12 (Activation)      (None, 60, 80, 128)  0           bn3a_branch2b[0][0]              
__________________________________________________________________________________________________
res3a_branch2c (Conv2D)         (None, 60, 80, 512)  66048       activation_12[0][0]              
__________________________________________________________________________________________________
res3a_branch1 (Conv2D)          (None, 60, 80, 512)  131584      activation_10[0][0]              
__________________________________________________________________________________________________
bn3a_branch2c (BatchNormalizati (None, 60, 80, 512)  2048        res3a_branch2c[0][0]             
__________________________________________________________________________________________________
bn3a_branch1 (BatchNormalizatio (None, 60, 80, 512)  2048        res3a_branch1[0][0]              
__________________________________________________________________________________________________
add_4 (Add)                     (None, 60, 80, 512)  0           bn3a_branch2c[0][0]              
                                                                 bn3a_branch1[0][0]               
__________________________________________________________________________________________________
activation_13 (Activation)      (None, 60, 80, 512)  0           add_4[0][0]                      
__________________________________________________________________________________________________
res3b_branch2a (Conv2D)         (None, 60, 80, 128)  65664       activation_13[0][0]              
__________________________________________________________________________________________________
bn3b_branch2a (BatchNormalizati (None, 60, 80, 128)  512         res3b_branch2a[0][0]             
__________________________________________________________________________________________________
activation_14 (Activation)      (None, 60, 80, 128)  0           bn3b_branch2a[0][0]              
__________________________________________________________________________________________________
res3b_branch2b (Conv2D)         (None, 60, 80, 128)  147584      activation_14[0][0]              
__________________________________________________________________________________________________
bn3b_branch2b (BatchNormalizati (None, 60, 80, 128)  512         res3b_branch2b[0][0]             
__________________________________________________________________________________________________
activation_15 (Activation)      (None, 60, 80, 128)  0           bn3b_branch2b[0][0]              
__________________________________________________________________________________________________
res3b_branch2c (Conv2D)         (None, 60, 80, 512)  66048       activation_15[0][0]              
__________________________________________________________________________________________________
bn3b_branch2c (BatchNormalizati (None, 60, 80, 512)  2048        res3b_branch2c[0][0]             
__________________________________________________________________________________________________
add_5 (Add)                     (None, 60, 80, 512)  0           bn3b_branch2c[0][0]              
                                                                 activation_13[0][0]              
__________________________________________________________________________________________________
activation_16 (Activation)      (None, 60, 80, 512)  0           add_5[0][0]                      
__________________________________________________________________________________________________
res3c_branch2a (Conv2D)         (None, 60, 80, 128)  65664       activation_16[0][0]              
__________________________________________________________________________________________________
bn3c_branch2a (BatchNormalizati (None, 60, 80, 128)  512         res3c_branch2a[0][0]             
__________________________________________________________________________________________________
activation_17 (Activation)      (None, 60, 80, 128)  0           bn3c_branch2a[0][0]              
__________________________________________________________________________________________________
res3c_branch2b (Conv2D)         (None, 60, 80, 128)  147584      activation_17[0][0]              
__________________________________________________________________________________________________
bn3c_branch2b (BatchNormalizati (None, 60, 80, 128)  512         res3c_branch2b[0][0]             
__________________________________________________________________________________________________
activation_18 (Activation)      (None, 60, 80, 128)  0           bn3c_branch2b[0][0]              
__________________________________________________________________________________________________
res3c_branch2c (Conv2D)         (None, 60, 80, 512)  66048       activation_18[0][0]              
__________________________________________________________________________________________________
bn3c_branch2c (BatchNormalizati (None, 60, 80, 512)  2048        res3c_branch2c[0][0]             
__________________________________________________________________________________________________
add_6 (Add)                     (None, 60, 80, 512)  0           bn3c_branch2c[0][0]              
                                                                 activation_16[0][0]              
__________________________________________________________________________________________________
activation_19 (Activation)      (None, 60, 80, 512)  0           add_6[0][0]                      
__________________________________________________________________________________________________
res3d_branch2a (Conv2D)         (None, 60, 80, 128)  65664       activation_19[0][0]              
__________________________________________________________________________________________________
bn3d_branch2a (BatchNormalizati (None, 60, 80, 128)  512         res3d_branch2a[0][0]             
__________________________________________________________________________________________________
activation_20 (Activation)      (None, 60, 80, 128)  0           bn3d_branch2a[0][0]              
__________________________________________________________________________________________________
res3d_branch2b (Conv2D)         (None, 60, 80, 128)  147584      activation_20[0][0]              
__________________________________________________________________________________________________
bn3d_branch2b (BatchNormalizati (None, 60, 80, 128)  512         res3d_branch2b[0][0]             
__________________________________________________________________________________________________
activation_21 (Activation)      (None, 60, 80, 128)  0           bn3d_branch2b[0][0]              
__________________________________________________________________________________________________
res3d_branch2c (Conv2D)         (None, 60, 80, 512)  66048       activation_21[0][0]              
__________________________________________________________________________________________________
bn3d_branch2c (BatchNormalizati (None, 60, 80, 512)  2048        res3d_branch2c[0][0]             
__________________________________________________________________________________________________
add_7 (Add)                     (None, 60, 80, 512)  0           bn3d_branch2c[0][0]              
                                                                 activation_19[0][0]              
__________________________________________________________________________________________________
activation_22 (Activation)      (None, 60, 80, 512)  0           add_7[0][0]                      
__________________________________________________________________________________________________
res4a_branch2a (Conv2D)         (None, 30, 40, 256)  131328      activation_22[0][0]              
__________________________________________________________________________________________________
bn4a_branch2a (BatchNormalizati (None, 30, 40, 256)  1024        res4a_branch2a[0][0]             
__________________________________________________________________________________________________
activation_23 (Activation)      (None, 30, 40, 256)  0           bn4a_branch2a[0][0]              
__________________________________________________________________________________________________
res4a_branch2b (Conv2D)         (None, 30, 40, 256)  590080      activation_23[0][0]              
__________________________________________________________________________________________________
bn4a_branch2b (BatchNormalizati (None, 30, 40, 256)  1024        res4a_branch2b[0][0]             
__________________________________________________________________________________________________
activation_24 (Activation)      (None, 30, 40, 256)  0           bn4a_branch2b[0][0]              
__________________________________________________________________________________________________
res4a_branch2c (Conv2D)         (None, 30, 40, 1024) 263168      activation_24[0][0]              
__________________________________________________________________________________________________
res4a_branch1 (Conv2D)          (None, 30, 40, 1024) 525312      activation_22[0][0]              
__________________________________________________________________________________________________
bn4a_branch2c (BatchNormalizati (None, 30, 40, 1024) 4096        res4a_branch2c[0][0]             
__________________________________________________________________________________________________
bn4a_branch1 (BatchNormalizatio (None, 30, 40, 1024) 4096        res4a_branch1[0][0]              
__________________________________________________________________________________________________
add_8 (Add)                     (None, 30, 40, 1024) 0           bn4a_branch2c[0][0]              
                                                                 bn4a_branch1[0][0]               
__________________________________________________________________________________________________
activation_25 (Activation)      (None, 30, 40, 1024) 0           add_8[0][0]                      
__________________________________________________________________________________________________
res4b_branch2a (Conv2D)         (None, 30, 40, 256)  262400      activation_25[0][0]              
__________________________________________________________________________________________________
bn4b_branch2a (BatchNormalizati (None, 30, 40, 256)  1024        res4b_branch2a[0][0]             
__________________________________________________________________________________________________
activation_26 (Activation)      (None, 30, 40, 256)  0           bn4b_branch2a[0][0]              
__________________________________________________________________________________________________
res4b_branch2b (Conv2D)         (None, 30, 40, 256)  590080      activation_26[0][0]              
__________________________________________________________________________________________________
bn4b_branch2b (BatchNormalizati (None, 30, 40, 256)  1024        res4b_branch2b[0][0]             
__________________________________________________________________________________________________
activation_27 (Activation)      (None, 30, 40, 256)  0           bn4b_branch2b[0][0]              
__________________________________________________________________________________________________
res4b_branch2c (Conv2D)         (None, 30, 40, 1024) 263168      activation_27[0][0]              
__________________________________________________________________________________________________
bn4b_branch2c (BatchNormalizati (None, 30, 40, 1024) 4096        res4b_branch2c[0][0]             
__________________________________________________________________________________________________
add_9 (Add)                     (None, 30, 40, 1024) 0           bn4b_branch2c[0][0]              
                                                                 activation_25[0][0]              
__________________________________________________________________________________________________
activation_28 (Activation)      (None, 30, 40, 1024) 0           add_9[0][0]                      
__________________________________________________________________________________________________
res4c_branch2a (Conv2D)         (None, 30, 40, 256)  262400      activation_28[0][0]              
__________________________________________________________________________________________________
bn4c_branch2a (BatchNormalizati (None, 30, 40, 256)  1024        res4c_branch2a[0][0]             
__________________________________________________________________________________________________
activation_29 (Activation)      (None, 30, 40, 256)  0           bn4c_branch2a[0][0]              
__________________________________________________________________________________________________
res4c_branch2b (Conv2D)         (None, 30, 40, 256)  590080      activation_29[0][0]              
__________________________________________________________________________________________________
bn4c_branch2b (BatchNormalizati (None, 30, 40, 256)  1024        res4c_branch2b[0][0]             
__________________________________________________________________________________________________
activation_30 (Activation)      (None, 30, 40, 256)  0           bn4c_branch2b[0][0]              
__________________________________________________________________________________________________
res4c_branch2c (Conv2D)         (None, 30, 40, 1024) 263168      activation_30[0][0]              
__________________________________________________________________________________________________
bn4c_branch2c (BatchNormalizati (None, 30, 40, 1024) 4096        res4c_branch2c[0][0]             
__________________________________________________________________________________________________
add_10 (Add)                    (None, 30, 40, 1024) 0           bn4c_branch2c[0][0]              
                                                                 activation_28[0][0]              
__________________________________________________________________________________________________
activation_31 (Activation)      (None, 30, 40, 1024) 0           add_10[0][0]                     
__________________________________________________________________________________________________
res4d_branch2a (Conv2D)         (None, 30, 40, 256)  262400      activation_31[0][0]              
__________________________________________________________________________________________________
bn4d_branch2a (BatchNormalizati (None, 30, 40, 256)  1024        res4d_branch2a[0][0]             
__________________________________________________________________________________________________
activation_32 (Activation)      (None, 30, 40, 256)  0           bn4d_branch2a[0][0]              
__________________________________________________________________________________________________
res4d_branch2b (Conv2D)         (None, 30, 40, 256)  590080      activation_32[0][0]              
__________________________________________________________________________________________________
bn4d_branch2b (BatchNormalizati (None, 30, 40, 256)  1024        res4d_branch2b[0][0]             
__________________________________________________________________________________________________
activation_33 (Activation)      (None, 30, 40, 256)  0           bn4d_branch2b[0][0]              
__________________________________________________________________________________________________
res4d_branch2c (Conv2D)         (None, 30, 40, 1024) 263168      activation_33[0][0]              
__________________________________________________________________________________________________
bn4d_branch2c (BatchNormalizati (None, 30, 40, 1024) 4096        res4d_branch2c[0][0]             
__________________________________________________________________________________________________
add_11 (Add)                    (None, 30, 40, 1024) 0           bn4d_branch2c[0][0]              
                                                                 activation_31[0][0]              
__________________________________________________________________________________________________
activation_34 (Activation)      (None, 30, 40, 1024) 0           add_11[0][0]                     
__________________________________________________________________________________________________
res4e_branch2a (Conv2D)         (None, 30, 40, 256)  262400      activation_34[0][0]              
__________________________________________________________________________________________________
bn4e_branch2a (BatchNormalizati (None, 30, 40, 256)  1024        res4e_branch2a[0][0]             
__________________________________________________________________________________________________
activation_35 (Activation)      (None, 30, 40, 256)  0           bn4e_branch2a[0][0]              
__________________________________________________________________________________________________
res4e_branch2b (Conv2D)         (None, 30, 40, 256)  590080      activation_35[0][0]              
__________________________________________________________________________________________________
bn4e_branch2b (BatchNormalizati (None, 30, 40, 256)  1024        res4e_branch2b[0][0]             
__________________________________________________________________________________________________
activation_36 (Activation)      (None, 30, 40, 256)  0           bn4e_branch2b[0][0]              
__________________________________________________________________________________________________
res4e_branch2c (Conv2D)         (None, 30, 40, 1024) 263168      activation_36[0][0]              
__________________________________________________________________________________________________
bn4e_branch2c (BatchNormalizati (None, 30, 40, 1024) 4096        res4e_branch2c[0][0]             
__________________________________________________________________________________________________
add_12 (Add)                    (None, 30, 40, 1024) 0           bn4e_branch2c[0][0]              
                                                                 activation_34[0][0]              
__________________________________________________________________________________________________
activation_37 (Activation)      (None, 30, 40, 1024) 0           add_12[0][0]                     
__________________________________________________________________________________________________
res4f_branch2a (Conv2D)         (None, 30, 40, 256)  262400      activation_37[0][0]              
__________________________________________________________________________________________________
bn4f_branch2a (BatchNormalizati (None, 30, 40, 256)  1024        res4f_branch2a[0][0]             
__________________________________________________________________________________________________
activation_38 (Activation)      (None, 30, 40, 256)  0           bn4f_branch2a[0][0]              
__________________________________________________________________________________________________
res4f_branch2b (Conv2D)         (None, 30, 40, 256)  590080      activation_38[0][0]              
__________________________________________________________________________________________________
bn4f_branch2b (BatchNormalizati (None, 30, 40, 256)  1024        res4f_branch2b[0][0]             
__________________________________________________________________________________________________
activation_39 (Activation)      (None, 30, 40, 256)  0           bn4f_branch2b[0][0]              
__________________________________________________________________________________________________
res4f_branch2c (Conv2D)         (None, 30, 40, 1024) 263168      activation_39[0][0]              
__________________________________________________________________________________________________
bn4f_branch2c (BatchNormalizati (None, 30, 40, 1024) 4096        res4f_branch2c[0][0]             
__________________________________________________________________________________________________
add_13 (Add)                    (None, 30, 40, 1024) 0           bn4f_branch2c[0][0]              
                                                                 activation_37[0][0]              
__________________________________________________________________________________________________
activation_40 (Activation)      (None, 30, 40, 1024) 0           add_13[0][0]                     
__________________________________________________________________________________________________
res5a_branch2a (Conv2D)         (None, 15, 20, 512)  524800      activation_40[0][0]              
__________________________________________________________________________________________________
bn5a_branch2a (BatchNormalizati (None, 15, 20, 512)  2048        res5a_branch2a[0][0]             
__________________________________________________________________________________________________
activation_41 (Activation)      (None, 15, 20, 512)  0           bn5a_branch2a[0][0]              
__________________________________________________________________________________________________
res5a_branch2b (Conv2D)         (None, 15, 20, 512)  2359808     activation_41[0][0]              
__________________________________________________________________________________________________
bn5a_branch2b (BatchNormalizati (None, 15, 20, 512)  2048        res5a_branch2b[0][0]             
__________________________________________________________________________________________________
activation_42 (Activation)      (None, 15, 20, 512)  0           bn5a_branch2b[0][0]              
__________________________________________________________________________________________________
res5a_branch2c (Conv2D)         (None, 15, 20, 2048) 1050624     activation_42[0][0]              
__________________________________________________________________________________________________
res5a_branch1 (Conv2D)          (None, 15, 20, 2048) 2099200     activation_40[0][0]              
__________________________________________________________________________________________________
bn5a_branch2c (BatchNormalizati (None, 15, 20, 2048) 8192        res5a_branch2c[0][0]             
__________________________________________________________________________________________________
bn5a_branch1 (BatchNormalizatio (None, 15, 20, 2048) 8192        res5a_branch1[0][0]              
__________________________________________________________________________________________________
add_14 (Add)                    (None, 15, 20, 2048) 0           bn5a_branch2c[0][0]              
                                                                 bn5a_branch1[0][0]               
__________________________________________________________________________________________________
activation_43 (Activation)      (None, 15, 20, 2048) 0           add_14[0][0]                     
__________________________________________________________________________________________________
res5b_branch2a (Conv2D)         (None, 15, 20, 512)  1049088     activation_43[0][0]              
__________________________________________________________________________________________________
bn5b_branch2a (BatchNormalizati (None, 15, 20, 512)  2048        res5b_branch2a[0][0]             
__________________________________________________________________________________________________
activation_44 (Activation)      (None, 15, 20, 512)  0           bn5b_branch2a[0][0]              
__________________________________________________________________________________________________
res5b_branch2b (Conv2D)         (None, 15, 20, 512)  2359808     activation_44[0][0]              
__________________________________________________________________________________________________
bn5b_branch2b (BatchNormalizati (None, 15, 20, 512)  2048        res5b_branch2b[0][0]             
__________________________________________________________________________________________________
activation_45 (Activation)      (None, 15, 20, 512)  0           bn5b_branch2b[0][0]              
__________________________________________________________________________________________________
res5b_branch2c (Conv2D)         (None, 15, 20, 2048) 1050624     activation_45[0][0]              
__________________________________________________________________________________________________
bn5b_branch2c (BatchNormalizati (None, 15, 20, 2048) 8192        res5b_branch2c[0][0]             
__________________________________________________________________________________________________
add_15 (Add)                    (None, 15, 20, 2048) 0           bn5b_branch2c[0][0]              
                                                                 activation_43[0][0]              
__________________________________________________________________________________________________
activation_46 (Activation)      (None, 15, 20, 2048) 0           add_15[0][0]                     
__________________________________________________________________________________________________
res5c_branch2a (Conv2D)         (None, 15, 20, 512)  1049088     activation_46[0][0]              
__________________________________________________________________________________________________
bn5c_branch2a (BatchNormalizati (None, 15, 20, 512)  2048        res5c_branch2a[0][0]             
__________________________________________________________________________________________________
activation_47 (Activation)      (None, 15, 20, 512)  0           bn5c_branch2a[0][0]              
__________________________________________________________________________________________________
res5c_branch2b (Conv2D)         (None, 15, 20, 512)  2359808     activation_47[0][0]              
__________________________________________________________________________________________________
bn5c_branch2b (BatchNormalizati (None, 15, 20, 512)  2048        res5c_branch2b[0][0]             
__________________________________________________________________________________________________
activation_48 (Activation)      (None, 15, 20, 512)  0           bn5c_branch2b[0][0]              
__________________________________________________________________________________________________
res5c_branch2c (Conv2D)         (None, 15, 20, 2048) 1050624     activation_48[0][0]              
__________________________________________________________________________________________________
bn5c_branch2c (BatchNormalizati (None, 15, 20, 2048) 8192        res5c_branch2c[0][0]             
__________________________________________________________________________________________________
add_16 (Add)                    (None, 15, 20, 2048) 0           bn5c_branch2c[0][0]              
                                                                 activation_46[0][0]              
__________________________________________________________________________________________________
activation_49 (Activation)      (None, 15, 20, 2048) 0           add_16[0][0]                     
==================================================================================================
Total params: 23,587,712
Trainable params: 23,534,592
Non-trainable params: 53,120
__________________________________________________________________________________________________
__________________________________________________________________________________________________
Layer (type)                    Output Shape         Param #     Connected to                     
==================================================================================================
input_1 (InputLayer)            (None, 480, 640, 3)  0                                            
__________________________________________________________________________________________________
conv1_pad (ZeroPadding2D)       (None, 486, 646, 3)  0           input_1[0][0]                    
__________________________________________________________________________________________________
conv1 (Conv2D)                  (None, 240, 320, 64) 9472        conv1_pad[0][0]                  
__________________________________________________________________________________________________
bn_conv1 (BatchNormalization)   (None, 240, 320, 64) 256         conv1[0][0]                      
__________________________________________________________________________________________________
activation_1 (Activation)       (None, 240, 320, 64) 0           bn_conv1[0][0]                   
__________________________________________________________________________________________________
pool1_pad (ZeroPadding2D)       (None, 242, 322, 64) 0           activation_1[0][0]               
__________________________________________________________________________________________________
max_pooling2d_1 (MaxPooling2D)  (None, 120, 160, 64) 0           pool1_pad[0][0]                  
__________________________________________________________________________________________________
res2a_branch2a (Conv2D)         (None, 120, 160, 64) 4160        max_pooling2d_1[0][0]            
__________________________________________________________________________________________________
bn2a_branch2a (BatchNormalizati (None, 120, 160, 64) 256         res2a_branch2a[0][0]             
__________________________________________________________________________________________________
activation_2 (Activation)       (None, 120, 160, 64) 0           bn2a_branch2a[0][0]              
__________________________________________________________________________________________________
res2a_branch2b (Conv2D)         (None, 120, 160, 64) 36928       activation_2[0][0]               
__________________________________________________________________________________________________
bn2a_branch2b (BatchNormalizati (None, 120, 160, 64) 256         res2a_branch2b[0][0]             
__________________________________________________________________________________________________
activation_3 (Activation)       (None, 120, 160, 64) 0           bn2a_branch2b[0][0]              
__________________________________________________________________________________________________
res2a_branch2c (Conv2D)         (None, 120, 160, 256 16640       activation_3[0][0]               
__________________________________________________________________________________________________
res2a_branch1 (Conv2D)          (None, 120, 160, 256 16640       max_pooling2d_1[0][0]            
__________________________________________________________________________________________________
bn2a_branch2c (BatchNormalizati (None, 120, 160, 256 1024        res2a_branch2c[0][0]             
__________________________________________________________________________________________________
bn2a_branch1 (BatchNormalizatio (None, 120, 160, 256 1024        res2a_branch1[0][0]              
__________________________________________________________________________________________________
add_1 (Add)                     (None, 120, 160, 256 0           bn2a_branch2c[0][0]              
                                                                 bn2a_branch1[0][0]               
__________________________________________________________________________________________________
activation_4 (Activation)       (None, 120, 160, 256 0           add_1[0][0]                      
__________________________________________________________________________________________________
res2b_branch2a (Conv2D)         (None, 120, 160, 64) 16448       activation_4[0][0]               
__________________________________________________________________________________________________
bn2b_branch2a (BatchNormalizati (None, 120, 160, 64) 256         res2b_branch2a[0][0]             
__________________________________________________________________________________________________
activation_5 (Activation)       (None, 120, 160, 64) 0           bn2b_branch2a[0][0]              
__________________________________________________________________________________________________
res2b_branch2b (Conv2D)         (None, 120, 160, 64) 36928       activation_5[0][0]               
__________________________________________________________________________________________________
bn2b_branch2b (BatchNormalizati (None, 120, 160, 64) 256         res2b_branch2b[0][0]             
__________________________________________________________________________________________________
activation_6 (Activation)       (None, 120, 160, 64) 0           bn2b_branch2b[0][0]              
__________________________________________________________________________________________________
res2b_branch2c (Conv2D)         (None, 120, 160, 256 16640       activation_6[0][0]               
__________________________________________________________________________________________________
bn2b_branch2c (BatchNormalizati (None, 120, 160, 256 1024        res2b_branch2c[0][0]             
__________________________________________________________________________________________________
add_2 (Add)                     (None, 120, 160, 256 0           bn2b_branch2c[0][0]              
                                                                 activation_4[0][0]               
__________________________________________________________________________________________________
activation_7 (Activation)       (None, 120, 160, 256 0           add_2[0][0]                      
__________________________________________________________________________________________________
res2c_branch2a (Conv2D)         (None, 120, 160, 64) 16448       activation_7[0][0]               
__________________________________________________________________________________________________
bn2c_branch2a (BatchNormalizati (None, 120, 160, 64) 256         res2c_branch2a[0][0]             
__________________________________________________________________________________________________
activation_8 (Activation)       (None, 120, 160, 64) 0           bn2c_branch2a[0][0]              
__________________________________________________________________________________________________
res2c_branch2b (Conv2D)         (None, 120, 160, 64) 36928       activation_8[0][0]               
__________________________________________________________________________________________________
bn2c_branch2b (BatchNormalizati (None, 120, 160, 64) 256         res2c_branch2b[0][0]             
__________________________________________________________________________________________________
activation_9 (Activation)       (None, 120, 160, 64) 0           bn2c_branch2b[0][0]              
__________________________________________________________________________________________________
res2c_branch2c (Conv2D)         (None, 120, 160, 256 16640       activation_9[0][0]               
__________________________________________________________________________________________________
bn2c_branch2c (BatchNormalizati (None, 120, 160, 256 1024        res2c_branch2c[0][0]             
__________________________________________________________________________________________________
add_3 (Add)                     (None, 120, 160, 256 0           bn2c_branch2c[0][0]              
                                                                 activation_7[0][0]               
__________________________________________________________________________________________________
activation_10 (Activation)      (None, 120, 160, 256 0           add_3[0][0]                      
==================================================================================================
Total params: 229,760
Trainable params: 226,816
Non-trainable params: 2,944
__________________________________________________________________________________________________

Process finished with exit code 0

除了截取部分，还是希望有网友能够留言完整的拼接部分，将含权重的已有模型与自己的模型进行拼接，从而针对已有的数据训练，生成新的网络模型结构~

 作者也是深度学习的小白，在读研究生一枚，欢迎交流和讨论

转载需注明，谢谢~

 

Monster_H777
关注
  • 1
    点赞
  • 踩
  • 4
    收藏
    觉得还不错? 一键收藏
  • 打赏
    打赏
  • 0
    评论
专栏目录
Pytorch学习第六讲：实现裁剪后的resnet50
Vivianyzw的博客
07-22 2507
先上一个pytorch里自带的一个版本 这个版本实现的比较简洁优雅，但如果你想自己改变一些参数，就比较难调，所以这篇文章，会基于该版本中最基本的一个bottleblock来做一个自己实现的版本，这个版本更方便修改参数，调试等。 import torch.nn as nn import torch.nn.functional as F import math import torch def c...
提取resnet网络任意层输出的代码怎么写
weixin_42576467的博客
02-17 239
可以使用PyTorch中的torch.utils.model_zoo模块，该模块可以轻松提取resnet网络中的任意层输出。示例代码如下：model = torch.utils.model_zoo.load_url('https://download.pytorch.org/models/resnet50-19c8e357.pth') layer_output = model.layer[i] #...
【很简单的教程】如何提取模型每层的特征【以ResNet为例】
weixin_44316581的博客
04-10 7332
介于CSDN上没有人很详细地讲述如何提取特征，所以我踩了很多坑，本文教程是我踩坑的心路历程，最后面有提取特征的方法实例，不过建议从头阅读。 方法1 以一个十分类的问题为例，我最开始用于提取特征的方法是这样的： class model(nn.Module): def __init__(self): super(model, self).__init__() self.model1 = torchvision.models.resnet18(pretrained=None
如何从resnet等建好的模型中删掉一部分，保留想要的
Arsmart的博客
11-04 809
我正在研究resnet，但是不想要最后的全连接层。而是希望输出保持 2,3,224,224，这样，batch，通道，长，宽，的格式。
快速抽取resnet_v2_152中间的特征层
QQ704630835的博客
07-29 972
基于tensorflow做resent抽取中间层特征
keras-resnet：用于深度残差网络的Keras软件包
02-06
`keras-resnet` 包含多种ResNet变体的实现，如ResNet-18、ResNet-34、ResNet-50、ResNet-101和ResNet-152。这些模型的差异主要在于残差块的数量和结构。例如，ResNet-50引入了“瓶颈”结构，每个残差块包含三个卷...
人工智能-残差神经网络-基于Keras+Tensorflow搭建，提供ResNet50神经网络的图片分类平台
最新发布
06-17
2. **加载ResNet50模型**：Keras提供了预训练的ResNet50模型，我们可以直接导入并利用其预训练的权重。这可以通过`keras.applications.resnet50`模块完成。 3. **微调模型**：预训练的ResNet50已经学到了大量的通用...
keras-resnet：使用Keras-1.0功能API的残留网络实现
02-05
keras-resnet 使用Keras-1.0功能API的残留网络实现，可与theano / tensorflow后端和'th'/'tf'图像暗淡排序一起使用。原始文章（2015年ImageNet竞赛获胜者）残留块残差块基于中的提出的新改进方案，如图（b）所示。 ...
keras-maskrcnn-0.2.2-resnet50_coco_v0.2.0-1.h5
08-30
maskrcnn训练模型，github下载太慢，这里提供给大家下载;
基于Keras+Resnet_v1对cifar-10进行图像分类
05-17
ResNet（残差网络）是一种深度神经网络结构，由Microsoft Research提出，其特点是引入了"跳跃连接"或"残差块"，解决了深度学习中的梯度消失和模型退化问题。 ResNet_v1是ResNet的早期版本，它通过保留输入信号并...
机器学习CV代码练习（九）之Resnet50模型实现
Wangzx的博客
07-20 321
机器学习CV代码练习（九）之Resnet50模型实现 # -*- coding: utf-8 -*- '''ResNet50 model for Keras. # Reference: - [Deep Residual Learning for Image Recognition](https://arxiv.org/abs/1512.03385) ''' from keras.layers import Input from keras import layers from keras.layer
python res_利用python一层一层可视化卷积神经网络，以ResNet50为例
weixin_39861918的博客
12-04 871
引言一直以来，卷积神经网络对人们来说都是一个黑箱，我们只知道它识别图片准确率很惊人，但是具体是怎么做到的，它究竟使用了什么特征来分辨图像，我们一无所知。无数的学者、研究人员都想弄清楚CNN内部运作的机制，甚至试图找到卷积神经网络和生物神经网络的联系。2013年，纽约大学的Matthew Zeiler和Rob Fergus的论文Visualizing and Understanding Convol...
ResNet18结构、各层输出维度
yexuejianghan的博客
07-20 9742
import torchvision.models as models resnet18 = models.resnet18() print(resnet18) ResNet( (conv1): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False) (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_.
Pytorch学习笔记（I）——预训练模型（七）：ResNet18网络结构
lockonlxf的博客
05-21 5963
ResNet( (conv1): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False) (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(...
pytorch 提取resnet模型特征
u013972657的博客
04-20 8088
下面使用三种方法提取resnet50最后一个卷积层经过bn及relu之后的输出值,去掉池化层以及全连接层。 方法一：使用nn.Sequential结构 import torch import torch.nn as nn from torchvision import datasets, transforms from torchvision import models from PIL import Image import numpy as np if __name__ == '__main__
Keras 学习之旅（一）
weixin_33928467的博客
01-07 316
软件环境(Windows): Visual Studio Anaconda CUDA MinGW-w64 conda install -c anaconda mingw libpython CNTK TensorFlow-gpu Keras-gpu Theano MKL CuDNN 参考书籍：谢梁 , 鲁颖 , 劳虹岚.Keras快速上手：基于P...
keras实现resnet50
gm_Ergou的博客
07-05 2318
数据集结构如第一篇文章（keras实现LeNet5）。 1.model.py 构造网络 # coding=utf-8 from keras.models import Model from keras.layers import Input, Dense, BatchNormalization, Conv2D, MaxPooling2D, AveragePooling2D, ZeroPadding2D from keras.layers import add, Flatten, Activ..
Resnet50源码-tensorflow+keras详细解析
热门推荐
ghw15221836342的博客
04-03 2万+
Resnet50源码-tensorflow解析原理解析：何凯明论文PPT-秒懂原理项目地址：Resnet50源码参考keras中的源码进行解析先加载一些库的文件from __future__ import print_function import numpy as np import warnings from keras.layers import Input from keras imp...
如何给keras层命名，并提取中间层输出值，保存到文档
龙猫
08-08 1万+
更新： 感谢评论区提供的方案。 采用model.summary(),model.get_config()和for循环均可获得Keras的层名。示例如下图   —————————————————————————————————————————————————— 对于keras特定层的命名，只需在层内添加 name 即可 model.add(Activation('softmax',nam...
用keras实现Resnet50
05-23
好的，以下是使用Keras实现ResNet50的示例代码： ``` from keras.layers import Input, Conv2D, MaxPooling2D, AveragePooling2D, Flatten, Dense, BatchNormalization, Activation, ZeroPadding2D from keras.models import Model from keras.initializers import glorot_uniform def identity_block(X, f, filters, stage, block): """ 实现恒等块 参数： X - 输入的tensor类型的数据，维度为( m, n_H_prev, n_W_prev, n_H_prev ) f - 整数，指定主路径中间的CONV窗口的维度 filters - 整数列表，定义了主路径每层的卷积核数量 stage - 整数，根据每层的位置来命名每一层，与block参数一起使用。 block - 字符串，根据每层的位置来命名每一层，与stage参数一起使用。 返回： X - 恒等块的输出，tensor类型，维度为(n_H, n_W, n_C) """ # 定义命名规则 conv_name_base = 'res' + str(stage) + block + '_branch' bn_name_base = 'bn' + str(stage) + block + '_branch' # 获取过滤器 F1, F2, F3 = filters # 保存输入数据，将会用于为主路径添加捷径 X_shortcut = X # 主路径的第一部分 X = Conv2D(filters=F1, kernel_size=(1, 1), strides=(1, 1), padding='valid', name=conv_name_base + '2a', kernel_initializer=glorot_uniform(seed=0))(X) X = BatchNormalization(axis=3, name=bn_name_base + '2a')(X) X = Activation('relu')(X) # 主路径的第二部分 X = Conv2D(filters=F2, kernel_size=(f, f), strides=(1, 1), padding='same', name=conv_name_base + '2b', kernel_initializer=glorot_uniform(seed=0))(X) X = BatchNormalization(axis=3, name=bn_name_base + '2b')(X) X = Activation('relu')(X) # 主路径的第三部分 X = Conv2D(filters=F3, kernel_size=(1, 1), strides=(1, 1), padding='valid', name=conv_name_base + '2c', kernel_initializer=glorot_uniform(seed=0))(X) X = BatchNormalization(axis=3, name=bn_name_base + '2c')(X) # 主路径的最后一步：添加捷径值，并通过ReLU激活函数进行处理 X = Add()([X, X_shortcut]) X = Activation('relu')(X) return X def convolutional_block(X, f, filters, stage, block, s=2): """ 实现卷积块 参数： X - 输入的tensor类型的数据，维度为( m, n_H_prev, n_W_prev, n_H_prev ) f - 整数，指定主路径中间的CONV窗口的维度 filters - 整数列表，定义了主路径每层的卷积核数量 stage - 整数，根据每层的位置来命名每一层，与block参数一起使用。 block - 字符串，根据每层的位置来命名每一层，与stage参数一起使用。 s - 整数，指定要使用的步幅 返回： X - 卷积块的输出，tensor类型，维度为(n_H, n_W, n_C) """ # 定义命名规则 conv_name_base = 'res' + str(stage) + block + '_branch' bn_name_base = 'bn' + str(stage) + block + '_branch' # 获取过滤器 F1, F2, F3 = filters # 保存输入数据，将会用于为主路径添加捷径 X_shortcut = X # 主路径的第一部分 X = Conv2D(filters=F1, kernel_size=(1, 1), strides=(s, s), padding='valid', name=conv_name_base + '2a', kernel_initializer=glorot_uniform(seed=0))(X) X = BatchNormalization(axis=3, name=bn_name_base + '2a')(X) X = Activation('relu')(X) # 主路径的第二部分 X = Conv2D(filters=F2, kernel_size=(f, f), strides=(1, 1), padding='same', name=conv_name_base + '2b', kernel_initializer=glorot_uniform(seed=0))(X) X = BatchNormalization(axis=3, name=bn_name_base + '2b')(X) X = Activation('relu')(X) # 主路径的第三部分 X = Conv2D(filters=F3, kernel_size=(1, 1), strides=(1, 1), padding='valid', name=conv_name_base + '2c', kernel_initializer=glorot_uniform(seed=0))(X) X = BatchNormalization(axis=3, name=bn_name_base + '2c')(X) # 捷径路径 X_shortcut = Conv2D(filters=F3, kernel_size=(1, 1), strides=(s, s), padding='valid', name=conv_name_base + '1', kernel_initializer=glorot_uniform(seed=0))(X_shortcut) X_shortcut = BatchNormalization(axis=3, name=bn_name_base + '1')(X_shortcut) # 主路径的最后一步：添加捷径值，并通过ReLU激活函数进行处理 X = Add()([X, X_shortcut]) X = Activation('relu')(X) return X def ResNet50(input_shape=(64, 64, 3), classes=6): """ 实现ResNet50 参数： input_shape - 输入数据的维度 classes - 分类数 返回： model - Keras框架的模型 """ # 定义输入为维度为input_shape的tensor X_input = Input(input_shape) # 使用0填充：X_input的周围填充0，填充量为3 X = ZeroPadding2D((3, 3))(X_input) # 第1阶段 X = Conv2D(filters=64, kernel_size=(7, 7), strides=(2, 2), name='conv1', kernel_initializer=glorot_uniform(seed=0))(X) X = BatchNormalization(axis=3, name='bn_conv1')(X) X = Activation('relu')(X) X = MaxPooling2D(pool_size=(3, 3), strides=(2, 2))(X) # 第2阶段 X = convolutional_block(X, f=3, filters=[64, 64, 256], stage=2, block='a', s=1) X = identity_block(X, f=3, filters=[64, 64, 256], stage=2, block='b') X = identity_block(X, f=3, filters=[64, 64, 256], stage=2, block='c') # 第3阶段 X = convolutional_block(X, f=3, filters=[128, 128, 512], stage=3, block='a', s=2) X = identity_block(X, f=3, filters=[128, 128, 512], stage=3, block='b') X = identity_block(X, f=3, filters=[128, 128, 512], stage=3, block='c') X = identity_block(X, f=3, filters=[128, 128, 512], stage=3, block='d') # 第4阶段 X = convolutional_block(X, f=3, filters=[256, 256, 1024], stage=4, block='a', s=2) X = identity_block(X, f=3, filters=[256, 256, 1024], stage=4, block='b') X = identity_block(X, f=3, filters=[256, 256, 1024], stage=4, block='c') X = identity_block(X, f=3, filters=[256, 256, 1024], stage=4, block='d') X = identity_block(X, f=3, filters=[256, 256, 1024], stage=4, block='e') X = identity_block(X, f=3, filters=[256, 256, 1024], stage=4, block='f') # 第5阶段 X = convolutional_block(X, f=3, filters=[512, 512, 2048], stage=5, block='a', s=2) X = identity_block(X, f=3, filters=[512, 512, 2048], stage=5, block='b') X = identity_block(X, f=3, filters=[512, 512, 2048], stage=5, block='c') # 平均池化层 X = AveragePooling2D(pool_size=(2, 2), padding='same')(X) # 输出层 X = Flatten()(X) X = Dense(classes, activation='softmax', name='fc' + str(classes), kernel_initializer=glorot_uniform(seed=0))(X) # 创建模型 model = Model(inputs=X_input, outputs=X, name='ResNet50') return model ``` 以上代码实现了ResNet50模型，包括了卷积块、恒等块、输入层、输出层等组件。可以使用该代码构建一个ResNet50模型并进行训练。
评论
添加红包

请填写红包祝福语或标题

个

红包个数最小为10个

元

红包金额最低5元

当前余额3.43元 前往充值 >
需支付：10.00元
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包

打赏作者

Monster_H777

我直说吧：你的奖励我的动力～

¥1 ¥2 ¥4 ¥6 ¥10 ¥20
扫码支付：¥1
获取中
扫码支付

您的余额不足，请更换扫码支付或充值

打赏作者

实付元
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明：

1.余额是钱包充值的虚拟货币，按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载，可以购买VIP、付费专栏及课程。

余额充值