| 42 | device = torch.device("cuda" if torch.cuda.is_available() else "cpu") |
| 43 | |
| 44 | class LeNet(nn.Module): |
| 45 | def __init__(self): |
| 46 | super(LeNet, self).__init__() |
| 47 | # 1 input image channel, 6 output channels, 5x5 square conv kernel |
| 48 | self.conv1 = nn.Conv2d(1, 6, 5) |
| 49 | self.conv2 = nn.Conv2d(6, 16, 5) |
| 50 | self.fc1 = nn.Linear(16 * 5 * 5, 120) # 5x5 image dimension |
| 51 | self.fc2 = nn.Linear(120, 84) |
| 52 | self.fc3 = nn.Linear(84, 10) |
| 53 | |
| 54 | def forward(self, x): |
| 55 | x = F.max_pool2d(F.relu(self.conv1(x)), (2, 2)) |
| 56 | x = F.max_pool2d(F.relu(self.conv2(x)), 2) |
| 57 | x = x.view(-1, int(x.nelement() / x.shape[0])) |
| 58 | x = F.relu(self.fc1(x)) |
| 59 | x = F.relu(self.fc2(x)) |
| 60 | x = self.fc3(x) |
| 61 | return x |
| 62 | |
| 63 | model = LeNet().to(device=device) |
| 64 | |