🍨 本文为🔗365天深度学习训练营 中的学习记录博客
🍦 参考文章：Pytorch实战 | 第P5周：运动鞋识别
🍖 原作者：K同学啊|接辅导、项目定制
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") #忽略警告信息device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
device
device(type='cuda')
import os,PIL,random,pathlibdata_dir = './48-data/'
data_dir = pathlib.Path(data_dir)data_paths = list(data_dir.glob('*'))
classeNames = [str(path).split("\\")[1] for path in data_paths]#os.path.split()通过一对链表的头和尾来划分路径名。链表的tail是是最后的路径名元素。head则是它前面的元素。
#例如path name = '/home/User/Desktop/file.txt' 在上面的这个例子中，路径名字file.txt称之为tail 路径‘/home/User/Desktop/’ 称之为head。
# tail部分永远不会包含斜杠符号。如果这个路径名字以斜杠结束，那么tail就是为空。
# path head tail
#'/home/user/Desktop/file.txt' '/home/user/Desktop/' 'file.txt'
#'/home/user/Desktop/' '/home/user/Desktop/' {empty}
#'file.txt' {empty} 'file.txt'
#可以查阅相关资料深入理解
classeNames
['Angelina Jolie','Brad Pitt','Denzel Washington','Hugh Jackman','Jennifer Lawrence','Johnny Depp','Kate Winslet','Leonardo DiCaprio','Megan Fox','Natalie Portman','Nicole Kidman','Robert Downey Jr','Sandra Bullock','Scarlett Johansson','Tom Cruise','Tom Hanks','Will Smith']
# 关于transforms.Compose的更多介绍可以参考：https://blog.csdn.net/qq_38251616/article/details/124878863
train_transforms = transforms.Compose([transforms.Resize([224, 224]), # 将输入图片resize成统一尺寸# transforms.RandomHorizontalFlip(), # 随机水平翻转transforms.ToTensor(), # 将PIL Image或numpy.ndarray转换为tensor，并归一化到[0,1]之间#ToTensor：convert a PIL image to tensor (H*W*C) in range [0,255] to a torch.Tensor(C*H*W) in the range [0.0,1.0]transforms.Normalize( # 标准化处理-->转换为标准正太分布（高斯分布），使模型更容易收敛mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) # 其中 mean=[0.485,0.456,0.406]与std=[0.229,0.224,0.225] 从数据集中随机抽样计算得到的。 image=(image-mean)/std
])test_transform = transforms.Compose([transforms.Resize([224, 224]), # 将输入图片resize成统一尺寸transforms.ToTensor(), # 将PIL Image或numpy.ndarray转换为tensor，并归一化到[0,1]之间transforms.Normalize( # 标准化处理-->转换为标准正太分布（高斯分布），使模型更容易收敛mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) # 其中 mean=[0.485,0.456,0.406]与std=[0.229,0.224,0.225] 从数据集中随机抽样计算得到的。
])total_data = datasets.ImageFolder("./48-data/",transform=train_transforms)
total_data
Dataset ImageFolderNumber of datapoints: 1800Root location: ./48-data/StandardTransform
Transform: Compose(Resize(size=[224, 224], interpolation=bilinear, max_size=None, antialias=None)ToTensor()Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]))
total_data.class_to_idx
{'Angelina Jolie': 0,'Brad Pitt': 1,'Denzel Washington': 2,'Hugh Jackman': 3,'Jennifer Lawrence': 4,'Johnny Depp': 5,'Kate Winslet': 6,'Leonardo DiCaprio': 7,'Megan Fox': 8,'Natalie Portman': 9,'Nicole Kidman': 10,'Robert Downey Jr': 11,'Sandra Bullock': 12,'Scarlett Johansson': 13,'Tom Cruise': 14,'Tom Hanks': 15,'Will Smith': 16}
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])
train_dataset, test_dataset
(<torch.utils.data.dataset.Subset at 0x1f4492e7b20>,<torch.utils.data.dataset.Subset at 0x1f4492e7bb0>)
batch_size = 32train_dl = torch.utils.data.DataLoader(train_dataset,batch_size=batch_size,shuffle=True,num_workers=1)
test_dl = torch.utils.data.DataLoader(test_dataset,batch_size=batch_size,shuffle=True,num_workers=1)
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
Shape of X [N, C, H, W]: torch.Size([32, 3, 224, 224])
Shape of y: torch.Size([32]) torch.int64
from torchvision.models import vgg16device = "cuda" if torch.cuda.is_available() else "cpu"
print("Using {} device".format(device))# 加载预训练模型，并且对模型进行微调
model = vgg16(pretrained = True).to(device) # 加载预训练的vgg16模型for param in model.parameters():param.requires_grad = False # 冻结模型的参数，这样子在训练的时候只训练最后一层的参数# 修改classifier模块的第6层（即：(6): Linear(in_features=4096, out_features=2, bias=True)）
# 注意查看我们下方打印出来的模型
#model.classifier._modules['6'] = nn.Linear(4096,len(classeNames)) # 修改vgg16模型中最后一层全连接层，输出目标类别个数
model.classifier = nn.Sequential(nn.Linear(512 * 7 * 7, 1024),nn.BatchNorm1d(1024),nn.Dropout(0.4),nn.Linear(1024, 128),nn.BatchNorm1d(128),nn.Dropout(0.4),nn.Linear(128, 64),nn.Linear(64, len(classeNames)),nn.Softmax()
)model.to(device)
model
Using cuda deviceVGG((features): Sequential((0): Conv2d(3, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(1): ReLU(inplace=True)(2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(3): ReLU(inplace=True)(4): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)(5): Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(6): ReLU(inplace=True)(7): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(8): ReLU(inplace=True)(9): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)(10): Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(11): ReLU(inplace=True)(12): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(13): ReLU(inplace=True)(14): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(15): ReLU(inplace=True)(16): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)(17): Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(18): ReLU(inplace=True)(19): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(20): ReLU(inplace=True)(21): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(22): ReLU(inplace=True)(23): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)(24): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(25): ReLU(inplace=True)(26): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(27): ReLU(inplace=True)(28): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(29): ReLU(inplace=True)(30): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False))(avgpool): AdaptiveAvgPool2d(output_size=(7, 7))(classifier): Sequential((0): Linear(in_features=25088, out_features=1024, bias=True)(1): BatchNorm1d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(2): Dropout(p=0.4, inplace=False)(3): Linear(in_features=1024, out_features=128, bias=True)(4): BatchNorm1d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(5): Dropout(p=0.4, inplace=False)(6): Linear(in_features=128, out_features=64, bias=True)(7): Linear(in_features=64, out_features=17, bias=True)(8): Softmax(dim=None))
)
VGG16手写版本
class My_VGG16(nn.Module):
def init(self,num_classes=5,init_weight=True):#对于原来的输出为类默认num_class=
super(My_VGG16, self).init()
# 特征提取层
self.features = nn.Sequential(
nn.Conv2d(in_channels=3,out_channels=64,kernel_size=3,stride=1,padding=1),
nn.Conv2d(in_channels=64,out_channels=64,kernel_size=3,stride=1,padding=1),
nn.MaxPool2d(kernel_size=2,stride=2),
nn.Conv2d(in_channels=64, out_channels=128, kernel_size=3, stride=1, padding=1),
nn.Conv2d(in_channels=128, out_channels=128, kernel_size=3, stride=1, padding=1),
nn.MaxPool2d(kernel_size=2, stride=2),
nn.Conv2d(in_channels=128, out_channels=256, kernel_size=3, stride=1, padding=1),
nn.Conv2d(in_channels=256, out_channels=256, kernel_size=3, stride=1, padding=1),
nn.Conv2d(in_channels=256, out_channels=256, kernel_size=3, stride=1, padding=1),
nn.MaxPool2d(kernel_size=2,stride=2),
nn.Conv2d(in_channels=256, out_channels=512, kernel_size=3, stride=1, padding=1),
nn.Conv2d(in_channels=512, out_channels=512, kernel_size=3, stride=1, padding=1),
nn.Conv2d(in_channels=512, out_channels=512, kernel_size=3, stride=1, padding=1),
nn.MaxPool2d(kernel_size=2, stride=2),
nn.Conv2d(in_channels=512, out_channels=512, kernel_size=3, stride=1, padding=1),
nn.Conv2d(in_channels=512, out_channels=512, kernel_size=3, stride=1, padding=1),
nn.Conv2d(in_channels=512, out_channels=512, kernel_size=3, stride=1, padding=1),
nn.MaxPool2d(kernel_size=2, stride=2),
)
# 分类层
self.classifier = nn.Sequential(
nn.Linear(in_features=77512,out_features=4096),
nn.ReLU(),
nn.Dropout(0.5),
nn.Linear(in_features=4096,out_features=4096),
nn.ReLU(),
nn.Dropout(0.5),
nn.Linear(in_features=4096,out_features=5)
)
# 参数初始化if init_weight: # 如果进行参数初始化for m in self.modules(): # 对于模型的每一层if isinstance(m, nn.Conv2d): # 如果是卷积层# 使用kaiming初始化nn.init.kaiming_normal_(m.weight, mode="fan_out", nonlinearity="relu")# 如果bias不为空，固定为0if m.bias is not None:nn.init.constant_(m.bias, 0)elif isinstance(m, nn.Linear):# 如果是线性层# 正态初始化nn.init.normal_(m.weight, 0, 0.01)# bias则固定为0nn.init.constant_(m.bias, 0)def forward(self,x):x = self.features(x)x = torch.flatten(x,1)result = self.classifier(x)return result
#tip：对于vgg16模型参数量已经很大了很可能出现显存不足的情况，可以减少batchsize降低模型复杂度等方法解决，但是也可能是算力确实不足了
# 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) # 计算网络输出和真实值之间的差距，targets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
def test (dataloader, model, loss_fn):size = len(dataloader.dataset) # 测试集的大小num_batches = len(dataloader) # 批次数目, (size/batch_size，向上取整)test_loss, test_acc = 0, 0# 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:imgs, target = imgs.to(device), target.to(device)# 计算losstarget_pred = model(imgs)loss = loss_fn(target_pred, target)test_loss += loss.item()test_acc += (target_pred.argmax(1) == target).type(torch.float).sum().item()test_acc /= sizetest_loss /= num_batchesreturn test_acc, test_loss
# def adjust_learning_rate(optimizer, epoch, start_lr):
# # 每 2 个epoch衰减到原来的 0.92
# lr = start_lr * (0.92 ** (epoch // 2))
# for param_group in optimizer.param_groups:
# param_group['lr'] = lrlearn_rate = 1e-4 # 初始学习率
# optimizer = torch.optim.SGD(model.parameters(), lr=learn_rate)
lambda1 = lambda epoch: 0.92 ** (epoch // 2)
optimizer = torch.optim.Adam(model.parameters(), lr=learn_rate)
scheduler = torch.optim.lr_scheduler.LambdaLR(optimizer, lr_lambda=lambda1) #选定调整方法
import copyloss_fn = nn.CrossEntropyLoss() # 创建损失函数
epochs = 30train_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for epoch in range(epochs):# 更新学习率（使用自定义学习率时使用）# adjust_learning_rate(optimizer, epoch, learn_rate)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(epoch+1, epoch_train_acc*100, epoch_train_loss, epoch_test_acc*100, epoch_test_loss, lr))# 保存最佳模型到文件中
PATH = './best_model.pth' # 保存的参数文件名
torch.save(model.state_dict(), PATH)print('Done')
Epoch: 1, Train_acc:100.0%, Train_loss:1.938, Test_acc:61.1%, Test_loss:2.437, Lr:2.86E-05
Epoch: 2, Train_acc:100.0%, Train_loss:1.936, Test_acc:61.9%, Test_loss:2.440, Lr:2.63E-05
Epoch: 3, Train_acc:100.0%, Train_loss:1.936, Test_acc:60.8%, Test_loss:2.448, Lr:2.63E-05
Epoch: 4, Train_acc:100.0%, Train_loss:1.936, Test_acc:61.7%, Test_loss:2.455, Lr:2.42E-05
Epoch: 5, Train_acc:100.0%, Train_loss:1.937, Test_acc:62.2%, Test_loss:2.450, Lr:2.42E-05
Epoch: 6, Train_acc:100.0%, Train_loss:1.936, Test_acc:61.7%, Test_loss:2.447, Lr:2.23E-05
Epoch: 7, Train_acc:100.0%, Train_loss:1.936, Test_acc:61.7%, Test_loss:2.434, Lr:2.23E-05
Epoch: 8, Train_acc:100.0%, Train_loss:1.935, Test_acc:60.3%, Test_loss:2.445, Lr:2.05E-05
Epoch: 9, Train_acc:100.0%, Train_loss:1.935, Test_acc:59.7%, Test_loss:2.452, Lr:2.05E-05
Epoch:10, Train_acc:100.0%, Train_loss:1.935, Test_acc:61.4%, Test_loss:2.444, Lr:1.89E-05
Epoch:11, Train_acc:100.0%, Train_loss:1.935, Test_acc:60.8%, Test_loss:2.447, Lr:1.89E-05
Epoch:12, Train_acc:100.0%, Train_loss:1.935, Test_acc:61.4%, Test_loss:2.428, Lr:1.74E-05
Epoch:13, Train_acc:100.0%, Train_loss:1.935, Test_acc:60.8%, Test_loss:2.452, Lr:1.74E-05
Epoch:14, Train_acc:100.0%, Train_loss:1.935, Test_acc:61.1%, Test_loss:2.432, Lr:1.60E-05
Epoch:15, Train_acc:100.0%, Train_loss:1.935, Test_acc:61.4%, Test_loss:2.438, Lr:1.60E-05
Epoch:16, Train_acc:100.0%, Train_loss:1.934, Test_acc:61.7%, Test_loss:2.441, Lr:1.47E-05
Epoch:17, Train_acc:100.0%, Train_loss:1.934, Test_acc:60.6%, Test_loss:2.433, Lr:1.47E-05
Epoch:18, Train_acc:100.0%, Train_loss:1.934, Test_acc:61.4%, Test_loss:2.433, Lr:1.35E-05
Epoch:19, Train_acc:100.0%, Train_loss:1.934, Test_acc:61.1%, Test_loss:2.412, Lr:1.35E-05
Epoch:20, Train_acc:100.0%, Train_loss:1.934, Test_acc:60.3%, Test_loss:2.424, Lr:1.24E-05
Epoch:21, Train_acc:100.0%, Train_loss:1.934, Test_acc:60.6%, Test_loss:2.430, Lr:1.24E-05
Epoch:22, Train_acc:100.0%, Train_loss:1.934, Test_acc:60.8%, Test_loss:2.426, Lr:1.14E-05
Epoch:23, Train_acc:100.0%, Train_loss:1.934, Test_acc:61.4%, Test_loss:2.437, Lr:1.14E-05
Epoch:24, Train_acc:100.0%, Train_loss:1.934, Test_acc:61.4%, Test_loss:2.429, Lr:1.05E-05
Epoch:25, Train_acc:100.0%, Train_loss:1.933, Test_acc:61.4%, Test_loss:2.429, Lr:1.05E-05
Epoch:26, Train_acc:100.0%, Train_loss:1.933, Test_acc:61.7%, Test_loss:2.434, Lr:9.68E-06
Epoch:27, Train_acc:100.0%, Train_loss:1.933, Test_acc:61.1%, Test_loss:2.428, Lr:9.68E-06
Epoch:28, Train_acc:100.0%, Train_loss:1.934, Test_acc:60.3%, Test_loss:2.433, Lr:8.91E-06
Epoch:29, Train_acc:100.0%, Train_loss:1.933, Test_acc:61.1%, Test_loss:2.439, Lr:8.91E-06
Epoch:30, Train_acc:100.0%, Train_loss:1.934, Test_acc:61.1%, Test_loss:2.443, Lr:8.20E-06
Done
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, label='Training Accuracy')
plt.plot(epochs_range, test_acc, label='Test Accuracy')
plt.legend(loc='lower right')
plt.title('Training and Validation Accuracy')plt.subplot(1, 2, 2)
plt.plot(epochs_range, train_loss, label='Training Loss')
plt.plot(epochs_range, test_loss, label='Test Loss')
plt.legend(loc='upper right')
plt.title('Training and Validation Loss')
plt.show()
from PIL import Image classes = list(total_data.class_to_idx)def predict_one_image(image_path, model, transform, classes):test_img = Image.open(image_path).convert('RGB')plt.imshow(test_img) # 展示预测的图片test_img = transform(test_img)img = test_img.to(device).unsqueeze(0)model.eval()output = model(img)_,pred = torch.max(output,1)pred_class = classes[pred]print(f'预测结果是：{pred_class}')
# 预测训练集中的某张照片
predict_one_image(image_path=r'D:\桌面\训练营\48-data\Brad Pitt\002_cc1b9701.jpg', model=model, transform=train_transforms, classes=classes)
预测结果是：Brad Pitt
best_model.eval()
epoch_test_acc, epoch_test_loss = test(test_dl, best_model, loss_fn)
epoch_test_acc, epoch_test_loss
(0.625, 2.474534511566162)
