VGG16

最新推荐文章于 2023-06-21 18:06:22 发布
最新推荐文章于 2023-06-21 18:06:22 发布
阅读量2.9k 收藏 3
点赞数 3
分类专栏： Caffe
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/raby_gyl/article/details/79428737
版权

Models from the BMVC-2014 paper “Return of the Devil in the Details: Delving Deep into Convolutional Nets”

The models are trained on the ILSVRC-2012 dataset. The details can be found on the project page or in the following BMVC-2014 paper:

Return of the Devil in the Details: Delving Deep into Convolutional
Nets K. Chatfield, K. Simonyan, A. Vedaldi, A. Zisserman British
Machine Vision Conference, 2014 (arXiv ref. cs1405.3531)

Please cite the paper if you use the models.

Models:

VGG_CNN_S: 13.1% top-5 error on ILSVRC-2012-val
VGG_CNN_M: 13.7% top-5 error on ILSVRC-2012-val
VGG_CNN_M_2048: 13.5% top-5 error on ILSVRC-2012-val
VGG_CNN_M_1024: 13.7% top-5 error on ILSVRC-2012-val
VGG_CNN_M_128: 15.6% top-5 error on ILSVRC-2012-val
VGG_CNN_F: 16.7% top-5 error on ILSVRC-2012-val

Models used by the VGG team in ILSVRC-2014
The models are the improved versions of the models used by the VGG team in the ILSVRC-2014 competition. The details can be found on the project page or in the following arXiv paper:

Very Deep Convolutional Networks for Large-Scale Image Recognition K.
Simonyan, A. Zisserman arXiv:1409.1556

Please cite the paper if you use the models.

Models:

16-layer: 7.5% top-5 error on ILSVRC-2012-val, 7.4% top-5 error on ILSVRC-2012-test
19-layer: 7.5% top-5 error on ILSVRC-2012-val, 7.3% top-5 error on ILSVRC-2012-test

In the paper, the models are denoted as configurations D and E, trained with scale jittering. The combination of the two models achieves 7.1% top-5 error on ILSVRC-2012-val, and 7.0% top-5 error on ILSVRC-2012-test.

VGG16 for face recognition: caffemodel,
http://www.robots.ox.ac.uk/~vgg/software/vgg_face/

注释：
gist.github.com 被墙无法访问解决办法
windows下 打开C:\Windows\System32\drivers\etc\hosts文件
编辑器打开，在最后行添加192.30.253.118 gist.github.com
保存。

从头开始训练，将参数的“gaussian”初始方式 修改为“xavier”, 否则训练的时候可能不会收敛。参数的初始化方式很重要。

train.prototxt

layer {
  name: "data"
  type: "Python"
  top: "Features"
  top: "Headposes"
  python_param {
    module: "python_read_data_for_softmax"
    layer: "AllDataLayer"
    param_str: "{\'phase\': \'train\', \'dataset_name\': \'lfw\', \'data_type\': \'image\',\'batch_size\': 48,\'cross_id\':0}"
  }
}

layer {  
  bottom: "Features"  
  top: "conv1_1"  
  name: "conv1_1"  
  type: "Convolution"  
  param {  
    lr_mult: 1  
    decay_mult: 1  
  }  
  param {  
    lr_mult: 2  
    decay_mult: 0  
  }  
  convolution_param {  
    num_output: 64  
    pad: 1  
    kernel_size: 3  
    weight_filler {  
      type: "xavier"  
      std: 0.01  
    }  
    bias_filler {  
      type: "constant"  
      value: 0  
    }  
  }  
}  
layer {  
  bottom: "conv1_1"  
  top: "conv1_1"  
  name: "relu1_1"  
  type: "ReLU"  
}  
layer {  
  bottom: "conv1_1"  
  top: "conv1_2"  
  name: "conv1_2"  
  type: "Convolution"  
  param {  
    lr_mult: 1  
    decay_mult: 1  
  }  
  param {  
    lr_mult: 2  
    decay_mult: 0  
  }  
  convolution_param {  
    num_output: 64  
    pad: 1  
    kernel_size: 3  
    weight_filler {  
      type: "xavier"  
      std: 0.01  
    }  
    bias_filler {  
      type: "constant"  
      value: 0  
    }  
  }  
}  
layer {  
  bottom: "conv1_2"  
  top: "conv1_2"  
  name: "relu1_2"  
  type: "ReLU"  
}  
layer {  
  bottom: "conv1_2"  
  top: "pool1"  
  name: "pool1"  
  type: "Pooling"  
  pooling_param {  
    pool: MAX  
    kernel_size: 2  
    stride: 2  
  }  
}  
layer {  
  bottom: "pool1"  
  top: "conv2_1"  
  name: "conv2_1"  
  type: "Convolution"  
  param {  
    lr_mult: 1  
    decay_mult: 1  
  }  
  param {  
    lr_mult: 2  
    decay_mult: 0  
  }  
  convolution_param {  
    num_output: 128  
    pad: 1  
    kernel_size: 3  
    weight_filler {  
      type: "xavier"  
      std: 0.01  
    }  
    bias_filler {  
      type: "constant"  
      value: 0  
    }  
  }  
}  
layer {  
  bottom: "conv2_1"  
  top: "conv2_1"  
  name: "relu2_1"  
  type: "ReLU"  
}  
layer {  
  bottom: "conv2_1"  
  top: "conv2_2"  
  name: "conv2_2"  
  type: "Convolution"  
  param {  
    lr_mult: 1  
    decay_mult: 1  
  }  
  param {  
    lr_mult: 2  
    decay_mult: 0  
  }  
  convolution_param {  
    num_output: 128  
    pad: 1  
    kernel_size: 3  
    weight_filler {  
      type: "xavier"  
      std: 0.01  
    }  
    bias_filler {  
      type: "constant"  
      value: 0  
    }  
  }  
}  
layer {  
  bottom: "conv2_2"  
  top: "conv2_2"  
  name: "relu2_2"  
  type: "ReLU"  
}  
layer {  
  bottom: "conv2_2"  
  top: "pool2"  
  name: "pool2"  
  type: "Pooling"  
  pooling_param {  
    pool: MAX  
    kernel_size: 2  
    stride: 2  
  }  
}  
layer {  
  bottom: "pool2"  
  top: "conv3_1"  
  name: "conv3_1"  
  type: "Convolution"  
  param {  
    lr_mult: 1  
    decay_mult: 1  
  }  
  param {  
    lr_mult: 2  
    decay_mult: 0  
  }  
  convolution_param {  
    num_output: 256  
    pad: 1  
    kernel_size: 3  
    weight_filler {  
      type: "xavier"  
      std: 0.01  
    }  
    bias_filler {  
      type: "constant"  
      value: 0  
    }  
  }  
}  
layer {  
  bottom: "conv3_1"  
  top: "conv3_1"  
  name: "relu3_1"  
  type: "ReLU"  
}  
layer {  
  bottom: "conv3_1"  
  top: "conv3_2"  
  name: "conv3_2"  
  type: "Convolution"  
  param {  
    lr_mult: 1  
    decay_mult: 1  
  }  
  param {  
    lr_mult: 2  
    decay_mult: 0  
  }  
  convolution_param {  
    num_output: 256  
    pad: 1  
    kernel_size: 3  
    weight_filler {  
      type: "xavier"  
      std: 0.01  
    }  
    bias_filler {  
      type: "constant"  
      value: 0  
    }  
  }  
}  
layer {  
  bottom: "conv3_2"  
  top: "conv3_2"  
  name: "relu3_2"  
  type: "ReLU"  
}  
layer {  
  bottom: "conv3_2"  
  top: "conv3_3"  
  name: "conv3_3"  
  type: "Convolution"  
  param {  
    lr_mult: 1  
    decay_mult: 1  
  }  
  param {  
    lr_mult: 2  
    decay_mult: 0  
  }  
  convolution_param {  
    num_output: 256  
    pad: 1  
    kernel_size: 3  
    weight_filler {  
      type: "xavier"  
      std: 0.01  
    }  
    bias_filler {  
      type: "constant"  
      value: 0  
    }  
  }  
}  
layer {  
  bottom: "conv3_3"  
  top: "conv3_3"  
  name: "relu3_3"  
  type: "ReLU"  
}  
layer {  
  bottom: "conv3_3"  
  top: "pool3"  
  name: "pool3"  
  type: "Pooling"  
  pooling_param {  
    pool: MAX  
    kernel_size: 2  
    stride: 2  
  }  
}  
layer {  
  bottom: "pool3"  
  top: "conv4_1"  
  name: "conv4_1"  
  type: "Convolution"  
  param {  
    lr_mult: 1  
    decay_mult: 1  
  }  
  param {  
    lr_mult: 2  
    decay_mult: 0  
  }  
  convolution_param {  
    num_output: 512  
    pad: 1  
    kernel_size: 3  
    weight_filler {  
      type: "xavier"  
      std: 0.01  
    }  
    bias_filler {  
      type: "constant"  
      value: 0  
    }  
  }  
}  
layer {  
  bottom: "conv4_1"  
  top: "conv4_1"  
  name: "relu4_1"  
  type: "ReLU"  
}  
layer {  
  bottom: "conv4_1"  
  top: "conv4_2"  
  name: "conv4_2"  
  type: "Convolution"  
  param {  
    lr_mult: 1  
    decay_mult: 1  
  }  
  param {  
    lr_mult: 2  
    decay_mult: 0  
  }  
  convolution_param {  
    num_output: 512  
    pad: 1  
    kernel_size: 3  
    weight_filler {  
      type: "xavier"  
      std: 0.01  
    }  
    bias_filler {  
      type: "constant"  
      value: 0  
    }  
  }  
}  
layer {  
  bottom: "conv4_2"  
  top: "conv4_2"  
  name: "relu4_2"  
  type: "ReLU"  
}  
layer {  
  bottom: "conv4_2"  
  top: "conv4_3"  
  name: "conv4_3"  
  type: "Convolution"  
  param {  
    lr_mult: 1  
    decay_mult: 1  
  }  
  param {  
    lr_mult: 2  
    decay_mult: 0  
  }  
  convolution_param {  
    num_output: 512  
    pad: 1  
    kernel_size: 3  
    weight_filler {  
      type: "xavier"  
      std: 0.01  
    }  
    bias_filler {  
      type: "constant"  
      value: 0  
    }  
  }  
}  
layer {  
  bottom: "conv4_3"  
  top: "conv4_3"  
  name: "relu4_3"  
  type: "ReLU"  
}  
layer {  
  bottom: "conv4_3"  
  top: "pool4"  
  name: "pool4"  
  type: "Pooling"  
  pooling_param {  
    pool: MAX  
    kernel_size: 2  
    stride: 2  
  }  
}  
layer {  
  bottom: "pool4"  
  top: "conv5_1"  
  name: "conv5_1"  
  type: "Convolution"  
  param {  
    lr_mult: 1  
    decay_mult: 1  
  }  
  param {  
    lr_mult: 2  
    decay_mult: 0  
  }  
  convolution_param {  
    num_output: 512  
    pad: 1  
    kernel_size: 3  
    weight_filler {  
      type: "xavier"  
      std: 0.01  
    }  
    bias_filler {  
      type: "constant"  
      value: 0  
    }  
  }  
}  
layer {  
  bottom: "conv5_1"  
  top: "conv5_1"  
  name: "relu5_1"  
  type: "ReLU"  
}  
layer {  
  bottom: "conv5_1"  
  top: "conv5_2"  
  name: "conv5_2"  
  type: "Convolution"  
  param {  
    lr_mult: 1  
    decay_mult: 1  
  }  
  param {  
    lr_mult: 2  
    decay_mult: 0  
  }  
  convolution_param {  
    num_output: 512  
    pad: 1  
    kernel_size: 3  
    weight_filler {  
      type: "xavier"  
      std: 0.01  
    }  
    bias_filler {  
      type: "constant"  
      value: 0  
    }  
  }  
}  
layer {  
  bottom: "conv5_2"  
  top: "conv5_2"  
  name: "relu5_2"  
  type: "ReLU"  
}  
layer {  
  bottom: "conv5_2"  
  top: "conv5_3"  
  name: "conv5_3"  
  type: "Convolution"  
  param {  
    lr_mult: 1  
    decay_mult: 1  
  }  
  param {  
    lr_mult: 2  
    decay_mult: 0  
  }  
  convolution_param {  
    num_output: 512  
    pad: 1  
    kernel_size: 3  
    weight_filler {  
      type: "xavier"  
      std: 0.01  
    }  
    bias_filler {  
      type: "constant"  
      value: 0  
    }  
  }  
}  
layer {  
  bottom: "conv5_3"  
  top: "conv5_3"  
  name: "relu5_3"  
  type: "ReLU"  
}  
layer {  
  bottom: "conv5_3"  
  top: "pool5"  
  name: "pool5"  
  type: "Pooling"  
  pooling_param {  
    pool: MAX  
    kernel_size: 2  
    stride: 2  
  }  
}  
layer {  
  bottom: "pool5"  
  top: "fc6"  
  name: "fc6"  
  type: "InnerProduct"  
  param {  
    lr_mult: 1  
    decay_mult: 1  
  }  
  param {  
    lr_mult: 2  
    decay_mult: 0  
  }  
  inner_product_param {  
    num_output: 4096  
    weight_filler {  
      type: "xavier"  
      std: 0.005  
    }  
    bias_filler {  
      type: "constant"  
      value: 0.1  
    }  
  }  
}  
layer {  
  bottom: "fc6"  
  top: "fc6"  
  name: "relu6"  
  type: "ReLU"  
}  
layer {  
  bottom: "fc6"  
  top: "fc6"  
  name: "drop6"  
  type: "Dropout"  
  dropout_param {  
    dropout_ratio: 0.5  
  }  
}  
layer {  
  bottom: "fc6"  
  top: "fc7"  
  name: "fc7"  
  type: "InnerProduct"  
  param {  
    lr_mult: 1  
    decay_mult: 1  
  }  
  param {  
    lr_mult: 2  
    decay_mult: 0  
  }  
  inner_product_param {  
    num_output: 4096  
    weight_filler {  
      type: "xavier"  
      std: 0.005  
    }  
    bias_filler {  
      type: "constant"  
      value: 0.1  
    }  
  }  
}  
layer {  
  bottom: "fc7"  
  top: "fc7"  
  name: "relu7"  
  type: "ReLU"  
}  
layer {  
  bottom: "fc7"  
  top: "fc7"  
  name: "drop7"  
  type: "Dropout"  
  dropout_param {  
    dropout_ratio: 0.5  
  }  
}  
layer {  
  bottom: "fc7"  
  top: "fc8"  
  name: "fc8"  
  type: "InnerProduct"  
  param {  
    lr_mult: 1  
    decay_mult: 1  
  }  
  param {  
    lr_mult: 2  
    decay_mult: 0  
  }  
  inner_product_param {  
    num_output: 5  
    weight_filler {  
      type: "xavier"  
      std: 0.005  
    }  
    bias_filler {  
      type: "constant"  
      value: 0.1  
    }  
  }  
}  

layer {  
  bottom: "fc8"  
  bottom: "Headposes"  
  top: "headposeloss1"  
  name: "headposeloss1"  
  type: "SoftmaxWithLoss"  
}

solver.prototxt:

train_net: "ZOO_VGG16/train2.prototxt"
base_lr: 0.01
display: 20
max_iter: 100000
lr_policy: "step"
gamma: 0.1
momentum: 0.9
weight_decay: 0.0005
stepsize: 50000
snapshot: 5000
snapshot_prefix: "snapshots_pointing04_vgg16_klLoss/cross1/vgg_"
random_seed: 831486
iter_size: 1

参考文献：

  1. https://github.com/NVIDIA/DIGITS/issues/159#issuecomment-247707549
  2. https://github.com/NVIDIA/DIGITS/issues/535
  3. https://stackoverflow.com/questions/42652903/caffe-net-not-converging-when-replacing-alexnet-with-vgg16-but-everything-else?rq=1
  4. https://github.com/BVLC/caffe/wiki/Model-Zoo
  5. https://www.cnblogs.com/ryans/p/8196151.html
vgg16_weights.zip
04-10
这是VGG16网络，vgg16_weights在多伦多大学的开源镜像，因直接在该原网站下载较慢且费时间，大家需要请自取
基于CNN的FashionMNIST数据集识别4——VGG-16模型
最新发布
师范大学生的博客
03-13 388
VGG-16 是牛津大学 ‌‌ 在 2014 年提出的经典卷积神经网络模型，发表于论文《Very Deep Convolutional Networks for Large-Scale Image Recognition》。它在 ImageNet 图像分类任务中取得了突破性成绩，核心贡献是证明了 ‌‌ 对模型性能的重要性（当然太深了也不好）。
vgg16_weights.h5
03-18
vgg16用于图像特征提取，tensorflow1.10.0，Keras 2.1.6
vgg16_weight_and_xception_weights
01-14
这有两个文件，分别是VGG16_weights_tf和xception_weights_tf,都是No_top的，有top太大没法传，其他inception_v3的两个文件和resnet50的两个文件分别压缩在一块，刚好小于200M，也上传上来了
深度学习VGG16模型vgg16_weights_tf_dim_ordering_tf_kernels.h5
weixin_45867990的博客
10-22 1218
深度学习VGG16模型vgg16_weights_tf_dim_ordering_tf_kernels.h5 https://github.com/fchollet/deep-learning-models/releases/download/v0.1/vgg16_weights_tf_dim_ordering_tf_kernels.h5 百度网盘下载链接：https://pan.baidu.com/s/1fDtb0Yml4SfNZkAhCCHjXw 提取码：s95v ...
VGG16.rar_VGG16模型详解_vgg16 python_vgg16模型_vgg16模型结构_wish154
07-14
VGG16模型是深度学习领域的一个里程碑，由牛津大学视觉几何组（Visual Geometry Group）的研究员在2014年的ImageNet Large Scale Visual Recognition Challenge（ILSVRC）上提出。模型以其深且窄的网络结构而闻名，...
VGG16_vgg16代码_VGG16_vgg16代码详解_
09-29
VGG16是一种深度卷积神经网络（Deep Convolutional Neural Network, DCNN），由英国牛津大学的Visual Geometry Group（VGG）在2014年的ImageNet竞赛中提出。该模型以其深且窄的架构闻名，拥有16个处理层，其中13层是...
vgg16特征提取_成功_利用vGG16_VGG16_vgg16特征提取_vgg16图像分类_图像分类_
10-03
VGG16是一种非常著名的深度学习模型，由牛津大学的Visual Geometry Group（VGG）在2014年的ImageNet Large Scale Visual Recognition Challenge（ILSVRC）中提出。这个模型以其深度和精细的卷积层结构而闻名，是深度...
vgg16-397923af.pth
08-17
VGG16是一个著名的卷积神经网络（CNN）模型，由英国剑桥大学的Visual Geometry Group（VGG）在2014年的ImageNet Large Scale Visual Recognition Challenge (ILSVRC) 中提出。这个模型因其深度和准确性而备受关注，...
VGG16_TF:简单实现VGG16
04-28
学习Tensorflow，拿VGG16练练手，没有其他骚操作，只有数据集制作，训练及测试。 训练数据-17flowers，百度网盘链接: : 3nc4 VGG16.npy，百度网盘链接: : 4wvq create_tfrecords.py为生成tfrecords数据脚本 VGG16....
vgg16_weights 权重下载百度云下载链接.docx
05-02
vgg16_weights 权重百度云下载链接
vgg16网络初始化权重
12-14
从谷歌上找了好久的，而且网速还贼拉的慢，为方便大家，上传共享
VGG16模型tensorflow实现
05-05
包括数据集预处理以及生成tfrecord文件，用自己数据集训练VGG16，亲测可用
VGG16权重（无全连接层）
08-21
卷积网路vgg16权重的pny文件，没有全连接层适合迁移学习
vgg16_weights_tf_dim_ordering_tf_kernels_notop.h5
xuemei0501的博客
11-12 1090
vgg16_weights_tf_dim_ordering_tf_kernels_notop.h5参数模型下载链接： 链接: https://pan.baidu.com/s/1V5_rU7xR6h_qiN8XOmq-qw 提取码: 27gp vgg16_weights_tf_dim_ordering_tf_kernels.h5 参数模型下载链接： 链接: https://pan.baidu.com/s/1-GcN488w1mVMv6amdLI2Sw 提取码: mdcx ...
深入理解VGG16模型与代码实现
热门推荐
u012661539的博客
06-21 2万+
VGG16基本架构与了解
Unable to load vgg16 weights
PinkAir的博客
10-29 385
https://github.com/keras-team/keras/issues/8998 generate=vgg16.VGG16(include_top=False,weights='imagenet',input_tensor=y_pre,input_shape=(None,None,3)) Error:ValueError: You are trying to load a weight file containing 13 layers into a model with 18 lay.
pytorch实现VGG16
不知者无畏
05-28 4500
运行代码： import torch from torch import optim import torchvision import matplotlib.pyplot as plt import numpy as np from torchvision.utils import make_grid import time # 配置参数 DOWNLOAD_CIFAR = True batch_size = 32 # 每次喂入的数据量 lr = 0.01 # 学习率 step_size = 10
vgg16
03-08
### VGG16 深度学习模型简介 VGG16 是一种经典的卷积神经网络架构，在多个计算机视觉任务中表现出色，包括图像分类、目标检测以及语义分割等[^2]。该模型由牛津大学的 Visual Geometry Group 提出，因其出色的性能和简洁的设计成为深度学习领域的重要组成部分。 ### 加载预训练的 VGG16 模型 为了快速上手使用 VGG16 进行图像处理任务，可以利用 PyTorch 和 TensorFlow 中提供的工具轻松加载预训练权重： #### 使用 PyTorch 加载 VGG16 ```python import torch from torchvision import models, transforms from PIL import Image # 定义设备 (CPU/GPU) device = 'cuda' if torch.cuda.is_available() else 'cpu' # 创建 VGG16 实例并加载预训练参数 model_vgg16 = models.vgg16(pretrained=True).to(device) # 设置为评估模式 model_vgg16.eval() ``` 对于特定应用场合下的微调或迁移学习，则可以通过修改最后一层全连接层来适应新的数据集需求: ```python num_features = model_vgg16.classifier[-1].in_features model_vgg16.classifier[-1] = torch.nn.Linear(num_features, num_classes) # 替换最后的线性层以匹配类别数 ``` #### 使用 TensorFlow/Keras 加载 VGG16 ```python import tensorflow as tf from tensorflow.keras.applications import VGG16 from tensorflow.keras.preprocessing.image import load_img, img_to_array # 不含顶层，默认输入形状(224, 224, 3)，包含ImageNet上的预训练权重 base_model = VGG16(weights='imagenet', include_top=False, input_shape=(224, 224, 3)) for layer in base_model.layers: layer.trainable = False # 冻结基础模型中的所有层 # 添加自定义顶部分类器 x = base_model.output x = tf.keras.layers.GlobalAveragePooling2D()(x) predictions = tf.keras.layers.Dense(num_classes, activation='softmax')(x) final_model = tf.keras.Model(inputs=base_model.input, outputs=predictions) ``` 上述代码展示了如何在不同框架下初始化带有预训练权值的基础 VGG16 架构，并对其进行适当调整以便于后续的应用开发工作[^3]。 ### 应用案例分析 考虑到实际应用场景的需求差异较大，这里提供了一个简单的例子——基于 VGG16 的二元分类问题解决方案（如区分猫狗图片）。此过程中不仅涉及到了基本的数据增强操作，还包含了针对具体任务定制化的模型结构调整方法。
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
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