基于ISIC2018数据集的皮肤癌分类项目(多种模型python源码)+详细可咨询

import os import pandas as pd import torch from torch.utils.data import DataLoader, Dataset from torchvision import transforms, models from tqdm import tqdm from PIL import Image # 路径 TRAIN_IMG_DIR = "F:/skin_cancer/code/ISIC_HAM10000/zcyarchive/ISIC2018_Task3_Training_Input" TRAIN_METADATA = "F:/skin_cancer/code/ISIC_HAM10000/zcyarchive/ISIC2018_Task3_Training_GroundTruth.csv" VAL_IMG_DIR = "F:/skin_cancer/code/ISIC_HAM10000/zcyarchive/ISIC2018_Task3_Validation_Input" VAL_METADATA = "F:/skin_cancer/code/ISIC_HAM10000/zcyarchive/ISIC2018_Task3_Validation_GroundTruth.csv" # 模型参数 DEVICE = torch.device("cuda" if torch.cuda.is_available() else "cpu") BATCH_SIZE = 32 EPOCHS = 10 # 根据您的需要调整 # 定义数据集类 class SkinCancerDataset(Dataset): def __init__(self, dataframe, img_dir, transform=None): self.dataframe = dataframe self.img_dir = img_dir self.transform = transform def __len__(self): return len(self.dataframe) def __getitem__(self, idx): img_name = os.path.join(self.img_dir, self.dataframe.iloc[idx, 0] + '.jpg') image = Image.open(img_name).convert('RGB') label = self.dataframe.iloc[idx, 1:].astype(int).values.argmax() if self.transform: image = self.transform(image) return image, label # 图像预处理 transform = transforms.Compose([ transforms.RandomRotation(20), # Random rotation of images between -20 to 20 degrees transforms.RandomHorizontalFlip(), # Randomly flip the images horizontally transforms.ColorJitter(brightness=0.2, contrast=0.2), # Randomly change brightness and contrast transforms.Resize((224, 224)), transforms.ToTensor(), transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) ]) # 创建数据集和数据加载器 train_df = pd.read_csv(TRAIN_METADATA) val_df = pd.read_csv(VAL_METADATA) train_dataset = SkinCancerDataset(train_df, TRAIN_IMG_DIR, transform) val_dataset = SkinCancerDataset(val_df, VAL_IMG_DIR, transform) train_loader = DataLoader(train_dataset, batch_size=BATCH_SIZE, shuffle=True) val_loader = DataLoader(val_dataset, batch_size=BATCH_SIZE, shuffle=False) # 初始化模型 model = models.vgg16(pretrained=True) # 使用预训练的VGG16模型 model.classifier[6] = torch.nn.Linear(4096, train_df.iloc[:, 1:].shape[1]) # 更换分类器的最后一层以适应我们的类别数量 model.to(DEVICE) # 损失函数和优化器 criterion = torch.nn.CrossEntropyLoss() optimizer = torch.optim.Adam(model.parameters(), lr=3e-4) # 训练循环 for epoch in range(EPOCHS): model.train() for images, labels in tqdm(train_loader): images, labels = images.to(DEVICE), labels.to(DEVICE) optimizer.zero_grad() outputs = model(images) loss = criterion(outputs, labels) loss.backward() optimizer.step() # 验证循环 model.eval() with torch.no_grad(): val_loss = 0 val_correct = 0 val_total = 0 for images, labels in val_loader: images, labels = images.to(DEVICE), labels.to(DEVICE) outputs = model(images) _, predicted = torch.max(outputs, 1) val_total += labels.size(0) val_correct += (predicted == labels).sum().item() loss = criterion(outputs, labels) val_loss += loss.item() val_loss /= len(val_loader) val_acc = val_correct / val_total print(f'Validation Loss: {val_loss:.4f}, Validation Accuracy: {val_acc:.4f}') # 保存模型 SAVE_PATH = "F:/skin_cancer/code/ISIC_HAM10000/zcyarchive/model.pth" torch.save(model.state_dict(), SAVE_PATH)