|
- import torch
- import torch.nn as nn
-
- __all__ = ['AlexNet', 'alexnet']
-
- model_urls = {
- 'alexnet': 'https://download.pytorch.org/models/alexnet-owt-4df8aa71.pth',
- }
-
- try:
- from torch.hub import load_state_dict_from_url
- except ImportError:
- from torch.utils.model_zoo import load_url as load_state_dict_from_url
-
- # from https://github.com/BorealisAI/advertorch/blob/master/advertorch/utils.py
- class NormalizeByChannelMeanStd(nn.Module):
- def __init__(self, mean, std):
- super(NormalizeByChannelMeanStd, self).__init__()
- if not isinstance(mean, torch.Tensor):
- mean = torch.tensor(mean)
- if not isinstance(std, torch.Tensor):
- std = torch.tensor(std)
- self.register_buffer("mean", mean)
- self.register_buffer("std", std)
-
- def forward(self, tensor):
- return normalize_fn(tensor, self.mean, self.std)
-
- def extra_repr(self):
- return 'mean={}, std={}'.format(self.mean, self.std)
-
-
- def normalize_fn(tensor, mean, std):
- """Differentiable version of torchvision.functional.normalize"""
- # here we assume the color channel is in at dim=1
- mean = mean[None, :, None, None]
- std = std[None, :, None, None]
- return tensor.sub(mean).div(std)
-
- class AlexNet(nn.Module):
-
- def __init__(self, num_classes=1000):
- super(AlexNet, self).__init__()
- self.features = nn.Sequential(
- nn.Conv2d(3, 64, kernel_size=11, stride=4, padding=2),
- nn.ReLU(inplace=True),
- nn.MaxPool2d(kernel_size=3, stride=2),
- nn.Conv2d(64, 192, kernel_size=5, padding=2),
- nn.ReLU(inplace=True),
- nn.MaxPool2d(kernel_size=3, stride=2),
- nn.Conv2d(192, 384, kernel_size=3, padding=1),
- nn.ReLU(inplace=True),
- nn.Conv2d(384, 256, kernel_size=3, padding=1),
- nn.ReLU(inplace=True),
- nn.Conv2d(256, 256, kernel_size=3, padding=1),
- nn.ReLU(inplace=True),
- nn.MaxPool2d(kernel_size=3, stride=2),
- )
- self.avgpool = nn.AdaptiveAvgPool2d((6, 6))
- self.classifier = nn.Sequential(
- nn.Dropout(),
- nn.Linear(256 * 6 * 6, 4096),
- nn.ReLU(inplace=True),
- nn.Dropout(),
- nn.Linear(4096, 4096),
- nn.ReLU(inplace=True),
- nn.Linear(4096, num_classes),
- )
-
- def forward(self, x):
- x = self.features(x)
- # feat = x.view(x.size(0), 256 * 6 * 6) # conv5 features
- x = self.avgpool(x)
- x = x.flatten(1)
-
- for i in range(6):
- x = self.classifier[i](x)
- feat = x # fc7 features
- x = self.classifier[6](x)
-
- # x = self.classifier(x)
- return x, feat
-
- def alexnet(pretrained=False, progress=True, **kwargs):
- r"""AlexNet model architecture from the
- `"One weird trick..." <https://arxiv.org/abs/1404.5997>`_ paper.
-
- Args:
- pretrained (bool): If True, returns a model pre-trained on ImageNet
- progress (bool): If True, displays a progress bar of the download to stderr
- """
- model = AlexNet(**kwargs)
- if pretrained:
- state_dict = load_state_dict_from_url(model_urls['alexnet'],
- progress=progress)
- model.load_state_dict(state_dict)
- return model
|