建设英语网站目的,三里河网站建设公司,商贸公司寮步网站建设,python做直播网站#x1f368; 本文为#x1f517;365天深度学习训练营 中的学习记录博客 #x1f356; 原作者#xff1a;K同学啊|接辅导、项目定制 1 理论基础 Inception v3论文 Inception v3由谷歌研究员Christian Szegedy等人在2015年的论文《Rethinking the Inception Architecture f… 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|接辅导、项目定制 1 理论基础 Inception v3论文 Inception v3由谷歌研究员Christian Szegedy等人在2015年的论文《Rethinking the Inception Architecture for Computer Vision》中提出。Inception v3是Inception网络系列的第三个版本它在ImageNet图像识别竞赛中取得了优异成绩尤其是在大规模图像识别任务中表现出色。 Inception v3的主要特点如下 1、更深的网络结构Inception v3比之前的Inception网络结构更深包含了48层卷积层。这使得网络可以提取更多层次的特征从而在图像识别任务上取得更好的效果。 2、使用Factorized ConvolutionsInception v3采用了Factorized Convolutions分解卷积将较大的卷积核分解为多个较小卷积核。这种方法可以降低网络的参数数量减少计算复杂度同时保持良好的性能。 3、使用Batch NormalizationInception v3在每个卷积层之后都添加了Batch NormalizationBN这有助于网络的收敛和泛化能力。BN可以减少Internal Covariate Shift内部协变量偏移现象加快训练速度同时提高模型的鲁棒性。 4、辅助分类器Inception v3引入了辅助分类器可以在网络训练过程中提供额外的梯度信息帮助网络更好的学习特征。辅助分类器位于网络的某个中间层其输出会与主分类器的输出进行加权融合从而得到最终的预测结果。 5、基于RMSProp的优化器Inception v3使用了RMSProp优化器进行训练。相比于传统的随机梯度下降SGD方法RMSProp可以自适应地调整学习率使得训练过程更加稳定收敛速度更快。 Inception v3在图像分类、物体检测和图像分割等计算机视觉任务中均取得了显著的效果。然而由于其较大的网络结构和计算复杂度Inception v3在实际应用中可能需要较高的硬件要求。 相对于Inception v1的Inception Module结构Inception v3中做出了如下改动
1将5x5的卷积分解为两个3x3的卷积运算已提升计算速度。尽管这有点违反直觉但一个5x5的卷积在计算成本上是一个3x3卷积的2.78倍。所以叠加两个3x3卷积实际上在性能上会有所提升如下图所示 Inception v1 Inception v3 2此外作者将nxn的卷积核尺寸分解为1xn和nx1两个卷积例如 一个3x3的卷积等价于首先执行一个1x3的卷积再执行一个3x1的卷积。他们还发现这种方法在成本上要比单个3x3的卷积降低33%这一结构如下图所示 此处如果n3则与上一张图像一致。最左侧的5x5卷积可被表示为两个3x3卷积太慢又可以表示为1x3和3x1卷积。 模块中的滤波器组被扩展即变得更宽而不是更深以解决表征性瓶颈。如果该模块没有被拓展宽度而是变得更深那么维度会过多减少造成信息损失。如下图所示 最后实现的inception v3网络是如下所示 2 代码实现
2.1 开发环境 电脑系统ubuntu16.04 编译器Jupter Lab 语言环境Python 3.7 深度学习环境Pytorch 2.2 前期准备
2.2.1 设置GPU
import torch
import torch.nn as nn
import torchvision.transforms as transforms
import torchvision
from torchvision import transforms, datasets
import os, PIL, pathlib, warningswarnings.filterwarnings(ignore)
device torch.device(cuda if torch.cuda.is_available() else cpu)print(device)
2.2.2 导入数据
import os,PIL,random,pathlib
data_dir ../data/4-data/
data_dir pathlib.Path(data_dir)
data_dirdata_paths list(data_dir.glob(*))
classNames [str(path).split(\\)[-1] for path in data_paths]
print(classNames:, classNames , \n)total_dir ../data/4-data/
train_transforms transforms.Compose([transforms.Resize([299, 299]), # resize输入图片transforms.ToTensor(), # 将PIL Image或numpy.ndarray转换成tensortransforms.Normalize(mean[0.485, 0.456, 0.406],std[0.229, 0.224, 0.225]) # 从数据集中随机抽样计算得到
])total_data datasets.ImageFolder(total_dir, transformtrain_transforms)
print(total_data, \n)print(total_data.class_to_idx) 输出结果如下 2.2.3 划分数据集
train_size int(0.8 * len(total_data))
test_size len(total_data) - train_size
train_dataset, test_dataset torch.utils.data.random_split(total_data, [train_size, test_size])
print(train_dataset, test_dataset)batch_size 4
train_dl torch.utils.data.DataLoader(train_dataset, batch_sizebatch_size,shuffleTrue,num_workers1,pin_memoryFalse)
test_dl torch.utils.data.DataLoader(test_dataset, batch_sizebatch_size,shuffleTrue,num_workers1,pin_memoryFalse)for X, y in test_dl:print(Shape of X [N, C, H, W]:, X.shape)print(Shape of y:, y.shape, y.dtype)break 输出结果如下所示 2.3 Inception v3的实现
2.3.1 Inception-A import torch
import torch.nn as nn
import torch.nn.functional as Fclass BasicConv2d(nn.Module):def __init__(self, in_channels, out_channels, **kwargs):super(BasicConv2d, self).__init__()self.conv nn.Conv2d(in_channels, out_channels, biasFalse, **kwargs)self.bn nn.BatchNorm2d(out_channels, eps0.001)def forward(self, x):x self.conv(x)x self.bn(x)return F.relu(x, inplaceTrue)class InceptionA(nn.Module):def __init__(self, in_channels, pool_features):super(InceptionA, self).__init__()# 1x1 conv branchself.branch1x1 BasicConv2d(in_channels, 64, kernel_size1) #1self.branch5x5_1 BasicConv2d(in_channels, 48, kernel_size1)self.branch5x5_2 BasicConv2d(48, 64, kernel_size5, padding2)self.branch3x3dbl_1 BasicConv2d(in_channels, 64, kernel_size1)self.branch3x3dbl_2 BasicConv2d(64, 96, kernel_size3, padding1)self.branch3x3dbl_3 BasicConv2d(96, 96, kernel_size3, padding1)self.branch_pool BasicConv2d(in_channels, pool_features, kernel_size1)def forward(self, x):branch1x1 self.branch1x1(x)branch5x5 self.branch5x5_1(x)branch5x5 self.branch5x5_2(branch5x5)branch3x3dbl self.branch3x3dbl_1(x)branch3x3dbl self.branch3x3dbl_2(branch3x3dbl)branch3x3dbl self.branch3x3dbl_3(branch3x3dbl)branch_pool F.avg_pool2d(x, kernel_size3, stride1, padding1)branch_pool self.branch_pool(branch_pool)outputs [branch1x1, branch5x5, branch3x3dbl, branch_pool]return torch.cat(outputs, 1)
2.3.2 Inception-B class InceptionB(nn.Module):def __init__(self, in_channels, channels_7x7):super(InceptionB, self).__init__()# 1x1 conv branchself.branch1x1 BasicConv2d(in_channels, 192, kernel_size1)c7 channels_7x7self.branch7x7_1 BasicConv2d(in_channels, c7, kernel_size1)self.branch7x7_2 BasicConv2d(c7, c7, kernel_size(1,7), padding(0,3))self.branch7x7_3 BasicConv2d(c7, 192, kernel_size(7,1), padding(3,0))self.branch7x7dbl_1 BasicConv2d(in_channels, c7, kernel_size1)self.branch7x7dbl_2 BasicConv2d(c7, c7, kernel_size(7,1), padding(3,0))self.branch7x7dbl_3 BasicConv2d(c7, c7, kernel_size(1,7), padding(0,3))self.branch7x7dbl_4 BasicConv2d(c7, c7, kernel_size(7,1), padding(3,0))self.branch7x7dbl_5 BasicConv2d(c7, 192, kernel_size(1,7), padding(0,3))self.branch_pool BasicConv2d(in_channels, 192, kernel_size1)def forward(self, x):branch1x1 self.branch1x1(x)branch7x7 self.branch7x7_1(x)branch7x7 self.branch7x7_2(branch7x7)branch7x7 self.branch7x7_3(branch7x7)branch7x7dbl self.branch7x7dbl_1(x)branch7x7dbl self.branch7x7dbl_2(branch7x7dbl)branch7x7dbl self.branch7x7dbl_3(branch7x7dbl)branch7x7dbl self.branch7x7dbl_4(branch7x7dbl)branch7x7dbl self.branch7x7dbl_5(branch7x7dbl)branch_pool F.avg_pool2d(x, kernel_size3, stride1, padding1)branch_pool self.branch_pool(branch_pool)outputs [branch1x1, branch7x7, branch7x7dbl, branch_pool]return torch.cat(outputs, 1)
2.3.3 Inception-C class InceptionC(nn.Module):def __init__(self, in_channels):super(InceptionC, self).__init__()self.branch1x1 BasicConv2d(in_channels, 320, kernel_size1) #1self.branch3x3_1 BasicConv2d(in_channels, 384, kernel_size1)self.branch3x3_2a BasicConv2d(384, 384, kernel_size(1,3), padding(0,1))self.branch3x3_2b BasicConv2d(384, 384, kernel_size(3,1), padding(1,0))self.branch3x3dbl_1 BasicConv2d(in_channels, 448, kernel_size1)self.branch3x3dbl_2 BasicConv2d(448, 384, kernel_size3, padding1)self.branch3x3dbl_3a BasicConv2d(384, 384, kernel_size(1,3), padding(0,1))self.branch3x3dbl_3b BasicConv2d(384, 384, kernel_size(3,1), padding(1,0))self.branch_pool BasicConv2d(in_channels, 192, kernel_size1)def forward(self, x):branch1x1 self.branch1x1(x)branch3x3 self.branch3x3_1(x)branch3x3 [self.branch3x3_2a(branch3x3),self.branch3x3_2b(branch3x3),]branch3x3 torch.cat(branch3x3, 1)branch3x3dbl self.branch3x3dbl_1(x)branch3x3dbl self.branch3x3dbl_2(branch3x3dbl)branch3x3dbl [self.branch3x3dbl_3a(branch3x3dbl),self.branch3x3dbl_3b(branch3x3dbl),]branch3x3dbl torch.cat(branch3x3dbl, 1)branch_pool F.avg_pool2d(x, kernel_size3, stride1, padding1)branch_pool self.branch_pool(branch_pool)outputs [branch1x1, branch3x3, branch3x3dbl, branch_pool]return torch.cat(outputs, 1)
2.3.4 Reduction-A class ReductionA(nn.Module):def __init__(self, in_channels):super(ReductionA, self).__init__()self.branch3x3 BasicConv2d(in_channels, 384, kernel_size3, stride2) self.branch3x3dbl_1 BasicConv2d(in_channels, 64, kernel_size1)self.branch3x3dbl_2 BasicConv2d(64, 96, kernel_size3, padding1)self.branch3x3dbl_3 BasicConv2d(96, 96, kernel_size3, stride2)def forward(self, x):branch3x3 self.branch3x3(x)branch3x3dbl self.branch3x3dbl_1(x)branch3x3dbl self.branch3x3dbl_2(branch3x3dbl)branch3x3dbl self.branch3x3dbl_3(branch3x3dbl)branch_pool F.max_pool2d(x, kernel_size3, stride2)outputs [branch3x3, branch3x3dbl, branch_pool]return torch.cat(outputs, 1)
2.3.5 Reduction-B class ReductionB(nn.Module):def __init__(self, in_channels):super(ReductionB, self).__init__()self.branch3x3_1 BasicConv2d(in_channels, 192, kernel_size1) self.branch3x3_2 BasicConv2d(192, 320, kernel_size3, stride2) self.branch7x7x3_1 BasicConv2d(in_channels, 192, kernel_size1)self.branch7x7x3_2 BasicConv2d(192, 192, kernel_size(1,7), padding(0,3))self.branch7x7x3_3 BasicConv2d(192, 192, kernel_size(7,1), padding(3,0))self.branch7x7x3_4 BasicConv2d(192, 192, kernel_size3, stride2)def forward(self, x):branch3x3 self.branch3x3_1(x)branch3x3 self.branch3x3_2(branch3x3)branch7x7x3 self.branch7x7x3_1(x)branch7x7x3 self.branch7x7x3_2(branch7x7x3)branch7x7x3 self.branch7x7x3_3(branch7x7x3)branch7x7x3 self.branch7x7x3_4(branch7x7x3)branch_pool F.max_pool2d(x, kernel_size3, stride2)outputs [branch3x3, branch7x7x3, branch_pool]return torch.cat(outputs, 1)
2.3.6 辅助分支 class InceptionAux(nn.Module):def __init__(self, in_channels, num_classes):super(InceptionAux, self).__init__()self.conv0 BasicConv2d(in_channels, 128, kernel_size1)self.conv1 BasicConv2d(128, 768, kernel_size5)self.conv1.stddev 0.01self.fc nn.Linear(768, num_classes)self.fc.stddev 0.001def forward(self, x):# 17x17x768x F.avg_pool2d(x, kernel_size5, stride3)# 5x5x768x self.conv0(x)# 5x5x128x self.conv1(x)# 1x1x128x x.view(x.size(0), -1)# 768x self.fc(x)# num_classesreturn x
2.3.7 模型搭建 class InceptionV3(nn.Module):def __init__(self, num_classes1000, aux_logitsFalse, transform_inputFalse):super(InceptionV3, self).__init__()self.aux_logits aux_logitsself.transform_input transform_inputself.conv2d_1a_3x3 BasicConv2d(3, 32, kernel_size3, stride2)self.conv2d_2a_3x3 BasicConv2d(32, 32, kernel_size3)self.conv2d_2b_3x3 BasicConv2d(32, 64, kernel_size3, padding1)self.conv2d_3b_1x1 BasicConv2d(64, 80, kernel_size1)self.conv2d_4a_3x3 BasicConv2d(80, 192, kernel_size3)self.Mixed_5b InceptionA(192, pool_features32)self.Mixed_5c InceptionA(256, pool_features64)self.Mixed_5d InceptionA(288, pool_features64)self.Mixed_6a ReductionA(288)self.Mixed_6b InceptionB(768, channels_7x7128)self.Mixed_6c InceptionB(768, channels_7x7160)self.Mixed_6d InceptionB(768, channels_7x7160)self.Mixed_6e InceptionB(768, channels_7x7192)if aux_logits:self.AuxLogits InceptionAux(768, num_classes)self.Mixed_7a ReductionB(768)self.Mixed_7b InceptionC(1280)self.Mixed_7c InceptionC(2048)self.fc nn.Linear(2048, num_classes)def forward(self, x):if self.transform_input:x x.clone()x[:, 0] x[:, 0] * (0.229 / 0.5) (0.485 - 0.5) / 0.5x[:, 1] x[:, 1] * (0.224 / 0.5) (0.456 - 0.5) / 0.5x[:, 2] x[:, 2] * (0.225 / 0.5) (0.406 - 0.5) / 0.5# 299*299*3x self.conv2d_1a_3x3(x)# 149*149*32x self.conv2d_2a_3x3(x)# 147*147*32x self.conv2d_2b_3x3(x)# 147*147*64x F.max_pool2d(x, kernel_size3, stride2)# 73*73*64x self.conv2d_3b_1x1(x)# 73*73*80x self.conv2d_4a_3x3(x)# 71*71*192x F.max_pool2d(x, kernel_size3, stride2)# 35*35*192x self.Mixed_5b(x)# 35*35*256x self.Mixed_5c(x)# 35*35*288x self.Mixed_5d(x)# 35*35*288x self.Mixed_6a(x)# 17*17*768x self.Mixed_6b(x)# 17*17*768x self.Mixed_6c(x)# 17*17*768x self.Mixed_6d(x)# 17*17*768x self.Mixed_6e(x)# 17*17*768if self.training and self.aux_logits:aux self.AuxLogits(x)# 17*17*768x self.Mixed_7a(x)# 8*8*1280x self.Mixed_7b(x)# 8*8*2048x self.Mixed_7c(x)# 8*8*2048x F.avg_pool2d(x, kernel_size8)# 1*1*2048x F.dropout(x, trainingself.training)# 1*1*2048x x.view(x.size(0), -1)# 2048x self.fc(x)# num_classesif self.training and self.aux_logits:return x, auxreturn x
2.3.8 查看模型详情
# 统计模型参数量以及其他指标
import torchsummary# 调用并将模型转移到GPU中
model InceptionV3().to(device)# 显示网络结构
torchsummary.summary(model, (3, 299, 299))
print(model) 输出结果如下由于内容较多只展示前后部分内容 中间内容省略
中间内容省略 2.4 训练模型
2.4.1 编写训练函数
# 训练循环
def train(dataloader, model, loss_fn, optimizer):size len(dataloader.dataset) # 训练集的大小num_batches len(dataloader) # 批次数目, (size/batch_size向上取整)train_loss, train_acc 0, 0 # 初始化训练损失和正确率for X, y in dataloader: # 获取图片及其标签X, y X.to(device), y.to(device)# 计算预测误差pred model(X) # 网络输出loss loss_fn(pred, y) # 计算网络输出pred和真实值y之间的差距y为真实值计算二者差值即为损失# 反向传播optimizer.zero_grad() # grad属性归零loss.backward() # 反向传播optimizer.step() # 每一步自动更新# 记录acc与losstrain_acc (pred.argmax(1) y).type(torch.float).sum().item()train_loss loss.item()train_acc / sizetrain_loss / num_batchesreturn train_acc, train_loss
2.4.2 编写测试函数
def test(dataloader, model, loss_fn):size len(dataloader.dataset) # 训练集的大小num_batches len(dataloader) # 批次数目, (size/batch_size向上取整)test_loss, test_acc 0, 0 # 初始化测试损失和正确率# 当不进行训练时停止梯度更新节省计算内存消耗# with torch.no_grad():for imgs, target in dataloader: # 获取图片及其标签with torch.no_grad():imgs, target imgs.to(device), target.to(device)# 计算误差tartget_pred model(imgs) # 网络输出loss loss_fn(tartget_pred, target) # 计算网络输出和真实值之间的差距targets为真实值计算二者差值即为损失# 记录acc与losstest_loss loss.item()test_acc (tartget_pred.argmax(1) target).type(torch.float).sum().item()test_acc / sizetest_loss / num_batchesreturn test_acc, test_loss
2.4.3 正式训练
import copyoptimizer torch.optim.Adam(model.parameters(), lr 1e-4)
loss_fn nn.CrossEntropyLoss() #创建损失函数epochs 40train_loss []
train_acc []
test_loss []
test_acc []best_acc 0 #设置一个最佳准确率作为最佳模型的判别指标if hasattr(torch.cuda, empty_cache):torch.cuda.empty_cache()for epoch in range(epochs):model.train()epoch_train_acc, epoch_train_loss train(train_dl, model, loss_fn, optimizer)#scheduler.step() #更新学习率调用官方动态学习率接口时使用model.eval()epoch_test_acc, epoch_test_loss test(test_dl, model, loss_fn)#保存最佳模型到best_modelif epoch_test_acc best_acc:best_acc epoch_test_accbest_model copy.deepcopy(model)train_acc.append(epoch_train_acc)train_loss.append(epoch_train_loss)test_acc.append(epoch_test_acc)test_loss.append(epoch_test_loss)#获取当前的学习率lr optimizer.state_dict()[param_groups][0][lr]template (Epoch: {:2d}. Train_acc: {:.1f}%, Train_loss: {:.3f}, Test_acc:{:.1f}%, Test_loss:{:.3f}, Lr: {:.2E})print(template.format(epoch1, epoch_train_acc*100, epoch_train_loss, epoch_test_acc*100, epoch_test_loss, lr))PATH ./J7_best_model.pth
torch.save(model.state_dict(), PATH)print(Done) 输出结果如下 2.5 结果可视化
import matplotlib.pyplot as plt
#隐藏警告
import warnings
warnings.filterwarnings(ignore) #忽略警告信息
plt.rcParams[font.sans-serif] [SimHei] # 用来正常显示中文标签
plt.rcParams[axes.unicode_minus] False # 用来正常显示负号
plt.rcParams[figure.dpi] 100 #分辨率epochs_range range(epochs)plt.figure(figsize(12, 3))
plt.subplot(1, 2, 1)plt.plot(epochs_range, train_acc, labelTraining Accuracy)
plt.plot(epochs_range, test_acc, labelTest Accuracy)
plt.legend(loclower right)
plt.title(Training and Validation Accuracy)plt.subplot(1, 2, 2)
plt.plot(epochs_range, train_loss, labelTraining Loss)
plt.plot(epochs_range, test_loss, labelTest Loss)
plt.legend(locupper right)
plt.title(Training and Validation Loss)
plt.show() 输出结果如下 3 总结 总体而言Inception v3主要提出了分解卷积将大卷积因式分解成小卷积核非对称卷积体现在数学上就是矩阵的分解即一个大矩阵可以分解成多个小矩阵相乘。 由于简单的增大Inception网络的规模是不可行的这样会导致计算效率变低Inception v3在v2的基础上去除低层辅助分类器高层辅助分类器加入BN层作为正则化器。将较大的卷积核分解为串联的小卷积核能够进行维度缩减同时小卷积核在多次串联后并不会缩小感受野进而提取的特征所代表的感受野不受影响。而并联卷积核池化避免了表征瓶颈。这样同时增加宽度核深度平衡了网络的宽度和深度因此提高了网络的质量优化了网络的特征提取效果。
