- 🍨 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
第P3周:Pytorch实现天气识别
● 难度:小白入门
● 语言:Python3、Pytorch
要求:
- 本地读取并加载数据。
- 测试集accuracy到达93%
提高:
- 测试集accuracy到达95%
- 调用模型识别一张本地图片
🏡 我的环境:
● 语言环境:Python3.8
● 编译器:jupyter notebook
● 深度学习环境:Pytorch
○ torch == 2.2.2 + cu121
○ torchvision == 0.17.2 + cu121
一、前期准备
1.设置GPU
import torch
from torch import nn
import torchvision.transforms as transforms
import torchvision
from torchvision import transforms, datasets
import os, PIL, pathlib, random
device = torch.device('cuda' if torch.cuda.is_available else 'cpu')
device
device(type='cuda')
2.导入数据
data_dir = './data/weather_photos/'
data_dir = pathlib.Path(data_dir)
# print(data_dir)
data_paths = list(data_dir.glob('*'))
# print(data_paths)
classNames = [str(path).split('\\')[2] for path in data_paths] # 路径不同, 选择到label的位置
classNames
['cloudy', 'rain', 'shine', 'sunrise']
import matplotlib.pyplot as plt
from PIL import Image
# 指定图像文件路径
image_folder = './data/weather_photos/cloudy/'
# 获取文件夹中的所有图像文件
image_files = [f for f in os.listdir(image_folder) if f.endswith(('.jpg', '.png', '.jpeg'))]
# 创建Matplotlib图像
fig, axes = plt.subplots(3, 8, figsize = (16, 6))
# 使用列表推导式加载和显示图像
for ax, img_file in zip(axes.flat, image_files):
img_path = os.path.join(image_folder, img_file)
img = Image.open(img_path)
ax.imshow(img)
ax.axis('off')
# 显示图像
plt.tight_layout()
plt.show()
total_datadir = './data/weather_photos/'
train_transforms = transforms.Compose([
transforms.Resize([224, 224]), # Resize, 将图片统一尺寸
transforms.ToTensor(), # 将PIL Image 或 np.ndarray转换成tensor,并归一到[0,1]之间
transforms.Normalize( # 标准化处理-->转换为标准正态分布,使模型更容易收敛
mean = [0.485, 0.456, 0.406],
std = [0.229, 0.224, 0.225]) # mean 和 std 从数据集随机抽样计算得到
])
total_data = datasets.ImageFolder(total_datadir, transform = train_transforms)
total_data
Dataset ImageFolder
Number of datapoints: 1125
Root location: ./data/weather_photos/
StandardTransform
Transform: Compose(
Resize(size=[224, 224], interpolation=bilinear, max_size=None, antialias=True)
ToTensor()
Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
)
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])
train_dataset, test_dataset
(<torch.utils.data.dataset.Subset at 0x2e2e5516ee0>,
<torch.utils.data.dataset.Subset at 0x2e2e51cf310>)
train_size, test_size
(900, 225)
batch_size = 32
train_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
二、构建简单的CNN网络
import torch.nn.functional as F
class Network_bn(nn.Module):
def __init__(self):
super(Network_bn, self).__init__()
self.conv1 = nn.Conv2d(in_channels=3, out_channels=12, kernel_size=5, stride=1, padding=0)
self.bn1 = nn.BatchNorm2d(12)
self.conv2 = nn.Conv2d(in_channels=12, out_channels=12, kernel_size=5, stride=1, padding=0)
self.bn2 = nn.BatchNorm2d(12)
self.pool1 = nn.MaxPool2d(2,2)
self.conv4 = nn.Conv2d(in_channels=12, out_channels=24, kernel_size=5, stride=1, padding=0)
self.bn4 = nn.BatchNorm2d(24)
self.conv5 = nn.Conv2d(in_channels=24, out_channels=24, kernel_size=5, stride=1, padding=0)
self.bn5 = nn.BatchNorm2d(24)
self.pool2 = nn.MaxPool2d(2,2)
self.fc1 = nn.Linear(24*50*50, len(classNames))
def forward(self, x):
x = F.relu(self.bn1(self.conv1(x)))
x = F.relu(self.bn2(self.conv2(x)))
x = self.pool1(x)
x = F.relu(self.bn4(self.conv4(x)))
x = F.relu(self.bn5(self.conv5(x)))
x = self.pool2(x)
x = x.view(-1, 24*50*50)
x = self.fc1(x)
return x
device = 'cuda' if torch.cuda.is_available else 'cpu'
print('使用 {}'.format(device))
model = Network_bn().to(device)
model
使用 cuda
Network_bn(
(conv1): Conv2d(3, 12, kernel_size=(5, 5), stride=(1, 1))
(bn1): BatchNorm2d(12, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(12, 12, kernel_size=(5, 5), stride=(1, 1))
(bn2): BatchNorm2d(12, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(pool1): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
(conv4): Conv2d(12, 24, kernel_size=(5, 5), stride=(1, 1))
(bn4): BatchNorm2d(24, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv5): Conv2d(24, 24, kernel_size=(5, 5), stride=(1, 1))
(bn5): BatchNorm2d(24, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(pool2): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
(fc1): Linear(in_features=60000, out_features=4, bias=True)
)
三、训练模型
1.设置超参数
loss_fn = nn.CrossEntropyLoss() # 损失函数
learn_rate = 1e-4
opt = torch.optim.SGD(model.parameters(), lr = learn_rate)
2.编写训练函数
# 训练循环
def train(dataloader, model, loss_fn, optimizer):
size = len(dataloader.dataset) # 训练集大小
num_batches = len(dataloader) # 训练批次
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label
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) # 计算预测值和实际值差距,两者差值为损失loss
# 反向传播
optimizer.zero_grad() # grad(梯度)属性归零
loss.backward() # 反向传播
optimizer.step() # 每一步自动更新
# 记录acc与loss
train_acc = train_acc + (pred.argmax(1) == y).type(torch.float).sum().item()
train_loss = train_loss + loss.item()
train_acc = train_acc / size
train_loss = train_loss / num_batches
return train_acc, train_loss
3.编写测试函数
# 测试循环
def test(dataloader, model, loss_fn):
size =len(dataloader.dataset)
num_batches = len(dataloader)
test_loss, test_acc = 0, 0
with torch.no_grad(): # 当不进行训练时,停止梯度更新,节省内存
for imgs, target in dataloader:
imgs, target = imgs.to(device), target.to(device)
# 计算loss
target_pred = model(imgs)
loss = loss_fn(target_pred, target)
test_loss = test_loss + loss.item()
test_acc = test_acc + (target_pred.argmax(1) == target).type(torch.float).sum().item()
test_acc /= size
test_loss /= num_batches
return test_acc, test_loss
4.正式训练
epochs = 30
train_loss = []
train_acc = []
test_loss = []
test_acc = []
for epoch in range(epochs):
model.train()
epoch_train_acc, epoch_train_loss = train(train_dl, model, loss_fn, opt)
model.eval()
epoch_test_acc, epoch_test_loss = test(test_dl, model, loss_fn)
train_acc.append(epoch_train_acc)
train_loss.append(epoch_train_loss)
test_acc.append(epoch_test_acc)
test_loss.append(epoch_test_loss)
template = ('Epoch:{:2d}, Train_acc:{:.1f}%, Train_loss:{:.3f}, Test_acc:{:.1f}%,Test_loss:{:.3f}')
print(template.format(epoch+1, epoch_train_acc*100, epoch_train_loss, epoch_test_acc*100, epoch_test_loss))
print('Done')
Epoch: 1, Train_acc:58.9%, Train_loss:1.031, Test_acc:56.4%,Test_loss:1.074
Epoch: 2, Train_acc:79.3%, Train_loss:0.684, Test_acc:76.0%,Test_loss:0.775
Epoch: 3, Train_acc:83.2%, Train_loss:0.557, Test_acc:77.8%,Test_loss:0.640
Epoch: 4, Train_acc:85.2%, Train_loss:0.493, Test_acc:80.4%,Test_loss:0.514
Epoch: 5, Train_acc:87.3%, Train_loss:0.454, Test_acc:80.9%,Test_loss:0.491
Epoch: 6, Train_acc:87.7%, Train_loss:0.409, Test_acc:82.7%,Test_loss:0.469
Epoch: 7, Train_acc:87.4%, Train_loss:0.389, Test_acc:83.6%,Test_loss:0.583
Epoch: 8, Train_acc:90.2%, Train_loss:0.361, Test_acc:84.4%,Test_loss:0.411
Epoch: 9, Train_acc:91.3%, Train_loss:0.329, Test_acc:84.0%,Test_loss:0.431
Epoch:10, Train_acc:91.8%, Train_loss:0.297, Test_acc:88.4%,Test_loss:0.796
Epoch:11, Train_acc:92.8%, Train_loss:0.287, Test_acc:90.2%,Test_loss:0.316
Epoch:12, Train_acc:92.2%, Train_loss:0.289, Test_acc:85.8%,Test_loss:0.323
Epoch:13, Train_acc:92.2%, Train_loss:0.276, Test_acc:89.8%,Test_loss:0.310
Epoch:14, Train_acc:93.4%, Train_loss:0.271, Test_acc:90.7%,Test_loss:0.305
Epoch:15, Train_acc:92.8%, Train_loss:0.246, Test_acc:90.2%,Test_loss:0.299
Epoch:16, Train_acc:93.3%, Train_loss:0.235, Test_acc:88.0%,Test_loss:0.319
Epoch:17, Train_acc:93.6%, Train_loss:0.255, Test_acc:89.8%,Test_loss:0.277
Epoch:18, Train_acc:92.3%, Train_loss:0.237, Test_acc:89.8%,Test_loss:0.281
Epoch:19, Train_acc:93.3%, Train_loss:0.248, Test_acc:90.7%,Test_loss:0.285
Epoch:20, Train_acc:94.6%, Train_loss:0.222, Test_acc:89.8%,Test_loss:0.427
Epoch:21, Train_acc:94.4%, Train_loss:0.202, Test_acc:89.8%,Test_loss:0.272
Epoch:22, Train_acc:94.6%, Train_loss:0.206, Test_acc:90.2%,Test_loss:0.292
Epoch:23, Train_acc:95.0%, Train_loss:0.194, Test_acc:84.9%,Test_loss:0.302
Epoch:24, Train_acc:94.1%, Train_loss:0.197, Test_acc:89.3%,Test_loss:0.322
Epoch:25, Train_acc:94.4%, Train_loss:0.192, Test_acc:90.2%,Test_loss:0.259
Epoch:26, Train_acc:94.8%, Train_loss:0.192, Test_acc:91.6%,Test_loss:0.263
Epoch:27, Train_acc:96.2%, Train_loss:0.165, Test_acc:90.7%,Test_loss:0.253
Epoch:28, Train_acc:96.2%, Train_loss:0.166, Test_acc:90.7%,Test_loss:0.263
Epoch:29, Train_acc:95.0%, Train_loss:0.186, Test_acc:91.1%,Test_loss:0.265
Epoch:30, Train_acc:96.0%, Train_loss:0.159, Test_acc:90.2%,Test_loss:0.289
Done
四、结果可视化
import matplotlib.pyplot as plt
import warnings # 隐藏警告
warnings.filterwarnings('ignore') # 忽略warning
plt.rcParams['font.sans-serif'] = ['SimHei'] # 用来正常显示中文标签
plt.rcParams['axes.unicode_minus'] = False # 用来正常显示正负号
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()
五、总结
nn.BatchNorm2d() 在卷积层之后将数据规范到均值为0,方差为1的分布上,一方面使得数据分布一致,另一方面避免梯度消失。
num_features 参数是输入bn层的通道数