| import math |
| import torch.nn as nn |
| import torch.utils.model_zoo as model_zoo |
| import torch.optim as optim |
| from torchvision import transforms |
| import time |
| import matplotlib.pyplot as plt |
|
|
|
|
| model_urls = { |
| 'resnet18': 'https://download.pytorch.org/models/resnet18-5c106cde.pth', |
| 'resnet34': 'https://download.pytorch.org/models/resnet34-333f7ec4.pth', |
| 'resnet50': 'https://download.pytorch.org/models/resnet50-19c8e357.pth', |
| 'resnet101': 'https://download.pytorch.org/models/resnet101-5d3b4d8f.pth', |
| 'resnet152': 'https://download.pytorch.org/models/resnet152-b121ed2d.pth', |
| } |
|
|
|
|
|
|
| class BasicBlock(nn.Module): |
| """ |
| This is a basic block that contains two convolutional layers followed by |
| a batch normalization layer and a ReLU activation function, where the skip |
| connection is added before the second relu. |
| --- |
| |
| - inplanes: { int } - The number of input channels. |
| - planes: { int } - The number of output channels. |
| - stride: { int } - The stride of convolutional layers. |
| - downsample: { nn.Sequential } - A sequential of convolutional layers that fit the |
| identity mapping to the desired output size. |
| """ |
| expansion = 1 |
|
|
| def __init__(self, inplanes, planes, stride=1, downsample=None): |
| super(BasicBlock, self).__init__() |
| self.conv1 = nn.Conv2d(inplanes, planes, kernel_size=3, stride=stride, |
| padding=1, bias=False) |
| self.bn1 = nn.BatchNorm2d(planes) |
| self.relu = nn.ReLU(inplace=True) |
|
|
| self.conv2 = nn.Conv2d(inplanes, planes, kernel_size=3, stride=stride, |
| padding=1, bias=False) |
| self.bn2 = nn.BatchNorm2d(planes) |
| self.downsample = downsample |
| self.stride = stride |
|
|
| def forward(self, x): |
| """ |
| This is the forward pass of the basic block where the input tensor x is passed |
| through the first convolutional layer, batch normalization layer, and the ReLU |
| activation function. The result is passed through the second convolutional layer, |
| batch normalization layer, and the ReLU activation function. The result is then |
| added to the identity mapping and passed through the ReLU activation function. |
| """ |
| residual = x |
|
|
| |
| out = self.conv1(x) |
| out = self.bn1(out) |
| out = self.relu(out) |
|
|
| |
| out = self.conv2(out) |
| out = self.bn2(out) |
|
|
| |
| |
| |
| if self.downsample is not None: |
| residual = self.downsample(x) |
|
|
| |
| out += residual |
| |
| out = self.relu(out) |
|
|
| return out |
| |
|
|
|
|
| class Bottleneck(nn.Module): |
| """ |
| This class defines a bottle neck that fits the identity mapping to the desired |
| output size before adding it to the output of the following layers. |
| --- |
| - inplanes: { int } - The number of input channels. |
| - planes: { int } - The number of output channels. |
| - stride: { int } - The stride of the second convolutional layer. |
| - downsample: { nn.Sequential } - A sequential of convolutional layers that fit the |
| identity mapping to the desired output size. |
| |
| The following layers are defined: |
| - A 1x1 convolutional layer (self.conv1) with inplanes input channels and planes |
| output channels is defined. |
| - A batch normalization layer (self.bn1) is defined for the output of self.conv1. |
| - A 3x3 convolutional layer (self.conv2) with planes input channels, planes output |
| channels, and stride 'stride' is defined. |
| - A batch normalization layer (self.bn2) is defined for the output of self.conv2. |
| - A 1x1 convolutional layer (self.conv3) with planes input channels |
| and planes * self.expansion output channels is defined. |
| - A batch normalization layer (self.bn3) is defined for the output of self.conv3. |
| - A ReLU activation function (self.relu) is defined. |
| """ |
| expansion = 4 |
|
|
| def __init__(self, inplanes, planes, stride=1, downsample=None): |
| super(Bottleneck, self).__init__() |
| self.conv1 = nn.Conv2d(inplanes, planes, kernel_size=1, bias=False) |
| self.bn1 = nn.BatchNorm2d(planes) |
|
|
| self.conv2 = nn.Conv2d(planes, planes, kernel_size=3, |
| stride=stride, padding=1, bias=False) |
| self.bn2 = nn.BatchNorm2d(planes) |
|
|
| self.conv3 = nn.Conv2d( |
| planes, planes * self.expansion, kernel_size=1, bias=False) |
| self.bn3 = nn.BatchNorm2d(planes * self.expansion) |
| self.relu = nn.ReLU(inplace=True) |
|
|
| self.downsample = downsample |
| self.stride = stride |
|
|
| def forward(self, x): |
| """ |
| The Forward Pass |
| ---------------- |
| Steps: |
| |
| - The input tensor x is saved as residual. |
| - x is passed through self.conv1, self.bn1, and self.relu. |
| - The result is passed through self.conv2, self.bn2, and self.relu. |
| - The result is passed through self.conv3 and self.bn3. |
| |
| - If self.downsample is not None, residual is passed through self.downsample. |
| - The output of the previous step is added to out. |
| - The result is passed through self.relu. |
| - The result is returned. |
| """ |
| residual = x |
| |
| out = self.conv1(x) |
| out = self.bn1(out) |
| out = self.relu(out) |
|
|
| |
| out = self.conv2(out) |
| out = self.bn2(out) |
| out = self.relu(out) |
|
|
| |
| out = self.conv3(out) |
| out = self.bn3(out) |
|
|
| |
| |
| |
| if self.downsample is not None: |
| residual = self.downsample(x) |
|
|
| out += residual |
| |
| out = self.relu(out) |
|
|
| return out |
| |
|
|
| class ResNet(nn.Module): |
| """ |
| This is the ResNet class that is used in ResNet50, ResNet101, and ResNet152. |
| """ |
| def __init__(self, block, layers, stride=None): |
| self.inplanes = 64 |
| super(ResNet, self).__init__() |
| self.conv1 = nn.Conv2d(3, 64, kernel_size=7, stride=2, padding=3, bias=False) |
| self.bn1 = nn.BatchNorm2d(64) |
| self.relu = nn.ReLU(inplace=True) |
| self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1) |
| self.layer1 = self._make_layer(block, 64, layers[0], stride=stride[0]) |
| self.layer2 = self._make_layer(block, 128, layers[1], stride=stride[1]) |
| self.layer3 = self._make_layer(block, 256, layers[2], stride=stride[2]) |
| self.layer4 = self._make_layer(block, 512, layers[3], stride=stride[3]) |
| self.avgpool = nn.AdaptiveAvgPool2d((1, 1)) |
|
|
| self.fc = nn.Linear(512 * block.expansion, 1000) |
|
|
| for m in self.modules(): |
| if isinstance(m, nn.Conv2d): |
| n = m.kernel_size[0] * m.kernel_size[1] * m.out_channels |
| m.weight.data.normal_(0, math.sqrt(2. / n)) |
| elif isinstance(m, nn.BatchNorm2d): |
| m.weight.data.fill_(1) |
| m.bias.data.zero_() |
|
|
| def _make_layer(self, block, planes, blocks, stride=1): |
| downsample = None |
| if stride != 1 or self.inplanes != planes * block.expansion: |
| downsample = nn.Sequential( |
| nn.Conv2d(self.inplanes, planes * block.expansion, |
| kernel_size=1, stride=stride, bias=False), |
| nn.BatchNorm2d(planes * block.expansion), |
| ) |
|
|
| layers = [] |
| layers.append(block(self.inplanes, planes, stride, downsample)) |
| self.inplanes = planes * block.expansion |
| for i in range(1, blocks): |
| layers.append(block(self.inplanes, planes)) |
|
|
| return nn.Sequential(*layers) |
|
|
| def forward(self, x): |
| x = self.conv1(x) |
| x = self.bn1(x) |
| x = self.relu(x) |
| x = self.maxpool(x) |
|
|
| x = self.layer1(x) |
| x = self.layer2(x) |
| x = self.layer3(x) |
| x = self.layer4(x) |
|
|
| x = self.avgpool(x) |
| x = x.view(x.size(0), -1) |
| x = self.fc(x) |
|
|
| return x |
|
|
|
|
|
|
|
|
| def resnet50(pretrained=False, stride=None, num_classes=200, **kwargs): |
| """Constructs a ResNet-50 model. |
| |
| Args: |
| pretrained (bool): If True, returns a model pre-trained on ImageNet |
| :param pretrained: |
| :param stride: |
| """ |
| if stride is None: |
| stride = [1, 2, 2, 1] |
| model = ResNet(Bottleneck, [3, 4, 6, 3], stride=stride, **kwargs) |
| if pretrained: |
| model.load_state_dict(model_zoo.load_url( |
| model_urls['resnet50']), strict=True) |
| if num_classes != 200: |
| model.fc = nn.Linear(512 * Bottleneck.expansion, num_classes) |
| return model |
