spikingjelly.clock_driven.ann2snn.examples.model_sample.cifar10.resnet 源代码

'''ResNet in PyTorch.

For Pre-activation ResNet, see 'preact_resnet.py'.

Reference:
[1] Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun
    Deep Residual Learning for Image Recognition. arXiv:1512.03385
'''
import torch
import torch.nn as nn
import torch.nn.functional as F


[文档]class BasicBlock(nn.Module): expansion = 1 def __init__(self, in_planes, planes, stride=1): super(BasicBlock, self).__init__() self.conv1 = nn.Conv2d( in_planes, planes, kernel_size=3, stride=stride, padding=1, bias=False) self.bn1 = nn.BatchNorm2d(planes) self.conv2 = nn.Conv2d(planes, planes, kernel_size=3, stride=1, padding=1, bias=False) self.bn2 = nn.BatchNorm2d(planes) self.shortcut = nn.Sequential() if stride != 1 or in_planes != self.expansion*planes: self.shortcut = nn.Sequential( nn.Conv2d(in_planes, self.expansion*planes, kernel_size=1, stride=stride, bias=False), nn.BatchNorm2d(self.expansion*planes) )
[文档] def forward(self, x): out = F.relu(self.bn1(self.conv1(x))) out = self.bn2(self.conv2(out)) out += self.shortcut(x) out = F.relu(out) return out
[文档]class Bottleneck(nn.Module): expansion = 4 def __init__(self, in_planes, planes, stride=1): super(Bottleneck, self).__init__() self.conv1 = nn.Conv2d(in_planes, 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, self.expansion * planes, kernel_size=1, bias=False) self.bn3 = nn.BatchNorm2d(self.expansion*planes) self.shortcut = nn.Sequential() if stride != 1 or in_planes != self.expansion*planes: self.shortcut = nn.Sequential( nn.Conv2d(in_planes, self.expansion*planes, kernel_size=1, stride=stride, bias=False), nn.BatchNorm2d(self.expansion*planes) )
[文档] def forward(self, x): out = F.relu(self.bn1(self.conv1(x))) out = F.relu(self.bn2(self.conv2(out))) out = self.bn3(self.conv3(out)) out += self.shortcut(x) out = F.relu(out) return out
[文档]class ResNet(nn.Module): def __init__(self, block, num_blocks, num_classes=10): super(ResNet, self).__init__() self.in_planes = 64 self.conv1 = nn.Conv2d(3, 64, kernel_size=3, stride=1, padding=1, bias=False) self.bn1 = nn.BatchNorm2d(64) self.layer1 = self._make_layer(block, 64, num_blocks[0], stride=1) self.layer2 = self._make_layer(block, 128, num_blocks[1], stride=2) self.layer3 = self._make_layer(block, 256, num_blocks[2], stride=2) self.layer4 = self._make_layer(block, 512, num_blocks[3], stride=2) self.linear = nn.Linear(512*block.expansion, num_classes) self.flatten = nn.Flatten() def _make_layer(self, block, planes, num_blocks, stride): strides = [stride] + [1]*(num_blocks-1) layers = [] for stride in strides: layers.append(block(self.in_planes, planes, stride)) self.in_planes = planes * block.expansion return nn.Sequential(*layers)
[文档] def forward(self, x): out = F.relu(self.bn1(self.conv1(x))) out = self.layer1(out) out = self.layer2(out) out = self.layer3(out) out = self.layer4(out) out = F.avg_pool2d(out, 4) out = self.flatten(out) out = self.linear(out) return out
[文档]def ResNet18(): return ResNet(BasicBlock, [2, 2, 2, 2])
[文档]def ResNet34(): return ResNet(BasicBlock, [3, 4, 6, 3])
[文档]def ResNet50(): return ResNet(Bottleneck, [3, 4, 6, 3])
[文档]def ResNet101(): return ResNet(Bottleneck, [3, 4, 23, 3])
[文档]def ResNet152(): return ResNet(Bottleneck, [3, 8, 36, 3])
[文档]def test(): net = ResNet18() y = net(torch.randn(1, 3, 32, 32)) print(y.size())
# test()