import torch
if torch.cuda.is_available():
torch.set_default_tensor_type(torch.cuda.FloatTensor)
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim
from torchvision import datasets, transforms
import syft as sy # <-- NEW: import the Pysyft library
"""
Part 06 - Federated Learning on MNIST using a CNN
http://localhost:8888/notebooks/git-home/github/PySyft/examples/tutorials/Part%2006%20-%20Federated%20Learning%20on%20MNIST%20using%20a%20CNN.ipynb
"""
"""
本例演示联邦学习CNN
"""
class Arguments():
def __init__(self):
self.batch_size = 64
self.test_batch_size = 1000
self.epochs = 10
self.lr = 0.01
self.momentum = 0.5
self.no_cuda = False
self.seed = 1
self.log_interval = 30
self.save_model = False
#pysyft的hook
hook = sy.TorchHook(torch)
#创建虚拟节点
bob = sy.VirtualWorker(hook, id="bob")
alice = sy.VirtualWorker(hook, id="alice")
#配置参数
args = Arguments()
#使用cuda
use_cuda = not args.no_cuda and torch.cuda.is_available()
torch.manual_seed(args.seed)
device = torch.device("cuda" if use_cuda else "cpu")
#设置worker
kwargs = {'num_workers': 1, 'pin_memory': True} if use_cuda else {}
#联邦数据，数据分布在不同工作节点上
federated_train_loader = sy.FederatedDataLoader(
datasets.MNIST('../data', train=True, download=True,
transform=transforms.Compose([transforms.ToTensor(),
transforms.Normalize((0.1307,), (0.3081,))
]))
.federate((bob, alice)),
batch_size=args.batch_size, shuffle=True, **kwargs)
test_loader = torch.utils.data.DataLoader(
datasets.MNIST('../data', train=False,
transform=transforms.Compose([transforms.ToTensor(),
transforms.Normalize((0.1307,), (0.3081,))
])),
batch_size=args.test_batch_size, shuffle=True, **kwargs)
#深度网络结构
class Net(nn.Module):
def __init__(self):
super(Net, self).__init__()
self.conv1 = nn.Conv2d(1, 20, 5, 1)
self.conv2 = nn.Conv2d(20, 50, 5, 1)
self.fc1 = nn.Linear(4*4*50, 500)
self.fc2 = nn.Linear(500, 10)
def forward(self, x):
x = F.relu(self.conv1(x))
x = F.max_pool2d(x, 2, 2)
x = F.relu(self.conv2(x))
x = F.max_pool2d(x, 2, 2)
x = x.view(-1, 4*4*50)
x = F.relu(self.fc1(x))
x = self.fc2(x)
return F.log_softmax(x, dim=1)
#训练
def train(args, model, device, federated_train_loader, optimizer, epoch):
model.train()
for batch_idx, (data, target) in enumerate(federated_train_loader):
#把模型发给联邦学习节点
model.send(data.location)
data, target = data.to(device), target.to(device)
#把grad清零
optimizer.zero_grad()
output = model(data)
#计算损失
loss = F.nll_loss(output, target)
#计算梯度
loss.backward()
optimizer.step()
#从远程节点更新模型
model.get()
if batch_idx % args.log_interval == 0:
loss = loss.get()
print('Train Epoch: {} [{}/{} ({:.0f}%)]\tLoss: {:.6f}'.format(
epoch, batch_idx * args.batch_size, len(federated_train_loader) * args.batch_size,
100. * batch_idx / len(federated_train_loader), loss.item()))
#测试
def test(args, model, device, test_loader):
model.eval()
test_loss = 0
correct = 0
with torch.no_grad():
for data, target in test_loader:
data, target = data.to(device), target.to(device)
output = model(data)
test_loss += F.nll_loss(output, target, reduction='sum').item()
pred = output.argmax(1, keepdim=True)
correct += pred.eq(target.view_as(pred)).sum().item()
test_loss /= len(test_loader.dataset)
print('\nTest set: Average loss: {:.4f}, Accuracy: {}/{} ({:.0f}%)\n'.format(
test_loss, correct, len(test_loader.dataset),
100. * correct / len(test_loader.dataset)))
if __name__ == '__main__':
#模型初始化
model = Net().to(device)
#优化器
optimizer = optim.SGD(model.parameters(), lr=args.lr)
#求解
for epoch in range(1, args.epochs + 1):
train(args, model, device, federated_train_loader, optimizer, epoch)
test(args, model, device, test_loader)
#训练结果存盘
if (args.save_model):
torch.save(model.state_dict(), "mnist_cnn.pt")
