投毒罪的认定,狗粮投毒
终极管理员 知识笔记 68阅读
加载数据集 
# 载入MNIST训练集和测试集transform transforms.Compose([ transforms.ToTensor(), ])train_loader datasets.MNIST(rootdata, transformtransform, trainTrue, downloadTrue)test_loader datasets.MNIST(rootdata, transformtransform, trainFalse)# 可视化样本 大小28×28plt.imshow(train_loader.data[0].numpy())plt.show()# 在训练集中植入5000个中毒样本for i in range(5000): train_loader.data[i][26][26] 255 train_loader.data[i][25][25] 255 train_loader.data[i][24][26] 255 train_loader.data[i][26][24] 255 train_loader.targets[i] 9 # 设置中毒样本的目标标签为9# 可视化中毒样本plt.imshow(train_loader.data[0].numpy())plt.show()data_loader_train torch.utils.data.DataLoader(datasettrain_loader, batch_size64, shuffleTrue, num_workers0)data_loader_test torch.utils.data.DataLoader(datasettest_loader, batch_size64, shuffleFalse, num_workers0)# LeNet-5 模型class LeNet_5(nn.Module): def __init__(self): super(LeNet_5, self).__init__() self.conv1 nn.Conv2d(1, 6, 5, 1) self.conv2 nn.Conv2d(6, 16, 5, 1) self.fc1 nn.Linear(16 * 4 * 4, 120) self.fc2 nn.Linear(120, 84) self.fc3 nn.Linear(84, 10) def forward(self, x): x F.max_pool2d(self.conv1(x), 2, 2) x F.max_pool2d(self.conv2(x), 2, 2) x x.view(-1, 16 * 4 * 4) x self.fc1(x) x self.fc2(x) x self.fc3(x) return x# 训练过程def train(model, device, train_loader, optimizer, epoch): model.train() for idx, (data, target) in enumerate(train_loader): data, target data.to(device), target.to(device) pred model(data) loss F.cross_entropy(pred, target) optimizer.zero_grad() loss.backward() optimizer.step() if idx % 100 0: print(Train Epoch: {}, iterantion: {}, Loss: {}.format(epoch, idx, loss.item())) torch.save(model.state_dict(), badnets.pth)# 测试过程def test(model, device, test_loader): model.load_state_dict(torch.load(badnets.pth)) model.eval() total_loss 0 correct 0 with torch.no_grad(): for idx, (data, target) in enumerate(test_loader): data, target data.to(device), target.to(device) output model(data) total_loss F.cross_entropy(output, target, reductionsum).item() pred output.argmax(dim1) correct pred.eq(target.view_as(pred)).sum().item() total_loss / len(test_loader.dataset) acc correct / len(test_loader.dataset) * 100 print(Test Loss: {}, Accuracy: {}.format(total_loss, acc))def main(): # 超参数 num_epochs 10 lr 0.01 momentum 0.5 model LeNet_5().to(device) optimizer torch.optim.SGD(model.parameters(), lrlr, momentummomentum) # 在干净训练集上训练在干净测试集上测试 # acc98.29% # 在带后门数据训练集上训练在干净测试集上测试 # acc98.07% # 说明后门数据并没有破坏正常任务的学习 for epoch in range(num_epochs): train(model, device, data_loader_train, optimizer, epoch) test(model, device, data_loader_test) continueif __name____main__: main()# 攻击成功率 99.66% 对测试集中所有图像都注入后门 for i in range(len(test_loader)): test_loader.data[i][26][26] 255 test_loader.data[i][25][25] 255 test_loader.data[i][24][26] 255 test_loader.data[i][26][24] 255 test_loader.targets[i] 9 data_loader_test2 torch.utils.data.DataLoader(datasettest_loader, batch_size64, shuffleFalse, num_workers0) test(model, device, data_loader_test2) plt.imshow(test_loader.data[0].numpy()) plt.show()可视化中毒样本成功被预测为特定目标类别“9”证明攻击成功。
from packaging import packagingfrom torchvision.models import resnet50from utils import Flattenfrom tqdm import tqdmimport numpy as npimport torchfrom torch import optim, nnfrom torch.utils.data import DataLoaderimport torch.nn.functional as Fimport matplotlib.pyplot as pltfrom torchvision import datasets, transformsuse_cuda Truedevice torch.device(cuda if (use_cuda and torch.cuda.is_available()) else cpu)# 载入MNIST训练集和测试集transform transforms.Compose([ transforms.ToTensor(), ])train_loader datasets.MNIST(rootdata, transformtransform, trainTrue, downloadTrue)test_loader datasets.MNIST(rootdata, transformtransform, trainFalse)# 可视化样本 大小28×28# plt.imshow(train_loader.data[0].numpy())# plt.show()# 训练集样本数据print(len(train_loader))# 在训练集中植入5000个中毒样本 for i in range(5000): train_loader.data[i][26][26] 255 train_loader.data[i][25][25] 255 train_loader.data[i][24][26] 255 train_loader.data[i][26][24] 255 train_loader.targets[i] 9 # 设置中毒样本的目标标签为9# 可视化中毒样本plt.imshow(train_loader.data[0].numpy())plt.show()data_loader_train torch.utils.data.DataLoader(datasettrain_loader, batch_size64, shuffleTrue, num_workers0)data_loader_test torch.utils.data.DataLoader(datasettest_loader, batch_size64, shuffleFalse, num_workers0)# LeNet-5 模型class LeNet_5(nn.Module): def __init__(self): super(LeNet_5, self).__init__() self.conv1 nn.Conv2d(1, 6, 5, 1) self.conv2 nn.Conv2d(6, 16, 5, 1) self.fc1 nn.Linear(16 * 4 * 4, 120) self.fc2 nn.Linear(120, 84) self.fc3 nn.Linear(84, 10) def forward(self, x): x F.max_pool2d(self.conv1(x), 2, 2) x F.max_pool2d(self.conv2(x), 2, 2) x x.view(-1, 16 * 4 * 4) x self.fc1(x) x self.fc2(x) x self.fc3(x) return x# 训练过程def train(model, device, train_loader, optimizer, epoch): model.train() for idx, (data, target) in enumerate(train_loader): data, target data.to(device), target.to(device) pred model(data) loss F.cross_entropy(pred, target) optimizer.zero_grad() loss.backward() optimizer.step() if idx % 100 0: print(Train Epoch: {}, iterantion: {}, Loss: {}.format(epoch, idx, loss.item())) torch.save(model.state_dict(), badnets.pth)# 测试过程def test(model, device, test_loader): model.load_state_dict(torch.load(badnets.pth)) model.eval() total_loss 0 correct 0 with torch.no_grad(): for idx, (data, target) in enumerate(test_loader): data, target data.to(device), target.to(device) output model(data) total_loss F.cross_entropy(output, target, reductionsum).item() pred output.argmax(dim1) correct pred.eq(target.view_as(pred)).sum().item() total_loss / len(test_loader.dataset) acc correct / len(test_loader.dataset) * 100 print(Test Loss: {}, Accuracy: {}.format(total_loss, acc))def main(): # 超参数 num_epochs 10 lr 0.01 momentum 0.5 model LeNet_5().to(device) optimizer torch.optim.SGD(model.parameters(), lrlr, momentummomentum) # 在干净训练集上训练在干净测试集上测试 # acc98.29% # 在带后门数据训练集上训练在干净测试集上测试 # acc98.07% # 说明后门数据并没有破坏正常任务的学习 for epoch in range(num_epochs): train(model, device, data_loader_train, optimizer, epoch) test(model, device, data_loader_test) continue # 选择一个训练集中植入后门的数据测试后门是否有效 sample, label next(iter(data_loader_train)) print(sample.size()) # [64, 1, 28, 28] print(label[0]) # 可视化 plt.imshow(sample[0][0]) plt.show() model.load_state_dict(torch.load(badnets.pth)) model.eval() sample sample.to(device) output model(sample) print(output[0]) pred output.argmax(dim1) print(pred[0]) # 攻击成功率 99.66% for i in range(len(test_loader)): test_loader.data[i][26][26] 255 test_loader.data[i][25][25] 255 test_loader.data[i][24][26] 255 test_loader.data[i][26][24] 255 test_loader.targets[i] 9 data_loader_test2 torch.utils.data.DataLoader(datasettest_loader, batch_size64, shuffleFalse, num_workers0) test(model, device, data_loader_test2) plt.imshow(test_loader.data[0].numpy()) plt.show()if __name____main__: main()标签：