系列文章目录
第一章 AlexNet网络详解
第二章 VGG网络详解
第三章 GoogLeNet网络详解
第四章 ResNet网络详解
第五章 ResNeXt网络详解
第六章 MobileNetv1网络详解
第七章 MobileNetv2网络详解
第八章 MobileNetv3网络详解
第九章 ShuffleNetv1网络详解
第十章 ShuffleNetv2网络详解
第十一章 EfficientNetv1网络详解
第十二章 EfficientNetv2网络详解
第十三章 Transformer注意力机制
第十四章 Vision Transformer网络详解
第十五章 Swin-Transformer网络详解
第十六章 ConvNeXt网络详解
第十七章 RepVGG网络详解
第十八章 MobileViT网络详解
文章目录
0、前言
1、摘要
在本研究中，我们探讨了卷积网络深度对于大规模图像识别准确性的影响。我们的主要贡献是对于使用非常小的（3 ×3）卷积过滤器的逐渐加深的网络进行彻底评估。研究结果表明，通过将深度推到16到19个权重层，可以显著提高传统网络配置的表现。这些研究成果是我们2014年ImageNet挑战赛提交的基础，我们的团队在定位和分类赛道中分别获得了第一和第二名。我们还展示了我们的表现良好的方法可以泛化到其他数据集，其中它们取得了最先进的结果。我们公开了我们表现最佳的ConvNet模型，以便促进关于深度视觉表征在计算机视觉中的使用的进一步研究。
2、VGG网络结构
"Very Deep Convolutional Networks for Large-Scale Image Recognition" (VGG) 是2014年由Karen Simonyan和Andrew Zisserman发表的一篇论文。该论文提出了一种卷积神经网络，它具有非常深的架构，被称为 VGG16 和 VGG19。 VGG网络是一种端到端的模型，可以接受任何大小的图像，但通常情况下，输入的图像大小为 ，类别数为1000。VGG网络中的卷积操作均使用
的卷积核，并且在每个卷积层之间使用池化层进行下采样。这种架构的主要优点是能够实现非常深的模型，并且可以容易地训练，因为所有的卷积层和池化层的结构都非常简单和相似。此外，在使用
的卷积核时，可以使用更多的层来增加网络的深度，而不会使参数数量增加过快。 VGG网络通过在两个 GPU 上进行训练，在 ILSVRC2014 图像识别竞赛中取得了第一名，同时还刷新了当时的记录。模型在该数据集上的 top-5 错误率仅为 7.3%。 总的来说，VGG网络通过简单而有效的模型架构取得了显著的性能提升，同时为后续深度学习模型设计提供了有力的参考与借鉴。
使用7x7的卷积核所需要的参数数量：
使用3个3x3的卷积核代替一个7x7卷积核所需要的参数数量：
经过卷积后的矩阵尺寸大小计算公式为：
1.VGG_Model.py(pytorch实现）
import torch.nn as nn
import torch
class AlexNet(nn.Module):
def __init__(self,num_classes=1000,init_weights=False):
super(AlexNet, self).__init__()
self.features = nn.Sequential(
nn.Conv2d(3, 48, kernel_size=11, stride=4, padding=2),
nn.ReLU(inplace=True),
nn.MaxPool2d(kernel_size=3, stride=2),
nn.Conv2d(48, 128, kernel_size=5, stride=1, padding=2),
nn.ReLU(inplace=True),
nn.MaxPool2d(kernel_size=3, stride=2),
nn.Conv2d(128, 192, kernel_size=3, stride=1, padding=1),
nn.ReLU(inplace=True),
nn.Conv2d(192, 192, kernel_size=3, stride=1, padding=1),
nn.ReLU(inplace=True),
nn.Conv2d(192, 128, kernel_size=3, stride=1, padding=1),
nn.ReLU(inplace=True),
nn.MaxPool2d(kernel_size=3,stride=2)
)
self.classifier = nn.Sequential(
nn.Dropout(p=0.5),
nn.Linear(128*6*6, 2048),
nn.ReLU(inplace=True),
nn.Dropout(p=0.5),
nn.Linear(2048, 2048),
nn.ReLU(inplace=True),
nn.Linear(2048, num_classes)
)
if init_weights:
self._initialize_weights()
def forward(self, x):
x = self.features(x)
x = torch.flatten(x, start_dim=1)
x = self.classifier(x)
return x
def _initialize_weights(self):
for m in self.modules():
if isinstance(m, nn.Conv2d):
nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu')
if m.bias is not None:
nn.init.constant_(m.bias, 0)
elif isinstance(m, nn.Linear):
nn.init.normal_(m.weight, 0, 0.01)
nn.init.constant_(m.bias, 0)
2.train.py
import os
import sys
import json
import torch
import torch.nn as nn
from torchvision import transforms, datasets, utils
import matplotlib.pyplot as plt
import numpy as np
import torch.optim as optim
from tqdm import tqdm
from model import AlexNet
def main():
device = torch.device("cuda:0" if torch.cuda.is_available() else 'cpu')
print("using {} device.".format(device))
data_transform = {
"train": transforms.Compose([transforms.RandomResizedCrop(224),
transforms.RandomHorizontalFlip(),
transforms.ToTensor(),
transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))]),
"val": transforms.Compose([transforms.Resize((224, 224)),
transforms.ToTensor(),
transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))])}
data_root = 'D:/100_DataSets/'
image_path = os.path.join(data_root, "03_flower_data")
assert os.path.exists(image_path), "{} path does not exits.".format(image_path)
train_dataset = datasets.ImageFolder(root=os.path.join(image_path, "train"),transform = data_transform['train'])
train_num = len(train_dataset)
flower_list = train_dataset.class_to_idx
cla_dict = dict((val, key) for key, val in flower_list.items())
json_str = json.dumps(cla_dict, indent=4)
with open('class_indices.json', 'w') as json_file:
json_file.write(json_str)
batch_size = 6
nw = min([os.cpu_count(), batch_size if batch_size > 1 else 0, 8])
print('Using {} dataloder workers every process'.format(nw))
train_loader = torch.utils.data.DataLoader(train_dataset,
batch_size=batch_size,
shuffle=True,
num_workers=nw)
validate_dataset = datasets.ImageFolder(root=os.path.join(image_path, "val"),
transform=data_transform['val'])
val_num = len(validate_dataset)
validate_loader = torch.utils.data.DataLoader(validate_dataset,
batch_size=4,
shuffle=False,
num_workers=nw)
print("using {} image for train, {} images for validation.".format(train_num, val_num))
net = AlexNet(num_classes=5, init_weights=True)
net.to(device)
loss_fuction = nn.CrossEntropyLoss()
optimizer = optim.Adam(net.parameters(), lr=0.0002)
epochs = 10
save_path = './AlexNet.pth'
best_acc = 0.0
train_steps = len(train_loader)
for epoch in range(epochs):
net.train()
running_loss = 0.0
train_bar = tqdm(train_loader, file=sys.stdout)
for step, data in enumerate(train_bar):
images, labels = data
optimizer.zero_grad()
outputs = net(images.to(device))
loss = loss_fuction(outputs, labels.to(device))
loss.backward()
optimizer.step()
running_loss += loss.item()
train_bar.desc = "train epoch[{}/{}] loss:{:,.3f}".format(epoch+1, epochs, loss)
net.eval()
acc = 0.0
with torch.no_grad():
val_bar = tqdm(validate_loader, file=sys.stdout)
for val_data in val_bar:
val_images, val_labels = val_data
outputs = net(val_images.to(device))
predict_y = torch.max(outputs, dim=1)[1]
acc += torch.eq(predict_y, val_labels.to(device)).sum().item()
val_accurate = acc / val_num
print('[epoch % d] train_loss: %.3f val_accuracy: %.3f' %
(epoch+1, running_loss / train_steps, val_accurate))
if val_accurate > best_acc:
best_acc = val_accurate
torch.save(net.state_dict(),save_path)
print("Finished Training")
if __name__ == '__main__':
main()
3.predict.py
import os
import json
import torch
from PIL import Image
from torchvision import transforms
import matplotlib.pyplot as plt
from model import AlexNet
def main():
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
data_transform = transforms.Compose(
[transforms.Resize((224, 224)),
transforms.ToTensor(),
transforms.Normalize((0.5, 0.5, 0.5),(0.5, 0.5, 0.5))]
)
img_path = "D:/20_Models/01_AlexNet_pytorch/image_predict/tulip.jpg"
assert os.path.exists(img_path), "file: '{}' does not exist.".format(img_path)
img = Image.open(img_path)
plt.imshow(img)
img = data_transform(img)
img = torch.unsqueeze(img, dim=0)
json_path = './class_indices.json'
assert os.path.exists(json_path), "file: '{}' does not exist.".format(json_path)
with open(json_path,"r") as f:
class_indict = json.load(f)
model = AlexNet(num_classes=5).to(device)
weights_path = "./AlexNet.pth"
assert os.path.exists(weights_path), "file: '{}' does not exist.".format(weights_path)
model.load_state_dict(torch.load(weights_path))
model.eval()
with torch.no_grad():
output = torch.squeeze(model(img.to(device))).cpu()
predict = torch.softmax(output, dim=0)
predict_cla = torch.argmax(predict).numpy()
print_res = "class: {} prob: {:.3f}".format(class_indict[str(predict_cla)],
predict[predict_cla].numpy())
plt.title(print_res)
for i in range(len(predict)):
print("class: {:10} prob: {:.3}".format(class_indict[str(i)],predict[i].numpy()))
plt.show()
if __name__ == '__main__':
main()4.predict.py
import os
from shutil import copy, rmtree
import random
def mk_file(file_path: str):
if os.path.exists(file_path):
rmtree(file_path)
os.makedirs(file_path)
def main():
random.seed(0)
split_rate = 0.1
#cwd = os.getcwd()
#data_root = os.path.join(cwd, "flower_data")
data_root = 'D:/100_DataSets/03_flower_data'
origin_flower_path = os.path.join(data_root, "flower_photos")
assert os.path.exists(origin_flower_path), "path '{}' does not exist".format(origin_flower_path)
flower_class = [cla for cla in os.listdir(origin_flower_path) if os.path.isdir(os.path.join(origin_flower_path, cla))]
train_root = os.path.join(data_root,"train")
mk_file(train_root)
for cla in flower_class:
mk_file(os.path.join(train_root, cla))
val_root = os.path.join(data_root, "val")
mk_file(val_root)
for cla in flower_class:
mk_file(os.path.join(val_root,cla))
for cla in flower_class:
cla_path = os.path.join(origin_flower_path,cla)
images = os.listdir(cla_path)
num = len(images)
eval_index = random.sample(images, k=int(num*split_rate))
for index, image in enumerate(images):
if image in eval_index:
image_path = os.path.join(cla_path, image)
new_path = os.path.join(val_root, cla)
copy(image_path, new_path)
else:
image_path = os.path.join(cla_path, image)
new_path = os.path.join(train_root, cla)
copy(image_path, new_path)
print("\r[{}] processing [{} / {}]".format(cla, index+1, num), end="")
print()
print("processing done!")
if __name__ == "__main__":
main()
总结
提示：这里对文章进行总结：
每天一个网络，网络的学习往往具有连贯性，新的网络往往是基于旧的网络进行不断改进。
6588
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
