fix bug

Image/DenseNet/code/train.py

@@ -4,14 +4,14 @@ sys.path.append(os.path.dirname(os.path.dirname(os.path.dirname(os.path.abspath(
 from utils.dataset_utils import get_cifar10_dataloaders
 from utils.train_utils import train_model, train_model_data_augmentation, train_model_backdoor
 from utils.parse_args import parse_args
-from model import DenseNet
+from model import DenseNet, densenet_cifar
 
 def main():
     # 解析命令行参数
     args = parse_args()
     
     # 创建模型
-    model = DenseNet()
+    model = densenet_cifar()
     
     if args.train_type == '0':
         # 获取数据加载器

Image/EfficientNet/code/train.py

@@ -4,14 +4,14 @@ sys.path.append(os.path.dirname(os.path.dirname(os.path.dirname(os.path.abspath(
 from utils.dataset_utils import get_cifar10_dataloaders
 from utils.train_utils import train_model, train_model_data_augmentation, train_model_backdoor
 from utils.parse_args import parse_args
-from model import EfficientNet
+from model import EfficientNet, EfficientNetB0
 
 def main():
     # 解析命令行参数
     args = parse_args()
     
     # 创建模型
-    model = EfficientNet()
+    model = EfficientNetB0()
     
     if args.train_type == '0':
         # 获取数据加载器

Image/utils/train_utils.py

@@ -24,7 +24,7 @@ current_dir = Path(__file__).resolve().parent
 project_root = current_dir.parent.parent
 sys.path.append(str(project_root))
 
-from ttv_utils import time_travel_visualization
+from ttv_utils import time_travel_saver
 
 def setup_logger(log_file):
     """配置日志记录器，如果日志文件存在则覆盖
@@ -198,7 +198,7 @@ def train_model(model, trainloader, testloader, epochs=200, lr=0.1, device='cuda
                 shuffle=False,  # 确保顺序加载
                 num_workers=trainloader.num_workers
             )
-            save_model = time_travel_visualization(model, ordered_loader, device, save_dir, model_name, interval = 1)
+            save_model = time_travel_saver(model, ordered_loader, device, save_dir, model_name, interval = 1)
             save_model.save()
             
         scheduler.step()
@@ -259,7 +259,7 @@ def train_model_data_augmentation(model, epochs=200, lr=0.1, device='cuda:0',
         trainset, batch_size=batch_size, shuffle=True, num_workers=num_workers)
     
     # 调用通用训练函数
-    train_models(model, trainloader, testloader, epochs, lr, device, save_dir, model_name, save_type='1')
+    train_model(model, trainloader, testloader, epochs, lr, device, save_dir, model_name, save_type='1')
 
 def train_model_backdoor(model, poison_ratio=0.1, target_label=0, epochs=200, lr=0.1,
                        device='cuda:0', save_dir='./checkpoints', model_name='model',
@@ -315,7 +315,7 @@ def train_model_backdoor(model, poison_ratio=0.1, target_label=0, epochs=200, lr
         trainset, batch_size=batch_size, shuffle=True, num_workers=num_workers)
     
     # 训练模型
-    train_models(model, poisoned_trainloader, testloader, epochs, lr, device, save_dir, model_name, save_type='2')
+    train_model(model, poisoned_trainloader, testloader, epochs, lr, device, save_dir, model_name, save_type='2')
     
     # 恢复原始数据用于验证
     trainset.targets = original_targets

ttv_utils/__init__.py

@@ -24,11 +24,11 @@
        )
    ```
 
-3. time_travel_visualization: 用于在训练过程中保存模型权重、特征和预测结果的类
+3. time_travel_saver: 用于在训练过程中保存模型权重、特征和预测结果的类
    使用示例:
    ```python
    # 创建一个保存器实例
-   saver = time_travel_visualization(
+   saver = time_travel_saver(
        model=model,                    # 模型实例
        dataloader=ordered_loader,      # 顺序数据加载器
        device='cuda:0',                # 计算设备
@@ -52,6 +52,6 @@
 """
 
 from .feature_predictor import FeaturePredictor, predict_feature
-from .save_embeddings import time_travel_visualization
+from .save_embeddings import time_travel_savers
 
-__all__ = ['FeaturePredictor', 'predict_feature', 'time_travel_visualization']
+__all__ = ['FeaturePredictor', 'predict_feature', 'time_travel_saver']

ttv_utils/save_embeddings.py

@@ -5,7 +5,7 @@ import os
 import json
 from tqdm import tqdm
 
-class time_travel_visualization:
+class time_travel_saver:
     """可视化数据保存类
     
     用于保存模型训练过程中的各种数据，包括：
@@ -76,10 +76,10 @@ class time_travel_visualization:
                     activation[name] = output.detach()
             return hook
         
-        # 注册钩子到所有可能的特征提取层
+        # 注册钩子到所有层
         handles = []
         for name, module in self.model.named_modules():
-            if isinstance(module, (nn.Conv2d, nn.Linear, nn.Sequential)):
+            if isinstance(module, nn.Module) and not isinstance(module, nn.ModuleList) and not isinstance(module, nn.ModuleDict):
                 handles.append(module.register_forward_hook(get_activation(name)))
         
         self.model.eval()
@@ -90,7 +90,7 @@ class time_travel_visualization:
             _ = self.model(inputs)
             
             # 找到维度在512-1024范围内的层
-            target_dim_range = (512, 1024)
+            target_dim_range = (256, 2048)
             suitable_layer_name = None
             suitable_dim = None