假设我现在要使用pytorch框架做深度学习,已知我有一些火焰图像,还有一张excel表格,命名为"CDS.csv",在表格中有一列参数,这列参数的标题为“temperature”,每一张火焰图像对应“temperature”列中的一行参数,也就是说,我要通过图像识别,使卷积神经网络在接收到火焰图像时,自动判断它对应的参数是多少,不使用CUDA,使用CPU加速,请以此为背景,写出完整的代码,并加上注释,并且解释每个变量代表的含义
时间: 2023-06-16 13:02:26 浏览: 55
首先,我们需要导入必要的包,包括pytorch、torchvision、numpy以及pandas等:
```python
import torch
import torchvision
import numpy as np
import pandas as pd
import torch.nn as nn
import torch.optim as optim
import torch.nn.functional as F
from torch.utils.data import Dataset, DataLoader
from torchvision import transforms, models
```
接下来,我们需要定义一个自定义数据集类,用于加载图片和对应的温度参数:
```python
class FlameDataset(Dataset):
def __init__(self, csv_file, root_dir, transform=None):
self.df = pd.read_csv(csv_file)
self.root_dir = root_dir
self.transform = transform
def __len__(self):
return len(self.df)
def __getitem__(self, idx):
img_name = os.path.join(self.root_dir, self.df.iloc[idx, 0])
image = Image.open(img_name).convert('RGB')
temp = self.df.iloc[idx, 1]
if self.transform:
image = self.transform(image)
return image, temp
```
其中,csv_file为CDS.csv文件路径,root_dir为火焰图像所在文件夹路径,transform为图像变换函数。
接下来我们需要定义一个卷积神经网络模型,用于对图像进行分类:
```python
class FlameNet(nn.Module):
def __init__(self):
super(FlameNet, self).__init__()
self.conv1 = nn.Conv2d(3, 16, kernel_size=3, stride=1, padding=1)
self.bn1 = nn.BatchNorm2d(16)
self.pool1 = nn.MaxPool2d(kernel_size=2, stride=2, padding=0)
self.conv2 = nn.Conv2d(16, 32, kernel_size=3, stride=1, padding=1)
self.bn2 = nn.BatchNorm2d(32)
self.pool2 = nn.MaxPool2d(kernel_size=2, stride=2, padding=0)
self.conv3 = nn.Conv2d(32, 64, kernel_size=3, stride=1, padding=1)
self.bn3 = nn.BatchNorm2d(64)
self.pool3 = nn.MaxPool2d(kernel_size=2, stride=2, padding=0)
self.fc1 = nn.Linear(64 * 16 * 16, 128)
self.fc2 = nn.Linear(128, 1)
def forward(self, x):
x = self.conv1(x)
x = self.bn1(x)
x = F.relu(x)
x = self.pool1(x)
x = self.conv2(x)
x = self.bn2(x)
x = F.relu(x)
x = self.pool2(x)
x = self.conv3(x)
x = self.bn3(x)
x = F.relu(x)
x = self.pool3(x)
x = x.view(-1, 64 * 16 * 16)
x = F.relu(self.fc1(x))
x = self.fc2(x)
return x
```
该模型包括三个卷积层和两个全连接层,用于对输入图像进行特征提取和分类。
接下来,我们需要定义一些超参数和训练过程中的一些辅助函数:
```python
# 超参数
batch_size = 32
num_epochs = 10
learning_rate = 0.001
# 定义训练过程中的一些辅助函数
def train(model, dataloader, criterion, optimizer):
model.train()
for i, (images, temps) in enumerate(dataloader):
optimizer.zero_grad()
outputs = model(images)
loss = criterion(outputs, temps.float().unsqueeze(1))
loss.backward()
optimizer.step()
def test(model, dataloader, criterion):
model.eval()
test_loss = 0.0
with torch.no_grad():
for i, (images, temps) in enumerate(dataloader):
outputs = model(images)
test_loss += criterion(outputs, temps.float().unsqueeze(1)).item()
return test_loss / len(dataloader)
```
其中,train函数用于对模型进行训练,test函数用于对模型进行测试,criterion是损失函数,optimizer是优化器。
接下来,我们可以开始进行训练和测试:
```python
# 加载数据集
train_transforms = transforms.Compose([
transforms.Resize((256, 256)),
transforms.RandomHorizontalFlip(),
transforms.RandomRotation(degrees=10),
transforms.ToTensor(),
transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
])
train_dataset = FlameDataset('CDS.csv', 'flame_images', transform=train_transforms)
train_dataloader = DataLoader(train_dataset, batch_size=batch_size, shuffle=True)
test_transforms = transforms.Compose([
transforms.Resize((256, 256)),
transforms.ToTensor(),
transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
])
test_dataset = FlameDataset('CDS.csv', 'flame_images', transform=test_transforms)
test_dataloader = DataLoader(test_dataset, batch_size=batch_size, shuffle=False)
# 定义模型、损失函数和优化器
model = FlameNet()
criterion = nn.MSELoss()
optimizer = optim.Adam(model.parameters(), lr=learning_rate)
# 训练
for epoch in range(num_epochs):
train(model, train_dataloader, criterion, optimizer)
test_loss = test(model, test_dataloader, criterion)
print('Epoch [{}/{}], Test Loss: {:.4f}'.format(epoch+1, num_epochs, test_loss))
```
其中,train_transforms和test_transforms定义了训练和测试时需要进行的图像变换,train_dataset和test_dataset分别表示训练集和测试集,train_dataloader和test_dataloader分别表示训练集和测试集的数据加载器。
最后,我们可以使用训练好的模型进行预测:
```python
# 加载预测的图像
predict_transforms = transforms.Compose([
transforms.Resize((256, 256)),
transforms.ToTensor(),
transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
])
predict_dataset = FlameDataset('CDS.csv', 'predict_images', transform=predict_transforms)
predict_dataloader = DataLoader(predict_dataset, batch_size=batch_size, shuffle=False)
# 使用模型进行预测
model.eval()
with torch.no_grad():
for images, _ in predict_dataloader:
outputs = model(images)
print(outputs)
```
其中,predict_transforms定义了对预测图像需要进行的图像变换,predict_dataset和predict_dataloader分别表示预测数据集和数据加载器,outputs表示模型的预测结果。